Lipidomics: Volume 1 (Methods in Molecular Biology, v579)

Methods

in

Molecular Biology™

Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK



For other titles published in this series, go to www.springer.com/series/7651

1

2 3 4 5

Lipidomics Volume 1: Methods and Protocols

Edited by

Donald Armstrong University at Buffalo, Buffalo, NY, USA University of Florida, Gainesville, FL, USA

Editor Donald Armstrong University of Buffalo Buffalo, NY, USA and University of Florida Gainesville, FL, USA

The cover image are overlapping snapshots of a single peroxidized lipid, taken from computer stimulations. Peroxidation mofidies the local conformational preferences of acyl chains and increases their mobility, with implications for structural and dynamic properties of the membrane. ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-60761-321-3 e-ISBN 978-1-60761-322-0 DOI 10.1007/978-1-60761-322-0 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2009927725 © Humana Press, a part of Springer Science+Business Media, LLC 2009 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or ­dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, ­neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Cover illustration: Snapshots of a single PLPC and a peroxidized analogue (13-tc), taken at 5 ns intervals. Molecules are oriented along the membrane and superimposition was done on the phosphorus and oxygen atoms. Images were obtained from Chapter 18, Vol. 1. Cover design: Karen Schulz Printed on acid-free paper Humana Press is part of Springer Science+Business Media (www.springer.com)

sdfsdf

Preface Lipidomics is a sub-discipline of metabolomics and is defined as the large-scale study of non-water-soluble metabolites (lipids and lipidome) that utilize system-level analysis to characterize lipids and their interacting moieties (1). A literature search at the end of 2008 showed there were 200 articles published on lipidomics encompassing glycerophospholipids, sphingolipids, polyunsaturated fatty acids, glycolipids, sterol lipids, and proteolipids. It has been predicted that the combination of these lipid classes totals between 1,000 and 2,000 molecules. Lipids can also act as s­econd messengers, or mitogens, and participate in profiling and signaling via specialized microdomains that have large amounts of lipids. When lipids are disturbed, their metabolites probably contribute to disease. In prostate cancer, for example, cyclooxygenase and lipooxygenase are upregulated reducing angiogenesis and tumor growth. Early separation and identification of lipids started with TLC, and as technology advanced, it progressed to the use of GC and HPLC. Technical improvements to HPLC include reversed-phase methods, ESI, evaporative light scattering, electrochemical detection, APCI, suppressed conductivity and multi-dimensional electrophoresis. Other ­technologies coupled to chromatographic methods, such as MS/MS-MSn/MALDI/ TOF, NMR, and MRM, provide a powerful approach to the global analysis of complex lipid mixtures, understanding structural changes through biophysical approaches and the effects of lipids on physiology, i.e. atherosclerosis. This has given us a clearer understanding of human and animal pathology, i.e. diabetes, cancer, neurodegeneration, and infectious disease. A new approach to measure oxidized lipids, referred to as “oxidative lipidomics,” has recently been described which provides methodology for separation and identification of these highly reactive lipids, especially in mitochondria. Many novel techniques are described in these volumes, including an imaging lipidomics approach. For another lipidomic approach of a lipid-derived radical technique, the reader is referred to Iwabashi, H., 2008. Advanced Protocols in Oxidative Stress I, volume 477, Chapter 6, Humana Press. In that same volume, a lipidomics technique for sphingolipids is also described, i.e. Wilder, AJ and Cowart, LA, Chapter 28. The present volumes have taken a “shotgun” approach and are divided into seven parts in order to include as many different varieties of technology as possible. Chapters by international experts present a wide variety of reviewed as well as unpublished data on ­isolation techniques, structural analysis, lipid rafts, lipid trafficking and profiling, biomarkers, lipid peroxidation, biostatistics applied to lipids, software tools, and bioinformatics. These studies range from simple systems, such as in yeast, to complex biological models. The ever increasing utilization of lipidomics will lead to more powerful technology, improved diagnostic–prognostic capabilities for medical disorders, and for the identification of new classes of lipids.

v

vi

Preface

I thank my son, Dennis Armstrong, and my grandson, David Armstrong of On-Staff Technology, Inc., for assistance with technical support, information technology, and multi-media services. Buffalo, NY Gainesville, FL

Donald Armstrong

Reference 1. Wenk, M.R. 2005. The emerging field of lipidomics. Nature Reviews 4: 595–610.





vii

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v ix

Part I  Shotgun and Global Lipidomics   1  Examination of the Brain Mitochondrial Lipidome Using Shotgun Lipidomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Michael A. Kiebish, Xianlin Han, and Thomas N. Seyfried   2  Tracking the Glycerophospholipid Distribution of Docosahexaenoic Acid by Shotgun Lipidomics . . . . . . . . . . . . . . . . . . . . . . . . Todd W. Mitchell   3  Global Analysis of Retina Lipids by Complementary Precursor Ion and Neutral Loss Mode Tandem Mass Spectrometry . . . . . . . . . . . Julia V. Busik, Gavin E. Reid, and Todd A. Lydic   4  Combining Lipidomics and Proteomics of Human Cerebrospinal Fluids . . . . . . . Alfred N. Fonteh and Rachel D. Fisher

3

19

33 71

Part II Analytical Approaches   5  Lipid Profiling Using Two-Dimensional Heteronuclear Single Quantum Coherence NMR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Engy A. Mahrous, Robin B. Lee, and Richard E. Lee   6  Capabilities and Drawbacks of Phospholipid Analysis by MALDI-TOF Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Beate Fuchs, Ariane Nimptsch, Rosmarie Süß, and Jürgen Schiller   7  Lipidomics of the Red Cell in Diagnosis of Human Disorders . . . . . . . . . . . . . . . Peter J. Quinn , Dominique Rainteau , and Claude Wolf   8  HPLC/MS/MS-Based Approaches for Detection and Quantification of Eicosanoids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Susanna L. Lundström, Fabio L. D´Alexandri, Kasem Nithipatikom, Jesper Z. Haeggström Åsa M. Wheelock, and Craig E. Wheelock   9  Brain Phosphoinositide Extraction, Fractionation, and Analysis by MALDI-TOF MS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Roy A. Johanson and Gerard T. Berry 10  Lipidomic Analysis of Biological Samples by Liquid Chromatography Coupled to Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . Giuseppe Astarita, Faizy Ahmed, and Daniele Piomelli 11  Lipidomic Analysis of Human Meibum Using HPLC–MS . . . . . . . . . . . . . . . . . . Igor A. Butovich

vii

89

103 127

161

189

201 221

viii

Contents

12  Lipid Geographical Analysis of the Primate Macula by Imaging Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Timothy J. Garrett and William W. Dawson 13  A Novel Role for Nutrition in the Alteration of Functional Microdomains on the Cell Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wooki Kim , Robert S. Chapkin, Rola Barhoumi, and David W.L. Ma 14  Lipidomic Analysis of Prostanoids by Liquid Chromatography–Electrospray Tandem Mass Spectrometry . . . . . . . . . . . . . . . . . Anna Nicolaou, Mojgan Masoodi, and Adnan Mir 15  Qualitative and Quantitative Analyses of Phospholipids by LC–MS for Lipidomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hiroki Nakanishi, Hideo Ogiso, and Ryo Taguchi 16  Determination of Fatty Acid Profiles and TAGs in Vegetable Oils by MALDI-TOF/MS Fingerprinting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zeev Wiesman and Bishnu P. Chapagain 17  Dynamics of Adipose Tissue Development by 2 H 2O Labeling . . . . . . . . . . . . . . . Etienne Pouteau, Carine Beysen, Nabil Saad, and Scott Turner 18  Analysis of Lipid Particles from Yeast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Melanie Connerth, Karlheinz Grillitsch, Harald Köfeler, and Günther Daum 19  Mammalian Fatty Acid Elongases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Donald B. Jump 20  Membrane Lipidomics and the Geometry of Unsaturated Fatty Acids : From Biomimetic Models to Biological Consequences . . . . . . . . . . . . . . . . . . . . . Carla Ferreri and Chryssostomos Chatgilialoglu 21  OnLine Ozonolysis Methods for the Determination of Double Bond Position in Unsaturated Lipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Michael C. Thomas, Todd W. Mitchell, and Stephen J. Blanksby 22  Comprehensive Quantitative Analysis of Bioactive Sphingolipids High-Performance Liquid Chromatography-Tandem Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jacek Bielawski, Jason S. Pierce, Justin Snider, Barbara Rembiesa, Zdzislaw M. Szulc, and Alicja Bielawska 23  The Use of Charged Aerosol Detection with HPLC for the Measurement of Lipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Marc Plante, Bruce Bailey, and Ian Acworth

247

261

271

287

315 337 359

375

391

413

443

469

Part III Lipid Maps 24  Non-invasive Mapping of Lipids in Plant Tissue Using Magnetic Resonance Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 485 Thomas Neuberger, Hardy Rolletschek, Andrew Webb, and Ljudmilla Borisjuk



Contents

ix

25  Mapping the Lipolytic Proteome of Adipose Tissue Using Fluorescent Suicide Inhibitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 497 Maximilian Schicher, Manfred Kollroser, and Albin Hermetter 26  Identifying the Spatial Distribution of Vitamin E, Pulmonary Surfactant and Membrane Lipids in Cells and Tissue by Confocal Raman Microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 513 J. Renwick Beattie and Bettina C. Schock Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537

sdfsdf

Contributors Ian Acworth  •  ESA Biosciences, Inc., Chelmsford, MA, USA Faizy Ahmed  •  Agilent Technologies, Irvine/Agilent Analytical Discovery Facility, University of California – Irvine, Irvine, CA, USA Giuseppi Astarita  •  Department of Pharmacology, University of California – Irvine, Irvine, CA, USA Bruce Bailey  •  Applications Department, ESA Biosciences, Inc., Chelmsford, MA, USA Rola Barhoumi  •  Department of Nutritional Sciences, University of Toronto Faculty of Medicine, Toronto, ON, Canada; Center for Environmental and Rural Health, Texas A&M University, College Station, TX, USA J. Renwick Beattie  •  School of Medicine, Dentistry and Biomedical Sciences, Queen’s University, Belfast, UK Gerard T. Berry  •  Division of Genetics, Children’s Hospital, Boston, MA, USA Carine Beysen  •  KineMed, Inc., Emeryville, CA, USA Alicja Bielawska  •  Department of Biochemistry and Molecular Biology, Lipidomic Core Mass Spectrometry Lab, Medical University of South Carolina, Charleston, SC, USA Jacek Bielawski  •  Department of Biochemistry and Molecular Biology, Lipidomic Core Mass Spectrometry Lab, Medical University of South Carolina, Charleston, SC, USA Stephen J. Blanksby  •  School of Chemistry, University of Wollongong, Wollongong, NSW, Australia Ljudmilla Borisjuk  •  Leibniz-Institut fur Pflanzengenetik und Kulturpflanzenforschung (IPK), Gaterslaben, Germany Julia V. Busik  •  Department of Physiology, Michigan State University, East Lansing, MI, USA Igor A. Butovich  •  Department of Ophthalmology, University of Texas Southwestern Medical Center, Dallas, TX, USA Bishnu P. Chapagain  •  The Phyto-Lipid Biotechnology Laboratory, Department of Biotechnology Engineering, The Institutes for Applied Research, Ben-Gurion University of the Negev, Beer-Sheva, Israel Robert S. Chapkin  •  Program in Integrative Nutrition & Complex Diseases, Genomics & Bioinformatics Facility Core Center for Environmental and Rural Health, Kieberg Biotechnology Center, Texas A&M University, College Station, TX, USA Chryssostomos Chatgilialoglu  •  ISOF, Consiglio Nazionale delle Ricerche, Bologna, Italy Melanie Connerth  •  Institute of Biochemistry, Graz University of Technology, Austria

xi

xii

Contributors

Fabio Luiz D’Alexandri  •  Department of Medical Biochemistry and Biophysics, Division of Physiological Chemistry II, Karolinska Institutet, Stockholm, Sweden; Department of Parasitology, Department of Biochemical Sciences, University of San Paulo, San Paulo, Brazil Günther Daum  •  Institute of Biochemistry, Graz University of Technology, Graz, Austria William W. Dawson  •  Department of Ophthalmolgy, College of Medicine, University of Florida Health Sciences Center, Gainesville, FL, USA Carla Ferreri  •  ISOF-BioFree Radicals, Consiglio Nazionale delle Riceriche, Bologna, Italy Rachel D. Fisher  •  Scripps College, Claremont, CA, USA Alfred N. Fonteh  •  Molecular Neurology Program, Huntington Medical Research Institutes, Pasadena, CA, USA Beate Fuchs  •  Medical Department, Institute of Medical Physics and Biophysics, University of Leipzig, Leipzig, Germany Timothy J. Garrett  •  Department of Medicine, College of Medicine, University of Florida Health Science Center, Gainesville, FL, USA Karlheinz Grillitsch  •  Institute of Biochemistry, Graz University of Technology, Graz, Austria Jesper Z. Haeggström  •  Department of Medical Biochemistry and Biophysics, Division of Physiological Chemistry II, Karolinska Institutet, Stockholm, Sweden Xianlin Han  •  Department of Internal Medicine, Washington University School of Medicine, St. Louis, MO, USA Albin Hermetter  •  Institute of Biochemistry, Graz University of Technology, Graz, Austria Roy A. Johanson  •  Department of Neurology, Thomas Jefferson University Medical Center, Philadelphia, PA, USA Donald B. Jump  •  Department of Nutrition and Exercise Sciences, The Linus Pauling Institute, Oregon State University, Corvallis, OR, USA Michael A. Kiebish  •  Biology Department, Boston College, MA, USA Wooki Kim  •  Department of Nutritional Sciences, University of Toronto Faculty of Medicine, Toronto, ON, Canada; Center for Environmental and Rural Health, Texas A&M University, College Station, TX, USA Harald Köfeler  •  Institute of Biochemistry, Graz University of Technology, Graz, Austria Manfred Kollroser  •  Institute of Forensic Medicine, Medical University of Graz, Graz, Austria Richard E. Lee  •  Department of Pharmaceutical Sciences, College of Pharmacy, University of Tennessee Health Science Center, Memphis, TN, USA Robin B. Lee  •  Department of Pharmaceutical Sciences, College of Pharmacy, University of Tennessee Health Sciences Center, Memphis, TN, USA Susanna L. Lundström  •  Department of Medical Biochemistry and Biophysics, Division of Physiological Chemistry II, Karolinska Institute, Stockholm, Sweden Todd A. Lydic  •  Department of Physiology, Michigan State University, MI, USA David W.L. Ma  •  Department of Nutritional Sciences, University of Toronto Faculty of Medicine, Toronto, ON, Canada; Center for Environmental and Rural Health, Texas A&M University, College Station, TX, USA



Contributors

xiii

Engy A. Mahrous  •  Department of Pharmaceutical Sciences, College of Pharmacy, University of Tennessee Health Sciences Center, Memphis, TN, USA Mojgan Masoodi  •  Division Pharmaceutics and Pharmaceutical Chemistry, School of Pharmacy, Life Sciences, University of Bradford, West Yorkshire, UK Adnan Mir  •  Division Pharmaceutics and Pharmaceutical Chemistry, School of Pharmacy , Life Sciences, University of Bradford, West Yorkshire, UK Todd W. Mitchell  •  School of Health Sciences, University of Wollongong, Wollongong, NSW, Australia Hiroki Nakanishi  •  Department of Metabolome, Graduate School of Medicine, The University of Tokyo and Core Research for Evolutional Science and Technology, Saitama, Japan Thomas Neuberger  •  Department of Bioengineering, Pennsylvania State University, University Park, PA, USA Anna Nicolaou  •  Division of Pharmaceutics and Pharmacological Chemistry, School of Pharmacy, Life Sciences, University of Bradford, West Yorkshire, UK Ariane Nimptsch  •  Medical Department, Institute of Medical Physics and Biophysics, University of Leipzig, Leipzig, Germany Kasem Nithipatikom  •  Department of Pharmacology and Toxicology, Medical College of Wisconsin, WI, USA Hideo Ogiso  •  Department of Metabolome, Graduate School of Medicine, The University of Tokyo and Core Research for Evolutional Science and Technology, Saitama, Japan Jason S. Pierce  •  Department of Biochemistry and Molecular Biology, Lipidomic Core Mass Spectrometry Lab, Medical University of south Carolina, Charleston, SC, USA Daniele Piomelli  •  Department of Pharmacology, University of California – Irvine, Irvine, CA, USA Marc Plante  •  Applications Department , ESA Biosciences, Inc., Chelmsford, MA, USA Etienne Pouteau  •  Nutrition and Health Department, Nestle Research Center, Lausanne, Switzerland Paul J. Quinn  •  Life Sciences, King’s College, London, UK Dominique Rainteau  •  UMRS 893-Faculte de Medicine, Pierre et Marie Curie, CHU Saint Antoine, Paris, France Gavin E. Reid  •  Departments of Chemistry, Biochemistry and Molecular Biology, Michigan State University, East Lansing, MI, USA Barbara Rembiesa  •  Department of Biochemistry and Molecular Biology, Lipidomics Core Mass Spectrometry Lab, Medical University of South Carolina, Charleston, SC, USA Hardy Rolletschek  •  Leibniz-Institut fur Pflanzengenetik und Kulturpflanzenforschung ( IPK ), Gaterslaben, Germany Nabil Saad  •  KineMed, Inc., Emeryville, CA, USA Maximilian Schicher  •  Institute of Biochemistry, University of Graz, Graz, Austria Jürgen Schiller  •  University of Leipzig Medical Faculty, Leipzig, Germany Bettina C. Schock  •  School of Medicine, Dentistry and Biomedical Sciences, Queen’s University of Belfast, Belfast, UK

xiv

Contributors

Thomas N. Seyfried  •  Department of Biology, Boston College, Chestnut Hill, MA, USA Justin Snider  •  Department of Biochemistry and Molecular Biology, Lipidomics Core Mass Spectrometry Lab, Medical University of South Carolina, Charleston, SC, USA Rosmarie Süß  •  Medical Department, Institute of Medical Physics and Biophysics, University of Leipzig, Leipzig, Germany Zdzislaw M. Szulc  •  Department of Biochemistry and Molecular Biology, Lipidomics Core Mass Spectrometry Lab, Medical University of South Carolina, Charleston, SC, USA Ryo Taguchi  •  Department of Metabolome, Graduate School of Medicine, The University of Tokyo and Core Research for Evolutional Science and Technology, Saitama, Japan Michael C. Thomas  •  School of Chemistry and Health Sciences, University of Wollongong, Wollongong, NSW, Australia Scott Turner  •  KineMed, Inc., Emeryville, CA, USA Andrew Webb  •  Department of Bioengineering, Pennsylvania State University, PA, USA Åsa M. Wheelock  •  Lung Research Lab L4:01, Respiratory Medicine Unit, Department of Medicine, Karolinska Institutet, Stockholm, Sweden Craig E. Wheelock  •  Department of Medical Biochemistry and Biophysics, Division of Physiological Chemistry II, Karolinska Institutet, Stockholm, Sweden Zeev Wiesman  •  Phyto-Lipid Biotechnology Lab, Department of Biotechnology Engineering, The Institutes for Applied Research, Ben Gurion University of the Negrev, Beer-Sheva, Israel Claude Wolf  •  HDR, Laboratoire de Spectrometrie de Masse, Faculte de Medicine, Pierre et Marie Curie, Universite Paris, Paris, France

Part I Shotgun and Global Lipidomics

sdfsdf

Chapter 1 Examination of the Brain Mitochondrial Lipidome Using Shotgun Lipidomics Michael A. Kiebish, Xianlin Han, and Thomas N. Seyfried Summary  Contamination from subcellular organelles and myelin has hindered attempts to characterize the lipidome of brain mitochondria. A high degree of mitochondrial purity is required for accurate measurements of the content and molecular species composition of mitochondrial lipids. We devised a discontinuous Ficoll and sucrose gradient procedure for the isolation and purification of brain mitochondria free from any detectable contamination. Shotgun lipidomics was used to analyze the lipid composition of the brain mitochondria. These procedures can be used to determine whether intrinsic lipid abnormalities underlie mitochondrial dysfunction associated with neurological and neurodegenerative diseases. Key words:  Nonsynaptic, Synaptic, Mitochondria, Brain, Lipidome

1. Introduction The brain contains two major populations of mitochondria that can be isolated by discontinuous gradients. These populations include the nonsynaptic (NS) mitochondria, which are largely derived from neuronal and glial cell bodies, and the synaptic (Syn) mitochondria, which originate from the synaptic nerve terminal of neurons (1, 2). These mitochondria populations differ in calcium homeostasis, enzymes involved in the TCA cycle, electron transport chain activities, and glutamate metabolism (2–5). Differences between NS and Syn mitochondria may underlie the complexity of neural metabolism (1). A better understanding of brain bioenergetics can be obtained with accurate information on the mitochondrial lipidome. In contrast to extensive research on mitochondria DNA and proteins in disease pathology, little attention has been focused on Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, DOI 10.1007/978-1-60761-322-0_1, © Humana Press, a part of Springer Science + Business Media, LLC 2009

3

4

Kiebish, Han, and Seyfried

the role of mitochondrial membrane lipids (6). Mitochondrial lipids can influence electron transport chain activities and create a membrane environment conducive for an efficient proton gradient (7). In addition, mitochondrial lipids regulate membrane fluidity and can affect the morphological structure of mitochondria (8, 9). Also, lipid composition can influence numerous mitochondrial enzyme activities to include creatine kinase, adenine nucleotide transporter, voltage dependent ion channel, and carnitine acyltransferase (7, 10, 11). Thus, the maintenance of the mitochondrial lipidome is critical for mitochondrial functionality. Shotgun lipidomics uses a high-throughput platform which allows for the rapid quantitation of lipid molecular species using multidimensional mass spectrometry (12). This lipidomic approach has the capabilities of detecting glycerophospholipid subclasses which contain alkyl ether, vinyl ether, or ester linkages (13–15). In addition, shotgun lipidomics has demonstrated increased detection sensitivity to quantitating cardiolipin, a mitochondrial specific phospholipid (16, 17). Cardiolipin is a complex phospholipid, containing four acyl chains, three glycerol moieties, and two phosphate groups (8). We also used shotgun lipidomics and tandem MS to measure all detectable molecular species of cardiolipin in the C57BL/6J mouse brain (1). Difficulty in removing subcellular contamination from brain mitochondrial populations has hindered the accurate analysis of mitochondrial lipid composition. This is especially the case for lipid rich myelin membranes, which can remain bound to brain mitochondria thus distorting analysis of lipid composition and content. Mitochondria can be isolated by Percoll, Nycodenz, metrizamide, Ficoll, and sucrose discontinuous gradients, however, the purity of these mitochondrial preparations differ depending on the intended use of these mitochondrial fractions (18–24). We used discontinuous Ficoll and sucrose gradients to isolate NS and Syn brain mitochondria free from detectable contamination. This procedure will be useful for analyzing differences in brain mitochondria in a variety of neurodegenerative and neurological disorders that involve alterations in energy metabolism.

2. Materials 2.1.Equipment (See Also Note 1)

  1. Ultracentrifuge.

2.2. Reagents

  1. Mitochondrial isolation buffer. 0.32 M sucrose, 10 mM Tris–HCl, and 1 mM EDTA–K (pH 7.4).

2.2.1. Mitochondrial Isolation

  2. Mass Spectrometer.

  2. Nanopure deionized water.



Examination of the Brain Mitochondrial Lipidome Using Shotgun Lipidomics

5

3. TE buffer (1 mM EDTA–K and 10 mM Tris–HCl, pH 7.4). 4. 20% Ficoll stock solution made with MIB, 12% and 7.5% Ficoll working solutions. 5. 1.6 M sucrose stock containing 1 mM EDTA–K and 10 mM Tris–HCl (pH 7.4); 0.8 M, 1.0 M, 1.3 M, and 1.6 M sucrose gradient working solutions. 6. Bovine serum albumin (BSA). 7. Mitochondrial isolation buffer containing 0.5 mg/mL BSA. 8. 6 mM Tris–HCl (pH 8.1). 2.2.2. Mass Spectrometry

1. 1,2-dimyristoleoyl-sn-glycero-3-phosphocholine (14:1–14:1 PtdCho). 2. 1,2-dipalmitoleoyl-sn-glycero-3phosphoethanolamine (16:1–16:1 PtdEtn). 3. 1,2-dipentadecanoyl-sn-glycero-3-phosphoglycerol (15:0–15:0 PtdGro). 4. 1,2-dimyristoyl-sn-glycero-3-phosphoserine (14:0–14:0 PtdSer). 5. N-lauroryl sphingomyelin (N12:0 CerPCho). 6. 1,1¢,2,2¢-tetramyristoyl cardiolipin (T14:0 Ptd2Gro). 7. Heptadecanoyl ceramide (N17:0 Cer). 8. 1-heptadecanoyl-2-hydroxy-sn-glycero-3-phosphocholine (17:0 LysoPtdCho). 9. Deuterated cholesterol (d6-chol). 10. Chloroform. 11. Methanol. 12. Lithium chloride.

3. Methods 3.1. Brain Mitochondrial Isolation

Mice of the C57BL/6 strain (4 months of age) were sacrificed by cervical dislocation and the cerebral cortex was dissected. Mitochondria were isolated in a cold room (4°C) and all reagents were kept on ice. The isolation procedure employed a combination of gradients and strategies as previously described (2–4, 25–28) (Fig. 1). The cerebral cortexes (a pool of 6 per sample) were diced on an ice-cold metal plate and then placed in 12 mL of mitochondria isolation buffer (MIB; 0.32 M sucrose, 10 mM Tris–HCl, and 1 mM EDTA–K (pH 7.4)). The pooled cerebral cortexes were homogenized using a Potter Elvehjem homogenizer with a Teflon coated pestle attached to a hand-held drill.

6

Kiebish, Han, and Seyfried

Fig. 1. Procedure used for the isolation and purification of NS and Syn mitochondria from mouse cerebral cortex.



Examination of the Brain Mitochondrial Lipidome Using Shotgun Lipidomics

7

Samples were homogenized using 15 up and down strokes at 500 rpm. The homogenate was centrifuged at 1,000 × g for 5 min. The supernatant was collected and the pellet was washed twice by centrifugation, collecting the supernatants each time. The supernatants were pooled and centrifuged at 1,000 × g for 5 min. The collected supernatant was then spun at 14,000 × g for 15 min. The supernatant was discarded and the pellet, which contained primarily NS mitochondria, synaptosomes, and myelin, was resuspended in 12 mL MIB and was layered on a 7.5%/12% discontinuous Ficoll gradient. Each Ficoll gradient layer contained 12 mL for a total volume of 36 mL. The Ficoll gradients were made from a 20% Ficoll stock with MIB. The gradient was centrifuged at 73,000 × g for 36 min (4°C) in a Sorvall SW 28 rotor with slow acceleration and deceleration (Optima L-90K Ultracentrifuge). The centrifugation time used permitted sufficient acceleration and deceleration to achieve maximum g force (28), and to prevent synaptosomal contamination of the mitochondrial fraction below the 12% Ficoll layer. Crude myelin collected at the MIB/7.5% Ficoll interface was discarded. Synaptosomes were collected at the 7.5%/12% interface and were resuspended in MIB and centrifuged at 16,000 × g for 15 min. The Ficoll gradient-purified NS mitochondria (FM) were collected as a pellet below 12% Ficoll. 3.2. Mitochondria Purification 3.2.1. Purification of NS Mitochondria

3.2.2. Purification of Syn Mitochondria

The FM pellet, containing NS mitochondria, was resuspended in MIB containing 0.5 mg/mL bovine serum albumin (BSA) and was centrifuged at 12,000 × g for 15 min. The resulting pellet was collected and resuspended in 6 mL of MIB. The resuspended FM pellet was layered on a discontinuous sucrose gradient containing 0.8 M/ 1.0 M/ 1.3 M/ 1.6 M sucrose. The volumes for the sucrose gradient were 6 mL/ 6 mL/ 10 mL/ 8 mL, respectively. The gradients were made from a 1.6 M sucrose stock containing 1 mM EDTA–K and 10 mM Tris–HCl (pH 7.4). The discontinuous sucrose gradient was centrifuged at 50,000 × g for 2 h (4°C) in a Sorvall SW 28 rotor using slow acceleration and deceleration to prevent disruption of the gradient. Purified NS mitochondria were collected at the interface of 1.3 M and 1.6 M sucrose. NS mitochondria were collected and resuspended in (1:3, v/v) TE buffer (1 mM EDTA–K and 10 mM Tris–HCl, pH 7.4) containing 0.5 mg/mL BSA and centrifuged at 18,000 × g for 15 min. The pellet was then resuspended in MIB and centrifuged at 12,000 × g for 10 min. The pellet was again resuspended in MIB and centrifuged at 8,200 × g for 10 min. Synaptosomes were burst by homogenization in 6 mM Tris–HCl (pH 8.1) using five up and down strokes. The homogenized synaptosomes were transferred to a 15 mL conical tube and then placed on a rocker for 1 h (4°C). The burst synaptosomes were

8

Kiebish, Han, and Seyfried

centrifuged at 10,000 × g for 10 min. The pellet was resuspended in 6 mL of MIB. The resuspended pellet was layered on a discontinuous sucrose gradient and centrifuged following the same procedure as described above for NS mitochondria. 3.3. Mass Spectrometry 3.3.1. Sample Preparation for Mass Spectrometric Analysis

3.3.2. Instrumentation and Mass Spectrometry

An aliquot of the mitochondrial preparation was transferred to a disposable culture borosilicate glass tube (16 × 100 mm). Internal standards were added based on protein concentration and included 16:1–16:1 PtdEtn (100 nmol/mg protein), 14:1–14:1 PtdCho (45 nmol/mg protein), T14:0 Ptd2Gro (3 nmol/mg protein), 15:0–15:0 PtdGro (7.5 nmol/mg protein), 14:0–14:0 PtdSer (20 nmol/mg protein), 17:0 LysoPtdCho (1.5 nmol/mg protein), N12:0 CerPCho (20 nmol/mg protein), N17:0 Cer (5 nmol/mg protein). This allowed the final quantified lipid content to be normalized to the protein content and eliminated potential loss from the incomplete recovery. The molecular species of internal standards were selected because they represent < 0.1% of the endogenous cellular lipid mass as demonstrated by ESI/MS lipid analysis. A modified Bligh and Dyer procedure was used to extract lipids from each mitochondrial preparation as previously described (29). Each lipid extract was reconstituted with a volume of 500 mL/mg protein (which was based on the original protein content of the samples as determined from protein measurement) in CHCl3/MeOH (1:1, v/v). The lipid extracts were flushed with nitrogen, capped, and stored at –20°C for ESI/MS analysis. Each lipid solution was diluted approximately 50-fold immediately prior to infusion and lipid analysis. A triple-quadrupole mass spectrometer (Thermo Scientific TSQ Quantum Ultra, Plus, San Jose, CA, USA), equipped with an electrospray ion source and Xcalibur system software, was utilized as previously described (30). The first and third quadrupoles serve as independent mass analyzers using a mass resolution setting of peak width 0.7 Thomson while the second quadrupole serves as a collision cell for tandem MS. The diluted lipid extract was directly infused into the ESI source at a flow rate of 4 mL/min with a syringe pump. Lipid classes were analyzed in three different modes: negative-ion ESI, negative-ion ESI plus lithium hydroxide, and positive-ion ESI plus lithium hydroxide (Fig. 2). Typically, a 2-min period of signal averaging in the profile mode was employed for each mass spectrum. For tandem MS, a collision gas pressure was set at 1.0 mTorr, but the collision energy varied with the classes of lipids as described previously (14, 30). Typically, a 2- to 5-min period of signal averaging in the profile mode was employed for each tandem MS spectrum. All the mass spectra and tandem MS spectra were automatically acquired by a customized sequence subroutine operated under



Examination of the Brain Mitochondrial Lipidome Using Shotgun Lipidomics

9

Fig. 2. Schematic of the shotgun lipidomics procedure used for the analysis of the mitochondrial lipidome (modified from (12)).

Xcalibur software. Data processing of 2D MS analyses including ion peak selection, data transferring, peak intensity comparison, and quantitation was conducted using self-programmed MicroSoft Excel macros (30). 3.4. Results 3.4.1. Mitochondrial Purity

Ficoll as well as sucrose discontinuous gradients were used to purify NS and Syn mitochondria (Fig. 1). As Ficoll gradientpurified NS mitochondria (FM) contained markers for cytoskeletal (b-actin) and membrane (SNAP25, PCNA, tuberin, PLP, and calnexin) contamination, we further purified mitochondria using a discontinuous sucrose gradient (see also Note 2). None of these markers were present in the Ficoll and sucrose discontinuous gradient-purified mitochondria, which contained only mitochondrial-enriched markers representing the inner mitochondrial membrane (Complex IV, subunit IV) and the outer mitochondrial membrane (MAO-A) (Fig. 3a). Cholera toxin b immunostaining is a sensitive procedure for detecting gangliosides with the GM1a structure in cells and tissues (31). The toxin can have slight cross reactivity with GD1a. GM1a and a low amount of GD1a was found in the TH, My, and FM fractions, indicating the presence of myelin and microsomal membranes in these subcellular fractions (Fig. 3b). Only a trace amount of GM1a was detected in the NS mitochondria and no GM1a was detected in Syn mitochondria. These findings attest to the high degree of mitochondrial purity achieved with the isolation procedure.

10

Kiebish, Han, and Seyfried

Fig. 3. Distribution of protein markers on Western blots (a) and of gangliosides on thin-layer chromatography (b) in subcellular fractions from mouse cerebral cortex. Subcellular fractions included total homogenate (TH), crude myelin (My), Ficoll gradient-purified NS mitochondria (FM), Ficoll and sucrose gradient purified nonsynaptic mitochondria (NS), and Ficoll and sucrose gradient-purified synaptic mitochondria (Syn). Western blots were performed to determine the distribution of specific protein markers for the inner mitochondria membrane (complex IV, subunit IV), outer mitochondrial membrane (monoamine oxidase A), myelin (proteolipid protein), synaptosomal membrane (SNAP25), cytoskeleton (β-actin), nuclear membrane (proliferating cell nuclear antigen), Golgi membrane (Tuberin), and microsomal membrane (calnexin). GM1a was visualized on TLC plates with cholera toxin b immunostaining as described in Methods. Std. is GM1a.

3.4.2. Mitochondrial Lipid Composition

We used shotgun lipidomics to evaluate lipid content and distribution of fatty acid molecular species in the NS and Syn mitochondria. The lipid classes were listed according to their relative abundance (Table 1) (see also Note 3). Although the content of most lipids was similar in the NS and Syn mitochondria, the content of Ptd2Gro was lower whereas the content of PtdSer and Cer were higher in the Syn mitochondria than in the NS mitochondria. The myelin-enriched lipids, sulfatides, and cerebrosides, were not detected in either NS or Syn mitochondria. No major changes in lipid molecular species were found between NS and Syn mitochondria (see also Note 4). The major molecular species (>2%) of anionic, weak anionic, and weak polar lipids are found in Tables 2–4.



Examination of the Brain Mitochondrial Lipidome Using Shotgun Lipidomics

11

Table 1  Lipid composition of C57BL/6J mouse brain mitochondria Lipid

Non-synaptic

Synaptic

Ethanolamine glycerophospholipids   Phosphatidylethanolamine   Plasmenylethanolamine

187.4 ± 12.1 164.9 ± 10.0 22.5 ± 2.2

211.7 ± 21.3 184.6 ± 20.3 27.0 ± 1.0

Choline glycerophospholipids   Phosphatidylcholine   Plasmenylcholine   Plasmanylcholine

129.9 ± 7.7 119.6 ± 5.3 1.2 ± 0.1 9.1 ± 3.2

156.3 ± 26.1 137.4 ± 17 2.4 ± 1.1 16.5 ± 8.5

Cholesterol

139.0 ± 46.7

126.7 ± 31.2

Cardiolipin

52.7 ± 4.5

39.9 ± 3.4

Phosphatidylinositol

9.4 ± 0.8

10.2 ± 0.9

Phosphatidylglycerol

7.1 ± 0.5

6.4 ± 0.7

Sphingomyelin

5.3 ± 1.2

6.5 ± 0.6

Phosphatidylserine

4.6 ± 1.5

14.1 ± 3.0

Lysophosphatidylcholine

2.7 ± 0.6

3.3 ± 0.4

Ceramide

0.7 ± 0.2

1.6 ± 0.2

Values are expressed as mean nmol/mg protein ±S.D (N = 3) Significantly different values from NS mitochondria at *: P < 0.02; **: P < 0.005 as determined from the two-tailed t-test

Table 2 Mass content of major molecular species of anionic lipids as detemined by shotgun lipidomics [M − 2H]−

Major species

Non-synaptic

Synaptic

713.0

18:2–18:1–18:1–16:1 18:1–18:1–18:1-16:2 18:2–18:1–18:0–16:2 18:2–18:2–18:0–16:1

0.71 ± 0.02

0.53 ± 0.04

714.0

18:1–18:1–18:1–16:1 18:2–18:1–18:1–16:0

1.47 ± 0.23

1.03 ± 0.12

715.0

18:1–18:1–18:1–16:0 8:0–18:1–18:1–16:1

0.68 ± 0.04

0.61 ± 0.08

724.0

20:4–18:2–18:1–16:1

0.95 ± 0.08

0.66 ± 0.08

Cardiolipin

(continued)

12

Kiebish, Han, and Seyfried

Table 2  (continued) [M − 2H]−

Major species

Non-synaptic

Synaptic

725.0

20:4–18:2–18:1–16:0 20:4–18:1–18:1–16:1

2.22 ± 0.17

1.56 ± 0.22

726.0

20:4–18:1–18:1–16:0 20:3–18:1–18:1–16:1 20:3–18:1–18:1–16:1

1.74 ± 0.13

1.31 ± 0.09

727.0

18:2–18:1–18:1–18:1

2.48 ± 0.33

1.59 ± 0.18

728.0

18:1–18:1–18:1–18:1

3.72 ± 0.72

2.21 ± 0.15

735.0

20:4–20:4–18:2–16:1

0.56 ± 0.06

0.39 ± 0.02

736.0

20:4–20:4–18:1–16:1

1.41 ± 0.16

1.10 ± 0.12

737.0

20:4–20:4–18:1–16:0 22:6–18:1–18:1–16:1 22:6–18:2–18:1–16:0

2.11 ± 0.07

1.54 ± 0.17

738.0

20:4–18:2–18:1–18:1 22:6–18:1–18:1–16:0

2.68 ± 0.37

2.02 ± 0.12

739.0

20:4–18:1–18:1–18:1

3.30 ± 0.45

2.48 ± 0.23

740.0

20:4–18:1–18:1–18:0 20:3–18:1–18:1–18:1

1.07 ± 0.14

0.75 ± 0.03

748.0

20:4–20:4–18:2–18:2 20:4–20:4–20:4–16:0 22:6–20:4–18:1–16:1 22:6–22:6–16:0–16:0 22:6–18:2–18:2–18:2

1.38 ± 0.09

1.11 ± 0.17

749.0

20:4–20:4–18:2–18:1

1.99 ± 0.19

1.46 ± 0.20

750.0

20:4–20:4–18:1–18:1

3.24 ± 0.24

2.31 ± 0.28

751.0

22:6–18:1–18:1–18:1 20:4–20:3–18:1–18:1

2.74 ± 0.10

2.36 ± 0.26

752.0

22:6–18:1–18:1–18:0

0.54 ± 0.17

0.46 ± 0.07

760.0

22:6–20:4–18:2–18:2 22:6–20:4–20:4–16:0 22:6–22:6–18:1–16:1

0.76 ± 0.05

0.58 ± 0.09

761.0

22:6–20:4–18:2–18:1

1.72 ± 0.08

1.27 ± 0.05

762.0

22:6–20:4–18:1–18:1

2.94 ± 0.23

2.36 ± 0.13

763.0

22:6–20:4–18:1–18:0 22:6–20:3–18:1–18:1

1.03 ± 0.05

0.78 ± 0.07

773.0

22:6–20:4–20:4–18:1

1.42 ± 0.12

1.10 ± 0.10

Cardiolipin

(continued)



Examination of the Brain Mitochondrial Lipidome Using Shotgun Lipidomics

13

Table 2  (continued) [M − 2H]−

Major species

Non-synaptic

Synaptic

774.0

22:6–22:6–18:1–18:1 22:6–20:4–20:3–18:1

1.81 ± 0.09

1.56 ± 0.17

786.0

22:6–22:6–20:4–18:1 22:6–20:4–20:4–20:3

1.33 ± 0.08

1.11 ± 0.16

797.0

22:6–22:6–22:6–18:1 22:6–22:6–20:4–20:3

0.50 ± 0.04

0.50 ± 0.05

Cardiolipin

Phosphatidylinositol 857.5

16:0–20:4

1.12 ± 0.09

1.17 ± 0.11

881.5

18:2–20:4 16:0–22:6

0.23 ± 0.01

0.26 ± 0.06

885.5

18:0–20:4

6.68 ± 0.41

7.04 ± 0.69

909.5

18:0–22:6

0.37 ± 0.02

0.48 ± 0.03

Phosphatidylglycerol 719.5

16:0–16:1

0.29 ± 0.02

0.31 ± 0.02

721.5

16:0–16:0

0.49 ± 0.04

0.49 ± 0.03

743.5

16:1–18:2

0.22 ± 0.04

0.19 ± 0.04

745.5

16:0–18:2

0.54 ± 0.10

0.48 ± 0.05

747.5

16:0–18:1

3.41 ± 0.13

3.07 ± 0.43

769.5

18:2–18:2

0.82 ± 0.04

0.64 ± 0.10

771.5

18:1–18:2

0.29 ± 0.03

0.26 ± 0.02

773.5

18:1–18:1

0.47 ± 0.12

0.38 ± 0.02

775.5

18:0–18:1

0.55 ± 0.10

0.60 ± 0.14

Phosphatidylserine 786.5

18:0–18:2

0.29 ± 0.07

0.47 ± 0.11

788.5

18:0–18:1

0.24 ± 0.21

1.61 ± 0.77

810.5

18:0–20:4

0.10 ± 0.08

0.35 ± 0.15

834.5

18:0–22:6

3.13 ± 1.22

9.59 ± 1.60

836.5

18:0–22:5

0.22 ± 0.27

0.31 ± 0.20

838.6

20:0–20:4

0.28 ± 0.07

0.79 ± 0.04

18:0–22:4

14

Kiebish, Han, and Seyfried

Table 3  Mass content of major molecular species of weak anionic lipids as detemined by shotgun lipidomics [M − H]−

Major species

Non-synaptic

Synaptic

Ethanolamine Glycerophospholipids 716.5

D16:0–18:1

5.11 ± 1.14

4.82 ± 0.68

742.5

D18:0–18:2 D18:1–18:1 D16:0–20:2

5.07 ± 0.62

5.09 ± 0.88

746.5

P16:0–22:6 D18:0–18:0 P18:2–20:4

9.03 ± 1.39

9.89 ± 1.72

762.5

D16:0–22:6

9.39 ± 0.81

11.08 ± 1.00

764.5

D16:0–22:5 D18:1–20:4

8.78 ± 0.52

8.35 ± 0.86

766.5

D18:0–20:4 D16:0–22:4

37.49 ± 2.42

36.39 ± 4.86

768.6

D18:1–20:2 D16:0–22:3 D18:0–20:3

11.49 ± 2.19

16.42 ± 1.11

774.5

P18:0–22:6 P18:1–22:5 D18:0–20:0

10.37 ± 1.44

13.25 ± 1.63

776.6

P18:0–22:5 P18:1–22:4

7.29 ± 2.19

8.24 ± 2.24

788.5

D18:1–22:6

5.95 ± 0.44

6.47 ± 0.55

790.5

D18:0–22:6 D18:1–22:5

58.18 ± 1.87

58.66 ± 8.62

792.6

D18:1–22:4 D18:0–22:5

3.62 ± 0.29

4.90 ± 0.38

794.6

D20:0–20:4 D18:1–22:3 D18:0–22:4

6.53 ± 0.46

15.63 ± 2.58

796.6

D20:0–20:3 D18:0–22:3

6.85 ± 1.34

9.14 ± 0.48

564.5

N18:0

0.66 ± 0.15

1.54 ± 0.19

592.6

N20:0

0.04 ± 0.02

0.03 ± 0.00

646.6

N24:1

0.02 ± 0.02

0.04 ± 0.02

648.6

N24:0

0.02 ± 0.00

0.01 ± 0.01

Ceramide



Examination of the Brain Mitochondrial Lipidome Using Shotgun Lipidomics

15

Table 4  Mass content of major molecular species of weak polar lipids as detemined by shotgun lipidomics [M + Li]+

Major species

Non-synaptic

Synaptic

Choline glycerophospholipids 740.6

D16:0–16:0

10.00 ± 1.27

13.58 ± 1.32

766.6

D16:0–18:1

33.31 ± 1.31

37.34 ± 3.07

768.6

D16:0–18:0

2.47 ± 0.26

3.36 ± 0.93

782.7

A16:0–20:0

7.42 ± 2.81

13.18 ± 7.40

788.6

D18:2–18:2 D16:0–20:4

23.37 ± 1.28

23.13 ± 2.55

792.6

D18:0–18:2 D18:1–18:1

4.15 ± 0.42

3.89 ± 0.28

794.6

D18:0–18:1

4.39 ± 0.46

6.43 ± 0.99

812.6

D16:0–22:6 D18:2–20:4

13.94 ± 0.79

14.08 ± 1.45

814.6

D18:1–20:4 D16:0–22:5

5.66 ± 0.72

5.51 ± 1.07

816.6

D18:2–20:2 D18:0–20:4

11.36 ± 0.90

12.21 ± 2.03

840.6

D18:0–22:6 D20:2–20:4

2.07 ± 1.96

4.14 ± 1.02

709.6

N16:0

0.27 ± 0.32

0.27 ± 0.13

735.6

N18:1

0.26 ± 0.02

0.29 ± 0.08

737.6

N18:0

1.99 ± 0.38

2.76 ± 0.17

765.6

N20:0

2.33 ± 0.41

2.61 ± 0.32

793.7

N22:0

0.16 ± 0.27

0.32 ± 0.13

821.7

N24:0

0.20 ± 0.14

0.08 ± 0.02

Sphingomyelin

Lysophosphatidylcholine 490.3

14:0

0.06 ± 0.01

0.04 ± 0.02

504.3

A16:0

0.06 ± 0.03

0.04 ± 0.01

516.3

16:1

0.09 ± 0.04

0.13 ± 0.07

518.3

16:0

0.89 ± 0.21

1.22 ± 0.14

542.3

18:2

0.06 ± 0.02

0.03 ± 0.02

544.3

18:1

0.41 ± 0.13

0.54 ± 0.12 (continued)

16

Kiebish, Han, and Seyfried

Table 4  (continued) [M + Li]+

Major species

Non-synaptic

Synaptic

Lysophosphatidylcholine 546.4

18:0

0.32 ± 0.08

0.44 ± 0.08

566.3

20:4

0.16 ± 0.09

0.16 ± 0.03

574.4

20:0

0.04 ± 0.02

0.07 ± 0.09

590.3

22:6

0.24 ± 0.12

0.27 ± 0.03

592.3

22:5

0.08 ± 0.03

0.09 ± 0.03

4. Notes 1. Additional equipments and supplies needed include a 4°C temperature-controlled room, Potter Elvehjem homogenizer with a Teflon coated pestle (10 mL and 25 mL), handheld drill (max speed 500 rpm), table top centrifuge, 1.5 mL Eppendorf tubes, 15 mL and 50 mL conical tubes, Sorvall SW 28 rotor, Ultra-Clear™ ultracentrifuge tubes (Part 344058 Beckman Coulter), and culture borosilicate glass tube (16 × 100 mm). The mass spectrometer and ultracentrifuge used were a triple-quadrupole mass spectrometer (Thermo Scientific TSQ Quantum Ultra, Plus, San Jose, CA, USA), equipped with an electrospray ion source and Xcalibur system software and an Optima L-90K ultracentrifuge, respectively. 2. The analysis of highly purified brain mitochondria by shotgun lipidomics presents an innovated high-throughput approach to analyzing the mitochondrial lipidome. The major obstacle that arises from analyzing brain mitochondria is obtaining a highly purified fraction free from contamination. Although numerous types of discontinuous gradients can be utilized, we found that a Ficoll gradient in addition to a sucrose discontinuous gradient could obtain a highly purified brain mitochondria preparation suitable for lipidomic analysis (1). 3. Differences can exist in content, molecular species, glycerophospholipid subclass, or total fatty acid distribution in mitochondrial lipids. These differences can be readily detected using a shotgun lipidomics platform with minimal sample processing (14). Changes in any or all aspects of the mitochondrial



Examination of the Brain Mitochondrial Lipidome Using Shotgun Lipidomics

17

lipidome will likely change membrane fluidity, efficiency of the proton gradient, regulation of specific enzyme activities, as well as the overall bioenergetic efficiency of mitochondria (7). Alterations in brain energy metabolism in neurological and neurodegenerative diseases is well established (32–34). Analysis of the brain mitochondrial lipidome in diseased tissues or mouse models can provide new insight into the role of mitochondria lipid alterations during disease pathogenesis. 4. The development of a high-throughput approach to analyze the brain mitochondrial lipidome is a new tool to study the cause(s) for altered energy metabolism in diseased brain (35). We have also used this approach to provide lipidomic evidence supporting the Warburg theory of cancer in a series of mouse brain tumors (36). It is our opinion that new insight on brain function and pathogenesis will be realized with further investigations of the brain mitochondrial lipidome. References 1. Kiebish M.A., Han X., Cheng H., Lunceford A., Clarke C.F., Moon H., Chuang J.H. and Seyfried T.N. (2008) Lipidomic analysis and electron transport chain activities in C57BL/6J mouse brain mitochondria. J Neurochem 106, 299–312. 2. Lai J.C., Walsh J.M., Dennis S.C. and Clark J.B. (1977) Synaptic and non-synaptic mitochondria from rat brain: isolation and characterization. J Neurochem 28, 625–631. 3. Brown M.R., Sullivan P.G. and Geddes J.W. (2006) Synaptic mitochondria are more susceptible to Ca2+overload than nonsynaptic mitochondria. J Biol Chem 281, 11658– 11668. 4. Dagani F., Gorini A., Polgatti M., Villa R.F. and Benzi G. (1983) Synaptic and non-synaptic mitochondria from rat cerebral cortex. Characterization and effect of pharmacological treatment on some enzyme activities related to energy transduction. Farmaco [Sci] 38, 584–594. 5. Villa R.F., Gorini A., Geroldi D., Lo Faro A. and Dell’Orbo C. (1989) Enzyme activities in perikaryal and synaptic mitochondrial fractions from rat hippocampus during development. Mech Ageing Dev 49, 211– 225. 6. Wallace D.C. (2001) A mitochondrial paradigm for degenerative diseases and ageing. Novartis Found Symp 235, 247–263; discussion 263–266. 7. Daum G. (1985) Lipids of mitochondria. Biochim Biophys Acta 822, 1–42.

8. Hoch F.L. (1992) Cardiolipins and biomembrane function. Biochim Biophys Acta 1113, 71–133. 9. Stuart J.A., Gillis T.E. and Ballantyne J.S. (1998) Remodeling of phospholipid fatty acids in mitochondrial membranes of estivating snails. Lipids 33, 787–793. 10. Rostovtseva T.K. and Bezrukov S.M. (2008) VDAC regulation: role of cytosolic proteins and mitochondrial lipids. J Bioenerg Biomembr 40, 163–170. 11. Campbell A.M. and Chan S.H. (2008) Mitochondrial membrane cholesterol, the voltage dependent anion channel (VDAC), and the Warburg effect. J Bioenerg Biomembr 40, 193–197. 12. Han X. and Gross R.W. (2005) Shotgun lipidomics: electrospray ionization mass spectrometric analysis and quantitation of cellular lipidomes directly from crude extracts of biological samples. Mass spectrometry reviews 24, 367–412. 13. Yang K., Zhao Z., Gross R.W. and Han X. (2007) Shotgun lipidomics identifies a paired rule for the presence of isomeric ether phospholipid molecular species. PLoS ONE 2, e1368. 14. Han X. and Gross R.W. (2005) Shotgun lipidomics: multidimensional MS analysis of cellular lipidomes. Expert review of proteomics 2, 253–264. 15. Han X., Holtzman D.M. and McKeel D.W., Jr. (2001) Plasmalogen deficiency in early Alzheimer’s disease subjects and in animal models: molecular characterization using electrospray

18

Kiebish, Han, and Seyfried

ionization mass spectrometry. J Neurochem 77, 1168–1180. 16. Cheng H., Mancuso D.J., Jiang X., Guan S., Yang J., Yang K., Sun G., Gross R.W. and Han X. (2008) Shotgun lipidomics reveals the temporally dependent, highly diversified cardiolipin profile in the mammalian brain: temporally coordinated postnatal diversification of cardiolipin molecular species with neuronal remodeling. Biochemistry 47, 5869–5880. 17. Han X., Yang K., Yang J., Cheng H. and Gross R.W. (2006) Shotgun lipidomics of cardiolipin molecular species in lipid extracts of biological samples. J Lipid Res 47, 864–879. 18. Zischka H., Lichtmannegger J., Jagemann N., Jennen L., Hamoller D., Huber E., Walch A., Summer K.H. and Gottlicher M. (2008) Isolation of highly pure rat liver mitochondria with the aid of zone-electrophoresis in a free flow device (ZE-FFE). Methods Mol Biol 424, 333–348. 19. Sims N.R. (1990) Rapid isolation of metabolically active mitochondria from rat brain and subregions using Percoll density gradient centrifugation. J Neurochem 55, 698–707. 20. Sims N.R. and Anderson M.F. (2008) Isolation of mitochondria from rat brain using Percoll density gradient centrifugation. Nat Protoc 3, 1228–1239. 21. Dagani F., Zanada F., Marzatico F. and Benzi G. (1985) Free mitochondria and synaptosomes from single rat forebrain. A comparison between two known subfractionation techniques. J Neurochem 45, 653–656. 22. Taylor S.W., Warnock D.E., Glenn G.M., Zhang B., Fahy E., Gaucher S.P., Capaldi R.A., Gibson B.W. and Ghosh S.S. (2002) An alternative strategy to determine the mitochondrial proteome using sucrose gradient fractionation and 1D PAGE on highly purified human heart mitochondria. J Proteome Res 1, 451–458. 23. Stocco D.M. and Hutson J.C. (1980) Characteristics of mitochondria isolated by rate zonal centrifugation from normal liver and Novikoff hepatomas. Cancer Res 40, 1486–1492. 24. Graham J.M. (2001) Purification of a crude mitochondrial fraction by density-gradient centrifugation. Curr Protoc Cell Biol Chapter 3, Unit 3 4. 25. Lai J.C. and Clark J.B. (1976) Preparation and properties of mitochondria derived from synaptosomes. Biochem J 154, 423–432.

26. Mena E.E., Hoeser C.A. and Moore B.W. (1980) An improved method of preparing rat brain synaptic membranes. Elimination of a contaminating membrane containing 2´,3´-cyclic nucleotide 3´-phosphohydrolase activity. Brain Res 188, 207–s31. 27. Rendon A. and Masmoudi A. (1985) Purification of non-synaptic and synaptic mitochondria and plasma membranes from rat brain by a rapid Percoll gradient procedure. J Neurosci Methods 14, 41–51. 28. Battino M., Bertoli E., Formiggini G., Sassi S., Gorini A., Villa R.F. and Lenaz G. (1991) Structural and functional aspects of the respiratory chain of synaptic and nonsynaptic mitochondria derived from selected brain regions. J Bioenerg Biomembr 23, 345–363. 29. Cheng H., Guan S. and Han X. (2006) Abundance of triacylglycerols in ganglia and their depletion in diabetic mice: implications for the role of altered triacylglycerols in diabetic neuropathy. J Neurochem 97, 1288–300. 30. Han X., Yang J., Cheng H., Ye H. and Gross R.W. (2004) Toward fingerprinting cellular lipidomes directly from biological samples by two-dimensional electrospray ionization mass spectrometry. Anal Biochem 330, 317–331. 31. Brigande J.V., Platt F.M. and Seyfried T.N. (1998) Inhibition of glycosphingolipid biosynthesis does not impair growth or morphogenesis of the postimplantation mouse embryo. J Neurochem 70, 871–882. 32. Petrozzi L., Ricci G., Giglioli N.J., Siciliano G. and Mancuso M. (2007) Mitochondria and neurodegeneration. Biosci Rep 27, 87–104. 33. Beal M.F. (2005) Mitochondria take center stage in aging and neurodegeneration. Ann Neurol 58, 495–505. 34. Calabrese V., Scapagnini G., Giuffrida Stella A.M., Bates T.E. and Clark J.B. (2001) Mitochondrial involvement in brain function and dysfunction: relevance to aging, neurodegenerative disorders and longevity. Neurochem Res 26, 739–64. 35. Bowling A.C. and Beal M.F. (1995) Bioenergetic and oxidative stress in neurodegenerative diseases. Life Sciences 56, 1151–1171. 36. Kiebish M.A., Han X., Cheng H., Chuang J.H. and Seyfried T.N. (2008) Cardiolipin and electron transport chain abnormalities in mouse brain tumor mitochondria: Lipidomic evidence supporting the Warburg theory of cancer. J Lipid Res 49, 2545–2556

Chapter 2 Tracking the Glycerophospholipid Distribution of Docosahexaenoic Acid by Shotgun Lipidomics Todd W. Mitchell Summary Docosahexaenoic acid (DHA, 22:6 n-3) is an omega-3 fatty acid with a 22 carbon acyl chain containing six cis double bonds and is predominantly found in membrane glycerophospholipids. Dietary consumption of DHA has been positively linked with the prevention of numerous pathologies and consequently, it has been the focus of extensive research over the last four decades. Nevertheless, our understanding of its molecular mode of action is not well understood. One likely mechanism is through DHA’s influence on cell membranes and the proteins embedded within them. This influence may be altered depending on the glycerophospholipid head group DHA is esterified to and its fatty acid partner, i.e., the specific glycerophospholipid molecule. Accordingly, an understanding of the exact glycerophospholipid distribution of DHA within a tissue is important if we wish to gain further insight into its role in the prevention of disease. In this chapter a rapid, shotgun lipidomic approach for identifying the molecular glycerophospholipid distribution of DHA is described. Key words:  Docosahexaenoic acid, ESI–MS, Phospholipid, Shotgun lipidomics, Lipid

1. Introduction Docosahexaenoic acid (DHA, 22:6 n-3) is a long-chain polyunsaturated omega-3 fatty acid with a 22 carbon acyl chain that contains six cis double bonds (Fig. 1). It is found in most animals, particularly in glycerophospholipids (GPLs) and is abundant in fish. DHA has been the focus of a large amount of research over the last few decades with interest in this essential fatty acid initiated by the famous work of Bang and Dyerberg in the 1970s (1). From this pioneering work, a link between the consumption of omega-3 fatty acids, in particular DHA and eicospentaenoic acid Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, DOI 10.1007/978-1-60761-322-0_2, © Humana Press, a part of Springer Science + Business Media, LLC 2009

19

20

Mitchell

Fig. 1. Structure of dochosahexaenoic acid.

(EPA, 20:5 n-3) and a reduced risk of cardiovascular disease was identified. Since that time research has implicated DHA in the prevention of several prevalent chronic diseases affecting modern society, e.g., cardiovascular disease (2), depression (3), type 2 diabetes (4, 5), obesity (6), and cancer (7). In fact, dietary consumption of this simple lipid has been associated with the prevention of a myriad of pathologies (8). Although DHA’s mechanism/s of action are not well known, its high preference for phospholipids has ensured that its effect on membrane physics is of particular interest. The high number of possible structural conformations of DHA (9, 10) leads to a reduction in membrane packing and stability (11). DHA is also known to increase membrane permeability (12) and propensity for fusion (13). With such an extensive influence on the physical properties of membranes it is not surprising that phospholipid DHA content has been linked to the activity of numerous membrane proteins, e.g., Na + K + ATPase (14). Nevertheless, studies on protein kinase C suggest that, at least in some cases, DHA’s precise molecular glycerophospholipid distribution may be of more importance than its influence on bulk membrane properties (15, 16). In 2000, Farkas and co workers were able to identify several phosphatidylcholine (GPCho) and phosphatidylethanolamine (GPEtn) molecules containing DHA in vertebrate brains (17). Although successful, the method employed for tracking the molecular GPL distribution of DHA was extremely laborious, requiring numerous chromatographic and derivatization steps (18). More recently, the relatively simple approach of shotgun lipidomics has been applied to the identification of DHA-containing glycerophospholipids in various tissues from mice and naked mole rats (19). Shotgun lipidomics describes the ability to determine the lipid content, from classes to subclasses and even individual molecules directly from a crude lipid extract by electrospray-ionization mass spectrometry (ESI–MS) without need for prior separation or derivatization (20–22). This technique is highly sensitive, specific, and can produce quantitative data with the addition of appropriate internal standards to the lipid extract (23). In this technique, GPLs are initially separated by their charge polarity, i.e., by selecting either the positive or negative ion mode. These molecules are easily viewed using a simple MS scan, in



Tracking the Glycerophospholipid Distribution of Docosahexaenoic Acid

21

which all GPLs carrying a common charge are selected from a lipid extract. The power in ESI–MS is unveiled in its ability to select one molecular ion, and through collision-induced dissociation (CID), observe the compositional fragments of that ion. From these fragments the phospholipid head group as well as the carbon chain length and unsaturation of each fatty acid (FA) can be determined (24). The production of these characteristic fragments forms the basis of the separation methods exploited by the shotgun lipidomics approach. A triple quadrupole or a quadrupole time of flight mass spectrometer (with enhanced duty cycle) permits precursor ion scanning allowing the analysis of specific phospholipids based on a characteristic fragment ion. An example of such is the glycerol backbone of GPLs at m/z 153.0 that identifies phosphtidylserine (GPSer), Phosphatidic acid (GPA), phosphatidylglycerol (GPGro), phosphatidylinositol (GPIns), and cardiolipin in negative ions, or a FA scan to separate out GPLs based on their FA moieties, e.g., DHA (20, 25). Alternatively, triple quadrupole mass spectrometers also provide the ability to perform neutral loss scans to identify a common neutral fragment, such as the loss of didehydroalanine (87 Da) from GPSer in negative ions or phosphoethanolamine (141 Da) from GPEtn in positive ions. Such scans focusing on the head group fragments reduce the incidence of isobaric interference from other GPLs and are used for the quantification of GPLs by comparing to an internal standard, and varying the collision energy depending on the headgroup (25). In this chapter, a step-by-step description of how to utilize these various mass spectrometric scan techniques to identify the molecular glycerophospholipid distribution of DHA within tissues will be presented.

2. Materials 2.1. Equipment

1. Glass–glass homogenizers. 2. Tube rotator. 3. Vortex mixer. 4. UV 1601 spectrophotometer (Shimadzu Scientific Instruments, Colombia, U.S.A.). 5. Waters QuattroMicro™ triple quadrupole mass spectrometer (Waters, Manchester, U.K.).

2.2. Reagents and Solvents

All solvents must be of minimum HPLC grade. 1. Ammonium acetate.

22

Mitchell

2. Ammonium molybdate. 3. Butylated hydroxyltoluene (BHT). 4. Chloroform. 5. Hydrochloric acid. 6. Methanol. 7. Phospholipid Standards (all purchased from Avanti Polar Lipids):

(a)  Dinonadecanoyl phosphatidylcholine, GPCho (19:0/ 19:0).



(b) Diheptadecanoyl phosphatidylserine, GPSer (17:0/17:0).



(c) Diheptadecanoyl phosphatidylethanolamine, GPEtn (17:0/17:0).



(d) Diheptadecanoyl phosphatidylglycerol, GPGro (17:0/ 17:0).



(e) Diheptadecanoyl phosphatidic acid, GPA (17:0/17:0).



(f) Heptadecanoyl eicosatetraenoyl phosphatidylinositol, GPIns (17:0/5Z,8Z,11Z,14Z-20:4).

8. Potassium dihydrogen phosphate. 9. Perchloric acid. 10. Stannous chloride. 2.3. Supplies

1. Pasteur pipettes. 2. Pyrex screw cap 15 mL test tubes. 3. Plastic cuvettes.

3. Methods A workflow outlining the procedures required for the identification and relative quantification of glycerophospholipids containing DHA is shown in Fig. 2. 3.1. Total Lipid Extraction

Total lipids are extracted from tissues according to traditional methods (26) with slight modifications to enhance the compatibility of the extracts with mass spectrometric analysis as described previously (27). In detail: 1. Weigh tissue and homogenize in 2 mL methanol:chlorofrorm (1:2 v/v) containing 0.01% butylated hydroxytoluene (BHT) using a glass–glass homogenizer. 2. Add internal standard mixture at 4 mL /g tissue (see Note 1).



Tracking the Glycerophospholipid Distribution of Docosahexaenoic Acid

23

Fig. 2. Workflow for tracking the glycerophospholipid distribution of docosahexaenoic acid.

24

Mitchell

3. Add further methanol:chlorofrorm (1:2 v/v) containing 0.01% BHT to ensure that the total methanol:chlorofrorm volume is 20× the tissue weight. 4. Vortex and mix in a test tube rotator for a minimum of 4 h or preferably overnight at 4°C. 5. Add 500 mL 0.15 M ammonium acetate and vortex for at least 15 s. 6. Centrifuge for 5 min at 2,000 × g. There should be two phases with chloroform at the bottom containing the lipids. Proteins float in the water/methanol phase or interface. 7. Aspirate a glass pipette with a small amount of chloroform and expel. Insert glass pipette gently to the bottom of the tube and take out the chloroform without removing any of the water/methanol phase or proteins. Expel the chloroform/ lipid mixture into a new test tube. 8. Add 2 mL methanol:chloroform (1:2 v/v) to original homogenate tube and vortex for at least 15 s. 9. Centrifuge for 5 min at 2,000 × g. 10. Aspirate glass pipette with a small amount of chloroform and expel. Insert glass pipette gently to the bottom of the tube and take out the chloroform without removing any of the water/methanol phase or proteins, combine with first chloroform extract. 11. Add 500 mL of 0.15 M ammonium acetate to the combined chloroform extract and vortex for at least 15 s. 12. Centrifuge for 5 min at 2,000 × g. 13. Insert glass pipette gently into the aqueous phase without removing any of the organic phase, and discard this aqueous phase and protein layer as waste. 14. Dry down under nitrogen at 37°C. 15. Resuspend phospholipids in 2 mL methanol:chloroform (2:1 v/v) and vortex. 16. Store in glass vial at −80°C. 3.2. Phosphorus Assay

The total phospholipid concentration of lipid extracts is determined by phosphorous assay (28). The concentration of phosphorous in the lipid extract is quantified by comparison of absorbance values (680 nm) with a standard reference curve (see Note 2). 1. Take 100 mL aliquots of each lipid extract (in duplicate), dry under nitrogen and resuspend in 0.8 mL of 72% (w/v) perchloric acid. 2. Heat at 190°C for 45 min. 3. Place on ice and add 5 mL of water and 500 mL each of ammonium molybdate (8%, w/v) and stannous chloride (0.005% dilution of 40% (w/v) SnCl2 in HCl).



Tracking the Glycerophospholipid Distribution of Docosahexaenoic Acid

25

4. Make up to 10 mL by the addition of water and allow color to develop for 10 min. 5. Measure absorbance at 680 nm. 6. Calculate phospholipid content using the following equation:

Phospholipid content =

µg phosphorous × 780 , 30.97

where 780 is the assumed average mass of phospholipids in grams and 30.97 is the molecular weight of phosphorous in grams. 3.3. Mass Spectrometry 3.3.1. Instrumentation 3.3.2. Identification of Phospholipids Containing DHA

The following discussion is based on the use of a Waters QuattroMicro™ triple quadrupole mass spectrometer (Waters, Manchester, U.K.) equipped with a z-spray electrospray ion source and controlled by Micromass Masslynx version 4.0 software. Performing this step will identify all phospholipids containing a DHA moiety, providing a targeted approach for later quantitative analysis. To identify anionic glycerophospholipids (GPA, GPGro, GPSer, GPIns, and GPEtn; see Note 3) containing DHA by precursor ion scanning in negative ion mode: 1. Dilute lipid extracts to a final phospholipid concentration of 40 mM with the addition of methanol:chloroform (2:1 v/v). 2. Set capillary voltage to 3,000 V, source temperature to 80°C, desolvation temperature to 120°C and Cone voltage to 50 V. Nitrogen drying gas is used at a flow rate of 320 L/h. 3. Infuse samples into the electrospray ion source at a flow rate of 10 mL/min using the instrument’s on-board syringe pump. 4. Set the argon collision gas at a pressure of 3 mTorr and accelerate the ions at a collision energy offset of 35 eV 5. Set quadrupole 3 (Q3) to m/z 327.3 and scan quadrupole 1 (Q1) over a mass range of m/z 740–920 For the identification of GPCho containing DHA by neutral loss scanning in positive ion mode: 1. Dilute lipid extracts to a final phospholipid concentration of 40 mM with the addition of methanol:chloroform (2:1 v/v) and add aqueous lithium acetate (to a final concentration of 200 mM). 2. Set capillary voltage to 3,000 V, source temperature 80 °C, desolvation temperature 120°C and Cone voltage to 35 V. Nitrogen drying gas is used at a flow rate of 320 L/h. 3. Infuse samples into the electrospray ion source at a flow rate of 10 mL/min using the instrument’s on-board syringe pump. 4. Set the argon collision gas at a pressure of 3 mTorr and accelerate the ions at a collision energy offset of 35 eV

Fig. 3. (a) A precursor ion scan for anionic glycerophospholipids containing DHA (m/z 327.3) and (b) a neutral loss scan for GPCho containing DHA (387.3 Da; GPCho – trimethylamine – DHA) in total lipids extracts from mouse skeletal muscle. GPA phosphatidic acid, GPEtn phosphatidylglycerol, GPSer phosphatidylserine, GPCho phosphatidylcholine, DHA docosahexaenoic acid.

26 Mitchell



Tracking the Glycerophospholipid Distribution of Docosahexaenoic Acid

27

5. Scan Q1 over a mass range of m/z 750–850 and scan Q3 at an m/z offset of Q1–387.3. This offset corresponds to the loss of both trimethylamine and DHA (see Note 4). 3.4. Results 3.4.1. Structural Characterization of DHA Containing Glycerophospholipids

The above scans provide spectra of all anionic phospholipid and GPCho molecules containing DHA as shown in Fig. 3a, b respectively. Product ion spectra can now be obtained from each of the DHAcontaining phospholipid molecules to complete their structural characterization. These spectra are obtained using the same instrument settings and sample preparations as described for both positive and negative ion analysis above with the exception that: 1. Q1 is set to the m/z of the deprotonated or lithiated molecular ion, and 2. Q3 is scanned over an appropriate range to identify all fatty acid moieties present, i.e., m/z 200–350 in negative ion mode for deprotonated ions or an m/z range that is between 400 and 200 Da less than the lithiated molecular ion in positive ion mode. If GPEtn or GPCho ethers are identified as containing DHA, their identity can be confirmed by the presence of ions indicative of the loss of the ether-linked acyl chain as an alcohol from lithiated ions as described by Hsu and Turk (29, 30). Alternatively, the identity of ether lipids can be confirmed by ozone-induced dissociation (OzID) (31) that is described in detail in Chapter 21.

3.4.2. Quantification of DHA Containing Glycerophospholipids

In order to remove isobaric interference across glycerophospholipid classes, quantification of molecular phospholipids is performed using precursor ion and neutral loss scans of head group-specific fragments (25). It is unlikely that all GPL classes will contain DHA and therefore the specific head group scans required are determined by the previous identification of molecular GPLs containing DHA. GPA, GPGro, and GPIns specific scans are performed in negative ion mode: 1. Dilute lipid extracts to a final phospholipid concentration of 40 mM with the addition of methanol:chloroform (2:1 v/v). 2. Set capillary voltage to 3,000 V, source temperature 80°C, desolvation temperature 120°C and Cone voltage to 50 V. Nitrogen drying gas is used at a flow rate of 320 L/h. 3. Infuse Samples into the electrospray ion source at a flow rate of 10 mL/min using the instrument’s on-board syringe pump. 4. Set the argon collision gas at a pressure of 3 mTorr. 5. Q1 and Q3 settings and collision energies are set as listed in Table 1. GPCho, GPSer, and GPEtn specific scans are performed in positive ion mode:

28

Mitchell

Table 1 Mass spectrometer scan parameters used for the relative quantification of GPA, GPGro, and GPIns GPL class

Scan type

Q1 scan range (m/z) Q2

Collision energy offset (eV)

GPA/GPGro Precursor ion 650–830

m/z 153.0 50

GPIns

m/z 241.0 45

Precursor ion 860–920

GPL glycerophospholipid, GPA phosphatidic acid, GPGro phosphatidylglycerol, GPIns phosphatidylinositol, m/z mass-to-charge ratio

Table 2 Mass spectrometry scan parameters used for the relative quantification of GPCho, GPEtn, and GPSer GPL class Scan type

Q1 scan range (m/z) Q2

Collision energy offset (eV)

GPCho

Precursor ion 780–840

m/z 184.1

35

GPEtn

Neutral loss

700–800

Q1–141.5 Da 25

GPSer

Neutral loss

750–840

Q1–185.4 Da 22

GPL glycerophospholipid, GPCho phosphatidylcholine, GPEtn phosphatidylethanolamine, GPSer phosphatidylserine; m/z, mass-to-charge ratio

1. Dilute lipid extracts to a final phospholipid concentration of 40 mM with the addition of methanol:chloroform (2:1 v/v). The formation of protonated GPSer and GPEtn ions can be enhanced by the addition aqueous ammonium acetate (to a final concentration of approximately 50 mM) (32). 2. Set capillary voltage to 3,000 V, source temperature 80°C, desolvation temperature 120°C, and Cone voltage to 35 V. Nitrogen drying gas is used at a flow rate of 320 L/h. 3. Infuse samples into the electrospray ion source at a flow rate of 10 mL/min using the instrument’s on-board syringe pump. 4. Set the argon collision gas at a pressure of 3 mTorr. 5. Q1 and Q3 settings and collision energies are set as listed in Table 2. Glycerophospholipids are then quantified by comparing their peak areas, obtained from averaging a minimum of 100 scans with the appropriate internal standard for each class after ­correction



Tracking the Glycerophospholipid Distribution of Docosahexaenoic Acid

29

Table 3 Molecular glycerophospholipids containing DHA identified in mouse skeletal muscle using shotgun lipidomics GPL

Proportion of Total GPL (%)

GPCho (16:0/22:6)

16.5 ± 0.4

GPCho (16:1/22:6)

0.6 ± 0.1

GPCho (18:0/22:6)

2.2 ± 0.2

GPCho (18:1/22:6)

0.9 ± 0.1

GPCho (18:2/22:6)

0.6 ± 0.1

GPEtn (16:0/22:6)

4.6 ± 0.3

GPEtn (18:0/22:6)

8.3 ± 0.5

GPEtn (18:1/22:6)

2.1 ± 0.1

GPEtn (18:2/22:6)

0.8 ± 0.1

GPSer (18:0/22:6)

3.6 ± 1.1

GPA (18:0/22:6)

0.2 ± 0.0

Total DHA-containing GPL

40.5 ± 1.3

GPL glycerophospholipid, GPCho phosphatidylcholine, GPEtn phosphatidylethanolamine, GPSer phosphatidylserine, GPA phosphatidic acid. Data are presented as mean ± SE (n = 4)

for isotope contributions as described by Deeley et al. (27) (see Note 5). To achieve this, the isotopic ion distribution of each phospholipid can be calculated from isotope models and the area of the monoisotopic peak multiplied by the calculated correction factor. This calculation must start with the smallest observed phospholipid so that any contribution of its isotope peaks to the area of phospholipids of greater m/z can be subtracted before subsequent isotope calculations are performed. Where two or more isomeric phospholipids are identified the relative abundance of each isomer can be determined from comparison of the combined abundances of the two fatty acid carboxylate ions arising from each lipid with the combined peak area of all carboxylate anions present in the product ion spectrum. ­Neither the relative position of the acyl chains on the glycerol backbone (33) (often called the sn-position) nor the position of double bonds (31, 34, 35) can be rigorously assigned from these data. A list of DHA-containing glycerophospholipids detected in mouse skeletal muscle using this technique is shown in Table 3.

30

Mitchell

4. Notes 1. An internal standard mixture in methanol:chloroform (2:1, v/v) should be prepared with the concentration of each internal standard reflecting the concentration of each phospholipid class within the tissue, e.g., for skeletal muscle use GPCho (19:0/19:0), 250 mM; GPEtn (17:0/17:0), 188 mM; GPSer (17:0/17:0), 125 mM; GPA (17:0/17:0), 25 mM; GPGro (17:0/17:0), 25 mM, and GPIns (17:0/20:4), 25 mM. 2. The standard used for the phosphorous assay is potassium dihydrogen phosphate (KH2PO4) at 20 mg/mL and the curve constructed using 1, 2, 5, and 10 mg of phosphorous. 3. GPEtn is technically a zwitterionic phospholipid, however it can be easily deprotonated (particularly at elevated pH) to form an anion. 4. Other ions characteristic of the neutral loss of DHA are also produced by the collision-induced dissociation of lithated GPCho; however, the neutral loss described here produces the most abundant of these ions under the described conditions. 5. Isotope corrections are required to account for (a) The greater contribution of isotopic ions to the total abundance of larger phospholipids. (b) The contribution of isotope peaks of one phospholipid to the area of the monoisotopic peak of a larger one. References   1. Dyerberg J, Bang HO, Hjorne N. (1975) Fatty acid composition of the plasma lipids in Greenland Eskimos. Am J Clin Nutr. 28(9):958–66.   2. McLennan P, Abeywardena M. (2005) Membrane basis for fish oil effects on the heart: Linking natural hibernators to prevention of human sudden cardiac death. J Membr Biol. 206(2):102.   3. Hibbeln JR, Salem N, Jr. (1995) Dietary polyunsaturated fatty acids and depression: when cholesterol does not satisfy. Am J Clin Nutr. 62(1):1–9.   4. Andersen G, Harnack K, Erbersdobler HF, Somoza V. (2008) Dietary eicosapentaenoic acid and docosahexaenoic acid are more effective than alpha-linolenic acid in improving insulin sensitivity in rats. Ann Nutr Metab. 52(3):250–6.   5. Carpentier YA, Portois L, Malaisse WJ. (2006) n-3 Fatty acids and the metabolic syndrome. Am J Clin Nutr. 83(6):S1499–504.   6. Li J-J, Huang CJ, Xie D. (2008) Anti-obesity effects of conjugated linoleic acid, docosahex-

aenoic acid, and eicosapentaenoic acid. Mol Nutr Food Res. 52(6):631–45.   7. Chapkin RS, Seo J, McMurray DN, Lupton JR. (2008) Mechanisms by which docosahexaenoic acid and related fatty acids reduce colon cancer risk and inflammatory disorders of the intestine. Chem Phys Lipids. 153(1):14–23.   8. Stillwell W. (2008) Docosahexaenoic acid: a most unusual fatty acid. Chem Phys Lipids. 153(1):1–2.   9. Feller SE. (2008) Acyl chain conformations in phospholipid bilayers: a comparative study of docosahexaenoic acid and saturated fatty acids. Chem Phys Lipids. 153(1):76–80. 10. Feller SE, Gawrisch K, MacKerell AD, Jr. (2002) Polyunsaturated fatty acids in lipid bilayers: intrinsic and environmental contributions to their unique physical properties. J Am Chem Soc. 124(2):318–26. 11. Stillwell W, Wassall SR. (2003) Docosahexaenoic acid: membrane properties of a unique fatty acid. Chem Phys Lipids. 126(1):1–27.



Tracking the Glycerophospholipid Distribution of Docosahexaenoic Acid

12. Huster D, Arnold K, Gawrisch K. (1998) Influence of Docosahexaenoic Acid and Cholesterol on Lateral Lipid Organization in Phospholipid Mixtures. Biochemistry. 37(49):17299–308. 13. Ehringer W, Belcher D, Wassall SR, Stillwell W. (1990) A comparison of the effects of linolenic (18:3[Omega]3) and docosahexaenoic (22:6[Omega]3) acids on phospholipid bilayers. Chem Phys Lipids. 54(2):79–88. 14. Turner N, Else PL, Hulbert AJ. (2003) Docosahexaenoic acid (DHA) content of membranes determines molecular activity of the sodium pump: implications for disease states and metabolism. Naturwissenschaften. 90(11):521–3. 15. Giorgione JR, Kraayenhof R, Epand RM. (1998) Interfacial membrane properties modulate Protein Kinase C Activation: Role of the Position of Acyl Chain Unsaturation. Biochemistry. 37(31):10956–60. 16. Slater SJ, Kelly MB, Taddeo FJ, Ho C, Rubin E, Stubbs CD. (1994) The modulation of protein kinase C activity by membrane lipid bilayer structure. J Biol Chem. 269(7):4866–71. 17. Farkas T, Kitajka K, Fodor E, et  al. (2000) Docosahexaenoic acid-containing phospholipid molecular species in brains of vertebrates. Proc Natl Acad Sci USA. 97(12):6362–6. 18. Takamura H, Kito M. (1991) A highly sensitive method for quantitative analysis of phospholipid molecular species by high-performance liquid chromatography. J Biochem. 109:436–9. 19. Mitchell TW, Buffenstein R, Hulbert AJ. (2007) Membrane phospholipid composition may contribute to exceptional longevity of the naked mole-rat (Heterocephalus glaber); a comparative study using shotgun lipidomics. Exp Gerontol. 42(11):1053–62. 20. Ekroos K, Chernushevich IV, Simons K, Shevchenko A. (2002) Quantitative profiling of phospholipids by multiple precursor ion scanning on a hybrid quadrupole time-of-flight mass spectrometer. Anal Chem. 74:941–9. 21. Han X, Gross RW. (2003) Global analyses of cellular lipidomes directly from crude extracts of biological samples by ESI mass spectrometry: a bridge to lipidomics. J Lipid Res. 44(6): 1071–9. 22. Han X, Gross RW. (2005) Shotgun lipidomics: electrospray ionization mass spectrometric analysis and quantitation of cellular lipidomes directly from crude extracts of biological samples. Mass Spectrom Rev. 24:367–412. 23. Han X, Yang K, Yang J, Fikes KN, Cheng H, Gross RW. (2006) Factors influencing the electrospray intrasource separation and selective ionization of glycerophospholipids. J Am Soc Mass Spectrom. 17(2):264–74.

31

24. Pulfer M, Murphy RC. (2003) Electrospray mass spectrometry of phospholipids. Mass Spectrom Rev. 22:332–64. 25. Brugger B, Erben G, Sandhoff R, Wieland FT, Lehmann WD. (1997) Quantitative analysis of biological membrane lipids at the low picomole level by nano-electrospray ionisation tandem mass spectrometry. Proc Natl Acad Sci U S A. 94:2339–44. 26. Folch J, Lees M, Sloane-Stanley GH. (1957) A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem. 226:497–509. 27. Deeley JM, Mitchell TW, Nealon JR, et  al. (2008) Human lens lipids differ markedly from those of commonly used experimental animals. Biochim Biophys Acta. 1781(6–7):288–98. 28. Mills GL, Lane PA, Weech PK. (1984) A guide to lipoprotein technique. In: Burdon RH, Kippenberg PHv, eds. Laboratory Techniques in Biochemistry and Molecular Biology. Elsevier Science, New York, pp. 240–1. 29. Hsu FF, Turk J. (2000) Characterization of phosphatidylethanolamine as a lithiated adduct by triple quadrupole tandem mass spectrometry with electrospray ionization. J Mass Spectrom. 35(5):595–606. 30. Hsu FF, Turk J, Thukkani AK, Messner MC, Wildsmith KR, Ford DA. (2003) Characterization of alkylacyl, alk-1-enylacyl and lyso subclasses of glycerophosphocholine by tandem quadrupole mass spectrometry with electrospray ionization. J Mass Spectrom. 38(7): 752–63. 31. Thomas MC, Mitchell TW, Harman DG, Deeley JM, Nealon JR, Blanksby SJ. (2008) Ozone-induced dissociation: Elucidation of double bond position within mass-selected lipid ions. Anal Chem. 80(1):303–11. 32. Thai TP, Rodemer C, Worsch J, Hunziker A, Gorgas K, Just WW. (1999) Synthesis of plasmalogens in eye lens epithelial cells. FEBS Lett. 456(2):263–8. 33. Ekroos K, Ejsing Christer S, Bahr U, Karas M, Simons K, Shevchenko A. (2003) Charting molecular composition of phosphatidylcholines by fatty acid scanning and ion trap MS3 fragmentation. J Lipid Res. 44(11):2181–92. 34. Thomas MC, Mitchell TW, Blanksby SJ. (2006) Ozonolysis of phospholipid double bonds during electrospray ionization: A new tool for structure determination. J Am Chem Soc. 128(1):58–9. 35. Thomas MC, Mitchell TW, Harman DG, Deeley JM, Murphy RC, Blanksby SJ. (2007) Elucidation of double bond position in unsaturated lipids by ozone electrospray ionization mass spectrometry. Anal Chem. 79(13):5013–22.

sdfsdf

Chapter 3 Global Analysis of Retina Lipids by Complementary Precursor Ion and Neutral Loss Mode Tandem Mass Spectrometry Julia V. Busik, Gavin E. Reid, and Todd A. Lydic Summary Despite an increasing recognition of the causative and diagnostic role of lipids in the onset and progression of retinal disease, information on the global lipid profile of the normal retina is quite limited. Here, a “shotgun” tandem mass spectrometry approach involving the use of multiple lipid class-specific precursor ion and neutral loss scan mode experiments has been employed to analyze lipid extracts from normal rat retina, obtained with minimal sample handling prior to analysis. Redundant information for the identification and characterization of molecular species in each lipid class was obtained by complementary analysis of their protonated or deprotonated precursor ions, or by analysis of their various ionic adducts (e.g., Na+, NH4+, Cl–, CH3OCO2–). Notably, “alternative” precursor ion or neutral loss scan mode MS/ MS experiments are introduced that were used to identify rat retina lipid molecular species that were not detected using “conventional” scan types typically employed in large-scale lipid-profiling experiments. This chapter outlines the principles and advantages of utilizing complementary/redundant identification of lipid species as a strategy to overcome inherent challenges and limitations of shotgun lipid analysis, and provides examples of the application of this strategy in the analysis of the retina lipidome. Key words: Retina, Lipidomics, Lipid analysis, Electrospray ionization, Tandem mass spectrometry

1. Introduction Retina has a unique fatty acid profile with the highest levels of long chain polyunsaturated fatty acids (LCPUFA), including docosahexaenoic acid (DHA22:6n3), and arachidonic acid (ARA20:4n6), observed in the body (1–5). Of these, DHA22:6n3 is the most abundant fatty acid in both neural and vascular elements of the retina (2), retinal pigment epithelial cells (6, 7) and retinal

Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, DOI 10.1007/978-1-60761-322-0_3, © Humana Press, a part of Springer Science + Business Media, LLC 2009

33

34

Busik, Reid, and Lydic

photoreceptor outer segment disc membranes (8, 9). Extensive studies clearly demonstrate the important role of lipids in retinal health and disease (5). However, most studies to date have focused on the role of LCPUFA’s, and DHA in particular, measuring total fatty acid levels without obtaining structural information. The reason for this is likely methodological – LCPUFA are very abundant in the retina and relatively easy to measure by well developed high-performance liquid chromatography (HPLC) or gas chromatography (GC) techniques. The limitations of the traditional techniques have precluded comprehensive complex lipid analysis from the limited amount of retinal material that can be obtained from animal models such as rats and mice. Thus, there is surprisingly little information available regarding the lipid composition of the normal retina, and only limited information describing changes in global lipid profiles between normal and diseased retinal tissue (See Note 1). Recent advances in the application of electrospray ionization (ESI) and matrix assisted laser desorption/ionization (MALDI) (10–20) techniques, coupled with the use of tandem mass spectrometry methods employing selective precursor ion and neutral loss scan mode analysis strategies, have enabled the development of “shotgun” lipidomics approaches for rapid and sensitive monitoring of the molecular compositions and abundances of individual lipid species in unfractionated lipid extracts (10–17). While shotgun approaches allow for high-throughput analysis of multiple lipid classes without prior chromatographic separation of lipid analytes, the large number of lipid molecular species present in crude extracts presents a significant challenge for the analyst. In addition to possible overlap of the molecular ions and the 13C isotope peaks of numerous lipid species, even greater extract complexity can arise due to the presence of individual lipid species in multiple ionic forms, via adduction with a variety of cationic (e.g., +H+, +Na+, +NH4+) or anionic (e.g., −H–, +Cl–) species in positive or negative ion modes, respectively, that may be present in small amounts following extraction of lipids from tissues or cells. Additional complexity may also be observed for certain lipids due to the presence of adducts formed by reaction of the solvents and buffer additives that are commonly employed for sample analysis. For example, phosphocholine-containing lipids are readily adducted with methylcarbonate (CH3OCO2–) anions (21), formed by the in vitro reaction of hydroxide or bicarbonate salts (22, 23) with methanol, effectively increasing the number of lipid species observed in negative ion mode analysis. In our studies, relatively limited tissue availability has prompted us to develop strategies for attaining a thorough accounting of the global lipid composition of retina without requirement for multiple sample fractionation or processing steps, such as chromatographic separation, or destruction of glycerophospholipids



Global Analysis of Retina Lipids

35

for enhanced sphingolipid analysis. Utilizing a triple quadrupole mass spectrometer to perform multiple precursor ion and neutral loss scan mode MS/MS experiments, we have found that complementary/redundant detection of a given lipid class based on the unique fragmentation behaviors of various ionic forms (e.g.,[M + H]+, [M − H]−, [M + Na]+, [M + NH4]+, [M + Cl]−, [M + CH3OCO2]−) of various lipid classes facilitates (a) the ability to identify molecular lipid species that may not be detected when only one scan mode is used, (b) a more thorough accounting of the various lipid species that may be present in multiple ionic forms when “absolute quantification” is desired, and (c) simplification of relative quantification by selection of an MS/MS scan mode in which lipid species are present in only one ionic form. Furthermore, the ability to identify lipids in more than one ionic form greatly increases the confidence for peak identification, even for ions observed at low (<5%) relative abundance. Here, we outline the general approach and the results obtained by performing complementary precursor ion and neutral loss scan mode tandem mass spectrometry analyses. Although presented within the context of defining the normal rat retina lipidome, the analytical principles could benefit the analysis of virtually any tissue or cell lipid extract. In this chapter, our discussion will be limited to analysis of rat retina sphingomyelin (SM), glycerophosphatidylcholine (GPCho), glycerophosphatidylethanolamine (GPEtn), glycerophosphatidylinositol (GPIns), glycerophosphatidylserine (GPSer), and diacylglycerol (DG) and triacylglycerol (TG) molecular lipid species. Complementary sets of precursor ion and neutral loss scan mode MS/MS experiments will be described for the analysis of these lipid classes. Analogous sets of complementary precursor ion and neutral loss experiments will also be suggested for additional lipid classes including ceramides, cholesteryl esters, glycerophosphatidylglycerol (GPGro), glycerophosphatidic acid (GPA), and monoacylglycerols. The results of the analyses for these additional lipid classes in rat retina lipid extracts may be obtained upon request from the authors.

2. Materials 2.1. Equipment

1. Teflon/glass tissue homogenizer (Glas-Col®, Sigma-Aldrich St. Louis, MO). 2. Vortex. 3. Centrifuge with swinging bucket rotor (Sorvall RC 5B Plus, Thermo Fisher Scientific Inc., Waltham, MA).

36

Busik, Reid, and Lydic

4. Evaporating Unit (Reacti-Vap, Pierce, Thermo Fisher Scientific Inc., Waltham, MA). 5. SpeedVac SC110 (Savant, Thermo Fisher Scientific Inc., Waltham, MA). 6. Chip-based nano-electrospray ionization (nESI) source (Advion NanoMate, Ithaca, NY). 7. Thermo Scientific model TSQ Quantum Ultra triple quadrupole mass spectrometer (San Jose, CA). 8. Anesthesia System Isoflurane Vapor 19.1 (G.A.S. Inc., Bridgeville). 2 .2. Reagents and Chemicals

1. Methanol (MeOH), HPLC grade, J.T. Baker (Phillipsburg, NJ). 2. Deionized water, HPLC grade, J.T. Baker (Phillipsburg, NJ). 3. Sodium chloride (NaCl), HPLC grade, J.T. Baker (Phillipsburg, NJ). 4. Ammonium hydroxide (NH4OH), EMD Chemicals (Gibbstown, NJ). 5. Chloroform (CHCl3), EMD Chemicals (Gibbstown, NJ). 6. Isopropanol, Fisher Scientific (Pittsburgh, PA). 7. Lipid standards from Avanti Polar Lipids (Alabaster, AL).

2.3. Animals

Male Wistar rats were maintained on Harlan-Teklad laboratory chow (#8640) and water ad libitum. At 12 weeks of age, rats were sacrificed under anesthesia (isoflurane/Vapomatic), and eyeballs were removed and used for retinal preparations. All procedures for the use and care of animals for laboratory research were approved by the All University Committee for Animal Use and Care at Michigan State University.

3. Methods 3 .1. Lipid Extraction from Rat Retinal Tissue

Lipids were extracted using a modified method of Folch (24). Whole retinas (~12 mg tissue) from 12-week old male Wistar rats were placed in an ice-chilled Teflon/glass tissue grinder and homogenized on ice in 50 mL/mg tissue of 40% MeOH in H2O. Retina homogenates were extracted with 200 mL/mg tissue of CHCl3:MeOH (2:1 v/v) by vortexing for 1.0 min. After centrifugation at 3,000 × g for 10 min, the lower organic phase was recovered and transferred to a new glass tube. The aqueous upper phase was re-extracted with 200 mL/mg tissue of CHCl3, then vortexed and centrifuged as before. The lower organic phase was collected and combined with the lower phase from the first extraction. The pooled organic phases were evaporated under



Global Analysis of Retina Lipids

37

nitrogen and further dried in a speedvac overnight. Dried lipid extracts were resuspended in 50 mL/mg tissue of isopropanol: methanol:chloroform (4:2:1v/v/v) and stored under nitrogen in glass vials in the dark at −80°C until further use. 3 .2. nESI-MS and MS/ MS Conditions

Immediately prior to analysis, aliquots of retina lipid extracts were diluted 1:20 in a solution of isopropanol: methanol: chloroform (4: 2: 1 v/v/v) containing 20 mM NH4OH. Synthetic lipid standards were diluted in the same solution to a final concentration of 2 mM. For some experiments, samples were analyzed in the presence of 0.5 mM NaCl. All samples were centrifuged, loaded into Whatman Multichem 96-well plates (Fisher Scientific, Pittsburgh, PA) and sealed with Teflon Ultra-Thin Sealing Tape (Analytical Sales and Services, Pompton Plains, NJ). Lipids were then introduced to a Thermo Scientific model TSQ Quantum Ultra triple quadrupole mass spectrometer (San Jose, CA) via a chip-based nano-electrospray ionization (nESI) source (Advion NanoMate, Ithaca, NY) operating in infusion mode, using an ESI HD_A chip, a spray voltage of 1.4 kV, a gas pressure of 0.3 psi, and an air gap of 2 mL. The ion transfer tube of the mass spectrometer was maintained at 150°C. All MS and MS/MS spectra were acquired using methods controlled by the Xcalibur software (Thermo, San Jose, CA). In MS mode, the Q1 peak width was maintained at 0.5 u. For neutral loss and precursor ion MS/MS scans, the peak widths of both Q1 and Q3 were maintained at 0.5 u. For product ion scan mode MS/MS experiments, Q1 and Q3 were operated with peak widths of 1.0 u. The Q2 collision gas pressure was set at 0.5 mTorr. MS and MS/ MS scans were typically acquired using scan rates of 500m/z/s. Collision energies were individually optimized for each product ion, neutral loss, or precursor ion scan mode MS/MS experiment using commercially available lipid standards. The MS/MS spectra shown were typically averaged over 100–300 scans.

3 .3. Identification of Lipid Molecular Species

Peak finding and correction for 13C isotope effects was performed on complementary sets of precursor ion and/or neutral loss scan mode MS/MS data using the Lipid Mass Spectrum Analysis (LIMSA) v.1.0 software (25) peak model fit algorithm in conjunction with an expanded user-defined database of hypothetical lipid compounds. Assignment of glycerophospholipid acyl chain constituents was achieved by precursor ion scans to monitor for the formation of specific m/z product ions corresponding to deprotonated fatty acid anions in negative ion mode; these scans identified glycerophospholipids as [M − H]−, [M + Cl]−, and [M + CH3OCO2]− ions. Assingment of diacylglycerol and triacylglycerol acyl chain constituents was achieved by neutral loss scans to monitor for the loss of m/z values corresponding to the loss of a fatty acid + NH3 from DG and TG [M + NH4]+ ions in positive ion mode.

38

Busik, Reid, and Lydic

3 .4. Application of Complementary Precursor Ion and Neutral Loss Mode Tandem Mass Spectrometry to Analysis of Rat Retina Lipids 3 .4.1. Mass Spectrum of a Crude Rat Retina Lipid Extract

The use of an animal model, such as the rat, for studying changes in global retina lipid profiles as a function of the onset and progression of disease is attractive for several reasons. These include the ability to control or systematically evaluate the effect of a number of variables, such as genetic background, environment, age, and diet that are known to alter the presence and abundance of particular lipid species at any given time. In order to obtain an initial “snapshot” of the global lipid profile of the normal rat retina, a crude lipid extract isolated from a whole retina was subjected to analysis by using nanoelectrospray ionization coupled to a triple quadrupole mass spectrometer in both positive and negative ionization modes. The mass spectra obtained from these analyses are shown in Fig. 1a, b respectively. Analysis of lipid extracts from three separate rat retinas resulted in essentially identical spectra to those shown in Fig. 1a, b, indicating that minimal biological variability was observed between these animals (data not shown). This was expected given that these samples were obtained from animals with identical genetic backgrounds that were maintained under essentially identical conditions. The resolution and mass accuracy associated with the mass spectra shown in Fig. 1 precluded the identification of individual lipid species on the basis of their masses alone. Indeed, in many cases, the product ion scan mode MS/MS experiments and the precursor ion and neutral loss scan mode MS/MS experiments discussed below indicated that multiple lipid species having the same nominal mass values, as well as multiple lipids with isobaric masses, were present at a given m/z value (see Notes 2 and 3). Furthermore, the presence and the identities of low abundance

Fig. 1. Mass spectra obtained from unfractionated rat retina lipid extract by scanning from m/z 400 to m/z 1,200 in (a) positive ion or (b) negative ion mode.



Global Analysis of Retina Lipids

39

lipid species, such as mono-, di-, and tri-acylglycerols, ceramides, and cholesteryl esters that were present at or below the level of chemical noise in these spectra could not be determined. 3 .4.2. Complementary Sets of Precursor Ion and Neutral Loss Scans for Global Lipid Analysis

In order to obtain a more comprehensive profile of the lipid composition of the crude rat retina lipid extract, a series of 27 precursor ions and neutral loss scan mode MS/MS experiments were performed in both positive and negative ionization modes (11, 13, 14, 21, 26–30). A complete description of these scans, including the specific lipid class to be identified, the precursor ion type selected for each scan (e.g., [M + H]+, [M + Na]+, [M + NH4]+, [M − H]−, [M + Cl]− and [M + CH3OCO2]−), the mass of the specific product ion or the neutral loss that was detected, and their identities, is given in Table 1 (see Notes 4, 5, and 6). Specific examples of these experiments, for determining the presence of SM, GPCho, GPEtn, GPIns, GPSer, DG, and TG lipids, are discussed below. In each case, multiple precursor ion and neutral loss scan mode experiments were used to identify and characterize the individual components of a particular lipid class. The use of multiple scans enabled the compositions of the overlapping ions present at a particular m/z value in either the positive or negative ionization mode mass spectra to be more fully elucidated, and allowed the list of spectral features (i.e., molecular ions and their various adducts) that were observed in both the positive and negative ionization mode mass spectra to be significantly expanded. Particularly noteworthy was an observation that the use of complementary precursor ion and neutral loss scans enabled the identification and characterization of molecular species that were not detected using the ‘typical’ scan types that have been most commonly employed previously in the literature for lipid-profiling experiments.

3 .4.3. Complementary Precursor Ion and Neutral Loss Scan Mode Analysis of Rat Retina SM and GPCho Lipids

The identification of abundant SM and GPCho species in negative ion mode is useful when it is necessary to (a) identify spectral features of a negative ion mass spectrum; (b) perform fatty acid analysis of lipid species by negative ion product ion tandem mass spectrometry or precursor ion scanning of m/z corresponding to deprotonated fatty acyls; and (c) identify m/z at which one or more lipid species may be present in a negative ion mass spectrum or tandem mass spectrum. Analysis of these two lipid classes is often complicated by overlap of a GPCho first 13C isotope peak with a putative SM molecular ion. This can be overcome by destruction of GPCho (and all glyerophospholipids) by alkaline hydrolysis (29). This approach is not useful, however, when analysis of both SM and GPCho species are needed from limited amounts of tissue. While algorithms to correct for 13 C isotope abundances have been used to distinguish putative

3.4.3.1. Negative Ion Analysis of SM and GPCho

Neutral phosphocholine

NLS 183

+

[M + Na] [M + H]+, M + Na]+ [M − H] [M − H]− [M + NH4]+ [M + H]+, [M + Na]+ [M + Na]+ [M − H]−

  7 SM, GPCho

  8 GPEtn

  9 GPEtn

10 GPIns

11 GPIns

12 GPSer

13 GPSer

14 GPSer

−

Sodium cholinephosphate

NLS 205

[M + Na]+

  6 GPCho

NLS 87

PIS 208

NLS 185

NLS 277

Serine

Sodium phosphoserine

Phosphoserine

Phosphoinositol+NH3

Dehydrated phosphoinositol

Glycerol phosphoethanolamine derivative

PIS 196 PIS 241

Phosphoethanolamine

NLS 141

Sodium cyclophosphane

PIS 147

[M + Na]

  5 SM, GPCho

Phosphocholine

[M + H]+

  4 SM, GPCho

PIS 184

CH3OCO2+(CH3)3 NCHCH2)

NLS 161

[M + CH3OCO2]

  3 SM, GPCho

+

CH3OCO2H+(CH3)3N

NLS 135

−

CH3Cl

[M + CH3OCO2]−

NLS 50

MS/MS scan type Product ion / Neutral loss

  2 SM, GPCho

Precursor ion [M + Cl]−

Lipid class

  1 SM,GPCho

#

(13)

–

(11)

(38)

(13)

(13)

(14)

(37)

(13)

(37)

(13)

(21)

(21)

(21)

Ref

Table 1 Description of the precursor ion and neutral loss scan mode MS/MS experiments employed for the class-specific identification of lipid molecular species

40 Busik, Reid, and Lydic

Cyclic glycerophosphate derivative Sphinganine derivative Sphingosine derivative Sphingadienine derivative 16:0 aldehyde 16:1 aldehyde

NLS 115 PIS 153 PIS 266 PIS 264 PIS 262 NLS 258 NLS 256

+

[M + NH4] [M − H]− +

[M + H]+ +

[M + H] [M − H]− −

[M + H]

[M − H] [M − H]−

20 GPA

21 GPGro, GPA

22 18:0-based ceramides

23 18:1-based ceramides

24 18:2-based ceramides

25 18:0-based ceramides

26 18:1-based ceramides

27 18:2-based ceramides

PIS parent ion scan, NLS neutral loss scan

NH3+H3PO4

NLS 189

[M + NH4]+

19 GPGro

NLS 254

Phosphoglycerol+NH3

PIS (m/z varies)

16:2 aldehyde

Fatty acyl anion

H2O+NH3

[M − H] , [M + Cl] , [M + CH3OCO2] −

18 Glycerophospholipids −

NLS 35

−

[M + NH4]+

17 Mono-/ Diacylglycerols

Cholestane cation

PIS 369

[M + NH4]

Fatty acid+NH3

NLS (m/z varies)

16 Cholesteryl esters +

[M + NH4]+

15 Di-/Triacylglycerols, Cholesteryl esters

(30)

(30)

(30)

(27)

(29)

(29)

(13)

–

(38)

(13)

–

(13)

(28)

Global Analysis of Retina Lipids 41

42

Busik, Reid, and Lydic

SM molecular ions from GPCho first 13C isotope peaks (25, 31), the extremely low abundance of retinal SM molecular ions detected by typical positive ion analysis (see Subheading 3.3.2) has precluded the successful reliance upon deisotoping algorithms alone for thorough analysis of rat retina SM species. The ability of SM and GPCho species to form adducts with chloride (Cl−) and methyl carbonate (CH3OCO2−) anions enabled a convenient method for negative ion mode detection of SM and GPCho molecular species. Chloride ions are often present in small amounts in lipid extracts. While Cl− ions may be minimized by multiple sample washing and desalting steps, we have found the presence of Cl− to be beneficial for negative ion analysis of SM and GPCho. Figure 2a demonstrated negative ion mode collisionally induced dissociation (CID) of the [M + Cl]− ion of a synthetic GPCho(14:0/14:0) species in the presence of 0.5 mM NaCl. Collisionally induced dissociation of chloride adducts of GPCho species led to an abundant neutral loss of 50 Da, corresponding to loss of methyl chloride. Dissociation of this species also resulted in ions corresponding to the loss of the GPCho 14:0 fatty acid moieties as neutral ketene species, observed at m/z 452, as well as abundant ions at m/z 227 corresponding to the deprotonated fatty acids. By contrast, CID of chloride-adducted synthetic SM(d18:1/12:0) revealed the abundant neutral loss of 50 Da was the primary product ion observed

Fig. 2. Negative ion mode product ion MS/MS spectra of [M + Cl]− ions of synthetic GPCho(14:0/14:0) (a), SM(d18:1/12:0) (b), and [M + CH3OCO2]− ions of synthetic GPCho(14:0/14:0) (c), and SM(d18:1/12:0) (d).



Global Analysis of Retina Lipids

43

over a wide range of collision energies (Fig. 2b). Dissociation of GPCho(14:0/14:0) [M + CH3OCO2]− ions resulted in the spectra shown in Fig. 2c. Abundant neutral losses of 76 Da (loss of CH3OCO2 + H), 135 Da (loss of CH3OCO2+ (CH3)3N), and 161 Da (loss of CH3OCO2+ (CH3)3 NCHCH2) were observed, as well as ions representing the loss of the neutral 14:0 fatty acid (m/z 391) and the fatty acid anion (m/z 227). Fragmentation of the methyl carbonate adduct of synthetic SM(d18:1/12:0) by CID Fig. (2d) revealed phosphocholine-derived ions similar to those observed for GPCho, however at significantly different abundances. Abundant neutral losses of 76, 135, and 161 Da were all observed, yet here the loss of 161 Da dominated, as opposed to the loss of 135 Da that was seen for GPCho. The negative ion mode gas-phase fragmentation behavior of SM and GPCho species, as demonstrated in Fig. 2, enable the use of neutral loss scan mode MS/MS for class-specific detection of SM and GPCho. Performing the conventionally used neutral loss scan (NLS) of 50 Da (scan number 1 in Table 1) in rat retina lipid extracts in the presence of 0.5 mM chloride resulted in detection of both SM and GPCho species (Fig. 3a). From the NLS

Fig. 3. Complementary negative ion mode MS/MS analysis of rat retina SM and GPCho species. Bold text indicates m/z of identified SM species. (a) Detection of SM and GPCho as [M + Cl]− ions by neutral loss scanning of 50 Da in the presence of 0.5 mM NaCl. (b) Detection of SM and GPCho as [M + CH3OCOO]− ions by neutral loss scanning of 135 Da. (c) Detection of SM and GPCho as [M + CH3OCOO]− ions by neutral loss scanning of 161 Da.

44

Busik, Reid, and Lydic

50 scan a total of eight SM peaks, highlighted with bold text in Fig. 3a, were readily discernable from GPCho species. As the neutral loss of 135 Da was the most abundant loss observed for GPCho [M + CH3OCO2]− ions across a wide range of collision energies, it enabled negative ion detection of rat retina GPCho species as methyl carbonate adducts by performing NLS 135 (Fig. 3b, scan number 2 in Table 1) when ammonium hydroxide was added to the rat retina lipid extract at a concentration of 20 mM. Five low abundance SM peaks, highlighted with bold text in Fig. 3b, were also detected with this scan. We next employed the neutral loss of 161 Da (scan number 3 in Table 1) to detect SM and GPCho [M + CH3OCO2]− ions in a rat retina lipid extract, which resulted in the spectrum shown in Fig. 3c. Here, seven additional SM peaks could be detected that were not observed in the NLS 135 scan, as well as four SM peaks not detected by NLS 50. Additionally, the relative abundances of SM species at m/z 777 and 805 (corresponding to SM(d18:1/16:0) and SM(d18:1/18:0) [M + CH3OCO2]− ions) were increased approximately tenfold and 20-fold, respectively, compared to their relative abundances in the NLS 135 scan. The differential fragmentation behavior of methylcarbonate-adducted anions of SM and GPCho was thus exploited to preferentially detect either a predominance of GPCho ions via the neutral loss of 135 Da, or a predominance of SM ions via the neutral loss of 161 Da. Both chloride and methylcarbonate adducts of GPCho species were suitable for performing negative ion mode product ion scans to confirm GPCho fatty acid assignments. Likewise, GPCho species were detected as both chloride and methylcarbonate-adducted anions when precursor ion scanning of m/z corresponding to deprotonated fatty acyls was performed in rat retina lipid extracts (scan number 18 in Table 1, data not shown.) Therefore, it was advantageous to assess rat retinal GPCho content by utilizing both NLS 135 and NLS 50 for detection of GPCho species as [M + CH3OCO2]− and [M + Cl]− ions, respectively. Performing the additional NLS 161 for detection of SM [M + CH3OCO2]− ions allowed for the most thorough identification of rat retinal SM species. Under the conditions employed here, negligible formate or acetate adducts of PC and SM lipids were observed (11, 32). 3 .4.3.2. Positive Ion Analysis of SM and GPCho

Collisionally induced dissociation of GPCho and SM [M + H]+ ions yields an abundant characteristic product ion at m/z 184 corresponding to the protonated phosphocholine headgroup. Product ion spectra of synthetic GPCho(14:0/14:0) and SM(d18:1/12:0) are shown in Fig. 4a, b, respectively. As observed for chloride and methylcarbonate adducts of GPCho and SM, product ion spectra of synthetic GPCho(14:0/14:0) and SM(d18:1/12:0) [M + Na]+ ions yielded similar product ions, though at differing relative abundances (Fig. 4b, c). ­Abundant



Global Analysis of Retina Lipids

45

Fig. 4. Positive ion mode product ion MS/MS spectra of [M + H]+ ions of synthetic GPCho(14:0/14:0) (a) and SM(d18:1/12:0) (b) and [M + Na]+ ions of synthetic GPCho(14:0/14:0) (c) and SM(d18:1/12:0) (d).

ions corresponding to loss of 59 Da (N(CH3)3), 183 Da (neutral phosphocholine), and an ion corresponding to sodiated cyclophosphane at m/z 147 were observed in the product ion spectra of both GPCho and SM. In the case of GPCho, the ion at m/z 147 was the most abundant ion observed in the product ion spectrum. In the case of SM, however, the neutral losses of 183 and 59 Da were more abundant than the ion at m/z 147. Additionally, the loss of 205 Da (corresponding to the neutral loss of sodium choline phosphate) was abundant from dissociation of GPCho [M + Na]+ ions, yet was present at less than 1% relative abundance upon dissociation of SM [M + Na]+ ions. The spectra obtained from various positive ionization mode precursor ion and neutral loss scan mode MS/MS experiments for the identification of SM and GPCho lipid ions from crude whole rat retina lipid extract are shown in Fig. 5. The spectrum obtained from the conventional precursor ion scan (PIS) mode MS/MS experiment (scan number 4 in Table 1), to monitor for the formation of the characteristic phosphocholine product ion at m/z 184 from dissociation of SM and GPCho [M + H]+ ions, is shown in Fig. 5a. After isotopic correction, only four SM peaks were identified (m/z 703, 731, 759, and 815, corresponding to d18:1/16:0, d18:1/18:0, d18:1/20:0, and d18:1/24:0 species of SM, respectively) at low (<5%) relative abundance. However, PIS 184 provided excellent sensitivity for GPCho detection, as it provided the highest signal/noise observed for all methods of GPCho detection.

46

Busik, Reid, and Lydic

Fig. 5. Complementary positive ion mode MS/MS analysis of rat retina SM and GPCho species. Bold text indicates m/z of identified SM species. (a) Detection of SM and GPCho as [M + H]+ ions by precursor ion scanning of m/z 184. (b) Detection of SM and GPCho as [M + Na]+ ions by precursor ion scanning of m/z 147. (c) Detection of GPCho as [M + Na]+ ions by neutral loss scanning of 205 Da. (d) Detection of SM and GPCho as [M + Na]+ ions by neutral loss scanning of 183 Da.

As sodium (Na+) ions are often present in lipid extracts due to carryover from the aqueous phase during lipid extraction, sample washing steps, and the mass spectrometer matrix, additional precursor ion and neutral loss scan mode MS/MS analyses of SM and GPCho [M + Na]+ ions were also performed (see Fig. 5b–d; scan numbers 5–7 in Table 1.) Notably, each scan type varied greatly in terms of utility for SM analysis, consistent with the analysis of product ion spectra from GPCho and SM [M + Na]+ ions. We found that PIS 147 (Fig. 5b, scan number 5 in Table 1), resulted in relative abundances of SM and GPCho species that were overall very similar to those obtained by the PIS 184 scan; however, only two SM species could be identified with this MS/ MS scan. Figure 5c demonstrates the virtually exclusive detection of GPCho species as [M + Na]+ ions with no detection of SM [M + Na]+ ions. This was achieved by utilizing the constant neutral loss of 205 Da (scan number 6 in Table 1). Exclusive detection



Global Analysis of Retina Lipids

47

of GPCho species by NLS 205 was consistent with the expected result based on the product ion spectra of GPCho and SM [M + Na]+ ions. Conversely, the product ion spectra of GPCho and SM [M + Na]+ ions suggested that the neutral loss of 183 Da could allow enhanced detection of SM ions relative to GPCho ions. The use of NLS 183 in rat retina lipid extracts resulted in the spectrum shown in Fig. 5d. Here, four SM species could be detected, and the relative abundances of SM and GPCho were similar to those observed for PIS 184. The same results were obtained when NLS 183 was performed across a wide range of collision energies, which implies that this MS/MS scan mode does not afford a significant improvement over PIS 184 in terms of detection of SM species. Additionally, the ion at m/z 856 (corresponding to GPCho(18:0/22:6) was observed at greater relative abundance in NLS 183 relative to the other MS/ MS scans utilized, which may implicate an acyl chain contribution to the fragmentation behavior of these ions. All of the scans used to detect SM and GPCho as [M + Na]+ ions were equally useful in providing redundant identification of GPCho molecular species, yet none of these scans matched the PIS 184 MS/MS scan in terms of overall sensitivity. As observed for negative ion analysis, complementary precursor ion and neutral loss scan mode MS/MS analysis of SM and GPCho in the positive ion mode enabled redundant detection of SM and GPCho species, and the use of NLS 205 enabled specific detection of only GPCho without lipid extract fractionation prior to analysis. The ability to redundantly detect multiple molecular species, especially in the case of GPCho, by using multiple lipid class-specific product ions increased confidence in peak identification for low abundance ions. It should be noted, however, that negative ion analysis of SM species as either [M + CH3OCO2]− or [M + Cl]− ions proved more useful than positive ion analysis in terms of readily identifying the greatest number of distinct SM molecular species. In contrast, positive ion analysis was more effective for identification of GPCho species in terms of overall sensitivity (signal/noise); however, negative ion analysis of GPCho was useful for determining fatty acid substituents by product ion scanning and/or precursor ion scanning of deprotonated fatty acyl species. If “absolute quantitation” of retinal SM and GPCho were to be attempted here, it would be necessary to analyze all ionic forms of SM and GPCho relative to internal standards and sum the contributions of each ionic form of SM or GPCho molecular species. Relative comparisons of lipid species across numerous samples could be achieved by selecting optimal scan modes for the desired lipid classes (i.e., NLS 161 for analysis of SM and PIS 184 for analysis of GPCho). The MS/MS spectra collected for analysis of SM and GPCho species described here were analyzed between m/z 700–1,200 for

48

Busik, Reid, and Lydic

positive ion mode, and m/z 700–1,300 for negative ion mode. To allow these spectra to be viewed in greater detail, smaller mass ranges are shown in Figs. 3 and 5. However, it should be noted that complementary analysis of rat retina SM and GPCho species verified the presence of GPCho species containing 32:6 and 34:6 very long chain polyunsaturated fatty acids; in each case, a 22:6 fatty acid was identified at the remaining glycerol position of the GPCho species. These species were observed as abundant ions in the positive ion MS spectrum Fig. (1a) at m/z 1,018 and 1,046, respectively. These species were identified by all positive ion MS/MS scan modes employed here for the analysis of SM and GPCho. These high molecular weight GPCho species were also identified by NLS 50 as [M + Cl]− ions, which subsequently were used for product ion mode MS/MS analysis to verify the identities of the fatty acid moieties. However, these GPCho species were not detected as methylcarbonate adducts by either NLS 135 or NLS 161. By combining both positive and negative ion mode shotgun MS/MS analyses of all phosphocholine-containing lipids, identification of 13 SM and 29 GPCho lipids was achieved in rat retina crude lipid extracts with a significant degree of redundancy (see Note 7). A summary of the SM and GPCho molecular species detected by all MS/MS scan modes described above is provided in Table 2. 3 .4.4. Complementary Precursor Ion and Neutral Loss Scan Mode Analysis of Rat Retina GPEtn and GPIns Lipids

Pairs of complementary neutral loss and precursor ion scans were also employed for the identification and characterization of GPEtn and GPIns lipid species. The detection of GPEtn lipids has typically been achieved by monitoring for the neutral loss of 141 Da (i.e., phosphoethanolamine) from their [M + H]+ precursor ions in positive ion mode MS/MS analysis (scan number 8 in Table 1). The spectrum obtained from analysis of the crude rat retina lipid extracts using this scan is shown in Fig. 6a. After correction for 13C isotope contributions, a total of 20 GPEtn lipid species were identified in this experiment. The use of a complementary precursor ion scan to monitor for the presence of a characteristic m/z 196 product ion, corresponding to a glycerol phosphoethanolamine derivative Fig. (6b) from the [M − H]− precursor ions of GPEtn (scan number 9 in Table 1), enabled the detection of three additional GPEtn lipids. The observed discrepancy in identified GPEtn lipids between the two scan types was due, in part, to the fragmentation behavior of GPEtn alkenyl species. While collisionally induced dissociation of GPEtn alkenyl [M − H]− ions gives rise to the GPEtn-specific fragment at m/z 196, the neutral loss of 141 Da is virtually absent from CID of GPEtn alkenyl [M + H]+ ions (38). GPEtn alkenyl species detected by the negative ion PIS 196 included those at m/z 748, 750, and 774 in Fig. 6b (see Table 3). Although scanning for the constant neutral loss of 141 Da affords the most

Scan #1 NLS 50 (m/z)

–

737

738

–

763

765

766

768

–

–

–

–

–

793

794

796

–

Lipid speciesa

SM(d18:1/16:1)

SM(d18:1/16:0)

GPCho(14:1/16:0)

GPCho(14:0/16:0)

SM(d18:1/18:1)

SM(d18:1/18:0)

GPCho(16:0/16:1)

GPCho(16:0/16:0)

SM(d18:1/19:0)

GPCho(O-18:0/16:0)

SM(d18:1/20:1)

GPCho(16:1/18:1)

GPCho(16:0/18:2)

SM(d18:1/20:0)

GPCho(16:0/18:1)

GPCho(16:0/18:0)

GPCho(O-18:0/18:0)

–

836

834

833

832

832

–

–

–

808

806

805

–

–

–

777

–

Scan #2 NLS 135 (m/z)

Table 2 Rat retina SM and GPCho molecular species

–

836

834

833

832

832

831

820

819

808

806

805

803

780

778

777

775

Scan #3 NLS 161 (m/z)

774

762

760

759

758

758

–

746

–

734

732

731

–

706

704

703

–

Scan #4 PIS 184 (m/z)

–

784

782

–

780

780

–

–

–

756

754

753

–

728

–

725

–

Scan #5 PIS 147 (m/z)

–

784

782

–

–

780

–

–

–

756

–

–

–

–

–

–

–

Scan #6 NLS 205 (m/z)

– (continued)

784

782

781

780

780

–

–

–

756

754

753

751

–

726

725

–

Scan #7 NLS 183 (m/z)

Global Analysis of Retina Lipids 49

Scan #1 NLS 50 (m/z)

816

–

–

820

820

822

824

840

840

841

–

844

845

847

849

–

–

Lipid speciesa

GPCho(16:0/20:4)

SM(d18:1/22:2)

SM(d18:1/22:1)

GPCho(18:1/18:1)

GPCho(18:0/18:2)

GPCho(18:0/18:1)

GPCho(18:0/18:0)

GPCho(16:0/22:6)

GPCho(16:1/22:5)

SM(d18:1/24:4)

GPCho(38:5)

GPCho(18:0/20:4)

SM(d18:1/24:2)

SM(d18:1/24:1)

SM(d18:1/24:0)

GPCho(18:1/20:0)

GPCho(18:1/22:6)

Table 2 (continued)

906

–

–

–

–

884

–

881

880

880

864

862

860

860

–

857

856

Scan #2 NLS 135 (m/z)

–

890

889

887

–

884

–

881

880

880

864

862

860

860

859

857

856

Scan #3 NLS 161 (m/z)

832

816

815

–

–

810

808

–

806

806

790

788

786

786

–

–

782

Scan #4 PIS 184 (m/z)

854

–

–

–

–

832

–

–

828

828

812

810

808

808

–

–

804

Scan #5 PIS 147 (m/z)

854

–

–

–

–

832

830

–

828

828

812

810

808

808

–

–

804

Scan #6 NLS 205 (m/z)

854

–

–

–

–

832

830

–

828

828

812

810

808

808

–

–

804

Scan #7 NLS 183 (m/z)

50 Busik, Reid, and Lydic

868

868

–

–

–

912

1,052

1,080

GPCho(18:0/22:6)

GPCho(18:1/22:5)

GPCho(40:4)

GPCho(42:10)

GPCho(42:4)

GPCho(22:6/22:6)

GPCho(22:6/32:6)

GPCho(22:6/34:6)

–

–

952

–

–

–

908

908

–

–

952

–

–

–

908

908

1,046

1,018

878

866

854

838

834

834

1,068

1,040

900

–

876

860

856

856

1,068

1,040

900

–

876

860

856

856

1,068

1,040

900

–

876

860

856

856

Global Analysis of Retina Lipids 51

52

Busik, Reid, and Lydic

Fig. 6. Complementary positive and negative ion mode MS/MS analysis of rat retina GPEtn and GPIns species. (a) Detection of GPEtn as [M + H]+ and [M + Na]+ ions by positive ion mode neutral loss scanning of 141 Da. (b) Detection of GPEtn as [M − H]− ions by precursor ion scanning of m/z 196. (c) Detection of GPIns as [M − H]− ions by negative ion mode precursor ion scanning of m/z 241. (d) Detection of GPIns as [M + NH4]+ ions by positive ion mode neutral loss scanning of 277 Da.

sensitivity for detection of GPEtn species in terms of signal/noise ratio and absolute ion abundance (data not shown), the utility of the positive ion NLS 141 is effectively limited to detection of diacyl GPEtn species. Furthermore, GPEtn [M + Na]+ ions, if present, are also detected by NLS 141, further complicating spectrum interpretation and quantification against an internal standard. m/z 814 and 858 in Fig. 6a represent abundant ions corresponding to sodium adducts of GPEtn(18:0/22:6) and GPEtn(22:6/22:6), respectively. It is possible to decrease the abundance of [M + Na]+ ions in this scan mode by using low collision energy CID (<30 V under the conditions employed); however, the use of lower collision energies in the analysis of rat retina GPEtn lipids resulted in underrepresentation of GPEtn species containing 22:6 fatty acid constituents due to the lower fragmentation efficiency of these species. While the negative ion precursor of m/z 196 scan affords detection of both diacyl and alkyl/acyl GPEtn species, this scan mode may also detect GPEtn [M + Cl]− ions formed when chloride is present in a lipid extract (m/z 826 and 870 in Fig. 6b). It should be noted that the signal-to-noise associated with the spectrum obtained by negative ion mode PIS 196 was approximately two orders of magnitude lower than that of the positive ion mode neutral loss scan, thereby placing a limitation on the utility of this scan mode



Global Analysis of Retina Lipids

53

Table 3 Rat retina GPEtn species detected by complementary positive and negative ion mode MS/MS analysis Lipid speciesa

Scan #8 NLS 141 (m/z)

Scan #9 PIS 196 (m/z)

GPEtn(16:0/18:1)

718

716

GPEtn(16:0/18:0)

720

718

GPEtn(O-16:1/20:4)

–

722

GPEtn(16:0/20:4)

740

738

GPEtn(18:1/18:1)

744

–

GPEtn(18:0/18:1)

746

744

GPEtn(O-16:1/22:6)

–

746

GPEtn(O-16:0/22:6)

750

748

GPEtn(O-18:0/20:4)

752

750

GPEtn(16:0/22:6)

764

762

GPEtn(18:1/20:4)

766

764

GPEtn(18:0/20:4)

768

766

GPEtn(O-18:2/22:6)

774

772

GPEtn(O-18:1/22:6)

776

774

GPEtn(O-18:1/22:5) GPEtn(O-18:0/22:6)

778

776

GPEtn(O-20:1/20:4)

–

778

GPEtn(18:2/22:6)

788

786

GPEtn(18:1/22:6)

790

788

GPEtn(18:0/22:6)

792

790

GPEtn(20:0/20:4)

796

794

GPEtn(20:4/22:6)

812

810

GPEtn(22:6/22:6)

836

834

GPEtn(22:4:/22:6)

840

838

a  GPEtn fatty acid substituents were determined by negative ion mode MS/MS experiments as described in Subheading 3. Alkenyl species are denoted by O- preceding the acyl chain. A dash indicates a species that was not detected.

for the detection of additional low abundance GPEtn lipid ions. Due to the potential complications arising from the presence of various ionic adducts, as well as the differential fragmentation

54

Busik, Reid, and Lydic

behavior of alkenyl GPEtn species in positive vs. negative ionization mode, the use of both of these MS/MS scan modes to redundantly detect GPEtn lipids allowed for more thorough evaluation of GPEtn lipids and greater confidence in peak identification (see Note 2). The GPEtn lipids identified in rat retina are summarized in Table 3. Fatty acid constituents were analyzed in negative ion mode as described in Subheading 3.3. In this analysis, a total of 24 GPEtn species were identified in rat retina after accounting for isobaric species, with redundant identification achieved for all diacyl species. The detection of GPIns lipids is typically achieved using a negative ion precursor ion scan mode MS/MS experiment, by monitoring for the formation of a characteristic product ion at m/z 241 (corresponding to a dehydrated phosphoinositol) from their [M − H]− ions (scan number 10 in Table 1). The spectrum resulting from this scan, shown in Fig. 6c, enabled the identification of 14 GPIns lipids. The use of a complementary scan (number 11 in Table 1) to monitor for the characteristic neutral loss of 277 Da (corresponding to the combined losses of phosphoinositol + NH3) from the [M + NH4]+ adducts of GPIns Fig. (6d) enabled nine of the lipids observed in the negative ion mode precursor ion scan experiment to be confirmed. This was especially useful in confirming ions of relatively low abundance. For example, the ions at m/z 909 and 929 in the spectrum shown in Fig. 6c represent putative GPIns [M − H]− (Total carbon: total double bond) 40:6 and 42:10 species, respectively. These ions were confirmed as likely GPIns species by the presence of their corresponding [M + NH4]+ ions after conducting NLS 277 (m/z 928 and 948 in Fig. 6d). Further complementary negative ion mode fatty acid analysis by scanning for precursor ions corresponding to deprotonated fatty acyl groups (scan number 18 in Table 1) confirmed the fatty acid constituents of these GPIns species as diacyl 18:0/22:6 and 20:4/22:6, respectively (data not shown.) Taken together, these complementary sets of data enabled confident identification of the ions in question despite their low relative abundance. A compilation of all rat retina GPIns species identified by positive and negative ion mode analysis is provided in Table 4. 3 .4.5. Complementary Precursor Ion and Neutral Loss Scan Mode Analysis of Rat Retina GPSer Lipids

Detection of glycerophosphatidylserine (GPSer) lipid species is typically achieved in either positive or negative ion mode by the use of neutral loss scan mode MS/MS experiments. Figure 7a demonstrates the collisionally induced dissociation of the [M + H]+ ion of synthetic GPSer(14:0/14:0) at m/z 680. The fragmentation of this ion is dominated by the neutral loss of 185 Da corresponding to loss of the phosphoserine headgroup. Figure 7b shows the spectrum obtained from collisionally induced dissociation of the [M + Na]+ ion of the same



Global Analysis of Retina Lipids

55

Table 4 Rat retina GPIns species detected by complementary positive and negative ion mode MS/MS analysis Lipid speciesa

Scan #10 PIS 241 (m/z)

Scan #11 NLS 277 (m/z)

GPIns(16:0/18:1)

835

–

GPIns(16:1/20:4)

855

874

GPIns(16:0/20:4)

857

876

GPIns(18:2/20:4)

881

900

GPIns(16:0/22:6)

881

900

GPIns(18:1/20:4)

883

902

GPIns(18:0/20:4)

885

904

GPIns(18:2/22:6)

905

–

GPIns(18:1/22:6)

907

–

GPIns(18:0/22:6)

909

928

GPIns(20:0/20:4)

913

932

GPIns(20:4/22:6)

929

948

GPIns(20:4/22:6)

931

–

GPIns(22:6/22:6)

953

–

a  GPIns fatty acid substituents were determined by negative ion mode MS/MS experiments as described in Subheading 3. A dash indicates a species that was not detected.

GPSer(14:0/14:0) species. While a neutral loss of 185 Da was observed, the fragmentation of the GPSer(14:0/14:0) [M + Na]+ ion was dominated by the formation of the ion at m/z 208, corresponding to a sodium phosphoserine cation. Additionally, loss of neutral sodium phosphoserine (207 Da) was observed at m/z 495. Figure 7c demonstrates CID of the GPSer(14:0/14:0) [M − H]− ion observed at m/z 678. The most abundant ion observed in this spectrum results from the loss of 87 Da, corresponding to a serine derivative. Additional abundant ions were observed at m/z 381 and 363, corresponding to losses of the 14:0 fatty acids as ketene and neutral species, respectively; m/z 227, corresponding to the deprotonated fatty acid ions; and m/z 153, corresponding to a cyclic glycerolphosphate derivative. As the neutral loss of 185 Da was observed from dissociation of both GPSer [M + H]+ and [M + Na]+ ions, performing the typical positive ion mode analysis of rat retina GPSer by utilizing NLS 185 (scan number 12 in Table 1), resulted in

56

Busik, Reid, and Lydic

Fig. 7. Product ion mode MS/MS spectrum of synthetic GPSer(14:0/14:0) [M + H]+ ions (a), [M + Na]+ ions (b), and [M − H]− ions (c).

detection of both GPSer [M + H]+ and [M + Na]+ ions (see Fig. 8a). As observed in the positive ion analysis of GPEtn lipids, the presence of GPSer species in two different ionic forms within the same spectrum complicated interpretation of the spectrum, as ions that differ by 22 mass units could represent a difference of an adducted sodium ion, or the mass difference represented by the addition of two fatty acid methylene groups and three double bonds. The spectrum in Fig. 8a contains several sets of ions that differ by 22 mass units, such as m/z 790 and m/z 812; m/z 812 and m/z 834; m/z 836 and m/z 858; m/z 858 and m/z 880; and m/z 880 and m/z 902. In each case, the ion observed at an increase of 22 mass units could represent either a sodium-adducted species of the ion at the lower nominal mass, or a unique GPSer molecular species with one or more fatty acids that contain additional carbons and double bonds relative to the GPSer species at the lower nominal mass. Additionally, the spectrum obtained by NLS 185 contained several ions of very low relative abundance (denoted by bold type), including m/z 762, m/z 790, and m/z 936. In order to provide confirmation of low abundance rat retina GPSer species and clarify the identities and ionic forms of ions observed in the NLS 185 spectrum, we employed two additional complementary



Global Analysis of Retina Lipids

57

Fig. 8. Complementary positive and negative ion mode MS/MS analysis of rat retina GPSer species. Bold text indicates m/z of species at very low relative abundance that were redundantly detected in multiple ion forms. (a) Detection of GPSer as [M + H]+ and [M + Na]+ ions by positive ion mode neutral loss scanning of 185 Da. (b) Detection of GPSer as [M + Na]+ ions by positive ion mode precursor ion scanning of m/z 208. (c) Detection of GPSer as [M − H]− ions by negative ion mode neutral loss scanning of 87 Da.

GPSer-specific MS/MS scans. Figure 8b demonstrates use of a precursor ion scan for m/z 208 to detect formation of the sodium phosphoserine ion from GPSer [M + Na]+ ions (scan number 13 in Table 1.) The spectrum obtained by performing this scan contained ions of only a single ionic form and elucidated m/z at which sodium-adducted GPSer species contributed to total ion current of the NLS 185 spectrum. The use of this complementary precursor ion scan would be useful in any positive ion mode analysis of GPSer in order to confirm the presence or absence of GPSer [M + Na]+ ions, or to simplify relative quantitation of GPSer species against an internal standard. Furthermore, the use of PIS 208 enabled redundant detection of the low abundance ions observed in the spectrum obtained by scanning for the neutral loss of 185 Da. The ions in bold in Fig. 8b at m/z 784, m/z 812, and m/z 958 represent sodiated adducts of the protonated GPSer ions observed at m/z 762, m/z 790, and m/z 936, respectively, detected by the neutral loss of 185 Da.

58

Busik, Reid, and Lydic

Negative ion mode analysis of rat retina GPSer was achieved by the commonly used neutral loss of 87 Da (scan number 14 in Table 1). The spectrum obtained from this neutral loss mode MS/MS experiment is shown in Fig. 8c. Performing NLS 87 results in detection of GPSer molecular species in only one ionic form (i.e., [M − H]−), which offers the same advantage as the positive ion mode PIS 208 in terms of spectrum simplification and ease of relative quantification against an internal standard. The detection of GPSer [M − H]− ions by NLS 87 offers an additional advantage in that ions identified by this MS/ MS scan may in turn be used for direct analysis of esterified fatty acid constituents by product ion mode MS/MS experiments, or by precursor ion scanning of individual fatty acid anions. The use of NLS 87 to detect rat retina GPSer also provided another means of confirming the presence of low abundance ions detected by positive ion GPSer analysis. The ions in bold in Fig. 8c at m/z 760, m/z 788, and m/z 934 represent deprotonated equivalents of the low-abundance ions detected by NLS 185 and PIS 208. Subsequent analysis of the fatty acid constituents of these GPSer species from their deprotonated ions revealed that the identities of these low abundance ions were diacyl GPSer(16:0/18:1), GPSer(18:0/18:1), and GPSer(24:6/24:6). Notably, an abundant ion at m/z 810 was observed in the spectrum obtained from the negative ion NLS 87 MS/MS experiment. This ion represents the deprotonated form of the protonated ion observed at m/z 812 in the NLS 185 MS/MS experiment. However, an ion at m/z 812 was also observed in the PIS 208 MS/MS experiment, indicating that a small contribution of the ion current for m/z 812 in the NLS 185 spectrum was made by a GPSer sodium adduct of the ion at m/z 790. The quantification of m/z 812 in the NLS 185 spectrum against an internal standard would necessitate accommodation for the presence of underlying the GPSer [M + Na]+ ion; however, the presence of the underlying GPSer [M + Na]+ ion would not have been detected without performing the PIS 208 MS/MS scan. Furthermore, complementary positive and negative ion mode analysis of GPSer utilizing only NLS 185 and NLS 87, respectively, would have falsely confirmed that the ion detected at m/z 812 in the NLS 185 MS/MS scan consisted entirely of GPSer [M + H]+ species. This analysis underscores the advantages of employing lipid class-specific MS/MS detection methods in which lipid species are detected as only one ionic form. The ability to redundantly detect low abundance rat retina GPSer species in multiple ionic forms enabled increased confidence in peak calling and subsequent inclusion of these ions in analysis of esterified fatty acid groups. The ability to identify both protonated and sodium-adducted GPSer molecular ions in positive ion mode MS/MS analysis expanded the number of spectral features



Global Analysis of Retina Lipids

59

that could be assigned in the spectrum obtained from the initial MS analysis. Importantly, the use of a precursor ion scan of m/z 208 to identify GPSer [M + Na]+ ions simplified positive ion GPSer analysis by clarifying the identities of lipid species at m/z where potential overlap existed between protonated and sodiated GPSer ions of differing fatty acid composition. Use of the negative ion NLS 87 scan for detection of GPSer enabled the identification of spectral features of the negative ion mode MS spectra, and provided additional confirmation of GPSer species observed at very low abundance. Relative quantification of GPSer would be most straightforward when using either the negative ion NLS 87 MS/MS experiment, or the positive ion PIS 208 MS/MS experiment when known amounts of exogenous salts have been added to a lipid extract; this scan is also useful for determining the presence of endogenous salts that may be present following lipid extraction. After taking into account the redundant lipid species identified in both the negative ion scan mode neutral loss experiment and the positive ion scan mode neutral loss and precursor ion experiments, a total of 29 GPSer molecular species could be identified in rat retina, with 22 redundant identifications. See Table 5 for a detailed catalog of all identified rat retina GPSer species. This example clearly highlights the utility of alternative or complementary MS/MS scans that are not commonly employed in lipidome-profiling studies to provide more detailed insights into the composition of complex lipid mixtures. 3 .4.6. Complementary Neutral Loss Scan Mode Analysis of Rat Retina DG and TG Lipids

Nonpolar lipids, such as diacylglycerols (DG) and triacylglycerols (TG), have typically required derivatization and/or chromatographic separation prior to ESI-MS analysis (33–36). This limits the ability to analyze both polar and nonpolar lipid classes from limited sample amounts. We have applied the principle of complementary neutral loss scan mode MS/MS to the analysis of DG and TG species in the same unfractionated rat retina extract that was used for analysis of the polar lipid classes described above. The addition of ammonium hydroxide (other other ammonium salts) to lipid extracts prior to mass spectrometry analysis provides abundant ammonium cations that readily adduct neutral lipids such as DG’s and TG’s. Redundant detection of DG and TG [M + NH4]+ ions was achieved by neutral loss scanning for multiple m/z corresponding to the loss of a neutral fatty acid + NH3 (scan number 15 in Table 1.) By comparing spectra obtained from neutral loss scans corresponding to multiple naturally occurring fatty acids, the fatty acid moieties esterified to the glycerol backbone of species at each m/z were elucidated. Additionally, we have found that mono- and diacylglycerol [M + NH4]+ ions are readily identified by neutral loss scanning of 35 Da, corresponding to an abundant loss of NH3 + H2O (scan number 17 in Table 1.) This neutral loss scan was advantageous when

Table 5 Rat retina GPSer molecular species detected by complementary positive and negative ion mode MS/MS analysis Lipid species a

Scan #12 NLS 185 (m/z)

Scan #13 PIS 208 (m/z)

Scan #14 NLS 87 (m/z)

GPSer(16:0/18:1)

762

784

760

GPSer(16:1/18:0)

762

784

760

GPSer(16:0/18:0)

–

–

762

GPSer(18:0/18:1)

790

812

788

GPSer(16:1/20:4)

–

–

808

GPSer(16:0/20:4)

812

834

810

GPSer(16:0/22:0)

–

842

818

GPSer(18:0/20:0)

–

842

818

GPSer(O-18:0/22:6)

–

842

818

GPSer(18:1/22:6)

834

856

832

GPSer(18:0/22:6)

836

858

834

GPSer(18:0/22:5)

838

860

836

GPSer(18:0/22:4)

840

862

838

GPSer(20:0/20:4)

840

862

838

GPSer(20:4/22:6)

856

878

854

GPSer(20:4/22:5)

858

–

856

GPSer(20:4/22:4)

–

–

858

GPSer(20:0/22:6)

–

886

862

GPSer(20:0/22:5)

866

–

864

GPSer(22:6/22:6)

880

902

878

GPSer(22:5/22:6)

882

904

880

GPSer(22:4/22:6)

884

906

882

GPSer(22:3/22:6)

–

908

884

GPSer(22:6/24:6)

908

930

906

GPSer(22:5/24:6)

910

932

908

GPSer(22:4/24:6)

912

934

910

GPSer(22:3/24:6)

–

–

912

GPSer(24:6/24:6)

936

958

934

GPSer(24:5/24:6)

938

–

936

a  GPSer fatty acid substituents were determined by negative ion mode MS/MS experiments as described in the Subheading 3. Alkenyl species are denoted by O-preceding the acyl chain. A dash indicates a species that was not detected.



Global Analysis of Retina Lipids

61

simultaneous comparison of all MG/DG species to an internal standard was desired. Use of the NLS 35 MS/MS experiment was also useful for identifying DG species present in the ammoniumadducted fatty acid neutral loss spectra, as scanning for loss of a fatty acid + NH3 may also detect cholesteryl esters present at very low abundance in rat retina. Thus, NLS 35 enabled differentiation between putative DG and cholesteryl ester ions detected by positive ion mode fatty acid analysis of neutral lipid species. This neutral loss mode MS/MS scan also provided another means of verifying low abundance putative DG peaks identified in one or more fatty acid neutral loss scan. The limitation of NLS 35 was that DG species were only identified in terms of total carbons: total double bonds, which could be deduced from the nominal mass of a given species. Figure 9a shows the spectrum obtained from simultaneous analysis of all abundant rat retina DG species by scanning for a neutral loss of 35 Da between m/z 580 and 700. m/z 614 was detected as a putative diglyceride in this scan, consistent with the mass of a 34:0 (TC:TDB) diglyceride as an [M + NH4]+ ion. The fatty acid compositions of DG and TG species were then determined by scanning for the neutral loss of a 16:0 fatty acid + NH3 (Fig. 9b), loss of an 18:0 fatty acid + NH3 (Fig. 9c), and the loss of a 22:6 fatty acid + NH3 (Fig. 9d), over a wider mass range (m/z 580–980). m/z 614 was redundantly detected in scans for the neutral loss of 16:0 and 18:0 fatty acids, confirming the identification of this m/z as DG(16:0/18:0). Analysis of the high mass range of the spectra shown in Fig. 9b–d also revealed the presence of m/z 924 in all three fatty acid neutral loss scans. This led to the identification of the TG species as TG(16:0/18:0/22:6). However, an ion at low relative abundance was observed at m/z 670 in the spectrum obtained from scanning for a neutral loss of an 18:0 fatty acid (Fig. 9c) that was not observed in the NLS 35 MS/MS experiment used to identify DG species. While m/z 670 is within the expected mass range for putative DG molecular species, the lack of an ion at m/z 670 in the NLS 35 spectrum cast doubt that this particular ion represented an 18:0 fatty acid-containing diglyceride. Direct product ion MS/MS analysis of m/z 670 was precluded by the exceedingly low abundance of this ion relative to polar lipid ion abundances in the unfractionated rat retina lipid extract. As the expected mass range for cholesteryl esters is similar to the expected mass range of diglycerides, we conducted an additional precursor ion scan of m/z 369 (scan number 16 in Table 1) to detect cholesteryl esters as [M + NH4]+ ions. A peak at m/z 670 was observed in the PIS 369 spectrum, corresponding to an 18:0 cholesteryl ester (data not shown). This was in agreement with the original observation of m/z 670 in a neutral loss scan of 18:0 fatty acid neutral species. Hence, the use of complementary precursor ion and neutral loss scan mode MS/MS experiments enabled detailed

62

Busik, Reid, and Lydic

Fig. 9. Complementary positive ion mode MS/MS analysis of rat retina DG and TG species. (a) Simultaneous detection of all abundant DG species as [M + NH4]+ ions by neutral loss scanning of 35 Da. (b) Detection of all 16:0-containing DG and TG as [M + NH4]+ ions by neutral loss scanning of 273 Da. (c) Detection of all 18:0-containing DG and TG as [M + NH4]+ ions by neutral loss scanning of 301 Da. (d) Detection of all 22:6-containing DG and TG as [M + NH4]+ ions by neutral loss scanning of 345 Da.

analysis and redundant confirmation of low abundance ions corresponding to rat retina neutral lipids without chromatographic separation of polar and nonpolar analytes. Note that the fatty acids selected for analysis by neutral loss scanning presented here were for demonstrative purposes only; a thorough analysis was conducted by scanning for neutral losses of all naturally occurring fatty acids. From this set of complementary analyses, we delineated a list of 29 diacylglycerol and 111 triacylglycerol distinct molecular species (see Table 6).

Lipid species

DG(16:0/16:1)

DG(16:0/16:0)

DG(16:0/18:1)

DG(16:1/18:0)

DG(16:0/18:0)

DG(16:0/20:4)

DG(18:1/18:1)

DG(16:1/20:0)

DG(18:0/18:1)

DG(18:0/18:0)

DG(16:0/20:0)

DG(16:0/22:6)

DG(18:1/20:4)

DG(16:0/22:5)

DG(18:0/20:4)

DG(18:0/20:3)

m/z

584

586

612

612

614

634

638

640

640

642

642

658

660

660

662

664

848

846

846

846

846

844

844

844

844

824

824

822

822

820

820

820

m/z

TG(14:0/18:1/18:1)

TG(18:1/14:1/18:1)

TG(18:1/14:0/18:2)

TG(16:1/16:1/18:1)

TG(16:0/18:2/16:1)

TG(18:2/14:0/18:2)

TG(18:1/18:2/14:1)

TG(16:1/16:1/18:2)

TG(16:0/16:1/18:3)

TG(16:0/16:0/16:0)

TG(14:0/16:0/18:0)

TG(16:0/16:0/16:1)

TG(14:0/16:0/18:1)

TG(16:0/16:1/16:1)

TG(14:1/16:1/18:0)

TG(14:0/16:0/18:2)

Lipid species

898

898

898

898

896

896

896

894

894

894

894

880

878

878

876

876

m/z

TG(16:0/20:3/18:2)

TG(16:0/18:2/20:3)

TG(16:0/18:1/20:4)

TG(16:0/16:0/22:5)

TG(18:2/18:2/18:2)

TG(16:0/18:2/20:4)

TG(16:0/16:0/22:6)

TG(18:1/14:1/22:5)

TG(18:0/14:1/22:6)

TG(16:1/18:3/20:3)

TG(16:0/22:6/16:1)

TG(16:0/18:0/18:0)

TG(16:1/16:0/20:0)

TG(16:1/18:0/18:1)

TG(16:1/16:1/20:0)

TG(16:0/18:0/18:1)

Lipid species

956

952

950

950

948

948

946

944

928

926

926

926

924

924

924

922

m/z

(continued)

TG(18:0/20:0/20:4)

TG(18:0/18:0/22:6)

TG(18:0/18:2/22:5)

TG(18:0/18:1/22:6)

TG(18:1/18:1/22:6)

TG(18:0/18:2/22:6)

TG(18:1/20:4/20:4)

TG(18:1/22:6/18:3)

TG(18:0/18:0/20:4)

TG(18:0/18:1/20:4)

TG(16:0/20:1/20:4)

TG(16:0/18:0/22:5)

TG(16:1/18:0/22:5)

TG(16:0/18:0/22:6)

TG(16:0/16:0/24:6)

TG(18:1/20:4/18:2)

Lipid species

Table 6 Rat retina DG and TG molecular species detected as [M + NH4]+ ions by complementary positive ion mode neutral loss scanning of m/z corresponding to the mass of naturally occurring fatty acids+NH3 (scan #15)

Global Analysis of Retina Lipids 63

Lipid species

DG(18:2/20:0)

DG(18:1/20:0)

DG(18:0/20:0)

DG(20:4/20:4)

DG(18:1/22:6)

DG(18:0/22:6)

DG(18:0/22:5)

DG(20:0/20:4)

DG(20:0/20:0)

DG(20:4/22:6)

DG(20:0/22:6)

DG(20:0/22:3)

DG(22:6/22:6)

TG(14:0/14:0/16:0)

TG(14:0/14:0/18:1)

TG(14:0/14:1/18:0)

m/z

666

668

670

682

684

686

688

690

698

706

714

720

730

768

794

794

Table 6 (continued)

872

872

TG(16:1/18:0/18:3)

TG(16:0/20:3/16:1)

TG(16:0/18:2/18:2)

TG(16:0/18:1/18:3)

872 872

TG(16:0/16:0/20:4)

TG(18:2/18:3/16:0)

TG(16:1/18:3/18:1)

TG(16:1/18:2/18:2)

TG(16:1/16:1/20:3)

TG(16:0/20:4/16:1)

TG(16:0/16:0/18:0)

TG(18:0/14:0/18:1)

TG(16:0/18:0/16:1)

TG(16:0/16:0/18:1)

TG(18:1/14:0/18:1)

TG(16:0/16:1/18:1)

TG(16:0/16:0/18:2)

Lipid species

872

870

870

870

870

870

852

850

850

850

848

848

848

m/z

906

906

906

904

904

904

902

902

902

900

900

900

900

900

898

898

898

m/z

TG(18:0/18:0/18:1)

TG(16:0/18:0/20:1)

TG(16:0/18:1/20:0)

TG(18:0/18:1/18:1)

TG(18:0/18:0/18:2)

TG(16:0/20:1/18:1)

TG(18:2/20:1/16:0)

TG(18:1/18:2/18:0)

TG(18:1/18:1/18:1)

TG(18:1/18:1/18:2)

TG(18:0/18:2/18:2)

TG(18:0/18:1/18:3)

TG(16:0/18:1/20:3)

TG(16:0/18:0/20:4)

TG(18:2/18:2/18:1)

TG(16:1/20:3/18:1)

TG(16:1/20:1/18:3)

Lipid species

978

974

972

970

968

968

968

966

m/z

TG(18:2/22:5/20:0)

TG(18:0/20:4/22:5)

TG(18:0/20:4/22:6)

TG(16:1/22:5/22:5)

TG(22:6/22:6/16:0)

TG(18:3/20:3/22:6)

TG(18:2/22:6/20:4)

TG(20:4/20:4/20:5)

Lipid species

64 Busik, Reid, and Lydic

TG(14:0/16:1/18:2)

TG(16:1/14:1/18:1)

TG(16:1/16:1/16:1)

TG(14:0/16:0/18:2)

TG(14:0/16:1/18:1)

818

818

818

820

820

876

876

874

874

874

872

TG(16:0/18:1/18:1)

TG(16:0/18:0/18:2)

TG(16:1/18:1/18:1)

TG(16:1/18:0/18:2)

TG(16:0/18:1/18:2)

TG(16:1/18:1/18:2)

Stereochemistry of fatty acids has not been verified

TG(14:0/16:1/16:0)

794

922

922

922

922

922

922

TG(16:1/18:1/22:5)

TG(16:1/20:2/20:4)

TG(16:1/18:0/22:6)

TG(16:0/16:1/24:6)

TG(16:0/20:3/20:4)

TG(16:0/18:1/22:6)

Global Analysis of Retina Lipids 65

66

Busik, Reid, and Lydic

4. Notes 1. Although disruption of lipid metabolism is known to play a key role in the onset and progression of a variety of retinal diseases, the use of retinal lipid profiles has not been sufficiently explored. This lack of knowledge is largely due to the diversity of lipids and to limits in the sensitivity of previously available analytical methods. The methodology described in this chapter for lipid profile analysis of normal retina will be further applied to various diseased retinal states to aid in developing effective therapeutic strategies for retinal disorders. 2. From the data described above, it is clear that crude lipid extracts from whole rat retinas contain an extremely complex mixture of molecular species, representing a wide array of lipid classes at varying abundances. The observation of multiple ionic adducts of a given lipid class in both positive (e.g., [M + H]+, [M + Na]+, and [M + NH4]+) and negative (e.g., [M − H]−, [M + Cl]− and [M + CH3OCO2]−) ionization modes adds further complexity, effectively increasing the number of “spectral features” to the resultant mass spectra obtained by analysis of these extracts. However, the ability to identify specific lipid species in both positive and negative ionization modes, as well as in their multiple adducted forms, enables unambiguous lipid identification, particularly important when lipids were present at low abundance. 3. In many cases, multiple lipid species were found to be present at a single nominal m/z value. For example, m/z 834 in the positive ion mode MS spectrum shown in Fig. 1a, was found to contain diacyl GPCho(18:0/22:6) [M + H]+, GPSer (18:1/22:6) [M + H]+, and GPSer(18:0/20:4) [M + Na]+ ions. Likewise, m/z 834 in the negative ionization mode spectrum shown in Fig. 1b was found to consist of [M − H]− ions of GPSer(18:0/22:6) and GPEtn(22:6/22:6), as well as the methyl carbonate adduct ([M + CH3OCO2]−) of GPCho(16:0/18:1). Detection of lipid classes in multiple ionization forms was necessary to fully elucidate the identities of ions present in both positive and negative ion mass spectra at numerous m/z. This point becomes especially important when comparisons must be made across numerous samples, such as comparisons between normal and disease states, in which case any observed change in ion abundance between two or more groups of samples must be accurately characterized. 4. To assist in the implementation of complementary analysis of major lipid classes within the lipidome, we have assembled multiple pairings of possible precursor ion and neutral loss scan mode MS/MS experiments to enable redundant detec-



Global Analysis of Retina Lipids

67

tion of each lipid class. The advantages and disadvantages of many of these scan types have been discussed, and in several cases, the use of redundant detection of a lipid class minimized the shortcomings of one or more MS/MS experiments. For example, positive ion analysis of GPEtn [M + H]+ ions by neutral loss scan mode MS/MS was found to be more sensitive than negative ion mode precursor ion analysis of GPEtn [M − H]− ions in terms of signal/noise and absolute ion abundance. The more sensitive positive ion neutral loss scan is therefore optimal for analysis of low abundance GPEtn ions. However, the additional presence of [M + Na]+ ions, even at very low abundance, complicates analysis of the resultant positive ion MS/MS spectra and necessitates complementary analysis of GPEtn [M − H]− ions for clarity. Negative ion MS/ MS analysis of GPEtn enables the identification of alkenyl species that are not observed in positive ion GPEtn MS/MS analysis; yet the significantly lower sensitivity of the negative ion mode GPEtn precursor ion scan precludes analysis of lipid species that may be present at relatively low abundance. 5. Most of the individual MS/MS experiments we have paired for use in complementary analysis of rat retina lipid classes were obtained from the available literature. In several cases we have employed novel precursor ion and neutral loss scans derived from previously published gas-phase fragmentation pathways to detect lipid classes in various ionic forms or modification states. To our knowledge, this analysis of the rat retina lipidome was the first analysis to include the negative ion neutral loss of 135 Da for detection of SM and GPCho species as methylcarbonate adducts; likewise, it was the first analysis to explore differential detection of SM and GPCho as methylcarbonate adducts by use of the negative ion neutral loss of 161 Da for preferential detection of SM species; and it was the first analysis to include detection of sodiated GPCho and GPSer species by positive ion mode precursior ion scanning of m/z 147 and m/z 208, respectively. Additionally, we introduced the positive ion mode neutral loss of 35 Da for simultaneous detection of diglyceride species as ammonium adducts. In this chapter we have provided a preliminary outline of the possible benefits and shortcomings of each of these novel MS/MS experiments, although a more rigorous examination of each proposed scan would be beneficial for assessment of the viability of these scans for use in large-scale lipid-profiling experiments. 6. We have provided, as a reference, pairings of MS/MS experiments in Table 1 that can be used in the complementary analysis of lipid classes that were not discussed in this chapter, including cholesteryl esters, glycerophosphatidic acid (GPA), glycerophosphatidylglycerol (GPGro), and ceramides. As these lipid classes are of low abundance in many tissues, and

68

Busik, Reid, and Lydic

especially in rat retina, their analysis certainly benefits from complementary analysis of multiple ion forms where possible, or by redundant detection of the same ion form by multiple scans. The pairings suggested for complementary analysis of lipid classes are certainly not exhaustive, and other sets of complementary pairs could be substituted for analysis of various lipid classes. Furthermore, as the understanding of gasphase fragmentation pathways of lipid classes and their ionic adducts progresses, it will expand the ability to perform complementary analysis to lipid classes that have not been addressed in this chapter. 7. Clearly, the use of redundant identification of rat retina lipid classes from multiple precursor ion types enabled the identification of molecular species that would not have been observed if only one MS/MS scan mode had been employed in the analysis of each lipid class. This was especially true in the analysis of SM and GPCho species, in which only four rat retina SM species could be readily detected by the conventional positive ion precursor ion scan of m/z 184; four SM species were detected by neutral loss of 183 Da from sodiated SM ions; two SM species were detected by PIS 147 from sodiated SM ions; and 12 SM species were detected by the negative ion neutral loss of 161 Da from SM methylcarbonate adducts.

References 1. Anderson RE. Lipids of ocular tissues. IV. A comparison of the phospholipids from the retina of six mammalian species. Exp Eye Res 1970;10(2):339–44. 2. Fliesler SJ, Anderson RE. Chemistry and metabolism of lipids in the vertebrate retina. Prog Lipid Res 1983;22(2):79–131. 3. Lecomte M, Paget C, Ruggiero D, Wiernsperger N, Lagarde M. Docosahexaenoic acid is a major n-3 polyunsaturated fatty acid in bovine retinal microvessels. J Neurochem 1996;66(5):2160–7. 4. O’Brien JS, Sampson EL. Fatty acid and fatty aldehyde composition of the major brain lipids in normal human gray matter, white matter, and myelin. J Lipid Res 1965;6(4):545–51. 5. SanGiovanni JP, Chew EY. The role of omega-3 long-chain polyunsaturated fatty acids in health and disease of the retina. Prog Retin Eye Res 2005;24(1):87–138. 6. Wang N, Anderson RE. Synthesis of docosahexaenoic acid by retina and retinal pigment epithelium. Biochem (Mosc) 1993;32(49): 13703–9.

7. Wang N, Anderson RE. Transport of 22:6n-3 in the plasma and uptake into retinal pigment epithelium and retina. Exp Eye Res 1993; 57(2):225–33. 8. Soubias O, Teague WE, Gawrisch K. Evidence for specificity in lipid–rhodopsin interactions. J Biol Chem 2006;281(44):33233–41. 9. Suh M, Wierzbicki AA, Clandinin MT. Dietary fat alters membrane composition in rod outer segments in normal and diabetic rats: impact on content of very-long-chain (C > or = 24) polyenoic fatty acids. Biochim biophys Acta 1994;1214(1):54–62. 10. Watson AD. Lipidomics: a global approach to lipid analysis in biological systems. J Lipid Res 2006;47(10):2101–11. 11. Brugger B, Erben G, Sandhoff R, Wieland FT, Lehmann WD. Quantitative analysis of biological membrane lipids at the low picomole level by nano-electrospray ionization tandem mass spectrometry. Proc Natl Acad Sci U S A 1997;94(6):2339–44. 12. Ejsing CS, Duchoslav E, Sampaio J, et  al. Automated identification and quantification

of glycerophospholipid molecular species by multiple precursor ion scanning. Anal Chem 2006;78(17):6202–14. 13. Han X, Gross RW. Shotgun lipidomics: electrospray ionization mass spectrometric analysis and quantitation of cellular lipidomes directly from crude extracts of biological samples. Mass Spectrom Rev 2005;24(3):367–412. 14. Pulfer M, Murphy RC. Electrospray mass spetrometry of phospholipids. Mass Spectrom Rev 2003;22(5):332–64. 15. Schwudke D, Oegema J, Burton L, et al. Lipid profiling by multiple precursor and neutral loss scanning driven by the data-dependent acquisition. Anal Chem 2006;78(2):585–95. 16. Welti R, Wang X, Williams TD. Electrospray ionization tandem mass spectrometry scan modes for plant chloroplast lipids. Anal Biochem 2003;314(1):149–52. 17. Welti R, Shah J, LeVine S, Esch SW, Williams TD, Wang X. High throughput lipid profiling to identify and characterize genes involved in lipid metabolism, signaling, and stress response. In: L. F, G.D. P, eds. Functional Lipidomics. New York: Marcel Dekker; 2005. 18. Jones JJ, Batoy SM, Wilkins CL. A comprehensive and comparative analysis for MALDI FTMS lipid and phospholipid profiles from biological samples. Comput Biol Chem 2005; 29(4):294–302. 19. Jones JJ, Borgmann S, Wilkins CL, O’Brien RM. Characterizing the phospholipid profiles in mammalian tissues by MALDI FTMS. Anal Chem 2006;78(9):3062–71. 20. Jones JJ, Stump MJ, Fleming RC, Lay JO, Jr., Wilkins CL. Strategies and data analysis techniques for lipid and phospholipid chemistry elucidation by intact cell MALDI-FTMS. J Am Soc Mass Spectrom 2004;15(11):1665– 74. 21. Zhang X, Reid GE. Multistage tandem mass spectrometry of anionic phosphatidylcholine lipid adducts reveals novel dissociation pathways. Int J Mass Spectrom 2006;252(3):242–55. 22. Koivusalo M, Haimi P, Heikinheimo L, Kostiainen R, Somerharju P. Quantitative determination of phospholipid compositions by ESI-MS: effects of acyl chain length, unsaturation, and lipid concentration on instrument response. J Lipid Res 2001;42(4):663–72. 23. Zacarias A, Bolanowski D, Bhatnagar A. Comparative measurements of multicomponent phospholipid mixtures by electrospray mass spectroscopy: relating ion intensity to concentration. Anal Biochem 2002;308(1):152–9. 24. Folch J, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem 1957;226(1):497–509.

Global Analysis of Retina Lipids

69

25. Haimi P, Uphoff A, Hermansson M, Somerharju P. Software tools for analysis of mass spectrometric lipidome data. Anal Chem 2006; 78(24):8324–31. 26. Callender HL, Forrester JS, Ivanova P, Preininger A, Milne S, Brown HA. Quantification of diacylglycerol species from cellular extracts by electrospray ionization mass spectrometry using a linear regression algorithm. Anal Chem 2007;79(1):263–72. 27. Colsch B, Afonso C, Popa I, et al. Characterization of the ceramide moieties of sphingoglycolipids from mouse brain by ESI-MS/MS: identification of ceramides containing sphingadienine. J Lipid Res 2004;45(2):281–6. 28. Murphy RC, James PF, McAnoy AM, Krank J, Duchoslav E, Barkley RM. Detection of the abundance of diacylglycerol and triacylglycerol molecular species in cells using neutral loss mass spectrometry. Anal Biochem 2007;366(1): 59–70. 29. Merrill AH, Sullards MC, Allegood JC, Kelly S, Wang E. Sphingolipidomics: High-throughput, structure-specific, and quantitative analysis of sphingolipids by liquid chromatography tandem mass spectrometry. Methods 2005; 36(2):207–24. 30. Han X. Characterization and direct quantitation of ceramide molecular species from lipid extracts of biological samples by electrospray ionization tandem mass spectrometry. Anal Biochem 2002;302(2):199–212. 31. Song H, Hsu F, Ladenson J, Turk J. Algorithm for processing raw mass spectrometric data to identify and quantitate complex lipid molecular species in mixtures by data-dependent scanning and fragment ion database searching. J Am Soc Mass Spectrom 2007;18:1848–58. 32. Houjou T, Yamatani K, Nakanishi H, Imagawa M, Shimizu T, Taguchi R. Rapid and selective identification of molecular species in phosphatidylcholine and sphingomyelin by conditional neutral loss scanning and MS3. Rapid Commun Mass Spectrom 2004;18(24):3123– 30. 33. Liebisch G, Binder M, Schifferer R, Langmann T, Schulz B, Schmitz G. High throughput quantification of cholesterol and cholesteryl ester by electrospray ionization tandem mass spectrometry (ESI-MS/MS). Biochim Biophys Acta 2006;1761(1):121–8. 34. Hvattum E. Analysis of triacylglycerols with non-aqueous reversed-phase liquid chromatography and positive ion electrospray tandem mass spectrometry. Rapid Commun Mass Spectrom 2001;15(3):187–90. 35. Li YL, Su X, Stahl PD, Gross ML. Quantification of diacylglycerol molecular species in biological samples by electrospray ionization mass

70

Busik, Reid, and Lydic

spectrometry after one-step derivatization. Anal Chem 2007;79(4):1569–74. 36. Cai SS, Syage JA. Comparison of atmospheric pressure photoionization, atmospheric pressure chemical ionization, and electrospray ionization mass spectrometry for analysis of lipids. Anal Chem 2006;78(4):1191–9.

37. Han XL, Gross RW. Structural determination of picomole amounts of phospholipids via electrospray ionization tandem mass spectrometry. J Am Soc Mass Spectr 1995;6(12):1202–10. 38. Taguchi R, Houjou T, Nakanishi H, et  al. Focused lipidomics by tandem mass spectrometry. J Chromatogr 2005;823(1):26–36.

Chapter 4 Combining Lipidomics and Proteomics of Human Cerebrospinal Fluids Alfred N. Fonteh and Rachel D. Fisher Summary  Lipids from dietary sources or from de novo synthesis are transported while bound to proteins to other tissues where they are used for cell membrane synthesis or stored for energy generation. In cell membranes or in plasma, lipids can undergo several modifications that are important in cell function. Several proteins orchestrate the transport, biosynthesis, and modification of lipids. Thus, the intersection of lipids and proteins is important in human metabolic pathways. Recent advances in mass spectrometry and bioinformatics have made it possible to obtain compositional (structural and functional) data of lipid molecular species and proteins in biological samples. This combination of lipidomics and proteomics is advantageous because it allows us to better define biochemical pathways, discover new drug targets, and better understand the pathophysiology of several diseases. Key words:  Lipidomics, Proteomics, Shotgun sequencing, Mass spectrometry, Liquid chromatography, Isotope dilution, Cerebrospinal fluid

1. Introduction Lipid metabolic pathways are complex, comprising diverse substrates and products (1). Generally, most unsaturated lipids are obtained from dietary sources while de novo fatty acid synthesis accounts for saturated fats. Thus, there are several enzymes involved in the uptake, transport, biosynthesis, and modification of lipids (Fig. 1). Once ingested, lipids may be subjected to lipolysis before adsorption in the gut. Once absorbed, lipids may be modified in the liver and transported to other parts of the body while bound to albumin or to apolipoprotein particles of different sizes (2–4). Lipids are incorporated into cellular membranes or are used as a major storage form of energy in adipose tissue. Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, DOI 10.1007/978-1-60761-322-0_4, © Humana Press, a part of Springer Science + Business Media, LLC 2009

71

72

Fonteh and Fisher

Fig. 1. The intersection of lipidomics and proteomics – Lipids from diets are subjected to lipolysis before adsorption in the gut. Absorbed lipids are modified in the liver and transported to other parts of the body bound to several transport proteins. In other tissues, lipids are incorporated into cellular membranes or are stored in adipose tissues as a major form of energy. Several proteins can modify plasma or cell membrane-bound lipids. Modified lipids often signify physiologic changes characteristic of inflammatory, immune, or other cell responses. We also identify four major classes of proteins that intersect with lipids that are listed at the bottom of Fig. 1.

In cell membranes or when bound to plasma proteins, lipids and several proteins interact, resulting in the modifications of lipids. Oxidative, inflammatory, or immune response usually results in the modification of lipids (5–14). The most studied pathway involves the initial digestion of lipids by lipases (phospholipase A2,



Combining Lipidomics and Proteomics of Human Cerebrospinal Fluids

73

phospholipase C, phospholipase D, lipases) (15–37). A combination of lipase and oxygenase activities results in the generation of bioactive lipids. For example, PLA2 activity on phospholipids results in the release of fatty acids and lysophospholipids. When the fatty acid is arachidonic acid (AA, 20;4, n-6), it is further metabolized by cyclooxygenase, lipoxygenase, or epoxygenases to form lipid mediators of inflammation collectively known as eicosanoids (38–43). The lysophospholipid generated by the same PLA2 action can also be acetylated to a potent lipid mediator of inflammation known as platelet-activating factor (44–46). From this example, it is clear that lipids intersect with proteins and this intersection can have ramifications on cellular function. Thus, a better understanding of how lipids are modified in diseases should encompass a targeted study of proteins which modify these lipids. Currently, understandings of these biochemical pathways involve metabolic labeling experiments in combination with chromatographic separations to quantify changes in lipid substrates or products. Detection of enzyme expression involves activity assays using fluorescent or colorimetric substrates in combination with antibody-based assays (western analysis, ELISA) or pharmacologic assays that employ different inhibitors (47–49). While these methods have increased our knowledge of lipid biochemical pathways, only a limited number of substrates, products or proteins can be measured in the same assay. In cases where sample amounts are limiting, a comprehensive approach involving lipidomic and proteomic approaches will allow us to elucidate biochemical pathways important in regulating lipid metabolism (1). Our approach is to use liquid chromatography in combination with mass spectrometry to catalog lipids and proteins involved in the pathophysiology of brain diseases.

2. Materials 2.1. Equipment 2.1.1. Equipment for Lipidomics

1. Borosilicate glass tubes (clean with chloroform and heat at 100°C for at least 1 h). 2. HPLC system (HP 1100, Hewlett Packard) fitted with a Binary pump, autosampler and column heater. 3. HPLC column – (Luna Silica, 150 × 3.0 mm) from Phenomenex (Torrance, CA). 4. Triple quadrupole mass spectrometer (TSQ Quantum, Thermo Fisher, San Jose, CA). 5. XCaliber acquisition and Quan Browser software (Thermo Fisher, San Jose, CA).

74

Fonteh and Fisher

2.1.2. Equipment for Proteomics

1. Heat block at 57°C. 2. Aluminum foil. 3. 5 kDa Vivaspin tubes (Millipore, Bedford, MA). 4. Centrifuge (Maximum force: ~12,000 × g). 5. Water bath maintained at 37°C. 6. Vacuum centrifuge. 7. HPLC systems. For ion trap studies, we use a Surveyor system fitted with a pump and autosampler (Thermo Fisher, San Jose, CA). For quantitative proteomic studies, we use an HP 1100 system (Hewlett Packard, Foster City, CA) fitted with a Binary pump, autosampler and column heater (40°C). 8. HPLC columns. For ion trap protein sequencing studies, we use a Luna C18 column (1 mm × 250 mm) from Phenomenex (Torrance, CA). For quantitative proteomic studies, we use a hydrophilic interaction liquid chromatography column (TSK-Gel Amide-80, 2 × 150 mm from Tosoh Biosciences, Montgomeryville, PA). 9. Ion trap mass spectrometer (LCQ, Thermo Fisher, San Jose, CA). 10. Triple quadrupole mass spectrometer (TSQ Quantum, Thermo Fisher, San Jose, CA). 11. XCaliber acquisition and Quan Browser software (Thermo Fisher, San Jose, CA).

2.2. Reagents 2.2.1. Reagents for Lipidomics

1. Solvents, chloroform, methanol (VWR). Aqueous ammonium hydroxide (28%) and glacial formic acid were purchased from Sigma (St Louis, MO). Make a 9% aqueous solution of formic acid using double-distilled deionized water (VWR, West Chester, PA). 2. Phospholipid standards (phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS), phosphatidylinositol (PI), phosphatidic acid (PA), phosphatidylglycerol (PG) and internal standards (C11 PC, C11 PE, C11 PA)) from Avanti Polar Lipids (Alabaster, AL). Soy bean PI, ceramide (CM), sphingomyelin (SPM), 1-palmitoyl-2-sn-arachidonoyl glycerophoscholine, 1-palmitoyl-2-sn-docosahexaenoyl glycerophoscholine and 1-palmitoyl-2-sn-eicosatetraenoyl glycerophoscholine were purchased from Sigma. 3. HPLC solvents: Solvent A. CHCl3:CH3OH:NH4OH (80:19.5: 0.5, v/v/v); Solvent B: CHCl3:CH3OH:H2O:NH4OH (60:34: 5.5:0.5, v/v/v/v). HPLC sample buffer (Hexane:isopropanol:water, 40:50:3, v/v/v) 4. Make a solution (5 mg/mL) of butylated hydroxyl toluene (BHT) in methanol.



Combining Lipidomics and Proteomics of Human Cerebrospinal Fluids

2.2.2. Reagents for Proteomics

75

1. 100 mM ammonium bicarbonate, pH 8.0. 2. Rapigest™ [1% (w/v)] (Waters, Milford, MA) stock in ammonium bicarbonate. Rapigest™ is stable for 1 week at 2–8°C. 3. Urea (add salt directly to CSF or peptide standard solutions to make an 8 M solution). 4. Dithiothreitol (DTT, 100 mg/mL) in ammonium bicarbonate. 5. Iodoacetamide (30 mg/mL) in ammonium bicarbonate made fresh each time and protected from light. 6. Sequencing grade modified porcine trypsin (Princeton Separations, Aldelphia, NJ) in water, stored in a 4°C refrigerator or on ice until use. 7. Glacial formic acid. 8. HPLC solvents. Solvent A: H2O with 0.1% (v/v) formic acid and 5 mM ammonium bicarbonate. Solvent B: Acetonitrile with 0.1% (v/v) formic acid and 5 mM ammonium bicarbonate. 9. Isotope-labeled peptides used as internal standards synthesized with AQUA peptide synthesis (Dr. John Rush, Cell Signaling Technologies, Boston, MA).

3. Methods 3.1. Extraction of CSF Lipids

1. Add BHT solution to samples to a final concentration of 1 mg/mL (see Note 1). 2. Extract lipids using a modified method of Bligh and Dyer previously described in detail in Methods in Molecular Biology (50, 51). 3. Dry lipid extracts under a stream of N2 and resuspend in 100 mL HPLC sample buffer.

3.2. Detection and Analyses of CSF Lipids

1. Inject lipid extract onto the HPLC column maintained at 40°C. 2. Start with 50% HPLC solvent B for 14 min, then increase to 100% B over 10 min, maintain at 100% B for 15 min and then return to 50% B. 3. Equilibrate column for 15 min (5 times column volume) before next sample injection. 4. Directly interphase LC eluent to the electrospray ionization (ESI) source of the triple quadrupole mass spectrometer (TSQ Quantum, Thermo Fisher, San Jose, CA). 5. For positive ions, tune the mass spectrometer for parent ion scanning of 184.14, 141.32, 264.27 for PC, PE, CM, and SPM, respectively. Tune for neutral ion lose of 86.86 for PS (see Note 2).

76

Fonteh and Fisher

6. For negative ions, tune the mass spectrometer for parent ion scanning of 240.961, 196.31, 171.14, 168.17, and 153.05/96.99 for PI, PE, PG, SPM and all glycerophospholipids/phosphate, respectively. Tune the mass spectrometer for neutral ion loss of 86.86 and 50.1 for PS and PC, respectively (see Note 2). 7. Tune instrument for collision energy (CE) and collisioninduced dissociation (CID) for each phospholipid class and program a segment for each phospholipid using the XCalibur acquisition software (Thermo Fisher, San Jose, CA). 8. Acquire data and process using Quan Browser software (Thermo Fisher, San Jose, CA) (see Note 3). 3.3. ShotgunSequencing of CSF Proteins

1. Add 45 mL of CSF into 1.7 mL microtube and vortex briefly (see Note 4) 2. Add 5 mL of Rapigest™ stock a sample (see Note 5) 3. Add 0.67 mL of DTT stock a sample and vortex briefly 4. Briefly centrifuge (30 s, 1,000 × g) and incubate samples on a heat block at 56°C for 30 min 5. Add 2.67 mL of freshly made iodoacetamide stock to each sample and centrifuge briefly (30 s, 1,000 × g). Protect samples from light by wrapping with aluminum foil 6. Leave at room temperature in the dark for 30 min 7. Clean-up excess iodoactamide by adding 2.67 mL of DTT, vortex briefly and centrifuge for 30 s 8. Leave at room temperature for 30 min 9. Remove excess denaturing solution by adding the entire solution to a 500 mL Vivaspin (5 kDa cutoff) tube. Centrifuge at 8,000 x g until all solution has passed through the membrane filter 10. Add 50–100 mL of ammonium bicarbonate to the Vivaspin tube and centrifuge at 8,000 x g until all solution has passed through the membrane filter 11. For on filter digestion, add 100 mL of ammonium bicarbonate to the CSF concentrate and add 5 mL of trypsin stock (peptide/protein to trypsin ratio should be ~50:1) (see Note 6). 12. Incubate in a water bath overnight at 37°C 13. Stop all reactions by adding 5 mL of glacial formic acid 14. Centrifuge on-filter digest at 8,000 x g until all solution has passed through the membrane filter and use the filtrate for LC/MS2 studies



Combining Lipidomics and Proteomics of Human Cerebrospinal Fluids

77

3.4. Liquid Chromatography of Peptides

1. Start with 0% HPLC solvent B for 30 min (Surveyor MS pump), then increase to 60% B at 110 min, and then increase to 80% B at 114 min. Maintain at 80% B until 128 min and then return to 0% B at 132 min. Equilibrate column until 162 min before injecting another sample (see Note 7).

3.5. Mass Spectrometry of Peptides

1. Using an LCQ ion trap mass spectrometer (Thermo Fisher, San Jose, CA), set the acquisition time to 160 min with four different scan events. 2. Program the mass spectrometer for positive ions of four scan events. For scan event 1 (300.0–1,600.0); scan event 2 (Dependent MS/MS of most intense ion from scan event 1); scan event 3 (Dependent MS/MS of second most intense ion from scan event 1); scan event 4 (Dependent MS/MS of third most intense ion from scan event 1). Use default charge state of +2, isolation width of 2.0, normalized collision energy of 35.0, activation time of 30, and a minimum signal of 1,000 (see Note 8). For global data dependent acquisition, set exclusion mass widths to 1.0 and 1.5 for low and high masses, respectively.

3.6. Data Search and Analyses

1. Search MS spectra against a SwissProt database (release 7459 or later release) using the Sequest algorithm implementation of BioWorks 3.1 (Thermo Fisher, San Jose, CA) (52). 2. Use the Washburn criteria (53) for identification of CSF proteins and export protein list to an Excel spreadsheet. 3. Search the spreadsheet using key words for lipids to identify all lipid-metabolizing proteins.

3.7. Quantification of Lipid-Metabolizing Proteins in CSF

Several metabolic-labeling methods can be used to quantify changes in proteins in biological samples (54, 55). We favor isotope dilution tandem mass spectrometry for our quantitative proteomics (56). In this method, peptides identified from shotgun sequencing or mechanistically determined to be important in lipid metabolism are designed and synthesized with stable isotopes (see Note 8). Samples are spiked with known amounts (50 pmol) of each stable isotope standard and then processed as described below.

3.7.1. Denaturation and Tryptic Digestion of CSF Proteins for Isotope Dilution Studies

1. Add 45 mL of CSF into 1.7 mL microtube and vortex briefly. 2. Add 5 mL of Rapigest™ stock a sample containing 50 pmol stable isotope-labeled peptide internal standards (see Note 5). 3. Add 0.67 mL of DTT stock a sample and vortex briefly, and then centrifuge (30 s, 1,000 × g). 4. Leave in heat block at 56°C for 30 min.

78

Fonteh and Fisher

5. Add 2.67 mL of freshly made iodoacetamide mL to each sample. Wrap the tubes in aluminum foil to keep in the dark and mix using Vortexer, and centrifuge for 30 s 6. Leave at room temperature in the dark for 30 min. 7. Add 2.67 mL of DTT and vortex briefly, and centrifuge for 30 s. 8. Leave at room temperature for 30 min. 9. Add entire solution to 500 mL Vivaspin tube. Centrifuge at 8,000 x g until all solution has passed membrane. 10. Add 50 mL of ammonium bicarbonate to the 5 kDa Vivaspin tube and centrifuge at 8,000 x g until all solution has passed membrane. 11. Make a stock solutions containing 50 pmol of each stable isotope ([15N],[13C]) labeled internal standard a sample. Dry in vacuum centrifuge and resuspend in 50 mL of ammonium bicarbonate solution. 12. On-filter digestion. Add 100 mL of ammonium bicarbonate on the filter. Add stable isotope standards to filter and add 5 mL of trypsin stock solution. 13. Incubate in heat block overnight at 37°C. 14. On day 2, stop all reactions with 5 mL of formic acid. 15. Centrifuge on-filter digestion at 8,000 x g until all solution has passed membrane and use the filtrate for LC/MS2 studies. 3.7.2. Quantification of Peptides by LC/MS

1. Dry samples and resuspend in 85% acetonitrile with 15% H2O and 0.1% formic acid. 2. Condition the column at 100% solvent A. Run the samples at flow rate 0.3 mL/min (HP 1100 HPLC system) using the following sequence (%A/min): 100/0.0, 100/5.0, 0/37.5, 0/45.0, 100/47.5, 100/65.0. 3. Perform MS2 using a triple quadrupole mass spectrometer (TSQ-Quantum, Thermo Fisher, San Jose, CA). Isolate parent ions for each peptide and perform tandem MS to obtain the most abundant product ion. Verify products ions using Mass prospector (http://prospector.ucsf.edu/). 4. Optimize parent ion-product ion transitions for collision energy and tube lens voltage and perform multiple reaction monitoring (MRM) of unlabeled and stably labeled peptides of interest (see Note 8).

3.7.3. Data Analyses for Isotope Dilution LC/MS Studies

1. For isotope dilution studies, serially dilute (0–100 pmol) unlabeled peptides or protein standards if available. Recombinant proteins from several manufacturers can be used.



Combining Lipidomics and Proteomics of Human Cerebrospinal Fluids

79

2. Spike unlabeled peptides/proteins with 50 pmol [13C],[15N]labeled peptide. 3. Once MRM are performed, analyze data using Quan Browser of the XCalibur software (Thermo Fisher). 4. Also spike CSF samples with 50 pmol [13C],[15N]-labeled peptide standards, denature, digest, and subject to LC tandem mass spectrometry. 5. Determine amounts of proteins in biological samples using standard curves obtained by plotting concentrations of [12C]peptide against the ratio of peak intensity [12C]-peptide to peak intensity of [13C],[15N]-labeled peptide. 3.8. Results 3.8.1. Design and Synthesis of Stable Isotope Peptide Standards

1. Select peptides of interest from an extensive LTQ database of CSF peptides, or design peptides in silico using Peptidemass (http://www.expasy.org/tools/peptide-mass.html). UniProt Knowledgebase names of some lipid-metabolizing proteins (Table 1) of interest can be used to get tryptic peptides. An example of Peptidemass digest of apolipoprotein A1 (APOA1_HUMAN) is shown on Table 2. 2. The major criteria for peptide selection included size (500– 1,500 Da for m/z = 1, or 1,000–3,000 Da for m/z = 2) and the presence of amino acids that can be substituted with a stable isotopes ([13C9],[15N]-Phe, ([13C6],[15N]-Leu, [13C5],[15N]Val, [13C5],[15N]-Pro, [13C9],[15N]-Tyr). 3. Perform BLAST (http://www.expasy.ch/tools/blast/) on all peptides to confirm structure and specificity before AQUA peptide synthesis by Dr. John Rush (Cell Signaling Technologies, Boston, MA). 4. An advantage of isotope dilution studies is that post-translational modifications (PTMs) of proteins can be studied. For example, a peptide with a phosphorylation site can be synthesized with phosphorylated stable isotopes and used to quantify phosphorylation of that specific site. Similarly, variants can be studied by synthesizing peptide standards containing the modifications found in the variants. Several variants of APOA1_HUMAN (Table 2) can be studies by synthesizing peptides with the substitutions found in these variants.

3.9. Conclusions and Future Directions

Using LC/MS, we can catalog most lipid species in biological samples (1) (see Note 2–3). Shotgun sequencing allows us to determine the composition of proteins in CSF (52, 57). Several lipid-metabolizing proteins are detected in human CSF. These include proteins that bind and transport lipids, and proteins that synthesize or break down lipids. Once these proteins are detected in CSF, quantitative proteomics using isotope dilution methodologies can be applied to determine the levels of several lipid-metabolizing enzymes.

UniProt Knowledgebase name of human proteins APOA_HUMAN, APOA1_HUMAN, AIBP_HUMAN, A6XGK9_HUMAN, Q5T3I3_HUMAN, Q5T3I4_HUMAN, A6XGL0_HUMAN, COT2_HUMAN, APOA2_HUMAN, Q76EI7_HUMAN, APOA4_HUMAN,APOA5_HUMAN, B0YIV9_HUMAN, B0YIW1_HUMAN, APOB_HUMAN, AB48R_HUMAN, Q9NS13_HUMAN, ABEC1_HUMAN, B0QYD2_HUMAN, B0QYD3_HUMAN, B2CML4_HUMAN, ABC3F_HUMAN, B0QYD4_HUMAN, Q05JX5_HUMAN, ABC3H_HUMAN, B0QYP0_HUMAN, B0QYP1_HUMAN, ABEC4_HUMAN, APOC1_HUMAN, APOC2_HUMAN, APOC3_HUMAN, B0YIW2_HUMAN, A3KPE2_HUMAN, APOC4_HUMAN, A5YAK2_HUMAN, APOD_HUMAN, APOE_HUMAN, LRP1_HUMAN, LRP8_HUMAN, B1AMT6_HUMAN, B1AMT7_HUMAN, B1AMT8_HUMAN, APOF_HUMAN, APOH_HUMAN, CLUS_HUMAN, APOL1_HUMAN, A5PLQ4_HUMAN, APOL2_HUMAN, B0QYK7_HUMAN, APOL3_HUMAN, B1AHI4_HUMAN, B1AHI5_HUMAN, B1AHI6_HUMAN, B1AHI7_HUMAN, Q5U5N4_HUMAN, APOL4_HUMAN, B1AHJ3_HUMAN, Q6ICI8_HUMAN, Q9BRG6_HUMAN, APOL5_HUMAN, APOL6_HUMAN, APLD1_HUMAN, A0AVN6_HUMAN, APOM_HUMAN, Q5SRP5_HUMAN, Q5SRP4_HUMAN, B0UX98_HUMAN, APOO_HUMAN, APOOL_HUMAN, LPAL2_HUMAN, Q6ZU85_HUMAN AT8A1_HUMAN, AT8A2_HUMAN, AT8B1_HUMAN, AT8B2_HUMAN, AT11B_HUMAN, AT11C_HUMAN, AT11A_HUMAN, AT8B3_HUMAN, AT8B4_HUMAN, ATP9A_HUMAN, ATP9B_HUMAN, AT10A_HUMAN, AT10B_HUMAN, AT10D_HUMAN, AT93_HUMAN CETP_HUMAN, GLTD1_HUMAN, GLTD2_HUMAN, GLTP_HUMAN, LCAT_HUMAN, PLTP_HUMAN DGKA_HUMAN, DGKB_HUMAN, A4D117_HUMAN, A4D116_HUMAN, DGKD_HUMAN, DGKE_HUMAN, A1L4Q0_HUMAN, DGKG_HUMAN, Q2M1H4_HUMAN, DGKH_HUMAN, DGKI_HUMAN, DGKK_HUMAN, DGKQ_HUMAN, Q6P3W4_HUMAN, DGKZ_HUMAN, Q8IVW9_HUMAN FAS_HUMAN, SAST_HUMAN, ACBG1_HUMAN, ACBG2_HUMAN, ACSL1_HUMAN, ACSL3_HUMAN, ACSL4_HUMAN, ACSL5_HUMAN, ACSL6_HUMAN, S27A2_HUMAN, S27A5_HUMAN, A6GV77_HUMAN

Protein group

Apolipoproteins

Phospholipid transporting ATPase

Other lipid transfer proteins

Diacylgycerol kinases

Fatty Acid snythetases

Table 1  List of lipid-metabolizing proteins of interest from UniProt Knowledgebase (Swiss-Prot: http://www.expasy.ch/tools/)

80 Fonteh and Fisher

PLCB1_HUMAN, Q17RQ6_HUMAN, PLCB2_HUMAN, Q9BVT6_HUMAN, PLCB3_HUMAN, A5PKZ6_HUMAN, PLCB4_HUMAN, PLCD1_HUMAN, PLCD3_HUMAN, PLCD4_HUMAN, PLCE1_HUMAN, Q5VWL4_HUMAN, PLCG1_HUMAN, A2A284_HUMAN, PLCG2_HUMAN, PLCH1_HUMAN, PLCH2_HUMAN, Q86YW0_HUMAN, NSMA_HUMAN, PLCX1_HUMAN, PLCX2_HUMAN, PLCX3_HUMAN, PLCL2_HUMAN, PLCL1_HUMAN, SOCS7_HUMAN PLD1_HUMAN, PLD2_HUMAN, PLD3_HUMAN, PLD4_HUMAN, PLD5_HUMAN, ENPP2_HUMAN, PHLD_HUMAN, NAPEP_HUMAN, DDHD1_HUMAN, DDHD2_HUMAN, YQ008_HUMAN PGH1_HUMAN, PGH2_HUMAN, Q5T7T7_HUMAN, Q5T7T8_HUMAN, Q3HY30_HUMAN, Q3HY29_HUMAN, Q3HY28_HUMAN, Q6ZYK7_HUMAN, Q8IZA9_HUMAN, PD2R_HUMAN, PTGD2_HUMAN, PTGES_HUMAN, PGES2_HUMAN, A6NHH0_HUMAN, FEM1A_HUMAN, CBR1_HUMAN, PE2R1_HUMAN, PE2R2_HUMAN, PE2R3_HUMAN, PE2R4_HUMAN, Q6TTN3_HUMAN, Q5CZ60_HUMAN, Q5CZ62_HUMAN, Q5CZ57_HUMAN, O00325_HUMAN, TEBP_HUMAN, AK1C3_HUMAN, PF2R_HUMAN, FPRP_HUMAN, Q6RYQ6_HUMAN, PGDH_HUMAN, PGH1_HUMAN, PGH2_HUMAN, Q5T7T7_HUMAN, Q5T7T8_HUMAN, PTGDS_HUMAN, PTGIS_HUMAN, PI2R_HUMAN, SO2A1_HUMAN LOXE3_HUMAN, LOX5_HUMAN, AL5AP_HUMAN, LOXP_HUMAN, LX12B_HUMAN, LOX1_HUMAN, LX15B_HUMAN, LOXH1_HUMAN

Phospholipase C

Phospholipase D

Prostaglandin enzymes

Leukotrienes

Protein groups identified in human CSF by shotgun sequencing were searched using the UniProt Knowledgebase (UPK). All human proteins in the UPK were listed. Each name can be used for the in silico design of tryptic peptides (http://www.expasy.ch/tools/peptide-mass.html). These peptides can be used for quantitative proteomics using isotope dilution MS. Other protein groups not included in this list include proteins involved with sphingomyelin/ceramide metabolism and the cytochrome P450 enzymes involved in cholesterol and polyunsaturated fatty acid metabolism.

PA21B_HUMAN, PA2GA_HUMAN, PA2GC_HUMAN, PA2GD_HUMAN, B1AEL9_HUMAN, PA2GE_HUMAN, PA2GF_HUMAN, PA2G3_HUMAN, PA24A_HUMAN, B1AKG4_HUMAN, Q14064_HUMAN, PA24B_HUMAN, Q19KD6_HUMAN, Q19KD5_HUMAN, PA24C_HUMAN, PA24D_HUMAN, PA24E_HUMAN, PA24F_HUMAN, A6H8K5_HUMAN, A5PKZ7_HUMAN, PA2G5_HUMAN, PLPL2_HUMAN, PLPL8_HUMAN, ADPN_HUMAN, PRDX6_HUMAN, B0QYE8_HUMAN, PA2G6_HUMAN, PA2GX_HUMAN, Q14DU3_HUMAN, PG12A_HUMAN, PG12B_HUMAN, PAFA_HUMAN, PAFA2_HUMAN, Q6DN23_HUMAN, OC90_HUMAN, PLB1_HUMAN, PLA2R_HUMAN, B2RTU9_HUMAN, PLAP_HUMAN, ANXA1_HUMAN, LYPA2_HUMAN, Q5QPN9_HUMAN, Q5QPQ0_HUMAN, Q5QPQ2_HUMAN, Q5QPQ3_HUMAN, LYPA3_HUMAN

Phospholipase A2

Combining Lipidomics and Proteomics of Human Cerebrospinal Fluids 81

82

Fonteh and Fisher

Table 2 Peptidemass digest of apolipoprotein A1 (APOA1_HUMAN) Position

Peptide sequence

M/z = 1

m/z = 2

Variants

25–34

DEPPQSPWDR

1,226.5436

613.775

27:P->H(in Munster-3C). 27: 28:P->R(in Munster-3B). 34:

37–47

DLATVYVDVLK

1,235.6881

618.348

52–64

DYVSQFEGSALGK

1,400.6692

700.838

65–69

QLNLK

615.3824

308.195

70–83

LLDNWDSVTSTFSK

1,612.7853

806.896

86–101

EQLGPVTQEFWDNLEK

1,932.9337

966.971

102–107

ETEGLR

704.3573

352.682

108–112

QEMSK

622.2865

311.647

113–118

DLEEVK

732.3774

366.692

113:

121–130

VQPYLDDFQK

1,252.6208

626.814

126: 127:

132–140

WQEEMELYR

1,283.5725

642.29

132: 134:E->K(in Fukuoka).

143–147

VEPLR

613.3668

307.187

148–155

AELQEGAR

873.4424

437.225

158–164

LHELQEK

896.4836

448.745

160:E->K(in Norway). 163:

165–173

LSPLGEEMR

1,031.519

516.263

167: 168:L->R(in Zaragoza). 171:

178–184

AHVDALR

781.4315

391.219

180: 184:R->P(indbSNP

185–195

THLAPYSDELR

1,301.6484

651.328

189:P->R.

202–206

LEALK

573.3606

287.184

207–212

ENGGAR

603.2845

302.146

213–219

LAEYHAK

831.4359

416.222

220–230

ATEHLSTLSEK

1,215.6215

608.314

231–239

AKPALEDLR

1,012.5785

506.793

240–250

QGLLPVLESFK

1,230.7092

615.858

251–262

VSFLSALEEYTK

1,386.7151

693.861

61:A->T

92:

222:

APOA1_HUMAN was subjected to tryptic digestion using Peptidemass (http://www.expasy.org/ tools/peptide-mass.html). The number of missed cleavages was set at 0 and all cysteines were treated with iodoacetamide to form carbamidomethyl-cysteine. Methionines were oxidized to methionine sufoxide and only peptide masses >500 Da are displayed. Monoisotopic masses ([M + H+], m/z = 1) and the masses with a charge state (m/z) of 2 (default for the ion trap mass spectrometer) are shown. All known variants are also shown.



Combining Lipidomics and Proteomics of Human Cerebrospinal Fluids

83

This approach allows for the complete understanding of biological pathways and should be useful in detecting changes in protein expression for brain diseases. Any change in lipids in combination with proteins should better illuminate strategies for biomarker discovery or molecular targets for drugs.

4. Notes 1. Lipids with polyunsaturated fatty acids are susceptible to oxidation. Enzymes at room temperature may also rapidly degrade some lipid mediators and signaling lipids. It is important to protect lipids from oxidation by adding BHT and by storing samples under inert conditions. We flush air out of our sample tubes before storage at −80°C. Once lipids are extracted, they are similarly stored in organic solvents until required for LC/MS2. In cases where specific enzyme inhibitors are known, these can be incorporated into the samples to avoid degradation of lipids. 2. While ESI is normally used for lipids, others have reported efficient use of other ionization methods. These include chemical ionization or photo ionization under atmospheric conditions (58). We have used the triple quadrupole mass spectrometer for most of our studies. However, iontrap mass spectrometers and high resolution linear FT/MS have also been used for lipidomic studies (59). 3. Several databases can be used to search mass spectrometer data of lipids. So far, these are proprietary and not universally available. These may also be instrument- and/or method-specific especially when attempts are made to identify molecular species of lipids. With accurate mass detection and increase in sensitivity of instruments, it may be possible to link individual lipidomic studies with the lipid maps database (http://www. lipidmaps.org/) for identification of lipid molecular species. 4. High-abundant proteins may be removed from samples to increase the dynamic range. This allows detection of the less abundant proteins in CSF. 5. Sample preparation is a key step in shotgun sequencing of proteins. For denaturation, urea or guanidine HCl can be used in place of Rapigest™. However, these have to be removed before mass spectrometry since these compounds may reduce the ionization of peptides of interest. 6. Other proteolytic enzymes and other buffer conditions can also be used instead of, or in combination, with trypsin. Such combinations may increase sequence coverage and increase the number of proteins detected in a sequencing experiment (52).

84

Fonteh and Fisher

7. 2-Dimensional liquid chromatography can be used to improve the separation and detection of peptides. Normally, the first dimension is a cation exchange resin that is followed by a reverse phase column. The Pepfinder kit (ThermoFisher, San Jose, CA) has been effectively used to improve extraction of peptides, removal of salts, and thus enables detection of thousands of proteins in human CSF without prior removal of abundant proteins (52). 8. The default charge state of the ion trap mass spectrometer is 2. Ions of charge state 1 or 2 may appear when a full scan is performed on the triple quadrupole mass spectrometer. For some peptides, the major charge state may be 1 or it may be 2 for others. We use the more intense ion for MS/MS and MRM studies.

Acknowlegments We thank Susan Onami and Elena Oborina for technical help. We appreciate the support and critical discussion with Drs. Michael Harrington, Andreas Hulmer, and Roger Biringer. This work was supported in part by NIH grants RO1# NS43295; institutional support (HMRI), the Norris, Jamison and Glide Foundations and donations from the Dunlevey, Hezlep and Posthuma families. References 1. Fonteh, A. N., Harrington0, R. J., Huhmer, A. F., Biringer, R. G., Riggins, J. N., and Harrington, M. G. (2006) Identification of disease markers in human cerebrospinal fluid using lipidomic and proteomic methods. Dis Markers 22(1–2), 247–272. 2. Albers, J. J. (1978) Effect of human plasma apolipoproteins on the activity of purified lecithin:cholesterol acyltransferase. Scand J Clin Lab Invest Suppl 150, 48–52. 3. Beffert, U., Danik, M., Krzywkowski, P., Ramassamy, C., Berrada, F., and Poirier, J. (1998) The neurobiology of apolipoproteins and their receptors in the CNS and Alzheimer’s disease. Brain Res Brain Res Rev 27(2), 119–142. 4. Haffner, S., Applebaum-Bowden, D., Wahl, P. W.et al. (1985) Epidemiological correlates of high density lipoprotein subfractions, apolipoproteins A-I, A-II, and D, and lecithin cholesterol acyltransferase. Effects of smoking, alcohol, and adiposity. Arteriosclerosis 5(2), 169–177.

5. Bazan, N. G., and Allan, G. (1996) Plateletactivating factor is both a modulator of synaptic function and a mediator of cerebral injury and inflammation. Adv Neurol 71, 475–482. 6. Brady, H. R., and Serhan, C. N. (1996) Lipoxins: putative braking signals in host defense, inflammation and hypersensitivity. Curr Opin Nephrol Hypertens 5(1), 20–27. 7. Fonteh,vv A. N., and Harrington, M. G. (2004) Remodeling of arachidonate and other polyunsaturated fatty acids in Alzheimer’s disease. In: Fonteh AN, Wykle RL, editors. Arachidonate Remodeling and Inflammation. Birkhauser Verlag 145–168. 8. Ford-Hutchinson, A. W. (1990) Leukotriene B4 in inflammation. Crit Rev Immunol 10(1), 1–12. 9. Lefkowith, J. B. (1988) Essential fatty acid deficiency inhibits the in  vivo generation of leukotriene B4 and suppresses levels of resident and elicited leukocytes in acute inflammation. J Immunol 140(1), 228–233.



Combining Lipidomics and Proteomics of Human Cerebrospinal Fluids

10. O’Banion, M. K. (1999) COX-2 and Alzheimer’s disease: potential roles in inflammation and neurodegeneration. Expert Opin Investig Drugs 8(10), 1521–1536. 11. Peroutka, S. J. (2005) Neurogenic inflammation and migraine: implications for the therapeutics. Mol Interv 5(5), 304–311. 12. Pruzanski, W., and Vadas, P. (1989) Phospholipase A2 and inflammation. Ann Rheum Dis 48(11), 962–963. 13. Serhan, C. N. (1996) Inflammation. Signalling the fat controller. Nature 384(6604), 23–24. 14. Zurier, R. B. (1993) Fatty acids, inflammation and immune responses. Prostaglandins Leukot Essent Fatty Acids 48(1), 57–62. 15. Axelrod, J. (1995) Phospholipase A2 and G proteins. Trends Neurosci 18(2), 64–65. 16. Bazan, N. G., Allan, G., and Rodriguez de Turco, E. B. (1993) Role of phospholipase A2 and membrane-derived lipid second messengers in membrane function and transcriptional activation of genes: implications in cerebral ischemia and neuronal excitability. Prog Brain Res  96, 247–257. 17. Bonventre, J. V. (1992) Phospholipase A2 and signal transduction. J Am Soc Nephrol 3(2), 128–150. 18. Clark, J. D., Schievella, A. R., Nalefski, E. A., and Lin, L. L. (1995) Cytosolic phospholipase A2. J Lipid Mediat Cell Signal 12(2–3), 83–117. 19. Dennis, E. A. (2000) Phospholipase A2 in eicosanoid generation. Am J Respir Crit Care Med 161(2 Pt 2), S32–S35. 20. Diez, E., Chilton, F. H., Stroup, G., Mayer, R. J., Winkler, J. D., and Fonteh, A. N. (1994) Fatty acid and phospholipid selectivity of different phospholipase A2 enzymes studied by using a mammalian membrane as substrate. Biochem J 301(Pt 3), 721–726. 21. Farooqui, A. A., Antony, P., Ong, W. Y., Horrocks, L. A., and Freysz, L. (2004) Retinoic acid-mediated phospholipase A2 signaling in the nucleus. Brain Res Brain Res Rev 45(3), 179–195. 22. Fonteh, A. N., Bass, D. A., Marshall, L. A., Seeds, M., Samet, J. M., and Chilton, F. H. (1994) Evidence that secretory phospholipase A2 plays a role in arachidonic acid release and eicosanoid biosynthesis by mast cells. J Immunol 152(11), 5438–5446. 23 Kramer, R. M., Stephenson, D. T., Roberts, E.F., and Clemens, J. A. (1996) Cytosolic phospholipase A2 (cPLA2) and lipid mediator release in the brain. J Lipid Mediat Cell Signal 14(1–3), 3–7.

85

24. Kudo, I., and Murakami, M. (2002) Phospholipase A2 enzymes. Prostaglandins Other Lipid Mediat 68–69, 3–58. 25. Cocco, L., Martelli, A. M., Gilmour, R. S., Rhee, S. G., and Manzoli, F. A. (2001) Nuclear phospholipase C and signaling. Biochim Biophys Acta 1530(1), 1–14. 26. Cockcroft, S. (2006) The latest phospholipase C, PLCeta, is implicated in neuronal function. Trends Biochem Sci 31(1), 4–7. 27. Manzoli, L., Billi, A. M., Martelli, A. M., and Cocco, L. (2004) Regulation of nuclear phospholipase C activity. Acta Biochim Pol 51(2), 391–395. 28. Matsushima, H., Shimohama, S., Kawamata, J., Fujimoto, S., Takenawa, T., and Kimura, J. (1998) Reduction of platelet phospholipase C-delta1 activity in Alzheimer’s disease associated with a specific apolipoprotein E genotype (epsilon3/epsilon3). Int J Mol Med 1(1), 91–93. 29. Minke, B. (2001) The TRP channel and phospholipase C-mediated signaling. Cell Mol Neurobiol 21(6), 629–643. 30. Nakahara, M., Shimozawa, M., Nakamura, Y., et  al. (2005) A novel phospholipase C, PLC(eta)2, is a neuron-specific isozyme. J Biol Chem 280(32), 29128–29134. 31. Rhee, S. G. (2001) Regulation of phosphoinositide-specific phospholipase C. Annu Rev Biochem 70, 281–312. 32. Shimohama, S., Sasaki, Y., Fujimoto, S.et  al. (1998) Phospholipase C isozymes in the human brain and their changes in Alzheimer’s disease. Neuroscience 82(4), 999–1007. 33. Stewart, A. J., Mukherjee, J., Roberts, S. J., Lester, D., and Farquharson, C. (2005) Identification of a novel class of mammalian phosphoinositol-specific phospholipase C enzymes. Int J Mol Med 15(1), 117–121. 34. Cockcroft, S. (2001) Signalling roles of mammalian phospholipase D1 and D2. Cell Mol Life Sci 58(11), 1674–1687. 35. Jenkins, G. M., and Frohman, M. A. (2005) Phospholipase D: a lipid centric review. Cell Mol Life Sci 62(19–20), 2305–2316. 36. Klein, J. (2005) Functions and pathophysiological roles of phospholipase D in the brain. J Neurochem 94(6), 1473–1487. 37. Xie, Z., Ho, W. T., Spellman, R., Cai, S., and Exton, J. H. (2002) Mechanisms of regulation of phospholipase D1 and D2 by the heterotrimeric G proteins G13 and Gq. J Biol Chem 277(14), 11979–11986. 38. Balazy, M. (2004) Eicosanomics: targeted lipidomics of eicosanoids in biological systems. Prostaglandins Other Lipid Mediat 73(3–4), 173–180.

86

Fonteh and Fisher

39. Fitzpatrick, F. A., and Murphy, R. C. (1988) Cytochrome P-450 metabolism of arachidonic acid: formation and biological actions of “epoxygenase”-derived eicosanoids. Pharmacol Rev  40(4), 229–241. 40. Lands, W. E. (1993) Eicosanoids and health. Ann N Y Acad Sci 676, 46–59. 41. Murphy, R. C. (2001) Free-radical-induced oxidation of arachidonoyl plasmalogen phospholipids: antioxidant mechanism and precursor pathway for bioactive eicosanoids. Chem Res Toxicol 14(5), 463–472. 42. Serhan, C. N., Lu, Y., Hong, S., and Yang, R. (2007) Mediator lipidomics: search algorithms for eicosanoids, resolvins, and protectins. Methods Enzymol 432, 275–317. 43. Smith, W. L. (1989) The eicosanoids and their biochemical mechanisms of action. Biochem J 259(2), 315–324. 44. Bazan, N. G., Squinto, S. P., Braquet, P., Panetta, T., and Marcheselli, V. L. (1991) Platelet-activating factor and polyunsaturated fatty acids in cerebral ischemia or convulsions: intracellular PAF-binding sites and activation of a fos/jun/AP-1 transcriptional signaling system. Lipids 26(12), 1236–1242. 45. Benveniste, J., Chignard, M., Le Couedic, J. P., and Vargaftig, B. B. (1982) Biosynthesis of platelet-activating factor (PAF-ACETHER). II. Involvement of phospholipase A2 in the formation of PAF-ACETHER and lyso-PAFACETHER from rabbit platelets. Thromb Res 25(5), 375–385. 46. Snyder, F. (1994) Metabolic processing of PAF. Clin Rev Allergy 12(4), 309–327. 47. Farooqui, A. A., Litsky, M. L., Farooqui, T., and Horrocks, L. A. (1999) Inhibitors of intracellular phospholipase A2 activity: their neurochemical effects and therapeutical importance for neurological disorders. Brain Res Bull 49(3), 139–153. 48. Berenbaum, F. (1995) Phospholipase A2 inhibitors: a challenge for the future. Rev Rhum Engl Ed 62(6), 409–414. 49. Glaser, K. B., Lock, Y. W., and Chang, J. Y. (1991) PAF and LTB4 biosynthesis in the human neutrophil: effects of putative inhibitors of phospholipase A2 and specific inhibitors of 5- lipoxygenase. Agents Actions 34(1–2), 89–92. 50. Bligh, E. A., and Dyer, W. T. (1959) A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37, 911–917.

51. Fonteh, A. N. (1999) Assessment of arachidonic acid distribution into phospholipids of inflammatory cells. Methods Mol Biol 120, 77–89. 52. Biringer, R. G., Amato, H., Harrington, M. G., Fonteh, A. N., Riggins, J. N., and Huhmer, A. F. (2006) Enhanced sequence coverage of proteins in human cerebrospinal fluid using multiple enzymatic digestion and linear ion trap LC-MS/MS. Brief Funct Genomic Proteomic 5(2), 144–153. 53. Wolters, D. A., Washburn, M. P., and Yates, III J.R. (2001) An automated multidimensional protein identification technology for shotgun proteomics. Anal Chem 73(23), 5683–5690. 54. Amanchy, R., Kalume, D. E., and Pandey, A. (2005) Stable isotope labeling with amino acids in cell culture (SILAC) for studying dynamics of protein abundance and posttranslational modifications. Sci STKE 267, l2. 55. Romijn, E. P., Christis, C., Wieffer, M. et al. (2005) Expression clustering reveals detailed co-expression patterns of functionally related proteins during B cell differentiation: a proteomic study using a combination of onedimensional gel electrophoresis, LC-MS/MS, and stable isotope labeling by amino acids in cell culture (SILAC). Mol Cell Proteomics 4(9), 1297–1310. 56. Fonteh, A. N., Biringer, R. G., Huhmer, A. F., Rush, J. D., and Harrington, M. G. (2005) Use of [13C],[15N]-peptide standards to quantify enzymes of the Cyclooxygenase and Lipoxygenase pathways in human cerebrospinal fluids. American Society for Mass Spectrometry: Conference Proceeding 5–6(http://www. asms.org/asms05pdf/A052054.pdf). 57. Huhmer, A. F., Biringer, R. G., Amato, H., Fonteh, A. N., and Harrington, M. G. (2006) Protein analysis in human cerebrospinal fluid: Physiological aspects, current progress and future challenges. Dis Markers 22(1–2), 211–234. 58. Lee, S. H., and Blair, I. A. (2007) Targeted chiral lipidomics analysis by liquid chromatography electron capture atmospheric pressure chemical ionization mass spectrometry (LC-ECAPCI/ MS). Methods Enzymol 433, 159–174. 59. Schwudke, D., Liebisch, G., Herzog, R., Schmitz, G., and Shevchenko, A. (2007) Shotgun lipidomics by tandem mass spectrometry under data-dependent acquisition control. Methods Enzymol 433, 175–191.

Part II Analytical Approaches

sdfsdf

Chapter 5 Lipid Profiling Using Two-Dimensional Heteronuclear Single Quantum Coherence NMR Engy A. Mahrous, Robin B. Lee, and Richard E. Lee Summary  The use of NMR spectroscopy in lipid research has been traditionally reserved for the analysis and structural elucidation of discrete lipid molecules. Although NMR analysis of organic molecules provides a plethora of structural information that is normally unattainable by most other techniques, its use for global analysis of mixed lipid pools has been hampered by its relatively low sensitivity and overlapping of signals in the spectrum. However, the last few decades have witnessed great advancements in NMR spectroscopy that generally resulted in greater sensitivity and offered more flexibility in sampling techniques. The method discussed in this chapter describes the use of NMR for global lipidome analysis. This methodology benefits from the quantitative nature of this technique together with the abundance of the structural information it can offer, while partially overcoming the problems of low sensitivity and overlapping signals through isotope-enrichment and the use of multidimensional NMR, respectively. We have applied this method successfully to the mycobacterial lipidome as an example of an organism with a very complex and chemically diverse lipid pool. The same concept is applicable to a wide range of prokaryotes that can grow in the laboratory in well-defined growth media. Key words:  NMR spectroscopy, Mycobacteria, 2D-HSQC, Cell wall lipids, Polyketides, Glycolipids

1. Introduction The genus Mycobacteria is characterized by a unique cell wall rich in complex lipids, polyketides, and polysaccharides. Historically, numerous studies of individual mycobacterial lipids and the way they participate in pathogenesis and virulence have been conducted (1–3). Early attempts for global analysis of the mycobacterial lipidome began in the 1960s as a search for a taxonomical tool to assign newly discovered species in their right taxonomical order and to help identify clinical isolates at Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, DOI 10.1007/978-1-60761-322-0_5, © Humana Press, a part of Springer Science + Business Media, LLC 2009

89

90

Mahrous, Lee, and Lee

the species level (4–7). A major breakthrough in this field was achieved by Dobson and Minnikin who reported five 2D-TLC systems in the 1980s to analyze the mycobacterial lipidome (8). Despite its wide application the methodology remained less than satisfactory. Problems associated with poor staining, low sensitivity, and limited resolution on one hand and the recent advancement of bioanalytical tools on the other, have fueled a research effort for new methodologies to study the mycobacterial lipidome. Recently, two new methodologies based on mass spectroscopy have been described. The first uses Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) (9). The second uses prior separation with HPLC followed by analysis with electrospray ionization HPLC/ ESI-MS (10). These analytical techniques are highly sensitive, however, they require significant sample manipulation to overcome the phenomena of ion suppression within lipid mixtures thus are best used for the analysis of readily ionizable molecules (9, 10). The method discussed in this chapter focuses on the application of NMR spectroscopy as a complementary procedure that can be utilized to rapidly analyze the composition of the mycobacterial lipid pool directly from crude lipid extracts (11). It offers a great tool for investigating the mode of action of antituberculosis drugs, determining gene function, understanding the stringent response of Mycobacteria under stress conditions that resemble the in  vivo environment, and investigating differences between species or strains of Mycobacteria. To maximize its benefit, this lipid-profiling technique can be used in a complementary fashion with other genomic and proteomic methods. As it is the case with other “omics” techniques, the great utility of lipidomics methods is that they provide a wealth of information regarding the total composition of the lipid pools, from which a small number of candidate molecules can then be defined for more comprehensive follow-up studies through purification and isolation of such molecules of interest. For introduction of this method, we applied the technique to the lipid pool of Mycobacterium tuberculosis, the etiological agent of tuberculosis, which currently infects one-third of the world population and claimed 1.7 million lives worldwide in 2006 (12). The cell wall of M. tuberculosis, with its associated lipids, is believed to play an important role in its pathogenesis and virulence. The currently accepted structural and functional model of the mycobacterial cell wall describes it as a three-layer structure (13). The innermost layer consists of the plasma membrane, followed to the outside by the peptidoglycan layer, which is then covalently attached to the outer mycolyl arabinogalactan complex. The unusual long chain mycolic acids form a ­permeability barrier where a mixture of lipids, proteins, and glycolipids are loosely bound to the outside of the cell wall. Some of these molecules are surface



Lipid Profiling Using Two-Dimensional Heteronuclear Single Quantum

91

exposed and are free to interact with the host immune system (14–16), which may contribute to virulence and pathogenicity. This population of lipids can be easily extracted and analyzed by NMR spectroscopy. As NMR measures properties of nuclei rather than the physical property of the entire molecule, NMR signals are less likely to be affected by the different physicochemical properties exhibited by different classes of lipids (molecular weight, polarity, ionization potential, etc.). This uniformity of response is advantageous because it limits artifacts and confers quantitative representation of all lipids regardless of their structure. However, several challenges exist for using NMR to study complex pools of metabolites. First is the low sensitivity associated with NMR analysis. This can be partially overcome by enriching the cell lipidome with 13 C isotope through the replacement of the carbon source in the growth media with 13C-labeled precursors. Second, is the complexity and frequent overlapping of the NMR signals in the onedimensional spectrum. This problem can be addressed through the use of two- and three-dimensional NMR. Two-dimensional 1 H-13C heteronuclear single quantum coherence (HSQC) NMR can be used to rapidly generate lipid profiles, while more sophisticated three-dimensional experiments can be used to resolve the lipid profile and help assign signals when necessary. These experiments require no specialized equipment and can be performed using a standard high-field NMR spectrometer. Third, data analysis of such complex data requires extra consideration and efforts. To ensure data quality, signals in the NMR need to be referenced in both their chemical shift and relative intensity to an internal reference that produces a discrete signal in the NMR spectrum. To enhance reproducibility, external references are required to calibrate the NMR magnet and account for any observed variation due to instability of the magnet overtime. The data must then be normalized based on these internal and external standards using NMR software that confers robust analysis of large data sets. Specific NMR acquisition and data processing methods will not be discussed here as they are highly dependent on the instrumentation and software available. The following sections describe methods for sample preparation and analysis of mycobacterial lipid profiles using 2D NMR spectroscopy.

2. Materials 2.1. Equipment

1. High-field nuclear magnetic resonance spectrometer. 2. NMR processing software.

92

Mahrous, Lee, and Lee

2.2. Reagents and Supplies 2.2.1. Cell Culture

1. Sterile solution of Difco Middlebrook 7H9 Becton Dickinson and company, Sparks, MD. 2. Sterile 10× ADG supplement (2% glucose, 2% glycerol, 0.8% sodium chloride, 5% Bovine serum albumin (BSA) fraction V, Calbiochem, La Jolla, CA). 3. Sterile 10× solution of 5% Bovine serum albumin/0.8% sodium chloride. 4. Sterile solution of 10% 13C6 glucose Cambridge Isotope Laboratories, Andover, MA. 5. Sterile solution of 10% 13C3 glycerol Cambridge Isotope Laboratories, Andover, MA. 6. Sterile solution of 10% Tween-80, Calbiochem, La Jolla, CA.

2.2.2. Lipid Extraction

1. Deuterium oxide D2O 99% Cambridge Isotope Laboratories, Andover, MA. 2. Deuterated chloroform CDCl3 99% Cambridge Isotope Laboratories, Andover, MA. 3. Deuterated methanol CD3OD Cambridge Isotope Laboratories, Andover, MA.

2.2.3. Mycolic Acid Extraction

1. 1 M methanolic solution of KOH. 2. Diethyl ether. 3. Concentrated HCl.

3. Methods 3.1. Cell Culture and Preparation of Cell Pellet

To generate a good 2D HSQC spectrum in a short acquisition time, it is necessary to enrich the cell lipidome with 13C isotope. This can be achieved through the replacement of the carbon source in the growth media with 13C-labeled lipid precursors such as U-13C2 acetate, U-13C6 glucose, or U-13C3 glycerol. To attain a well-enriched sample, it is best to use a defined medium and limit any carbon sources to 13C as much as possible. Culture can also be passed in 13 C medium to further increase enrichment. Although the procedure discussed here focuses mainly on profiling the overall free cell wall lipid populations, the method can be easily modified to target a specific lipid population. The global 13C-enrichement strategy using 13C glucose and glycerol can be replaced by a biosynthetically guided approach that uses specific 13C-labeled precursors in order to enrich the signals for certain lipid targets. Cells can be grown or treated under various conditions to probe the response of the bacteria to different environmental conditions such as



Lipid Profiling Using Two-Dimensional Heteronuclear Single Quantum

93

nutrient starvation, drug treatment, pH, oxidative stress, or media composition. The time 13C label is incorporated can be varied based on the conditions tested. For example, studies involving the response of bacteria to drug treatment may use either cells fully labeled prior to drug treatment to observe changes in the complete lipidome or the 13C precursors may be added at the time of drug treatment to observe incorporation of label, and hence determine which lipids are synthesized in the presence of the drug. After cells have been grown under the desired conditions, the cell pellet must be washed with D2O, prior to extraction of the lipids, to ensure the removal of water and media components that may interfere with signals in the NMR spectrum. The methods below describe the standard protocol for the preparation of a mycobacterial sample (see Note 1). Generally, a 20–50 ml culture grown to an OD600 of 0.4–0.6 is sufficient to generate a good lipid profile. 1. 50 ml of 13C-labeled Middlebrook 7H9 media is inoculated from a midlog culture to an OD600 = 0.05 for slow growing species or OD600 = 0.005 for fast growing species (see Notes 1–3). 2. The cells are incubated at 37°C with continuous shaking at 250 rpm until an OD600 of 0.4–0.6 is reached. The cells are harvested by centrifugation at 3,700 × g for 10 min. 3. The supernatant is carefully discarded and the cells are washed twice with a volume D2O equal to 20% of the original culture volume (10 ml for a 50 ml culture). Centrifuge at 3,700 × g for 10 min and discard the supernatant. 4. Use 1 ml of D2O to resuspend the pellet and transfer the sample to a preweighed microcentrifuge tube. Centrifuge the cells at 3,700 × g for 10 min. 5. Discard the supernatant and carefully remove all remaining D2O from the cell pellet. Determine the weight of the wet pellet. 3.2. Extraction and Analysis of the Total Free Lipid Pool

In these experiments, the use of deuterated solvents is necessary to suppress solvent peaks and optimize the signal to noise ratio. The lipid pool of mycobacterial cells are extracted directly from the washed cell pellet by a 2:1 mixture of CDCl3:CD3OD, a solvent system commonly used to extract surface lipids of mycobacteria. Internal standards are incorporated into the extraction mixture to facilitate NMR data processing and allow normalization of data collected at different times. Ideally, compounds are selected that provide two sets of discrete internal reference peaks located in an uncongested area of the spectrum. For mycobacterial lipid profiling, dimethyl sulfoxide (DMSO) and/or 1,3,5 trimethyl benzene can be added to the CDCl3:CD3OD mixture prior to extraction. A lipid profile is generated directly from the

94

Mahrous, Lee, and Lee

crude extract using a 2D-HSQC pulse sequence to resolve the signals into two dimensions (the F1 proton dimension and the F2 carbon dimension). This inverse detection sequence helps maximize sensitivity to heteronuclei and increase signal to noise ratio. The isotope enrichment strategy greatly enhances the signal intensity and confers a good spectrum in a relatively short acquisition time. A good quality 2D spectrum is attainable in 20 min. However, observation of some less abundant lipid species benefits from a longer acquisition time of 90 min. Importantly, such maps are not attainable from normally grown (unlabeled) cells using larger quantities of crude lipid in the same volume as these lipids precipitate at high concentrations in the NMR solvents. After generation of a baseline lipid profile, to facilitate interpretation of the data, it is necessary to define biomarker signals of known lipid species using purified standards. In the lipid profile, each molecule is represented by multiple signals. Although many signals in the 2D spectrum still overlap, each lipid molecule generally has at least one discrete signal present. These discrete signals are designated as biomarker peaks. Therefore, the presence or absence and the quantity of a certain molecule can be readily determined by the identification of these discrete and diagnostic biomarker peaks within the 2D spectrum. When possible it is useful to identify and utilize multiple biomarker signals for molecules of interest to confer a greater degree of confidence in the data interpretation and hence limit the possibility of artifacts from unidentified overlapping species. Due to the complexity of the spectra and incomplete spectral assignments in the literature, it is important to purify a set of standards and analyze them under the same conditions as the lipid profile, in order to assign biomarkers effectively. For some molecules, their abundance in M. tuberculosis is not sufficient to purify enough sample. This can often be resolved by utilizing the increased expression of the targeted lipids in other mycobacterial species, from which they can then be isolated in suitable amounts. If necessary 3D HCCH TOCSY and 3D HCCH COSY can then be used to confirm assignment of biomarker signals in the total lipid profile. This is especially useful for confirming the shift of any signals that may vary slightly from standards that were purified from other mycobacterial species, which occurs due to slight variations in structural composition or substitution. After biomarkers have been assigned in the total lipid profile, spectral sets can be compared and lipids quantified by processing the data using software such as Felix or NMR Pipe. 1. Add 2 ml of CDCl3 and 1 ml of CD3OD per 1 mg of wet pellet weight. Mix the sample to homogeneity and extract at 37°C with continuous shaking at 250 rpm for 90 min.



Lipid Profiling Using Two-Dimensional Heteronuclear Single Quantum

95

2. After extraction, add an extra 1 ml of CD3OD for every milligram of original pellet weight and centrifuge at 3,700 × g for 10 min. Due to the water released from the cells during lysis, the additional CD3OD is necessary to allow the cell debris to pellet and to form a single phase solution. If the solution is still biphasic after centrifugation, CD3OD can be added dropwise with mixing until a single phase is formed after centrifugation. 3. Remove the supernatant to a clean microcentrifuge tube, being careful not to disrupt the pellet. Recentrifuge at 3,700 × g for 10 min and carefully transfer the supernatant to a fresh tube. 4. Transfer an appropriate volume of the supernatant to a clean dry NMR tube and acquire a 2D HSQC spectrum according to the NMR spectrometer used (Notes 4 and 5). 3.3. Preparation of the Mycolic Acid Pool

One important population of lipids in Mycobacteria is the mycolic acids. These unusual long chain fatty acids form a pseudomembrane layer where a mixture of free lipids, proteins, and glycolipids are embedded in a parallel inclusion. They are tethered to the cell wall by the arabinogalactan and therefore additional steps are required to isolate them from the cell pellet to complete the lipid profile in global lipid analysis studies. 13C enrichment significantly reduces the amount of sample required and they can easily be isolated from the left over cell debris after extraction of the free lipids. 1. The remaining pellet left after free lipid extraction is washed with 1 ml of D2O, which is subsequently removed by centrifugation and decantation. 2. A solution of 1 M methanolic KOH is added to the washed pellet (2 ml per 100 mg of original weight of the wet cell pellet). The methanolic solution is then transferred to a glass screw capped tubes and stirred for 16 h at 80°C. 3. The methanolysate is then cooled down and acidified by adding concentrated HCl dropwise to reach pH 4, to precipitate the free mycolic acids. The acidified methanolic solution is then diluted 3× with deionized water and extracted with one volume of diethyl ether. The top ether layer is carefully removed and transferred to another tube. 4. The extraction step is repeated and the ether extract is pooled and washed collectively with one volume of deionized water. The water is finally removed and the ether layer is dried under an atmosphere of inert gas. 5. The lipid film produced upon dryness is then dissolved in 1:1 solution of CD3OD: CDCl3 and transferred to a standard NMR tube.

96

Mahrous, Lee, and Lee

3.4. Alternative Extraction Methods

Although it is advantageous to monitor the whole cell lipidome at once, in some cases lipids of low abundance are better observed in a more limited subpopulation. The general extraction method discussed previously in Subheading 3.2 can be easily modified, through the incorporation of simple extraction or fractionation processes, to focus on specific or less abundant subpopulations. For example, methods include successive extraction of cells with deuterated solvents of increasing polarity, use of scavenger resins to pull down and fractionate anionic lipids and cationic lipids (17, 18), and the use of solvent partitions to remove very abundant signals or to separate the polar and apolar lipid pools. Separation of the polar and apolar lipids can be easily achieved through a simple modification to this method. As water is released from cells during extraction with 2:1 CDCl3:CD3OD, the relative ratio of chloroform to methanol to water in the solvent mixture results in partitioning of the sample into two phases as described in Subheading 3.2. The two phases can simply be collected and analyzed separately rather than adding more CD3OD to achieve a monophasic solution. 1. Add 2 ml of CDCl3 and 1 ml of CD3OD per 1 mg of wet pellet weight. Mix the sample to homogeneity and extract at 37°C with continuous shaking at 250 rpm for 1 h, 30 min. After extraction, centrifuge the sample at 3,700 × g for 10 min. A 2-layer suspension will be formed with the cellular debris at the interface. 2. Remove the aqueous upper layer carefully and transfer to a clean tube. This constitutes the bulk of the polar cell wall lipids. 3. Transfer the lower organic to a clean microcentrifuge tube. Add 2 ml of CDCl3 per original cell pellet weight to the extraction mixture. Vortex the tube for 30 s, then centrifuge the mixture at 3,700 × g for 10 min. 4. Transfer the chloroformic extract (lower layer) to a clean NMR tube. This constitutes the apolar lipid pool.

3.5. Results

1. Figure 1 shows the 2D HSQC lipid profile of M. bovis BCG. Signals in the HSQC spectrum can be categorized into three main regions; the most up-field region (d1H,–0.5–3.0 ppm) representing mainly the aliphatic chain of the lipid molecules, the far down-field region (d1H, 5.2–8.5 ppm) representing unsaturated and aromatic substructures and the middle region (d1H, 3.2–5.4 ppm), representing signals from sugars attached to glycolipids and phospholipids. Two sets of biomarker signals for menaquinone and trehalose dimycolates are labeled on the HSQC spectrum for illustration.



Lipid Profiling Using Two-Dimensional Heteronuclear Single Quantum

97

Fig. 1. 2D HSQC lipid profile of M. bovis BCG: 1H-13C HSQC spectrum of the 2:1 v/v CDCl3:CD3OD extract of M. bovis BCG. The extract was obtained as described in the text Subheading 3.2. For illustration, three biomarker signals for menaquinones are shown (signals Ia–c), and two biomarker signals are shown for trehalose dimycolate (signals IIa,b) including the diastereotopic CH2. Signal III indicates the signal for DMSO, which was included as an internal standard during the extraction. The total sample preparation time including extraction was 2 h. Total NMR acquisition time was 1 h and 40 min (ni = 256, nt = 10).

2. Figure 2 demonstrates the utility of this technique to probe gene function by detecting changes in the lipid pool as a result of gene knockout. In this experiment the lipid profiles of the hypervirulent M. tuberculosis HN-878 was compared to that of HN878-▵Dpks, a mutant strain where the pks 1–15 gene responsible for the production of the virulence factor phenolicglycolipid (PGL) was effectively disrupted. The three biomarker signals corresponding to M. tuberculosis PGL can be observed in the lipid profile of the clinical isolate HN-878, but are missing in the lipid profile of the mutant strain as a result of pks1–15 gene knockout. PGL is generally a difficult molecule to isolate from M. tuberculosis due to its low abundance, which makes application of this technique particularly useful for PGL studies. Indeed, this method was found to be applicable for rapidly screening clinical isolates for the presence or absence of PGL expression. 3. Figure 3 shows the 2D 1H-13C HSQC spectrum of the mycolic acid pool extracted from M. bovis BCG grown in Middlebrook 7H9 supplemented with 13C glucose and 13C glycerol under normal growth conditions. Each class of

98

Mahrous, Lee, and Lee

a

c

e

b

d

f

Fig. 2. Demonstration of lipid profiling to probe gene function. Biomarker signals for phenolicglycolipid (PGL) were defined using the 1H-13C HSQC spectrum acquired from a sample of PGL purified from M. liflandii [shown as circled cross peaks in spectra (a) and (b)]. These signals were readily detected in extracts from wild type M tuberculosis: HN878 (c) and (d) but not in the knockout strain of HN878 ▵∆ pks (e) and (f). For simplicity, relevant smaller regions are shown from the whole HSQC spectra: two biomarker aromatic signals are shown in solid circles in panels (a), (c), (e) (dH 6.7–7.4, dC 115–133) and the benzylic CH2 signal is shown in dashed circles in panels (b), (d), (f) (dH 2.3–3.4, dC 33.8–49.4). Due to differences in the glycosidation pattern between PGL molecules produced by M. tuberculosis and M. liflandii, a small difference in the chemical shift of the aromatic protons at the ortho-position to the glycosidic linkage was observed in their respective 2D-HSQC spectrum. Adapted from (11) with permission.

mycolic acids exhibit unique biomarker signals, which can be used to detect the relative abundance of a certain class in the mycolic acid pool. 4. Figure 4 demonstrates the differences in the polar and apolar lipid profiles of M. bovis BCG compared to the total lipid profile: total lipids A, apolar lipids B, polar lipids C. By acquiring the 2D HSQC spectra of the polar and apolar lipid fractions separately, some overlapping signals were effectively resolved. Noticeably, the abundant triglycerides and phosphatidyl inositol mannosides signals were eliminated from the polar lipid pool allowing the detection of many glycolipids of relatively low abundance.



Lipid Profiling Using Two-Dimensional Heteronuclear Single Quantum

99

Fig. 3. 2D-HSQC mycolic acid profiles for M. bovis BCG. 1H-13C HSQC spectrum of mycolic acids, extracted from M. bovis BCG as described in text Subheading 3.3. Examples for signals common to all mycolic acid are shown by arrow [(a), bCH–OH (b), aCH–COOH)]. Signals indicative to specific classes of mycolic acids are circled in dashed lines.

4. Notes 1. The handling of mycobacterial species should follow the biosafety recommendation for each specific species. Mycobacterium tuberculosis must be handled at Biosafety Level 3. Handling of extracts should be performed in accordance with institutional guidelines. The sterilization ability of the extraction methods should be verified before samples are removed from the BSL3 laboratory. In this case, it was found that extraction of 200 mg of cells with 600 ml of a 2:1 mixture of CDCl3:CD3OD for 90 min followed by two centrifugation steps at 3,700 × g for 10 min resulted in a sterile sample. Samples were prepared in triplicate, dried under sterile conditions, resuspended in 7H9 and the entire contents of a single preparation were plated on 7H11 media or inoculated into 7H9 broth. After 6 weeks, no growth on plates or in broth was observed. 2. Composition of 13C-7H9 Media (a) 45 ml Sterile Middlebrook 7H9 media (7H9, 0.05% Tween 80) (b) 5 ml Filter sterilized solution of BSA (5%) and NaCl (0.8%) (c) 1 ml Filter sterilized solution of 13C-glucose 0.1 gm/ml (d) 1 ml Filter sterilized solution of 13C-glycerol 0.1 gm/ml (e) the growth

100

a

Mahrous, Lee, and Lee

b

c

Fig. 4. Separation of the polar and apolar lipid profiles of M. bovis BCG. 1H-13C HSQC spectra generated from M. bovis BCG grown in 13C-labeled 7H9 (a) total lipid, (b) apolar lipids, and (c) polar lipids. Expanded views showing detail of the biomarkers for mycobactins, mycoside B, (dashed box, dH 6.68–7.28, dC 113–132.3) and glycerol-containing phospholipids (solid box, dH 5.0–5.44, dC 67–78.8) have been shown in these spectra as examples of how lipids can be separated through simple fractionation. Only mycoside B is present in the apolar lipid spectrum while mycobactin signals are observed in the polar spectrum only. The glycerol-containing phospholipids are overlapping in the total lipid pool, however, when the abundant triglyceride and PIMs are removed into the apolar solvent, less-abundant species were better presented in the polar lipid spectrum.

rate of bacteria is not affected by the use of a labeled carbon source. 3. Examples of fast growing species include the saprophytic mycobacterial species M. smegmatis and M. phlei. Examples of slow growing species include M. tuberculosis, M. bovis, and M avium.



Lipid Profiling Using Two-Dimensional Heteronuclear Single Quantum

101

4. For the NMR data presented in this chapter, a Varian Inova 500 MHz NMR spectrometer was used. The spectrometer was equipped with a triple resonance trpfg 5 mm probe, which was tuned, matched, and shimmed for each individual sample. The 1H-13C HSQC spectra were acquired with a standard pulse sequence with 10 or 12 transients and 256 increments. 5. Extraction of lipids from a 20–50 ml culture usually generates approximately 0.5–0.7 ml of extract, which is appropriate for use with 5 mm sample probe and standard 5 mm NMR tubes. However, microsampling inserts and other low volume NMR probes require far less sample volume and may be used if analysis of smaller samples is required.

Acknowledgments We thank Dr Clifton Barry III, (National Institutes of Health) for providing M. tuberculosis HN-878 and HN878-▵D pks strains, Dr. Pamela L.C. Small, (University of Tennessee, Knoxville) for providing both M. marinum and M. liflindii strains, and Dr. Wei Li (University of Tennessee, Memphis) for technical assistance. We acknowledge financial support for this work from National Institutes of Health grant AI076938. References 1. Goren, M. B., O. Brokl, and W. B. Schaefer. 1974. Lipids of putative relevance to virulence in Mycobacterium tuberculosis: phthiocerol dimycocerosate and the attenuation indicator lipid. Infection and immunity 9:150–158. 2. Goren, M. B., O. Brokl, and W. B. Schaefer. 1974. Lipids of putative relevance to virulence in Mycobacterium tuberculosis: correlation of virulence with elaboration of sulfatides and strongly acidic lipids. Infection and immunity 9:142–149. 3. Kato, M., M. Kusunose, K. Miki, K. Matsunaga, and Y. Yamamura. 1959. The mechanism of the toxicity of cord factor. The American review of respiratory disease 80:240–248. 4. Etemadi, A. H. 1967. [Mycolic acids. Structure, biogenesis and phylogenetic value]. Exposes annuels de biochimie medicale 28:77–109. 5. Etemadi, A. H. 1967. [Structural and biogenetic correlations of mycolic acids in relation to the phylogenesis of various genera

of Actinomycetales]. Bulletin de la Societe de chimie biologique 49:695–706. 6. Lechevalier, M. P., A. C. Horan, and H. Lechevalier. 1971. Lipid composition in the classification of nocardiae and mycobacteria. Journal of bacteriology 105:313–318. 7. Minnikin, D. E., L. Alshamaony, and M. Goodfellow. 1975. Differentiation of Mycobacterium, Nocardia, and related taxa by thin-layer chromatographic analysis of wholeorganism methanolysates. Journal of general microbiology 88:200–204. 8. Dobson, G., D. E. Minnikin, S. M. Minnikin, M. Parlett, M. Goodfellow, M. Ridell, and M. Magnusson. 1985. Systematic analysis of complex mycobacterial lipids. In Chemical methods in bacterial systematics. M. Goodfellow, D. E. Minnikin(eds). Academic, London. 237–2654. 9. Jain, M., C. J. Petzold, M. W. Schelle, M. D. Leavell, J. D. Mougous, C. R. Bertozzi, J. A. Leary, and J. S. Cox. 2007. Lipidomics reveals control of Mycobacterium tuberculosis

102

Mahrous, Lee, and Lee

virulence lipids via metabolic coupling. Proceedings of the National Academy of Sciences of the United States of America 104:5133–5138. 10. Shui, G., A. K. Bendt, K. Pethe, T. Dick, and M. R. Wenk. 2007. Sensitive profiling of chemically diverse bioactive lipids. Journal of lipid research 48:1976–1984. 11. Mahrous, E. A., R. B. Lee, and R. E. Lee. 2008. A rapid approach to lipid profiling of mycobacteria using 2D HSQC NMR maps. Journal of lipid research 49:455–463. 12. 2008. WHO report 2008 Global tuberculosis control surveillance, planning, financing. WHO. 17–20. 13. Brennan, P. J. and H. Nikaido. 1995. The envelope of mycobacteria. Annual review of biochemistry 64:29–63. 14. Puzo, G. 1990. The carbohydrate- and lipidcontaining cell wall of mycobacteria, phenolic glycolipids: structure and immunological properties. Critical reviews in microbiology 17:305–327. 15. Karakousis, P. C., W. R. Bishai, and S. E. Dorman. 2004. Mycobacterium tuberculosis cell envelope

lipids and the host immune response. Cellular microbiology 6:105–116. 16. Ortalo-Magne, A., A. Lemassu, M. A. Laneelle, F. Bardou, G. Silve, P. Gounon, G. Marchal, and M. Daffe. 1996. Identification of the surface-exposed lipids on the cell envelopes of Mycobacterium tuberculosis and other mycobacterial species. Journal of bacteriology 178:456–461. 17. Villeneuve, C., G. Etienne, V. Abadie, H. Montrozier, C. Bordier, F. Laval, M. Daffe, I. Maridonneau-Parini, and C. Astarie-Dequeker. 2003. Surface-exposed glycopeptidolipids of Mycobacterium smegmatis specifically inhibit the phagocytosis of mycobacteria by human macrophages. Identification of a novel family of glycopeptidolipids. The Journal of biological chemistry 278:51291–51300. 18. Villeneuve, C., M. Gilleron, I. MaridonneauParini, M. Daffe, C. Astarie-Dequeker, and G. Etienne. 2005. Mycobacteria use their surface-exposed glycolipids to infect human macrophages through a receptor-dependent process. Journal of lipid research 46:475–483.

Chapter 6 Capabilities and Drawbacks of Phospholipid Analysis by MALDI-TOF Mass Spectrometry

1

2 3

Beate Fuchs, Ariane Nimptsch, Rosmarie Süß, and Jürgen Schiller

4

Summary

5

The important roles of lipids particularly certain phospholipids in signal transduction processes and as important disease markers are becoming increasingly evident. Unfortunately, however, sensitive methods of lipid analysis are established to a much lesser extent than, e.g., methods of protein analysis. Mass spectrometry (MS) is an increasingly used technique of lipid analysis and electrospray ionization (ESI) MS is the so far most established ionization method. Although matrix-assisted laser desorption and ionization time-of-flight mass spectrometry (MALDI-TOF MS) was so far primarily used for protein analysis, however, this method has itself proven to be very useful in the field of lipid analysis, too. This chapter gives an overview of methodological aspects of MALDI-TOF MS in lipid research and summarizes the specific advantages and drawbacks of this soft-ionization method. In particular, suppression effects of some lipid classes, especially those with quaternary ammonia groups such as phosphatidylcholine, will be highlighted and possible ways to overcome this problem (use of different matrices, separation of the relevant lipid mixture prior to analysis) will be discussed on the example of an organic liver extract.

6 7 8 9 10 11 12 13 14 15 16 17 18

Key words: Glycerophospholipids, Lipid analysis, MALDI-TOF MS, Lipid extracts, Liver, Matrix, Thin-layer chromatography, Mass spectrometry

19 20

1. Introduction

21

Over decades lipids were primarily considered as the cellular “packing material” of more important contents, for instance, enzymes, nucleic acids, etc., and as energy-rich “fuel” in nutrition (1). Nowadays, however, lipids such as diacylglycerols and particularly phospholipids (PLs) such as phosphatidic acids or ­phosphoinositides are known to represent important secondmessenger molecules involved in cellular communication (2). Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_6, © Humana Press, a part of Springer Science + Business Media, LLC 2009

103

22 23 24 25 26 27 28

104 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76

Fuchs et al.

Additionally, lipids such as lysophosphatidylcholines (LPCs) were also recognized as important disease markers, for instance, in atherosclerosis or rheumatoid arthritis (3, 4). Accordingly, in addition to terms as “proteomics” or “genomics”, the term “lipidomics” (5, 6) was also recently introduced. In order to make a long story short: The interest in lipids and their analysis has significantly increased during the last decade. Surprisingly, however, in comparison to the analysis of proteins, there are currently only much less developed protocols of lipid analysis. One potential reason is the considerable diversity of lipids as “lipid” – in the most general definition – relates to all compounds that may be isolated from body fluids or biological tissues by extraction with organic solvents due to their apolar character (1). This diversity of phospholipids is not only coming from differences in the headgroup (e.g., choline or ethanolamine) but also from the linkage type between the apolar alkyl chains and the glycerol (diacyl, alkyl-acyl-, or alkenyl-acyl) and, finally, the large variability of potential fatty acyl residues (7). Therefore, complex lipid patterns with thousands of different lipid species can be expected if crude organic extracts from biological samples are analyzed. The best analytical approach to assess this diversity is still an open issue (8): Although chromatographic techniques are highly established, they have the disadvantage – equally if liquid chromatography, in particular, high-performance liquid chromatography (HPLC) (9) or thin-layer chromatography (TLC) (10) is used – that several runs are normally required in order to obtain complete compositional information: Normal phase chromatography is needed for the separation of lipids according to their headgroups, whereas reversed phase chromatography is usually used for the differentiation of lipids according to their fatty acyl compositions (8). Another problem is the detection of the lipids within the obtained fractions. If, for instance, a UV detector is used, lipids with unsaturated fatty acyl residues are primarily detected, whereas this kind of detector is not suitable for the detection of lipids with saturated fatty acyl residues as they lack the UV absorption of olefinic residues (1). Mass spectrometry (MS) is increasingly used for the analysis of complex lipid mixtures (11), in particular, for lipidomic studies (5). Although methods such as atmospheric pressure chemical ionization (APCI) (12) or electrospray ionization (ESI) (13) are nowadays considered to be the methods of choice for lipids, there is growing evidence that matrix-assisted laser desorption and ionization time-of-flight mass spectrometry (MALDI-TOF MS) is also capable of providing important information about the composition of an unknown mixture. The particular advantages of MALDI MS are the simple performance and the high sensitivity (14) down to a few attomoles of the analyte. The use



1.1. Advantages and Disadvantages of MALDI-TOF MS in the Lipid Field

Capabilities and Drawbacks of Phospholipid Analysis

105

of MALDI-TOF MS for lipid and phospholipid analysis has been recently reviewed (15, 16) and compared with other methods of lipid analysis (17).

77

Although a comprehensive treatise of the fundamentals of MALDI-TOF MS is clearly outside the scope of this chapter (for a detailed description see (18)), some important aspects representing the most important differences between MALDI MS and other MS methods have to be shortly discussed: 1. A solid sample is used in MALDI-TOF MS. The lipid sample is mixed directly on the sample plate (the “target”) or prior to its deposition with the matrix solution and is subsequently allowed to crystallize (15). The matrix is required to absorb the energy emitted from the laser and enables the ionization of the analyte, even if this analyte does not exhibit a major absorption at the laser wavelength (often 337 nm). Although there are UV- and IR lasers in use (18), the majority of commercially available MALDI devices are equipped with UV lasers and, therefore, IR lasers will not be considered here to a larger extent. However, completely different matrix compounds in comparison to UV lasers are required if IR lasers are used. One common IR MALDI matrix is, for instance, glycerol that provides intense IR absorption (19). Equally if IR or UV MALDI is used: The actual analyte is not the sample as such but cocrystals between the matrix and the sample of interest. As these cocrystals are never completely homogenous, the reproducibility of MALDI measurements is often doubted (20). However, as many laser shots, each leading to an individual mass spectrum, are averaged for the final mass spectrum, these inhomogeneities may be significantly minimized. Additionally, the matrix, that normally comprises an aromatic, rather hydrophobic residue, as well as the lipid are both readily soluble in organic solvents and there is no need to use water as additional solvent (as, for instance, in the case of water-soluble analytes). Therefore, the homogeneity of matrix/lipid crystals is higher than that of matrix/protein crystals (21). Finally, the use of solid samples enables to record spatially resolved mass spectra. This “mass spectrometric imaging” is an important topic of modern research and attracts currently significant interest in medical diagnosis, for instance, in order to differentiate different types of cancer (22).

80

2. Soft ionization MS techniques as MALDI provide – in contrast to traditional electron impact (EI) MS – no radical cations by the abstraction of an electron from the analyte of interest (cf. Fig. 1), but “quasimolecular” ions. Thereby the analyte is cationized by the addition of a proton or an alkali metal ion (positive ions) or the loss of a proton (negative ions) (23). Protons are normally provided by the matrix that is often an

117

78 79

81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 118 119 120 121 122 123

106

Fuchs et al.

Fig. 1. Schema showing the events of ion formation upon electron impact ionization and quasimolecular ion generation that is most relevant to the MALDI ionization process. Please note that radical cations are generated upon EI ionization. 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148

organic acid, whereas alkali metal ions are present as such in virtually all biological samples. The tendency of the analyte to add or loose a proton depends, however, on the basicity of the analyte in comparison to the acidity of the used matrix or vice versa. Although 2,5-dihydroxybenzoic acid (DHB) is the most common matrix for the analysis of lipids (24), it should be noted that for special problems, other matrices as paranitroaniline (PNA) can be advantageously used (25). This will be illustrated below in more detail. A survey of the characteristics of electron impact ionization and positive as well as negative quasimolecular ion formation is given in Fig. 1. In the light of these aspects it is evident that providing information on the detection limits of MALDI-TOF MS makes sense only if the applied matrix is simultaneously mentioned (26). 3. Although MALDI is relatively robust against impurities of the sample as salts or buffer components (23), great care is needed if detergents such as Triton or Brij, for instance, are used for the extraction of lipids from the biological material as these molecules are also rather apolar and cannot be easily separated from the lipids of interest. The use of commercially available “detergent removal” kits normally fails in the case of lipids because these kits were primarily developed for the purification of proteins that are normally more polar in comparison to the applied detergent. Due to that problem, the use of organic solvents or mixtures is recommended for lipid



Capabilities and Drawbacks of Phospholipid Analysis

107

extraction. Additionally, the extraction with organic solvents removes simultaneously salts and other low molecular compounds of high polarity that might affect the ion yield.

149

4. MALDI-TOF MS offers an extremely high sensitivity and detection limits in the atto- (10−18) and even low zeptomole (10−21) range were already reported in the case of lipids – in particular for lyso-phosphatidylcholine (14). Although these detection limits seem only realistic for special samples in combination with highly sophisticated matrices, sensitivities in the femtomole (10−15) range can be easily achieved by standard MALDI-TOF MS. This means that 1 ml of a 1 mg/ml lipid sample is absolutely sufficient to detect the lipid of interest (26). This high sensitivity clearly requires the use of extremely pure chemicals: It is much simpler to detect very small amounts of the lipid of interest than – vice versa – NOT to detect impurities in the solvents, the matrix compounds, etc.

152 153

2. Materials 2.1. Equipment and Supplies (See Also Note 1)

2.2. Reagents

150 151

154 155 156 157 158 159 160 161 162 163 164 165

166

1. Bruker “Autoflex” MALDI-TOF mass spectrometer (Bremen, Germany) equipped with reflectron, “delayed extraction facility” and N2 laser emitting at 337 nm. The capability to record positive and negative ion spectra should be available.

167 168

2. MALDI targets made from stainless steel or from aluminum with gold-coated surface (see Note 2).

171

3. Micropipetts (see Note 3).

173

4. Glass (Hamilton) syringes of different sizes.

174

5. Small glass vessels for single use for mixing matrix and sample or for diluting stock solutions of lipids (available, for instance, from Knauer, Berlin, Germany).

175 176

6. TLC silica gel 60 plates (Merck, Darmstadt, Germany) and TLC trough (CAMAG, Muttenz, Switzerland).

178 179

7. Heat gun or common hair dryer (see Note 4).

180

8. Standard laboratory centrifuge.

181

1. Some selected phospholipid samples of known composition and with known concentration should be used as the first samples. As palmitoyl and oleoyl residues are highly abundant in samples of biological origin (16), it is recommended to use stock solutions of 1-palmitoyl-2-oleoyl-sn-phosphatidylcholine, -ethanolamine, and -glycerol. These are available from Avanti Polar Lipids, Alabaster, AL, USA. These compounds

182

169 170 172

177

183 184 185 186 187 188

108

Fuchs et al.

are normally abbreviated by POPC, POPE, and POPG. A stock concentration of 1 mg/ml lipid in chloroform is recommended for best performance and to optimize instrumental parameters.

189 190 191 192 194

2. 2,5-Dihydroxybenzoic acid (DHB) and para-nitroaniline (PNA) of highest available purity (see Note 5).

195 196

3. Chloroform, methanol, ethanol, triethylamine, and distilled water of highest commercially available quality (see Note 6).

197

4. A small piece of fresh bovine liver that is best obtained from a local butcher or slaughterhouse.

193

198

199

200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230

3. Methods 3.1. Sample Processing 3.1.1. Artificial Lipid Samples of Known Composition

1. If sufficient amounts of samples are available, the use of stock solutions in the mg/ml concentration range is recommended. Therefore, dilute the available lipid samples to about 1 mg/ml with CHCl3. Mix one equivalent of the lipid standard solution with one equivalent of DHB matrix solution (0.5 mol/l corresponding to about 77 mg/ml in CH3OH). Trifluoroacetic acid (TFA) is often suggested as an additive because it enhances the yield of H+ adducts due to its considerable acidity (23). However, TFA may also have deleterious effects because lipids containing alkenyl ether groups (“plasmalogens”) are readily hydrolyzed in the presence of TFA (27, 28). Therefore, the use of TFA is discouraged – particularly if unknown lipid mixtures are to be analyzed. Prepare the same samples in the presence of PNA (0.17 mol/l in chloroform/methanol (2:1, v/v)). 2. Apply the prepared samples to the MALDI target. Do not be surprised if a very large spot is formed and the sample “spreads” out over the sample plate: In contrast to water, organic solvents possess only a rather small surface tension (H2O: ~73 × 10−3 N/m; CH3OH ~23 × 10−3 N/m). This prevents the formation of a small spot on the metal surface of the sample plate. Do not be worried! Simply leave some space between the individual samples in order to avoid mixing of the different samples during deposition onto the MALDI target. Avoid touching the MALDI target with the pipette tip or the needle of the Hamilton syringe in order not to affect homogeneous crystallization. The MALDI target should also not be touched with the fingers. 3. Evaporate the solvent quickly by drying the sample plate with a hair dryer and load the prepared sample plate directly into the mass spectrometer. Avoid long-term exposition of the



3.1.2. Preparation of the Liver Extract

3.1.3. Separation of the Crude Lipid Extract by Thin-Layer Chromatography

Capabilities and Drawbacks of Phospholipid Analysis

109

TLC plate to air: Due to the large surface of the lipid film on the MALDI target, lipids may be easily oxidized by air. This is particularly a problem if diluted samples are analyzed. It has also been shown that the content of residual water in the organic solvents influences the quality of the mass spectra (29). This may be surprising as completely “dry” samples are investigated by MALDI-TOF MS under high vacuum conditions. However, there is increasing evidence that the solvent or solvent mixture determines the homogeneity of the crystallization between the matrix and the analyte. Therefore, the quality of mass spectra depends dramatically on the used solvent system.

231

1. Cut the liver tissue into small pieces with a scalpel. Add the 20-fold amount (by weight) of Bligh & Dyer solvent (30) mixture (CHCl3:CH3OH:H2O = 1:1:0.9) and stir or vortex the resulting mix vigorously for a few minutes. Subsequently centrifuge the sample (10 min, ca. 1000 g, 20°C) in order to improve the separation of the organic (bottom) and the H2O/CH3OH phase (see also Note 7).

243

2. Remove the organic (lower) phase by a Hamilton syringe and transfer it to another unused vial. Do not try to get the complete organic phase in order to avoid the introduction of impurities from the aqueous layer. This organic phase is used for all further experiments and the lipid concentration is of the order of 6–8 mg/ml. The lipid yield can be easily determined by evaporation of the solvent under reduced pressure and subsequent weighing of the residual material. If needed, a bovine liver extract is also commercially available from AVANTI Polar Lipids, Alabaster, AL, USA. Another established method to determine the PL concentration is the classical phosphate determination according to Bartlett (31).

250

3. The obtained lipid extract may be directly used for MALDITOF MS by simply diluting the organic solution with the prepared matrix solution (1:10, v/v).

262 263

1. For the separation of phospholipids according to differences of the headgroup normal phase chromatography is used, whereas reversed phase chromatography is used for the separation according to differences in the fatty acyl compositions (8). Use HPTLC silica gel 60 plates (10 cm × 10 cm in size with aluminum or glass backs) (Merck, Darmstadt, Germany). Prerun the plates with the needed solvent mixture in order to remove impurities. Dry the plate carefully.

265 266

2. Apply 0.5–1 ml (corresponding to a total lipid amount of 4–8 mg) of the liver extract by using a standard Hamilton syringe as small spots with 1 cm space between the individual spots and

273

232 233 234 235 236 237 238 239 240 241 242

244 245 246 247 248 249 251 252 253 254 255 256 257 258 259 260 261

264

267 268 269 270 271 272 274 275

110

Fuchs et al.

at a distance of 1.5 cm from the bottom edge of the TLC plate. Also apply a lipid mix of known composition (about 1–2 mg per PL). Dry the plate carefully prior to development.

276 277 278

3. Develop the plates in a commercially available vertical ascending TLC chamber (CAMAG, Switzerland) with chamber saturation using chloroform, ethanol, water, triethylamine (30:35:7:35, v/v/v/v) as solvent mixture (32). This should result in good separation quality of the individual PL classes. The required time of development is about 35 min: The total length of the run is about 6 cm under these conditions and is a good compromise between resolution and the required time. The separation is performed at room temperature (22 ± 2°C) and 50 ± 5% relative humidity.

279 280 281 282 283 284 285 286 287 288

4. Subsequent to development, TLC plates are dried under a stream of warm air and the lipids are visualized by spraying with a solution of primuline (Direct Yellow 59) in acetone according to (33). Upon irradiation by UV light individual PLs become detectable as violet spots and can be easily identified by comparison with the reference mixture. The binding between the dye and the lipid is noncovalent and does not affect the quality of the mass spectra. The monoisotopic mass of primuline is 475.0 and, therefore, a very small signal is sometimes detected at m/z = 498 (Na+ adduct) in the positive ion mass spectra.

289 290 291 292 293 294 295 296 297 298 299

5. Mark the spots of interest (under UV light) by circling with a soft pencil.

300 301

6. The spots of interest are carefully scratched off from the TLC plate into different small glass vessels and the PLs are eluted from the silica gel by the addition of a mixture of 75 ml CHC13, 75 ml methanol, and 75 ml 0.9% NaCl in water and intense vortexing (see Note 8). Afterward, samples are centrifuged (ca. 1000 g) as described above to enhance phase separation. The organic layers are evaporated to dryness and the residual material redissolved in 20 ml matrix solution (DHB or PNA) and directly used for MALDI-TOF MS.

302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320

3.2. Recording MALDI-TOF Mass Spectra

1. Due to the large number of MALDI devices that are available from different suppliers, it is impossible to describe the necessary experimental parameters in detail. Therefore, please consult the manual of your device for a suitable data file to start with. You should start with “delayed extraction conditions” (DE) and use the reflectron of the device. This results in higher resolution and mass accuracy than the common linear mode that provides, however, higher sensitivity. 2. It is recommended to start in all cases with the analysis of a known sample of known concentration in order to check if the



3.3. Quantitative Data Analysis

Capabilities and Drawbacks of Phospholipid Analysis

111

device is properly working and all parameters are adequately set. This known sample may also be used to check the mass accuracy, i.e., the quality of the applied – often default – calibration, as well as the resolution achievable by the used instrumental settings. Please note, however, that the applied laser intensity has the most pronounced effect on spectral quality.

321

3. Always take care that the MALDI target is carefully dried before it is inserted into the mass spectrometer in order to avoid a significant decrease of the vacuum – and the delay for waiting until high vacuum is re-established.

327

4. It is an incorrect assumption that spectral quality, in particular the signal-to-noise ratio, may be enhanced by increasing the laser intensity. Although absolute signal intensities may be enhanced at elevated laser intensities, the quality of the baseline simultaneously gets poor and the achievable resolution is simultaneously diminished. Therefore, always set the laser intensity as high as needed but as low as possible.

331 332

5. Try to move the laser randomly over the sample plate in order to average nonhomogeneous spots resulting from the sample preparation. It is, however, not possible to improve the signal-to-noise ratio by averaging a larger number of individual laser shots (19) because the level of unspecific chemical background noise forms the limiting criterion.

338 339

6. Try to optimize the required parameters always with a wellknown sample and use these parameters afterward for the unknown samples, i.e., the organic liver extract in our case.

344

7. DHB is completely stable under high vacuum conditions, while PNA tends to sublime at longer times in vacuum (25). Therefore, it is advisable to perform measurements with the PNA matrix relatively quickly. If no signal is detectable, try to add somewhat additional PNA matrix.

347

1. Quantitative evaluation of MALDI mass spectra is still a challenging task. However this does not only hold for MALDI MS but also for other soft-ionization techniques: The ion yield in EI mass spectra primarily depends on the ionization potential of the functional groups present in the analyte of interest. The ionization energy is rather constant for e.g., carbonyl groups – independent of their chemical environment. Therefore, the concentration of different compounds containing carbonyl groups can be easily determined by using the signal intensities. The comparison of different lipid classes with different acidities is particularly tricky if lipid mixtures are analyzed as the ion yield may be very different for the individual lipid classes (26). The very best way to obtain quantitative compositional information is normally the separation of the individual lipid

352

322 323 324 325 326 328 329 330

333 334 335 336 337

340 341 342 343 345 346 348 349 350 351

353 354 355 356 357 358 359 360 361 362 363 364 365

112

Fuchs et al.

classes and to add an internal standard to each fraction. As the structure of this internal standard should be as similar as possible to the analyte of interest, deuterated lipids are the reference compounds of choice (34). Unfortunately, however, this approach requires some prior knowledge about the composition of the sample to warrant that the standard is added in a suitable amount and particularly to avoid the addition of an excess of the standard that might result in severe signal suppression. Due to these problems, other simpler methods of quantitative analysis were also suggested.

366 367 368 369 370 371 372 373 374 375

2. Absolute peak intensities can hardly be used as quantitative concentration measures in MALDI-TOF mass spectra because they are affected by many different parameters, for instance, the applied laser intensity, the matrix to analyte ratio, the presence of impurities, etc.

376 377 378 379 380

3. Comparison of the intensity of the peak of interest to a defined matrix peak. Although common matrices have normally much lower masses than lipids, nearly all matrices, at least the ionic ones, tend to undergo photochemical reactions upon laser irradiation in the gas phase leading to peaks at higher m/z ratios (24). Although some applications of this method to apolar lipids as triacylglycerols (16) were described, the applicability of this method seems quite limited because the matrix peak pattern is influenced by many parameters.

381 382 383 384 385 386 387 388 389

4. Using the signal-to-noise (S/N) ratio. The achievable S/N ratio increases with the lipid concentration over a large concentration range. Only by using very large amounts of lipids, i.e., an inappropriate ratio between lipid and matrix, the S/N ratio decreases. Although this method seems to be applicable for all substances, it was so far only used for rather polar species, in particular lysophospholipids (35) and phosphoinositides (36). The majority of MALDI devices calculates the S/N ratio directly from the obtained spectra. Therefore, this is an easily accessible parameter.

390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411

3.4. Results

1. A coarse overview about the shape of the positive (left) and negative (right) ion MALDI-TOF mass spectra of different phospholipids (as well as their characteristic headgroups) in the presence of different matrix compounds is shown in Fig. 2. 1-Palmitoyl-2-oleoyl-sn-phosphatidylcholine (POPC, (2a,2e)), -phosphatidylethanolamine (POPE, (2b, 2f)), and -phosphatidylglycerol (POPG, (2c, 2g)) were chosen as defined PLs because they are rather abundant in biological materials but differ in their charge states (1): POPC and POPE are zwitterionic PLs, whereas POPG is a negatively charged compound at physiological pH. In (2d, 2h) the spectra of a 1:1:1 mixture of these three PLs are shown in order to illustrate the problems of mixture analysis by



Capabilities and Drawbacks of Phospholipid Analysis

113

Fig. 2. Positive (left) and negative ion (right) MALDI-TOF mass spectra of 1-palmitoyl-2-oleoyl-sn-phosphatidylcholine (a), (e), 1-palmitoyl-2-oleoyl-sn-phosphatidylethanolamine (b), (f), 1-palmitoyl-2-oleoyl-sn-phosphatidylglycerol (c), (g) and a 1:1:1 mixture of these three compounds (d), (h). Positive ion spectra were recorded with a 0.5 M solution of DHB in methanol, whereas a 0.17 M solution of PNA (in CHCl3 and CH3OH) was used as matrix for the negative ion spectra. Lipid sample solutions (1 mg/ml) were diluted 1:1 (v/v) with the corresponding matrix and afterwards spotted onto the MALDI target. All peaks are marked according to their m/z ratios and the structures of both matrices are shown in the figure. Please note that the PC can not be detected as negative ion and the presence of a characteristic fragment ion in the POPE and POPG spectra (b, c).

MALDI-TOF MS. It is evident that the spectra differ significantly: POPC gives – in the presence of DHB – two signals at m/z = 760.6 and 782.6 in the positive ion mode according to the generation of the H+ and the Na+ adduct (1a) (21). Due to the permanent positive charge of the quaternary ammonia group (37), PC cannot be detected as negative ion under the applied conditions (1e).

Please note that the PC does not give any fragmentation products under these conditions – even the loss of the choline moiety may be prevented by careful control of the laser intensity (38). In contrast, the PE (1b) as well as the PG (1c) gives significant yields of a fragment ion that corresponds to the loss of the polar headgroup. As the fatty acyl composition is the same, both compounds give the same fragment at m/z = 577.5 (38). The structure of this fragment is also shown in (2c). Additionally, it is evident that the POPE with the monoisotopic mass of 717.5 g/mol gives only rather small yields of the H+ adduct (m/z = 718.5) but significant amounts of the Na+ adduct (m/z = 740.5) as well

412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429

114 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477

Fuchs et al.

as the Na+ adduct subsequent to the exchange of one H+ by one Na+ (m/z = 762.5). This is caused by the exchangeable protons of the –NH3+ group. The reason why POPE and POPG give more pronounced fragmentation than POPC is not yet known. However, it might be possible that the stability of the H+ adduct of POPE is not high enough to allow these ions to reach the detector. A similar observation was already made in the case of triacylgylcerols (39), where no H+ adducts were detectable at all. Using PNA as an alkaline matrix, POPE is also detectable as negative ion at m/z = 716.5 (2f). Please note that the detectability of POPE as negative ion (data not shown) in the presence of DHB would be extremely weak due to the acidic properties of this matrix. POPG can be detected as negative ion in the presence of DHB (data not shown) as well as PNA (2g, m/z = 747.5) as negative ion because of its enhanced acidity in comparison to POPE. Regarding mixture analysis significant differences are obtained between the positive and the negative ion spectra: In the positive ion mode (2d), the spectrum of the mixture is clearly dominated by the POPC. The POPE is only detectable with low intensity although it is present in the same amount as the POPC. POPG is not detectable at all under these conditions. Therefore, the presence of PCs prevents the detection of other PL species in mixtures (37). Due to this problem, a simple method to remove PC from lipid mixtures has been recently suggested (40). In contrast, the POPC is completely absent in the negative ion spectrum (2h) and only POPE and POPG are detectable in the mixture. Although this is indeed a very simple example, it is evident that mixture analysis by MALDI-TOF MS must be regarded with great caution. 2. The problem of signal suppression is still more evident if the organic liver extract is investigated (Fig. 3). The liver extract was chosen as an educational example because it can be easily prepared with good reproducibility and is even commercially available. The left hand spectra represent the positive ion spectra, while the spectra at the right hand represent the negative ion spectra. Spectra (3a) and (3c) were recorded with DHB as matrix, while PNA was used for the spectra shown in (3b) and (3d). It is obvious that the positive ion spectra are, beside minor intensity differences of the individual adducts, virtually identical equally if DHB or PNA is used as matrix (3a, 3b). Both spectra are dominated by different PC species and a more detailed assignment of the individual peaks is given in Table 1. The negative spectra, however, differ significantly and it is evident at the first glance that the spectrum recorded with the DHB matrix (3c) exhibits rather poor quality. Although peaks of a few PI species (the signal at m/z = 885.5, for instance, corresponds to PI 18:0/20:4) can be easily identified, there are no further lipids detectable.

Fig. 3. Positive (left) and negative ion (right) MALDI-TOF mass spectra of an organic bovine liver extract. Traces (a) and (c) were recorded with DHB, whereas (b) and (d) were recorded in the presence of PNA. Due to the large variety of individual lipid classes in the extract, a relatively concentrated lipid solution was used (about 6 mg/ml) and diluted 1:10 with the matrix. All peaks are marked according to their m/z ratios. Note the poor quality of the negative ion spectrum recorded with DHB as matrix.

Capabilities and Drawbacks of Phospholipid Analysis 115

116

Fuchs et al.

Table 1 Overview of the positive ions detected in the MALDI-TOF mass spectra of the liver extract and the corresponding assignments Peak position (m/z)

Assignment of molecular mass

577.5

Fragment of PE 16:0/18:1/PG 16:0/18:1

603.5

Fragment of PE 16:0/18:1

702.5

Oxidation product of PS 18:0/18:1 (+H+)

703.6

SM 16:0 (+H+)

715.0

4 DHB - 2 H+ + Na+ + 2 K+

721.0

4 DHB - 3 H+ + 3 Na+ + K+

721.5

Oxidation product of PI 18:0/20:4 (+H+)

725.6

SM 16:0 (+Na+)

727.0

4 DHB - 4 H+ + 5 Na+

743.5

Oxidation product of PI 18:0/20:4 (+Na+)

753.5

Oxidation product of PI 18:0/18:1

760.6

PC 16:0/18:1 (+H+)

766.5

PE 18:0/18:2 (+Na+)

768.5

PE 18:0/18:1 (+Na+)

771.5

PG 16:0/18:1 (+H+)

777.5

Oxidation product of PI 18:0/18:1 + H+

782.6

PC 16:0/18:1 (+Na+)

786.6

PC 18:0/18:2 (+H+)

788.5

PE 18:0/18:2 (−H+ + 2 Na+)

788.6

PC 18:0/18:1 (+H+) or PE 18:1/20:4 (+Na+)

790.5

PE 18:0/20:4 (+Na+)

793.5

PG 16:0/18:1 (+Na+)

808.6

PC 18:0/18:2 (+Na+)

810.6

PC 18:0/18:1 (+Na+) and PC 18:0/20:4 (+H+) or PS 18:0/18:2 (+H+)

812.5

PE 18:0/20:4 (−H+ + 2Na+)

812.6

PS 18:0/18:1 (+H+) and PC 18:0/20:3 (+H+)

832.6

PC 18:0/20:4 (+Na+)

834.6

PC 16:0/22:6 (+H+) or PS 18:0/18:1 (+ Na+)

(continued)



Capabilities and Drawbacks of Phospholipid Analysis

117

Table 1 (continued) Peak position (m/z)

Assignment of molecular mass

856.6

PC 16:0/22:6 (+Na+) or PS 18:0/18:1 (−H+ + 2 Na+)

885.5

PI 18:0/18:2 (+ H+)

887.5

PI 18:0/18:1 (+ H+)

909.5

PI 18:0/20:4 (+ H+) and PI 18:0/18:1 (+ Na+)

911.5

PI 18:0/20:3 (+ H+)

931.5

PI 18:0/20:4 (+ Na+)

933.5

PI 18:0/20:3 (+ Na+)

PC phosphatidylcholine, PI phosphatidylinositol, PS phosphatidylserine, PE phosphatidylethanolamine Please note that all discussed PL– besides PC – contain functional groups showing exchange with the solvents and/or ions of the matrix solution leading to complex peak patterns. Please note that in the case of acidic PL (PS and PI) the corresponding sodium salt is considered as the neutral molecule.

The peaks between about m/z = 720 and 780 are stemming from the applied DHB matrix (24). The spectrum recorded in the presence of PNA (3d) is characterized by a much higher quality and provides the additional advantage that PE species are detectable simultaneously with the PI species. The additional signals at m/z = 742.5 and 766.5 correspond to PE 18:0/18:2 and PE 18:0/20:4. Please note that in the case of mixtures, negative ions have the considerable advantage that their assignment is less problematic as there is no superposition between the different proton and sodium adducts and differences of the fatty acyl composition. Each PE species results in a single peak. Of course, the differentiation of isobaric ions cannot be made by the provided simple mass spectra (MS1). For instance, the signal at m/z = 742.5 might not only represent PE 18:0/18:2 but also PE 18:1/18:1 that have the same masses. However, this differentiation can be easily made by recording the corresponding “postsource decay” (PSD) mass spectra, i.e., by observing the characteristic fragments generated during the travel of the ions from the source to the detector. Under these conditions the released fatty acyl residues can be observed as negative ions allowing the unequivocal differentiation of the above-mentioned compounds: The anion of stearic acid would be detected at m/z = 283, while oleic acid and linoleic acid are detected at m/z = 281 and 279, respectively (41). Three different PL species (PC, PE, and PI) may, therefore, already be identified from the analysis of the total extract without separation into the individual fractions. Nevertheless, for the

478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504

118 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524

Fuchs et al.

identification of minor components a previous separation is indispensable and can be performed by TLC or HPLC. We have chosen TLC as it is a quite simple and often used method of PL separation. 3. In Fig. 4 a typical TLC plate of the separated liver PLs subsequent to primuline staining is shown. The left lane (4a) corresponds to a standard mixture of known composition and the right lane (4b) corresponds to the liver extract. From the TLC it is evident that the liver extract consists of five lipid classes, namely sphingomyelin (SM), phosphatidylcholine (PC), phosphatidylethanolamine (PE), quite small amounts of phosphatidylinositol (PI), and very small amounts of phosphatidylserine (PS). Please note that the observed slight differences in the migration properties between the standard and the liver extract are caused by differences in the fatty acyl compositions of the individual lipid classes. After visual inspection of the TLC plate, the identified lipid classes were scraped off from the TLC plate, the lipids eluted from the silica gel, and the obtained fractions individually characterized by MALDITOF MS.

Fig. 4. Videoimage of a typical HPTLC plate of a reference PL mixture (a) and the liver extract (b) subsequent to primuline staining (33). 1.69 mg of each PL were applied onto the TLC plate as standard (a) and the total amount of liver lipids was 6 mg (b). The detailed reasons why the PS results in a relatively diffuse spot are yet unknown. LPC lyso-Phosphatidylcholine, PC phosphatidylcholine, PI phosphatidylinositol, PE phosphatidylethanolamine, PS phosphatidylserine, SM sphingomyelin.



Capabilities and Drawbacks of Phospholipid Analysis

119

4. The obtained spectra are shown in Fig. 5. It is evident that the quality of these mass spectra differs significantly. The best spectra regarding the signal-to-noise ratio are stemming from the SM (5a), PC (5b), and PE (5f). The qualities of the PI (5d,5e; negative and positive ion mode, respectively) and particularly the PS fraction (5c) are much worse. This is not only caused by their rather small contribution to the total amounts of lipids, but also by their high polarity and negative charge. Nevertheless it should be noted that the PS could not be identified neither in the positive nor the negative ion spectra of the total liver extract. This is a clear indication that separation is necessary if minor species of a mixture are of interest. All assignments are additionally provided in Table 1. Although the identification of an unknown compound is normally simpler as negative ion because the related adduct pattern is less complicated, the contribution of DHB matrix signals is much more pronounced in the negative ion spectrum (5d) and all the peaks at m/z = 845, 851, and 857 are stemming from the applied DHB matrix (24). This may lead to the suppression of minor species and is also the reason why no convincing negative ion spectrum of PS could be obtained. Please note that in some fractions oxidation of lipids is obvious. For instance the peak at m/z = 863.5 (5d) is caused by PI 18:0/18:1 and the intense peak at m/z = 753.5 corresponds to an oxidation product of that PI at the double bond of the oleoyl residue under generation of the corresponding aldehyde that is schematically shown in Fig. 5. The same mechanism is also valid in the case of the PS, where the peak at m/z = 702.5 is derived from the oxidation of PS 18:0/18:1 at m/z = 812.6. It is also not surprising that the minor fractions are most sensitive to oxidation. Details of this oxidation process are available in (16). Although oxidation is normally an unwanted process, it should be noted that these fragmentation products are sometimes also helpful because they allow the assignments of the positions of the double bonds. Of course, there are still open issues: First, the question of the most suitable matrix for lipid analysis by MALDI-TOF MS is not yet sufficiently answered. We have shown here that the matrix has a tremendous influence on spectral quality as well as the detectabilities of the individual lipid classes. Therefore, it is a strong hope that other still more suitable matrix compounds for the analysis of lipids will be discovered – maybe compounds that allow the detection of all lipid classes even in crude mixtures with higher sensitivity (14), lower matrix background (42), and improved reproducibility (43). Second, some caution is necessary if spectra are recorded from lipid fractions that were purified by means of chromatography (44). There is evidence that lipids differing in their fatty acyl compositions are not released from the TLC plate to the same extent and

525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572

Fig. 5. Typical MALDI-TOF mass spectra of the lipid extract shown in Fig. 3 subsequent to TLC separation into the individual lipid classes. Subsequent to the TLC run, the individual fractions were scraped off and re-eluted from the silica gel of the TLC plate for MALDI-TOF MS. Traces correspond to the following lipid classes: SM (a), PC (b), PS (c), PI (d), (e) and PE (f). All spectra are positive ion spectra, the only negative ion spectrum was obtained from the PI fraction (trace (5d)). Please note that the contribution of matrix peaks (marked by asterisks) is much higher in the negative than the positive ion mode. The most common oxidation pathway of lipids leading to characteristic fragmentation at the position of the double of an oleoyl residue is also shown.



Capabilities and Drawbacks of Phospholipid Analysis

121

saturated or moderately unsaturated lipid “stick” more tightly to the TLC plate. This is confirmed by the slight intensity differences of the PC species in trace (3a) and (5b). Therefore, some lipids may be lost during the extraction process. However, this problem may be overcome by the direct “scanning” of the TLC plate by MALDI-TOF MS without previous extraction of the separated fractions. This method has been recently reported by two different approaches (19, 32) and it is expected that it will develop significantly in the future. There are strong indications that this technique will also be applicable to complex lipids from, e.g., stem cells and other important cell lines (45).

4. Notes

573 574 575 576 577 578 579 580 581 582 583

584

1. Of course all commercially available MALDI-TOF devices can be used independent of the manufacturer. If product or company names are given here, this only means that the indicated products were used for performing the presented experiments. This is sometimes important as different nomenclatures may be used. For instance, “delayed extraction” and “energy lag focusing” are synonyms and mean the same. However, MALDI devices that are exclusively capable of recording linear mode spectra (23), are less suitable for lipid analysis because resolution as well as mass accuracy is reduced in comparison to reflector mode spectra. Isotopic resolution is, however, often important because this spectral feature enables the estimation of the molecular formula – at least in the case of smaller molecules.

585

2. Under the conditions described here, the matrix and the analyte give a relatively “thick” layer of hundreds of mm on the target. Only the upper layers are actually relevant for laser ablation. Therefore, the MALDI target material is less important than by using very thin layers. We are using by default gold-coated targets as they are expected to have a lower content of catalytically active transition metals than stainless steel (16).

598

3. Normally, the use of plastic pipettes (as well as other plastic material) is strongly discouraged because CHCl3 is a rather aggressive solvent that releases impurities as plasticers from the plastic material that may interfere with the signals of the analyte of interest (23). However according to our experience “grey” original Eppendorf pipette tips (up to a volume of 20 ml) may be used without problems. It is, however, recommended to check the potential contribution of impurities by using a sample of known composition and known concentration. Unfortunately, although CHCl 3 is

605

586 587 588 589 590 591 592 593 594 595 596 597 599 600 601 602 603 604 606 607 608 609 610 611 612 613 614 615

122 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661

Fuchs et al.

a hazardous solvent, all attempts to replace it, for instance, by hexane and isopropanol, were not very successful so far in lipid research. 4. Different solvents are normally required to solubilize matrix and lipid. In the majority of cases, the lipid stock solutions will be prepared in CHCl3 or CHCl3/CH3OH mixtures. Unfortunately, the most common MALDI matrix – DHB – is nearly insoluble in CHCl3, but highly soluble in CH3OH. Due to the different volatilities of both solvents (the boiling point of CHCl3 is 61.2°C and that of CH3OH 64.5°C), the lipid will crystallize prior to the DHB resulting in rather inhomogeneous cocrystals. This problem can be minimized by drying the native matrix/sample mixtures rapidly under a warm stream of air. The extent of lipid oxidation induced by air-drying of the sample is practically negligible. 5. Special attention should be paid to the salt content of the matrix as well as the solvents. Changes of the salt content may lead to changes of the peak patterns and affect the ratio between H+ and Na+ adducts. Additionally, the peaks stemming from the DHB matrix are also influenced by changes of the salt content (37). Please note that a spectrum of DHB crystallized from CH3OH differs from that in the presence of salts. 6. All used solvents should be of highest quality! Due to the high sensitivity of MS even very minor impurities are detected: The detection of analytes in small amounts is often a minor problem than NOT detecting impurities stemming from the solvents or the used reagents. 7. It is not implied that under the used experimental conditions all lipids are completely extracted: Some lipids may also stick to the proteins that precipitate at the interphase between the aqueous and the organic layer. If protein-rich samples are investigated, higher ionic strength, i.e., a high salt concentration is recommended in order to reduce the loss of lipids due to the binding to the protein. Complete extraction of lipids from biological tissues is a science of its own and the extraction must be optimized in dependence on the tissue or body fluid of interest. Under the recommended conditions of centrifugation using moderate g-values, special centrifuge glasses are not absolutely required, but common test tubes may also be used. 8. The addition of salt (NaCl) is recommended in order to enhance the dissociation of the lipids from the silica gel. The use of distilled water instead of NaCl solution may lead to the loss of lipids. This particularly holds for negatively charged lipids that are otherwise lost.



Capabilities and Drawbacks of Phospholipid Analysis

123

Abbreviations and Acronyms APCI Atmospheric Pressure Chemical Ionization DE Delayed Extraction DHB 2,5-Dihydroxybenzoic Acid EI Electron Impact ESI Electrospray Ionisation HPLC High-Performance Liquid Chromatography IR Infrared LPC Lyso-Phosphatidylcholine MALDI Matrix-Assisted Laser Desorption and Ionization MS Mass Spectrometry m/z mass over charge PC Phosphatidylcholine PE Phosphatidylethanolamine PG Phosphatidylglycerol PI Phosphatidylinositol PL Phospholipid PNA Para-Nitroaniline PO Palmitoyl-oleoylPS Phosphatidylserine PSD Post Source Decay rpm rotations per minute S/N Signal to Noise SM Sphingomyelin sn Stereospecific Numbering TFA Trifluoroacetic Acid TLC Thin-Layer Chromatography TOF Time-of-Flight UV Ultraviolet

Acknowledgments

662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691

692

This work was supported by the German Research Council (DFG Schi 476/5–1 and FU 771/1-1) and the Federal Ministry of Education and Research (Grant BMBF 0313836). The kind and helpful advice of Dr. Suckau and Dr. Schürenberg (Bruker Daltonics, Bremen) is particularly gratefully acknowledged.

699

References 1. Christie, W. W. (2003) Lipid Analysis, Oily Press, Bridgwater. 2. Wymann, M. P. and Schneiter, R. (2008) Lipid signalling in disease. Nat. Rev. Mol. Cell. Biol. 9, 162–176. 3. Matsumoto, T., Kobayashi, T., and Kamata, K. (2007) Role of lysophosphatidylcholine (LPC)

693 694 695 696 697 698

in atherosclerosis. Curr. Med. Chem. 14, 3209– 3220. 4. Fuchs, B., Schiller, J., Wagner, U., Häntzschel, H. and Arnold, K. (2005) The phosphatidylcholine/lysophosphatidylcholine ratio in human plasma is an indicator of the severity of rheumatoid arthritis: investigations by 31P

700 701 702 703 704 705 706 707

124 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766

Fuchs et al.

NMR and MALDI-TOF MS. Clin. Biochem. 38, 925–933. 5. Wenk, M.R. (2005) The emerging field of lipidomics. Nat. Rev. Drug Discov. 4, 594–610. 6. Han, X. (2007) An update on lipidomics: progress and application in biomarker and drug development. Curr. Opin. Mol. Ther. 9, 586–591. 7. Fahy, E., Subramaniam, S., Brown, H. A., Glass, C. K., Merrill, A. H. Jr, Murphy, R. C., Raetz, C. R., Russell, D. W., Seyama, Y., Shaw, W., Shimizu, T., Spener, F., van Meer, G., VanNieuwenhze, M. S., White, S. H., Witztum, J. L. and Dennis, E. A. (2005) A comprehensive classification system for lipids. J. Lipid Res. 46, 839–861. 8. Peterson, B. L. and Cummings, B. S. (2006) A review of chromatographic methods for the assessment of phospholipids in biological samples. Biomed. Chromatogr. 20, 227–243. 9. Patton, G. M. and Robins, S. J. (1998) Separation and quantitation of phospholipid classes by HPLC. Methods Mol. Biol. 110, 193–215. 10. Touchstone, J. C. (1995) Thin-layer chromatographic procedures for lipid separation. J. Chromatogr. B. 671, 169–195. 11. Wolf, C. and Quinn, P. J. (2008) Lipidomics: practical aspects and applications. Prog. Lipid Res. 47, 15–36. 12. Byrdwell, W. C. (2001) Atmospheric pressure chemical ionization mass spectrometry for analysis of lipids. Lipids 36, 327–346. 13. Pulfer, M. and Murphy, R. C. (2003) Electrospray mass spectrometry of phospholipids. Mass Spectrom. Rev. 22, 332–364. 14. Northen, T. R., Yanes, O., Northen, M. T., Marrinucci, D., Uritboonthai, W., Apon, J., Golledge, S. L., Nordström, A., and Siuzdak, G. (2007) Clathrate nanostructures for mass spectrometry. Nature 449, 1033–1036. 15. Schiller, J., Süß, R., Fuchs, B., Müller, M., Zschörnig, O. and Arnold, K. (2007) MALDI-TOF MS in lipidomics. Front. Biosci. 12, 2568–2579. 16. Schiller, J., Süß, R., Arnhold, J., Fuchs, B., Leßig, J., Müller, M., Petkovic, M., Spalteholz, H., Zschörnig, O. and Arnold, K. (2004) Matrix-assisted laser desorption and ionization time-of-flight (MALDI-TOF) mass spectrometry in lipid and phospholipid research. Prog. Lipid Res. 43, 449–488. 17. Roberts, L. D., McCombie, G., Titmana, C. M. and Griffin J. L. (2008) A matter of fat: An introduction to lipidomic profiling methods. J. Chromatogr. B. 871, 174–181.

18. Hillenkamp, F. and Peter-Katalinic, J. (2007) MALDI MS – A Practical Guide to Instrumentation. Methods and Application. Wiley, Weinheim. 19. Rohlfing, A., Müthing, J., Pohlentz, G., Distler, U., Peter-Katalinić, J., Berkenkamp, S. and Dreisewerd, K. (2007) IR-MALDI-MS analysis of HPTLC-separated phospholipid mixtures directly from the TLC plate. Anal. Chem. 79, 5793–5808. 20. Cohen, L. H. and Gusev, A. I. (2002) Small molecule analysis by MALDI mass spectrometry. Anal. Bioanal. Chem. 373, 571–586. 21. Schiller, J., Arnhold, J., Benard, S., Müller, M., Reichl, S. and Arnold, K. (1999) Lipid analysis by matrix-assisted laser desorption and ionization mass spectrometry: A methodological approach. Anal.Biochem. 267, 46–56. 22. McDonnell, L. A. and Heeren, R. M. (2007) Imaging mass spectrometry. Mass Spectrom. Rev. 26, 606–643. 23. Fuchs, B., Arnold, K. and Schiller, J. (2008) Mass spectrometry of biological molecules. In: Encyclopedia of Analytical Chemistry (Meyers, R. A., ed.), Wiley, Chichester, pp. 1–39. 24. Schiller, J., Süß, R., Fuchs, B., Müller, M., Petkovic, M., Zschörnig, O. and Waschipky, H. (2007) The suitability of different DHB isomers as matrices for the MALDI-TOF MS analysis of phospholipids: which isomer for what purpose? Eur. Biophys. J. 36, 517–527. 25. Estrada, R. and Yappert, M. C. (2004) Alternative approaches for the detection of various phospholipid classes by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. J. Mass Spectrom. 39, 412–422. 26. Gellermann, G. P., Appel, T. R., Davies, P. and Diekmann, S. (2006) Paired helical filaments contain small amounts of cholesterol, phosphatidylcholine and sphingolipids. Biol. Chem. 387, 1267–1274. 27. Murphy, R.C. (2002) Mass spectrometry of phospholipids: Tables of molecular and product ions. Illuminati Press, Denver. 28. Fuchs, B., Müller, K., Göritz, F., Blottner, S. and Schiller, J. (2007) Characteristic oxidation products of choline plasmalogens are detectable in cattle and roe deer spermatozoa by MALDI-TOF mass spectrometry. Lipids 42, 991–998. 29. Harvey, D. J. (1995) Matrix-assisted laser desorption/ionization mass spectrometry of phospholipids. J. Mass Spectrom. 30, 1333–1346. 30. Bligh, E. G. and Dyer, W. J. (1959) A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 3, 911–917.



Capabilities and Drawbacks of Phospholipid Analysis

31. Bartlett, G. R. (1959) Phosphorus assay in column chromatography. J. Biol. Chem. 234, 466–468. 32. Fuchs, B., Schiller, J., Süß, R., Schürenberg, M. and Suckau, D. (2007) A direct and simple method of coupling matrix-assisted laser desorption and ionization time-of-flight mass spectrometry (MALDI-TOF MS) to thin-layer chromatography (TLC) for the analysis of phospholipids from egg yolk. Anal. Bioanal. Chem. 389, 827–834. 33. White, T., Bursten, S., Frederighi, D., Lewis, R. A. and Nudelman, E. (1998) High-resolution separation and quantification of neutral lipid and phospholipid species in mammalian cells and sera by multi-one-dimensional thin-layer chromatography. Anal. Biochem. 10, 109–117. 34. Zschörnig, O., Richter, V., Rassoul, F., Süß, R., Arnold, K. and Schiller, J. (2006) Analysis of human blood plasma by MALDI-TOF MS – Evaluation of critical parameters. Anal. Lett. 39, 1101–1113. 35. Petkovic, M., Schiller, J., Müller, J., Müller, M., Arnold, K. and Arnhold, J. (2001) The signalto-noise ratio as the measure for the quantification of lysophospholipids by matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry. Analyst 126, 1042–1050. 36. Müller, M., Schiller, J., Petkovic, M., Oehrl, W., Heinze, R., Wetzker, R., Arnold, K. and Arnhold, J. (2001) Limits for the detection of (poly-)phosphoinositides by matrix-assisted laser desorption and ionization time-of-flight mass spectrometry (MALDI-TOF MS). Chem. Phys. Lipids 110, 151–164. 37. Petkovic, M., Schiller, J., Müller, M., Benard, S., Reichl, S., Arnold, K. and Arnhold, J. (2001) Detection of individual phospholipids in lipid mixtures by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry: phosphatidylcholine prevents the detection of further species. Anal. Biochem. 289, 202–216. 38. Al-Saad, K. A., Zabrouskov, V., Siems, W. F., Knowles, N. R., Hannan, R. M. and Hill, H.

125

H. Jr. (2003) Matrix-assisted laser desorption/ ionization time-of-flight mass spectrometry of lipids: ionization and prompt fragmentation patterns. Rapid Commun. Mass Spectrom. 17, 87–96. 39. Gidden, J., Liyanage, R., Durham, B. and Lay, J. O. Jr. (2007) Reducing fragmentation observed in the matrix-assisted laser desorption/ionization time-of-flight mass spectrometric analysis of triacylglycerols in vegetable oils. Rapid Commun. Mass Spectrom. 21, 1951–1957. 40. Johanson, R. A., Buccafusca, R., Quong, J. N., Shaw, M. A. and Berry, G. T. (2007) Phosphatidylcholine removal from brain lipid extracts expands lipid detection and enhances phosphoinositide quantification by matrixassisted laser desorption/ionization timeof-flight (MALDI-TOF) mass spectrometry. Anal. Biochem. 362, 155–167. 41. Fuchs, B., Schober, C., Richter, G., Süß, R. and Schiller, J. (2007) MALDI-TOF MS of phosphatidylethanolamines: different adducts cause different post source decay (PSD) fragment ion spectra. J. Biochem. Biophys. Methods 70, 689–692. 42. Guo, Z. and He, L. (2007) A binary matrix for background suppression in MALDI-MS of small molecules. Anal. Bioanal. Chem. 387, 1939–1944. 43. Li, Y.L., Gross, M.L. and Hsu, F.F. (2005) Ionic-liquid matrices for improved analysis of phospholipids by MALDI-TOF mass spectrometry. J. Am. Soc. Mass Spectrom. 16, 679–682. 44. DeLong, C. J., Baker, P. R., Samuel, M., Cui, Z. and Thomas, M. J. (2001) Molecular species composition of rat liver phospholipids by ESI-MS/MS: the effect of chromatography. J. Lipid Res. 42, 1959–1968. 45. Fuchs, B., Schiller, J., Süß, R., Zscharnack, M., Bader, A., Müller, P., Schürenberg, M., Becker, M., Suckau, D. (2008) Analysis of stem cell lipids by offline HPTLC-MALDI-TOF MS. Anal. Bioanal. Chem. 392, 849–860.

767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813

sdfsdf

Chapter 7 Lipidomics of the Red Cell in Diagnosis of Human Disorders

1

2 3

Peter J. Quinn, D. Rainteau, and C. Wolf

4

Summary

5

Applications of tandem mass spectrometry in the field of lipid clinical chemistry are considered. Haemato­ logical and biochemical advantages are presented favoring the choice of red blood cell membranes as a starting material in a wide variety of biomedical fields. Practical considerations are discussed with respect to methods of sampling, storage, and lipid extraction of red blood cells. The chapter describes the capabilities of a direct infusion of raw lipid extracts in the electro-spray ionization source compared with the more sophisticated method of high-performance liquid chromatography coupled with hybrid tandem mass spectrometry. Both methods have been evaluated and have been shown to be suitable for diagnosis and/or monitoring for a variety of human disorders.

6 7 8 9 10 11 12 13

Key words: Lipidomics, Diagnosis, Erythrocyte, Membrane, Essential fatty acid deficit, Peroxisomal defects, Plasmalogen, Oxidized phosphatidylethanolamines

14 15

1. Introduction

16

The human red blood cell (RBC) membrane is an object pathological investigation because it can provide a convenient and reliable indica­ tion of a range of pathological conditions. This work is devoted to the description of the development of routine methodologies for lipid profiling of the human red cell membrane. One develop­ ment is the routine profiling of the fatty acid substituents of the complex lipids of the erythrocyte membrane. Due to the rigorous and precise homeostatic regulatory mechanisms that operate in the membrane it is possible, for example, to detect essential fatty acid (EFA) deficit in children (i.e., exocrine pancreatic insufficiency of mucoviscidosis) as well as in adults (i.e. extensive enterectomy in advanced inflammatory bowel diseases). Other more subtle Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_7, © Humana Press, a part of Springer Science + Business Media, LLC 2009

127

17 18 19 20 21 22 23 24 25 26 27 28

128

Quinn, Rainteau, and Wolf

alterations in the molecular species of RBC phospholipids are also investigated for long-term nutritional consequences of sup­ plemented diets or biosynthetic defects of ether–lipids associated with peroxisomal enzyme blockade. The biosynthetic defects of ether–lipids together with very long chain fatty acid accumula­ tion are needed for the diagnosis of 12 out of 17 neurological conditions associated with one of the multiple forms of peroxisomal deficit (1). The complexity of the clinical diagnosis of peroxi­ somal disorders in children is an incentive for more widespread applications of lipidomic analysis. The analysis is also recommended to monitor the outcomes of diet and supplementation treatments because EFA profiling of the RBC membrane is a much more reliable index of deficit than profiling fatty acids of serum lipoproteins. This is because of a relatively slow turnover of fatty acyl components of complex membrane lipids as compared to renewal of neutral lipids and phos­ pholipids circulating in serum. Serum lipolytic enzymes (secretory PLA2, Lecithin-Cholesterol Acyl Transferase, Lipoprotein Lipase) are relatively active compared with low activities of phospholipase or acyltransferase in human erythrocyte membranes and the cells remain in circulation with a normal lifetime of 120 days. As a result of the relatively stable RBC fatty acid composition variations of a few percent of fatty acid molar ratios can be a significant indica­ tor of EFA deficit. For example, normal proportions of 20:4n-6 is 11–14% but values less than 10% may be considered as a deficit. The original analytic protocols of fatty acid profiling were based on analysis of hydrolyzed lipids and production of volatile methyl ester derivatives (FAME) using gas–liquid chromatography. The introduction of liquid chromatography coupled with mass spec­ trometry allows the identification of molecular species of mem­ brane phospholipids with relatively slow turnover rates thereby improving precision in indexing the EFA profile. Phosphatidylcho­ lines, for example, are subject to a different turnover rate compared with phosphatidylethanolamines populating the inner leaflet of the RBC membranes which normally are not accessible to circulating enzymes. However, phosphatidylethanolamines are subjected to enzymes from the inner side of the membrane. In addition phos­ phatidylcholine species are comprised of lower proportions of EFA than phosphatidylethanolamines which represent a reservoir enriched in EFA which expands under conditions of nutritional supplementation. Phosphatidyserines and phosphatidylinositols are enriched in the, so called, “distal” EFA such as 20:4n-6 and 22:6n3. These particular EFA are generated as the end products of bio­ synthetic pathways in the liver. The pathways comprised of acyl chain elongation and unsaturation lead to the interconversion of shorter chain and less unsaturated precursor EFA, such as 18:2n-6 and 18:3n-3, respectively. It is anticipated that future develop­ ments will enable distinction between nutritional deficits of EFAs from those derived by hepatic interconversion. Besides global liver



Lipidomics of the Red Cell in Diognosis of Human Disorders

129

f­ailure it is expected that specific elongase/desaturase steps are crit­ ically impaired in pathology such as insulin resistance and diabetes, obesity, and hepatosteatosis (2). New therapeutic developments are aimed to remedy these defects. The measurement of kinetics of refilling with EFA into particular molecular species after enriched diets may also provide useful clinical information. Other applica­ tions include immunomodulation correlated with a shift in the bal­ ance of n-3/n-6 EFA. The assumption that inflammation may be prompted by an inappropriate eicosanoid synthesis in which there is obvious medical interest has not been supported yet by total FA methyl ester profiling studies. However, it is anticipated that with application of recent lipidomics methods described in this Chap­ ter evidence of a significant linkage will be established between abnormalities in the FA composition of a few molecular species of precursor phospholipids and the inflammatory condition. Current lipidomic methods using mass spectrometry are able to resolve the molecular species composition of the five major phosphol­ ipid classes present in human erythrocyte membranes: phosphatidyl­ cholines (PC), sphingomyelins (SM), phosphatidylethanolamines (PE), phosphatidylserines (PS), and phosphatidylinositols (PI). Cho­ lesterol and detailed glycolipid composition can also be determined but discussion of this is outside the scope of this presentation.

2. Materials Lipidomics studies of patient erythrocyte membranes were conducted at CHU Saint Antoine. The MS facility at the hospital center includes two complementary tandem MS2 equipments (a) a hybrid tandem QTrap 2000 (Applied Biosystems/MDS SCIEX) is coupled with high-pressure liquid chromatography (HPLC Agilent series 1100). (b) The triple quadrupole (TQ) API3000 (Applied Biosystems/MDS SCIEX) system is presently used for direct infusion (Harvard Apparatus syringe pump 11 Plus). Mass spectra are quantified after translation of the propri­ etary file format.wiff (Analyst, Applied Biosystems) to .cdf format using the translation module included in the software. Then mass spectra are computed for peak identification and deisotopisation using the software LIMSA (3).

3. Methods 3.1. The Use of RBC Membranes for Medical Investigations

As a tissue indicator of metabolic or nutritional disturbance or disease states, the human erythrocyte offers a number of advan­ tages. The cells are easily obtained without inordinate invasion

130

Quinn, Rainteau, and Wolf

and can be drawn at suitable intervals for monitoring purposes. They are nondividing cells with a single, relatively stable cell membrane. This section details the processes of haematopoiesis and the origin and turnover of the constituent lipids of the mem­ brane. This is followed by a consideration of sampling, storage, and processing of lipids for lipidomic analysis. 3.1.1. Haematopoiesis and RBC Lifetime

Haematopoietic stem cells of the bone marrow give rise to all the cellular components of blood. The pool of these multipotent cells is maintained by retention of multipotency by a propor­ tion of the daughter cells. Differentiation of the stem cells takes place in a stepwise manner in which changes in gene expression that place constraints on cell destiny move the cells closer to toward a mature erythrocyte. These changes are reflected in the proteins detected on the cell surface. The differentiation of erythrocytes from myeloid progenitor cells is augmented by hematopoietic growth factors and principally, erythropoi­ etin, produced in the kidneys. Prior to and immediately after leaving the bone marrow, the differentiating erythrocytes pass through a reticulocyte stage. Reticulocytes represent about 1% of circulating red blood cells. This proportion of reticulocytes is commonly determined to be used as a measurement of RBC turnover rate. In turn, the percentage is an indication of RBC lifetime with consequences on membranogenesis and the lipid incorporation rate at the time of measurement. Erythrocytes develop from committed stem cells through reticulocytes to mature erythrocytes in about 7 days and survive in circulation for a total of about 120 days. Ageing erythrocytes become damaged due to irreversible changes in the cell membrane, some of them related to mem­ brane lipids such as the outer exposure of phosphatidylserines and phosphatidylethanolamines, which are recognized by macrophages in the spleen, bone marrow, and liver and are removed from the circulation by phagocytosis. Much of the important breakdown products (iron) are then recirculated in the body but little is known of recirculation of lipids and EFA. Some physiological conditions (e.g., gender and altitude acclimation) influence turnover of RBCs and this, in turn may interfere with the lipid composition of the membranes. These changes have not been yet detailed by the recent lipidomic methods described below. Adult humans have approximately 2–3 × 1013 red blood cells in circulation at any given time (women have about 4–5 × 106 erythrocytes/mL of blood and men about 5–6 × 106). Eryth­ rocytes from adults and children with reduced vitamin E intake and with low vitamin E serum levels that are relatively sensitive to oxidants in vitro have a reduced lifetime. The lifetime of eryth­ rocytes in premature infants is also shorter than those in normal infants (4).



Lipidomics of the Red Cell in Diognosis of Human Disorders

131

Polycythemias are characterized by an increased concentra­ tion of circulating erythrocytes, some such as polycythemia vera is an acquired stem cell disease and others are congenital and are caused by mutations of the Erythropoitin-receptor gene, haemo­ globin variants, 2,3-bisphosphoglycerate mutase deficiency, or by disturbances in renal oxygen sensing. Acquired polycythemias can occur secondary to hypoxia at high altitudes, or primarily through acquired mutations in the EPO-receptor signaling system (JAK2 mutations). Alternatively they may be caused by pulmonary or renal diseases which perturb fatty acid composition of RBC with numerous intricate causes. Because they are relatively long lived in the circulation, their membrane lipid composition is more representative of the longer term lipid homeostatic situation than is the lipids of the plasma. The latter varies in lipid composition according to short-term dietary and physiological circumstances. 3.1.2. Regulation of Phospholipid Synthesis During Membrane Biogenesis

The cell membrane of the mature erythrocyte is not synthesized de novo but it is transformed from its progenitor cells by a proc­ ess of membrane differentiation. The primary site of synthesis of membrane lipids and proteins is the endoplasmic reticulum where most of the enzymes required for these tasks are located. The biosynthesis of membrane phospholipids and their incorpo­ ration into the membrane takes place on the cytosolic monolayer of the progenitor endoplasmic reticulum. Phospholipid flippases are needed to translocate newly synthesized phospholipids to the opposite leaflet of the membrane (luminal in the progenitor to become extracellular in the mature RBC). Phosphatidylcho­ line transferase operates to concentrate molecular species of this phospholipid class to the protoplasmic leaflet of the membrane but there are no transferase specific for PE, PS, and PI, so these phospholipid classes are predominant in the cytoplasmic leaflet (to become intracellular in the mature RBC). This asymmetry is preserved when transition vesicles derived from the endoplas­ mic reticulum fuse with membranes of other organelles of the endomembrane system including, ultimately, the plasma membrane. In addition, soluble lipid exchange proteins which are specific for particular membrane lipids, such as glucosylceramides, are present which translocate these lipids from the site of synthesis to the cytoplasmic surface of the plasma membrane. During passage of membrane through the endomembrane system of precursor RBC cell, the lipid composition changes as it is transformed into different morphologically distinct membrane. This process involves retailoring the molecular species of lipids that comprises the bilayer matrix. This process can involve both complete turnover of the lipid where the lipid is removed from the membrane and replaced by another molecular species of the same lipid class and partial turnover, when only a component of

132

Quinn, Rainteau, and Wolf

the lipid is exchanged for another. Complete turnover of lipids is mediated by phospholipase C-type enzymes in which the products of hydrolysis no longer possess an amphipathic character (diacylg­ lycerols and water-soluble phosphate esters). Partial turnover is of two types. The first is where the fatty acyl chains are exchanged one for another to remodel the molecular species of phospholipid. These types of reaction involve phospholipase A-type enzymes with an intermediate lysophospholipid product that is strongly amphipathic in character. The hydrolysis of fatty acids and reacyla­ tion of the lysophospholipid is known as the Lands cycle. The sec­ ond type of partial turnover is catalyzed by phospholipase D-type enzymes in which the intermediate product is phosphatidic acid. The reaction involves an exchange of bases on phosphatidic acid and its significance is concerned mainly with the biosynthesis of phosphatidylserine from phosphatidylethanolamine. The partial turnover of molecular species of phosphatidylinositols may also be achieved by this mechanism. During differentiation of reticulocytes into mature eryth­ rocytes fatty acids provided from diet and presented from the digestive system and/or after interconversion in the liver are incorporated into membrane lipids. This process takes place during the many steps of cellular differentiation from the early erythroid progenitors (burst-forming-units-erythroid) to late erythroid progenitors (colony-forming units-erythroid) and to morphologically recognizable erythroid precursors. Ultimately the endomembrane system including the nucleus is lost and the mature erythrocyte is no longer in possession of the complete machinery for producing membrane lipids. The lipid composition of the erythrocyte membrane is thereon dependent on removal of lipids by metabolic turnover or the limited remodeling of the existing membrane lipids using substrates and circulating enzymes that are provided from the plasma (secretory PLA2, LCAT, etc.). 3.1.3. Incorporation of Fatty Acids into Membrane Phospholipids

The key building block of membrane phospholipids is diacylg­ lycerol. The fatty acids of the glycerides reflect, to a large extent, the nutritional and dietary circumstances of the patient and are modified by pathological conditions. There are two principal pathways for de novo phospholipid biosynthesis. The fatty acid of triacylglycerol at the C3 position is hydrolyzed to form 1, 2-diacylglycerol. This may react with CDP derivatives of choline or ethanolamine to form PC and PE, respectively. Alternatively, 1, 2-diacylglycerol may be phosphorylated at the sn-3 position by the action of diacylglycerol kinase to form phosphatidic acid. Phosphatidic acid, in turn, reacts with CTP to form CDP-dig­ lyceride. The product can react directly with serine or inositol for PS and PI, respectively. Thus the origin of the fatty acids origi­ nates from the metabolic stores of triacylglycerol. In the case of sphingolipid biosynthesis from sphingosine bases, the fatty acids



Lipidomics of the Red Cell in Diognosis of Human Disorders

133

amide bonded to the precursor to form ceramide are distinctive from those acylated to the phospholipid molecular species. These particularly long-chain, more saturated fatty acids are preferentially utilized for biosynthesis of the sphingolipids. 3.1.4. Remodeling of Phospholipids by a Deacylation/ Reacylation Cycle

As indicated above, remodeling of membrane phospholipids via acyl exchange reactions in the Lands cycle is an active pathway in most tissues. The molecular species of membrane phospholipids is a reflection of the nature of the fatty acid substrates and the physical signals generated by the bilayer structure which govern the type of fatty acids that can be incorporated. The Lands cycle is highly active and plays an important role in membrane differ­ entiation. Operation of the cycle requires phospholipase A2/CoA ligase/acyltransferase enzyme components. The activity of these enzymes in mature erythrocytes, however, is very low compared with those in other tissues such as the liver. Moreover, the rela­ tively high concentrations of serum albumin (>30 g/L) strongly bind lysoderivatives formed in the RBC membranes. This means that in erythrocytes the turnover of molecular species of phos­ pholipids is relatively slow so that reliable long-term homeostasis can be characterized from a lipidomic analysis of the erythrocyte membrane. Recently the apoptosis-associated changes in the glyc­ erophospholipid composition of hematopoietic progenitor cells have been monitored by lipidomic methods (MALDI-TOF and 31 P-NMR) (5). The study has shown that withdrawal of growth factor induces an increase in the ratio of ether-linked glycero­ phospholipids to diacyl-glycerophospholipids during apoptosis as well as the relative decrease in the membrane of diacyl-phosphati­ dylcholine (PC) levels.

3.1.5. Membrane Lipids in Human Hemoglobinopathies

The lipid molecular species composition of the red blood cell is preserved within relatively narrow limits throughout the lifetime of the cell. Retailoring of the constituent lipids is accomplished by uptake of fatty acids from the serum which, in turn, reflects the intake of lipids in the diet. This intake can be highly variable so that the process of lipid turnover must be selective in that the fatty acids incorporated into complex membrane lipids are tightly regulated. The biophysical sensors of membrane lipid composi­ tion are, as yet, unknown but modulation of the processes of acyl chain turnover must take place. Only a few members of the enzyme pathways involved have so far been identified, and little is known about how regulation of catalysis is achieved through interaction of the enzymes with their lipid environment that would clarify the mechanisms that govern this selectivity. Never­ theless, changes in membrane lipids provide useful indicators of a range of hemoglobinopathies (6). It may be expected that phospholipid turnover and repair is higher in hemoglobinopathies because damage to the mem­

134

Quinn, Rainteau, and Wolf

brane resulting from free radical reactions is often a feature of these conditions. A number of reports indicate evidence of lipid oxidation in RBCs from sickle cell or thalassemia patients, ((7, 8)) suggesting that phospholipid repair is not efficient enough to maintain the proper lipid molecular species composition in these cells. Analysis of the lipid molecular species of density fraction­ ated sickle cells reveals a complex picture. Fractions of greater density which are comprised of more saturated molecular species of phospholipids have more variable proportions of unsaturated phospholipids comparing one patient with another. This obser­ vation infers that the processes responsible for turnover and repair of membrane phospholipids differ from one cell to another because all cells share the same serum substrate pool. It is note­ worthy that when the exchange of phospholipids between RBCs is facilitated by lipid exchange, the densest fractions are those that undergo the greatest change of density (9). Erythrocytes are unable to replace proteins that are irrevers­ ibly damaged by reactive oxygen species. Thus, incorporation of fatty acids in erythrocyte membrane phospholipids is affected when cell suspensions are oxidatively stressed in  vitro. Since complex pathways are employed in lipid repair, turnover dam­ age to any one member of these pathways has repercussions in membrane lipid composition. The relative sensitivity of the indi­ vidual proteins toward oxidant stress must vary but not much is known of where the vulnerabilities occur. Other than direct oxidative damage of these proteins, alterations in the ionic balance of the cytosol may also play a role. The recently identified acylCoA:Lysophosphatidylcholine acyltransferase (LPCAT) family, for example, which exhibits an EF-hand motif at the C-terminus, was shown to be modulated by calcium and magnesium (10). Since both cytosolic calcium and oxidant stress are altered in sub­ populations of sickle RBCs, this suggests that LPCAT may act differently in subpopulations of sickle RBCs, leading to insuf­ ficient or miss repair of membrane lipids. Overall, oxidant stress and alterations in the cytosol of erythrocytes from patients with hemoglobinopathies will lead to the inability to maintain a proper lipid composition, which in turn leads to alterations in membrane function, including ion and water transport across the bilayer and altered red cell density distributions. The alteration in the sickle cell membrane lipids may also be understood as a consequence of the shorter erythrocyte lifespan and accelerated membranogenesis in the progenitor cells. 3.2. Practical Considerations for Lipidomic Analysis of Human Erythrocytes

Blood are the most convenient liquid “tissue” for analysis because obtaining samples is less invasive and samples can be taken at frequent intervals for monitoring purposes. The physiological state of the subject must be standardized because blood lipids vary accordingly between subpopulations. Erythrocytes must be



Lipidomics of the Red Cell in Diognosis of Human Disorders

135

i­solated from other blood cells and serum lipoproteins by differen­ tial centrifugation methods under conditions where lipid changes to the erythrocyte membrane are minimized. Once isolated, the erythrocytes can be stored before lipid extraction and analysis or alternatively the extraction of lipids may proceed immediately on isolation of the cells and the lipid extract stored until examined. This section deals with the practical aspects of sample handling prior to lipidomic analysis. 3.2.1. Blood Sample Collection and Erythrocyte Storage

The first question to be decided is what volume of blood needs to be collected and under what conditions to preserve the integrity of blood cells? Drawing of a 1 ml sample of blood is likely to be sufficient for lipidomic analysis. This volume can be reduced to 0.2 ml in new born babies without compromising the analysis. It is advisable to adapt the blood volume and the tube capacity to minimize the air in the container in contact with the sample. Collection of the sample into a vacutainer on a chelator like EDTA is recommended to reduce the availability of metal cations (Fe2+, Cu2+, etc.) activating lipid peroxidation. This prevents the chance of reactive oxygen species oxidizing unsaturated molecular species of lipid. The blood sample should not be shaken vigorously to minimize the damage to the erythrocytes; vesicles are known to be released from stored erythrocyte membranes after prolonged shaking and can be recovered in the plasma after centrifugation. Because the lipid composition of the released membranes is very different from the plasma the plasma composition is corrupted by minor contamination due to the presence of fragmented membrane vesicles. Likewise, the composition of membrane lipids can be compromised by contamination by lipoproteins adsorbed from plasma. For this reason it is good practice to “wash” the pelleted RBC with five volumes of isotonic sodium chloride, especially if the sample is of a small volume (under 1 ml). The storage conditions before analysis are important. Freezing whole blood causes extensive membrane fragmentation and the resulting plasma/RBC contamination after hemolysis. Freezing RBC only after separation of plasma from the blood cells is rec­ ommended in order to avoid exchange protein and enzyme activ­ ities which alter RBC membrane lipids. The activity of the serum enzymes is slow in EDTA-containing samples and a time of 2–4 h for transportation after drawing the sample is reasonable. If samples are to be stored for longer periods of time this should be at −20°C rather than −70°/−80°C. Storage at 4°C is not suffi­ cient to inhibit all interfering enzyme activities: the sharp decrease in the unesterified membrane cholesterol content under the influ­ ence of the enzyme L.C.A.T. upon storage for 21 days at −4°C under the standard blood bank conditions is an example. Unlike storage of samples for proteomic studies there is no advantage in storage at the lower temperatures in maintaining lipid integrity.

136

Quinn, Rainteau, and Wolf

The major cause of lipid deterioration is peroxidation and oxy­ gen gas is more soluble at low temperatures. Also oxygen-derived radicals are more stable at low temperatures, so very cold tempera­ tures are of no advantage. The oxygen should be chased from the sample by gassing with a dense inert gas like argon; nitrogen is a less satisfactory substitute because, being less dense, it leaks out of the tube unless the container has been flame-sealed. Nitrogen is also less effective in displacing oxygen than argon. An example of LCMS2 method to trace the oxidized phosphatidylethanolamines is given in the last paragraph of this chapter. 3.2.2. Lipid Extraction

Lipidomics studies often refer to studies of crude lipid extracts inferring that there is no requirement for the complete separation of protein, sugar, and lipid classes in the sample subject to analysis. Tandem mass spectrometry has considerably simplified the preanalytical steps as compared to relatively “low tech” procedures such as thin-layer chromatography which are more demanding in terms of purity. This is explained by the high specificity of MS and relatively weak influence of the matrix for monitoring the characteristic transition of a parent molecular ion to a fragment prod­ uct (tandem MS). Electrospray (ESI) with intrasource separation has led to strategies using 2D mass spectrometry in, so called, shotgun lipidomics and quantitation of cellular lipidomes directly from “crude” extracts of biological samples (11). The intrasource separation procedure is based on the activity of an external electric field to induce separation of cations from negative ions in the infusate while different ionization of molecular species that possess differential electrical propensities can be induced in either the positive- or negative-ion mode during the electrospray ioniza­ tion process. Intrasource separation and selective ionization are expected to simplify lipid purification prior to MS and result in greater accuracy of the analysis (12). Preanalytical separation steps and enrichment of lipid sample, however, are still required in many cases prior lipidomic final analyses. Ion suppression is a major complication with phospholipids. It can be viewed as the competition for ionization and transfer of molecules from the solvent droplets forming the spray to the gas phase. Because many different molecules are introduced simulta­ neously in the ESI source and because the volume of the drop­ lets is extremely small the positive or negative electrical charge in excess are scarce and will be captured by more affine molecular species. Similarly the surface of the droplet is extremely limited and surfactant molecules such as phospholipids are competing to be transferred to the gas phase and enter the mass spectrometer interface orifice. Phospholipids have a high propensity to suppress ionization of coeluting molecules such as drugs (13) or ganglio­ sides (14). In biological samples, abundant phospholipid molecu­ lar species compete with less abundant species and examination



Lipidomics of the Red Cell in Diognosis of Human Disorders

137

of diluted samples is recommended for characterization of minor species of phospholipids in the presence of abundant species. Moreover the ionization and transfer to the gas phase is dependent on the saturation and acyl chain length of phospholipid molecular species. This dependence and other ion suppression phenomenon are all increased as a function of the total lipid concentration (15). Inclusion in the infusate of ammonia or solvent-soluble trimethylamine (TEA), diisopropyl-amine (DIPA), or piperidine enhances phospholipid ionization in the negative mode and displaces Na+ and K+ counterions in the positive mode. However ion pairing of PC and SM with amines TEA or DIPA (+101) should be taken into account in the quantitation of the molecular species. To maintain equipment performance during long sampling lists, it is recommended that extracts are clarified. Indeed, insoluble lipid precipitates are frequently observed in solvent mixtures used for chromatography or infusion where phospholipid high concentra­ tion and limited solubility occurs in the solvent system required for HPLC separation. Precipitation of nonsoluble phospholipids can be reduced by heating the sample vial, connection tubes, and column with heating tape and in a thermostated sampler main­ tained at around 30–45°C. It should be noted that temperature may profoundly alter the performances of chromatography sys­ tems and modifies the retention times of components eluted from columns. On a reverse stationary phase grafted with octadecyl (C18) acyl chains the temperature reduces the troublesome ionic interactions and helps to maintain peak symmetry, so reducing the need for ionic pairing (16). On the other hand, retention is delayed on polar stationary phases (e.g., diol-silica) and useful degrees of resolution of phospholipid classes can be achieved at the expense of long analysis time. In erythrocyte membranes the proteins are associated with the lipids by both hydrophobic interactions and polar bonds. At the lipid/water interface hydrogen- and ionic-binding between headgroups are prevalent. Inside the bilayer core hydropho­ bic binding is responsible for the compactness and ordering of acyl chains and sterol rings. Therefore both polar and nonpolar solvents should be used for extraction of lipids bound to pro­ teins. Quantitative extraction requires a gradual procedure to separate the lipids from proteins by a method that avoids the rapid denaturation of the proteins and the trapping of lipids sur­ rounded by aggregated proteins. The clumping into aggregates of denaturated protein reduces lipid extractability into nonpolar nonwater-soluble solvents. A comparison of solvent activities and a description of the most common lipid extraction procedures can be found in the Cyberlipid website: (http://www.cyberlipid. org). Briefly, an aqueous suspension of lipid-containing biologi­ cal material is first mixed with a polar water-soluble solvent such as methanol (or isopropanol or ether). In a second step the less-

138

Quinn, Rainteau, and Wolf

polar nonwater-soluble solvent (chloroform, hexane) is added in a proportion which forms a single phase for the ternary system (water/water-soluble solvent/nonwater-soluble solvent). In this environment the proteins form a fine precipitate which does not tend to clump and does hinder lipid extraction from RBCs. Lipid­ omics studies by MS may be conducted directly from this ternary solvent mixture. However, a subsequent step is usually the parti­ tioning of the filtered previous extract mixture into two immis­ cible layers (17, 18) after the addition of a large extra volume of aqueous buffer. This partition helps to separate water-soluble constituents acting as an ESI ion-suppressor (proteins, sugar, salts) into an upper layer phase separated from the lower chlo­ roform layer containing the water-insoluble lipids. Because ionic strength and acidic buffer pH favors the desorption of anionic lipids (e.g., PIPx, gangliosides) bound to basic protein residues it is recommended for extraction protocols with partitioning of the polar lipids into the chloroform layer. Anionic lipids, however, tend to partition into the aqueous-alcoholic layer at pH greater than 4. Excessive acidification results in the cleavage of vinyl ether bond of plasmalogens. In tissues such as the red blood cell mem­ branes where both acid-resistant and plasmenyl lipids are to be examined, two separate extracts (one extraction protocol with acid and the other without acid addition) can be prepared. In solvent extraction mixtures with chloroform most of the lipids are finally extracted into the lower dense layer which separates from the aqueous-alcoholic upper layer. Proteins accumulate at the aqueous/chloroform layer interface with unfolded sequences oriented upward or downward according to their hydrophilic/ hydrophobic affinities. Indeed a turbid suspension is observed if abundant proteins or low ionic strength or other amphiphiles of the biological preparation stabilize the interface of chloroform droplets. When emulsions are formed (e.g., in some lipid extracts prepared in the presence of high bile acid concentrations), the separation of the upper/lower phases is long lasting and diffi­ cult. Protocols derived from the Folch procedure (17) have been devised to speed up the separation of layers such as centrifugation, cooling, alteration of chloroform/methanol ratio, or addition of saline buffer. When the lower chloroform layer is turbid it is rec­ ommended that the extract is percolated through a dehydrating agent to remove water. Barium oxide or anhydrous sodium sul­ fate is often employed for this purpose. Contamination by water jeopardizes the integrity of lipid extracts after solvent evaporation because a residue of acidic water remains which decomposes vinyl ether bonds. Lipid oxidation is also enhanced at the air interface because evaporation of water takes a long time. To minimize degradation and set up automatic preanalytical facilities for a simultaneous treatment of numerous samples, methods have been described using an automat robot with Solid



Lipidomics of the Red Cell in Diognosis of Human Disorders

139

Phase Extraction (SPE) protocols in which there is no need for partition the solvent/water mixture into two superposed layers. Extraction by SPE reduces the volume of solvent required. The principles of lipid SPE have been reviewed elsewhere (19). Station­ ary phases are designed for quantitative extraction of phospholip­ ids on reversed phase (20), or separation of neutral/phospholipid on straight phase (21). The recovery of anionic lipids has to be assessed against standards for polar stationary phases. A recent innovation for lipid extraction suited to highthroughput lipidomics using methyl tert-butyl ether has been described by Matyash et al. (22). Rigorous testing demonstrated that the protocol delivers similar or improved recoveries of spe­ cies of most of all major lipid classes compared with the ”goldstandard” Folch or Bligh and Dyer recipes. Moreover, because methyl tert-butyl ether forms the upper, less dense layer where lipids partition after separation, it allows faster and cleaner lipid recovery and is well suited for automated shotgun profiling using pipettage robots. 3.2.3. Storage of Lipid Extracts

Liquid chromatography–mass spectrometry (LC–MS) analysis has a high sample throughput relative to the methods described previously for the lipid extraction. Samples are pooled after extraction and stored before analysis as a series. The storage of lipids in organic solvent is critical if not protected against the chemical alteration by water and traces of oxygen. Intrinsic pro­ tective mechanisms and association with natural compounds act in living organism to defend the vulnerable double bonds of vinyl ether and polyunsaturated fatty acids against attack by reactive oxygen species. The removal of these lines of defense by isola­ tion and extraction of lipids requires addition of antioxidants. It is therefore usual to add a radical scavenger such as butylated hydroxy-toluene (BHT, 2,6-di-tert-butyl-p-cresol) and/or toco­ pherol to the extract. Being soluble in solvents of low polarity they can be easily added to the chloroform extract and their oxi­ dation can act as a potential index of damage to the biological lipids. Nevertheless, it is necessary to ensure that the concen­ tration of such agents does not exceed a threshold of 0.0001 (w/w) when antioxidants like BHT have been shown to act as prooxidants at higher concentrations. Oxidation can also be exac­ erbated by contamination with metal cations such as copper or/ and iron which catalyze production of free radicals in Fentontype reactions. Contamination can be avoided by using single-use glass vessels with PTFE fittings and it is recommended for all preanalytical steps in Lipidomics. Brown glass is recommended to prevent exposure of lipid extracts to UV light, for example, when performing operations on a UV-illuminated sterile bench, which can be particularly deleterious. Elevated temperatures also favor double-bond migration and diene conjugation. The isomeriza­

140

Quinn, Rainteau, and Wolf

tion proceeds after the initial abstraction of an electron from an intermediary − CH2 methylene group intercalated between two double-bonds. Odd electron radicals derived of lipids induce an extensive peroxidation to other lipids especially in concentrated solutions. It is ideally recommended that lipid extracts are stored at low temperatures (−20°C) diluted in a solvent solution contained in an opaque (brown glass) vessel purged with argon gas and sealed by PTFE tape. The anionic phospholipids such as PI and PS present in RBC membrane extracts are protected in the long term from the head-group and fatty acid cleavage in the form of ammonium salts. Evaporation of the solvent prior to analysis is performed under a stream of oxygen-free dry nitrogen at a temperature not exceeding 40°C. SpeedVac is an effective alternative to this method since numerous samples can be evapo­ rated simultaneously in vials which can be subsequently trans­ ferred to an automatic sampler module of an HPLC system. Up to 40 samples can be evaporated simultaneously before they are loaded into the cooled sampler module. Cooling of the injec­ tion module prevents evaporation of solvent after the septum is perforated by the sampling needle which allows the reinjection of sample for further analysis using different data acquisition modes. Caution should be exercised, however, as temperatures lower than about 10°C may result in precipitation of the fraction of the most insoluble lipids concentrated in the type of solvent system used for HPLC (heating instead of cooling the sampler may be an alternative if sphingomyelins of RBC membranes are studied). Insoluble lipid aggregation results in the splitting of the eluted peaks into micellar and monomeric compounds. More efficient solvents for lipid solubilization like chloroform/methanol binary mixtures (1/1 or 2/1) can interfere with chromatography selec­ tive retention, the metal tubing of the equipment may deteriorate due to exposure to HCl formed from chloroform and finally the transfer of molecules from the ESI spray to the gas phase can be decreased. Using double layered PTFE-sealed vials and extensive purg­ ing with argon extracts containing highly polyunsaturated glyc­ erophospholipids such as human RBC membranes were found to remain unchanged for up to 6 months storage at subzero tem­ peratures. Sphingolipids are expected to be stable for much longer periods of time since they are generally more saturated. Sterols, steroids, and bile acids are also innately resistant to chemical altera­ tion as indicated by analyses of paleofecal specimens (23). By con­ trast, compounds such as 7-dehydrocholesterol or ergosterol with conjugated double bonds in the B-ring require more cautious handling among which is recommended low illumination, inert atmosphere, and analysis as soon as possible after isolation. The last paragraph in the chapter shows the measurement of oxidized phosphatidylethanolamines by LCMS2 after long-term storage.



3.3. Results 3.3.1. Analysis of RBC Phospholipids by Tandem MS2

Lipidomics of the Red Cell in Diognosis of Human Disorders

141

In this section the use of simple tandem MS2 to analyze most of the phospholipids extracted from human RBC membrane following a straightforward protocol will be described. Examination of the raw lipid extracts obtained by the method of Rose-Oklander (24) and methyl tert butyl ether (22) offers many advantages in terms of time and manpower per sample. The method developed for basic research can be now automated and fits appropriately within the context of biomedical applications. The utility of the method can also be compared to methods requiring more sophisticated equipment such as hybrid tandem MS2, Q-Trap, and Q-TOF, coupled with HPLC which are described below. To illustrate the method, results obtained after the direct infusion of the total extract of RBC membrane phospholipids into the ElectroSpray Ionisation (ESI) source of a triple quad­ rupole (TQ) spectrometer are presented. More comprehensive analyses are obtained if an HPLC separation of phospholipid class and subclass is conducted prior to MS2. However, incorporation of a chromatography step extends the time required for analysis by the length of the chromatographic run. Furthermore, comple­ mentary HPLC equipment requires pumps, filters, and seals that must be carefully maintained for running long series of samples in clinical applications. Examining RBC extracts for some particular medical applications may require a resolution not achieved by a direct infusion setup. An example of such demanding applica­ tion is the separation of diacyl/ether phospholipids described in below. The direct infusion of raw phospholipid extracts into ESI–MS has replaced the multiple steps previously used for the separa­ tion of phospholipid classes and fatty acid composition: recovery of individual phospholipids from TLC plates; phosphorus assay, and fatty acid analysis by gas chromatography. The MS-based method allows detailed analysis of lipid species in the field of biomedicine. The principle applied in tandem MS2 is the cleavage of a spe­ cific product ion from the phospholipid parent ion, e.g., serine from phosphatidylserine (PS) or phosphorylcholine from phos­ phatidylcholine and sphingomyelin (SM). Figure 1 shows the specific product ions resulting of collision-induced dissociation (CID) in the tandem MS2. As compared to electron impact fragmentography (energy = 70 eV) the collision of accelerated parent ions through a quadru­ pole filled with argon or nitrogen has a relatively low energy (<20 eV). Only a few bonds are cleaved in the parent ion. The yield and position of the fragmentation depends on the excess internal energy and its distribution in the excited molecule. The fragmen­ tation yield varies widely among the phospholipid classes with product ion m/z = +184 (PC and SM) being the most abun­ dant in the RBC membrane lipid extract. The ionization of other

142

Quinn, Rainteau, and Wolf

Fig. 1. Specific cleavage of the head-group is produced by collision-induced dissociation to characterize the phospholipid class. Specific ions and neutral fragments derived from the head-group moiety are obtained from ionized phospholipid in the positive or negative mode. Ester-bonded fatty acyl groups are also cleaved from the negative phospholipid parent ions in the form of acyl carboxylates at a higher CID energy. The fatty acid composition which is derived characterizes the phospholipid molecular species.



Lipidomics of the Red Cell in Diognosis of Human Disorders

143

phospholipid classes can, to some extent, compensate for their low fragmentation yield. For instance, positively charged parent ions M+H+. are favored in acidified extracts. Formic or acetic acid are volatile acids used for this purpose. The small molecules do not compete with phospholipids spread on the surface of the spray droplets for transfer to the gas phase, a critical step before entering the MS entrance orifice. Indeed, the addition of vola­ tile acids at 0.1–2% (vol/vol) to the infusate helps to create the excess of protons favorable to M + H+ adducts in the ESI spray droplets. By contrast, negatively charged parent phospholipid ions (M − H)− are favored by addition of volatile basic amines which trap protons and shift the balance to (M − H)− ions. Tri­ ethylamine (TEA), pyridine, and di-isopropylamine (DIPA) are volatile amine bases used at low concentrations (<2% vol/vol) to improve the low yield of ionization in ESI source for negatively charged phospholipid ions (M − H)−. Unfortunately the organic amine base can also participate in the pollution of the source and interface (25). The modifier activity of ammonia for phospholi­ pids is of interest for both of the positive and negative ioniza­ tion modes but its activity depends on the water content of the spray which alters the balance of NH3/NH4+ (base/acid). Both NH4OH aqueous solution or NH3 dissolved in solvents are available as modifiers. The addition of TEA or DIPA (MW = 101) has the disadvantage of splitting mass peaks of PC + H+ (Fig. 2) and PE + H+ (not shown) into multiple adducts (M + 101), which may jeopardize the quantification protocol. The addition of formic acid improves the ionization yield of positively charged phospholipids. Acid modifiers (H+ and R3NH+) also displace sodium and potassium cations abundant in the biological tissues which bind to phospholipids. For instance, the adduct ion (M + Na)+ is m/z +22 relative to (M + H)+. This +22 shift is confusing since a molecular species having an acyl chain length with two extra methylene groups and three double bonds is also shifted to +22 (M + 28 − 6). Quantification requires dei­ sotopization of the mass to resolve the superimposed mass peaks of the isobaric compounds. Substitution of isotonic sodium chloride or potassium chlo­ ride in the original Folch procedure with acidified water (acetic or formic acid 1/1,000, vol/vol) is recommended to avoid this problem. However, caution is required because acidic treatment can result in difficulties in separation of emulsified solvents and aqueous phases of the Folch mixture. Moreover, acidic condi­ tions can also degrade plasmalogens present in the RBC extract. LiCl has been used as a substitute for NaCl or KCl to beneficial effect. It is also good practice to split the raw extract into two aliquots, one added with the base for optimization of negative ionization and the other with formic acid to measure the positive ions. Measurements from two such infusions of the extract greatly

Fig. 2. Mass spectrum of phosphatidylcholine and sphingomyelin in an extract of RBC membrane. Each M + 1 (M + H+) adduct is associated with a supplementary adduct M + 101 (M + TEA)+ formed in the presence of triethylamine (MW = 101). The TEA base was added previously into the infused membrane lipid extracts to improve detection of negatively charged phospholipids, e.g., PE (detected with a low sensitivity as a precursor of the negative product ion 196). The addition of triethylamine 0.5% (vol/vol) creates abundant M + 101 adducts labeled in the spectrum as SM-C16:0 +101 (m/z 804) and PC 32:0 +101 (m/z 835).

144 Quinn, Rainteau, and Wolf



Lipidomics of the Red Cell in Diognosis of Human Disorders

145

improve sensitivity of the method. Addition of ammonia (NH3/ NH4+) and water to change (NH3/NH4+) is also an option. Mass spectra of erythrocyte PE and PC accumulated during 2 min in the presence of triethylamine or formic acid, respectively, are shown in Fig. 3a, b. It shows 25–30 different molecular spe­ cies for each class in the RBC membrane, which is considered a sufficient resolution for many applications conducted with such a straightforward protocol and without a chromatographic step. 3.3.2. Analysis of RBC Phospholipids by Tandem Hybrid MS/MS

Tandem hybrid MS2 uses either an ion trap (Q-Trap) or a time-of-flight (Q-TOF) analyzer as the second mass filter. This achieves an enhanced resolution of ion products to less than width »0.3–0.1 amu for mass around m/z 700 compared to the mass analysis conducted in a quadrupolar filter (peak width at 50% »0.7 amu). The enhanced resolution of Q-Trap (QT) com­ pared to triple quadrupole spectrometers (TQ) has the advantage of a great selectivity for the assignment of structure from the mass spectrum. A particular advantage can be exploited in the case of phospholipid analysis by QT because as the phospholip­ ids are eluted from HPLC step ion-trap analysis is triggered for an “enhanced resolution” analysis during the data acquisition (Information Dependent Acquisition). When the CID mode of the parent ion is activated the product ions formed in the collision cell fill the trap to a preset appropriate optimum load. 20–50 ms are usual filling times to prevent the space charge effect in the ion trap (an excessive charge effect is minimized in a linear ion trap with a large volume). Then, an enhanced product ion (EPI) analysis is obtained. When the CID mode is off, the parent ions are filtered by the first quadrupole and the analysis of parent ions is repeated in the enhanced resolution (ER) mode in the ion trap. Triggering of the “enhanced mode” analysis of selected parent ions is depend­ ent on the signal intensity which is continuously monitored in the “standard” resolution mode. The criteria used by informa­ tion system for triggering the ER mode can include or exclude 13 C isotopes. An application of the enhanced mass resolution mode for QT coupled with HPLC is shown in Fig. 4a, b for phosphati­ dylethanolamines from the human red blood cell membrane. An important feature of the application is the discrimination of ether lipids which can be used to investigate peroxisomal defects in children with a neurological condition and with malformation syndrome. From a clinical viewpoint such results provide a ready means of diagnosis and for treatment monitoring. The uppermost panel in Fig. 4a, b shows the Total Ionisation Current (TIC) obtained by monitoring the loss of the neutral fragment 141 (phosphorylethanolamine) from the positively charged parent ions of phosphatidylethanolamines. Many different

Fig. 3. (a) Mass spectrum of phosphatidylethanolamine species accumulated after infusion for 2 min. 1/50 of the raw extract prepared from erythrocyte membranes is infused (0.5 ml RBC pellet is extracted). Thirty distinct molecular species of PE are detected as (M − H)– precursors of the negative product ion −196. The structure assignments indicated in the spectrum are obtained after CID products analysis (fatty acids and the lysoderivative moiety). In the negative mode the plasmalogen PEs (designated as p-PE) are detected with a

146 Quinn, Rainteau, and Wolf

Lipidomics of the Red Cell in Diognosis of Human Disorders

Fig. 3. (continued) high sensitivity. (b) Mass spectrum of phosphatidylcholines and sphingomyelins accumulated after the infusion for 2 min of 1/50 of the raw extract prepared from 0.5 ml pelleted RBC. Thirty distinct molecular species are detected as (M + H)+ precursors of the positive product ion +184. The structure assignments indicated in the spectrum are obtained after CID product analysis in the negative mode. The TEA adducts (+101) are also detected in the sample.

147

Fig. 4. (a) Q-Trap analysis of phosphatidylethanolamines extracted from human erythrocyte membranes. Phosphatidylethanolamines are eluted (50 µl/min) from the HPLC silica-diol column in the ESI source. The ternary solvent system (hexane/isopropanol/water) is added with ionization modifiers (formic acid and di-isopropylamine). Upper panel shows the Total Ion Current monitoring the neutral loss of 141 from the positively charged parent ions. At intervals the Information Dependent Acquisition (IDA) software triggers the high resolution analysis of negatively charged parent ions (middle panel shows Enhanced Resolution mass spectra triggered at 20.59 min for the overlapping plasmenyl species m/z 722 (pPE 16:0/20:4), m/z 750 (pPE 18:0/20:4), and m/z 748 (pPE 16:0/22:4). Structure assignments are obtained by product ion analysis after collision-induced dissociation (four lower panels: see text).

148 Quinn, Rainteau, and Wolf

Fig. 4. (continued) (b) Analysis of diacyl species of phosphatidylethanolamines extracted from human erythrocyte membranes. Upper panel shows the HPLC separation of plasmenyl(RT » 20min) and diacyl-(RT » 24 min) species. At interval the Information Dependent Acquisition (IDA) software triggers the high resolution analysis of negatively charged parent ions (middle panel shows mass spectra triggered at 24.4 min of four coeluting diacyl species). Structure assignments are obtained by product ion analysis after collision-induced dissociation (four lower panels: see text).

Lipidomics of the Red Cell in Diognosis of Human Disorders 149

150

Quinn, Rainteau, and Wolf

molecular species comprising this class are eluted between 19 and 28 min from a HPLC separation on a polar silica-diol column using a solvent mixture consisting of hexane/isopropanol/NH4 acetate 15 mM (60/30/2 vol/vol). More than 30 different molecular species are commonly authenticated in human RBC membranes with the high molecular weight and saturated species eluting before the short acyl chain species. In Fig. 4a, b the chromatographic separation of the subclass ether phosphatidylethanolamines (1 alkyl- and 1 alkenyl-PE, i.e. plasmanyl- and plasmenyl-PE) from the diester species is specifically improved at the expense of a sensitive detection favored by other solvent systems. Indeed the ether phosphatidylethanolamines are eluted clearly ahead (RT » 20 min) of the diester subclass (RT » 24 min.) with the ternary solvent mixture used for elution. Ether phosphatidylethanolamines are identified in Fig. 4a in the Enhanced Resolution mode for the parent ions (middle panel, ER) and structures are elucidated from the product ions spectra. For instance, the most prominent negative parent ions in human RBC membranes are eluted at RT » 20.6 min with m/z 722.6, 748.5, and 750.5. The parent ions are shown as negative ions (M − H)− in the middle panel (-ER) of Fig. 4a. The ER mode is triggered from the positive mode of ionization which serves for monitoring the neutral loss +141 (upper panel, TIC) to the nega­ tive mode of ionization for -ER (middle panel) in less than 1 s. Then the recording (»100 ms) in the enhanced resolution mode of the molecular species is triggered to produce -ER spectrum of the most abundant parent molecular species m/z 722.6, 748.5, and 750.5. The corresponding product ion mass spectrum (EPI) are also subsequently triggered (time duration »30 ms). The elu­ tion has to be sufficiently slow from the HPLC coupled to QTrap to take into account the time for subsequent recordings triggered in the negative mode before returning to the monitoring neutral loss of 141 from the positive ions. Four mass spectra of product ions obtained after CID of the most prominent peak mass detected in the middle panel are pre­ sented in the lower panel of Fig. 4a. The Enhanced Product Ions (EPI) spectrum which is triggered by the mass m/z 722.6 shows the mass peaks 303 and 259 corresponding to fatty acid arachido­ nate (20:4) and (20:4 – CO2), respectively. The ether bond of the fatty aldehyde at the position sn-1 is not cleaved by the low energy of CID (<50 eV). However the molecular species of ether PE corresponding to m/z 722.6 with 20:4 as the acyl group can be unambiguously assigned to alkenyl PE O-16:1 (1Z)-20:4. Similarly ether-PE of m/z 748.5 (Fig. 4a, lower right panel) shows two distinct fatty acid carboxylates in the EPI spectrum, 303 (20:4) and 329 (22:5). Taking account of the similar reten­ tion time, characteristically short for ether lipids, this suggests the assignment to two distinct alkenyl-PE with C38 and a total



Lipidomics of the Red Cell in Diognosis of Human Disorders

151

of six double bonds. The alkenyl PE O-18:2 (1Z, 9Z)-20:4 and alkenyl PE O-16:1 (1Z)-22:5 are the best candidate structures. The assignments are confirmed by mass peaks at m/z 436 and 462 corresponding to lysoderivatives M-H-22:5 and M-H-20:4 which are released from the candidate structures after cleavage of the sn-2 fatty acids. The EPI of mass peak m/z 750.8 also shows the cleavage of two product ions m/z 303 (20:4) and m/z 331 (22:4) consistent with the structure of two coeluting alkenyl Pes, O-18:1 (1Z)-20:4 and PE O-16:1 (1Z)-22:4. Finally, the EPI of m/z 723.7 is assigned to a structure similar to m/z 722.6 (alke­ nyl PE O-16:1 (1Z)-20:4) but enriched in a C13 carbon atom in the lyso derivative moiety (m/z 437). In this instance, the C13 carbon isotopomer is confirmatory of the structure previously described for m/z 722 (alkenyl PE O-16:1 (1Z)-20:4). Figure 4b shows the results obtained for a combination of the positive mode monitoring the neutral loss 141 specific for PE (M + H)+ (upper panel) and the analysis in the negative mode of enhanced resolution spectra around the retention time 24.4 min. At this time four species of the diester phosphatidylethanolamines are coeluted. The mass peaks are selected by the Information Dependent Acquisition software running on the data station and they are rescanned at an enhanced resolution in the ion trap. In the middle panel the mass m/z 716.6, 738.5, 764.6, and 766.6 are detected. Each of the peaks is associated with C13 isotopom­ ers (M + 1, M + 2, M + 3,…) of which an assignment and quan­ tification are queried after deisotopization of the spectrum. An example is peak m/z 766.6 the amplitude of which represents the sum of a C13 isotopomer of m/z 764.6 comprising 2 C13 atoms (M1+2) and another less unsaturated molecular species (five dou­ ble-bonds, M2 = 766.6). Deisotopisation software calculates the expected amplitude of the C13 isotopomer of 764.6 (PE 38:5) with two C13 atoms (9.6%) and it calculates after subtraction the remaining fraction of the mass peak m/z 766.5 due to PE38:4. The assignment of mass peaks in the diester subclass is usually straightforward because both of the two fatty acids substituting positions sn-1 and sn-2 of glycerol are detected in the EPI spec­ trum. For instance, m/z 716.6 (PE34:1) is cleaved into m/z 255 (palmitic acid, 16:0) and m/z 281 (oleic acid, m/z 281). m/z 738.6 corresponds to PE36:4. m/z 452 is the lyso derivative product ion corresponding to the cleavage of arachidonic acid. m/z 764.6 (PE38:5) is assigned to a mixture of two structures, one is PE16:0/22:5 (the EPI spectrum shows the peak for 16:0 (m/z 255) and 22:5 (m/z 329), and the other structure is 18:1 (m/z 281) and 20:4 (m/z 303 and 259 (303-CO2)). The resolution of such a mixture of isobars is a major advantage in phospholipid studies conducted with tandem MS2 coupled to HPLC. Hybrid tandem systems such as Q-Trap, which accumu­ lates parent ions during the HPLC elution for short periods, are

152

Quinn, Rainteau, and Wolf

useful for rapid solution of the isomeric composition of molecular species. It is explicit that no quantitative data can be determined without a careful calibration by internal standards. The amplitude of product ions (fatty acids) cannot be used for reliable quanti­ tation because of the widely different stability of the fatty acid carboxylate ions. Figure 5 shows the complete separation of ether- and diester subclasses of PEs achieved by a silica-diol HPLC. This chroma­ tographic separation has been optimized in the perspective of biomedical applications which are relevant to screening of per­ oxisomal biosynthetic defects in children. This investigation was previously developed for screening chondrodysplasia and neuro­ logical conditions in infants. However, recent developments are now scheduled in adult patients for dementia. PlsEtn levels were measured in five independent population collections comprising 400 clinically demented and 350 nondemented subjects. Circu­ lating PlsEtn levels were observed to be significantly decreased in serum from clinically and pathologically diagnosed Alzheimer dementia at all stages of dementia, and the severity of this decrease correlated with the severity of dementia (26). Figure 6 demonstrates a method for quantitation of RBC membrane phospholipids as a function of the fatty acid composition of the constituent molecular species. A potential application of this method is the detection of essential fatty acid deficit of the series n-3 (DHA, upper panel Fig. 6) and n-6 (AA and LA, middle and lower panel, respectively). The method is especially suited to detect early deficit of particular classes or molecular species of essential fatty acids that are subject to fast metabolic turnover. Potentially the method has a greater sensitivity than the total FA methyl ester profile. For example, EFA in PE-diester is expected to be altered earlier than in other phospholipid classes during the deficit and EFA resupplementation. The method should be appropriate for prediction of inflammatory and allergic pathologies where the bal­ ance of n-3/n-6 EFA series is assumed to play a modulatory role (27, 28). Because the collision energy required for the cleavage of the ester bonds is similar for most of the glyceroPLs classes (PI is a notable exception) a calibration with weighed standards allows the quantitative assay of EFA-enriched phospholipid species. An interesting application of lipidomics to RBC membranes is the investigation of oxidative damage to phospholipids. For example, the investigation is relevant to oxidative stress and selec­ tive mitochondrial cardiolipin damage (29), radiotherapy and selective cardiolipin and phosphatidylserine oxidation (30), and innate macrophage response (31). Presently, RBC membrane phosphatidylethanolamines serve as a simple biomarkers to monitor lipid oxidation in stored blood products. Figure 7a, b illustrate the consequences of RBC storage (ex vivo, 1 month at 4°C) on the phosphatidylethanolamine molecular species.

Fig. 5. Contour plot of phosphatidylethanolamine molecular species extracted from human erythrocyte membranes: HPLC separation of ether and diester subclasses on silica-diol (YMC Pack Diol 5 mm, 60A, 250 × 4 mm) is achieved by the quaternary mixture hexane/isopropanol/NH4 acetate 15 mM. Ether-PEs are eluted in the range 19.5–21 min. Ether-PEs are followed by the numerous species of diester-PEs in the range 21.7–28 min. The enhanced resolution mass spectra of the corresponding ranges are shown below the contour plot (left, ether-PEs, major mass peak is m/z 750 (p-PE 18:0/20:4; right, diester-PEs, major mass peaks are m/z 716 (PE 34:1), 738 (PE 36:4), 764 (PE 38:5), 766 (PE 38:4).

Lipidomics of the Red Cell in Diognosis of Human Disorders 153

Fig. 6. HPLC separation of red blood cell membrane phospholipids as a function of the fatty acid composition. (B) Upper panel shows the elution of phospholipids enriched in 22:6 (docosahexaenoic acid, DHA) monitored by the product ion m/z 327.2. The product ion is formed from the negative parent ions after collision-induced dissociation. The precursor PLs enriched in 22:6 are labeled in the chromatogram as ether PE (RT » 20.6 min) and diester-PE (RT » 23.1 min). PC enriched with DHA are eluted around 29.9 min from the silica-diol column. (b) Middle panel shows the elution of PLs enriched in 20:4 (arachidonic acid, AA) monitored by the product ion m/z 303.2. In addition to ether-PE and diester-PE the chromatogram shows the elution of AA containing species of phosphatidylinositol (PI, RT » 26.8 min.) and phosphatidylserine (PS, RT » 29.3 min). (c) Lower panel shows the elution around RT » 23.7 min of diester-PE species enriched with linoleic acid (18:2, LA) monitored by the product ion m/z 279.2.

154 Quinn, Rainteau, and Wolf

Fig. 7. (a) Separation of ether-, diacyl-, and oxidized RBC phosphatidylethanolamine molecular species is obtained on silica-diol (elution with hexane/isopropanol/ammonium acetate 15 mM: 70/20/10, vol/vol/vol). The elution of phosphatidylethanolamines is monitored in the negative mode of ionization by the class specific negative product ion m/z −196 (upper panel). The group “oxidized PE” accumulates during a long-term storage of the erythrocytes (only two groups were detected previously after a short-term storage as shown in Fig. 4). The most prominent mass peak (m/z 730) among the oxidized phosphatidylethanolamines (lower panel, right) is assigned by product analysis after CID to hydroxylated PE 34:2-OH.

Lipidomics of the Red Cell in Diognosis of Human Disorders 155

Fig. 7. (continued) (b) Structure assignment of oxidized phosphatidylethanolamines by product analysis after CID. Two prominent peaks eluting at RT »30 and 32 min are detected in the HPLC profile (upper panel). Peak at RT 30 min corresponds to the phosphatidylethanolamine m/z = 730 (middle panel, left) which is cleaved after CID into the product ion m/z 295 consistent with the mass of hydroxylated fatty acid 18:2-OH (middle panel, right). The mass peak at m/z 452 is also consistent with the lysoderivative formed from m/z 730 by the cleavage of the hydroxylated fatty acid m/z 295. The coelution of negative product ions m/z 730 and m/z 295 (lower panel) are indicative of the coexistence of two distinct positional isomers of the hydroxylated PE.

156 Quinn, Rainteau, and Wolf



Lipidomics of the Red Cell in Diognosis of Human Disorders

157

Using product analysis after CID the most abundant oxidized species of phosphatidylethanolamines in the RBC membrane found after long-term storage at 4°C are m/z 730 (PE 16:0/18:2 OH, characteristical hydroxylated product ion m/z 295), m/z 728 (PE 16:0/18:3 OH, characteristical hydroxylated product ion m/z 293), and m/z 754 (PE 16:0/20:4 OH, characteristical hydroxylated product ion m/z 319). A more detailed consideration of this topic can be found in the chapter by Kagan’s group in (32, 33).

4. Notes Analysis of the human red cell membrane provides a reliable method for diagnosis of a variety of human disorders of lipid metabolism and nutrition (Subheading 3.1). The utility of the method depends on convenient methods of lipid extraction and analysis (Subheading 3.2). The high sensitivity and resolution which are required for the measurement of minor lipid com­ pounds (metabolomics) is not a priority for the study of struc­ tural erythrocyte membrane lipids but practical considerations regarding manpower and throughput are of concern to develop the clinical applications. Mass spectrometry is the basis of the current analytic methods coupled with a direct infusion into the ESI source (Subheading 3.3.1) or coupled with a HPLC separation methods to deliver analytes. QTrap tandem MS (Subheading 3.3.2) provides the complete analysis of molecular species using the capability of ion-trap throughout the elution of phospholipids. TQ equipment is preferentially used for quantitative measurements of preset phospholipid molecular species. It is expected that the determination of fatty acid profiles which is currently offered by diagnosis center will be substituted favorably in the next future by MS-based methods resolving the whole lipid structures.

Acknowledgments This work was aided by a grant from the Human Frontier Science Programme (RGP0016/2005C).

158

Quinn, Rainteau, and Wolf

References 1. Fernandes, J., Saudubray, J., Berghe, G. V. D., and Walter, J. (Eds.) (2006) Inborn metabolic diseases, Springer, 4th Edn, New York. 2. Matsuzaka, T., Shimano, H., Yahagi, N., Kato, T., Atsumi, A., Yamamoto, T., Inoue, N., Ishikawa, M., Okada, S., Ishigaki, N., Iwasaki, H., Iwasaki, Y., Karasawa, T., Kumadaki, S., Matsui, T., Sekiya, M., Ohashi, K., Hasty, A. H., Nakagawa, Y., Takahashi, A., Suzuki, H., Yatoh, S., Sone, H., Toyoshima, H., Osuga, J., and Yamada, N. (2007) Crucial role of a longchain fatty acid elongase, Elovl6, in obesi­ ty-induced insulin resistance Nat Med 13, 1193–202. 3. Haimi, P., Uphoff, A., Hermansson, M., and Somerharju, P. (2006) Software tools for anal­ ysis of mass spectrometric lipidome data Anal Chem 78, 8324–31. 4. Chow, C. K. (1985) Vitamin E and blood World Rev Nutr Diet 45, 133–66. 5. Fuchs, B., Schiller, J., and Cross, M. A. (2007) Apoptosis-associated changes in the glycero­ phospholipid composition of hematopoietic progenitor cells monitored by 31P NMR spectroscopy and MALDI-TOF mass spec­ trometry Chem Phys Lipids 150, 229–38. 6. Kuypers, F. A. (2007) Membrane lipid altera­ tions in hemoglobinopathies Hematol Am Soc Hematol Educ Program 2007, 68–73. 7. Hebbel, R. P. (1984) Erythrocyte autoxida­ tion and the membrane abnormalities of sickle red cells Prog Clin Biol Res 159, 219–25. 8. Schrier, S. L., Centis, F., Verneris, M., Ma, L., and Angelucci, E. (2003) The role of oxidant injury in the pathophysiology of human thalassemias Redox Rep 8, 241–5. 9. van Deenen, L. L., Kuypers, F. A., Op den Kamp, J. A., and Roelofsen, B. (1987) Effects of altered phospholipid molecular species on erythrocyte membranes Ann N Y Acad Sci 492, 145–55. 10. Soupene, E., Fyrst, H., and Kuypers, F. A. (2008) Mammalian acyl-CoA:lysophosphatidylcholine acyltransferase enzymes Proc Natl Acad Sci U S A 105, 88–93. 11. Han, X., and Gross, R. W. (2005) Shotgun lipidomics: electrospray ionization mass spec­ trometric analysis and quantitation of cellu­ lar lipidomes directly from crude extracts of biological samples Mass Spectrom Rev 24, 367–412. 12. Han, X., Yang, K., Yang, J., Fikes, K. N., Cheng, H., and Gross, R. W. (2006) Factors influencing the electrospray intrasource sepa­ ration and selective ionization of glycero­ phospholipids J Am Soc Mass Spectrom 17, 264–74.

13. Shen, J. X., Motyka, R. J., Roach, J. P., and Hayes, R. N. (2005) Minimization of ion sup­ pression in LC-MS/MS analysis through the application of strong cation exchange solidphase extraction (SCX-SPE) J Pharm Biomed Anal 37, 359–67. 14. Tsui, Z. C., Chen, Q. R., Thomas, M. J., Samuel, M., and Cui, Z. (2005) A method for profiling gangliosides in animal tissues using electrospray ionization-tandem mass spec­ trometry Anal Biochem 341, 251–8. 15. Koivusalo, M., Haimi, P., Heikinheimo, L., Kostiainen, R., and Somerharju, P. (2001) Quantitative determination of phospholi­ pid compositions by ESI-MS: effects of acyl chain length, unsaturation, and lipid concen­ tration on instrument response J Lipid Res 42, 663–72. 16. Christie, W. W. (2003) Lipid analysis: isolation, separation, identification and structural analysis of lipids, Oily Press,3rd ed. Bridgwater. 17. Folch, J., Lees, M., and Stanley, G. H. S. (1957) A simple method for the isolation and purification of total lipids from animal tissues J Biol Chem 226, 497–509. 18. Bligh, E. G., and Dyer, W. J. (1959) A rapid method of total lipid extraction and puri­ fication Canadian J Biochem Physiol 37, 911–17. 19. Ruiz-Gutierrez, V., and Perez-Camino, M. C. (2000) Update on solid-phase extraction for the analysis of lipid classes and related com­ pounds J Chromatogr A 885, 321–41. 20. Carboni, E., Wieland, S., Lan, N. C., and Gee, K. W. (1996) Anxiolytic properties of endogenously occurring pregnanediols in two rodent models of anxiety Psychopharmacol­ ogy (Berl) 126, 173–8. 21. Persson, E., Lofgren, L., Hansson, G., Abra­ hamsson, B., Lennernas, H., and Nilsson, R. (2007) Simultaneous assessment of lipid classes and bile acids in human intestinal fluid by solid-phase extraction and HPLC methods J Lipid Res 48, 242–51. 22. Matyash, V., Liebisch, G., Kurzchalia, T. V., Shevchenko, A., and Schwudke, D. (2008) Lipid extraction by methyl-tert-butyl ether for high-throughput lipidomics J Lipid Res 49, 1137–46. 23. Sobolik, K. D., Gremillion, K. J., Whitten, P. L., and Watson, P. J. (1996) Sex determina­ tion of prehistoric human paleofeces Am J Phys Anthropol 101, 283–90. 24. Rose, H. G., and Oklander, M. (1965) Improved procedure for the extraction of lip­ ids from human erythrocytes J Lipid Res 6, 428–31.



Lipidomics of the Red Cell in Diognosis of Human Disorders

25. Rutters, H., Mohring, T., Rullkotter, J., Griep-Raming, J., and Metzger, J. O. (2000) The persistent memory effect of triethylamine in the analysis of phospholipids by liquid chro­ matography/mass spectrometry Rapid Com­ mun Mass Spectrom 14, 122–3. 26. Goodenowe, D. B., Cook, L. L., Liu, J., Lu, Y., Jayasinghe, D. A., Ahiahonu, P. W., Heath, D., Yamazaki, Y., Flax, J., Krenitsky, K. F., Sparks, D. L., Lerner, A., Friedland, R. P., Kudo, T., Kamino, K., Morihara, T., Takeda, M., and Wood, P. L. (2007) Peripheral ethanolamine plasmalogen deficiency: a logical causative fac­ tor in Alzheimer’s disease and dementia J Lipid Res 48, 2485–98. 27. Chilton, F. H., Rudel, L. L., Parks, J. S., Arm, J. P., and Seeds, M. C. (2008) Mechanisms by which botanical lipids affect inflammatory disorders Am J Clin Nutr 87, 498S–503S. 28. Surette, M. E., Stull, D., and Lindemann, J. (2008) The impact of a medical food con­ taining gammalinolenic and eicosapentaenoic acids on asthma management and the quality

159

of life of adult asthma patients Curr Med Res Opin 24, 559–67. 29. Lesnefsky, E. J., and Hoppel, C. L. (2008) Cardiolipin as an oxidative target in cardiac mitochondria in the aged rat Biochim Biophys Acta 1777, 1020–7. 30. Tyurina, Y. Y., Tyurin, V. A., Epperly, M. W., Greenberger, J. S., and Kagan, V. E. (2008) Oxidative lipidomics of gamma-irradiationinduced intestinal injury Free Radic Biol Med 44, 299–314. 31. Hazen, S. L. (2008) Oxidized phospholipids as endogenous pattern recognition ligands in innate immunity J Biol Chem 283, 15527–31. 32. Domingues, M. R., Reis, A., and Domingues, P. (2008) Mass spectrometry analysis of oxi­ dized phospholipids Chem Phys Lipids 156, 1–12. 33. Tyurin, V. A., Tyurina, Y. Y., Kochanek, P. M., Hamilton, R., DeKosky, S. T., Greenberger, J. S., Bayir, H., and Kagan, V. E. (2008) Oxi­ dative lipidomics of programmed cell death Methods Enzymol 442, 375–93.

sdfsdf

Chapter 8 HPLC/MS/MS-Based Approaches for Detection and Quantification of Eicosanoids Susanna L. Lundström, Fabio L. D'Alexandri, Kasem Nithipatikom, Jesper Z. Haeggström, Åsa M. Wheelock, and Craig E. Wheelock Summary Eicosanoids are oxygenated, endogenous, unsaturated fatty acids derived from arachidonic acid. Detection and quantification of these compounds are of great interest because they play important roles in a number of significant diseases, including asthma, chronic obstructive pulmonary disease (COPD), cardiovascular disease, and cancer. Because the endogenous levels of eicosanoids are quite low, sensitive and specific analytical methods are required to reliably quantify these compounds. High-performance liquid chromatography mass spectrometry (HPLC/MS) has emerged as one of the main techniques used in eicosanoid profiling. Herein, we describe the main LC/MS techniques and principles as well as their application in eicosanoid analysis. In addition, a protocol is given for extracting eicosanoids from biological samples, using bronchoalveolar lavage fluid (BALF) as an example. The method and instrument optimization procedures are presented, followed by the analysis of eicosanoid standards using reverse phase HPLC interfaced with an ion trap mass spectrometer (LC/MS/MS). This protocol is intended to provide a broad description of the field for readers looking for an introduction to the methodologies involved in eicosanoid quantification. Key words:  Eicosanoid, Oxylipin, Arachidonic acid, Mass spectrometry, Electrospray, Ion trap, LC/MS/MS, Bronchoalveolar lavage fluid, BALF

1. Introduction The biological relevance of arachidonic acid-derived mediators (eicosanoids) has been demonstrated in numerous studies (1–5). These metabolic products comprise several different classes including leukotrienes (LTs), prostaglandins (PGs), thromboxanes (TXs), hydroxyeicosatetraenoic acids (HETEs), epoxyeicosatrienoic acids (EETs), and oxo-fatty acids. Eicosanoid biosynthesis is accomplished via several distinct enzymatic pathways including Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_8, © Humana Press, a part of Springer Science + Business Media, LLC 2009

161

162

Lundström et al.

cyclooxygenase (COX), lipoxygenase (LOX), and cytochrome P450 (CYP). These three different pathways together are responsible for the biosynthesis of the entire suite of compounds collectively known as the arachidonic acid cascade, forming a complex interactive network that is present in almost all tissues (1, 2). The most well-known eicosanoids are the PGs, TXs, HETEs, and LTs, but more recently, the lipoxins (LXs) and EETs have attracted considerable attention. The distribution and synthesis of eicosanoids are key targets for studies involving a range of inflammatory pathologies including asthma and chronic obstructive pulmonary disease (COPD), (6, 7) nephritis, (8, 9) cardiovascular diseases (9, 10), and cancer (11, 12). Historically the detection and quantification of eicosanoids have mainly been performed by high-performance liquid chromatography (HPLC) or gas chromatography coupled to mass spectrometry (GC/MS), as well as the more sensitive radio- and enzyme-immunoassay techniques (RIA and EIA) (13, 14). RIA and EIA approaches have two limitations (1) generally, only a single product can be detected at a time and (2) cross reactivity can cause variability in sample quantification. HPLC and GC/ MS methods are generally less sensitive than RIA and EIA, but offer increased selectivity for the detection of multiple eicosanoids simultaneously (15–17). However, compounds must be both volatile and thermally stable in order to perform GC/MSbased analyses, which necessitates the use of chemical derivatization for the analysis of most eicosanoids. An initial limitation in the development of a combined LC/ MS method was the interface between the HPLC and the mass spectrometer. However, with the development of MS ionization techniques such as electrospray ionization (ESI) and atmospheric pressure chemical ionization (APCI), eicosanoids can now be analyzed directly from aqueous samples. A major advantage of this approach is the simplification of sample preparation (18). As a result, liquid chromatography tandem mass spectrometry (LC/ MS/MS or LC/MSn) has become one of the most powerful techniques to quantify large numbers of eicosanoids in a variety of biological matrices (3, 7, 19–21). The HPLC separates the compounds based upon physical properties, followed by unambiguous identification based upon the characteristic product ions in the MS. This is generally achieved by scanning in selective reaction monitoring (SRM) mode. The acquisition rate, sensitivity, and selectivity provided by SRM enables the acquisition of high quality data for more compounds at faster acquisition rates. The advent of ultra performance liquid chromatography (UPLC) methodologies will further increase this process based on increased chromatographic capacity (22). HPLC/MS/MS has consequently become the only platform capable of performing concurrent quantification of the large numbers of analytes



HPLC/MS/MS-Based Approaches for Detection and Quantification

163

required for studying the biological role of eicosanoids. The choice of chromatography type and mass spectrometer depend on the characteristics of the eicosanoids in the sample and the purpose of the study. Studies have been performed in which a few selected class-specific eicosanoids are analyzed as well as large metabolic-profiling studies of a range of eicosanoids. In this protocol we present some of the main LC/MS/MS techniques and methods used for the detection and quantification of eicosanoids in biological samples. A brief description of MS instrumentation, ionization methodologies, and mass analyzers is discussed followed by the principles and use of SRM in eicosanoid quantification. A method is also described for extracting eicosanoids from biological samples using bronchoalveolar lavage fluid (BALF) as an example. The steps for preparing eicosanoid standards, gradient solvents, and optimization of MS parameters for developing an LC/MS/MS-based quantification method for eicosanoid analysis with an ion trap mass spectrometer are then presented. Finally, the instrumentation parameters and preliminary results for method development are shown. It is expected that readers will gain an understanding of the range in technologies available for LC/MS/MS-based analyses of eicosanoids. 1.1. Analysis of Eicosanoids Using Reversed Phase HighPerformance Liquid Chromatography

Endogenous eicosanoids are present in low levels in biological tissues (~pmol/mg to fmol/mg range) (3, 23). Accordingly, the majority of analytical methods begin with a clean-up and/ or separation step to enrich the sample for the compounds of interest. While technically possible (24), direct injection of complex samples into the MS is challenging. Subsequently, the use of an HPLC interfaced with the MS results in improved detection limits and overall quality of the MS data (18). Separating the analytes reduces background noise and problems associated with ion suppression from coeluting compounds as well as minimizes isobaric interferences. The choice of column, particle size, and mobile phase composition depends on the quantity and chemical properties of the target compounds; however there is a limited variety of mobile phases compatible with MS interfaces. Because most eicosanoids are medium to nonpolar, reverse phase (RP) chromatography is most commonly used (3, 7, 20, 21, 25–31). Analytes elute in order of increasing hydrophobicity with a hydrophobic stationary phase (e.g., C18) and a gradient changing gradually from aqueous solvents (with the addition of volatile salts, acids, or bases that improve ionization and retention on the HPLC column) to organic solvents (usually methanol and acetonitrile). The gradient is most commonly run under acidic conditions using a small percentage (~0.01–0.1% v/v) of acetic or formic acid. However there are examples of gradients run under both neutral and basic conditions (32, 33).

164

Lundström et al.

1.2. Analysis of Oxylipin Stereoisomers by Liquid Chromatography Mass Spectrometry

The use of chiral columns for HPLC-based analysis is a valuable tool for the separation of lipid enantiomers. Chiral liquid chromatography has been extended to both LC/MS and LC/MS/MS for the determination of oxylipin enantiomers. In nature, several oxylipins including hydroperoxyeicosatetraenoic acids (HPETEs), hydroperoxyoctadecadienoic acids (HPODEs), hydroxyoctadecadienoic acids (HODEs), HETEs, and EETs are formed as racemic mixtures via nonenzymatic reactions (34, 35). On the other hand, when these lipids are oxidized enzymatically, enantiopure products are predominantly produced (36). Thus, chiral phase LC/MS is a valuable tool for diagnosing the mechanism and source of enantiomer formation. To improve the limit of detection (LOD), the carboxyl groups of these oxylipins can be derivatized to pentafluorobezyl esters (PFB) (37) and further analyzed by chiral phase liquid chromatography/electron capture/atmospheric pressure chemical ionization/mass spectrometry (chiral LC/EC/APCI/MS). A number of enantiomers of regioisomeric HODEs and HETEs have been analyzed in biological samples by chiral LC/EC/ APCI/MS (38) and the results have been reviewed elsewhere (39, 40). Although PFB-derivatization and chiral LC/EC/APCI/MS have improved the LOD of chiral oxylipins, the method has some disadvantages including extra derivatization steps and sample clean up as well as low yield for conversion to the corresponding PFB-esters. Attempts have been made to analyze oxylipins in their free acid forms. For example, chiral LC/MS/MS was used to analyze underivatized enantiomeric oxylipins and subsequently identified 12(S)-HETE as the major enantiomer in the urine of patients with diabetes mellitus (DM) (41). The effects of phenobarbital on the synthesis of enantiomers of cytochrome P450 metabolites of arachidonic acid and EETs in rat liver were determined by chiral LC/APCI/MS (42). At present, the main limitation of chiral LC/MS for the analysis of oxylipin enantiomers is the availability of chiral columns. For example, chiral columns are produced predominantly as normal phase, which is not compatible with ESI, but can be used with APCI under well-defined experimental conditions. However, even these limitations are being addressed as demonstrated by the recent development of a capillary tandem column chiral phase LC/MS/MS method for the analysis of EET enantiomers (43). This method can simultaneously quantify all eight EET enantiomers as the free acids with limits of quantification in the low pg range.

1.3. Mass Spectrometry Techniques Used for Eicosanoid HPLC/ MS/MS Analysis

The first step in mass spectrometry analysis is to convert analyte molecules into gas phase ions. Following ion production, the ions are separated by a mass analyzer that measures the mass to charge ratio (m/z). There are a number of different mass analyzers, which



HPLC/MS/MS-Based Approaches for Detection and Quantification

165

vary mainly by their mass range limits (the upper limit of the mass of the ion that can be measured); acquisition rate (the rate at which the mass analyzer measures scans over a particular mass range); transmission range (the ratio of the number of ions reaching the detector to the number of ions leaving the source); mass accuracy (accuracy of the ion mass measurement provided by the mass analyzer); and resolution (ability of a mass analyzer to yield 50% valley separation between distinct signals of two ions). This value is typically defined as the ratio of detected m/z to the full width at half-height maximum (FWHM) of the peak. Essentially all mass analyzers can be used for lipidomics or eicosanoid analysis; however, the field has demonstrated a preference for certain mass analyzers that are particularly suitable for quantifying lipids in complex samples. In this section we will discuss the main ionization techniques and mass analyzers currently used in eicosanoid HPLC/MS/MS analysis. 1.3.1. Ionization Techniques 1.3.1.1. Electrospray Ionization

Electrospray Ionization (ESI) was first proposed as a source of gas phase ions by Dole et al. (44) in 1968. However, the success of this technique began when Fenn et al. (45) used ESI in mass spectrometry and demonstrated that multiply charged ions were obtained from proteins. This enabled the determination of the molecular mass of a compound using instruments with limited mass range. In ESI the ionization process occurs by applying a strong electric field (kV), under atmospheric pressure, to a liquid passing through a capillary tube. This field induces a charge accumulation at the liquid surface located at the end of the capillary forming a Taylor cone (46, 47). The charge accumulation causes droplets that contain an excess positive or negative charge to detach from the capillary tip and move toward the mass analyzer. An uncharged carrier gas (e.g., nitrogen) is often used to help nebulize the liquid and evaporate the solvent in the droplets. As the solvent evaporates, the molecules are forced closer together which increases the Coulombic electrostatic repulsion (i.e., resulting in high surface charge densities on the droplets). This process continues until the Debye length is reached, after which the repulsion overcomes the surface tension of the droplets, which then break up forming ions in a process that is still not well understood. The main advantage of ESI/MS over other MS techniques is that ESI/MS overcomes the propensity of many biomolecules to fragment following ionization and enables the formation of multiply charged ions. Thus, ESI/MS is critical for the detailed structural analysis of large biomolecules, but it is also useful for the analysis of small polar molecules. Another feature of this technique is its relation with the concentration of the target compound, and not to the total quantity of sample injected in the source. This fact has led to the development of nanoflow rate ESI (nano-ESI). Consequently, ESI has been widely applied in the analysis of all

166

Lundström et al.

classes of eicosanoids (20, 23, 29, 48, 49). Moreover it is not necessary to chemically modify eicosanoids to enhance ionization efficiently when using this technique. Even when chemical modifications (such as the use of methoxime (MO) derivatization) are used to improve the chromatography it does not affect the ionization efficiency of the eicosanoids. The disadvantages of ESI are its low robustness to matrix interferences and changes of mobile phase and flow rates that can produce ion suppression (caused by ionization competition between compounds) and poor ionization of some compounds. 1.3.1.2. Atmospheric Pressure Chemical Ionization Mass Spectrometry

Atmospheric Pressure Chemical Ionization Mass Spectrometry (APCI/MS) has proven to be valuable for the analysis of lipids from a variety of classes (50, 51). APCI is a soft ionization technique, but unlike ESI, APCI usually produces some degree of fragmentation that can be useful for structural characterization. In contrast to ESI, solvent evaporation and analyte ionization are two separate processes. The latter being a chemical ionization at atmospheric pressure with the mobile phase acting as the reactant gas. As the evaporation of the mobile phase is performed at high temperatures (up to 500°C), APCI is less mild than ESI, thus the thermal stability of compounds is important. APCI normally produces singly charged ions through the addition or abstraction of a proton. This technique is frequently used in the LC/MS/MS analysis of eicosanoids (42, 52, 53). Compared to ESI methodologies, LC/MS analysis using an APCI source performs best using normal phase (NP) chromatography methods (54, 55). Under these conditions the ion detection capability of APCI methodology is greater due to the low interference of the mobile phase on compound ionization; however, the necessary safety precautions need to be taken.

1.3.1.3. Electron Capture/APCI

Electron Capture (EC)/APCI is a variation of APCI that can be used to improve ion production and subsequently the LOD. This technique requires sample derivatization with an electron capturing group that will enhance formation of negative ions (see Subheading 1.2). EC/APCI gives essentially equivalent results with either NP- or RP-based systems. The technique has not been used extensively for eicosanoid analysis (38, 39). However, it has great potential for use in the identification of eicosanoid enantiomers and regioisomers, one of the biggest challenges in eicosanoid analysis (34, 38). The main disadvantages are that it requires derivatization and the PFB derivatives give relatively weak signals under conventional positive APCI conditions when NP solvents are used.

1.3.1.4. Atmospheric Pressure Photo Ionization

Atmospheric Pressure Photo Ionization (APPI) is an ionization technique that extends the range of ionizable compounds (56, 57). APPI can ionize compounds that ionize poorly by ESI or APCI,



HPLC/MS/MS-Based Approaches for Detection and Quantification

167

especially nonpolar compounds. The APPI source is a modified APCI source that uses a discharge lamp that emits photons rather than a corona discharge needle that emits electrons. APPI uses photons to ionize gas phase molecules after the sample is vaporized by a heated nebulizer similar to the one used in APCI. The analytes interact with photons, which induces a series of gas phase reactions that lead to the ionization of the sample molecules. APPI gives cleaner spectra and is less susceptible to ion suppression, because it is not based on charge affinity. Cai and Sayge (58) used this technique to analyze the methyl esters of eicosapentaenoic acid (EPA) as well as other fatty acids and described in detail the potential mobile phase alterations (59). One drawback with the technique is that it highly depends on the chromatographic system applied (e.g., RP vs. NP) because the solvents can interfere with ion formation. 1.3.2. Mass Analyzers 1.3.2.1. Quadrupole

1.3.2.2. on Trap

The quadrupole analyzer consists of four parallel metal rods that use the stability of the trajectories in oscillating electric fields to separate ions according to their m/z ratios (60, 61). This mass analyzer can be used in either a single or triple quadrupole configuration, with the triple quadrupole used to perform MS/ MS experiments. In the triple quadrupole configuration, the first and the last quadrupole serve as mass filters and the center quadrupole (or hexapole) is a collision cell consisting of an RF quadrupole (or hexapole) with optimum ion collimating behavior. This geometry enables different types of MS/MS experiments to be conducted such as product, precursor, and constant neutral loss (CNL) scanning as well as SRM. Using a triple quadrupole LC/ MS system, Kingsley et al. (48) were able to quantify prostaglandin glyceryl esters from cells with a 25 fmol limit of quantification (LOQ) and Masoodi et al. (29) analyzed 20 mono and polyhydroxy-fatty acid derivatives of linoleic acid, AA, EPA, and docosahexaenoic acid (DHA) with a limit of quantitation with a LOQ of 60–177 fmol. Kita et al. (62) used a triple quadrupole system to develop a high-throughput method for the detection of 18 different eicosanoids from biological samples. Using a single quadrupole equipment, Blewett et al. were able to identify 23 eicosanoids in a single HPLC–MS run (20). However, despite these qualities, quadrupoles are fundamentally low resolution instruments and high mass accuracy can only be achieved at the expense of sensitivity. In addition, MSn experiments can only be performed by multiquadrupole instruments. Paul and Steinwedel first described the ion trap in 1960 (61, 63), followed by Stafford et al. (64) who modified it for use in a mass spectrometer. An ion trap uses an oscillating electric field to trap ions in either two or three dimensions. Thus, ion traps exist in two types: 2D ion traps (also referred to as linear ion traps) and

168

Lundström et al.

the 3D ion traps (also referred to as Paul ion traps). The 2D ion trap is based upon a four-rod quadrupole that ends in lenses that repel the ions inside the rods. The 3D ion trap is composed of a ring electrode and two cap electrodes in order to form a closed area where the ions are trapped. 2D and 3D ion traps work similarly; however, 2D ion traps have greater than tenfold higher ion trapping capacity compared with 3D traps. The 2D ion trap cannot perform all the scan functions present in 3D traps due to problems with ion ejection from the trap. The main disadvantage of the ion trap is that precursor ion and neutral loss scans cannot be conducted. Ion traps also have a poor dynamic range and quantification capacity, characteristics that limit their use in the quantification of complex samples. Ion trap mass analyzers exhibit high sensitivity and are most strongly characterized by the ability to perform multiple stages of mass spectrometry (MSn). Up to 12 stages of tandem mass spectrometry (MS12) have been performed using an ion trap (65), which greatly increases the amount of structural information obtainable for a given molecule. Ion traps are normally coupled to ESI or APCI ionization sources for the structural characterization of eicosanoids. Liminga and Oliw (66), performed quantitative analysis of LOX products in bovine tissues. They demonstrated that 5-, 12-, and 15-HPETE can be identified by scanning the carboxylated ion of the hydroperoxide and the anion of its dehydration product, [M − H − H2O]−. Using MS2 and MS3 scanning mode they elucidated the structures of these compounds, demonstrated the specificity of transition ions and showed how to use transition ions for quantification. Kiss et  al. (67) used capillary LC, photodiode array (PDA), and ESI/MS2 in negative mode in order to simultaneously identify and quantitate relevant CYP, sEH, LOX, and COX-derived eicosanoids (EETs, 20-HETE, LTs, LXs, PGs, TXs, HETEs, dihydro-xeicosatrienoic acids [DHETs], dihydroxyeicosatetraenoic acids [DHETEs], 12-hydroxyheptadecatrienoic acid [12-HHT]) and free isoprostanes (iPTs). This method was capable of identifying 44 eicosanoids in a 50 min run. Kiss et al. (67) also reported a detailed list of transition ions that can be used for SRM analysis of eicosanoids as well as several MS2 spectra of eicosanoid standards. The cysteinyl-containing leukotrienes (LTC4, LTD4, and LTE4) and the 5-oxo-7-glutathionyl-8,11,14-eicosatrienoic acid (FOG7) were analyzed by Hevko and Murphy (68) using an ESI/MS with a 3D ion trap and a triple quadrupole mass spectrometer. Both singly charged negative and positive ions could be observed. However, in contrast to the triple quadrupole, the collisional activation of the negative ions in the ion trap required high collision energy, and even then few product ions were produced. It is therefore advisable to analyze these compounds in positive ion mode if an ion trap is used. MS2 analyses



HPLC/MS/MS-Based Approaches for Detection and Quantification

169

in this mode showed a common series of fragment ions at m/z 319, 301, 283, 265, and 189, which can be observed for all cysteinyl leukotrienes (CysLTs). The ion at m/z 319 was formed by cleavage of the carbon–sulfur bond to afford a protonated epoxide cation and loss of the neutral cysteine. The other three ions result from the sequential losses of water, and the ion at m/z 189 represents the cleavage of the C6–C7 bond of the lipid backbone. For FOG7 the MS2 spectrum is strikingly simple with only one major fragment ion for cleavage of the g-glutamate amino acid (m/z 497). Accordingly, MS3 analysis was necessary to provide unambiguous structural characterization. Hong et al. (69) used electrospray low energy collision-induced dissociation (CID) tandem mass spectrometry analysis and reported the correlations between spectra and structure and the fragmentation mechanisms of resolvin D1, protectin D1, monohydroxy-DHA, and related hydroperoxy-DHA. 1.3.2.3. Time of Flight

This mass analyzer measures the time it takes ions of different masses to move from the ion source to the detector. Although the principle of Time of Flight (TOF) has been known since Thomson carried out his experiments on ionized particles in 1897 (70), its use in a mass spectrometer was not proposed until 1946 (71). In principle, a TOF has no upper mass range limit, which makes it especially suitable for soft ionization techniques. Other advantages of these instruments include their high transmission efficiency that leads to increased sensitivity and easy mass calibration. The disadvantages include the limited precursor ion selectivity for most MS/MS experiments and the inability to work in MSn mode. TOF is one of the most commonly used mass analyzers for protein analysis, but it can also be used for lipids and eicosanoids (18, 72). Dickinson and Murphy (73) explored the high resolution of a quadrupole time of flight (QTOF) instrument to elucidate the elemental composition of the ions generated by the collisional activation of the carboxylated anion of PGI2. Harks et al. (74) used a QTOF equipped with an ESI probe to identify the production of PGF2a (identified as the sodium adduct ion [M + Na]+) in normal rat kidney fibroblasts.

1.3.2.4. Fourier Transform Ion Cyclotron Resonance

Because biological activity is often regioisomer-specific, it is important to have techniques with the ability to identify complex isomeric oxylipins. Fourier transform ion cyclotron resonance (FTICR/MS) has emerged as a powerful tool for this application. FTICR/MS provides high mass accuracy and high mass resolution to obtain accurate molecular masses and elemental compositions (75, 76). MSn analysis of FTICR/MS can be achieved by sustained off resonance irradiation (SORI)/CID, infrared multiphoton dissociation (IRMPD), and electron capture dissociation (ECD) (77–81). Among these dissociation techniques, ECD provides a softer ionization and gives unique fragments for proteins and

170

Lundström et al.

peptides. SORI/CID and IRMPD yield similar fragmentation patterns, but each method gives specific spectral characteristics that can be useful for compound identification (49). The application of FTICR/MS has been demonstrated for the determination of regioisomers of trihydroxyeicosatrienoic acids (THETAs) (82–84) in biological samples (49). With three hydroxyl groups, liquid chromatographic separation of 11,12,15-, 11,14,15- and 13,14,15-THETA is more difficult and their MS/MS spectra are more complex than HETEs and DHETs. In general, the SORI/CID spectra of these THETAs are similar to IRMPD spectra, suggesting similar major fragmentation pathways for these compounds. The most abundant ions for 11,12,15-, 11,14,15-, and 13,14,15-THETA are m/z 197, 167, and 193, respectively. The secondary unique characteristic ions for these THETAs are m/z 157, 85, and 59, respectively. The different positions of the third hydroxyl group and the third double bond in the structures such as in 11,12,15-THETA and 11,14,15-THETA have a remarkable effect on their major fragmentation pathways (49). The intensities of the m/z 205 in the IRMPD spectrum for 11,14,15-THETA increased twofold while the abundance of m/z 85 decreased as compared with SORI/ CID. For 13,14,15-THETA, the abundance of the m/z 59 and 173 decreased tenfold in the IRMPD spectrum as compared with SORI/CID. LC/FTICR using SORI/CID and IRMPD (without a complete LC separation) identified 11,12,15-THETA and 11,14,15-THETA as arachidonic acid metabolites that were synthesized in rabbit aorta (49). MS/MS analysis of relatively simple HETEs and DHETs has been successfully achieved by ion trap and triple quadrupole mass spectrometers (18, 23, 85–88). In general FTICR/MS generates similar MS/MS spectra for these compounds. However, the high mass accuracy and resolution of FTICR/MS suggest some different molecular fragments for the oxylipins. Characteristic ions indicating the loss of the CO2 group is unique for FTICR/MS spectra. For example, the m/z 149 and 257 ions of 11-HETE are abundant when a FTICR/MS instrument is used. In contrast they have not previously been detected using a triple quadrupole mass spectrometer (18, 49, 86). One example of the powerful ability of FTICR/MS in identifying the fragments using mass accuracy and resolution is the identification of m/z 163 for 12-HETE, 11,12-DHET, and 14,15-DHET. This fragment was previously assigned as the same ion formed by the same fragmentation mechanistic pathway for all compounds (18). From the FTICR/MS spectra, the m/z 163 for 12-HETE and 11,12DHET has masses of 163.11239 and 163.11235, respectively. This ion has the elemental composition of [C11H15O]−. However, the m/z 163 for 14,15-DHET has the mass of 163.14947 with the elemental composition of [C12H19]−. These accurate masses



HPLC/MS/MS-Based Approaches for Detection and Quantification

171

indicate that the fragmentation mechanism of the m/z 163 for 11-HETE/11,12-DHET is different from 14,15-DHET and they are in fact different fragments (49). 1.3.2.5. Hybrid Analyzers

A recent trend in mass analyzer development is to combine different analyzers in sequence to form hybrid instruments. This increases the versatility and enables multiple experiments to be performed. Quadrupole TOF instruments, ion trap FT ion cyclotron resonance instruments, and quadrupole ion trap mass spectrometers, combine the strengths of each analyzer while often avoiding their weaknesses. Accordingly, improved performance can be obtained with a hybrid instrument relative to individual analyzers. Nowadays, the most popular hybrid MS used for eicosanoid analysis is the quadrupole ion trap mass spectrometer (Quad/trap), which provides excellent quantification, thus it is ideal for complex sample analysis. Using a hybrid quadrupole linear ion trap mass spectrometer, Harkewicz et al. (89) were able to detect 70 diverse eicosanoids produced from macrophage-like cells. Following metabolic labeling of the cells with deuterated arachidonic acid (AA-d8) and stimulation with endotoxin, they were able to identify a “doublet” pattern of eicosanoid products. The method enabled sensitive, comprehensive, and quantitative analysis of eicosanoid biology without any previous knowledge or assumptions of the molecular species involved. Using the same method, they were able to identify a series of dihomoprostaglandins produced by endotoxin-stimulated RAW cells (dihomo-(PGE2, PGD2, PGF2a, 11(R)-HETE, PGJ2, and 15-deoxy-∆12,14-PGD2)), further demonstrating the utility of this technique. See Note 1 for more information of hybrid instruments and applications.

1.4. Selective Reaction Monitoring of Eicosanoids

SRM is used in conjunction with LC/ESI/MS to improve the specificity and detection limit for quantification. SRM is a tandem mass spectrometry technique that delivers unique fragment ions that can be monitored and quantified in the midst of a complicated matrix. SRM spectra are highly specific and usually contain only a single peak. This characteristic makes the SRM signal ideal for quantification. In order to use this technique, it is necessary to know the fragmentation patterns of the compounds of interest. A fragment ion is selected from the known mass product spectrum and used for quantification. It is important not to choose a fragment peak ion close to the parent peak, such as dehydrated ions. These type of ions do not have structural significance and can easily be produced equally for isomers (e.g., the range of HETEs), or for unknown contaminants. It is also important to choose a stable peak that consistently appears scan after scan with a constant response.

1.5. Quantitation of Eicosanoids Using HPLC/MS/MS

One of the most important components in developing methods for the quantification of eicosanoids involves the preparation of standards. While beyond the scope of this chapter, we will

172

Lundström et al.

provide a brief summary of some of the more important points. Interested readers are directed to the literature (19, 90–92). 1. Internal standards (IS) are added prior to sample preparation in order to account for variable losses during the preparation steps. The choices of IS are critical. The ideal standard possesses characteristics similar to those of the eicosanoid during the entire experiment (i.e., extraction, chromatography, and mass spectrometry), but does not interfere with the target analyte(s). Many methods employ isotopically labeled versions of the analytes or odd chain fatty acids. However, since there are cost concerns with the use of multiple labeled standards, usually a single IS is employed for a group of eicosanoids with similar physical properties or structure. For example, the deuterated eicosanoid 5-HETE-d8 is an ideal IS for 5-HETE; however, it can also be used for other HETEs (19). Other examples include 10(11)-epoxyheptadecanoic acid (10(11)-EpHep), which is an odd chain length IS for EETs and 10,11-dihydroxydecanoic acid (10,11-DHN), which is appropriate for DHETEs (3, 7, 27). It is notable that the IS concentration does not need to be accurate, but that it is crucial that the exact same volume is used for both the samples and the primary standards (19). An important precaution is that sometimes deuterated standards can contain a small percent of nondeuterated analyte. Consequently, it is important to also run all IS separately, in order to obtain information on purity (19). To account for changes in volume, instrument variability, and matrix interferences, an additional technical standard (for example N-cyclohexyl-N'dodecanoic acid urea [CUDA]) can be added prior to analysis (7, 27). The technical standard should be a stable compound that does not interfere with other standards and analytes. 2. Primary standards (PS) are prepared for quantification purposes in order to create calibration curves (or standard curves). Consequently the concentration of PS must be known with high accuracy. Calibration curves are generated by the addition of defined amounts of analyte (unlabeled) followed by spiking with an aliquot of the IS (3, 19, 21, 90). The calibration curves are based on the ratio of standard peak area to the IS peak area in the primary standards plotted vs. the amount of primary standard. The concentrations of the standards are generally selected such that the resultant curve follows a linear regression in the desired detection range. The slope of the curve is a measurement of the sensitivity of the quantification method. The detection range is restricted on the low end by the detection limit of the MS and the upper limit is restricted by ion self-suppression and detection saturation issues. Standards should be run before and after each set of samples in order to monitor for changes in instrument performance during the run.



HPLC/MS/MS-Based Approaches for Detection and Quantification

173

The ratio of the unknown analyte peak area to IS peak area in each sample is then compared to the respective standard curve in order to calculate the amount of analyte. 3. In addition to using standards for quantification, compound recoveries have to be estimated during sample preparation and instrumental analysis. This can be achieved by preparing a set of samples where a known amount of the analytes is added to the matrix (e.g., phosphate buffered saline (PBS) solution in the protocol below). The samples are then analyzed by LC/ MS/MS and the eicosanoid peak intensities are compared to those standards that have not been through the sample preparation procedure. One can also compare the standards with samples in which only the matrix alone has gone through the sample preparation steps. The aliquot of the standard mixture is then added in the last step prior analysis. Recovery values in the range of 75–120% are preferred (19). 4. Even though the procedures described above limit the effect of systematic errors during quantification, random errors will still occur. Thus, there will always be a difference between the reported and “true” value, and since the possibility of error increases close to the detection limit a formal determination of the LOD is recommended. A common preconception in analytical chemistry is that the signal (S) has to be three times larger than the background noise (N), resulting in the LOD being equivalent to a S/N ratio of 3. However, the formal definition of the LOD is the concentration of analyte required to give a signal equal to the background noise (blank) plus three times the standard deviation (STD) of the blank (3). Similarly the LOQ is defined as the concentration of analyte required to give a signal equal to that of the blank plus ten times the standard deviation of the blank (3, 93). Importantly, sample preparations prior to analyses always involve extra factors of error and sample loss. Therefore the overall method detection limit (MDL) can be estimated by determining the STD across seven or more technical replicates of the sample for a concentration within the range or 1–5 times the expected LOD. Subsequently, the MDL can be calculated by multiplying the STD with the Student’s t value for the appropriate degree of freedom and the 99% confidence limit (2.143 for n = 7). A selection of t values can be found at http://www. itl.nist.gov/div898/handbook/eda/section3/eda3672.htm (92). The detection limits depend upon the instrument and configuration. Furthermore it depends on the number of eicosanoids and classes of eicosanoids that are being quantified. 5 and 20 fmol and LOQ values between 5 and 50 fmol have been reported for the quantification of prostaglandin glyceryl esters (48). Recently a ethod was presented where

174

Lundström et al.

over 70 diverse eicosanoids were screened with LODs in the range of 2–25,000 fmol (19, 89).

2. Materials 2.1. Equipment

1. Surveyor HPLC system (Thermo Finnigan, San Jose, CA, USA) equipped with a Surveyor Autosampler Plus and a Surveyor MS Pump Plus (four solvent channels). 2. LCQ Deca ion trap (Thermo Finnigan) working with an electrospray ionization source equipped with a stainless steel needle kit (Thermo Fisher Scientific, Waltan, MA, USA). 3. Optiplex 755 computer (Dell, Round Rock, TX, USA) equipped with a Windows XP operating system. The system software Xcalibur™ 2.0 SR2 (Thermo Finnigan) was used for data processing.

2.2. Reagents

1. LC/MS grade acetonitrile and methanol and HPLC grade glacial acetic acid were obtained from Fischer Scientific (Loughborough, LE, UK). All other solvents were acquired from Rathburn Chemicals Ltd. (Walkerburn, Scotland). 2. 20-Hydroxy-5Z,8Z,11Z,14Z-eicosatetraenoic acid (20-HETE) and N-cyclohexyl-N¢-dodecanoic acid urea (CUDA) were obtained from Cayman Chemical (Ann Arbor, MI, USA). All other standards were obtained from Biomol (Plymouth, PA, USA).

2.3. Supplies

1. Luna C18 column (2.0 × 150 mm, 5 mm) (Phenomenex, Allerod, Denmark). 2. C18 security guard cartridges (4 × 2 mm) (Phenomenex). 3. 2 mL amber glass autosampler vials with plastic screw caps (8 mm) and Teflon-faced seals (8 mm) (Chromacol, Thermo Fisher Scientific). 4. 150 mL polyspring insert for 2 mL autosampler vials (Thermo Fisher scientific). 5. 2 mL amber glass vials with plastic crew caps (8 mm) and Teflonfaced seals (8 mm) (Chromacol, Thermo Fisher Scientific). 6. Glass mL syringes (10, 100, and 500 mL) with PTFE-tipped plungers (Hamilton Company, Bonaduz, Switzerland). 7. 2 mL amber volumetric flasks (Volac) (Poulten & Graf Ltda, Barking, UK).



HPLC/MS/MS-Based Approaches for Detection and Quantification

175

8. Fluoropore™ PTFE membrane filters (47 mm, 0.2 mm) (Millipore, Billerica, MA, USA). 9. Nylon membrane filters (47 mm, 0.2 mm) (Millipore).

3. Methods 3.1. Protocol

Eicosanoids are frequently measured in a variety of biological matrices including urine, cell culture, plasma, and BALF(3, 7, 27, 32). Several different methods for sample preparation are available (3, 7, 21, 27, 32, 94–96). Generally sample preparation procedures are straight forward, with the exception of biological tissues, which require prior extraction steps (21, 23, 96, 97). A protocol optimized for the preparation of BALF samples for LC/ MS/MS analysis is given below, however the protocol can be readily adapted to other body fluids with slight modifications (7, 27). The extraction method is based upon methods developed by Hammock and coworkers (3, 5, 7, 27, 98) and is described in detail in a recent publication by Yang et al. (99). Subsequently, a set of oxylipin standards was analyzed under various conditions in order to provide an example of the experimental optimizations that need to be performed prior to analysis of biological samples (see Subheading 3.2). For limitations of this method see Note 2.

3.1.1. Offline Solid Phase Extraction

1. Clean Oasis HLB offline solid phase extraction (SPE) 60 mg columns with one column volume of ethyl acetate followed by two column volumes of methanol. Precondition columns with two column volumes of H2O/MeOH (95:5) in 0.1% acetic acid (wash solution). Elute the solvents by gravity and avoid letting the sorbents go dry. 2. Thaw BALF aliquots (4 mL) before extraction. Add 200 mL wash solution to columns, 10 mL of IS (400 nM/standard), and 10 mL antioxidant and enzyme inhibitor solution (0.2 mg/ mL of butylhydroxytoluene (BHT), ethylenediaminetetra acetic acid (EDTA), thiamine pyrophosphate (TPP), and indomethacin). Apply samples to the columns. In addition to samples, prepare matrix blanks (~2 blanks/30 samples) by using 0.1 M PBS instead of BALF. 3. Wash by gravity with two column volumes of wash solution, vacuum dry for 10 min at 5 psi. 4. Elute eicosanoids with 0.5 mL methanol followed by 1.5 mL ethyl acetate. Collect in 2 mL eppendorf tubes with 6 mL 30% glycerol in methanol. 5. Use speed vacuum or nitrogen stream to remove solvent until only the glycerol plug remains.

176

Lundström et al.

6. Cap and freeze samples for storage at −80°C or continue to step 7. 7. Prior to LC/MS analysis, reconstitute residues to 50 mL with 200 nM CUDA, centrifuge and transfer supernatant to glass autosampler vials. 3.1.2. Standards, Solvents, and MS Optimization Preparations

1. For method development purposes ionization optimization standard solutions were prepared for all analytes as follows: 1–10 mL of the standard stock solution (0.05–0.5 mg/mL) were transferred to 2 mL amber glass vials with screw caps and Teflon-faced seals and then diluted in methanol to a final volume of 1 mL (0.5 ng/mL). Similarly a chromatography optimization solution containing all standards was prepared by adding 2–20 mL/standard of the stock solutions to a 2 mL amber glass volumetric flask. The standard mixture was then diluted with methanol to a final concentration of 0.5 ng/mL. See Note 3 for further comments on sample treatments. 2. LC/MS grade acetonitrile and methanol were vacuum filtered (0.2-mm PTFE filter) and prepared in a ratio of 88:12. For preparation of the aqueous phase, MilliQ water was vacuum filtered (0.2-mm Nylon filter). In addition to the neutral aqueous (1) and organic (2) phases, an acidic phase (3) containing (1) and (2) in the ratio of 85:15 was prepared with an additive of 0.1% glacial HPLC graded acidic acid. See Note 4 for more comments on solvent treatments. 3. In order to optimize the MS parameters for optimal peak detection the ionization optimization standard solutions (0.5 ng/mL) were introduced into the MS source directly using a syringe pump. The flow rate was set to 10 mL/min and T-coupled to the HPLC in order to achieve a flow rate of 350 mL/min. The LC gradient was set to a similar ratio of solvent (1) and (2) as when the respective compound was known to elute from the column. By recording and comparing intensity changes in the molecular ion to changed parameters of the MS (e.g., heated capillary temperature, capillary voltages, sheath gas flow, etc.) optimal conditions could be obtained and saved (as tune files). Compounds were also fragmented in order to determine which fragment ions to select for SRM analysis (based on abundance and specificity). An instrument method was developed based on the compounds’ retention times, tune files, and selected fragment ions. The chromatographic run was divided into different time segments in order to reduce the number of scan events over time, subsequently improving the detection limit (see Note 5). The chromatography optimization solution was repeatedly run over time in sequences with up to 15 injections in order to determine the reproducibility of the method. It is vital that the system is



HPLC/MS/MS-Based Approaches for Detection and Quantification

177

reproducible, with little fluctuation in retention times and peak areas (see Note 6). 3.1.3. Instrument Operation Parameters

HPLC: Autosampler temperature: 10°C Column temperature: 40°C Mobile Phase: (1) MilliQ water 18 MW (2) Acetonitrile:Methanol (88:12) (3) Water:Acetonitrile:Methanol (85:13:2), 0.1% Acetic acid Gradient: Parameters are described in Table 1 for a set of 24 compounds. Flow rate: 350 mL/min Column temperature: 40°C LCQ Ion Trap: Capillary temperature: 300°C Collision energy (resonance energy, percent of 5 V): 40–50% Spray voltage: 4 kV Sheath gas flow rate: 70 arbitrary U Auxiliary gas flow rate: 20 arbitrary U Scan mode: Negative SRM: parameters are described in Table 2.

Table 1  Gradient parameters for Fig. 2 Time (min)

1 (%)a

2 (%)b

3 (%)c

0.0

85

15

0

0.5

85

15

0

2.0

70

30

0

8.0

45

55

0

28.0

25

75

0

28.5

0

100

0

34.0

0

100

0

37.0

0

0

100

45.0

0

0

100

1: Water. 2: Acetonitrile:Methanol (88:12). c 3: Water:Acetonitrile:Methanol (85:13:2), 0.1% Acetic acid. Solvent 3 is used to precondition the column prior to the next run. a

b

178

Lundström et al.

Table 2  Method parameters for analyzed eicosanoids Transition (m/z)

Collision energy (%)

LOD (pg/mL)

7.0

369 > 315

40

0.5

20-hydroxy-LTB4

7.4

351 > 195

40

0.5

20-carboxy-LTB4

7.4

365 > 347

40

2.5

TBX2

8.1

369 > 195

40

0.5

PGF2a

8.6

353 > 309

40

0.5

11-dehydro-TBX2

9.3

367 > 305

40

0.5

LXA4

10.4

351 > 115

40

5.0

LTB5

11.5

333 > 195

40

0.5

LTD4

11.7

495 > 177

40

0.5

LTE4

12.0

438 > 333

50

250

15-dPGJ2

16.9

315 > 271

40

25

20-HETE

17.0

319 > 275

40

5.0

13-HODE

18.6

295 > 195

40

0.5

9-HODE

18.9

295 > 171

40

0.5

15-HETE

19.4

319 > 219

40

2.5

15-KETE

20.6

317 > 113

40

0.5

12-HETE

21.0

319 > 179

40

0.5

5-HETE

22.6

319 > 115

40

0.5

14,15-EET

23.3

319 > 219

40

2.5

11,12-EET

25.0

319 > 167

40

25

5-KETE

25.1

317 > 273

40

25

8,9-EET

25.1

319 > 155

40

2.5

DHA

30.9

327 > 283

50

5.0

AA

31.3

303 > 259

40

5.0

Segment

Analyte

1

6-keto-PGF1a

2

3

4

3.2. Results

Rt (min)

1. We compared differences in separation and detection of eicosanoids under acidic (pH 4), neutral, and basic (pH 8) conditions (Fig. 1a–c, Table 3). Results clearly showed that analytes demonstrated improved performance (i.e., narrower peak shapes, reduced tailing, and improved baseline resolution) under weak acidic conditions. However, a basic mobile phase can potentially be useful for certain eicosanoid targets.



HPLC/MS/MS-Based Approaches for Detection and Quantification

179

Fig. 1. Chromatographic separation of eight eicosanoids ((1) 6-keto PFG1a, (2) LTB4, (3) 15-HETE, (4) 5-KETE, (5) 12-HETE, (6) 5-HETE, (7) 14,15-EET, and (8) 11,12-EET). Gradient parameters are described in Table 3. (a) Gradient run with an additive of 0.1% Acetic acid. (b) Gradient run with an additive of 0.01% NH4+solution (25%). *This peak contains all eight eicosanoids. (c) Gradient run without an additive (neutral conditions). (d) Gradient run without additive, but preconditioned under the same acidic conditions as in a.

Optimal resolution was achieved after preconditioning of the column at acidic pH followed by a neutral gradient (Fig. 1d). These conditions were therefore used for further development of the method. 2. Twenty-four arachidonic and linoleic acid metabolites were analyzed by LC/MS/MS using a 45 min pH gradient (Table 1) and scanning in SRM mode (Table 2). LOD values from 0.5 to 250 pg/mL were observed (Table 2). Figure 2 shows a total ion chromatogram of the analyzed compounds and ion chromatograms obtained for eicosanoids detected 6–10 min into the run. Compounds with similar properties (i.e., PGs, LTs, HETEs, HODEs, KETEs, and EETs) have close retention times but can readily be distinguished by differences in fragmentation pattern. An example of this is shown in Fig. 3

180

Lundström et al.

Table 3  Gradient parameters for Fig. 1a Time (min)

1 (%)b

2 (%)c

3 (%)d

0.0

85

15

0

0.5

85

15

0

1.0

70

30

0

5.0

45

55

0

10.0

25

75

0

10.5

0

100

0

15.0

0

100

0

16.0

85

e

15

100f

21.0

85e

15e

100f

e

a  Solvent (1) and (2) were run under neutral conditions or with an additive of either 0.1% Acetic acid or 0.01% NH4+(25% aqueous solution) depending on desired pH with a flow rate of 350 mL/min. b  1: Water. c  2: Acetonitrile:Methanol (88:12). d  3: Water:Acetonitrile:Methanol (85:13:2), 0.1% Acetic acid. e  Used parameters for chromatograms a–c in Fig. 1. f  Used parameters for chromatogram d in Fig. 1.

Fig. 2. ESI/MS/MS chromatogram obtained for the eicosanoids in Table 2. Details for all SRM parameters are provided in Table 2. (a) Total ion chromatogram. (b–g) Compounds detected in Segment 1: (b) 6-keto PGF1a, (c) 20-hydroxy-LTB4, (d) 20-carboxy-LTB4, (e) TXB2, (f) PGF2a, and (g) 11-dehydro-TXB2.



HPLC/MS/MS-Based Approaches for Detection and Quantification

181

Fig. 3. MS2 analysis of 5-HETE (m/z 319) (a) and of 12-HETE (m/z 319) (b). The fragment ions at m/z 115 and 179 are specific for 5-HETE and 12-HETE, respectively, and can be used in SRM analysis for compound identification and subsequent quantification.

in which the MS2 spectra of 5-HETE (a) and 12-HETE (b) are shown. The dehydrated ion at m/z 301 is present in both spectra, as well as the ion at m/z 257. However, scanning in SRM mode for the diagnostic fragment peaks at m/z 115 for 5-HETE and at m/z 179 for 12-HETE enabled the routine detection and quantification of these compounds.

4. Notes 1. The use of new instrumentation such as a quadrupole ion trap time of flight (QITTOF) or FTICR/MS will result in lower LODs in the range of attomol or even zeptomol

182

Lundström et al.

levels of constituent quantities. The use of these techniques for eicosanoid quantification can greatly increase our ability to detect and quantify a variety of eicosanoids in biological samples that have not been detected previously. This ability is important for clinical applications where the amount of sample available is often a limiting factor. Furthermore, the development of multi-isotope imaging mass spectrometry (MIMS) opens a new field of opportunities, enabling studies of the interaction, detection, and quantification of eicosanoids within a single cell. 2. There are specific challenges with the quantification of CysLTs, which are best analyzed directly by online SPE coupled to the HPLC/MS/MS run in positive mode (7, 95). This procedure is recommended because CysLTs are sensitive to oxidation and offline SPE extraction can result in decreased recoveries. 3. Eicosanoids are sensitive to oxidation and consequently standards should always be stored at low temperatures (e.g., £ −80°C) under an inert atmosphere (e.g., argon) in the absence of light, moisture, and active surfaces. Samples should be kept in a temperature-controlled environment both before as well as during the run (a common feature of many autosamplers). 4. It is important to use HPLC and MS grade solvents for the preparation of chromatographic mobile phases to reduce background noise and keep the instrument in a good working condition. The aqueous phase should be frequently replaced (once per week) in order to reduce the possibility of microorganism growth. In addition, the system should be regularly flushed with organic solvent (e.g., methanol). 5. The limit for the number of compounds that can be identified in a single run depends upon the HPLC separation as well as the ion dwell times (based on the number of analytes in each SRM, the interscan delays, and the peak width of the analytes). Accordingly, this number will vary greatly with different instruments. 6. System stability is particularly important when operated with large sample sets where the MS conducts multiple scans within a single run.

Acknowledgments We gratefully acknowledge the assistance of Malin Nording, Katrin Georgi, Jun Yang, and Pavel Aronov for helpful discussions and critical reading of the manuscript. This research was



HPLC/MS/MS-Based Approaches for Detection and Quantification

183

supported by The Åke Wibergs Stiftelse, the Fredrik and Ingrid Thurings Stiftelse, The Royal Swedish Academy of Sciences, The Swedish Council for Strategic Research, and The Swedish Research Council and the European Commission. C.E.W. was supported by a fellowship from the Centre for Allergy Research (Cfa). S. L. L. was supported by the Osher Initiative for severe Asthma. References 1. Funk CD. (2001) Prostaglandins and leukotrienes: advances in eicosanoid biology. Science 294, 1871–1875. 2. Conrad DJ. (1999) The arachidonate 12/15 lipoxygenases. A review of tissue expression and biologic function. Clin Rev Allergy Immunol 17, 71–89. 3. Newman JW, Watanabe T, Hammock BD. (2002) The simultaneous quantification of cytochrome P450 dependent linoleate and arachidonate metabolites in urine by HPLCMS/MS. J Lipid Res 43, 1563–1578. 4. Natarajan R, Reddy MA. (2003) HETEs/ EETs in renal glomerular and epithelial cell functions. Curr Opin Pharmacol 3, 198–203. 5. Schmelzer KR, Inceoglu B, Kubala L, et  al. (2006) Enhancement of antinociception by coadministration of nonsteroidal anti-inflammatory drugs and soluble epoxide hydrolase inhibitors. Proc Natl Acad Sci U S A 103, 13646–13651. 6. Samuelsson B, Dahlen SE, Lindgren JA, Rouzer CA, Serhan CN. (1987) Leukotrienes and lipoxins: structures, biosynthesis, and biological effects. Science 237, 1171–1176. 7. Schmelzer KR, Wheelock AM, Dettmer K, Morin D, Hammock BD. (2006) The role of inflammatory mediators in the synergistic toxicity of ozone and 1-nitronaphthalene in rat airways. Environ Health Perspect 114, 1354–1360. 8. Hoagland KM, Maier KG, Moreno C, Yu M, Roman RJ. (2001) Cytochrome P450 metabolites of arachidonic acid: novel regulators of renal function. Nephrol Dial Transplant 16, 2283–2285. 9. Roman RJ. (2002) P-450 metabolites of arachidonic acid in the control of cardiovascular function. Physiol Rev 82, 131–185. 10. Natarajan R, Nadler JL. (2004) Lipid inflammatory mediators in diabetic vascular disease. Arterioscler Thromb Vasc Biol 24, 1542–1548. 11. Hussey HJ, Tisdale MJ. (1996) Inhibition of tumour growth by lipoxygenase inhibitors. Br J Cancer 74, 683–687.

12. Nie D, Tang K, Szekeres K, Trikha M, Honn KV. (2000) The role of eicosanoids in tumor growth and metastasis. Ernst Schering Res Found Workshop 31, 201–217. 13. Kumlin M. (1996) Analytical methods for the measurement of leukotrienes and other eicosanoids in biological samples from asthmatic subjects. J Chromatogr A 725, 29–40. 14. Powell WS, Funk CD. (1987) Metabolism of arachidonic acid and other polyunsaturated fatty acids by blood vessels. Prog Lipid Res 26, 183–210. 15. Tsukamoto H, Hishinuma T, Mikkaichi T, et  al. (2002) Simultaneous quantification of prostaglandins, isoprostane and thromboxane in cell-cultured medium using gas chromatography-mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci 774, 205–214. 16. Tsikas D. (1998) Application of gas chromatography-mass spectrometry and gas chromatography-tandem mass spectrometry to assess in  vivo synthesis of prostaglandins, thromboxane, leukotrienes, isoprostanes and related compounds in humans. J Chromatogr B Biomed Sci Appl 717, 201–245. 17. Baranowski R, Pacha K. (2002) Gas chromatographic determination of prostaglandins. Mini Rev Med Chem 2, 135–144. 18. Murphy RC, Barkley RM, Zemski Berry K, et al. (2005) Electrospray ionization and tandem mass spectrometry of eicosanoids. Anal Biochem 346, 1–42. 19. Deems R, Buczynski MW, Bowers-Gentry R, Harkewicz R, Dennis EA. (2007) Detection and quantitation of eicosanoids via high performance liquid chromatography-electrospray ionization-mass spectrometry. Methods Enzymol 432, 59–82. 20. Blewett AJ, Varma D, Gilles T, Libonati JR, Jansen SA. (2008) Development and validation of a high-performance liquid chromatography-electrospray mass spectrometry method for the simultaneous determination of 23 eicosanoids. J Pharm Biomed Anal 46, 653–662.

184

Lundström et al.

21. Masoodi M, Nicolaou A. (2006) Lipidomic analysis of twenty-seven prostanoids and isoprostanes by liquid chromatography/electrospray tandem mass spectrometry. Rapid Commun Mass Spectrom 20, 3023–3029. 22. Wilson ID, Plumb R, Granger J, Major H, Williams R, Lenz EM. (2005) HPLC-MSbased methods for the study of metabonomics. J Chromatogr B Analyt Technol Biomed Life Sci 817, 67–76. 23. Zhang JH, Pearson T, Matharoo-Ball B, et  al. (2007) Quantitative profiling of epoxyeicosatrienoic, hydroxyeicosatetraenoic, and dihydroxyeicosatetraenoic acids in human intrauterine tissues using liquid chromatography/electrospray ionization tandem mass spectrometry. Anal Biochem 365, 40–51. 24. Han X, Gross RW. (2005) Shotgun lipidomics: electrospray ionization mass spectrometric analysis and quantitation of cellular lipidomes directly from crude extracts of biological samples. Mass Spectrom Rev 24, 367–412. 25. Wheelan P, Murphy RC. (1997) Quantitation of 5-lipoxygenase products by electrospray mass spectrometry: effect of ethanol on zymosan-stimulated production of 5-lipoxygenase products by human neutrophils. Anal Biochem 244, 110–115. 26. Yue H, Jansen SA, Strauss KI, et  al. (2007) A liquid chromatography/mass spectrometric method for simultaneous analysis of arachidonic acid and its endogenous eicosanoid metabolites prostaglandins, dihydroxyeicosatrienoic acids, hydroxyeicosatetraenoic acids, and epoxyeicosatrienoic acids in rat brain tissue. J Pharm Biomed Anal 43, 1122–1134. 27. Luria A, Weldon SM, Kabcenell AK, et  al. (2007) Compensatory mechanism for homeostatic blood pressure regulation in Ephx2 genedisrupted mice. J Biol Chem 282, 2891–2898. 28. Bylund J, Ericsson J, Oliw EH. (1998) Analysis of cytochrome P450 metabolites of arachidonic and linoleic acids by liquid chromatography-mass spectrometry with ion trap MS. Anal Biochem 265, 55–68. 29. Masoodi M, Mir AA, Petasis NA, Serhan CN, Nicolaou A. (2008) Simultaneous lipidomic analysis of three families of bioactive lipid mediators leukotrienes, resolvins, protectins and related hydroxy-fatty acids by liquid chromatography/electrospray ionisation tandem mass spectrometry. Rapid Commun Mass Spectrom 22, 75–83. 30. Hishinuma T, Suzuki K, Saito M, et al. (2007) Simultaneous quantification of seven prostanoids using liquid chromatography/tandem mass spectrometry: the effects of arachidonic acid on prostanoid production in mouse bone

marrow-derived mast cells. Prostaglandins Leukot Essent Fatty Acids 76, 321–329. 31. Oliw E, Wollheim FA. (2000) [New treatment of pain and fever in rheumatoid arthritis and arthrosis. The first cyclooxygenase-2 inhibitors show promising results]. Lakartidningen 97, 2942–2944. 32. Kempen EC, Yang P, Felix E, Madden T, Newman RA. (2001) Simultaneous quantification of arachidonic acid metabolites in cultured tumor cells using high-performance liquid chromatography/electrospray ionization tandem mass spectrometry. Anal Biochem 297, 183–190. 33. Yang P, Chan D, Felix E, et al. (2006) Determination of endogenous tissue inflammation profiles by LC/MS/MS: COX- and LOXderived bioactive lipids. Prostaglandins Leukot Essent Fatty Acids 75, 385–395. 34. Porter NA, Caldwell SE, Mills KA. (1995) Mechanisms of free radical oxidation of unsaturated lipids. Lipids 30, 277–290. 35. Yin H, Porter NA. (2005) New insights regarding the autoxidation of polyunsaturated fatty acids. Antioxid Redox Signal 7, 170–184. 36. Williams MV, Lee SH, Blair IA. (2005) Liquid chromatography/mass spectrometry analysis of bifunctional electrophiles and DNA adducts from vitamin C mediated decomposition of 15-hydroperoxyeicosatetraenoic acid. Rapid Commun Mass Spectrom 19, 849–858. 37. Hammonds TD, Blair IA, Falck JR, Capdevila JH. (1989) Resolution of epoxyeicosatrienoate enantiomers by chiral phase chromatography. Anal Biochem 182, 300–303. 38. Lee SH, Williams MV, DuBois RN, Blair IA. (2003) Targeted lipidomics using electron capture atmospheric pressure chemical ionization mass spectrometry. Rapid Commun Mass Spectrom 17, 2168–2176. 39. Lee SH, Blair IA. (2007) Targeted chiral lipidomics analysis by liquid chromatography electron capture atmospheric pressure chemical ionization mass spectrometry (LC-ECAPCI/ MS). Methods Enzymol 433, 159–174. 40. Lee SH, Williams MV, Blair IA. (2005) Targeted chiral lipidomics analysis. Prostaglandins Other Lipid Mediat 77, 141–157. 41. Suzuki N, Hishinuma T, Saga T, et  al. (2003) Determination of urinary 12(S)hydroxyeicosatetraenoic acid by liquid chromatography-tandem mass spectrometry with column-switching technique: sex difference in healthy volunteers and patients with diabetes mellitus. J Chromatogr B Analyt Technol Biomed Life Sci 783, 383–389.



HPLC/MS/MS-Based Approaches for Detection and Quantification

42. Wei S, Brittin JJ, Falck JR, et al. (2006) Chiral resolution of the epoxyeicosatrienoic acids, arachidonic acid epoxygenase metabolites. Anal Biochem 352, 129–134. 43. Kiss L, Bier J, Roder Y, Weissmann N, Grimminger F, Seeger W. (2008) Direct and simultaneous profiling of epoxyeicosatrienoic acid enantiomers by capillary tandem column chiralphase liquid chromatography with dual online photodiode array and tandem mass spectrometric detection. Anal Bioanal Chem 392, 717–726. 44. Dole M, Mack, L.L., Hines, R.L., Mobley, R.C., Ferguson, L.D. and Alice, M.B.. (1968) Molecular beams of macroions. J Chem Phys 49, 2240–2249. 45. Fenn JB, Mann M, Meng CK, Wong SF, Whitehouse CM. (1989) Electrospray ionization for mass spectrometry of large biomolecules. Science 246, 64–71. 46. Taylor G. (1964) Disintegration of water drops in an electric field. Proc R Soc London, A 280, 383–397. 47. Cech NB, Enke CG. (2001) Practical implications of some recent studies in electrospray ionization fundamentals. Mass Spectrom Rev 20, 362–387. 48. Kingsley PJ, Rouzer CA, Saleh S, Marnett LJ. (2005) Simultaneous analysis of prostaglandin glyceryl esters and prostaglandins by electrospray tandem mass spectrometry. Anal Biochem 343, 203–211. 49. Cui L, Isbell MA, Chawengsub Y, Falck JR, Campbell WB, Nithipatikom K. (2008) Structural characterization of monohydroxyeicosatetraenoic acids and dihydroxy- and trihydroxyeicosatrienoic acids by ESI-FTICR. J Am Soc Mass Spectrom 19, 569–585. 50. Horning EC, Horning MG, Carroll DI, Dzidic I, Stillwell RN. (1973) New picogram detection system based on a mass spectrometer with an external ionization source at atmospheric pressure. Anal Chem 45, 936–943. 51. Caroll DI, Dzidic I, Stillwell RN, Haegele KD, Horning EC. (1975) Atmospheric pressure ionization mass spectrometry: Corona discharge ion source for use in liquid chromatograph-mass spectrometer-computer analytical system. Anal Chem 47, 2369–2373. 52. Montuschi P, Martello S, Felli M, Mondino C, Chiarotti M. (2004) Ion trap liquid chromatography/tandem mass spectrometry analysis of leukotriene B4 in exhaled breath condensate. Rapid Commun Mass Spectrom 18, 2723–2729. 53. Koga D, Santa T, Fukushima T, Homma H, Imai K. (1997) Liquid chromatographicatmospheric pressure chemical ionization

185

mass spectrometric determination of anandamide and its analogs in rat brain and peripheral tissues. J Chromatogr B Biomed Sci Appl 690, 7–13. 54. Rosenberg E. (2003) The potential of organic (electrospray- and atmospheric pressure chemical ionisation) mass spectrometric techniques coupled to liquid-phase separation for speciation analysis. J Chromatogr A 1000, 841–889. 55. Cai SS, Syage JA. (2006) Comparison of atmospheric pressure photoionization, atmospheric pressure chemical ionization, and electrospray ionization mass spectrometry for analysis of lipids. Anal Chem 78, 1191–1199. 56. Locke DC, Dhingra BS, Baker AD. (1982) Liquid-phase photoionization detector for liquid chromatography. Anal Chem 54, 447–450. 57. Robb DB, Covey TR, Bruins AP. (2000) Atmospheric pressure photoionization: an ionization method for liquid chromatography-mass spectrometry. Anal Chem 72, 3653–3659. 58. Cai SS, Syage JA. (2006) Atmospheric pressure photoionization mass spectrometry for analysis of fatty acid and acylglycerol lipids. J Chromatogr A 1110, 15–26. 59. Short LC, Cai SS, Syage JA. (2007) APPIMS: effects of mobile phases and VUV lamps on the detection of PAH compounds. J Am Soc Mass Spectrom 18, 589–599. 60. Ferguson RE, MCCulloh KE, Rosenstock HM. (1965) Observation of the products of ionic collision processes and ion decomposition in a linear, pulsed time-of-flight mass spectrometer. J Chem Phys 42, 100. 61. Paul W, Steinwedel HS. (1960) US Patent, 2939952. 62. Kita Y, Takahashi T, Uozumi N, Shimizu T. (2005) A multiplex quantitation method for eicosanoids and platelet-activating factor using column-switching reversed-phase liquid chromatography-tandem mass spectrometry. Anal Biochem 342, 134–143. 63. Paul W. (1990) Electromagnetic traps for charged and neutral particles (nobel lecture). Angew Chem 29, 739–748. 64. Stafford Jr GC, Kelley PE, Syka JEP, Reynolds WE, Todd JFJ. (1984) Recent Improvements in Analytical Application of Advanced Ion-Trap Technology. Int J Mass Spectrom Ion Processes 60, 85–98. 65. Louris J, Brodbelt-Lustig J, Cooks R, Glish G, Van Berkel G, McLuckey S. (1990) Ion isolation and sequential stages of mass spectrometry in

186

Lundström et al.

a quadrupole ion trap mass spectrometer. Int J Mass Spectrom Ion Processes 96, 117–137. 66. Liminga M, Oliw E. (2000) Qualitative and quantitative analysis of lipoxygenase products in bovine corneal epithelium by liquid chromatography-mass spectrometry with an ion trap. Lipids 35, 225–232. 67. Kiss L, Roder Y, Bier J, Weissmann N, Seeger W, Grimminger F. (2008) Direct eicosanoid profiling of the hypoxic lung by comprehensive analysis via capillary liquid chromatography with dual online photodiode-array and tandem mass-spectrometric detection. Anal Bioanal Chem 390, 697–714. 68. Hevko JM, Murphy RC. (2001) Electrospray ionization and tandem mass spectrometry of cysteinyl eicosanoids: leukotriene C4 and FOG7. J Am Soc Mass Spectrom 12, 763–771. 69. Hong S, Lu Y, Yang R, Gotlinger KH, Petasis NA, Serhan CN. (2007) Resolvin D1, protectin D1, and related docosahexaenoic acidderived products: Analysis via electrospray/ low energy tandem mass spectrometry based on spectra and fragmentation mechanisms. J Am Soc Mass Spectrom 18, 128–144. 70. Thomson JJ. (1897) Cathode rays. Philos Mag 44, 29. 71. Stephens WE. (1946) Pulsed mass spectrometer with time dispersion. Phys Rev 69, 691. 72. Shui G, Bendt AK, Pethe K, Dick T, Wenk MR. (2007) Sensitive profiling of chemically diverse bioactive lipids. J Lipid Res 48, 1976–1984. 73. Dickinson JS, Murphy RC. (2002) Mass spectrometric analysis of leukotriene A4 and other chemically reactive metabolites of arachidonic acid. J Am Soc Mass Spectrom 13, 1227–1234. 74. Harks EG, Peters PH, van Dongen JL, van Zoelen EJ, Theuvenet AP. (2005) Autocrine production of prostaglandin F2alpha enhances phenotypic transformation of normal rat kidney fibroblasts. Am J Physiol Cell Physiol 289, C130–137. 75. Laude DA. Electrospray Ionization/Fourier Transform Ion Cyclotron Resonance Mass Spectrometry. In: Cole RB, ed. Elctrospray Ionization Mass Spectrometry: Fundamentals, Instrumentation, and Applications. New York: Wiley, 1997:291–320. 76. Page JS, Masselon CD, Smith RD. (2004) FTICR mass spectrometry for qualitative and quantitative bioanalyses. Curr Opin Biotechnol 15, 3–11. 77. Wu Q, Van Orden S, Cheng X, Bakhtiar R, Smith RD. (1995) Characterization of cytochrome c variants with high-resolution FTICR mass spectrometry: correlation of

fragmentation and structure. Anal Chem 67, 2498–2509. 78. Flora JW, Hannis JC, Muddiman DC. (2001) High-mass accuracy of product ions produced by SORI-CID using a dual electrospray ionization source coupled with FTICR mass spectrometry. Anal Chem 73, 1247–1251. 79. Hofstadler SA, Sannes-Lowery KA, Griffey RH. (1999) Infrared multiphoton dissociation in an external ion reservoir. Anal Chem 71, 2067–2070. 80. Stensballe A, Jensen ON, Olsen JV, Haselmann KF, Zubarev RA. (2000) Electron capture dissociation of singly and multiply phosphorylated peptides. Rapid Commun Mass Spectrom 14, 1793–1800. 81. Flora JW, Muddiman DC. (2004) Determination of the relative energies of activation for the dissociation of aromatic versus aliphatic phosphopeptides by ESI-FTICR-MS and IRMPD. J Am Soc Mass Spectrom 15, 121–127. 82. Pfister SL, Spitzbarth N, Nithipatikom K, Edgemond WS, Falck JR, Campbell WB. (1998) Identification of the 11,14,15- and 11,12, 15-trihydroxyeicosatrienoic acids as endothelium-derived relaxing factors of rabbit aorta. J Biol Chem 273, 30879–30887. 83. Pfister SL, Spitzbarth N, Zeldin DC, Lafite P, Mansuy D, Campbell WB. (2003) Rabbit aorta converts 15-HPETE to trihydroxyeicosatrienoic acids: potential role of cytochrome P450. Arch Biochem Biophys 420, 142–152. 84. Falck JR, Barma DK, Mohapatra S, et  al. (2004) Asymmetric synthesis of the stereoisomers of 11,12,15(S)-trihydroxyeicosa5(Z),8(Z),13(E)-trienoic acid, a potent endothelium-derived vasodilator. Bioorg Med Chem Lett 14, 4987–4990. 85. Kerwin JL, Torvik JJ. (1996) Identification of monohydroxy fatty acids by electrospray mass spectrometry and tandem mass spectrometry. Anal Biochem 237, 56–64. 86. Nakamura T, Bratton DL, Murphy RC. (1997) Analysis of epoxyeicosatrienoic and monohydroxyeicosatetraenoic acids esterified to phospholipids in human red blood cells by electrospray tandem mass spectrometry. J Mass Spectrom 32, 888–896. 87. Nakamura T, Henson PM, Murphy RC. (1998) Occurrence of oxidized metabolites of arachidonic acid esterified to phospholipids in murine lung tissue. Anal Biochem 262, 23–32. 88. Wheelan P, Zirrolli JA, Murphy RC. (1993) Low-energy fast atom bombardment tandem mass spectrometry of monohydroxy substituted unsaturated fatty acids. Biol Mass Spectrom 22, 465–473.



HPLC/MS/MS-Based Approaches for Detection and Quantification

89. Harkewicz R, Fahy E, Andreyev A, Dennis EA. (2007) Arachidonate-derived dihomoprostaglandin production observed in endotoxin-stimulated macrophage-like cells. J Biol Chem 282, 2899–2910. 90. Hall LM, Murphy RC. (1998) Electrospray mass spectrometric analysis of 5-hydroperoxy and 5-hydroxyeicosatetraenoic acids generated by lipid peroxidation of red blood cell ghost phospholipids. J Am Soc Mass Spectrom 9, 527–532. 91. Currie LA. (1995) Nomenclature in evaluation of analytical methods including detection and quantification capabilities. Pure and Appl Chem 67, 1699–1723. 92. Revised Assessment of Detection and Quantitation Approaches; 2004. Report No.: EPA821-B-04–005. 93. Assigning Values to Nondetected/Nonquantified Pesticide Residues in Human Health Dietary Exposure Assessments; 2000. Report No.: OPP-00570A 94. Song WL, Lawson JA, Wang M, Zou H, FitzGerald GA. (2007) Noninvasive assessment of the role of cyclooxygenases in cardiovascular health: a detailed HPLC/MS/MS method. Methods Enzymol 433, 51–72.

187

95. Kishi N, Mano N, Asakawa N. (2001) Direct injection method for quantitation of endogenous leukotriene E4 in human urine by liquid chromatography/electrospray ionization tandem mass spectrometry with a column-switching technique. Anal Sci 17, 709–713. 96. Saunders RD, Horrocks LA. (1984) Simultaneous extraction and preparation for highperformance liquid chromatography of prostaglandins and phospholipids. Anal Biochem 143, 71–75. 97. Folch J, Lees M, Sloane Stanley GH. (1957) A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem 226, 497–509. 98. Smith KR, Pinkerton KE, Watanabe T, Pedersen TL, Ma SJ, Hammock BD. (2005) Attenuation of tobacco smoke-induced lung inflammation by treatment with a soluble epoxide hydrolase inhibitor. Proc Natl Acad Sci U S A 102, 2186–2191. 99. Yang J, Schmelzer K, Georgi K, Hammock B. (2009) Quantitative profiling method for oxylipin metabolome by liquid chromatography electrospray ionization tandem mass spectrometry. Anal Chem submitted.

sdfsdf

Chapter 9 Brain Phosphoinositide Extraction, Fractionation, and Analysis by MALDI-TOF MS

1

2 3

Roy A. Johanson and Gerard T. Berry

4

Summary 

5

Matrix-assisted laser desorption and ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) can provide rapid, sensitive determinations of lipids from small tissue samples in both single determinations and automated high-throughput assays. MALDI-TOF MS is a sensitive, high-throughput technique for the determination of lipids such as the phosphoinositides, PtdIns (phosphatidylinositol), PIP (phosphatidylinositol-4-phosphate, and PIP2 (phosphatidylinositol-4,5-bisphosphate), but in crude extracts the signals are weak or not observed due in large part to ion suppression by phosphatidylcholine and other cationic lipids. A rapid separation step using a small column of a strong cation exchange (SCX) gel can be utilized easily and effectively to adsorb or capture cationic lipids from chloroform/methanol lipid extracts and provide substantially improved signals for the phosphoinositides. In this review, we describe the use of fractionation of a crude lipid extracts using cation exchange columns in conjunction with MALDI-TOF MS and appropriate internal standards to quantify the levels of phosphoinositides in small mammalian brain samples.

6 7 8 9 10 11 12 13 14 15 16 17

Key words:  MALDI-TOF MS, Phosphoinositides

1. Introduction

18

19

The utilization of mass spectrometry (MS) in studies of lipids has been expanding in recent years as instrumentation has improved and new methods of lipid sample preparation and handling have been developed. MS has been used increasingly to characterize changes in lipid composition related to lipid metabolism and lipid-mediated cellular signaling and structural changes (1–3). MALDI-TOF MS can provide rapid, sensitive determinations of lipids from small tissue samples in both single determinations and

Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_9, © Humana Press, a part of Springer Science + Business Media, LLC 2009

189

20 21 22 23 24 25 26 27

190

Johanson and Berry

automated high-throughput assays, and is relatively insensitive to impurities such as buffer salts compared with other MS methods (3–7). Both qualitative and quantitative determinations of lipids can be performed using MALDI-TOF MS (3). When MALDI-TOF MS is used to study crude lipid extracts from biological samples, signals for many of the lipids are weak or not observed. This is due in large part to ion suppression by phosphatidylcholine and other cationic lipids such as lysophosphatidylcholine which has been shown to markedly decrease the sensitivity and reproducibility of signals obtained for the noncationic lipids in a lipid mixture (6, 8). Fractionation of the compounds in crude extracts prior to MALDI-TOF MS analysis (8), such as by thin-layer chromatography (9), is necessary in order to overcome the effect of ion suppression with detection of lipids such as the phosphoinositides in crude lipid extracts from biological samples. The phosphoinositides, phosphatidylinositol (PtdIns), and its phosphorylated derivatives such as phosphatidylinositol-4,5-bisphosphate (PIP2), are important biologically because the polyphosphoinositides play key roles in intracellular signaling and in the localization of proteins to cellular membranes. Advances in analytical methods for phosphoinositides will improve the ability to study the metabolic disruptions and test hypotheses in appropriate biological models. The lipid extraction procedure described here was developed and optimized with the primary goal of determining the levels of the phosphoinositides, PtdIns, PIP, and PIP2 in small samples (10–250 mg) of brain tissue. Consequently, strong acid conditions (1 M HCl) are used in the extraction in order to recover the PIP2 in the organic extract phase (10). We have reported that a rapid separation step using a small column of a strong cation exchange (SCX) gel (RediSep® SCX silica gel) can be utilized easily and effectively to adsorb or capture cationic lipids including phosphatidylcholine species from chloroform/ methanol lipid extracts from mammalian brain (4). In this review we describe the extraction of lipids from tissue followed by fractionation of the crude lipid extracts using cation exchange columns and then MALDI-TOF MS with internal standards for the determination of the levels of PIP2 and other phosphoinositides in brain samples.

28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

67

2. Materials

68

2.1. Equipment

69

1. MALDI-TOF MS instrument. 2. Glass Teflon (Potter-Elvehjem) homogenizers.



2.2. Reagents

2.3. Supplies

Brain Phosphoinositided Extraction, Fractionation, and Analysis

191

1. MALDI–MS grade 2,5-dihydroxybenzoic acid (DHB) from Fluka (Switzerland).

70

2. Soybean PtdIns from Avanti Polar Lipids (Alabaster, AL).

72

3. Bovine liver PtdIns from Avanti Polar Lipids (Alabaster, AL).

73

4. Synthetic 1-stearoyl-2-arachidonoyl-phosphatidylinositol-4,5bisphosphate (PIP2) from Avanti Polar Lipids (Alabaster, AL).

74

5. Synthetic dipalmitoyl-PIP2 was obtained from both Sigma (St.Louis, MO) and Cayman Chemical (Ann Arbor, MI).

76 77

6. RediSep® SCX silica gel was from Teledyne Isco (Lincoln, NE).

78

C57BL/6 mice were obtained from Charles River (Wilmington, MA).

79

3. Methods 3.1. Tissue Harvest

3.2. Lipid Extraction 3.2.1. Solvent Preparation

71

75

80

81

The murine brains are harvested as quickly as possible, immediately frozen in liquid N2, and stored and maintained at −80°C until the homogenization is started in order to minimize the appearance of degradation products in the mass spectra (11–15). Adult murine brains were removed and clamp frozen in liquid N2 within 40–45 s after the mouse was sacrificed. The frozen brains were stored at −80°C. Embryonic day 18.5 (E18.5) brains are obtained from fetuses in pregnancies timed by the appearance of vaginal plugs following mating.

82

The extraction procedure is based on methods described by Bligh and Dyer (16) and Schacht (10) and is carried out using an extraction solvent containing a ratio of CHCl3:CH3OH:H2O:conc.HCl such that the final ratio of CHCl3:CH3OH:H2O:conc.HCl, including the H2O in the sample, was 10:20:7:1. Adult brains are estimated to be 80% H2O by weight and fetal brains are estimated to be 85% H2O in calculating how much solvent to use. 1. Extraction solvent CHCl3:CH3OH:HCl:H2O, 10:20:1:3

91

2. Wash solvent CH3OH:H2O:HCl, 20:19:1 3. SCX column solvent CHCl3:CH3OH:H2O, 20:9:1 4. SCX column salt elution CHCl3:CH3OH:500 mM NaCl in H2O, 20:9:1

83 84 85 86 87 88 89 90

92 93 94 95 96 97 98 99 100 101 102 103 104 105 106

192 107 108

Johanson and Berry

3.2.2. Preparation of SCX Column

109 110 111

1. A column containing Teledyne ISCO RediSep® SCX silica gel is packed in a glass–wool plugged 5¾ in. long Pasteur pipette. The column bed is packed to 5 cm total height, including taper (at full bore, 1 ml = approximately 3.7 cm). This column volume is satisfactory for extracts from up to 200 mg wet wt brain tissue.

114

2. Prior to use, the column is washed/regenerated with three column volumes freshly mixed acetonitrile/1 M HCl (100 ml HCl/ml acetonitrile) as recommended by the manufacturer.

115 116

3. The column is equilibrated using five column volumes of the column solvent, CHCl3:CH3OH:H2O, 20:9:1 (17–19).

117

4 After use, the column is washed with CH3OH prior to storage.

112 113

118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152

3.2.3. Extraction of Lipids

1. Aliquots of the internal standards, soybean PtdIns and/or dipalmitoyl PIP2, that are dissolved in CHCl3 and/or CH3OH, are first pipetted into glass–teflon homogenizers (see Note 1) and the solvent evaporated. To prevent degradation of phospholipids (11–15), the homogenizers are then chilled on dry ice. Frozen brain samples are weighed into the chilled homogenizers (see Note 2). Extraction solvent, 6.8 ml CHCl3:CH3OH:H2O:conc HCl (10:20:3:1)/mg wet wt adult brain (7.22 ml/mg wet wt for fetal brain), is added to the samples. The still-frozen murine brain samples are rapidly homogenized as the homogenizer and tissue are warmed by mechanical friction and hand warmth to above freezing temperature. Defrosting and homogenizing the brain tissue works best when the homogenizer capacity is 4–8 times the total volume of the sample plus extraction solvent. Care should be taken to minimize evaporative losses by minimizing exposure to air and transferring to glass centrifuge tubes (see Note 3) and capping for comparison, immediately after homogenization. All subsequent steps are carried out at room temperature. CHCl3 and H2O are added to obtain the separation of phases. First 2.0 ml CHCl3/mg wet wt (2.12 ml/mg wet wt for fetal brain) is added followed by moderate vortex mixing. Then 2.0 ml H2O/ mg wet wt (2.12 ml/mg wet wt for fetal brain) is added followed by mixing. The phases are separated by centrifugation at 100 × g and the lower CHCl3 phase recovered. If the volume of fluffy emulsion that fills more than ~10% of the upper portion of the lower phase, it can generally be reduced by rocking gently to break up the emulsion and recentrifuging to separate the phases (see Note 4). 2. The upper phase and precipitate are reextracted with 2.0 ml CHCl3/mg wet wt (2.12 ml/mg for fetal brain), and the lower phase recovered and combined with the first lower phase (see Note 5). The combined lower phases are then washed 2 times by shaking with 1.5 volumes CH3OH:H2O:HCl, 20:19:1, separating the phases by centrifugation, and recovering the lower phase. If unfractionated extract is to be saved, it should be stored at −80°C.

3.2.4. Lipid Fractionation

3.3. Mass Spectrometry

3.4. Data Analysis

Brain Phosphoinositided Extraction, Fractionation, and Analysis

193

1. Sufficient methanol and H2O is added to the washed chloroform phase lipid extract from the lipid extraction procedure to bring the CHCl3:CH3OH:H2O, ratio to 20:9:1. The sample (0.5–1.5 ml) is passed through a 1.3 ml column of RediSep ® SCX that is packed in a glass–wool plugged Pasteur pipette. Prior to use, the column should be washed/regenerated with freshly mixed acetonitrile/1 M HCl as recommended by the manufacturer, and equilibrated in CHCl3:CH3OH:H2O, 20:9:1 (17–19). The lipid containing flow through fraction is collected.

153

2. The H2O/CH3OH phase is separated from the CHCl3 phase by adding 26% volume H2O:HCl, 10:1.

163

3. The extracted sample should be dried thoroughly under a stream of N2 and stored under N2 at −80°C.

165 166

For an overview of mass spectrometry, see the review by Baldwin (20), and the description on the American Society for Mass Spectrometry website, http://www.asms.org/whatisms/. MALDI-TOF mass spectrometry was carried out using Ettan MALDI-ToF/Pro (GE Healthcare, Piscataway, NJ). The appropriate aliquots of dissolved samples were prepared, dried using N2, and then dissolved at 2.5–15 mg lipid/ml in 0.5 M DHB in CHCl3:CH3OH, 1:1. Aliquots of 0.1 ml were spotted on the target slides. The MALDI-TOF MS measurements were carried out at 20 kV or −20 kV in the reflectron mode using selective accumulation.

167

The S/N was calculated using the equation S/N = (I−(Nmax + Nmin)/2)/(Nmax−Nmin) where I = the maximum intensity of a peak, and Nmax and Nmin are the maximum and minimum of pure noise. Excel (Microsoft, Redmond, WA) and SigmaPlot (Systat Software, Inc, Point Richmond, CA) were used for graphing and linear regression analysis. Since the amounts of sample and internal standard mixed with DHB and loaded on the target spots are operationally maintained within a relatively narrow optimum range, the levels of 16:0–18:2 PtdIns and 18:0–20:4 PtdIns can be readily set up so that, in effect, the range of the samples being assayed will be within the range of the standards. A range of external standard levels was selected for calibration curves that spanned the range of 18:0–20:4 PtdIns levels found in the optimum range of sample levels with MALDITOF MS analysis. Since there is a peak in the mass spectra of the brain extract at m/z 834.5, it is best to set the 16:0–18:2 PtdIns in the soy PtdIns at the high end of the scale to minimize interference by the m/z 834.5 peak. When the ratio of I885/I833 is plotted vs. the increasing level of an external standard, it has been demonstrated that a linear plot is obtained (4).

178

154 155 156 157 158 159 160 161 162 164

168 169 170 171 172 173 174 175 176 177

179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197

194

Johanson and Berry

The first-order regression for the results can be expressed as I885/I833 = m(18:0−20:4 PtdIns)+b where I885 is the intensity at m/z 885 from variable amounts of 18:0–20:4 PtdIns in the external standard or sample, and I833,s is the intensity at m/z 833 for the amount of internal standard, 16:0–18:2 PtdIns, used in running the standard curve (s). For a different amount of internal standard (a) the relationship between the intensity at m/z 833 from a mg 16:0–18:2 PtdIns compared to the intensity at m/z 833 from s mg 16:0–18:2 PtdIns (I833,a) can be expressed as I833,a/ a = I833,s. By rearranging this equation to I833,s=(s/a)I833,a and then substituting into the regression equation and rearranging, Eq. 1 is obtained.

198 199 200 201 202 203 204 205 206 207 208 209

18 : 0 − 20 : 4 PtdIns = a (I 885 / I 883,a ) / ms − b / m.

210 211 212

3.5. Results

(1)

1. Mass spectra of the standards that were used are shown in Fig. 1. Peaks at several m/z were observed in the mass spectra

Fig. 1. Negative mode mass spectra of purified bovine liver PtdIns and soybean PtdIns. Separate samples (40 mg each) in CHCl3 were dried and dissolved in 10 ml of 0.5 M DHB in CHCl3:CH3OH, 1:1. Aliquots of 0.1 ml containing approximately 400 ng PtdIns (440 pmol bovine PtdIns or 470 pmol soybean PtdIns) were spotted on each spot of the target slides. The mass spectra for the bovine PtdIns and the soybean PtdIns are shown in Panel a and b, respectively.



Brain Phosphoinositided Extraction, Fractionation, and Analysis

195

of both the purified bovine PtdIns (Panel A) and purified soybean PtdIns (Panel B). The mass spectra of the bovine PtdIns showed peaks at m/z 836.5, 857.5, 859.5, 861.5, 863.5, 883.5, 885.5, and 887.5 and the averages of the results for the relative intensities of the peaks in five mass spectra were 3, 2, 3, 10, 9, 9, 37, and 26%, respectively. All of these peaks could be accounted for by the analysis of the acyl species provided by the vendor, Avanti Polar Lipids. A similar analysis of soybean PtdIns showed peaks at m/z 831.5, 833.5, 855.5, 857.5, 859.5, 861.5, and 883.5. The 16:0–18:2 PtdIns peak intensity at m/z 833.5 was 46% of the summed intensities of the peaks for soybean PtdIns and the S/N ratio of the peak was 72. For our calibration curve analyses, the external and internal standard masses for the bovine 18:0–20:4 PtdIns and the soybean 16:0–18:2 PtdIns were calculated using the fractions 0.37 and 0.46, respectively, and the manufacturer’s specified concentration, 10 mg/ml. The results obtained based on the specified concentration were consistent, lot to lot, and consistent with published values (4).

213

2. Mass spectra run in the positive ion mode showed that the cation exchange minicolumns effectively removed the cationic lipids including the phosphatidylcholine species from the lipid extracts. Mass spectra of brain extracts run in the negative ion mode are shown for extracts from adult murine brain (Fig. 2) and fetal brain (Fig. 3). Peaks for PIP and PIP2 were barely discernable in the extract from the adult brain prior to the removal of the cationic lipids, and after the cationic lipids were removed, clear peaks were obtained. A similar improvement in the mass spectra was obtained for the extracts from the fetal brain.

233

3. When mass spectra of a dilution series of PtdIns was run using a fixed amount of an added standard of closely similar chemistry and mass, the mass spectra peaks increased linearly with increasing PtdIns (see Fig. 4). When the samples were spiked with a fixed amount of murine brain extract, the curve obtained was displaced upward by the PtdIns in the extract. The curve in the presence of the added brain extract was parallel to that obtained without added extract indicating that the presence of nonphosphoinositide compounds in the extract did not alter the comparative responses observed for the internal standard and the analytes in MALDI-TOF MS.

244

214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 234 235 236 237 238 239 240 241 242 243 245 246 247 248 249 250 251 252 253 254 255

196

Johanson and Berry

Fig. 2. Mass spectra obtained in negative ion mode of lipid extract fractions from adult murine brains. The CHCl3:CH3OH:HCl lipid extracts were prepared and then fractionated using RediSep® SCX silica gel as described in Subheading 3. The internal standards (indicated by IS), 2.00 mg soybean PtdIns and 0.50 mg dipalmitoyl PIP2 per g wet wt brain, were added before the homogenization step for the extracts shown in panel a and b. Dried samples of the extracts were dissolved in 0.5 M DHB in CHCl3:CH3OH, 1:1 and 0.1 ml aliquots were spotted on the target slides. Panel a shows the mass spectra obtained in the negative ion mode of the unfractionated extract, and panel b shows the spectra obtained of the lipids remaining in the extract following passage through the SCX column.

Fig. 3. (continued) of unfractionated extract, and panel b shows the spectra obtained for an extract that had been treated with an SCX column. For the mass spectra shown in Panels c and d, soybean PtdIns containing 141 mg 16:0–18:4 PtdIns and 150 mg synthetic dipalmitoyl PIP2 were added as internal standards to a pool of 4 E18.5 brains with a total wet weight of 290 mg prior to the homogenization and extraction procedures. Panel c shows the mass spectra of a crude extract with internal standards and Panel d shows the SCX treated extract with internal standards. The inset in Panel d shows a portion of the spectra for an extract that was prepared using precautions to minimize contamination by sodium ion by using plastic ware and thoroughly rinsed glassware including all disposable glassware in the extraction procedure. For this extract, two internal standards, soybean PtdIns containing 38 mg 16:0–18:2 PtdIns and 40 mg synthetic dipalmitoyl PIP2, had been added to a 83 mg wet wt E18.5 brain before the homogenization and extraction procedures.



Brain Phosphoinositided Extraction, Fractionation, and Analysis

197

Fig. 3. Negative ion mode mass spectra of lipid extract fractions from E18.5 murine brains. The CHCl3:CH3OH:HCl lipid extracts were prepared and then fractionated using RediSep® SCX silica gel as described in Subheading 3. The dried samples of the extracts were dissolved in 0.5 M DHB in CHCl3:CH3OH, 1:1 and 0.1 ml aliquots were spotted on the target slides. Panel a shows the mass spectra obtained in the negative ion mode

198

Johanson and Berry

Fig. 4. PtdIns calibration curves. Samples containing a constant amount (9.2 mg) of the internal standard, soybean 16:0–18:2 PtdIns, and varying amounts of an external standard, bovine 18:0–20:4 PtdIns, were prepared without and with added SCX treated murine brain extract. The mixed samples were dried, dissolved in 20 ml 0.5 M DHB in CHCl3:CH3OH, 1:1 and 0.1 ml aliquots were spotted on the target slides. The maximum intensities for the mass spectral peaks at m/z 833.5 for 9.2 mg 16:0–18:2 PtdIns and m/z 885.5 for the 18:0–20:4 PtdIns were determined. The plots show the ratios of the signal intensities of the peaks at m/z 885.5 for variable amounts of 18:0–20:4 PtdIns (I885) to the intensities at m/z 833.5 from 9.2 mg 16:0–18:2 PtdIns (I833,9.2 mg) in the soybean PtdIns. The data points shown are the mean ± SD from five determinations. In Panel a, mixtures of the indicated amounts of bovine liver 18.0–20.4 PtdIns with 9.2 mg 16:0–18:2 PtdIns are shown by the filled circles and the results when an aliquot of SCX treated extract from one adult murine brain proportional to that obtained from 7 mg wet wt tissue was added to identical mixtures of bovine PtdIns and soybean PtdIns is shown by the open circles. In Panel b, the filled squares are the results for mixtures of



Brain Phosphoinositided Extraction, Fractionation, and Analysis

199

4. Notes

256

1. Glass Teflon homogenizers are recommended for homogenization because they work well in homogenizing the small, 50–500 mg, frozen tissue samples as they are thawing and also because they produce a minimal amount of aeration with concomitant changes in the solvent composition from the proportionally more rapid evaporation of CHCl3.

257 258

2. Since evidence of enzymatic degradation products of the lipids such as the appearance of lysophosphatidylcholine and increased diacylglycerols were evident in the mass spectra when the tissue was allowed to thaw even briefly before homogenization, the homogenizer should be maintained at dry ice temperatures until the homogenization is started in order to keep the time span to an absolute minimum between when the tissue becomes defrosted and when it is homogenized and solubilized in the acidic organic solvent.

263

3. When mass spectra are to be measured in the positive ion mode, glassware should be used throughout the preparation of the lipid extracts because of known problems with strong organic solvents dissolving or leaching various compounds such as plasticizers from plastic labware that appear as artifact peaks in mass spectra run in the positive ion mode (3). When only negative ion mode mass spectra are to be determined, plastic ware such as polypropylene tubes and “Eppendorf” tips are suitable, as well as other plastic ware that is compatible with CHCl3 keeping in mind that unusual peaks from compounds which come from the plastic may appear randomly.

272

4. When the phases are separated following the initial extraction, there is a tendency for a fluffy emulsion to form in the upper portion of the lower phase especially if the mixing is too vigorous. This fluffy emulsion can fill most of the lower phase. If the layers are gently rocked to disperse the emulsion and recentrifuged, the emulsion will break down and the fluffy layer will become smaller. This may take more than one repeat of the rocking and centrifuging to get this layer to a minimum width band in the top of the lower phase thus maximizing the recovery of the lower phase.

283

5. The upper phase can be recovered for determinations of metabolites such as inositol which partition to this phase.

292

Fig.4. (continued) bovine and soybean PtdIns identical to the samples shown by the filled circles in Panel a, and the open squares show the results when SCX treated extract from 12 mg wet wt E18.5 brain was added to the mixtures. The lines and equations are shown for the first order regressions where I885 is the signal intensity at m/z 885.5 from varying amounts of 18:0–20:4 PtdIns, I833,9.2 mg is the signal intensity at m/z 833.5 from 9.2 mg 16:0–18:2 PtdIns, and PtdIns is mg of 18:0–20:4 PtdIns.

259 260 261 262 264 265 266 267 268 269 270 271 273 274 275 276 277 278 279 280 281 282 284 285 286 287 288 289 290 291 293

200

Johanson and Berry

294

References

295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346

1. Han, X. and Gross, R. W. (2005) Shotgun lipidomics: electrospray ionization mass spectrometric analysis and quantitation of cellular lipidomes directly from crude extracts of biological samples. Mass Spectrom. Rev. 24, 367–412. 2. Han, X. and Cheng, H. (2005) Characterization and direct quantitation of cerebroside molecular species from lipid extracts by shotgun lipidomics. J. Lipid Res. 46, 163–175. 3. Schiller, J., Suss, R., Arnhold, J., Fuchs, B., Lessig, J., Muller, M., Petkovic, M., Spalteholz, H., Zschornig, O., and Arnold, K. (2004) Matrix-assisted laser desorption and ionization time-of-flight (MALDI-TOF) mass spectrometry in lipid and phospholipid research. Prog. Lipid Res. 43, 449–488. 4. Johanson, R. A., Buccafusca, R., Quong, J. N., Shaw, M. A., and Berry, G. T. (2007) Phosphatidylcholine removal from brain lipid extracts expands lipid detection and enhances phosphoinositide quantification by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry. Anal. Biochem. 362, 155–167. 5. Milne, S., Ivanova, P., Forrester, J., and Alex, Brown, H. (2006) Lipidomics: an analysis of cellular lipids by ESI-MS. Methods 39, 92–103. 6. Petkovic, M., Schiller, J., Muller, M., Benard, S., Reichl, S., Arnold, K., and Arnhold, J. (2001) Detection of individual phospholipids in lipid mixtures by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry: phosphatidylcholine prevents the detection of further species. Anal. Biochem. 289, 202–216. 7. Schiller, J., Arnhold, J., Benard, S., Muller, M., Reichl, S., and Arnold, K. (1999) Lipid analysis by matrix-assisted laser desorption and ionization mass spectrometry: A methodological approach. Anal. Biochem. 267, 46–56. 8. Muller, M., Schiller, J., Petkovic, M., Oehrl, W., Heinze, R., Wetzker, R., Arnold, K., and Arnhold, J. (2001) Limits for the detection of (poly-)phosphoinositides by matrix-assisted laser desorption and ionization time-of-flight mass spectrometry (MALDI-TOF MS). Chem. Phys. Lipids 110, 151–164. 9. Schiller, J., Suss, R., Fuchs, B., Muller, M., Zschornig, O., and Arnold, K. (2003)

Combined application of TLC and matrixassisted laser desorption and ionization timeof-flight mass spectrometry (MALDI-TOF MS) to phospholipid analysis of brain. Chromatographia 57, S297–S302. 10. Schacht, J. (1981) Extraction and purification of polyphosphoinositides. Methods Enzymol. 72, 626–631. 11. Arthur, G. and Sheltawy, A. (1980) The presence of lysophosphatidylcholine in chromaffin granules. Biochem. J. 191, 523–532. 12. Christie, W. W. (1993) (Ed.), Advances in Lipid Methodology – Two, Oily Press, Dundee, 1993, pp. 195–213. (available online at http:// www.lipidlibrar y.co.uk/topics/extract2/ index.htm) 13. Hauser, G., Eichberg, J., and Gonzalez-Sastre, F. (1971) Regional distribution of polyphosphoinositides in rat brain. Biochim. Biophys. Acta 248, 87–95. 14. Hauser, G. and Eichberg, J. (1973) Improved conditions for the preservation and extraction of polyphosphoinositides. Biochim. Biophys. Acta 326, 201–209. 15. Nishihara, M. and Keenan, R. W. (1985) Inositol phospholipid levels of rat forebrain obtained by freeze-blowing method. Biochim. Biophys. Acta 835, 415–418. 16. Bligh, E. G. and Dyer, W. J. (1959) A rapid method for total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 911–917. 17. Hendrickson, H. S. and Ballou, C. E. (1964) Ion exchange chromatography of intact brain phosphoinositides on diethylaminoethyl cellulose by gradient salt elution in a mixed solvent system. J. Biol. Chem. 239, 1369–1373. 18. Kiselev, G. V. (1982) Preparative isolation of polyphosphoinositide fractions from ox brain. Biochim. Biophys. Acta 712, 719–721. 19. Low, M. G. (1990) Purification of Phosphatidylinositol 4-Phosphate and Phosphatidylinositol 4,5-Bisphosphate by Column Chromatography. In: Methods in Inositide Research, (Irvine, R. F. ed.) Raven Press, New York, pp.145–151. 20. Baldwin, M. A. (2005) Mass spectrometers for the analysis of biomolecules. Methods Enzymol. 402, 3–48.

Chapter 10 Lipidomic Analysis of Biological Samples by Liquid Chromatography Coupled to Mass Spectrometry

1

2 3

Giuseppe Astarita, Faizy Ahmed, and Daniele Piomelli

4

Summary 

5

Lipidomics studies the large-scale changes in nonwater-soluble metabolites (lipids) accompanying perturbations of biological systems. Because lipids are involved in crucial biological mechanisms, there is a growing scientific interest in using lipidomic approaches to understand the regulation of the lipid metabolism in all eukaryotic and prokaryotic organisms. Lipidomics is a powerful tool in system biology that can be used together with genomics, transcriptomics, and proteomics to answer biological questions arising from various scientific areas such as environmental sciences, pharmacology, nutrition, biophysics, cell biology, physiology, pathology, and disease diagnostics. One of the main challenges for lipidomic analysis is the range of concentrations and chemical complexity of different lipid species. In this chapter, we present a lipidomic approach that combines sample preparation, chromatographic, and intrasource ionization separation coupled to mass spectrometry for analyzing a broad-range of lipid molecules in biological samples.

6 7 8 9 10 11 12 13 14 15 16

Key words:  Lipidomics, Lipids, Liquid chromatography mass spectrometry, Fatty acids, Phospholipids, Cholesterol, Lipid profile, Lipid biomarkers, Large-scale analysis

17 18

1. Introduction

19

Lipids are natural molecules that are insoluble or partially soluble in water. These hydrophobic or amphipathic molecules can be either biosynthesized or absorbed from the environment and are vital for the life of all eukaryotic and prokaryotic organisms. Lipids play crucial biological roles through three general mechanisms (1) they affect the cellular membrane structures and protein–membrane interactions, (2) they provide a source of energy through processes of oxidation, and (3) they serve as signaling molecules, binding to

Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_10, © Humana Press, a part of Springer Science + Business Media, LLC 2009

201

20 21 22 23 24 25 26 27

202

Astarita, Ahmed, and Piomelli

plasma membrane or nuclear receptors mediating transmembrane signaling and cell-to-cell communication (1). The development of mass spectro-metry (MS) techniques marked the beginning of a new era for the study of lipids, opening a series of unprecedented experimental opportunities. Indeed, the implementation of atmospheric-pressure ionization techniques such as electrospray ionization (ESI) and atmospheric pressure chemical ionization (APCI), capable of coupling liquid chromatography (LC) with MS, made it possible to separate and analyze even the most hydrophobic lipids with much greater accuracy than ever before possible. Such technological advances have contributed to the advancement of lipidomics, the discipline that studies the largescale changes in lipid composition accompanying perturbations of biological systems (see Note 1). The ultimate goal of lipidomics is to understand the role of lipids in the biology of living organisms. It represents a rapidly evolving tool in system biology, which integrates multidisciplinary sets of data derived from molecular-profiling techniques such as genomics, transcriptomics, and proteomics. Therefore, there is a growing scientific interest in using lipidomics to answer various biological questions, arising from living organisms with all degree of biological complexity, such as animals, plants, fungi, protists, bacteria, archaea, and viruses. Lipidomic approaches can be used to investigate the following main research areas:

28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54

1.1. Ecophysiology

Lipidomics can be applied to study the impact of both abiotic environmental factors (e.g., climate, radiation, toxins, gravity, CO2 and oxygen levels, insolation, light-dark cycle, habitat) and biotic environmental factors (e.g., plants, animals, pathogens, and micro-organisms) on lipid metabolism (2–5).

1.2. Nutrition

Lipidomics is applied to study the effect of nutrients (e.g., carbohydrates, fats, proteins, vitamins, minerals, water, and beverages), nutraceuticals (e.g., antioxidants, fibers, omega-3 fats), food additives and fertilizers on lipid metabolism (6, 7).

1.3. Pharmacology and Toxicology

Lipidomics can be utilized to study the effects of pharmacological treatments (e.g., medications, vaccinations) and other synthetic products (e.g., cosmetics, contaminants, drugs of abuse, chemicals) on lipid metabolism (9, 10).

1.4. Genetics, Transcriptomics, and Proteomics

Lipidomic approaches can be used to study the effects of genetic diversity (e.g., genotypes, epigenetic regulation, mutations, and polymorphism), messenger RNA expression profiles and protein diversity (e.g., isoforms, post-translational modifications, cofactors) on lipid metabolism. Furthermore, lipidomics can be applied to investigate the biological functions of genes and proteins by

55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71



Lipidomic Analysis of Biological Samples by Liquid Chromatography

1.5. Biophysics

1.6. Cell Biology

1.7. Biochemistry and Molecular Biology

1.8. Physiology and Psychology

1.9. Pathology and Disease Diagnostics

203

studying the lipid profiles associated with genetic manipulation (i.e., gene overexpression or knock down) in biological systems. Indeed, fluctuations in lipid composition can be used to uncover alterations in the transcriptome and proteome (11, 12).

72

Lipidomic approaches can be used to investigate the effects of lipid composition on biophysical parameters (e.g., fluidity, compressibility), the biological functions of membrane structures (e.g., lipid rafts), as well as lipid–protein and lipid–nucleic acid interactions (13).

76

Lipidomics can study the role of lipid metabolism in critical cellular processes (e.g., cell cycles, survival/death, morphology, organelles) together with the circadian regulation of biological processes (e.g., development, aging, hormone production, hunger, thirsty, sleep) (14–17).

81

Lipidomic approaches can be used to understand the biochemical mechanisms for the biosynthesis and the metabolism of lipids in living organisms. In this area of research, lipidomics may lead to the discovery of novel lipid molecular species and lipid-related biochemical pathways (e.g., enzymes, proteins, receptors, and genes) (18, 19).

86

Lipidomics can be applied to understand the physiological roles played by endogenous lipids in crucial biological processes (e.g., learning and memory, immune response, pain, and inflammation) (20, 21). Also, lipidomic approaches can unveil the role of lipids in the sensory perception (chemoreception, photoreception, mechanoreception, and thermoreception), in mental processes and behavior (e.g., cognition, emotion, personality, social, and sexual behaviors), and physical activity (e.g., exercise) (22).

92 93

Lipidomic strategies can be used to investigate the role of lipid metabolism in the pathology of plant and animal diseases. Indeed, epidemiological studies revealed that many human diseases are characterized by specific alterations in lipid metabolism (e.g., cancer, obesity, diabetes, insomnia, depression, stress, trauma, dementia, as well as infectious, cardiovascular, and neurodegenerative diseases) (23, 24). Therefore, lipidomics can be used to profile lipid composition of biological samples for disease diagnosis and drug discovery. In fact, lipid composition can provide a “snapshot” of the biological state of an organism and, consequently, be considerate as an index (biomarker) of healthy or diseased state (25, 26). Furthermore, such lipid biomarkers can serve also as indicators of pharmacologic responses to a therapeutic intervention (24).

100

73 74 75

77 78 79 80

82 83 84 85

87 88 89 90 91

94 95 96 97 98 99

101 102 103 104 105 106 107 108 109 110 111 112 113

204

Astarita, Ahmed, and Piomelli

One of the main challenges of lipidomic analysis is the range of concentrations and chemical complexity of lipid compounds in biological samples (27) (Scheme 1). In fact, a comprehensive lipidomic analysis is expected to take into consideration “structural lipids” (e.g., phospholipids, which serve both as building blocks of the cell membranes and as precursors for signaling lipids), “storage lipids” (e.g., triacylglycerols, which are hydrolyzed to produce either energy or signaling lipids) and the less abundant, but equally important, “signaling lipids” (e.g., fatty acids and their derivatives) (Scheme 1). Therefore, there is a need to develop analytical approaches that allow for the comprehensive analysis of structural, storage, and signaling lipids. In this chapter, we present current methodologies utilized in our laboratory for lipidomic analysis of biological samples. We describe in some detail an analytical approach that combines sample preparation, chromatographic, and intrasource ionization separation coupled to mass spectrometry for analyzing the lipid composition of cells, biological fluids and tissues (see Note 2).

114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

2. Materials 133

2.1. Equipment

1. Analytical balance.

134

2. Chemical fume-hood.

135

3. Homogenizer.

136

4. Vortex.

137

5. Centrifuge.

138 139

6. Pierce Reacti-Therm III Heating/Stirring Module Thermo Fisher Scientific (Somerset, NJ, USA).

140

7. Spectrophotometer for protein measurement.

141

8. Agilent 1200-LC system (with autosampler) coupled to IonTrap XCT or single quadrupole 1946D MS detectors and interfaced with ESI or APCI (Agilent Technologies).

142 143 144

9. Gas: ultra-high purity compressed helium (for MS fragmentation) and high-purity N2 (for drying samples and for atmospheric pressure ionization functioning).

145 146 147 148 149 150 151 152 153

2.2. Reagents

A representative list of internal standards may include the following lipids. 1. Fatty acyls Fatty acid: heptadecanoic acid from Nu-Chek Prep (Elysian, MN, USA); d8-arachidonic acid from Cayman Chemicals (Ann Arbor, MI, USA);

Lipidomic Analysis of Biological Samples by Liquid Chromatography

Scheme 1. Lipid classes. Chemical classification of lipids. (a) Fatty acyls are fatty acids and their derivatives (oxygenated, amides, esters). (b) Glycerolipids are fatty acid esters of glycerol and comprise mono-, di-, and tri-acylglycerols. (c) Glycerophospholipids contain phosphoric acid in ester form with a glycerolipid. (d) Sphingolipids contain a common sphingoid base moiety. (e) Sterol lipids contain a fused four-ring core (27).

205

206

Astarita, Ahmed, and Piomelli

Prostaglandin: d4-prostaglandin E2 from Cayman Chemicals (Ann Arbor, MI, USA). Fatty-acid ethanolamide: heptadecenoylethanolamide (synthesized as previously reported, see (28)).

154 155 156 157

2. Glycerolipids Triacylglycerol: Trinonadecenoin from Nu-Chek Prep; Diacylglycerol: dinonadienoyl-sn-glycerol from Nu-Chek Prep; Monoacylglycerol: monoheptadecanoyl-sn-glycerol from Nu-Chek Prep; d8-2-arachidonoyl-sn-glycerol from Cayman Chemicals.

158 159 160 161 162 163

3. Glycerophospholipids Phosphatidylethanolamine: 1,2-diheptadecanoyl-sn-glycero3-phosphoethanolamine from Avanti polar Lipids; Phosphatidylglycerol: 1,2-diheptadecanoyl-sn-glycero-3phosphoglycerol from Avanti Polar Lipids; Phosphatidylcholine: 1,2-diheptadecanoyl-sn-glycero-3phosphocholine from Avanti Polar Lipids; Phosphatidylserine: 1,2-diheptadecanoyl-sn-glycero-3-phosphoserine from Avanti Polar Lipids; Phosphatidylinositol: 1,2-dipalmitoyl-sn-glycero-3-phosphoinositol from Avanti Polar Lipids.

164 165 166 167 168 169 170 171 172 173 174

177 178

4. Sphingolipids Ceramide: N-Lauroyl-ceramide from Avanti Polar Lipids; Sphyngomyelin: N-Lauroyl-sphingomyelin from Avanti Polar Lipids.

179

5. Sterol lipids

180

Cholesterol: d7-cholesterol from Avanti Polar Lipids.

181

6. Solvents and chemicals Water, methanol, chloroform (HPLC grade) are purchased from Thermo Fisher Scientific (Somerset, NJ, USA). Acetic acid and ammonium acetate are from Sigma (Saint Louis, Missouri, USA).

175 176

182 183 184 185 186

2.3. Supplies

1. LC columns.

188

2. Glass Vials (8 ml, 40 ml, 1.5 ml for autosampler and LC analysis).

189

3. Glass pipettes (5 ml, 10 ml).

190

4. Glass Pasteur pipettes.

191

5. Caps with Teflon-liner.

192

6. Conical insert for reducing the volume of the autosampler vials.

193

7. Vial racks.

194

8. Vial trays.

195

9. Dry ice.

187



Lipidomic Analysis of Biological Samples by Liquid Chromatography

207

3. Methods 3.1. Sample Preparation for Lipidomics Analysis

3.1.1. Cells

196

Sample preparation includes the extraction of lipids from the biological matrix and the removal of any nonlipid contaminants from the extract (see Note 3). We use a modified Folch procedure for lipid preparation of biological samples from cells, biological fluids, and tissues (see Notes 4 and 5) (Scheme 2). Lipid molecular species are quantified by normalizing the individual molecular ion peak intensity with an internal standard for each lipid class. Therefore, a mixture of nonendogenous lipid species used as internal standards for each lipid class before the extraction process (see Subheading 2.2). These internal standards allow the lipid levels to be normalized for both extraction efficiency and instrument response.

197

1. Label 8-ml glass vials based on the number of tissue samples to analyze.

209 210

2. Wash cells with phosphate-buffered saline (PBS, 1×), remove all PBS.

211

3. Add 1 ml of methanol containing the internal standards to each well keeping the plate on ice.

213 214

4. Scrape and collect the cells in 8-ml glass vials (see Note 6).

215

5. Sonicate in ice for 10 s (5 pulses (×2) at 200 V).

216

6. Save 20 ml aliquot for protein measurements. Protein concentration is measured using the Bradford protein concentration assay (Bio-Rad Laboratories Inc., Hercules, CA) or the BCA protein assay (Pierce, Rockford, IL).

217

7. Add 2-ml chloroform and vortex for 10 s.

221

8. Wash with 0.75 ml of water (or better 0.7% KCl solution) and vortex for 10 s.

222 223

9. Centrifuge 1,000 × g for 15 min at 4°C to separate the mixture into two phases with a protein disk at their interface. The lower phase is mainly chloroform and contains most of the lipids; the upper phase is methanol and water containing more polar metabolites.

224

10. Take the organic (bottom) layer using a glass Pasteur pipette and transfer into another 8-ml glass vial. Discard the protein disk and the upper (aqueous) phase.

229

11. Dry down using a gentle N2 steam.

232

12. Resuspend in 50–100 ml chloroform/methanol (1:3, vol:vol).

233

13. Transfer into 1.5-ml glass vials with the 0.2-ml conical inserts and proceed to LC/MS analysis.

234

14. Normalize lipids for mg protein (mol/mg protein).

236

198 199 200 201 202 203 204 205 206 207 208

212

218 219 220

225 226 227 228 230 231

235

208

Astarita, Ahmed, and Piomelli

Scheme 2. Lipidomic approach. Flow chart of the strategy used for a broad-range analysis of lipids from biological sample. MAG monoacylglycerol, FAE fatty acid ethanolamide, LPC lysophosphatidylcholine, FA fatty acid, oxFA oxygenated fatty acids, DAG diacylglycerol, TAG triacylglycerol, SP sphingolipid, PC phosphatidylcholine, PE phosphatidylethanolamine PS phosphatidylserine, PI phosphatidylinoitol; PG phosphatidylglycerol; PA phosphatidic acid.

239

Tissues are rapidly collected and snap-frozen in liquid N2. 1. Label 8-ml glass vials according to the number of tissue samples to analyze.

240 241

2. Add 1 ml of methanol containing the internal standards in each vial, while keeping the vials in ice.

242

3. Weigh the frozen tissues (10–100 mg) (see Note 6) and transfer them into the previously prepared vials containing methanol with internal standards.

237 238

243 244

3.1.2. Tissues

246

4. Homogenize the tissues keeping the vials in an ice bath and collect 20 ml aliquots for protein measurements (see Note 7).

247

5. Add 2 ml of chloroform and vortex for 10 s.

248 249

6. Wash with 0.75 ml of water (or better 0.7% KCl solution) and vortex for 10 s (see Note 8).

250

7. Centrifuge at 1,000 × g for 10 min at 4°C to the mixture into two phases with a protein disk at their interface. The lower phase is mainly chloroform and contains most of the lipids; the upper phase is methanol and water containing more polar metabolites (see Note 9).

245

251 252 253 254



Lipidomic Analysis of Biological Samples by Liquid Chromatography

3.1.3. Biological Fluids Plasma and Serum

Cerebrospinal Fluid (CSF)

209

8. Prepare another set of 8-ml glass vials using the same labeling system as described before.

255

9. After centrifugation, collect the lower (organic) phase using a glass Pasteur pipette.

257

10. Re-extract the protein disk and the upper (aqueous/methanol) phase with 2-ml of chloroform.

259 260

11. Centrifuge and add together the two organic phases.

261

12. Evaporate the solvent to dryness in the vials using a gentle N2 stream.

262

13. Resuspend in 50–100 ml chloroform/methanol (1:3, vol:vol).

264 265

14. Transfer the resuspended lipid solution to the 1.5-ml vials with the 0.2-ml conical inserts and proceed to LC/MS analysis.

266

15. Normalize lipid amount per grams of tissue (mol/g) or per mg protein (mol/mg protein).

268 269

For plasma preparation, blood is centrifuged in EDTA-containing tubes at 1,000 × g for 10 min at 4°C, and the top layer (plasma) is recovered using a glass Pasteur pipette. For serum preparation, blood is immediately centrifuged in glass tubes at 1,000 × g for 10 min at room temperature and the top layer (serum) is recovered using a glass Pasteur pipette.

270

CSF samples are checked for blood contamination by measuring the total cell count, total protein, CSF/serum albumin and IgG quotients, and determination of oligoclonal bands by isoelectric focusing and silver staining. Normal cell counts, normal CSF/ serum albumin ratios, and no oligoclonal bands indicate healthy blood–brain barrier function and lack of intrathecal immunoglobulin G synthesis. 1. Label 8-ml glass vials according to the number of tissue samples to analyze.

276

2. Transfer 0.2 ml of plasma/serum/CSF samples into the 8-ml vial in ice.

285

3. Add three volumes of ice-cold acetone containing internal standards.

287 288

4. Vortex for 10 s.

289

5. Shake and refrigerate sample for 30 min.

290

6. Centrifuge at 1,000 × g for 10 min at 4°C to pellet out the precipitated proteins.

291

7. Take the supernatant and evaporate the excess acetone under N2 stream.

293 294

8. Add 1-ml of methanol and vortex for 10 s.

295

9. Add 2-ml of chloroform and vortex for 10 s.

296

256 258

263

267

271 272 273 274 275 277 278 279 280 281 282 283 284 286

292

210

Astarita, Ahmed, and Piomelli

298

10. Wash with 0.8 ml of water (or better 0.7% KCl) and vortex for 10 s.

299

11. 11 Centrifuge at 1,000 × g for 10 min at 4°C.

300 301

12. Collect the lower phase with a glass Pasteur pipette and transfer to a clean 8-ml glass vial.

302

13. Evaporate the elnates to dryness under N2 stream.

303 304

14. Resuspend in 50–80 ml of a solution chloroform/methanol (1:3, vol:vol).

305

15. Prepare 1.5-ml vials with conical glass inserts.

306 307

16. Transfer the resuspended lipid solution to the 1.5-ml vials with 0.2-ml conical insert and proceed to LC/MS analysis.

308

17. Normalize lipid amount per ml of biological fluid (mol/ml).

297

318 319

Blood can be fractionated in plasma and blood cells, which is made of white blood cells (WBCs) and red blood cells (RBCs). Blood cells are normally discarded when collecting the plasma. However, the same procedure used for plasma collection, also allows the recovery of the buffy coat (mainly WBCs) and the RBCs, which can be used to measure biomarkers for dietary fat (29) and diseases (30). 1. Fractionate whole blood samples by centrifuging in EDTA at 1,000 × g for 10 min at 4°C. This will separate the blood into an upper plasma layer, a lower RBCs layer, and a thin interface (buffy coat) containing the WBCs.

320

2. Recover the plasma.

321

3. Recover the WBCs, wash with PBS (1×) three times and centrifuge discarding the supernatant; freeze in distilled water (1:1, vol:vol).

309

Blood

310 311 312 313 314 315 316 317

322 323

4. Recover the RBCs, wash with PBS (1×) three times and centrifuge discarding the supernatant; freeze in distilled water (1:1, vol:vol).

324 325 326 327 328 329

3.2. LC/MS Analysis of Lipids

Mobile phase A is methanol containing 0.25% acetic acid and 5 mM ammonium acetate; mobile phase B is water containing 0.25% acetic acid and 5 mM ammonium acetate. Lipids are identified based on their retention times and MSn properties.

3.2.1. Small Lipids Analysis

Small lipid molecules are separated using a reversed-phase Zorbax XDB Eclipse C-18 column (50 × 4.6 mm i.d., 1.8 mm particle size, 80 Å of porous diameter, Agilent Technologies). Detection and analysis is controlled by Agilent Chemstation and Bruker Daltonics software. 1. Fatty acyls Lipids are eluted using a linear gradient from 90% A to 100% B in 2.5 min at a flow rate of 1.5 ml/min with column

330 331 332 333 334 335 336 337 338



Lipidomic Analysis of Biological Samples by Liquid Chromatography

211

Fig. 1. Analysis of small lipids by LC/MS. Representative LC/MS chromatograms of fatty acids extracted from biological samples.

3.2.2. Large Lipids Analysis

t­ emperature at 40°C. ESI is in the negative mode, capillary voltage is set at −4 kV and fragmentor voltage is 100 V. N2 is used as drying gas at a flow rate of 13 l/min and a temperature of 350°C. Nebulizer pressure is set at 60 psi. We use commercially available fatty acyls as reference standards. They are analyzed monitoring the mass-to-charge ratio (m/z) of the deprotonated molecular ions [M − H]− in the selected-ion monitoring mode (Fig. 1).

339

Large lipid molecules are separated using a reversed-phase Poroshell 300SB C-18 column (2.1 × 75 mm i.d., coating layer of 0.25 mm on total particle diameter of 5 mm, 300 Å of porous diameter, Agilent Technologies). Lipids are identified based on their retention times and MSn properties. Detection and analysis is controlled by Agilent Chemstation and Bruker Daltonics software. 1. Glycerolipids, glycerophospholipids, sphingolipids A linear gradient is applied from 85% A to 100% B in 5 min at a flow rate of 1.0 ml/min with column temperature set at 50°C.

346 347

340 341 342 343 344 345

348 349 350 351 352 353 354 355

212

Astarita, Ahmed, and Piomelli

The capillary voltage is set at 4.0 kV and skimmer voltage at 40 V. N2 is used as drying gas at a flow rate of 10 l/min, temperature at 350°C and nebulizer pressure at 60 psi. Helium is used as collision gas, and fragmentation amplitude is set at 1.2 V. MS detection is both in the positive and in the negative ionization modes. Ion charge control is on, smart target set at 50,000 and max accumulation time at 50 ms, scan range of 100–1,500 amu, 26,000 m/z per second.

356 357 358 359 360 361 362 363 364 365 366

3.2.3. Sterol Lipids Analysis

Lipids are separated using a linear gradient from 75% A to 100% B in 4-min period at a flow rate of 1.0 ml/min with column temperature at 50°C. APCI is set in positive mode. Drying gas is set at 350°C and a flow of 8 l/min. Nebulizer gas pressure is set at 30 psi and vaporizer temperature at 475°C. Capillary voltage is set at 300 V with the corona current set at 5 mA.

3.3. Results: Chromatographic and Intrasource Ionization Separation of Lipid Molecules

Lipids exist in nature in a wide variety of chemical complexities and dynamic range of concentrations (Scheme 1). In order to simplify the analysis in biological tissues, lipids are schematically divided into three main classes (a) small lipids, defined here as molecules containing one aliphatic group such as fatty acids and their derivatives (amides, esters, oxygenated compounds); (b) large lipids, molecules containing two or more aliphatic groups, such as phospholipids, diacylglycerols, triacylglycerols, sphingolipids; and (c) sterol lipids, molecules containing a rigid four-ring backbone such as cholesterol and its derivatives. Therefore, in order to analyze the different classes of lipids by LC/ MS, two separate chromatographic approaches are applied, using different reversed-phase C-18 stationary phases (Scheme 2). Furthermore, because lipid classes with different functional groups have characteristic ionization efficiencies, a combination of ESI set in either positive or negative mode, and APCI set in positive mode is used (Scheme 2).

3.3.1. Small Lipids Analysis

To separate lipids containing one fatty acyl group, a reversed-phase C-18 column packed with conventional porous silica particles of small spherical diameter (sub-2 mm) is used. Fatty acyl species are separated both by chain length and by degree of unsaturation of their fatty acid chains. For example, fatty acids containing shorter or more unsaturated acyl chains elute earlier than those with longer and more saturated chains (Fig. 1). Generally, in positive ESI mode small lipids are detected as protonated molecular ions or sodium and ammonium adducts. In contrast, in negative mode small lipids are detected as deprotonated molecular ions (Fig. 1).

3.3.2. Large Lipids and Sterol Lipids Analysis

To separate large and sterol lipids, a reversed-phase C-18 column packed with superficially porous particles (Poroshell,

367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399



Lipidomic Analysis of Biological Samples by Liquid Chromatography

213

Agilent-Technologies, coating layer of 0.25 mm on total particle diameter of 5 mm) is used (18, 31). This allows for fast flow rates and good peak shapes (Fig. 2). Usually, because of diffusion limits in totally porous silica, large lipid molecules give tailing peaks at high flow rates. However, superficially coated columns allow for faster diffusion at the surface, allowing high flow rates and good peak shape. Indeed, the thin shell allows the slowly diffusing hydrophobic macromolecules and the rigid structures of sterol lipids to rapidly penetrate the superficial packing material (since the solid core prevents further diffusion). To decrease the retention times, a high flow velocity is applied. To decrease mobile phase viscosity and avoid exceeding the column back-pressure limits, a relatively high column temperature is used. A combination of high temperature and high flow velocities improves the separation speed, resulting in better peak shape of the lipid analytes. Notably, lipids are stable at high temperatures using high flow rates (32). Although lipids are separated when differing in a single fatty acyl chain, their combinatorial nature makes only a partial separation of the isomeric species possible (Fig. 2). Therefore, to obtain more information on the lipid structure, LC separation is coupled with MSn fragmentation data (Fig. 3). Generally, large lipids are detected in the positive ESI mode as sodium or ammonium adducts or as deprotonated molecular ions in the negative mode. For sterol lipids, which are highly hydrophobic and hard to ionize, APCI is used in the positive mode and the protonated molecular ions are detected after loss of water.

Fig. 2. Analysis of large lipids by LC/MS. Representative LC/MS chromatograms of phosphatidylethanolamines (PE) extracted from biological samples.

400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426

Fig. 3. Analysis of large lipids by MS/MS. Representative MS/MS spectrum of a selected phosphatidylethanolamine (1-stearoyl,2-docosahexaenoyl-sn-glycero-3-phosphoethanolamine) derived from biological samples.

214 Astarita, Ahmed, and Piomelli



Lipidomic Analysis of Biological Samples by Liquid Chromatography

215

3.4. Maintenance and System Suitability Test for LC/MS Analysis

In the following sections we report a general procedure for the maintenance and system suitability testing used to validate the LC/MS lipidomic analysis.

427

3.4.1. LC/MS Maintenance

To avoid contaminations, routinely preventive maintenance is performed. 1. Replacing the spray needle and electron multiplier; cleaning ionization spray chamber or other accessible MS components.

430 431

2. Replacing inline filters and frits, the injector needle and capillaries or other accessible LC components; flushing the system with a mixture of cyclohexane/acetonitrile/isopropanol (1/1/2, v/v/v).

434

1. To check for the LC column status, assure that the column has a constant backpressure, which usually is a guarantee of good column performance. Increased pressure indicates column contamination or fouling.

438 439

2. To check for contaminations, blank samples are run before and between biological samples.

442

3. To check for accuracy of quantification, quality control samples are run at the end of the run (three concentrations that are representative of the concentration range of the analyte of interest).

444 445

4. To check for linearity of the detection response, calibration curves are run before running the samples.

448

5. To avoid sample cross-contaminations, the injector needle is washed automatically between each sample injection.

450 451

Lipid extracts are generally stored in a freezer at –80°C. They are solubilized in chloroform–methanol solutions using glass vials closed with Teflon-lined caps and secured with Parafilm. To prevent oxidation, air is removed by flushing the vials or tubes with N2 before closing them. It was shown that after storage up to 4 years at –80°C, the blood lipid composition is practically unchanged (33). If storage is brief, lipids can be stored at –20°C.

452

Contaminants can be detected as extra-peaks or high background noise in LC/MS chromatograms. They strongly affect the specificity and sensitivity of our analysis. During sample preparation, common sources of contamination are mineral oils, grease, detergents, and plasticizers from plastics, including lipid molecules such as oleamide (34). Plastic pipettes, tips, beakers, and vials can leach contaminants into organic solutions. Therefore, all operations are generally carried out in glass and all vials or tubes are closed with screw caps including a Teflon-covered liner. Furthermore, all

460 461

3.4.2. Quality Assurance

3.4.3. Storage of Lipid Extracts

3.4.4. Contaminations

428 429

432 433 435 436 437

440 441 443

446 447 449

453 454 455 456 457 458 459

462 463 464 465 466 467 468

216

Astarita, Ahmed, and Piomelli

operators must wear gloves during the procedures to prevent any contaminations by skin surface lipids. Change gloves frequently and keep vials closed or covered with aluminum foil.

469 470 471

472

473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495

4. Notes 1. Because lipids are a set of small-molecule metabolites, lipidomics is considered to be part of metabolomics, which is the large-scale study of all metabolites (both water-soluble and water-unsoluble) in biological organisms. The distinction originated as consequence of the metabolome (complete set of small-molecule metabolite) complexity, which required the development of analytical approaches specific for nonwater soluble metabolites (lipids) (35). 2. The described fast lipidomic approach is suitable for the determination of a broad-range of lipid alterations occurring in biological samples. The combination of the chromatographic resolving power in conjunction with the ionization source selection and the mass detection can be used to analyze even the lipids present at very low concentration. In contrast, the direct infusion of the lipid extract into the MS detector is subject to ionization suppression effects and loss of sensitivity and accuracy. Furthermore, because lipids may differ in mass by only two units, a partial chromatographic separation helps avoid the isotopic effects, which affect the actual mass abundance (36). 3. Particular attention should be given to sample preparation: It is worth remembering that there is no good LC/MS analysis without a good sample preparation.

507

4. Alternative extraction procedures that use less toxic organic solvents such as methyl-tert-butyl ether (37), hexane–isopropanol, and ethyl acetate/ethanol mixtures have been proposed for a wide range of tissues (38, 39). Surprisingly, it is not always made clear in the laboratory environment that methanol and chloroform are toxic and potentially carcinogenic (38, 40, 41). Furthermore, the methanol/ chloroform mixture is extremely irritating to skin and eyes. Therefore, it is particularly important to train students and new laboratory personnel to handle organic solvents with gloves in a chemical fume-hood, avoiding health-hazard by accidental spills, skin contact and breathing of vapors.

508 509

5. For the recovery of acidic phospholipids such as gangliosides and phosphoinositides, alternative extraction methods have

496 497 498 499 500 501 502 503 504 505 506



Lipidomic Analysis of Biological Samples by Liquid Chromatography

217

been suggested which use strong HCl solutions instead of water during the washing step of the Folch procedure42.

510

6. Lipid composition is altered during thawing at room temperature. Therefore, to avoid tissue degeneration (1) cells are kept on ice or (2) tissue samples are cut and weighted while still frozen.

512

7. Sometimes it is useful to normalize the lipid levels in tissue samples by protein amount. Indeed, very small amount of tissue are often difficult to weigh without thawing them and, consequently, altering the lipid composition. Therefore, the samples are directly added to methanol (without the weighting step) and prior to extraction, 20 ml aliquots from the homogenate solutions are taken for protein measurements, which can be conducted using the Bradford protein concentration assay (Bio-Rad Laboratories Inc., Hercules, CA, USA) or the BCA protein assay (Pierce, Rockford, IL, USA).

516 517

8. The lipid extraction requires a ratio of chloroform, methanol, and water of 8:4:3. In these conditions, after centrifugation and phase separation, the approximate proportion of chloroform, methanol, and water in the upper phase is 3:48:47 by volume. In the lower phase, the respective proportion is 86:14:1.

526

9. To avoid contaminations from the upper aqueous phase into the pipette tip during the recovery of the bottom phase, insert the glass Pasteur pipette through the upper phase with gentle positive-pressure (i.e., gentle bubbling). Also, carefully withdraw the bottom phase through the pipette from the bottom of the vial. Furthermore, to avoid the interface or upper phase, it is better not to recover the entire bottom phase, but leaving the last drops (5–10% of the total organic phase) in the vials.

532 533

Acknowledgments

511 513 514 515

518 519 520 521 522 523 524 525 527 528 529 530 531

534 535 536 537 538 539 540

541

The contribution of the Agilent Technologies/University of California Irvine Analytical Discovery Facility, Center for Drug Discovery and the Agilent Technologies Foundation are gratefully acknowledged. This work was supported by grants from the National Institute of Health (R21DA-022702, R01DK-073955, R01 DA-012413, R01DA-012447, RR274–297/3504008, RR274–305/3505998, 1RL1AA017538 to D.P.).

542 543 544 545 546 547 548

218

Astarita, Ahmed, and Piomelli

549

References

550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606

1. Piomelli D, Astarita G, Rapaka R. (2007) A neuroscientist’s guide to lipidomics. Nat. Rev. Neurosci. 8, 743–754. 2. Welti R, Shah J, Li W, Li M, Chen J, Burke JJ, Fauconnier ML, Chapman K, Chye ML, Wang X. (2007) Plant lipidomics: discerning biological function by profiling plant complex lipids using mass spectrometry. Front. Biosci. 12, 2494–2506. 3. Nicholson JK, Holmes E, Wilson ID. (2005) Gut microorganisms, mammalian metabolism and personalized health care. Nat. Rev. Micro. 3, 431–438. 4. Kell DB, Brown M, Davey HM, Dunn WB, Spasic I, Oliver SG. (2005) Metabolic footprinting and systems biology: the medium is the message. Nat. Rev. Micro. 3, 557–565. 5. Lin CY, Viant MR, Tjeerdema RS. (2006) Metabolomics: Methodologies and applications in the environmental sciences. J. Pest. S. 31, 245–251. 6. Whitfield PD, German AJ, Noble PM. (2004) Metabolomics: an emerging post-genomic tool for nutrition. Br J Nutr. 92, 549–555. 7. German JB, Roberts M, Watkins SM. (2003) Personal metabolomics as a next generation nutritional assessment. J. Nutr. 133, 4260–4266. 8. Han X. (2007) An update on lipidomics: progress and application in biomarker and drug development. Curr. Opin. Mol. Ther. 9, 586–591. 9. Rapaka RS. (2005) Targeted lipidomics and drug abuse research. Prostaglandins Other Lipid Mediat. 77, 219–222. 10. Jung K, Astarita G, Zhu C, Wallace M, Mackie K, Piomelli D. (2007) A key role for diacylglycerol lipase-alpha in metabotropic glutamate receptor-dependent endocannabinoid mobilization. Mol. Pharmacol. 72, 612–621. 11. Chen C, Shah YM, Morimura K, Krausz KW, Miyazaki M, Richardson TA, Morgan ET, Ntambi JM, Idle JR, Gonzalez FJ. (2008) Metabolomics reveals that hepatic stearoylCoA desaturase 1 downregulation exacerbates inflammation and acute colitis. Cell Metab. 7,135–147. 12. Epand RM. (2008) Proteins and cholesterolrich domains. Biochim. Biophys. Acta. 1778, 1576–1582. 13. Gaspar ML, Aregullin MA, Jesch SA, Nunez LR, Villa-García M, Henry SA. (2007) The emergence of yeast lipidomics. Biochim. Biophys. Acta. 1771, 241–254.

14. Rabinowitz JD. Cellular metabolomics of Escherchia coli. (2007) Expert Rev. Proteomics. 4, 187–98. 15. Meer GV. (2005) Cellular lipidomics. EMBO J. 24, 3159–3165. 16. Hunt AN. (2006) Dynamic lipidomics of the nucleus. J. Cell. Biochem. 97, 244–251. 17. Astarita G, Ahmed F, Piomelli D. (2008) Identification of biosynthetic precursors for the endocannabinoid anandamide in the rat brain. J. Lipid Res. 49, 48–57. 18. Astarita G, Rourke BC, Andersen JB, Fu J, Kim JH, Bennett AF, Hicks JW, Piomelli D. (2006) Postprandial increase of oleoylethanolamide mobilization in small intestine of the Burmese python (Python molurus). Am. J. Physiol. Regul. Integr. Comp. Physiol. 290, R1407–1412. 19. Walker JM, Krey JF, Chen JS, Vefring E, Jahnsen JA, Bradshaw H, Huang SM. (2005) Targeted lipidomics: fatty acid amides and pain modulation. Prostaglandins Other Lipid Mediat. 77, 35–45. 20. Serhan CN. (2005) Mediator lipidomics. Prostaglandins Other Lipid Mediat. 77, 4–14. 21. Kaddurah-Daouk R, McEvoy J, Baillie RA, Lee D, Yao JK, Doraiswamy PM, Krishnan KR. (2007) Metabolomic mapping of atypical antipsychotic effects in schizophrenia. Mol. Psychiatry. 12, 934–945. 22. Adibhatla R, Hatcher J, Dempsey R. (2006) Lipids and lipidomics in brain injury and diseases. AAPS J. 8, E314-E321. 23. Wenk MR. (2005) The emerging field of lipidomics. Nat. Rev. Drug Discov. 4, 594–610. 24. Ackermann B, Hale J, Duffin K. (2006) The role of mass spectrometry in biomarker discovery and measurement. Curr. Drug Metab. 7, 525–539. 25. Fonteh A, Harrington R, Huhmer A, Biringer RG, Riggins JN, Harrington MG. (2006) Identification of disease markers in human cerebrospinal fluid using lipidomic and proteomic methods. Dis Markers. 22, 39–64. 26. Fahy E, Subramaniam S, Brown HA, Glass CK, Merrill AH Jr, Murphy RC, Raetz CR, Russell DW, Seyama Y, Shaw W, Shimizu T, Spener F, van Meer G, VanNieuwenhze MS, White SH, Witztum JL, Dennis EA. (2005) A comprehensive classification system for lipids. J Lipid Res. 46, 839–861. 27. Fu J, Astarita G, Gaetani S, Kim J, Cravatt BF, Mackie K, Piomelli D. (2007) Food intake regulates oleoylethanolamide formation and



Lipidomic Analysis of Biological Samples by Liquid Chromatography

degradation in the proximal small intestine. J. Biol. Chem. 282, 1518–1528. 28. Fuhrman BJ, Barba M, Krogh V, Micheli A, Pala V, Lauria R, Chajes V, Riboli E, Sieri S, Berrino F, Muti P. (2006) Erythrocyte Membrane Phospholipid Composition as a Biomarker of Dietary Fat. Ann. Nutr. Met. 50, 95–102. 29. Keshavan MS, Mallinger AG, Pettegrew JW, Dippold C. (1993) Erythrocyte membrane phospholipids in psychotic patients. Psychiatry Res. 49, 89–95. 30. Kirkland JJ, Truszkowski FA, Dilks CH, Engel GS. (2000) Superficially porous silica microspheres for fast high-performance liquid chromatography of macromolecules. J. Chromatogr. A. 890, 3–13. 31. Barroso B, Bischoff R. (2005) LC-MS analysis of phospholipids and lysophospholipids in human bronchoalveolar lavage fluid. J. Chromatogr. B. 814, 21–28. 32. Hodson L, Skeaff CM, Wallace AJ, Arribas GLB. (2002) Stability of plasma and erythrocyte fatty acid composition during cold storage. Clin. Chim. Acta. 321, 63–67. 33. Lau O, Wong S. (2000) Contamination in food from packaging material. J. Chromatogr. A. 882, 255–270. 34. German JB, Gillies LA, Smilowitz JT, Zivkovic AM, Watkins SM. (2007) Lipidomics and lipid profiling in metabolomics. Curr. Opin. Lipidol. 18, 66–71.

219

35. Han X, Gross RW. (2005) Shotgun lipidomics: multidimensional MS analysis of cellular lipidomes. Expert Rev. Proteomics. 2, 253–264. 36. Matyash V, Liebisch G, Kurzchalia TV, Shevchenko A, Schwudke D. (2008) Lipid extraction by methyl-tert-butyl ether for high-throughput lipidomics. J. Lipid Res. 49, 1137–1146. 37. Hara A, Radin NS. (1978) Lipid extraction of tissues with a low-toxicity solvent. Anal. Biochem. 90, 420–426. 38. Lin J, Liu L, Yang M, Lee M. (2004) Ethyl acetate/Ethyl alcohol mixtures as an alternative to Folch reagent for extracting animal lipids. J. Agric. Food Chem. 52, 4984–4986. 39. Boorman GA. (1999) Drinking water disinfection byproducts: review and approach to toxicity evaluation. Environ. Health Perspect. 107, 207–217. 40. Greim H, Reuter U. (2001) Classification of carcinogenic chemicals in the work area by the German MAK Commission: current examples for the new categories. Toxicology. 166, 11–23. 41. Wenk M, Lucast L, Di Paolo G, Romanelli AJ, Suchy SF, Nussbaum RL, Cline GW, Shulman GI, McMurray W, De Camilli P. (2003) Phosphoinositide profiling in complex lipid mixtures using electrospray ionization mass spectrometry. Nat. Biotechnol. 21, 813–817.

607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640

sdfsdf

Chapter 11

1

Lipidomic Analysis of Human Meibum Using HPLC–MSn

2

Igor A. Butovich

3

Summary 

4

High-pressure liquid chromatography–mass spectrometry (HPLC–MS) has become a de facto standard analytical tool in lipidomic analyses of complex biological samples. This technique offers the best combination of selectivity and sensitivity among the currently available analytical methods, and provides not only the retention times of analytes, but also their m/z values, from which molecular masses of the compounds can be deduced. Further enhancement of the technique comes from the fact that some of the MS instruments (also known as ion traps, or MSn instruments) are capable of multistage fragmenting of the analytes, thus enabling the researcher to perform their structural elucidation. These capabilities make HPLC–MSn an ideal tool for analyzing small, complex lipid samples such as human meibum. Meibum is a very complex lipid mixture which is secreted onto the ocular surface by Meibomian glands. Meibum plays a critical role in the biochemistry and physiology of the human ocular surface. However, despite all efforts, its (bio)chemical composition remains elusive. In this chapter, several HPLC–MSn methods developed for lipidomic analysis of human meibum will be discussed. Considering the nature of analytes (all of which are hydrophobic compounds poorly soluble in water, and most of which are electroneutral), the only MS technique used in the study will be atmospheric pressure chemical ionization (APCI) MSn in both the positive and the negative ion modes. Electrospray ionization technique, though useful in phospholipid analyses, was found to be inadequate for analyzing less polar compounds, such as wax esters. As the data provided in this chapter will show that meibum is composed predominantly of nonpolar lipids of wax ester, cholesteryl ester, and triacylglycerol families with no appreciable amounts of more polar lipids present, APCI MSn seems to be a method of choice for lipidomic analysis of meibum and similar lipid mixtures.

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Key words:  HPLC, Mass spectrometry, Atmospheric pressure chemical ionization, Ion trap, Polar lipids, Nonpolar lipids, Wax esters, Cholesteryl esters, Cholesterol, Phospholipids, Ceramides, Oleamide, Triacylglycerols

24 25 26

1. Introduction

27

Meibomian glands (MG) that are located at the perimeters of eyelids of humans and other mammals were originally described by H. Meibom in 1666 (1). MG are a major source of lipid Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_11, © Humana Press, a part of Springer Science + Business Media, LLC 2009

221

28 29 30

222 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77

Butovich

material (often referred to as meibum (2)) that is considered to be an important part of a so-called tear film lipid layer (TFLL) – an outermost part of tear film (TF) which covers the entire ocular surface, including cornea and conjunctiva (3, 4). TF and TFLL play a critical role in protecting the ocular surface from dehydration (5). Among other functions of TF and TFLL are antimicrobial, nutritional, and lubricating ones. The TF is also important in maintaining visual acuity as its thickness and uniformity affects refractive properties of cornea. Lipid composition of human TF and TFLL has been a subject of intensive research efforts for decades (see, for example, (2, 6–19)). However, considering relatively small amounts of samples that can be collected from human subjects without harming them (typically, less than 1 mg of dry meibum), and their extremely complex nature, a complete lipidomic analysis of meibum is yet to be completed. Earlier efforts in the area produced evidence that the vast majority of lipids found in meibum were of nonpolar nature (2, 3, 7–16). Among those reported were wax esters, cholesterol (Chl) and Chl esters, hydrocarbons, tri- and di-acylglycerols, di- and triesters of very complex nature (12), and many unidentified compounds. The fraction of nonpolar compounds typically was reported to comprise 60% or more of the entire lipid pool. The rest of the compounds were proposed to be of a more polar nature and included ceramides, sphingosines, various phospholipids, free fatty acids and fatty acid amides, monoacyl glycerols, etc. (16–22). It is noteworthy that all this body of data was generated over a period of 40 years and came from laboratories with extremely varying backgrounds (ophthalmic, biochemical, and chemical). The techniques utilized in those studies also varied with most common approaches being gas chromatography and/or gas–liquid chromatography with flame ionization detection and/or mass spectrometric detection. Also popular in the earlier studies was thin-layer chromatography, which later has been replaced by high performance liquid chromatography (HPLC). Later, attempts were made to utilize nuclear magnetic resonance spectrometry to study phospholipids (23) and infrared spectroscopy to study the composition and conformations of certain nonpolar lipids (24, 25). In recent years, mass spectrometry (MS) with or without HPLC has become a de facto standard for lipidomic analysis of complex lipid mixtures (16, 17, 26–31). Various MS techniques have been tried in meibum analyses with varying degree of success. Currently, its most advanced incarnation is HPLC–MS. This technique combines the sensitivity and selectivity of MS with the ability of HPLC to separate complex lipid mixtures thus improving the chances of correct identification of unknowns.



Lipidomic Analysis of Human Meibum Using HPLC–MSn

223

Several MS methods compatible with HPLC are available for a lipid chemist to choose from when designing a study. Those include, among others, electrospray ionization (ESI), atmospheric pressure chemical ionization (APCI), and atmospheric pressure photo ionization (APPI) techniques. However, a choice of an MS technique may be a determining factor in detecting a compound (or a set of compounds) or missing it altogether. ESI, for example, is very effective in detecting relatively polar lipids which include phospholipids, ceramides, and glycolipids (32). However, it is less than perfect in detecting nonpolar lipids such as wax esters and hydrocarbons. APCI, on the other hand, was successfully implemented in analyses of nonpolar lipids of meibum (29–31) and aqueous tears (31), and is generally considered a better technique than ESI for analyzing nonpolar compounds. APPI, being a newer technique than both ESI and APCI, is gaining ground as an effective tool in analyzing nonpolar lipids (33). In terms of its specificity, it is closer to APCI than to ESI, and, theoretically, could be used for analyzing even more hydrophobic (nonpolar) compounds than APCI could. Yet, it is less robust than the APCI technique because special care must be taken to avoid damaging the ultraviolet lamp of an APPI ion source. Among the three, APCI seems to be better suited for analyzing lipids of human meibum by HPLC–MS as it is very robust, sensitive, and compatible with a wide variety of HPLC solvents, and tolerates relatively high flow rates of a typical HPLC system. Another important consideration is a choice between an ion trap and a triple quadrupole mass detector. Without going into a lengthy discussion on the subject, it is sufficient to mention that triple quadrupole instruments are better suited for quantitation of the analytes, while ion traps offer unsurpassed capabilities for structural elucidation of unknown compounds. In particular, triple quadrupole mass spectrometers are capable of only MS and MS/MS (or MS2) experiments, while ion traps can easily perform multistage fragmentation up to MS5 and above. This comes at a price of a lower sensitivity of ion traps, and their narrower dynamic range. However, these disadvantages were found to be a nonissue in the lipidomic analysis of meibum, because collected samples were large enough to offset both the lower sensitivity and the narrower dynamic range. Thus, the capability of an ion trap to perform MSn analyses – an indispensable tool in studying of unknown compounds present in complex mixtures – made it a clear winner in this “competition.” In this manuscript the author will describe the methods of analyses of human meibum based on the ion trap APCI technique in conjunction with normal phase HPLC (NP HPLC).

78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123

224

Butovich

124

2. Materials

125

2.1. Equipment

126 127 128 129 130

1. An Alliance 2695 HPLC Separations Module (Waters Corp., Milford, MA) consisting of a quaternary gradient HPLC pump, thermostated column compartment, vacuum degasser, pulse dumper, and a built-in autoinjector. The HPLC system is controlled through an Empower (v.1.0) software installed on a PC-compatible computer operated under Windows XP.

143

2. A Lichrosphere Diol (3.2 × 150 mm, 5 mm) HPLC column (Phenomenex, Torrance, CA) for NP HPLC experiments (see Note 1, below). A Lichrosphere Si-60 silica gel (3.2 × 150 mm, 5 mm) HPLC column (from Sigma-Aldrich) for NP HPLC separation of phospholipids. 3. A Finnigan LCQ Deca XP Max ion trap mass spectrometer capable of performing MSn analyses from ThermoFisher Scientific (formerly, ThermoElectron) (Waltham, MA). The spectrometer must be equipped with an APCI ion source (see Note 2, below). An Xcalibur software (v.1.4 SR1) is used to operate the spectrometer and analyze the MS and HPLC data.

144

4. An AB135-S microbalance from Mettler (Toledo, OH).

131 132 133 134 135 136 137 138 139 140 141 142

147

1. HPLC- or spectroscopy grade n-hexane, propan-2-ol, acetonitrile, chloroform, and ethanol, mostly from Burdick&Jackson (Muskegon, MI).

148

2. Glacial acetic acid.

149 150

3. MilliQ-grade deionized water (18 MW) or bottled HPLCgrade water for preparing HPLC mobile phases.

151

4. 5 mM ammonium formate in HPLC-grade water.

152

5. Lipid standards used in the study were purchased from NuChek Prep, Inc. (Elysian, MN), Avanti Polar Lipids (Alabaster, AL), and Sigma-Aldrich (St. Louis, MO).

145 146

2.2. Reagents and Solvents

153 154

2.3. Supplies

156

1. An Eppendorf Easypet electric pipettor with glass pipettes ranging from 1 to 25 mL.

157 158

2. Positive displacement Digital Microdispensers (25 and 100 mL) from Drummond Sci. Co. (Broomall, PA).

159 160

3. HPLC-style microsyringes (10, 50, and 100 mL capacity) manufactured by SGE (Australia).

161

4. Block sample heater (Lab-Line Instruments, Melrose Park, IL).

162 163

5. One-milliliter flat-bottom clear glass shell vials with polyethylene snap caps (Waters Corp).

155



Lipidomic Analysis of Human Meibum Using HPLC–MSn

225

6. Total Recovery® glass HPLC vials with Teflon® liners (Waters Corp).

164

7. Borosilicate glass vials (19 × 65 mm, 11.3 mL) with caps equipped with Teflon® liners (VWR, Batavia, IL).

166

8. Glass solvent bottles with either glass stoppers or caps with Teflon® liners.

168 169

9. High-purity compressed nitrogen.

170

10. High-purity liquid nitrogen in a high-pressure tank.

171

3. Methods

3.1. Sample Collection and Handling

165 167

172

For general information on handling and storing lipids, the reader is advised to visit an Avanti Web page (http://www.avantilipids. com/storageandhandlingoflipids.html).

173 174

Disclosure: Human meibum was collected in a clinical setting from healthy volunteers using a procedure approved by the University of Texas Southwestern Medical Center Institutional Review Board and conducted in accordance with the Declaration of Helsinki. 1. Meibum is expressed from an eyelid of a volunteer using a plastic conformer and a Q-tip. The plastic conformer, placed between an eyelid and the eye protects the ocular surface, while the Q-tip is used to press the eyelid against the conformer to extrude meibum from an orifice of a MG. Immediately, the excreta are picked up with a polished platinum spatula. Care must be taken to avoid contamination of the sample with aqueous tears and cosmetic products. Meibum has a melting point of about 32–33°C (29) and, upon contact with the spatula, immediately solidifies at room temperature to assume a waxy structure. Its color is, typically, off-whitish to pale yellowish. Thus, it is easy to see if the sample has been successfully collected. Both upper and lower eyelids of both the eyes are used to collect the sample.

176 177

2. The solidified sample collected from one eyelid is immediately transferred into a preweighed flat-bottom HPLC glass vial filled with ~1.5 mL of chloroform: methanol solvent mixture (2:1, v/v; solvent CM). Dry weight of the vial is measured with precision of 0.01 mg or better before adding chloroform. An average meibum sample readily dissolves in 1.5 mL of the solvent mixture. The procedure is repeated for all four eyelids. All the samples are transferred into the same vial.

195 196

175

178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194

197 198 199 200 201 202

226

Butovich

3. Next, the sample solution is brought to dryness under a stream of nitrogen. The temperature of the sample is maintained by a thermostated sample heater at no more than 35°C. The vial with dry sample is weighed on the same microbalance and its weight is used to calculate the weight of the sample in the vial. On average, ~0.7 mg of meibum could be collected from four eyelids of a volunteer. The repeatability of the weight determination should be better than 0.01 mg. When the samples are to be stored for an extended period, the vials with dry material are flashed with dry nitrogen, sealed, and stored at −80°C. The dry samples were shown to be stable for at least 6 months to a year.

203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221

3.2. Sample Analyses 3.2.1. Nonpolar Lipid Analyses Using Isocratic NP HPLC–APCI MS 3.2.1.1. HPLC Procedures

2. A 100 mg/mL stock solution of meibum is prepared by dissolving the dry meibum material in an appropriate volume of a n-hexane:propan-2-ol solvent mixture (1:1, v/v, HP) and stored in a borosilicate 19 × 65 mm glass vial at −80°C under nitrogen. Just before the HPLC–MS analysis, the sample (50– 100 mL) is placed into a TotalRecovery® HPLC vial. Note that the cap must not have a silicon backing to a Teflon® liner as the tiny silicon particles picked up by the autoinjector’s needle will contaminate the sample and produce strong HPLC and MS signals. (Remember: if in doubt, leave the cap out, or remove the liner and its backing altogether. This will not affect the results of experiments if the sample is to be immediately injected).

222 223 224 225 226 227 228 229 230 231 232 233 234

3. Routinely, between 1 and 10 mL of the sample is injected. It was found that larger volumes of the stock solution would impact the shape and retention times of the HPLC peaks, while higher concentrations of the meibum solutions could overwhelm the detector and contaminate the ion source and/ or the HPLC column. Note that vacuum degassing of the solvent is advised to achieve the most stable baseline and avoid problems with air bubbles, but is not necessary if the sample concentration is sufficiently high.

235 236 237 238 239 240 241 242 243 244 245 246 247 248

1. A n-hexane:propan-2-ol:acetic acid solvent mixture (94:5:1, v/v/v, HPA) is used (see Note 3, below). The Diol column is pre-equilibrated in this solvent at 30°C for 15 min or until the back pressure and the MS signal stabilized. A 30-min analysis is performed isocratically at the flow rate of 0.3 mL/min and a column temperature of 30°C. The entire effluent is directed to the APCI ion source.

3.2.1.2. Mass Spectrometric Procedures

The mass spectrometer equipped with the APCI ion source is operated in the positive ion mode. Before the meibum analyses, the spectrometer is tuned using a 1 mg/mL behenyl oleate (BO) stock solution in the HPA solvent mixture. The sample is infused



Lipidomic Analysis of Human Meibum Using HPLC–MSn

227

at a flow rate of ~10 mL/min using the mass spectrometer’s builtin syringe pump. The Autotune routine of the Xcalibur software is engaged to tune to a signal with m/z value of 591.5 (a proton adduct of BO). The final MS parameters should be as follows: Source voltage

3 kV

Source current

5 mA

Vaporizer temperature

350°C

Sheath gas flow rate

20 arb. units

Capillary voltage

12 V

Capillary temperature

350°C

Tube lens voltage

−40 V

Multipole 1 offset

−2 V

Lens voltage

−41 V

Multipole 2 offset

−11 V

Entrance lens

−75 V

The parameters should be saved in an Autotune file and are used for meibum analyses and in experiments with standard nonpolar lipids. 3.2.1.3. Results HPLC–MS Analyses of Nonpolar Lipid Standards

Human meibum was reported to contain nonpolar lipids of various classes. Therefore, a range of standard lipids should be tested to determine their retention times under the conditions of NP HPLC analysis on a Diol column. A typical test mixture contains 50 mM each of behenyl oleate (a wax ester, MW 590, BO), cholesteryl oleate (a cholesteryl ester, MW 650, Chl-O) and free cholesterol (MW 386, Chl), triolein (a triacylglycerol, MW 885, TO), dipalmitin (a diacylglycerol, MW 569, DP), C18-ceramide (MW 566, C18-Cer), oleamide (a fatty acid amide, MW 281, OA), and 1-monomyristoyl glycerol (a monoacyl glycerol, MW 303, MG). Based on the published data, discussed in Subheading 1, compounds of these classes are expected to be present in human meibum. The lipids may be dissolved in either chloroform or the HP solvent mixture. No difference in the retention times of the analytes were observed provided that the sample injection volume did not exceed 10 mL. Note that to accommodate for the higher flow rates typical of HPLC-style experiments, the flow of sheath gas needs be increased to 40 arbitrary units. The rest of the MS parameters should remain as optimized in the Autotune routine. A sample NP chromatogram of the mixture is presented in Fig. 1. The retention times of the test lipids were found to be

249 250 251 252

253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277

228

Butovich

Fig. 1.Total ion chromatogram of a mixture of standard lipids (TIC) and reconstructed chromatograms of its individual components. Depicted are: BO (ion m/z 591; retention time 3.4 min); Chl-O (m/z 369, 3.5 min); Chl (m/z 369; 6.9 min); TO (m/z 885; 3.6 min); 1,2-DP (m/z 551; 3.9 min) and 1,3-DP (m/z 551; 6.4 min); C18-Cer (m/z 548; 13.7 min); OA (m/z 282; 19.2 min), and 1-MG (m/z 285; 22.8 min).



HPLC–MS Analyses of Nonpolar Components of Meibum

Lipidomic Analysis of Human Meibum Using HPLC–MSn

229

Chl-O » TO » BO<1,2-DP<1,3-DP » Chl
278

1. The HPLC–MS procedures are performed exactly as described above for standard lipids. Between 1 and 10 mL of the 100 mg/ mL stock solution of meibum is injected. A 1–3 mL aliquot is sufficient for routine detection of major components, while larger volumes (up to 10 mL) of more concentrated stock solutions of meibum (up to 1 mg/mL) should be employed to detect the minor compounds. Injection volumes larger that 10 mL

317

279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 318 319 320 321 322 323

230

Butovich

Fig. 2. Mass spectra of a mixture of standard lipids and of its individual components. Depicted are: integrated spectrum of all peaks shown in Fig. 1 (panel A); BO + Chl-O + TO (peaks 3.4–3.6 min; panel B); 1,3-DP (peak 6.4 min; panel C); Chl (peak 6.9 min; panel D); C18-Cer (peak 13.7 min; panel E); OA (peak 19.2 min; panel F), and 1-MG (peak 22.8 min; panel G). Because of the partial overlapping of certain HPLC peaks, the ions of neighboring HPLC peaks are visible in some of the spectra.



Lipidomic Analysis of Human Meibum Using HPLC–MSn

231

should be avoided because of the column and the detector overloading, as well as peak distortion caused by the large volume of the injection solvent. To detect most of the meibum components, the meibum samples are analyzed using general (observation) scan mode in the m/z range of 200–2,000.

324

2. Initially, the results are plotted as total ion chromatograms (TIC). The meibum samples produce one large TIC peak with a retention time of about 3.6 min and a shoulder at 4.3 min (Fig. 3a). Note that no other HPLC peaks are seen on the TIC of human meibum beyond 5 min. The major HPLC peak with a retention time of 3.6 min coelutes with standard wax esters, Chl esters, and triacylglycerols. The shoulder with RT 4.3 min is close to the retention times of 1,2-diacylglycerols.

329

3. The observation mass spectrum of the main peak integrated between 2 and 10 min is presented in Fig. 3b (see Note 4, below). The spectrum shows a major signal m/z 369 and an abundance of compounds with m/z between 500 and 850. There are no obvious HPLC peaks in the region between 10 and 30 min. Integration of that portion of the chromatogram shows no recognizable MS peaks above the noise level (Fig. 3c).

337

4. Once the mass spectrum has been obtained, one can plot individual MS signals as extracted chromatograms simulating a single ion monitoring experiment (SIM). As an example, let

344

Fig. 3. A typical chromatogram of a human meibum sample (panel A) and the positive mode APCI mass spectra of the integrated chromatogram regions 3–10 min (panel B) and 10–25 min (panel C).

325 326 327 328 330 331 332 333 334 335 336 338 339 340 341 342 343 345 346

232

Butovich

Fig. 4. An extracted chromatogram of ion m/z 369 (panel A) and the corresponding mass spectrum of the dehydrated Chl ion m/z 369 (panel B). The first large HPLC peak is of Chl-E; the second (smaller) peak with the retention time of 6.7 min is of free Chl. 347 348 349 350 351 352

us plot a signal with m/z 369 which is to show us whether Chlesters and free Chl are present in the meibum sample, and if the are, then in which ratio (Fig. 4). It becomes obvious that the sample contains a large amount of Chl-E (major peak with a retention time of 3–5.5 min), and a relatively small amount of free Chl (6.7 min). Integration of the HPLC peaks using



Lipidomic Analysis of Human Meibum Using HPLC–MSn

233

the Xcalibur’s built-in Avalon integration algorithm allows us to estimate that the fraction of Chl-esters comprises about 95% of the overall SIM chromatogram, while the HPLC peak of free Chl is less than 5% for this particular sample of meibum. Additional evidence is that the compounds are indeed Chlesters and Chl can be obtained in fragmentation experiments (see below).

353

5. The relative intensity of the signals, though important, does not provide an answer to a question “What is the molar ratio of the analytes in the sample?”, because different analytes ionize with different efficiencies, especially if secondary reactions (like the neutral loss of fatty acid residues in case of Chl-esters) are involved. Indeed, there are numerous examples of when a major compound in a sample produces a weak signal, while a minor compound with higher efficiency of ionization is seen as a major HPLC–MS peak. Thus, the need for additional experiments becomes obvious. To compare the HPLC–MS signals of Chl-esters and free Chl, let us prepare and analyze a mixture of 50 mM Chl-O and 50 mM Chl. The experiment shows (Fig. 5) that the relative peak areas for Chl-O and Chl (respectively, 50.5 ± 3.2 and 49.5 ± 3.2, mean ± SD) remain remarkably constant over the range of injected volumes. Thus, one can conclude that the apparent (visible) ratio of the extracted signals of Chl-esters and Chl (Fig. 4a) is close to the actual molar ratio of the components in meibum. For this particular sample, the molar ratio of Chl-esters and Chl was estimated to be 95.2:4.8.

360

Fig. 5. HPLC–MS analysis of an equimolar mixture of Chl-O and Chl (50 mM each). The compounds were detected using an analytical ion m/z 369.

354 355 356 357 358 359 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378

234

Butovich

6. When needed, similar experiments may be conducted for each of the detected components in meibum, provided the experimenter has corresponding lipid standards. Interestingly, the tests show zero presence of oleamide (Fig. 6a, b) and of other fatty acid amides reported earlier to be the major components of human meibum (17). Equally absent from the mixtures are monoacyl glycerols (not shown) and simple and long-chain ceramides, which are easily detectable by this method in samples of mouse skin lipid extracts (Fig. 7a, b) (35), but not in human meibum (Fig. 7c, d).

379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397

Structural Elucidation of Nonpolar Lipids

The lipids present in human meibum are very diverse. Therefore, it is impossible to provide universal guidelines for their structural elucidation which would be equally applicable to all the components of human meibum. Instead, a protocol successfully implemented for structural elucidation of wax esters and Chlcontaining species will be discussed below. 1. A generic protocol for wax ester analysis utilizes direct infusion of the sample dissolved in chloroform:methanol:acetic acid = 70:30:1 (v/v) solvent mixture (CMA). As an example, let us consider analysis of a signal with m/z 619. At the

Fig. 6. HPLC–MS analysis of a 1 mM solution of authentic oleamide (5 mL injection volume; panel A) and human meibum (100 mg/mL solution; 7 mL injection volume; panel B). The compound was detected using an analytical ion m/z 282 (M + H)+.



Lipidomic Analysis of Human Meibum Using HPLC–MSn

235

Fig. 7. Comparative HPLC–MS analyses of mouse skin lipid extracts (panels A and B) and human meibum (panels C and D). Chromatograms of representative mouse skin long-chain (HPLC peak 11.9 min; panel A) and normal (HPLC peak 13.7 min; panel B) ceramides and their corresponding mass spectra. Note that no such compounds were detected in human meibum (see also Fig. 3).

beginning, its m/z ratio should be determined in an MS1 experiment utilizing the Zoom Scan mode of the LCQ Deca XP Max mass spectrometer. The scanning range is set to m/z± 5 a.m.u. The resulting spectrum (Fig. 8a) shows that the compound

398 399 400 401

236

Butovich

Fig. 8. Structural elucidation of a meibum compound m/z 619 using multi-stage APCI MS analysis in the positive ion mode. Panel A. Mass spectrum of ion m/z 619 obtained in the Zoom Scan mode (MS1). Panel B. MS/MS fragmentation of the precursor ion m/z 619 (MS2). Panel C. MS/MS fragmentation of the fragment ion m/z 283 (MS3). Panel D. MS/MS fragmentation of the fragment ion m/z 265 (MS4). Panel E. MS/MS fragmentation of the fragment ion m/z 247 (MS5). Note that exactly the same transitions were recorded for standard oleic acid-based wax esters.



Lipidomic Analysis of Human Meibum Using HPLC–MSn

237

has an m/z value of 619.5. Then, the compound is fragmented in an MS2 experiment at relative energy of 35 V, it produces a series of fragments 283, 265, 255, and 247 (Fig. 8b). Fragmentation of ion m/z 283 in an MS3 experiment leads to ions m/z 265 (M − H2O + H+) and 247 (M − 2H2O + H+) (Fig. 8c), while in an MS4 experiment ion m/z 265 fragments to produce a daughter ion m/z 247 (Fig. 8d). Finally, fragmentation of ion m/z 247 in an MS5 experiment resulted in a series of smaller fragments 205 through 57 (Fig. 8e).

402

2. Importantly, exactly the same fragmentation pattern is observed for authentic stearyl oleate (m/z 535) and behenyl oleate (m/z 591) (not shown). These observations, along with very close retention times for all the compounds, allows us to conclude that the meibum compound in question (m/z 619) was a wax ester composed of a C18:1 fatty acid and a C24:0 fatty alcohol. Based on the fact that the most abundant monounsaturated fatty acid in human tissue is oleic acid, it is reasonable to assume that the tested compound has an oleic acid residue in it. The exact geometry (cis or trans) and location of the double bond, as well as the nature of the saturated fatty alcohol (straight or branched) remain to be determined in future experiments.

411

3. Similarly, other major compounds of wax ester nature depicted in Fig. 2b produce evidence that they are also based on oleic acid esterified to a range of saturated fatty alcohols with C18 to C30 carbons. However, some of the MS signals depicted there with m/z 575, 589, 603, 617, 631, 645, and 659 produce fragments with m/z values of 281, 263, and 245, which are indicative of linoleic (C18:2) acid (not shown).

423

4. As discussed above, human meibum contains large amounts of compounds that coelute with authentic Chl-esters and Chl, and spontaneously produce a common product ion m/z 369 in a way authentic Chl-containing compounds do. To obtain additional confirmation that these compounds are indeed based on Chl, one can employ fragmenting of the ion and analyzing its fragments in an MS2 experiment. In preliminary fragmentation experiments with authentic Chl and Chl-O it was found that ion m/z 369 produced a series of fragments depicted in Fig. 9a and b, if collision energy was set to be 38 V. This unique fragmentation pattern can be used to identify Chl and its derivatives in complex mixtures. Under identical conditions, ion m/z 369 derived from natural components of human meibum fragments exactly as authentic compounds do (Fig. 9c; fraction of Chl-esters is shown). Thus, the evidence gathered in HPLC and MSn experiments clearly demonstrate that the compounds from human meibum that produce ion m/z are indeed Chl and Chl-esters.

403 404 405 406 407 408 409 410 412 413 414 415 416 417 418 419 420 421 422 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446

238

Butovich

Fig. 9. Fragmentation patterns of ions m/z 369@38 detected for authentic free Chl (panel A) and Chl-oleate (panel B), and Chl-esters fraction of human meibum (panel C).

447 448 449 450 451 452 453 454 455 456 457 458

3.2.2. Polar Lipid Analysis Using Gradient NP HPLC–APCI MS 3.2.2.1. HPLC Procedures

Results of preliminary experiments based on isocratic elution of a series of standard polar lipids (mostly, phospholipids) were found to be unsatisfactory because of a poor separation of standards, sample dilution, peak broadening, and/or excessively long retention times of the analytes. Therefore, a gradient protocol for NP HPLC separation of the polar lipids has been developed. A ternary solvent gradient is based on n-hexane (Solvent A), propan-2-ol (Solvent B), and 5 mM aqueous ammonium formate (Solvent C) (Table 1). The solvent should be constantly degassed using a built-in vacuum degasser of the Waters Alliance 2695 Separations module. Polar lipids are separated on a Lichrosphere Si-60 (3.2 × 150 mm, 5 mm) HPLC column. As a test polar lipid



Lipidomic Analysis of Human Meibum Using HPLC–MSn

239

Table 1  HPLC gradient for polar lipid analysis on a Lichrosphere Si-60 (3.2 × 150, 5 mm) silica gel columna Step

Time (min)

1

Flow (mL/min)

Propan-2-ol (%)

5 mM NH4COOH in H2O (%)

n-Hexane (%)

0.5

40

2

58

2

 5

0.5

40

2

58

3

20

0.5

40

5

55

4

40

0.5

40

5

55

5

41

1

40

5

55

6

45

1

40

5

55

7

46

0.5

40

2

58

8

60

0.5

40

2

58

Column temperature was 35 ± 1°C

a 

3.2.2.2. Mass Spectrometric Procedures

3.2.2.3. Results HPLC–MS Analyses of Phospholipid Standards

solution, a 10 mg/mL solution of each of 1,2-dimyristoyl-phosphatidylethanolamine (MMPE, MW 674), 1-palmitoyl-2-oleoylphosphatidic acid (POPA, MW 674), C16:0-sphingomyelin (C16-SM, MW 703), 1-palmitoyl-2-oleoyl-phosphatidylglycerol (POPG, MW 748), 1-palmitoyl-2-oleoyl-phosphatidylcholine (POPC, MW 760), and 1,2-distearoyl-phosphatidylserine (SSPS, MW 789) is recommended. The choice of the standards is based on the reported composition of human tears (see Subheading 1). The entire flow of the effluent is directed into the ion source of the mass spectrometer.

459

The analytes are monitored in the negative ion mode. The analysis of phospholipids and sphingomyelins in the negative ion mode is possible due to the presence of phosphoric acid residue in their structures. The spectrometer is autotuned using the direct infusion technique and a 10 mg/mL solution of POPC in propan2-ol:5 mM aqueous ammonium formate:n-hexane = 40:5:55 (v/v/v) solvent mixture. The following MS parameters were implemented:

468

Five well-resolved HPLC peaks can be easily detected on the chromatograms (Fig. 10). Let us examine their corresponding mass spectra. The peak with a retention time of 14 min produces an MS signal with m/z 747.7, which is a signal of POPG (M − H+). The next peak (at 16 min) is identified as MMPE because of its prominent ion m/z 634.6 (M − H+). The peak with a retention time of 27.4 min produces ion 788.5 indicative of

476

460 461 462 463 464 465 466 467 469 470 471 472 473 474 475 477 478 479 480 481 482

240

Butovich Source voltage

2 kV

Source current

6 mA

Vaporizer temperature

375°C

Sheath gas flow rate

40 arb. units

Capillary voltage

−15 V

Capillary temperature

300°C

Tube lens voltage

−30 V

Multipole 1 offset

4.75 V

Lens voltage

16 V

Multipole 2 offset

7V

Entrance lens

61 V

Fig. 10. A representative chromatogram of a mixture of six polar lipids analyzed using NP HPLC–APCI MS in the negative ion mode. The analyzed lipids were: POPG (retention time 14.0 min; m/z 747.7), MMPE (16 min; m/z 634), POPA (19.1 min; m/z 673.5), SSPS (27.4 min; m/z 788.5), POPC (31 min; m/z 744.7), and C16-SM (33.5 min; m/z 687.6).

483 484 485 486 487 488 489 490 491 492 493 494 495

SSPS (M − H+), while the remaining two peaks (at 31 and 33.5 min) are, correspondingly, POPC [m/z 744; (M − CH3 − H+)] and C16-SM [m/z 687; (M – CH3 − H+)]. Missing is an HPLC peak of POPA. However, upon closer examination of the chromatogram, a very small HPLC peak with a retention time of ~19 min can be found. It produces the previously unaccounted for signal of POPA with m/z value of 673 (M − H+). Knowing the analytical signals of all the lipids in the test mixture, their individual chromatograms can be reconstructed. Note that because of the noticeable differences in the efficiency of ionization of the compounds, the chromatograms of the individual ions should be normalized (Fig. 11). The retention time of the tested lipids are POPG<MMPE


Lipidomic Analysis of Human Meibum Using HPLC–MSn

241

Fig. 11. Reconstructed chromatograms of six individual polar lipids. The analyzed lipids were: POPG (retention time 14.0 min; m/z 747.7), MMPE (16 min; m/z 634), POPA (19.1 min; m/z 673.5), SSPS (27.4 min; m/z 788.5), POPC (31 min; m/z 744.7), and C16-SM (33.5 min; m/z 687.6). HPLC–MS Analyses of Polar Components of Meibum

1. The HPLC–MS procedures are performed as described above for standard lipids. Between 5 and 20 mL of a stock solution of meibum (~0.5 mg/mL in solvent CM) should be injected. The upper limit of injection (20 mL) is intentionally increased compared to the previously recommended

496 497 498 499 500

242

Butovich

10 mL limit of injection in an attempt to maximize signals of the minor components in meibum which may be present in the sample. This setting should be used for this purpose only. A typical total ion chromatogram of meibum is presented in Fig. 12. Routinely, there is only one noticeable HPLC peak detected, whose retention time of ~1.5 min precludes it from being a member of the phospholipid and sphingomyelin families. Indeed, in tested conditions standard polar lipids varying from acidic POPG to zwitterionic POPC and C16-SM have retention times in the range of 10–40 min (Fig. 11). Therefore, the meibum peak with retention time of 1.5 seems to be formed of much more nonpolar (or hydrophobic) compounds which do not interact with the silica gel matrix of the Lichrosphere Si-60 column.

501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546

2. Next, let us examine the rest of the meibum chromatogram. Considering the great variability of lipids and the absence of clearly visible HPLC peaks, looking for an unknown compound in the lipid mixture as complex as meibum is similar to looking for a needle in a haystack. Therefore, the following approach seems to be a logical choice. The chromatogram is divided into 30 2-min intervals (0–2 min, 2–4 min, etc.) and those are examined one after another using the moving selection tool of the Xcalibur software. This is achieved by pressing and holding down the left button of the mouse and selecting the desired portion of the chromatogram. Then, the red selection box can be freely moved along the chromatogram using the left and the right cursor keys.

To maximize the putative lipid signals, the background noise is minimized by using “Action → Subtract Spectra → 2 Ranges” sequence from the Xcalibur’s drop-down menu. The two ranges to be subtracted are those to the immediate left and to the immediate right from the region to be examined. For example, when examining a region between 10 and 12 min, the two subtracted regions should be 8–10 min and 12–14 min. This procedure allows accounting for ever- changing background noise caused by the HPLC solvent gradient. Of course, the interval can be set to any number between 0 and the entire duration of the experiment. However, considering an average width of the HPLC peaks (Fig. 10), the 2-min interval seems to be a practical choice.

3. The approach discussed above results in detection of a set of extremely weak MS signals presented in Fig. 12, which elute close to POPC and C16-SM (Fig. 11). The m/z values of the unknown compounds are indicative of a mixture of various phosphatidyl cholines and sphingomyelins. Note that the intensity of the corresponding HPLC peaks (plotted using the integrated MS signals in the m/z range of 700–840)



Lipidomic Analysis of Human Meibum Using HPLC–MSn

243

Fig. 12. A reconstructed chromatogram of polar lipids of human meibum. The chromatogram of integrated MS signals m/z 700–850 is shown. The first large peak with retention time of 1.5 min showed the presence of a series of nonpolar compounds, whose MS spectra are shown next to the HPLC peak. Note that the fastest-moving standard phospholipid, POPG, had a retention time of 14 min (Fig. 11). Two minor HPLC peaks of polar lipids with retention times between 27 and 29 min were detected in human meibum. Their retention times and mass spectra (shown next to the peak) were indicative of phosphatidyl cholines and sphingomyelins. The overall intensity of their signals was less than 0.1% of the intensity of the main peak.

3.3. Concluding Remarks

is about 0.15% (or less) of the peak with the retention time of 1.5 min. The minuscule amounts of these newly detected compounds make it very hard to evaluate them structurally and quantitatively. Experiments are in progress to address this issue. No other MS signals can be reliably detected above the noise level. This does not necessarily mean that no other minor components are present in the sample. However, their detection would most certainly necessitate performing more sensitive SIM experiments, which, in turn, require prior knowledge of the m/z values of the compounds in question.

547

The HPLC–MSn procedures highlighted above provide a solid basis for lipidomic analysis of human meibum and similar lipid mixtures. Compared to some previous studies, the use of HPLC– APCI MSn greatly facilitated analyses of nonpolar lipid components of meibum. However, the complexity of meibum calls for use of several different HPLC–MS protocols for lipid detection, each of which is tailored for a particular group of lipids. Even more complicated is structural evaluation of the meibum components, which requires designing and performing numerous HPLC–MSn experiments with unknown compounds and lipid standards. In this brief chapter, one could not possibly cover all the aspects

557

548 549 550 551 552 553 554 555 556

558 559 560 561 562 563 564 565 566 567

244

Butovich

of the lipidomic analysis of meibum. This area is undergoing a rapid development (24–31) and is bound to provide answers to questions about the composition and structure of the normal tear film, and its changes during an onset and development of various ocular diseases and pathological states such as ocular inflammation, dry eye, Sjögren’s syndrome, and others.

568 569 570 571 572 573

574

4. Notes

575

1. A Lichrosphere Diol (3.2 × 150 mm, 5 mm) HPLC column (Phenomenex, Torrance, CA) used in NP HPLC experiments can be substituted with an equivalent Merck LiChrospher® Diol product, on which it is based. Minor adjustments of the corresponding HPLC protocol might be needed to compensate for the differences in packing and geometry of the columns, and differences between different batches of the packing material.

576 577 578 579 580 581 582 583

2. HPLC–MSn routines described in this Chapter require using an APCI ion source. A much “softer” ESI ionization technique will produce inferior results when analyzing nonpolar lipids.

584 585 586

3. The composition of HPLC solvents is critical in detecting hydrophobic compounds such as wax esters: the best results are obtained when using n-hexane-propan-2-ol solvent mixtures with acetic acid as proton donor. In solvent mixtures based on aqueous ammonium formate, the efficiency of ionization of wax esters was much lower than in the presence of acetic acid. This observation partly explains the apparent absence of wax esters signals in the mass spectra of human meibum and tears published earlier.

587 588 589 590 591 592 593 594 595

4. Depending on the degree of contamination of meibum with aqueous tears, minor differences in the lipid analyses are expected (29–31). However, if collected properly, meibum samples are remarkably consistent in terms of their major lipid components.

596 597 598 599

600

601 602 603 604 605

Acknowledgments The author acknowledges support from Research to Prevent Blindness (New York, NY) and from NIH in the form of an unrestricted core grant EY-016664. The author thanks Drs. Wojciech Kedzierski and Ann McMahon for providing a sample of skin lipid extract.



Lipidomic Analysis of Human Meibum Using HPLC–MSn

245

References 1. Meibom, H. (1666) De vasis palpebrarum novis epistolae. H. Müller, Helmstädt. 2. Pes, O. (1897) Ricerche microchimiche sulla secrezione delle ghiandole sebacee palpebrali. Arch. Ottal. 5, 82–91. 3. Andrews, J. S. (1970) Human tear film lipids. I. Composition of the principal non-polar component. Exp Eye Res. 10, 223–227. 4. Brauninger, G. E., Shah, D. O., and Kaufman, H. E. (1972) Direct physical demonstration of oily layer on tear film surface. Am J Ophthalmol. 73, 132–134. 5. Bron, A. J., Tiffany, J. M., Gouveia, S. M., Yokoi, N., and Voon, L. W. (2004) Functional aspects of the tear film lipid layer. Exp Eye Res. 78, 347–360. 6. Linton, R. G., Curnow, D. H., and Riley, W. J. (1961) The meibomian glands: an investigation into the secretion and some aspects of the physiology. Br. J. Ophthalmol. 45, 718–723. 7. Nicolaides, N. (1965) Skin lipids. II. Lipid class composition of samples from various species and anatomical sites. J. Am. Oil Chem. Soc. 42, 691–702. 8. Cory, C. C., Hinks, W., Burton, J. L., and Shuster, S. (1973) Meibomian gland secretion in the red eyes of rosacea. Br. J. Dermatol. 89, 25–27. 9. Tiffany, J. M. (1978) Individual variations in human meibomian lipid composition. Exp. Eye Res. 27, 289–300. 10. Nicolaides, N., Kaitaranta, J. K., Rawdah, T. N., Macy, J. I., Boswell 3rd, F. M., and Smith, R. E. (1981) Meibomian gland studies: comparison of steer and human lipids. Invest. Ophthalmol. Vis. Sci. 20, 522–536. 11. Nicolaides, N., and Ruth, E. C. (1982–1983) Unusual fatty acids in the lipids of steer and human meibomian gland excreta. Curr. Eye Res. 2, 93–98. 12. Nicolaides, N. and Santos, E. C. (1985) The di- and triesters of the lipids of steer and human meibomian glands. Lipids. 20, 454–467. 13. Shine, W. E. and McCulley, J. P. (1993) Role of wax ester fatty alcohols in chronic blepharitis. Invest. Ophthalmol. Vis. Sci. 34, 3515–3521. 14. Shine, W. E. and McCulley, J. P. (1996) Meibomian gland triglyceride fatty acid differences in chronic blepharitis patients. Cornea. 15, 340–346. 15. McCulley, J. P. and Shine, W. E. (1998) Meibomian secretions in chronic blepharitis. Adv. Exp. Med. Biol. 438, 319–326. 16. Sullivan, B. D., Evans, J. E., Dana, M. R., and Sullivan, D. A. (2006) Influence of aging on the polar and neutral lipid profiles in human

606

meibomian gland secretions. Arch. Ophthalmol. 124, 1286–1292. 17. Nichols, K. K., Ham, B. M., Nichols, J . J., Ziegler, C., and Green-Church, K. B. (2007) Identification of fatty acids and fatty acid amides in human meibomian gland secretions. Invest Ophthalmol Vis Sci. 48, 34–39. 18. Dougherty, J. M. and McCulley, J. P. (1986) Analysis of the free fatty acid component of meibomian secretions in chronic blepharitis. Invest. Ophthalmol. Vis. Sci. 27, 52–56. 19. Shine, W. E., and McCulley, J. P. (2000) Association of meibum oleic acid with meibomian seborrhea. Cornea. 19, 72–74. 20. Shine, W. E., and McCulley, J. P. (1998) Keratoconjunctivitis sicca associated with meibomian secretion polar lipid abnormality. Arch. Ophthalmol. 116, 849–852. 21. Shine, W. E., and McCulley, J. P. (2003) Polar lipids in human meibomian gland secretions. Curr. Eye Res. 26, 89–94. 22. Shine, W. E., and McCulley, J. P. (2004) Meibomianitis: polar lipid abnormalities. Cornea. 23, 781–783. 23. Greiner, J. V., Glonek, T., Korb, D. R., and Leahy, C.D. (1996) Meibomian gland phospholipids. Curr Eye Res. 15, 371–375. 24. Borchman, D., Foulks, G. N., Yappert, M. C., and Ho, D. V. (2007) Temperature-induced conformational changes in human tearlipids hydrocarbon chains. Biopolymers. 87, 124–133. 25. Borchman, D., Foulks, G. N., Yappert, M. C., Tang, D., and Ho, D. V. (2007) Spectroscopic evaluation of human tear lipids. Chem. Phys. Lipids. 147, 87–102. 26. Ham, B.M., Jacob, J. T., Keese, M. M., and Cole, R. B. (2004) Identification, quantification and comparison of major non-polar lipids in normal and dry eye tear lipidomes by electrospray tandem mass spectrometry. J. Mass. Spectrom. 39: 1321–1336. 27. Joffre, C., Souchier, M., Grégoire, S., Viau, S., Bretillon, L., Acar, N., Bron, A. M. and Creuzot-Garcher, C. (2008) Differences in meibomian fatty acid composition in patients with meibomian gland dysfunction and aqueous-deficient dry eye. Br. J. Ophthalmol. 92, 116–119. 28. Souchier, M., Joffre, C., Grégoire, S., Bretillon, L., Muselier, A., Acar, N., Beynat, J., Bron, A., D’Athis, P., and Creuzot-Garcher, C. (2008) Changes in meibomian fatty acids and clinical signs in patients with meibomian gland dysfunction after minocycline treatment. Br. J. Ophthalmol. 92, 819–822.

607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665

246 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682

Butovich

29. Butovich, I. A., Uchiyama, E., and McCulley, J. P. (2007) Lipids of human meibum: massspectrometric analysis and structural elucidation. J. Lipid Res. 48, 2220–2235. 30. Butovich, I. A., Uchiyama, E., Di Pascuale, M. A, McCulley, J. P. (2007) Liquid chromatography-mass spectrometric analysis of lipids present in human meibomian gland secretions. Lipids. 42, 765–776. 31. Butovich, I. A. (2008) On the lipid composition of human meibum and tears: Comparative analysis of nonpolar lipids. Invest. Ophthalmol. Vis. Sci. 49, 3779–3789. 32. Levery, S. B. (2005) Glycosphingolipid structural analysis and glycosphingolipidomics. Methods Enzymol. 405, 300–369.

33. Cai, S. S., Short, L. C., Syage, J. A., Potvin, M., and Curtis, J. M. (2007) Liquid chromatography-atmospheric pressure photoionization-mass spectrometry analysis of triacylglycerol lipids–effects of mobile phases on sensitivity. J Chromatogr A. 1173, 88–97. 34. http://www.lipidlibrary.co.uk/Lipids/dg/ index.htm 35. McMahon, A., Butovich, I. A., Mata, N. L., Klein, M., Ritter, R. 3rd, Richardson, J., Birch, D. G., Edwards, A. O., and Kedzierski, W. (2007) Retinal pathology and skin barrier defect in mice carrying a Stargardt disease-3 mutation in elongase of very long chain fatty acids-4. Mol Vis. 13, 258–272.

Chapter 12 Lipid Geographical Analysis of the Primate Macula by Imaging Mass Spectrometry Timothy J. Garrett and William W. Dawson Summary Imaging mass spectrometry (IMS) provides a unique method to probe for chemical distributions within tissue sections with high chemical specificity. The direct analysis of tissue sections by mass spectrometry, which is the field of IMS, is relatively young, 10 year old; however, the techniques for mass spectrometric analysis are well known. Critical aspects of IMS then are the preparation of tissue specimens for insertion into a vacuum chamber and the interpretation of results with respect to disease studies. Here, we describe the methodologies for geographic localization of phospholipids in flat-mounted eye segments from rhesus monkey using IMS with matrix-assisted laser desorption/ionization (MALDI) and tandem mass spectrometry. Key words:  Imaging mass spectrometry, MALDI, Tandem MS, Macula, Phospholipids

1. Introduction According to the National Library of Medicine, over 9,000 clinical and scientific papers have been written on age-related macular eye disease (AMD) since 1950. AMD is a blinding disease that accounts for most of the visual loss in aging, western world, and the group 50 and above. Approximately 30% of the American population 70 years and older show some evidence of AMD (1, 2). The clinical hallmarks of AMD are drusen, which may begin as small white or yellowish spots with considerable variation. Progression may produce confluence areas often with pigmentary disturbances. Most drusen are visible during direct ophthalmoscopic examination of the fundus in humans or other primates (Fig. 1). The disease may be divided into the “dry” disease, Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_12, © Humana Press, a part of Springer Science + Business Media, LLC 2009

247

248

Garrett and Dawson

Fig. 1. Ophthalmoscopic image of a 30° posterior segment of an eye. Drusen appear as small pigment variations in the macular, M, region of the eye. F indicates the fovea, ONH indicates the optic nerve head (papilla), and TP indicates the temporal periphery.

which results in atrophy of the outer layers of the retina, or the “wet” disease, where there are hemorrhagic events that can produce blindness. Drusen are defined, by classical histopathology, as clumps of cell debris which exist between the retinal pigmented epithelium and Bruch’s membrane. That is, between a highly metabolic receptor cell layer and the major circulatory supply, the choriocapillaris. After 50 years of research, the foundational events leading to drusen expression are still poorly understood (3, 4). In a unique Florida monkey colony, we have recently discovered that the presence of clinically apparent drusen clumps corresponds well with the regions of clumped formations of lipid-filled pigment epithelium cells (3). In contrast, the classical drusen seen using routine histopathological techniques, correspond very poorly to the clinically apparent drusen in well-prepared primate tissue (3). This is probably due to the elimination of lipids by the organic solvents which are used in the dehydration of the tissue and in the preparation of plastic or paraffin infiltration. Imaging mass spectrometry (IMS) is a new analytical technique that offers the ability, for the first time, to identify the specific chemical composition and geographic distributions of these lipid-filled pigment epithelium cells by directly scanning the flat retina and identifying the compounds at each specific point (5). Imaging tissue sections by matrix-assisted laser desorption/ ionization mass spectrometry (MALDI–MS) is a young and rapidly growing field in mass spectrometry (6–8). This is related in part to the prospects of analyzing for very specific molecular species (9, 10) directly from sectioned tissue and the correlation of histological techniques with the results obtained from IMS (11, 12).



Lipid Geographical Analysis of the Primate Macula by Imaging Mass

249

Another significant opportunity is in the identification of known and unknown compounds from the tissue surface using tandem MS (13–16). This aspect is possible only using instruments capable of performing tandem mass spectrometry, which aids in the identification of an unknown using product ion spectra. The field of IMS pertains to the direct analysis of tissue using a focused laser, as in MALDI, or ion, as in secondary ion mass spectrometry (SIMS), beam (6, 17–19). In order to create an image, the sample is rastered in a discrete pattern with respect to the laser or ion beam. The typical process for generating images from fresh tissue sections by mass spectrometry using MALDI is displayed in Fig. 2. The process starts with the sectioning of a tissue sample, typically to 10 mm or less. The sample is then coated with a high ultra-violet absorbing small molecule, termed a matrix, to ensure ionization of intact species. After coating, the tissue is analyzed by mass spectrometry where each spot that the laser interrogates generates a mass spectrum, considered as a pixel with dimensions equal to the spot size of the laser and discrete x and y coordinates (a position-specific mass spectrum) (20, 21). After rastering across the entire tissue, images can then be generated

782.8

100

Extract to Generate image

90 80 70 60 50 40

Section the tissue

772.7 769.8

723.9

720

740

760

780 m/z

810.7

826.7

751.9

10 0

798.7

753.8

30 20

Position Specific Mass Spectrum

756.9

800

820

828.6 832.7

848.7

840

Send to MS

Place on microscope slide

Coat with matrix

Raster laser across tissue in MS

Fig. 2. Traditional imaging process for thin brain tissue section processing by mass spectrometry.

250

Garrett and Dawson

by extracting the intensity of an individual ion from every “pixel” to display the intensity of that ion with respect to position on the tissue. Spatial resolution in IMS with MALDI is limited to the size of the focused laser beam. Most commercial instruments have laser beam sizes from 60 to 120 mm. Smaller spot sizes are attainable, but not typically employed. Lipids, particularly phospholipids (PL), have been of considerable interest in IMS studies. Studies of PLs in rat brain have shown the ability of this technique coupled with tandem mass spectrometry to methodically characterize the PLs from a single brain tissue section using tandem mass spectrometry (both MS2 and MS3) (22). Phospholipids comprise a wide-range of compounds built from a glycerol backbone and separated into categories based on their head group. Each class of PL has two varying fatty acyl chains attached to the sn-1 and sn-2 carbons of the glycerol backbone. The use of mass spectrometry for lipid analysis enables the classification of individual PL species based on the arrangement of the fatty acyl chains, which is important for direct tissue analysis because of the variable distribution of each PL in a tissue section as has been shown previously in brain tissue (15, 23–25). Using tandem mass spectrometry, classes of lipids can be determined based on known fragmentation pathways allowing for a more complete understanding of lipid expression in the tissue sample. For example, all phophatidylcholine (PCho) species fragment to m/z 184 when they are protonated ([M + H]+) and produce a neutral loss of 59 when they are sodiated ([M + Na]+), which allows for identification of the lipid class and the ion of interest (24, 26, 27). Knowing the ion formed is important for direct tissue analysis because of the natural presence of ionizing species in the tissue, such as sodium or potassium. Recently, we have begun to explore relations between drusen (28) and lipid-filled pigmented epithelium (PE) (3) in the Florida Colony monkeys with clinical drusen. In old monkey and human flat-mounted donor retinal tissue, we have found, using IMS, that the mass spectra of lipids are geographically distributed in the regions where there were clinical drusen. Now it becomes important to better understand the pathophysiology of lipid-filled PE cells, their environment, and how they contribute to signs of AMD. With the ability to define the chemical constituents of the drusen environment using IMS in the retina, we can begin to geographically localize abnormality and elucidate the mechanisms of any underlying macular degeneration. Understanding the chemicals associated with AMD will aid in the development of noninvasive clinical strategies to diagnose (and perhaps treat) early macular abnormality. In this chapter, we present the techniques used to probe flat-mounted eye segments for lipid content using histological and mass spectral techniques.



Lipid Geographical Analysis of the Primate Macula by Imaging Mass

251

2. Materials 2.1. Equipment

1. Artistic airbrush (Aztek A470) from Testors (Rockford, IL). 2. Thermo LTQ with vMALDI ion source (see Note 1 for description).

2.2. Reagents and Supplies

1. Sodium acetate and 2,5-dihyroxybenzoic acid were purchased from Fisher Scientific (Fairlawn, NJ). 2. Methanol and water were HPLC grade and were purchased from Fisher Scientific. 3. Nitrocellulose (BioTrace™ NT) was purchased from Pall Life Sciences (East Hills, NY). 4. Double sided-tape (Scotch). 5. Half-inch 27–30 gauge needle.

3. Methods 3.1. Preparation of the Retina as a “Whole Mount” or “Flat-Mount” (Morning Glory) Style (See Note 2)

Figure 3 shows the IMS process for flat-mounted eye tissue. The process is identical to sections of tissue prepared on a cryostat except for the use of nitrocellulose for mounting (step 2, Fig. 3). The process of preparation is described in more detail in this section. 1. A stab wound 3–4 mm in length is made in and parallel to the ora serrata of the eye. 2. A half-inch 27–30 gauge syringe needle is inserted to the posterior-center of the vitreous body and 10 mL of 4% paraformaldehyde in buffered salt solution is slowly perfused into the posterior segment, displacing some vitreous. 3. With sharp microdissecting scissors the anterior segment is removed leaving a 120° cup where the optic nerve head is centered. 4. After 24 h in 50 mL of 4% paraformaldehyde, the posterior segment is saved in 0.5% paraformaldehyde or flat-mounted. 5. To flat-mount, the optic nerve and papilla are removed and a flat-mount “morning glory” is made by cutting the cup into 5–6 petals with the papilla opening at the center of the 40–50° uncut center disc. 6. The fovea is at the inner edge of the one marked petal. Using an operating microscope and floating the tissue, the sclera is teased away.

252

Garrett and Dawson 100 90

672.4

80

1

782.6

Position-specific spectrum 760.5

70

726.8 753.6

60 50

809.5

650.4

40

848.1 837.7

30 20 10 0

600

650

700

750

m/z

Prepare flat-mount

2

800

850

900

Send to MS and Raster tissue

4

Section cut away for reference point

Optic nerve head

3

Macula region Place flat-mount on nitrocellulose

Coat with matrix

Fig. 3. IMS process for analyzing flat-mounted eye segments. The process is identical to the typical process except for the addition of nitrocellulose. Nitrocellulose is needed (step 2) to ensure adhesion of the tissue to a surface that can be transported to the instrument target plate.

3.2. Mounting Tissue for Imaging Mass Spectrometry Analysis

1. Cut a small section of nitrocellulose paper, large enough to contain the entire flat-mount. (see Note 3) 2. Cut a small section out of the flat-mount to assist in image registration. We cut a small section out of the flap containing the macula. 3. Place nitrocellulose paper into petri dish with flat-mount and fixative (underneath the flat-mount). 4. Float wet eye flat-mount onto the nitrocellulose paper. 5. Remove excess fixative using a glass pipette. 6. Dry slightly under a stream on nitrogen (approximately 30 min). Do not dry completely, but do not leave visible fixative drops on the tissue (see Note 4).



Lipid Geographical Analysis of the Primate Macula by Imaging Mass

3.3. Applying Matrix Compound, 2,5Dihydroxybenzoic Acid, to Flat-Mount

253

1. Prepare a solution of 2,5-dihyroxybenzoic acid (DHB) at a concentration of 40 mg/mL in 70/30 methanol water. If sodiation of the lipids is desired, add sodium acetate at a concentration of 10 mM. 2. Adjust airbrush to be 15 cm from the tissue surface. 3. Airbrush should have nitrogen supplied for nebulization 4. Fill airbrush vial with matrix solution. 5. Place an empty glass slide under the airbrush to adjust deposition parameters. 6. Adjust spray settings to ensure that excessive wetting does not occur. 7. Pass empty glass slide under the spray at a slow but steady rate. 8. View the glass slide under light to evaluate an even coating. A gloved finger can be used to wipe the matrix coating. If matrix smears, the slide is too wet, and if the matrix comes off the slide, it is too dry. 9. Once settings are determined, pass flat-mount under the airbrush stream in a back and forth motion. After each back and forth coating, move the sample horizontal, but keeping a little overlap from the previous coating. Coat entire tissue. This is one pass. A total of 20 passes is typically needed to apply sufficient matrix for MS analysis. 10. After coating is complete, dry sample for a minimum of 30 min under nitrogen. 11. Place the sample on the instrument target plate using double sided tape.

3.4. Mass Spectrometric Analysis

1. Scan the image using instrument software. 2. Select the area of analysis. If the whole tissue section is desired, encircle all the tissue. 3. Set mass range. Typically, the mass range is set from 180 to 1,000. The PLs appear in the region from around m/z 650 to 900. Lysophospholipids are detected from m/z 400 to 600. 4. Set laser power and the number of laser shots by manually interrogating the sample in ten random spots. The total ion current should be close to 1 × 105 in order to avoid deleterious effects of space-charge. On the instrument used in these studies, if automatic gain control (AGC) is employed, the laser shots are not set. 5. Set the step size of the raster. This is set to the spot size of the laser.

254

Garrett and Dawson

We verified that fixation conditions used did not aversely affect the phospholipid content of tissue section by comparing two rat brain tissue sections (10 mm). One section was sliced from freshly frozen tissue, while the other was sliced from tissue fixed under the same conditions employed for eye flat-mounts. Figure 4 shows two mass spectra and two images for the fresh frozen tissue (top spectrum and image) and the fixed tissue (bottom spectrum and image). There are two key comparisons to make from this data. First, the mass-to-charge values (m/z) are very similar, which indicates that oxidation is not occurring and second, images generated from specific ion signals, show expected distributions in this tissue. This second point verifies that the analytes are not migrating in the tissue during the fixation procedure. The images are for the same ion, m/z 782, which is the sodiated ion of PCho (16:0, 18:1), and are not identical because the coronal sections were prepared from different areas of the brain. A key difference in the two spectra is the absence of potassiated adducts in the spectrum from the fixed tissue (bottom). It is believed that potassium is removed from the tissue during the fixation procedure. Sodium is added during the matrix deposition phase to induce production of sodiated adducts. Potassiated adducts are indicated with an asterisks in the top spectrum and appear at m/z values of 772.72 and 798.66. These are potassiated adducts of the ions at m/z 756.74 and 782.67, respectively.

3.5. Results 3.5.1. Verification of Fixation Conditions

Fresh frozen tissue

100

782.67

80

Relative Abundance

60 40 20

785.82

756.74

810.66 798.66 753.82 772.72 * 828.60 713.20723.74745.05 * 832.81 856.63

0 700

100

720

740

760

780

Paraformaldehyde fixed

820

840

860

880

900

782.90

756.96

80

800

Absence of potassiated adducts

60 40

727.40

754.08 761.01 735.04

20 0 700

720

740

760

780

785.93 810.89 804.85 800 m/z

832.82 820

840

858.74 880.72 860

880

900

Fig. 4. Comparison of fresh frozen tissue (top spectrum and image) and paraformaldehyde fixed tissue (bottom spectrum and image). The m/z values detected are similar in both cases indicating that the fixation procedure does not alter the phospholipid expression in brain tissue. The MS images are for the ion at m/z 782, which is sodiated PC ho (16:0, 18:1). The asterisks in the top spectrum indicated potassiated adducts, which are not detected in the fixed section. It is believed that postassium is removed during the fixation procedure.



Lipid Geographical Analysis of the Primate Macula by Imaging Mass

3.5.2. Image Processing and Interpretation

255

Data collected in a discrete manner is plotted as intensity vs. position. Developing of images is dependent on the instrument manufacturer software. For the most part, data at each individual point (x, y coordinates) is divided by the total ion current at that specific point, multiplied by 10,000, and then is plotted vs. position. Images prepared in this way normalize for changes in the overall signal that may be a result of an uneven matrix coating. Images are collected for ion signals detected across the tissue. On the instrument employed here, we use a data extraction program developed by ThermoScientific and then plot the images using Surfer 8.0 (Golden Software, Golden, CO). For data extraction, a window of 1 amu (± 0.5 amu) around the ion of interest is selected. The data extraction program exports the x and y coordinates and the intensity at each point as a text file. The text file is then read by a graphical spreadsheet program and normalized to data extracted for the total ion current. Figure 5 shows the distribution of three PCho ions, lysoPC ho 16:0, PCho (18:0, 18:1), and PCho (18:0, 22:6) from left to right, from one aged rhesus monkey (63 years old, human equivalent) flat-mounted eye. The optic nerve head (ONH) was removed and is indicated by the arrow, while the cut-away allows for identification of the flap containing the macula. As indicated in the figure, lyso PCho 16:0, left, shows a more uniform distribution in the tissue with a macular bias, while both PCho (18:0, 18:1), middle, and PCho (18:0, 22:6), right, show clumped distributions, which may indicate a predisposition for drusen. The specificity of imaging with mass spectrometry allows for the differentiation of these individual PCho species and the geographic localization within the tissue. A key comparison for identifying lipids or other potential compounds in drusen rich areas is the comparison to traditional

m /z 518, Lyso PCho 16:0

m/z 804, PCho (16:0 20:4)

m/ z 856, PCho (18:0, 22:6) 2000

1000

0

Cut away

ONH

Fig. 5. MS images of three ions from a rhesus monkey flat-mount. The ions shown are three different phosphatidycholine (PCho) species as indicated above each figure that were detected in the tissue. The cut-away indicates the flap containing the macula and is cut out during sample preparation. ONH indicates optic nerve head.

256

Garrett and Dawson

optical methods for drusen identification. A closer look at the flap containing the macula in comparison to a fundus images is shown in Fig. 6. The fundus image is in panel A, while the MS image for m/z 782.5 is in panel B. The small white arrows in panel A indicate regions containing clinically visible drusen, while the black areas identify the fovea, F, and ONH for additional orientation. A comparison of mass spectra from two regions of the tissue is shown at the right of the figure, where the top spectrum is from the fovea region and the bottom is from a drusen-rich region. As can be seen from the mass spectra, there are multiple ion signals detected in the drusen region that are not detected or are less abundant in the fovea, such as PCho (18:0, 22:6) at m/z 856.5 and PC (18:0, 18:1) at m/z 810.5. In addition, cholesterol is detected in the tissue at m/z 369.9 (top) and 369.4 (bottom) showing a relatively similar intensity in both regions. 3.5.3. Lipid Identification

Correct identification is a key component in mass spectrometric measurement. On the instrumentation used here, we employ tandem mass spectrometry for identification. In this approach, a precursor ion is isolated in the linear ion trap with a specified isolation window, typically 1.0 or 1.5 amu, and then subjected to collision-induced dissociation (CID), which causes the ion to gain energy until it breaks apart into product ions (termed MS/MS or MS2). Using an ion trap, further identification can be performed by selecting a specific product ion and perform CID again, termed MS3. Additionally, further stages can be performed depending on the ion of interest. An example of how PCho species are identified using tandem mass spectrometry is shown in Fig. 7 for the identification of PCho (16:0, 18:1) detected at m/z 782.6. Key indicators of a PCho species are the loss of 59, detected at m/z 723.4 and the losses of 183 and 205 at m/z 599.4 and 577.5 respectively. Two factors allow for the identification of this species as a sodiated PCho. First is the loss of 59, which only occurs when a PCho species is ionized by cation adduction, (15, 27, 29) and second is the difference between m/z 599.4 and 577.5. The difference between these two product ions is 22, which indicates sodium adduction. The ion at m/z 526.3 is the loss of palmitic acid, 16:0 ([M + Na−C16H31O2]+). The ions at m/z 500.3 and 478.4 correspond to losses of oleic acid, 18:1 ([M + Na−C18H34O2]+) and the sodium salt of oleate ([M-−C18H33O2Na]+), respectively. The joint losses of trimethylamine with palmitic acid (m/z 467.3) or oleic acid (m/z 441.3) are also identified. Correlating to previous studies of lithiated and sodiated PChos, the loss of the sn-1 fatty acid chain occurs more readily than the loss of the sn-2 fatty acyl chain (26). Therefore, the intensity of the loss of palmitic acid (m/z 526.3) is higher than the intensity of the loss of oleic acid (m/z 500.3) which indicates that palmitic acid is at the sn-1 position, while oleic acid is at sn-2

F

ONH B

300

0 300

400

391.8

450

449.9 500

496.9

350

400

450

600

616.7 650

500

551.0

550

518.4

700

600

650 m/z

700

750

800

900

950

1000

+

PCho (16:0, 18:1) [M+Na]

850

903.0

902.8

900

856.5 850

810.5

800

+

+

950

1000

958.2

PCho (18:0, 22:6) [M+Na]

PCho (18:0, 18:1) [M+Na]

782.5

726.8 760.6 723.4 672.4

+

+

750

704.5 727.2 760.9 782.8 810.7 672.8

PCho (16:0, 18:1) [M+H]

550

551.4

Lyso PCho 16:0, [M+Na]

375.0 352.9 397.0 449.4

330.7

369.4 Cholesterol

350

291.2 353.3

369.9

331.2

20 290.3

40

60

80

100

0

20

40

60

80

100

Fig. 6. A-fundus image of rhesus monkey. B-MS image of m/z 782 showing only the area containing the macula. The white arrows in A indicate drusen identified around the fovea, F. The MS image seems to show more clumping than indicated in A. A mass spectrum from F is shown at the top right and a mass spectrum from an apparent drusen clumping area is shown at the bottom right (black lines in B indicate the areas from which the spectra were taken). More lipid signals were identified in the drusen region, than in the fovea.

A

Relative Abundance

Lipid Geographical Analysis of the Primate Macula by Imaging Mass 257

258

Garrett and Dawson x50

100

723.4

CH3

H 3C

N

+

CH3

90 O

80

m/z 526.3

O

Na

O

+

3

O

70

2 O

Relative Abundance

m/z 500.3

O P

-

1

O O

60 50 40

526.3

30 20 10

H 3C

500.3 CH3

264.3

478.4 467.3

599.4 577.5

392.3

0 300

400

500

600

700

m/z

Fig. 7. Averaged MS/MS spectrum of m/z 782.6 showing the identification of this ion as [PCho (16:0, 18:1) + Na]+. The structure of this PCho species is shown in the inset. Note that the intensity from m/z 200 to m/z 550 is magnified ×50. NL denotes neutral loss.

and m/z 782.6 can be thus be classified as [PCho (16:0, 18:1) + Na] + rather than [PCho (18:1, 16:0) + Na] + , and more specific than [PCho 34:1 + Na] +. Additional products ions are also detected, but may be the result of isorbaric (compounds having the same nominal mass, but different structure) species, which are often detected in the low mass region in IMS.(15, 24).

4. Notes 1. The mass spectrometer employed is a linear ion trap fitted with a matrix-assisted laser desorption/ionization (MALDI) source operating at a vacuum pressure of 0.17 Torr (Finnigan LTQ with vMALDI from ThermoFisher San Jose, CA).



Lipid Geographical Analysis of the Primate Macula by Imaging Mass

259

This instrument has been previously described in detail (24). Briefly, the instrument consists of a nitrogen laser (337 nm) directed to the source with fiber optics and mirrors. The incident angle of the laser to the sample plate is 32° and the spot size is adjusted between 80 and 120 mm. 2. Preparation of the retina as a “whole mount” or “flat mount” (morning glory) (30) style for IMS is similar, in the early stages, to the lipid-friendly procedures used for frozen sections (3). This process assumes the eye has already been removed from the orbit. 3. Nitrocellulose was found to be the most reliable surface for use in preventing the tissue from falling off the sample plate when dried and inserted into the instrument. 4. We found it critical to ensure that the tissue was still slightly moist before matrix application to ensure adherence to the nitrocellulose during the coating process.

Acknowledgments The authors thank the Lion’s Eye Bank of Tampa, FL for donation of human tissue specimens. The Lion’s Club International Foundation of Jacksonville, FL is greatly acknowledged for funding. References 1. Abdelsalam, A., Del Priore, L., and Zarbin, M. A. (1999) Drusen in age-related macular degeneration: pathogenesis, matural course and laser photocoagulation induced regression, Survey Ophthal. 44, 1–29. 2. Donoso, L. A., Kim, D., Frost, A., Callahan, A., and Hageman, G. (2006) The role of inflammation in this pathogenesis of age-related macular disease, Survey of Ophthal. 51, 137–152. 3. Anderson, M., Dawson, W. W., GonzalezMartinez, J., and Curcio, C. A. (2006) Drusen and lipid-filled retinal pigment epithelium cells in a rhesus macula, Vet. Opthal. 9, 201–207. 4. Curcio, C. A., Presley, J. B., Millican, C. L., and Medeiros, N. E. (2005) Basal deposits and drusen in eyes with age-related maculopathy: evidence for solid lipid particles, Exp. Eye Res. 80, 761–775. 5. Garrett, T. J., Dawson, W., and Yost, R. A. (2007) ASMS Conference on Mass Spectrometry and Allied Topics, Indianapolis, IN.

6. Caprioli, R. M., Farmer, T. B., and Gile, J. (1997) Molecular imaging of biological samples: Localization of peptides and proteins using MALDI-TOF MS, Anal. Chem. 69, 4751–4760. 7. Chaurand, P., and Caprioli, R. M. (2002) Direct profiling and imaging of peptides and proteins from mammalian cells and tissue sections by mass spectrometry, Electrophoresis 23, 3125–3135. 8. Chaurand, P., Schwartz, S. A., and Caprioli, R. M. (2004) Assessing protein patterns in disease using imaging mass spectrometry, J. Proteome Res. 3, 245–252. 9. McDonnell, L. A., Piersma, S. R., Altelaar, A. F. M., Mize, T. H., Luxembourg, S. L., Verhaert, P. D. E. M., van Minnen, J., and Heeren, R. M. A. (2005) Subcellular imaging mass spectrometry of brain tissue, J. Mass Spectrom. 40, 160–168. 10. Monroe, E. B., Jurchen, J. C., Lee, J., Rubakhin, S. S., and Sweedler, J. V. (2005)

260

Garrett and Dawson

Vitamin E imaging and localization in the neuronal membrane, J. Am. Chem. Soc. 127, 12152–12153. 11. Chaurand, P., Schwartz, S. A., Billheimer, D., Xu, B. J., Crecelius, A., and Caprioli, R. M. (2004) Integrating histology and imaging mass spectrometry, Anal. Chem. 76, 1145–1155. 12. Maddalo, G., Petrucci, F., Iezzi, M., Pannellini, T., Del Boccio, P., Ciavardelli, D., Biroccio, A., Forli, F., Di Ilio, C., Ballone, E., Urbani, A., and Federici, G. (2005) Analytical assessment of MALDI-TOF imaging mass spectrometry on this histological samples. An insight into proteome investigation, Clin. Chim. Acta 357, 210–218. 13. Reyzer, M. L., Hsieh, Y., Ng, K., Korfmacher, W. A., and Caprioli, R. M. (2003) Direct analysis of drug candidates in tissue by matrixassisted laser desorption/ionization mass spectrometry, J. Mass Spectrom. 38, 1081–1092. 14. Jackson, S. N., Wang, H.-Y. J., and Woods, A. S. (2005) In situ structural characterization of phosphatidylcholines in brain tissue using MALDI-MS/MS, J. Am. Soc. Mass Spectrom. 16, 2052–2056. 15. Garrett, T. J., and Yost, R. A. (2009, submitted to Methods in Molecular Biology) Characterization of phospholipids in rat brain tissue by imaging tandem mass spectrometry with intermediate-pressure MALDI on a linear ion trap. 16. Drexler, D. M., Garrett, T. J., Cantone, J. L., Diters, R. W., Mitroka, J. G., Prieto-Conaway, M. C., Adams, S. P., Yost, R. A., and Sanders, M. (2007) Utility of imaging mass spectrometry (IMS) by matrix-assisted laser desoption ionization (MALDI) on an ion trap mass spectrometer in the analysis of drugs and metabolites in biological tissues, J. Pharm. and Toxicol. Methods 55, 279–288. 17. Stoeckli, M., Chaurand, P., Hallahn, D. E., and Caprioli, R. M. (2001) Imaging mass spectrometry: A new technology for the analysis of protein expression in mammalian tissues, Nature Med. 7, 493–496. 18. Winograd, N. (2003) Prospects for imaging TOF-SIMS: from fundamentals to biotechnology, Appl Surface Sci. 203–204, 13–19. 19. Todd, P. J., Schaaf, T. G., Chaurand, P., and Caprioli, R. M. (2001) Organic ion imaging of biological tissue with secondary ion mass spectrometry and matrix-assisted laser desorption/ ionization, J. Mass Spectrom. 36, 355–369.

20. Todd, P. J., Schaaf, T. G., Chaurand, P., and Caprioli, R. M. (2001) Organic ion imaging of biolocial tissue with secondary ion mass spectrometry and matrix-assisted laser desorption/ionization, J. Mass Spectrom. 36, 355–369. 21. Garrett, T. J. (2006) Chemistry, pp. 148, University of Florida, Gainesville, FL. 22. Garrett, T. J., Kovtoun, V. V., Buy, H., Izgarian, N., Prieto-Conaway, M. C., Stafford, G., and Yost, R. A. (2007) Imaging mass spectrometry with a new intermediate-pressure MALDI linear ion trap mass spectrometer, Int. J. Mass Spectrom. 263, 166–176. 23. Cha, S., and Yeung, E. S. (2007) Colloidal graphite-assisted laser desorption/ionization mass spectrometry and MSn of small molecules. 1. Imaging of cerebrosides directly from rat brain tissue, Anal. Chem. 79, 2373–2385. 24. Garrett, T. J., Prieto-Conaway, M. C., Kovtoun, V., Bui, H., Izgarian, N., Stafford, G. C., and Yost, R. A. (2007) Imaging of small molecules in tissue sections with a new intermediate-pressure MALDI linear ion trap mass spectrometer, Int. J. Mass Spectrom. 260, 166–176. 25. Jackson, S. N., Wang, H.-Y. J., and Woods, A. S. (2005) Direct profiling of lipid distribution in brain tissue using MALDI-TOFMS, Anal. Chem. 77, 4523–4527. 26. Han, X., and Gross, R. W. (1995) Structural determination of picomole amounts of phospholipids via electrospray ionization tandem mass spectrometry, J. Am. Soc. Mass Spectrom. 6, 1201–1210. 27. Hsu, F.-F., Bohrer, A., and Turk, J. (1997) Formation of lithiated adducts of glycerophosphocholine lipids facilitates their identification by electrospray ionization tandem mass spectrometry, J. Am. Soc. Mass Spectrom. 9, 516–526. 28. Dawson, W. W., Dawson, J. L. K., and Gonzalez-Martinez, J. (2008) Maculas, monkeys, models, AMD, and aging, Vis. Res. 48, 360–365.9 29. Ho, Y.-P., Huang, P.-C., and Deng, K.-H. (2003) Metal ion complexes in the structural analysis of phospholipids by electrospray ionization tandem mass spectrometry, Rapid Commun. Mass Spectrom. 17, 114–121. 30. Stone, J. (1981) The whole mount handbook, Maitland Publications, Sydney.

Chapter 13 A Novel Role for Nutrition in the Alteration of Functional Microdomains on the Cell Surface Wooki Kim, Robert S. Chapkin, Rola Barhoumi, and David W.L. Ma Summary Membrane rafts are ordered microdomains of the plasma membrane consisting of cholesterol, sphingolipids, and saturated fatty acids which appear to regulate many cellular signaling pathways. One such type of membrane raft is caveolae, which are cave-like invaginations of the plasma membrane. Interestingly, changes in the acyl composition of cellular membranes have been shown to alter the specific localization of membrane raft associated proteins and their function. This is noteworthy because modification of membrane acyl composition is readily accomplished through changes in dietary fat composition. Here we describe a common approach used to fractionate cell membranes to obtain an enriched preparation of caveolae and gas chromatographic techniques to determine fatty acyl composition. In addition, methods used to visualize and quantify lipid rafts using a fluorescent probe Laurdan in living cells will also be described. Key words: Caveolae, Fatty acids, Laurdan, Lipid rafts, Membrane rafts, Nutrition

1. Introduction Alteration of cell membrane structure, specifically the acyl fatty acids of phospholipids, is readily achieved through simple changes in dietary fat composition. This alteration is very rapid and can be readily detected within hours in cell culture or days in humans. The influence and impact of dietary fat composition on membrane structure has gained increasing interest due to advancements in techniques and refinements in our understanding of membrane rafts (1). Membrane rafts include structures known as caveolae, which are cave-like structures enriched in cholesterol, sphingolipids, and phospholipids containing saturated fatty acids. These membrane rafts are now implicated in a diversity of biological

Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, DOI 10.1007/978-1-60761-322-0_13, © Humana Press, a part of Springer Science + Business Media, LLC 2009

261

262

Kim et al.

Total Cell Lysate spin

Sonicate, Percoll Plasma OptiPrep Post-Nuclear Supernatant spin Membrane spin x2

Caveolae

Caveolae

Plasma Membrane Supernatant

Fig. 1. Shown in this figure are the bands obtained during the caveolae isolation procedure from cultured cells.

functions which impact on human health and disease (2–4). Studies have shown that alterations in dietary fat composition to include omega-3 fatty acids from marine oils can impact on the structure and function of colonocyte, immune, and cancer cell membranes (2, 5–7). Collectively, this body of work demonstrates that simple nutritional interventions can have significant impact on cell signaling in part mediated through membrane rafts. Concentrated caveolae can be isolated using a nondetergent approach following the methodology developed by Smart et al. (2, 8, 9). This protocol works well with isolated cells from cell culture and single cell populations from tissue. The procedure below describes the isolation of caveolae from cell culture. Depending upon the cell type and expression of caveolae, preliminary work is required to determine the optimum number of cells required to obtain sufficient caveolae. Start with 100 × 106 cells, which may be the equivalent of several T-175 flasks. An overview of the procedure and resultant fractions are shown in Fig. 1.

2. Materials 2.1. Equipment

1. Ultracentrifuge at 4°C. 2. Benchtop centrifuge. 3. Water bath at 37°C. 4. Gradient maker. 5. Stirrer plate.

2.2. Reagents

1. Optiprep (60% v/v solution of iodixanol). 2. Percoll.



A Novel Role for Nutrition in the Alteration of Functional Microdomains

263

3. Sucrose. 4. Na4EDTA (monohydrate, 380.2 MW). 5. Tricine.

3. Methods 3 .1. Fatty Acid Treatment Protocol

1. Adherent cells are first grown for 2 days in basal medium to acclimate cells (see Note 1). 2. On day 3, media are replaced with media containing supplemental fatty acids. 3. Stock nonesterified fatty acids (Nu-Chek Prep, Elysian Falls, MN, USA) are conveniently stored in ethanol at 1–10 mg/mL. 4. Prepare media containing 0–100 mM fatty acids. Note: concentrations exceeding 200 mM are toxic. Stock fatty acids are transferred to a 15-mL sterile Falcon tube and ethanol is evaporated under a gentle stream of nitrogen. 5. Fatty acids are bound to bovine serum albumin (fraction V) (0.1%w/v) by incubating in fetal bovine serum and basal medium in a shaking water bath at 37°C for 30 min. 6. Media is replaced. 7. Cells are grown until ~90% confluency. Media should be replaced every 3 days.

3.2. Caveolae Isolation 3 .2.1. Solutions to be Prepared in Advance

1× PBS. 500 mL warm (37°C), 500 mL cold (4°C) Buffer A: (stable for 1 week stored at 4°C)

Final Conc:

20 mL 2.5 M Sucrose

0.25 M

0.076 g Na4EDTA (monohydrate, 380.2 MW)

1 mM

0.716 g tricine

20 mM

Combine with 170 mL water, pH to 7.8, bring to final volume of 200 mL

2XA Buffer: (stable for 1 week stored at 4°C) 13.32 mL 2.5 M Sucrose

0.5 M

0.05 g Na4EDTA

2 mM

0.48 g tricine

40 mM

Combine with 40 mL of water, pH to 7.8, bring to final volume of 66.6 mL

264

Kim et al.

OptiPrep Diluent: (stable for 1 week stored at 4°C) 1 mL 2.5 M Sucrose

0.25 M

0.022 g Na4EDTA

6 mM

0.216 g tricine

120 mM

Combine with 7 mL water, pH to 7.8, bring to final volume of 10 mL

3.2.2. Solutions to be Prepared on Day of Experiment

30% Percoll: 30 mL Percoll 50 mL 2XA 20 mL water 50% OptiPrep: 36 mL OptiPrep 7.2 mL OptiPrep diluent OptiPrep (60% v/v solution of iodixanol) Solutions: 5% – 0.4mL 50% OptiPrep + 3.6mL Buffer A 10% – 2.94mL 50% OptiPrep + 11.74mL Buffer A 15% – 1.5mL 50% OptiPrep + 3.5mL Buffer A 20% – 5.86mL 50 OptiPrep + 8.8mL Buffer A 1. Harvest cells at 90% confluency then centrifuge at 100 × g for 3 min. After isolation, keep cells at 4°C or on ice. 2. Resuspend cells in 50 mL of cold 1× PBS and centrifuge at 100 × g for 3 min. 3. Discard supernatant and resuspend pellet in 30 mL of Buffer A and centrifuge at 100 × g for 3 min. 4. Discard supernatant and suspend pellet in 2 mL of Buffer A with protease inhibitor cocktail (Sigma) (40 mL/mL) and phosphatase inhibitor (orthovanadate) (10 mL/mL). 5. Homogenize 20 strokes on ice using Teflon /glass dounce homogenizer. Save aliquot (50 mL). 6. Spin at 1,000 × g, 10 min, 4°C. Meanwhile precool ultracentrifuge and make 30% Percoll. 7. Collect supernatant (postnuclear) and keep on ice. 8. Resuspend pellet in 1.5 mL of Buffer A, therefore final volume is 3.5 mL. Combine supernatants, save an aliquot (50 mL).



A Novel Role for Nutrition in the Alteration of Functional Microdomains

265

9. The combined PNS (3.5 mL) was overlaid on 9 mL of 34.86% (v/v) Percoll solution in 13.2 mL Ultra-Clear Beckman Centrifuge Tubes. The 34.86% Percoll solution is made by diluting Percoll with 46.53% (v/v) 2XA Buffer. 10. Centrifuge in Beckman SW41 Ti rotor at 84,000 × g for 30 min. Meanwhile make remaining OptiPrep Solutions 11. Collect 2 mL (up to 3 mL) of plasma membrane fraction (visible band 6 cm from bottom of tube). Place in 15 mL Beckman tubes. 12. The PM fraction is then sonicated on ice, three times for 2 min each, with 2 min intervals between each sonication (see Note 2) 13. Pour Opti gradient by adding 2.3 mL of 50% OptiPrep to the samples and adjusting the final volume to 5.0 mL using Buffer A as a diluent. 14. A linear 20–10% OptiPrep gradient (3 mL each) is overlaid on the new 23% OptiPrep solution using a gradient maker connected to a peristaltic pump set at a flow rate of 1 mL/min 15. Centrifuge at 52,000 × g (17,000 rpm) 90 min, 4°C in SW41 Ti rotar 16. Collect the top 5 mL into new tube, then prepare ~23% OptiPrep solution by adding 4 mL of 50% OptiPrep to sample and mix. 17. Overlay with a discontinuous gradient with 1 mL of 15% OptiPrep followed by 0.5 mL of 5% OptiPrep 18. Centrifuge at 52,000 × g (17,000 rpm) 90 min, 4°C. 19. Collect top white band (500–700 mL) within 5% OptiPrep layer, just above interface. 3.3. Fast GC Analysis

1. 100 mL aliquots of caveolae are transferred to a 15 mL glass screw cap tube with teflon cap. 2. Lipids are extracted in chloroform:methanol (2:1), 4 mL and 1 mL of KCl (0.88%w/v), vortex. 3. Centrifuge at 200 × g to separate phases. 4. Carefully transfer bottom chloroform fraction to a clean glass tube, with Teflon cap. 5. Add 2 mL of chloroform to top phase and repeat extraction. 6. Combine chloroform extracts and dry down under a gentle stream of nitrogen. 7. Saponify lipids by adding 2 mL of KOH–MeOH, then incubate at 100°C (water bath or oven) for 1 h.

266

Kim et al.

8. Cool for 5 min in fume hood, then add 2 mL of 14% BF3– MeOH and 2 mL of hexane, then incubate at 100°C (water bath or oven) for 1 h. 9. Cool for 5 min in fume hood and then add 2 mL of double distilled water. 10. Centrifuge at 200 × g to separate phases. 11. Transfer upper hexane phase to clean 2 mL gas chromatography vial. 12. Dry down hexane under gentle stream of nitrogen. 13. FAME are quantified on an Agilent 7890 flame-ionization detector gas chromatograph (Palo Alto, CA, USA) separated on a DB-FFAP fused silica column, 15 m × 0.10mmID × 0.10 mm (Agilent). Samples were injected in split mode, 200:1. The injector and detector ports should be set at 250°C. FAME are eluted using a temperature program set initially at 150°C for 0.25 min, increased at 35°C/min and held at 170°C for 3 min, increased at 9°C/min and held at 225°C for 0.5 min, increased at 80°C and held at 245°C for 2.2 min. Total run time is 12.9 min. The carrier gas, hydrogen, is set to a 0.5 mL/min constant flow rate. The makeup gas is nitrogen. Fatty acids are identified using authentic standards obtained from Nu-Chek Prep (#GLC-463) (see Note 3) (Fig. 2) 3.4. Protocol for Visualization of Lipid Rafts Using the Fluorescence Probe Laurdan

1. Dilute 0.1% (w/v) poly-l-lysine solution (Sigma) to 0.01% with sterile water.

3.4.1. Poly-l-lysine Precoating on Coverglass

4. Aspirate excess solution and allow the coverglass wells to air dry.

2. Fill the chambered coverglass (Nunc) wells with 0.01% polyl-lysine solution (1 mL). 3. Allow solution to sit for 30 min.

5. Sterilize the coverglass slides under UV light for 1 h. 3.4.2. Preparation of 10 mM Laurdan (Invitrogen, MW 353.55) in Ethanol

1. Measure 10 mg of Laurdan in a glass tube covered with aluminum foil (Laurdan is light sensitive). 2. Dissolve in 2.83 mL of ethanol to make 10 mM of Laurdan stock solution. 3. Vortex vigorously (~2 min) to dissolve Laurdan. 4. Keep at 4°C and prewarm at RT prior to use.

3.4.3. Cell Labeling of Laurdan

1. Cell concentration might differ by type of cells. 2. Start with 1 × 106 cells/mL, which has been shown to be optimal for CD4+ T cells (10). 3. Transfer 2 mL of cell suspension into a 15 mL-conical tube.

A Novel Role for Nutrition in the Alteration of Functional Microdomains

Fig. 2. Shown in this figure is a representative fast gas chromatographic separation of fatty acid methyl esters.

267

268

Kim et al.

4. Add 2 mL of (10 mM) Laurdan solution into 2 mL of cell suspension to dilute to 10 nM of Laurdan (see Note 4). 5. Vortex for 2 s. 6. Incubate cells at 37°C for 10 min. 7. Centrifuge cell suspension at 300× gravity for 5 min. 8. Resuspend cell pellets in 2 mL of Leibovitz’s medium. 9. Transfer 1 mL of cell suspension into a poly-l-lysineprecoated chambered glass. 10. Incubate cells in an appropriate condition, if necessary. 3.4.4. GP-Value Calculation

1. Capture Laurdan-labeled cell images using 2-photon microscopy. (Zeiss LSM 510 META NLO, 40× objective 1.3 NA oil. The coherent Chameleon femtosecond pulsed Ti:Sapphire laser should be set at an excitation wavelength of 770 nm) 2. Save the image files as 8-bit TIFF format. 3. Open the image file in an image analysis software package, such as Adobe Photoshop® CS (v.8.0). (The following protocol is using Adobe Photoshop®) 4. Using a marquee tool, make an oval to select a region of interest (Fig. 3b, c) 5. In “Histogram” panel, choose each channel of RGB. 6. Once blue channel is selected, the panel displays “mean intensity” of marquee-selected region. 7. A histogram will also show the pixels chosen by marquee tool. 8. If you change the channel to green, mean intensity is changed, while pixel is not. 9. Record green and blue intensities to calculate GP-values by formula GP = (IBlue – IGreen )/(IBlue + IGreen).

3.5. Results 3.5.1. Visualization of Lipid Rafts

There have been several methodologies developed to identify and visualize lipid rafts. For instance, flotation of DRM fractions and immunofluorescence microscopy by antibody patching have been used to identify putative raft association and raft proteins (11, 12). In addition, immunoelectron microscopy and single fluorophore tracking microscopy have been utilized to track the location of raft components (13). Fluorescence resonance energy transfer (FRET) has also been applied to determine whether two raft components are spatially impacted (14). However, due to the specificity of probes used, the visualization of heterogeneous lipid rafts have been difficult to determine. Recently, Gaus et  al. demonstrated that the fluorescent probe Laurdan can align itself parallel with the hydrophobic tails



A Novel Role for Nutrition in the Alteration of Functional Microdomains

a

RGB CD4+ T-cell

b

c

Blue

ROI

269

Green

ROI

Hybridoma Fig. 3. Representative (a) RGB images of Laurdan-labeled CD4+ T cells. Purified CD4+ T cells were labeled with 10 mM Laurdan and stimulated by anti-CD3 mAb expressing hybridoma cells at 37°C for 30 min. GP-values were calculated from values of fluorescence intensity identified using Adobe Photoshop® to draw regions of interest (ROI) in each (b) blue or (c) green channel at the immunological synapse.

of phospholipids in membranes (15–17). Owing to its ability to emit two different wavelengths according to the fluidity of the microenvironment, Laurdan can be used to measure and visualize membrane fluidity in living cells by two-photon microscopy (16). Generalized polarized values (GP-values) derived by the formula GP = (I400 – 460 – I470 – 530)/(I400 – 460 + I470 – 530)), where I stands for intensities of the blue and green channels, respectively, indicates the fluidity of cell membranes. Given that lipid rafts are highly enriched in cholesterol and sphingolipids, the liquid ordered state of the membrane detected by Laurdan labeling is assumed to be heterogeneous lipid rafts. As an example, representative images of Laurdan-labeled CD4+ T cells co-cultured with anti-CD3 mAb expressing hybridoma cells are presented in Fig. 3a.

4. Notes 1. Isolation of caveolae is best accomplished using a single cell type from cell culture. In addition, this method is also suitable for purified cells isolated from tissue. Caveolae are found in most cell types, however their function is also cell specific. Therefore, if whole tissue is used, it must be recognized that results represent an aggregate from different cell types. 2. The sonication step may introduce variability in the procedure dependent on the type of sonicator, power, and duration of cell disruption. Furthermore, interindividual variability contributes to differences in overall yield of caveolae.

270

Kim et al.

3. Although a fast-GC analysis approach is described, any fatty acid methyl ester method will be suitable. 4. Laurdan concentration and duration of labeling should be optimized for each cell type used.

Acknowledgments Supported in part by NIH grants CA59034, CA129444, DK071707, and P30ES09106 to R.S. Chapkin and a Natural Sciences and Engineering Research Council of Canada Discovery Grant to D.W.L. Ma. References 1. Pike, L. J. (2006) Rafts defined: a report on the Keystone symposium on lipid rafts and cell function. J. Lipid Res. 47: 1597–1598. 2. Ma, D. W., Seo, J., Davidson, L. A., Callaway, E. S., Fan, Y. Y., Lupton, J. R. & Chapkin, R. S. (2004) n-3 PUFA alter caveolae lipid composition and resident protein localization in mouse colon. FASEB J. 18: 1040–1042. 3. Ma, D. W. (2007) Lipid mediators in membrane rafts are important determinants of human health and disease. Appl. Physiol Nutr. Metab 32: 341–350. 4. Ma, D. W. L., Seo, J., Switzer, K. C., Fan, Y. Y., McMurray, D. N., Lupton, J. R. & Chapkin, R. S. (2004) n-3 PUFA and Membrane Microdomains: A New Frontier in Bioactive Lipid Research. J. Nutr. Biochem. 15: 700–706. 5. Fan, Y. Y., McMurray, D. N., Ly, L. H. & Chapkin, R. S. (2003) Dietary (n-3) polyunsaturated fatty acids remodel mouse T-cell lipid rafts. J. Nutr. 133: 1913–1920. 6. Fan, Y. Y., Ly, L. H., Barhoumi, R., McMurray, D. N. & Chapkin, R. S. (2004) Dietary docosahexaenoic acid suppresses T cell protein kinase C theta lipid raft recruitment and IL-2 production. J. Immunol. 173: 6151–6160. 7. Schley, P. D., Brindley, D. N. & Field, C. J. (2007) (n-3) PUFA alter raft lipid composition and decrease epidermal growth factor receptor levels in lipid rafts of human breast cancer cells. J. Nutr. 137: 548–553. 8. Michaely, P. A., Mineo, C., Ying, Y. S. & Anderson, R. G. (1999) Polarized distribution of endogenous Rac1 and RhoA at the cell surface. J. Biol. Chem. 274: 21430–21436. 9. Smart, E. J., Ying, Y. S., Mineo, C. & Anderson, R. G. (1995) A detergent-free method for purifying caveolae membrane from tissue

culture cells. Proc. Natl. Acad. Sci. U S A 92: 10104–10108. 10. Kim, W., Fan, Y. Y., Barhoumi, R., Smith, R., McMurray, D. N.,& Chapkin, R. S. (2008). n-3 polyunsaturated fatty acids suppress the localization and activation of signaling proteins at the immunological synapse in murine CD4+ T cells by affecting lipid raft formation. J. Immunol. 181: 6236–6243. 11. Harder, T., Scheiffele, P., Verkade, P. & Simons, K. (1998). Lipid domain structure of the plasma membrane revealed by patching of membrane components. J. Cell Biol. 141: 929–942. 12. Anderson, R. G. (1998). The caveolae membrane system. Annu. Rev. Biochem. 67: 199–225. 13. Wilson, B. S., Pfeiffer, J. R. & Oliver, J. M.. (2000). Observing FcepsilonRI signaling from the inside of the mast cell membrane. J. Cell Biol. 149: 1131–1142. 14. Kenworthy, A. K., Petranova, N. & Edidin M. (2000). High-resolution FRET microscopy of cholera toxin B-subunit and GPI-anchored proteins in cell plasma membranes. Mol. Biol. Cell 11: 1645–1655. 15. Gaus, K., Chklovskaia, E., Fazekas de St Groth, B., Jessup, W.,Harder, T.. (2005). Condensation of the plasma membrane at the site of T lymphocyte activation. J. Cell Biol. 171: 121–131. 16. Gaus, K., Zech, T.,& Harder, T. (2006). Visualizing membrane microdomains by Laurdan 2-photon microscopy. Mol. Membr. Biol. 23: 41–48. 17. Rentero, C., Zech, T., Quinn, C. M., Engelhardt, K., Williamson, D., Grewal, T., Jessup, W., Harder, T. & Gaus, K.. (2008). Functional implications of plasma membrane condensation for T cell activation. PLoS ONE 3: e2262.

Chapter 14 Lipidomic Analysis of Prostanoids by Liquid Chromatography–Electrospray Tandem Mass Spectrometry Anna Nicolaou, Mojgan Masoodi, and Adnan Mir Summary Lipidomics aim to generate qualitative and quantitative information on different classes of lipids and their species, and when applied in conjunction with proteomic and genomic assays, facilitate the comprehensive study of lipid metabolism in cellular, organ, or body systems. Advances in mass spectrometry have underpinned the expansion of lipidomic methodologies. Prostanoids are potent autacoids present in a plethora of cellular systems, known best for their intimate role in inflammation. Electrospray ionisation (ESI) allows the efficient ionisation of prostanoids in aqueous systems. ESI can be readily coupled to liquid chromatography (LC) followed by tandem mass spectrometry (MS/MS)-based detection, thus allowing the development of a potent and selective LC/ESI–MS/MS quantitative assays. The protocol we describe in this chapter outlines the steps we follow to (a) extract prostanoids from solid or liquid samples, (b) semi-purify the metabolites using solid phase extraction (c) set-up the HPLC separation using reverse phase chromatography and (d) set-up the MS/MS assay using a triple quadrupole mass spectrometer. The experimental details and notes presented here are based on the detailed protocols followed in our group. Key words:  Prostaglandins, Thromboxanes, Dihydro-prostaglandins, Electrospray ­ionisation, Tandem mass spectrometry, HPLC, Lipidomics

1. Introduction Prostanoids are a subclass of eicosanoids, a family of potent bioactive lipid mediators originating from polyunsaturated fatty acids (PUFA). The term is used to collectively describe prostaglandins, thromboxanes, and prostacyclins derived by dihomo-gamma linoleic (20:3, n-6) (DHGLA), arachidonic (20:4, n-6) (AA) and eicosapentaenoic (20:5, n-3) (EPA) acids (1). The precursor PUFA are liberated from membrane phospholipids by phospholipases A2 (2) and converted by cyclooxygenases (COX) Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_14, © Humana Press, a part of Springer Science + Business Media, LLC 2009

271

272

Nicolaou, Masoodi, and Mir

(3, 4) to an unstable endoperoxide indermediate (PGH) that is further metabolised to various prostanoids through the action of specific prostanoid synthases (5). The prevalence of those synthases results in organ/tissue/cell-specific prostanoid profiles. Prostanoids are unstable compounds that are rapidly metabolised in vivo and in vitro through enzymatic and non-enzymatic means, resulting in stable but inactive derivatives (6). Reliable and accurate analytical methodologies for the quantitative measurement of prostanoids and their derivatives in body fluids, tissue samples or cell culture media are important in evaluating the contribution of those potent bioactive lipids in physiological and pathophysiological processes, as well as to assess the bioactivities of their precursor omega-6 and omega-3 PUFA (7–10). Prostanoids can be measured by enzyme immunoassays (EIA) and radioimmuno-assays (RIA), liquid (LC) or gas chromatography (GC) coupled to fluorescence detection or mass spectrometry (MS), and liquid chromatography–tandem mass spectrometry (LC–MS/MS) (11–15). Although immunoassays are a very popular means of quantitating prostanoids, they can only be applied to the analysis of one mediator at a time and their application is limited to the relative few compounds that have commercially available antibodies. Simultaneous analyses of multiple prostanoids can be performed by GC-MS protocols although the need to form volatile derivatives and possible destruction of thermally labile compounds limits this application (16). LC–MS/MS approaches have overcome most of these technical limitations offering a powerful and versatile tool for the simultaneous qualitative and quantitative analysis of a large number of prostanoids (17–19). Lipidomic assays aim to generate qualitative and quantitative information on different classes of lipids and their species (17, 20–22). This approach allows the study of lipid profiles and, consequently, appreciation of the general picture of their metabolism in specific tissues, organ, or body systems. LC–MS/MS assays make use of mass spectrometers as extremely selective, sensitive, and definitive detectors that when coupled with the great analytical power and versatility of LC, yield multidimentional analyses which have enabled the development of lipidomics. The assay we describe in this chapter is based on the separation of prostanoids by reverse phase LC coupled to electrospray ionisation (ESI)–MS/MS detection using a triple quadrupole mass spectrometer (Fig. 1). ESI is the most widely applied ionisation method for the analysis of prostanoids which readily form negative molecular ion species ([M − H]−) (16). The ESI interface allows their ionisation directly as they elute from the LC system thus limiting the danger of thermal decomposition. Since the detection is based on the mass of the analyte of interest, the method allows the determination of isomeric compounds that are not resolved chromatographically. Furthermore, the application



Lipidomic Analysis of Prostanoids by Liquid Solid sample

273

Liquid sample

homogenize

15 % methanol pH 3

Prostanoid standards Calibration lines

C18 solid phase extraction

Semi-purified prostanoids Protein content

LC/ESI-MS/MS analysis Prostanoid (absolute quantitation)

Fig. 1. Schematic outline of the sample extraction and LC/ESI–MS/MS assay applied for the quantitative analysis of prostanoids.

of collision-induced dissociation (CID) permits the identification of a metabolite of interest to be made based not only on mass but also on structure: the multiple-reaction monitoring (MRM) mode of the MS allows characteristic precursor/product transitions to be followed during the course of the experiment. This setting allows the use of deuterated prostanoids as internal standards. Following the initial extraction, a semi-purification step based on solid phase extraction (SPE) serves to clean-up the biological extract and reduce matrix effects that may suppress the ionisation of the analytes of interest. Absolute quantitation is based on calibration lines constructed with commercially available standards. Overall, this lipidomic assay offers qualitative and quantitative analysis of 23 prostanoids in a single 20 min run (17). The assay is high throughput and we can typically analyse up to 60 samples in a single continuous run. The assay is versatile and can be tailored to the individual needs of many different projects in detecting metabolites from a variety of biological sources.

2. Materials 2.1. Equipment

1. Refrigerated centrifuge (500–4,000 x g). 2. 2 mL Dounce mini glass mini homogeniser with tight fitting pestle (Wheaton) (Fisher Scientific, Loughforough, UK).

274

Nicolaou, Masoodi, and Mir

3. SPE Vacuum Manifold (Phenomenex, Macclesfield, UK). 4. SPE Drying attachment for Vacuum Manifold (Phenomenex, Macclesfield, UK). 5. HPLC solvent filtering and degassing system (all glass funnel and reservoir). 6. Waters Alliance 2695 HPLC pump with auto sampler (Waters 2690) coupled to ESI triple quadrupole Quattro Ultima mass spectrometer (Waters, Elstree, UK). Instrument control and data acquisition were performed with MassLynxTM V4.0 software (Waters, Elstree, UK). 7. Vortex. 8. Ice box. 2.2. Reagents and Supplies 2.2.1. Prostanoid Solutions

1. The following standards were obtained from Cayman Chemicals (Ann Arbor, MI, USA): prostaglandin D1 (PGD1), prostaglandin E1 (PGE1), prostaglandin F1a (PGF1a), 6-keto-prostaglandin F1a (6-keto-PGF1a), prostaglandin B2 (PGB2), prostaglandin B2-d4 (PGB2-d4,) prostaglandin D2 (PGD2), prostaglandin E2 (PGE2), prostaglandin F2a (PGF2a), prostaglandin J2 (PGJ2), ▵D12-prostaglandin J2, (z12-PGJ2), 15-deoxy-D12,14prostaglandin J2 (15-deoxy-D12,14-PGJ2), prostaglandin D3 (PGD3), prostaglandin E3 (PGE3), prostaglandin F3a (PGF3a), thromboxane B2 (TXB2), thromboxane B3 (TXB3), 13,14-dihydro PGE1, 13,14-dihydro-15-keto PGE1, 13,14-dihydro-PGF1a, 13,14-dihydro-15-keto PGF1a, 13,14-dihydro- 15-keto PGE2, 13,14-dihydro-PGF2a, 13,14-dihydro-15-keto PGF2a. 2. Ethanol, HPLC grade (Sigma-Aldrich, Poole, UK). 3. 10, 50 and 250 mL glass syringes. 4. 1.5 mL amber glass vials with polypropylene caps. 5. Nitrogen gas cylinder. 6. Cryogenic storage boxes to hold 1.5 mL vials. 7. Sealing film (Parafilm®) (Fisher Scientific, Loughborough, UK).

2.2.2. Biological Sample Preparation

1. Long glass tubes (12 mL) with Teflon-lined caps. 2. Methanol, HPLC grade (Sigma-Aldrich, Poole, UK). 3. 0.1 M and 0.025 M Hydrochloric acid (Sigma-Aldrich, Poole, UK). 4. Narrow range (2.5–4.5) pH indicator paper strips (Merck, UK). 5. 2 mL glass Pasteur pipettes (long and short). 6. 20 mL Measuring cylinders. 7. 0.2 and 1.0 mL Adjustable pipettes.



Lipidomic Analysis of Prostanoids by Liquid

275

8. 1.5 mL Amber glass vials suitable for the HPLC autosampler, screw caps, septa, and 100 mL insert vials (Kinesis, Bedfordshire, UK). 2.2.3. Solid Phase Extraction

1. Methanol, n-hexane and ethanol (HPLC grade) and methyl formate (GC-grade) (Sigma-Aldrich, Poole, UK). 2. STRATA™ SPE cartridges (C18-E 500 mg, 6 mL) (Phenomenex, Macclesfield, UK). 3. SPE Manifold needles, stainless steel. 4. Long glass tubes (12 mL) with Teflon-lined caps. 5. Nitrogen cylinder. 6. 10, 50, and 250 mL glass syringes.

2.2.4. HPLC Solvents

1. Acetonitrile HPLC grade (Sigma-Aldrich, Poole, UK). 2. Deionised water. 3. Glacial acetic acid, HPLC grade (Sigma-Aldrich, Poole, UK). 4. Nylon filters (0.45 µm). 5. Volumetric flask (1 L) and measuring cylinders (0.5 and 1 L).

2.2.5. LC/ESI–MS/MS Analysis

1. C18 HPLC column (Luna, 5 mL, 150 × 2 mm) (Phenomenex, Macclesfield, UK). 2. SecurityGuard™ Cartridge holder (Phenomenex, Macclesfield, UK). 3. SecurityGuard™ Cartridges (C18 4 × 2.0 mm) (Phenomenex, Macclesfield, UK).

2.2.6. Protein Content

Protein Assay Kit using BSA as protein standard (BioRad, Hemel Hempstead, UK).

3. Methods 3.1. Solutions of Prostanoids 3.1.1. Stock and Working Solutions: Preparation and Storage

1. Prostanoid standards are provided either as dry material or solution in organic solvent (usually methyl acetate). If the material is solid, proceed to the next step. If the material is in solution carefully evaporate the solvent using a fine stream of nitrogen. You can now reconstitute the residue in the desired volume of ethanol and make stock solutions of the concentration of choice (see Note 1). 2. Add the appropriate volume of ethanol in each one of the prostanoid-containing vials to prepare stock solutions of 1 mg/mL

276

Nicolaou, Masoodi, and Mir

(e.g. use 1 mL ethanol to dissolve 1 mg, etc.). Flash the vials with nitrogen to create an inert atmosphere and seal them (see Note 2). These stock solutions should be kept at −20°C for a few weeks or −80°C if longer storage periods are needed. It is advisable to use parafilm to seal the vials: organic solvents evaporate even at low temperatures. It is also a good idea to keep the stock solutions in the original vial to avoid loses from changing vials (see Note 3). 3. Using a glass syringe transfer carefully 10 mL of each prostanoid stock solution (1 mg/mL) into a clean amber glass vial using a glass syringe, and add 990 mL of ethanol to make a final concentration of 10 ng/mL (see Note 4). These are the working solutions that will be used to construct composite standards for the calibration lines (see Note 5). 4. In order to prepare the internal standard solution take 10 mL PGB2-d4 (0.5 mg/mL commercially available solution) and dilute it to 500 mL using ethanol (final concentration 10 ng/ mL). This is the working solution of IS and can be kept at −20 or −80°C as describe before (see Note 6). 3.1.2. Calibration Lines

1. 40 mL of each standard (working solution 10 ng/mL) are transferred to a clean 1.5 mL amber glass vial using a glass syringe. The solution is made up to 1 mL with ethanol (final concentration of each standard is 400 pg/mL). This is the mixed standard working solution that will be used to make calibration lines. The mixed standard is prepared every few weeks, aliquoted into 200 mL aliquots and stored at −20°C (see Note 7). 2. The calibration lines are prepared by diluting the mixed standard solution (400 pg/mL) to the following final concentrations: 200 pg/mL, 100 pg/mL, 50 pg/mL, 20 pg/mL, 10 pg/mL, and 1 pg/mL. Internal standard PGB2-d4 is added to every one of these solutions (40 mL of the 1 ng/mL working IS solution, see Subheading 3.1.1, step 4) to a final concentration of 400 pg/mL. Typically, we use varied volumes of the mixed standard, 40 mL of internal standard and ethanol up to total volume of 100 mL. A fresh set of solutions is prepared for each experiment. At the end of the analysis all remaining solutions of the standards should be discarded.

3.2. Collection and Storage of Biological Samples

In principle, collect on ice, aliquot, and freeze the samples, as soon as possible. Do not add any preservatives. Store at −80°C and, if needed, transport on dry ice.

3.2.1. Plasma

Rat plasma is prepared in EDTA tubes, aliquoted to 500 mL aliquots in cryogenic vials and stored at −80°C (see Note 8).



Lipidomic Analysis of Prostanoids by Liquid

277

3.2.2. Urine

Rat urine was aliquoted to 500 mL aliquots in cryogenic vials and stored at −80°C.

3.2.3. Cell Culture Media

The number of cells needed for a successful experiment depends on their ability to produce prostanoids. Typically, we grow cells in T25 or T75 flasks and when they reach 80% confluency, we add 5 mL (T25 flask) or 10 mL (T57 flask) of fresh medium and the appropriate treatment (e.g. Ca-ionophore or arachidonic acid to stimulate prostanoid production). At the end of the treatment period the medium is collected in 12 mL polypropylene vials and stored at −80°C (see Note 9).

3.2.4. Solid Tissue Samples

The tissue samples (20–200 mg) should be collected and stored in dry ice to avoid any post-mortem changes. Tissue can then be kept at −80°C for a few months.

3.3. Sample Preparation

Perform all the following steps on ice using a large ice box and protect the resulting solutions from direct sunlight.

3.3.1. Plasma or Urine Samples

1. Defrost the sample on ice (see Note 10). 2. Transfer the liquid in a glass extraction vial using a glass Paster pipette. Keep a small amount (e.g. 10 mL) for protein content determination, if needed (see Note 11). 3. Add internal standard (40 mL of the freshly prepared 1 ng/mL PGB2-d4 solution; see Subheading 3.1.1, step 4). 4. Dilute plasma or urine samples with water and add methanol to 15% (v/v) (see Note 12). 5. Mix the solution gently and leave it on ice for 15 min. 6. Centrifuge (5 min, 4°C, 1,000 x g) to remove any precipitating material, if needed. 7. Transfer the clear supernatant in an appropriately labelled clean ice cold glass vial and acidify to pH 3.0 using 0.025 M HCl (see Note 13). 8. Immediately, apply the acidified solution to a preconditioned SPE cartridge (see Subheading 3.4 and Note 14).

3.3.2. Cell Culture Media

1. Defrost the medium as described in Subheading 3.3.1, step 1 (see Note 10). 2. Add internal standard (40 mL of the freshly prepared 1 ng/mL PGB2-d4 solution; see Subheading 3.1.1, step 4). 3. Add methanol to 15% (see Note 15). 4. Acidify to pH 3.0 using 0.1 M HCl (see Note 16). 5. Immediately, apply the acidified solution to a preconditioned SPE cartridge (see Subheading 3.4 and Note 14).

278

Nicolaou, Masoodi, and Mir

3.3.3. Solid Tissue Samples

1. Defrost the samples on ice and transfer it in the ice-cold glass mortar of the homogeniser (see Note 17). 2. Add 1 mL of ice-cold water into the mortar. Using the tightfitting pestle, use 35 up and down strokes to homogenise the sample. Work in an ice box. 3. Using a glass Pasteur pipette transfer the homogenate to a clean glass vial. Use a further 2 mL of the 15% methanol/ water solution to rinse the mortar and pestle and transfer the wash into the same sample-containing glass vial (final volume of the homogenate is 3 mL). 4. Keep a small amount of the sample (e.g. 10 mL) for protein content determination (see Note 11). 5. Add internal standard (40 mL of the freshly prepared 1 ng/mL PGB2-d4 solution; see Subheading 3.1.1, step 4). 6. Add methanol to 15% (v/v) (see Note 12). 7. Place the lids on the extraction vials and vortex for 30 s. 8. Allow the samples to stand on ice in the dark for 30 min. 9. Centrifuge the tubes at 1000 x g for 5 min at 4°C to remove the precipitating proteins. 10. Using a glass Pasteur pipette transfer the clear supernatant to a new clean appropriately labelled ice cold glass tube and acidify to pH 3.0 using 0.025 M HCl (see Note 13). 11. Immediately, apply the acidified solution to a preconditioned SPE cartridge (see Subheading 3.4 and Note 14).

3.4. Solid Phase Extraction 3.4.1. Cartridge Preconditioning

1. Attach the required number of SPE cartridges to the vacuum manifold and apply a pressure of approximately 3 mmHG. Place a waste bucket in the manifold to collect and discard the washes. 2. To activate the cartridges, wash them with 20 mL methanol, followed by 20 mL water (see Note 18 and 19). 3. Close the taps to stop the flow and turn the vacuum off. 4. Empty the waste bucket and replace it, in preparation of the next stage.

3.4.2. SPE Analysis of the Extracts

1. Transfer the acidified ice-cold sample onto one of the activated SPE cartridges using a long glass Pasteur pipette. Make sure the taps are switched off. 2. Turn the vacuum back on and let the liquid go through the SPE in a drop-wise manner. When all the liquid is through, switch off the tap. Take care not to allow any air through the sorbent at any point. Let the eluent run in the waste. 3. Repeat the process with the remaining samples (i.e. acidify and apply on the SPE), one at a time (see Note 14).



Lipidomic Analysis of Prostanoids by Liquid

279

4. Wash all the cartridges with the following solvents (in sequence): 15% methanol/water (20 mL), water (20 mL), and hexane (10 mL). Adjust the vacuum to attain a drop-wise flow. Let the solvents run in the waste (see Note 20). 5. Switch off the taps to stop the flow and turn off the vacuum. 6. Remove the waste basket from the manifold, and place clean, appropriately labelled empty glass tubes under each cartridge. 7. Elute the prostanoids with 10 mL methyl formate (see Note 21). 8. Carefully stop the vacuum, remove the extract-containing vials, put caps on each tube and transfer them to an ice box. 3.5. Sample Reconstitution and Preparation for LC/ ESI–MS/MS Analysis

1. Evaporate the solvent from the semi-purified samples (see Note 22). 2. Once the solvent has been completely removed, add 100 mL ethanol to re-dissolve the residue (see Note 23). 3. Set up and label the appropriate number of amber autosampler glass vials. Place an insert vial in each one of them and prepare lids by placing a Teflon disk over the hole in each cap. 4. Collect the ethanolic solution of the extract with the same glass syringe and transferred to the relevant insert vial. 5. Flash the vials with nitrogen, cap them and store them at −20°C. The samples should be analysed within 5 days (see Note 24).

3.6. HPLC Solvents

1. Solvent A: acetonitrile: water: glacial acetic acid (45:55:0.02; v/v/v). Using a measuring cylinder measure out 450 mL acetonitile and pour it in a volumetric flask (1 L). To this add 0.20 mL acetic acid and make up the volume to 1 L using deionised water. Mix thoroughly, vacuum filter, degas, and transfer to a 1 L Duran bottle suitable for HPLC (see Note 25). 2. Solvent B: acetonitrile: water: glacial acetic acid (90:10:0.02; v/v/v). Repeat the process described in step 1, this time using 900 mL acetonitrile (see Note 25). 3. Solvent C: acetonitile: water (50:50; v/v) with 0.3% acetic acid. This is used to wash the column at the end of each run. 4. Solvent D: acetonitile: water (50:50; v/v). This is the second wash step before storing the column. 5. Seal wash: acetonitrile: water (10:90; v/v). 6. Needle wash: acetonitrile: water (70:30; v/v).

280

Nicolaou, Masoodi, and Mir

3.7. LC/ESI–MS/MS Analysis

1. Condition the column running HPLC solvents A and B at a ratio of 50:50 at a flow rate 0.2 mL/min for at least 30 min. Ensure there is a stable base line (see Note 26). 2. Set-up the autosampler chamber temperature at 8°C. 3. Place the sample vials in the carousels making a note of the slots you have used so that you can set up the run sequence. 4. Set the following gradient system (MassLynx LC inlet method file): Time (min)

% Solvent A

% Solvent B

0.00

100

0

8.00

100

0

8.10

50

50

12.00

50

50

12.10

30

70

20.00

30

70

21.00

100

0

30.00

100

0

Run time: 30 min; Flow: 0.2 mL/min; Injection volume: 5 mL (see Note 27). 5. Set up the MS Method file (MassLynx method page) using the MRM and cone voltage settings listed in Table 1. The run time is 30 min (same duration as the HPLC inlet file), dwell time: 0.2 s and interscan delay: 0.10 s. 6. Check the MassLynx tune page for the following settings: Capillary voltage: −3.00 KV; Cone voltage: 35 eV; RF lens 1: 0.3; Aperture: 0.3; RF lens 2: 1.0; Source temperature: 120°C; Dessolvation temperature: 360°C; LM Res 1: 14.0; LM Res 2: 14.0; HM Res 1: 14.0; HM Res 2: 14.0; Ion Energy 1: 0.5; Ion energy 2: 1.0; Entrance: -1; Exit: 2. 7. Check that the API and collision gases are switched on. 8. Load the MS Method. The temperatures on the mass spec should begin to rise. 9. Programme the sequence of analysis on the MassLynx main page using the MS Method and Inlet Method files. Programme duplicate injections for all the samples and calibration lines (see Note 28). 10. At the end of each long run disconnect the column from the spectrometer and let it run in a waste-solvent bottle. Wash the MS inlet tube and capillary with methanol. Set up the



Lipidomic Analysis of Prostanoids by Liquid

281

Table 1 Experimental details for the ESI–LC–MS/MS lipidomic assay of prostanoids Prostanoid

MRM (m/z)

Collision energy (eV)

Indicative LC retention times (min)

PGD1

353 ®→317

15

4.79

PGE1

353 ®→317

15

4.47

PGF1a

355 ®→311

25

3.69

6-keto PGF1a

369 ®→163

23

2.83

PGB2

333 ®→175

20

9.96

PGB2-d4

337 ®→179

20

9.96

PGD2

351 ®→271

17

5.18

PGE2

351 ®→271

17

4.47

PGF2a

353 ®→193

25

3.77

PGJ2

333 ®→271

15

9.17

D12-PGJ2

333 ®→271

15

9.80

15-deoxy-D12,14 PGJ2

315 ®→271

15

18.10

PGD3

349 ®→269

15

4.08

PGE3

349 ®→269

15

3.85

PGF3a

351 ®→193

25

3.30

TXB2

369 ®→169

17

3.46

TXB3

367 ®→169

15

3.06

13,14-dihydro PGE1

355 ®→337

15

5.41

13,14-dihydro-15-keto PGE1

353 ®→335

12

7.61

13,14-dihydro-PGF1a

357 ®→113

38

4.94

13,14-dihydro-15-keto PGF1a

355 ®→193

32

6.67

13,14-dihydro-15-keto PGE2

351 ®→333

12

6.98

13,14-dihydro PGF2a

355 ®→311

30

4.79

13,14-dihydro-15-keto PGF2a

353 ®→113

28

6.12

column to wash at 0.2 mL/min with acidified acetonitrile– water (solvent C) for at least 30 min followed by a wash in acetonitrile–water (60 min or longer). The column can then be stored.

282

Nicolaou, Masoodi, and Mir

3.8. Data Analysis

1. Calculate the peak–area ratio of every standard prostanoid to PGB2-d4 (internal standard) and plot against the concentration of the calibration standards. Calibration lines are calculated by the least square regression method. 2. To calculate the concentration of each metabolite the peak– area ratio to PGB2-d4 should be calculated and read off the calibration lines. Table 1 presents indicative retention times of the prostanoids included in the assay. Although absolute retention times cannot be reported, these values indicate the relative elution position of prostanoids (see Note 29). Data should be normalised for protein content or sample volume depending on the application.

4. Notes 1. A glass Pasteur pipette attached to the nitrogen gas cylinder through a plastic tube makes a very useful device: you can direct the flow of the gas directly to the vial to evaporate the solvent; the pipette should be changed between samples to avoid contamination. Direct the flow of the gas directly to the surface of the solution; avoid strong flow that will create splashes. 2. Stock and working solutions should be kept in amber vials under nitrogen to avoid oxidation. The system described in Note 1 can be used to flash any prostanoid containing vials prior to storage. 3. Check the suppliers instructions regarding the storage condi­ tions needed for each compound: stability varies from 6 months to 2 years at −20°C depending on compound, e.g. thromboxane is stable for up to 6 months whilst most prostaglandins are stable for up to 2 years. In principle you should be keeping all stock solutions at −80°C, protected from light and under nitrogen. When you need to use any solution that has been kept cold, allow it to take room temperature before you open it. This way you will avoid water condensation in the cold vial. 4. Organic solvents do evaporate even at low temperatures. It is good practice to keep a log of the volume added in each vial when you make the stack solutions and then every time you use them. If in doubt you can dry and reconstitute using appropriate volumes of solvent. 5. Make a few aliquots of those solutions when you receive the prostanoid standards and store them in a separate box. This will stop you freezing and defrosting the stock solutions, thus minimizing the evaporation of the solvent and will help



Lipidomic Analysis of Prostanoids by Liquid

283

keeping the standards longer. We have noted that following multiple cycles of freezing/defrosting, the standards can “go-off” suddenly. In case of contamination the working solutions can be thrown away. 6. If the concentration of the commercially available standard is different, use the procedure described in step 1 to make the 10 ng/mL stock solutions. Every time you need internal standard take 100 mL of the 10 ng/mL stock solution and dilute it 1:10 (add 900 mL ethanol) to prepare 1 mL of 1 ng/mL solution: this will be used to spike the calibration lines and biological sample extracts. 7. Each vial contains enough material to prepare one set of calibration lines; any remaining material should be discarded: this helps to avoid freezing and defrosting the standards. 8. You may prepare aliquots of larger or smaller volumes but keep in mind that it is not advisable to re-freeze and keep sample that has been defrosted. It is better to prepare more aliquots of smaller volume and combine them up as needed. 9. Whenever possible, we perform cell treatments in serumfree media to avoid matrix effects. However, this depends entirely on the type of cells and experimental design, and appropriate controls are needed to examine the presence of any oxidation or other products present in the serum resulting in matrix effects. Furthermore, some drug treatments may interfere with the MS analysis: you should always plan control and blank experiments of the treatments to ensure that the MS analysis is reliable. 10. It is important to keep the samples cold at all times; do not use water bath to thaw the samples. You may warm them up using your hands but always make sure there is a little ice cub left in the sample indicating that the solution temperature is close to 0°C. 11. Protein assay protocols are normally provided with the commercially available kits; therefore, we will not give any details in the present protocol. 12. Typically, for plasma or urine samples up to 1 mL, we add water to a final volume of 3 mL. For a sample volume of 3 mL we add 0.530 mL methanol to bring the final solution to 15% methanol (v/v). 13. For a 3 mL solution you will need about 2–3 drops of 0.025 M HCL. After you add the acid, mix gently the solution, take a clean glass Pasteur pipette, dip it in the solution, take one drop and apply it to the indicator paper. Do not dip the indicator in the solution. Read the pH. If you need more acid, repeat this process carefully.

284

Nicolaou, Masoodi, and Mir

14. We recommend batches of no more than six samples at a time. This is a manageable number of samples for the SPE analysis. 15. Typically, we add 1.77 mL methanol in 10 mL medium. 16. For 10 mL medium we use 8–10 drops 0.1 M HCL. 17. It is a good idea to warm up the vial gently by rubbing in between your hands but do not allow it to completely melt. As soon as the sample has melted enough transfer it out to the homogeniser (always work on ice). You can use a long glass Pasteur pipette to help transfer the sample. 18. Adjust the vacuum to attain a steady drop-wise flow through the cartridge. Stop the flow of the solvent through the cartridges just as the meniscus touches the top of the sorbent bed taking care not to allow any air through – the SPE cartridges must not dry out. Once activated the cartridges should be kept wet and be used as soon as possible; we aim to use them within 10 min. 19. We have found it useful to have one 25-mL measuring cylinder “assigned” to each cartridge: we use it to measure and add the required solvents. This helps us to monitor the flow and progress of each cartridge/sample and respond to unforeseen experimental problems more efficiently; e.g. blocked cartridge, broken glass vial, vacuum problems, etc. 20. These washes are designed to remove unbound (15% methanol) and water-soluble (water) material, whilst hexane removes any remaining water prior to the elution of prostanoids with methyl formate. 21. Take care to maintain a slow drop-wise flow through the cartridges and collect the eluent in the appropriate glass tubes. Once all the solvent has passed through the cartridges allow air to pass through for a few seconds to ensure all the solvent has been collected. 22. We evaporate the organic solvent under nitrogen using the SPE manifold drying attachment and stainless steel needles to direct the flow of the gas into the glass tubes. Place the tubes in a rack so there is a needle directly above each one. Adjust the pressure of the gas cylinder so that there is a gentle flow of nitrogen over the surface of the solvent causing it to slightly ripple. The flow will need to be periodically readjusted as the methyl formate evaporates. 23. Using a glass syringe withdraw and spread the ethanol along the sides of the tube taking care to wet the entire inner surface of the tube; holding the tube at an angle will help. To collect any drops of ethanol stuck to the sides of the glass, centrifuge the tube 1,000 ´ g (20 s). Collect the solution with the same glass syringe and transferred to the relevant insert vial.



Lipidomic Analysis of Prostanoids by Liquid

285

24. Our usual practice is to prepare the samples a few days before the LC/ESI–MS/MS analysis and run them as soon as possible. Occasionally, we have stored the samples up to 2 weeks (under nitrogen, vials sealed with parafilm, at −20°C) without any problems. 25. Although practically all LC/MS systems are fit with solvent degassers, we find it useful to vacuum filter and degas all the solvents prior to using them. Make fresh solutions for every experiment; the volume of solvent described here will suffice to run the system for a week. 26. We tend to run 1–2 cycles with blank (ethanol) injections to condition the system. We suggest you run a quality control sample to ensure the appropriate response of the instrument. Before you start the analysis of the biological samples, make sure that you have a stable base line and reproducible retention times 27. This protocol includes wash and equilibration steps programmed in to prepare the system for the next injection. There is flexibility in increasing/decreasing these periods to suit your application. However, our advice is to invest in a longer wash/equilibration steps in between injections so that you can run more samples without needing to stop and clean the column. Our plan allows for up to 64 biological samples and the appropriate calibration lines to be run in one uninterrupted run lasting approximately 70 h. 28. We have found it very useful to analyse the calibration line standards at intervals during the run. For example, we inject the lowest concentration of the calibration line first, then programme a number of biological samples, then a blank, then the second lowest concentration of the calibration line, etc. This practice has been proven useful in cases of unforeseen experimental problems (power cuts, software problems, etc.) that can affect unsupervised overnight runs. 29. Some times the deuterated standards contain a small percent of non-deuterated analyte. If that is the case the non-deuterated amount should be estimated and subtracted from the corresponding values.

Acknowledgements This project has been supported in part by the Wellcome Trust Grant WT077714. We acknowledge the excellent technical support provided by Andrew Healey, Analytical Centre, University of Bradford.

286

Nicolaou, Masoodi, and Mir

References 1. Nicolaou, A. (2004) Prostanoids, Bioactive Lipids. Nicolaou, A., Kokotos, G. editors, 197–222. 2. Schaloske, R. H. & Dennis, E. A. (2006) The phospholipase A2 superfamily and its group numbering system, Biochim Biophys Acta. 1761, 1246–59. 3. Smith, W. L., DeWitt, D. L. & Garavito, R. M. (2000) Cyclooxygenases: structural, cellular, and molecular biology, Annu Rev Biochem. 69, 145–82. 4. Simmons, D. L., Botting, R. M. & Hla, T. (2004) Cyclooxygenase isozymes: the biology of prostaglandin synthesis and inhibition, Pharmacol Rev. 56, 387–437. 5. Helliwell, R. J., Adams, L. F. & Mitchell, M. D. (2004) Prostaglandin synthases: recent developments and a novel hypothesis, Prostaglandins Leukot Essent Fatty Acids. 70, 101–13. 6. Tai, H. H., Ensor, C. M., Tong, M., Zhou, H. & Yan, F. (2002) Prostaglandin catabolizing enzymes, Prostaglandins Other Lipid Mediat. 69, 483–93. 7. Wise, H., Wong, Y. H. & Jones, R. L. (2002) Prostanoid signal integration and cross talk, Neurosignals. 11, 20–8. 8. Samad, T. A., Sapirstein, A. & Woolf, C. J. (2002) Prostanoids and pain: unraveling mechanisms and revealing therapeutic targets, Trends Mol Med. 8, 390–6. 9. Bishop-Bailey, D., Calatayud, S., Warner, T. D., Hla, T. & Mitchell, J. A. (2002) Prostaglandins and the regulation of tumor growth, J Environ Pathol Toxicol Oncol. 21, 93–101. 10. Olson, D. M. (2003) The role of prostaglandins in the initiation of parturition, Best Pract Res Clin Obstet Gynaecol. 17, 717–30. 11. Battistini, B., Picard, S., Borgeat, P. & Sirois, P. (1998) Measurements of prostanoids, leukotrienes, and isoprostanes by enzyme immunoassays, Methods Mol Biol. 105, 201–7. 12. Kempen, E. C., Yang, P., Felix, E., Madden, T. & Newman, R. A. (2001) Simultaneous quantification of arachidonic acid metabolites in cultured tumor cells using high-performance liquid chromatography/electrospray ionization

tandem mass spectrometry, Anal Biochem. 297, 183–90. 13. Baranowski, R. & Pacha, K. (2002) Gas chromatographic determination of prostaglandins, Mini Rev Med Chem. 2, 135–44. 14. Yue, H., Strauss, K. I., Borenstein, M. R., Barbe, M. F., Rossi, L. J. & Jansen, S. A. (2004) Determination of bioactive eicosanoids in brain tissue by a sensitive reversed-phase liquid chromatographic method with fluorescence detection, J Chromatogr B Analyt Technol Biomed Life Sci. 803, 267–77. 15. Turk, J., Weiss, S. J., Davis, J. E. & Needleman, P. (1978) Fluorescent derivatives of prostaglandins and thromboxanes for liquid chromatography, Prostaglandins 16, 291–309. 16. Murphy, R. C., Barkley, R. M., Zemski Berry, K., Hankin, J., Harrison, K., Johnson, C., Krank, J., McAnoy, A., Uhlson, C. & Zarini, S. (2005) Electrospray ionization and tandem mass spectrometry of eicosanoids, Anal Biochem. 346, 1–42. 17. Masoodi, M. & Nicolaou, A. (2006) Lipidomic analysis of twenty-seven prostanoids and isoprostanes by liquid chromatography/electrospray tandem mass spectrometry, Rapid Commun Mass Spectrom. 20, 3023–9. 18. Margalit, A., Duffin, K. L. & Isakson, P. C. (1996) Rapid quantitation of a large scope of eicosanoids in two models of inflammation: development of an electrospray and tandem mass spectrometry method and application to biological studies, Anal Biochem. 235, 73–81. 19. Nithipatikom, K., Laabs, N. D., Isbell, M. A. & Campbell, W. B. (2003) Liquid chromatographic-mass spectrometric determination of cyclooxygenase metabolites of arachidonic acid in cultured cells, J Chromatogr B Analyt Technol Biomed Life Sci. 785, 135–45. 20. van Meer, G. (2005) Cellular lipidomics, Embo J. 24, 3159–65. 21. Balazy, M. (2004) Eicosanomics: targeted lipidomics of eicosanoids in biological systems, Prostaglandins Other Lipid Mediat. 73, 173–80. 22. Serhan, C. N. (2005) Mediator lipidomics, Prostaglandins Other Lipid Mediat. 77, 4–14.

Chapter 15 Qualitative and Quantitative Analyses of Phospholipids by LC–MS for Lipidomics Hiroki Nakanishi, Hideo Ogiso, and Ryo Taguchi Summary In this chapter we are going to mention about three different approaches in lipidomics and how to effectively profile or calculate the amounts of phospholipids from major molecular species up to minor ones. 1) Precise identification and profiling of individual molecular species of phospholipids by data-dependent LC–ESIMS/MS combination with “Lipid Search”. We have been using this method as a global analysis of phospholipid. We usually applied this method at least once for new biological samples. We constructed an automated search engine, “Lipid Search”, for identification and profiling of phospholipids. Once after applying this analysis, a specified retention time can be obtained for each elution peak of individual phospholipid molecular species. Thus, reproducible identification results can be effectively obtained by our search engine from the data obtained by single LC or combination of LC with specified head group survey by using precursor ion scanning or neutral loss scanning. 2) An effective analytical method of LC–ESIMS for the identification of acidic phospholipids such as phosphatidic acid and phosphatidylserine. This is an approach of how to obtain sharp chromatographic peaks for acidic lipids such as phosphatidic acid and phosphatidylserine that are normally detected as broad elution peaks. With this improvement very small amount of molecular species in minor acidic phospholipids were effectively obtained. 3) Identification and profiling of molecular species in focused phospholipids. Third one is a combination analysis of focused methods such as precursor ion scanning or neutral loss scanning and high efficient LC separation. As reported previously, different combinations of fatty acids on sn-1 and sn-2 can be mostly detected as separate peaks by reverse phase LC–ESIMS. Detection limit of precursor ion scanning or neutral loss scanning is more than ten times higher than that of the method without LC separation, because of decreased ion suppression. We will mention about application of this methods for focused analysis on phosphatidylethanolamine-plasmalogens. Key words:  Quantitative analysis, Phospholipid, Lipidomics, Liquid chromatography, Mass spectrometry, Search engine

Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, DOI 10.1007/978-1-60761-322-0_15, © Humana Press, a part of Springer Science + Business Media, LLC 2009

287

288

Nakanishi, Ogiso, and Taguchi

1. Introduction In lipidomics study, liquid chromatography electrospray ionization mass spectrometry (LC–ESIMS) is an effective technique (1–5). Normally quantitative analyses of metabolites by mass spectrometry were performed by LC–ESIMS/MS with the use of multiple reaction monitoring (MRM) or selected reaction monitoring (SRM) mode (6). It is true in some targeted lipidomics especially for identification of minor phospholipids (PLs) such as oxidized PLs (Nakanishi, H. et al. (2009) J. Chromatogr. B. 877, 1366–1374). Within recent approach in -omics studies in lipids, flow injection or direct injection have been commonly used. But as indicated in our previous work there are several different molecular species that can exist within a single molecular composition, also ion suppression is unevitable without preliminary separation with LC (7). For this reason combination of mass spectrometry with LC seems to be essential to find out and estimate the level of minor molecular PLs within many structural isomers. We essentially calculate the amount of each molecular species of PLs from intensity or area of each mass chromatographic peak. Here we will mention about our recent lipidomics approaches for quantitative profiling of PLs especially in acidic PLs and phosphatidylethanolamine (PE)-plasmalogens. Phosphatidic acid (PA) is not only an important intermediate in the synthesis of all PLs, but also is an intracellular signaling molecule in mammalian cells (8). Phosphatidylserine (PS) is mostly localized within the inner membrane leaflet. The appearance of PS at the cell surface is a distinctive hallmark of apoptosis and is crucial to the process of cell-death for the recognition and clearance of apoptotic debris (9). Lysophosphatidic acid (LPA) has several biological activities, such as platelet aggregation and cancer cell invasion, and is well known as a natural ligand for G-protein-coupled receptors (10, 11). Lysophosphatidylserine (LPS) reportedly acts as a mediator for the chemotaxis of fibroblast cells (12). Previous reverse phase (RP) LC approaches did not sufficiently address the simultaneous analyses of PA and PS with the other PLs (13–15). As a consequence, the PA and PS peak tailings on RPLC have often yielded inferior separations and low detection sensitivity of these acidic PLs. To increase the specificity of the PL analysis, RPLC methods for PA and PS analyses are required. Thus, we developed an LC condition that successfully reduced such peak tailings in the elution profiles of both PA and PS (16). When a PL mixture from biological samples was examined by this method, PA, PS, and their lyso-forms, as well as the other PLs, could be measured in a single chromatographic step. This method will be advantageous for lipidomics studies of



Qualitative and Quantitative Analyses of Phospholipids by LC–MS

289

cellular responses and pathology. In addition, this method would be useful to detect minor lipid mediators, such as LPA, LPS, and their precursors. While, precursor ion scanning and neutral loss scanning of the polar head groups of PLs are valuable techniques for detecting most molecular species within the same class of PLs (2, 5). For example, precursor ion scanning at m/z 184 is used for cholinecontaining PLs such as phosphatidylcholine (PC), lysoPC (LPC), and sphingomyelin (SM) in the positive ion mode. Neutral scanning of 141, 185, 189, and 277 Da are used for PE, PS, phosphatidylglycerol (PG), and phosphatidylinositol (PI), respectively. Additionally, precursor ion scanning of carboxylic anions was also very useful for identifying molecular species of PLs with specified fatty acyl chains. We recently constructed combination methods for these group-specific surveys and LC separation. This method is more sensitive than normal head group or fatty acid scanning without LC. Thus, a combination of group-specific survey scanning with reverse phase separation is particularly effective. Plasmalogens are unique of PLs, containing an O-alk-1¢-enyl aliphatic chain at the sn-1 position of glycerol (17, 18). Plasmalogen molecular species are predominantly present in choline and ethanolamine glycerophospholipids (glyceroPLs). These unique PLs are found in high concentration in cellular membrane of numerous mammalian tissues; especially PC- and PE-plasmalogens are abundant in heart and brain, respectively. Plasmalogens are considered as antioxidant molecules that protect cells from oxidative stress. Indeed, Plasmalogen-deficiency disorders have been implicated in various diseases such as Zellweger syndrome, Retinitis pigmentosa, Down syndrome, and Alzheimer’s disease (17–19). Thus, the variation analysis of plasmalogen content is very important to understand these physiological and pathological phenomena. Conventionally, analyses of plasmalogens have been performed using thin-layer chromatography (TLC) or highperformance liquid chromatography (HPLC). However, the precise molecular species of plasmalogens cannot be determined by these methods. Therefore, it is possible that important biological implications have been overlooked by these methods. Here, we present a method for specific detection of PE-plasmalogens using RPLC–ESIMS/MS (20–23).

2. Materials 2.1. Equipment

1. Q-TOF micro™ mass spectrometer (Waters, Milford, MA, USA).

2.1.1. Mass Spectrometry

2. LTQ-Orbitrap mass spectrometer (Thermo-Fisher Scientific, Waltham, MA, USA).

290

Nakanishi, Ogiso, and Taguchi

3. 4000 Q-TRAP® mass spectrometer (Applied Biosystems/ MDS Sciex, Concord, ON, Canada). 2.1.2. LC System

1. Shimadzu HPLC system. Two LC-10ADvp-micro pumps, a DGU-14AM degassing unit, a SCL-10Avp controller (Shimadzu, Kyoto, Japan). 2. UltiMate HPLC system (Dionex, Sunnyvale, CA, USA). 3. UltiMate 3000 HPLC system (Dionex, Sunnyvale, CA, USA). 4. ACQUITY Ultra Performance LC® system (Waters, Milford, MA, USA).

2.1.3. Autosampler

1. FAMOS autosampler (Dionex, Sunnyvale, CA, USA). 2. HTC PAL autosampler (CTC Analysis, Zwingen, Switzerland).

2.1.4. Column

1. Deverosil C30 (5 mm, 100 mm × 0.3 mm i.d.) column (Nomura Chemical, Aichi, Japan). 2. Waters X-Bridge C18 (3.5 mm, 150 mm × 1.0 mm i.d.) column (Waters, Milford, MA, USA). 3. ACQUITY UPLC™ BEH C18 (1.7 mm, 150 mm × 1.0 mm i.d.) column (Waters, Milford, MA, USA).

2.1.5. Other Instrument

1. Low-speed centrifuge (TOMY, Tokyo, Japan). 2. Dry-thermo bath (Tokyo Rikakikai, Tokyo, Japan). 3. Ultra-sonic cleaner (Aiwa, Tokyo, Japan).

2.2. Reagents

1. Methanol (LC–MS grade) (Wako Pure Chemicals, Osaka, Japan). 2. Acetonitrile (LC–MS grade) (Wako Pure Chemicals, Osaka, Japan). 3. Isopropanol (HPLC grade) (Wako Pure Chemicals, Osaka, Japan). 4. Butanol (Wako Pure Chemicals, Osaka, Japan). 5. Chloroform (Wako Pure Chemicals, Osaka, Japan). 6. Formic acid (Wako Pure Chemicals, Osaka, Japan). 7. Acetic acid (Wako Pure Chemicals, Osaka, Japan). 8. Phosphoric acid (Wako Pure Chemicals, Osaka, Japan). 9. 28% ammonium aqueous solution (Wako Pure Chemicals, Osaka, Japan). 10. 37% hydrochloric acid (Wako Pure Chemicals, Osaka, Japan). 11. 5% dimethyldichlorosilane toluene solution (Wako Pure Chemicals, Osaka, Japan).



Qualitative and Quantitative Analyses of Phospholipids by LC–MS

291

12. Sodium hydroxide (Wako Pure Chemicals, Osaka, Japan). 13. DEAE-cellulose (Wako Pure Chemicals, Osaka, Japan). 14. Ultra pure water was obtained from a Milli-Q water system (Millipore, Milford, MA, USA). 15. Phosphatidylcholine mixture from pig liver (Avanti Polar Lipids, Alabaster, AL, USA). 16. 17:0-lysophosphatidic acid sodium salt [LPA (17:0)] (Avanti Polar Lipids, Alabaster, AL, USA). 17. 14:0/14:0-phosphatidic acid sodium salt [PA (14:0/14:0)] (Sigma-Aldrich, St. Louis, USA). 18. 1-O-alk-1¢-octadecenyl-2-arachidonoyl-sn-glycero-3phosphoethanolamine, plasmalogen [p18:0/20:4 PE] (Avanti Polar Lipids, Alabaster, AL, USA). 19. 1, 2-oleoyl-sn-glycero-3-phosphoethanolamine [18:1/18:1 PE] (Avanti Polar Lipids, Alabaster, AL, USA). 2.3. Supplies

1. 20-mL, 200-mL and 1-mL adjustable pipets. 2. 1.1-mL Glass autosampler vials and caps. 3. 10-mL and 100-mL microsyringes. 4. 1-mL and 5-mL Glass pipets. 5. 10-mL and 50-mL Glass tubes and caps. 6. Pasteur pipettes. 7. Glass wools.

3. Methods 3.1. Untargeted and Global Lipidomics by LC–ESIMS/MS

Here we will mention this protocol very briefly, including the instruction our search engine “Lipid Search.” Most suitable system for this analysis is the combination of high separation LC–MS and accurate and high duty cycle data-dependent MS/ MS by hybrid of quadrupole (Q) and time-of-flight (TOF), such as Q-TOF or Q-Star (Applied Biosystems/MDS Sciex, Concord, ON, Canada) or hybrid ion trap (IT), and Fourier-transform (FT) or TOF, such as LTQ-Orbitrap and LC-IT-TOF (Shimadzu, Kyoto, Japan). A combination of LC and data-dependent shotgun MS/MS is a sort of global lipidomics (7). We normally used the negative ion mode to detect fatty acids as fragment ions from PLs by MS/ MS, and performed data-dependent scanning which automatically identifies molecular peaks with high intensity for subsequent

292

Nakanishi, Ogiso, and Taguchi

product ion scanning. A C30 reverse phase column was applied to separate minor molecular species. When this column was used, the influence of the fatty acyl chains was greater than that of the polar head group. Different molecules with the same mass value (i.e., the total number of carbons of the two fatty acyl chains and unsaturated bonds are the same) were detected at different elution times. One advantage of this method is a possibility that unexpected or new lipid metabolites might be detected with this method. 3.1.1. Results 3.1.1.1. HPLC Mobile Phases 3.1.1.2. Mobile Phase

1. Composition: Acetonitrile/methanol/triethyamine (25/75/1 v/v/v) supplemented with 0.3% formic acid. 2. Preparation: Mix 250 mL of acetonitrile, 750 mL of methanol, and 1 mL of triethyamine. Add 3 mL of formic acid. Add 28% aqueous ammonium and adjusted to pH 6.8. Degas by sonication for 5 min in a sonicator bath.

3.1.1.3. LC Separation Experiment

The ESIMS analysis was performed using a Q-TOF micro™ with an Ultimate HPLC system combined with a FAMOS autosampler. LC separation was achieved using a Deverosil C30 (5 mm, 100 mm × 0.3 mm i.d.) column, at room temperature (25°C). The flow rate was 3 mL/min for isocratic elution. Typically, samples (3 mL) were simply injected by the autosampler. The total chromatographic run time was 70 min.

3.1.1.4. Mass Spectrometry Conditions

The spectra were recorded in negative ion mode. The scan range of the instrument was set at m/z 200–950. The capillary voltage was set at 3.2 kV, cone voltage at 35 V, and source block temperature 100°C.

3.1.1.5. Full-Mass Scanning

Triggering MS to MS/MS switching in a data-dependent manner was based on ion intensity. Precursor ions with intensity more than 4 counts/s in the MS spectra were selected automatically for MS/MS. The collision gas used for MS/MS experiments was argon (pAr = 7.5 × 10–5 mbar), and the collision energy was set at 35 eV. MS/MS was switched back to MS after 3 s.

3.1.1.6. Data-Dependent Scanning

3.1.1.7. Identification of Phospholipids by Reverse Phase LC–ESIMS/MS

PL mixtures were analyzed by LC–ESIMS/MS with data-dependent scanning. When using a C30 reverse phase column, PLs were eluted in order from the hydrophilic molecules to the hydrophobic molecule (Fig. 1). In PLs, the length and the number of unsaturated fatty acyl chains is mainly influenced to the elution order in each molecular species. In this experiment, we used the negative ion mode to detect fatty acids as fragment ions from PLs by MS/MS, and performed data-dependent scanning which automatically identifies molecular peaks with high intensity for subsequent product ion scanning (Fig. 2). The ion intensities of each molecular related ion were used for quantitative profiling of molecular species of PLs after proper compensation by standard PLs of same classes with our identification and profiling tool “Lipid Search” (http://lipidsearch.jp) (Table 1).

34:3

36:4

38:6

34:2 32:1

38:5

32:0

38:4

40:5

Time (min)

36:3

40:6

34:1

36:2

38:3

40:4

36:1 34:0

38:2

TIC

Qualitative and Quantitative Analyses of Phospholipids by LC–MS

m/z

Fig. 1. 2D-map of phospholipids from rat liver obtained by RPLC–ESIMS using the C30 reverse phase column in negative ion mode. The 2D-map has the m/z value of [M + HCO2]− ions along the vertical axis and the retention time along the horizontal axis. When using a reverse phase column, phospholipids elute in order from the hydrophilic to the more hydrophobic molecules. In phospholipids, the length of fatty acyl chains mainly influence the elution order, i.e., in the order 32:1, 34:1, and 36:1. In addition, the number of double bonds in fatty acyl chains also mainly influence the elution order, i.e., in the order 34:3, 34:2, 34:1, and 34:0. (a) TIC, total ion chromatogram. (b) 2D-map. (From Ref. 7).

b

a

293

c

m/z

e’’

e’

m/z 802 (22.9 min) 16:0/18:2 PC

b’

m/z 804 (29.4 min) 16:0/18:1 PC

e

m/z 802 (20.9 min) 16:1/18:1 PC

d’

m/z 800 (19.2 min) 16:0/18:3 PC

a’

b

d

m/z 800 (16.9 min) 16:1/18:2 PC

a

m/z 828 (28.1 min) 18:0/18:3 PC

m/z 828 (25.1 min) 16:0/20:3 PC

m/z 828 (21.7 min) 18:1/18:2 PC

m/z 826 (21.5 min) 16:0/20:4 PC

m/z 826 (17.6 min) 18:2/18:2 PC

Fig. 2. MS/MS spectra of [M + HCO2]− ions of PCs from pig liver obtained using RPLC–ESIMS/MS in data-dependent scanning mode. When different molecules with the same mass value were separated, the MS/MS spectra of the other molecular species are shown with a prime, e.g., (a)¢, (b)¢, etc. (From Ref. 7).

Relative Intensity (%)

294 Nakanishi, Ogiso, and Taguchi



Qualitative and Quantitative Analyses of Phospholipids by LC–MS

295

Table 1  Molecular species of PC mixture from pig liver identified by RPLC–ESI/MS/MS m/z

Molecular species

Retention time (min)

m/z

Molecular species

Retention time (min)

764.55

a18:0/14:0

28.3

117

830.60

18:0/18:2

26.2

2,586

a16:0/16:0

45.8

200

16:0/20:2

31.9

15,362

774.57

16:1/16:1

17.4

334

18:1/18:1

31.9

15,362

776.55

14:0/18:1

21.3

184

16:0/20:1

43.1

179

16:0/16:1

21.4

1,897

18:0/18:1

44.2

23,995

778.56

16:0/16:0

33.2

2,515

840.56

18:0/20:4

25.8

1,054

790.62

a16:0/18:1

25.6

1,592

848.59

18:0/20:7

21.1

204

a18:1/16:0

37.6

363

850.59

16:0/22:6

16.1

1,034

792.62

a18:0/16:0

76.7

4,684

18:2/20:4

20.8

531

798.54

14:0/20:4

16.4

112

18:1/20:4

20.3

1,708

800.57

16:1/18:2

16.9

685

16:0/22:5

22.4

1,998

16:0/18:3

19.2

1,172

18:0/20:5

26.1

562

16:1/18:1

20.9

1,264

16:0/22:4

22.6

1,055

16:0/18:2

22.9

24,008

18:0/20:4

32.7

22,888

804.54

16:0/18:1

29.4

29,028

16:0/22:3

25.9

119

806.69

16:0/18:0

53.8

1,090

18:1/20:2

27.0

92

810.55

a16:1/20:4

24.9

129

18:0/20:3

29.3

3,191

812.56

a16:0/20:4

18.9

167

858.60

18:0/20:2

35.2

261

814.53

a18:0/18:3

18.6

293

860.58

18:0/20:1

67.4

124

a18:1/18:2

40.5

83

870.63

20:5/20:5

19.2

95

a18:0/18:2

17.8

296

878.61

18:1/22:5

11.5

99

a18:1/18:1

27.9

2,023

18:0/22:6

28.4

623

818.58

a18:0/18:1

13.6

232

880.69

18:0/22:5

31.5

1,348

820.61

a18:0/18:0

35.0

6,550

882.60

18:0/22:4

38.0

742

826.53

18:2/18:2

17.6

1,066

884.64

18:0/22:3

31.0

123

16:0/20:4

21.5

12,237

18:1/18:2

21.7

3,140

16:0/20:3

25.1

2,010

18:0/18:3

28.1

703

802.58

816.57

828.59

Intensity

832.62

852.66

854.62

856.60

a 1-O-alkyl aliphatic chain at the sn-1 position. From Ref. 7

Intensity

296

Nakanishi, Ogiso, and Taguchi

“Lipid Search,” a Search Engine for the Identification of Phospholipid Molecular Species from MS Raw Data

The latest version of “Lipid Search” was opened to the public in June 2008 (Fig. 3). This tool can directly handle raw MS data obtained from Q-TOF, LTQ, LTQ-Orbitrap, Quantum-Ultra (Thermo-Fisher Scientific, Waltham, MA, USA), 4000 Q-TRAP®, and LC-IT-TOF. The identification efficiency of “Lipid Search” highly depends on the qualities of MS data. Our database contains more than 200,000 theoretical m/z values of molecular weightrelated ions of individual lipid metabolites, and their theoretical fragment ions. Patterns of ion fragments have been theoretically

(http://lipidsearch.jp) Fig. 3. A window of “Lipid Search ver2.0 beta-version” search engine for lipid identification. User has to select the proper description in the each box of the indicated search condition.



Qualitative and Quantitative Analyses of Phospholipids by LC–MS

297

constructed and improved by using experimental data obtained from synthesized standards and many different natural samples. In latest version we set four different survey methods: the first is a molecular ion survey of only from the data on m/z value of molecular related ions (MS1); the second is a combination survey with m/z value of precursors and their fragments (MS2, MS3). The third one is a survey with precursor ion scanning (Prec). A fourth one is a survey with neutral loss scanning (NL) (http://lipidsearch.jp). The user must select the scan type used in the experiment, and enter each parameter, such as class, adduct ion, and tolerance. For example, the negative ion mode is normally selected for the identification of molecular species of glyceroPLs. The type of adduct ions are normally dependent on the eluting conditions and on the solvent used in the mass analysis. In our experiments, proton and ammonium are selected as positive adduct ions, and formate or acetate are selected as negative adduct ions, depending on the solvent used. In some cases – Na+ and K+ can be selected as positive adducts, and Cl  can be selected as a negative adduct. In addition, tolerance of mass data is mainly dependent on the MS machine used, and on the assay conditions. If a high tolerance value is selected, a large number of probable candidate molecules will be obtained, including those accidentally fitted to the search algorithm. Thus, appropriate tolerance values should be selected for each data set. Tolerances from 0.2 to 0.4 are normally recommended for the data obtained from quadrupole MS, and 0.02–0.1 are recommended for the data from TOF-MS. Tolerances from 0.01 to 0.001 are recommended for the data from FT-MS or FT-ICR (Ion Cyclotron Resonance)–MS. 3.2. Reverse Phase LC–ESIMS for Lipidomics, Improving Detection of Phosphatidic Acid and Phosphatidylserine

Although effective RPLC methods enabling the separation of PL molecular species had been developed, there were problems with the separation of PA and PS. These acidic PLs often eluted as extensively broad peaks, causing inferior separation, detection, and quantification, which was a sever limitation of the method. Thus, we have developed RPLC conditions that reduced the undesired peak tailings in the elution profiles of both PA and PS, by using a starting mobile phase containing a low concentration of phosphoric acid (5 mM) and a high percentage of water (40%). Our method sensitively analyzes PA, PS, and their lyso-forms, as well as the other phospholipids within a biological sample, in a single chromatographic step by an LC–MS method, and thus is suitable for lipidomics (16).

3.2.1. Pretreatments Glassware

All glassware, such as microsyringes, stoppered test tubes, sample vials, glass wools, and Pasteur pipettes, were silanized by immersing in a 5% dimethyldichlorosilane solution (see Note 1).

298

Nakanishi, Ogiso, and Taguchi

3.2.1.1. Metalware

To avoid adsorption of PA and LPA on the metal surface, microsyringes were rinsed with 10 mM phosphoric acid, followed by methanol rinsing before use.

3.2.1.2. Standard Lipid Solutions

As inner standards, 17:0 LPA and 14:0/14:0 PA were dissolved in methanol to prepare a 1 pmol each per mL solution. These standard solutions were stored at –80°C.

3.2.2. Sample Procedure

1. EL4 lymphoblastoma cells or A431 epidermoid carcinoma cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 100 units/mL penicillin, and 100 mg/mL streptomycin. After homogenization of the cultured cells (3 × 106~2 × 107), the lipids were extracted by the Bligh–Dyer method (24) (see Note 2). The extracted lipids, in 100 mL of methanol/chloroform (1/1 v/v), were stored at –30°C. For minor acidic PLs analyses, the extracted sample was applied to a DEAE-cellulose column (see Subheading “Sample Pretreatment Using a DEAECellulose Column”). Before the LC–MS analysis, the lipid solution was dried under nitrogen gas and redissolved in isopropanol/methanol/water (5/1/4 v/v/v) supplemented with 0.2% formic acid and 0.028% ammonia (abbreviated as IMW5 below).

3.2.2.1. Extraction with Bligh–Dyer Method

2. The extracted lipids from a mouse liver, in 100 mL of methanol/chloroform (1/1 v/v), were stored at –30°C. Before the LC–MS analysis, 10 mL of the extract was diluted with 90 mL of the IMW5 solvent. 3.2.2.2. Extraction with Butanol

1. For minor acidic PL analyses, A431 cells (3–6 × 106/100 mm plate) were scraped with 2 mL of ice-cold butanol, and 50 pmol each of inner standards [17:0 LPA and 14:0/14:0 PA etc.] were added. Then, 2 mL of water was added to each suspension and vigorously shacked for 3 min while cooling with ice. The following operations were done at room temperature. The sample was centrifuged at 1,000 × g for 5 min, and an upper layer was collected. After the addition of 1 mL of methanol and 1 mL of chloroform, the each sample was applied to a DEAE-cellulose column (see Subheading “Sample Pretreatment Using a DEAE-Cellulose Column”). 2. For total PL analyses, the butanol-extraction was performed as described above, and then the obtained butanol solution was dried under diminished pressure at 25°C, followed by re-dissolving with 0.5 mL of the IMW5 solvent.

3.2.3. Results Sample Pretreatment Using a DEAE-Cellulose Column

A 500-mL bed of DEAE-cellulose, packed in a Pasteur pipette, was sequentially washed with water, 1N HCl, water, 1N NaOH, water, methanol, acetic acid, methanol, methanol/chloroform (1/1 v/v), and chloroform. The sample (in chloroform) was



Qualitative and Quantitative Analyses of Phospholipids by LC–MS

299

applied to the DEAE-cellulose column (25). Stepwise elution was begun with chloroform (3 × 1 mL), to wash out neutral lipids. Among the bound lipids, the basic lipids, PC and PE, etc., were washed with chloroform/methanol (1/1 v/v) (3 × 1 mL). The acidic lipids, PA, PS, PI, PG, cardiolipin (CL), and their lyso-forms, etc., were then eluted with chloroform/methanol/ water/28% aqueous ammonium/acetic acid (200/100/10/6.7 v/v/v/v) (3 × 1 mL). After removing organic solvent from each eluate under nitrogen gas, 0.05 mL of water and 0.1 mL of isopropanol were added to about 0.05 mL of the residue. The fractionated samples were stored at –70°C when not in immediate use. 3.2.3.1. HPLC Mobile Phases 3.2.3.2. Mobile Phase A

1. Composition: isopropanol/methanol/water (5/1/4 v/v/v) supplemented with 0.2% formic acid, 0.028% ammonia, and 5 mM phosphoric acid. 2. Preparation: Mix 500 mL of isopropanol, 100 mL of methanol, and 400 mL of water. Add 2 mL of formic acid. Add 1 mL of 28% aqueous ammonium. Add 0.5 mL of 10 mM phosphoric acid aqueous solution. Degas by sonication for 5 min in a sonicator bath.

3.2.3.3. Mobile Phase B

1. Composition: Isopropanol supplemented with 0.2% formic acid and 0.028% ammonia. 2. Preparation: Mix 1 L of isopropanol, 2 mL of formic acid, and 1 mL of 28% aqueous ammonium. Degas by sonication for 5 min in a sonicator bath.

3.2.3.4. LC Separation Experiment

The LC–ESIMS/MS analysis was performed by using a LTQOrbitrap mass spectrometer. The LC system consisted of two Shimadzu LC-10Avp-micro pumps, a DGU-14AM degassing unit, an SCL-10Avp controller, and an HTC PAL autosampler equipped with a nonmetallic injector unit and a 20-mL PEEK tubing sample loop, in order to avoid the undesired interaction of PA with metallic materials. Lipids were separated on a Waters X-Bridge C18 (3.5 mm, 150 mm × 1.0 mm i.d.) column. The gradient elution program was indicated in Fig. 4. The flow rate was 20 mL/min, and the chromatography was performed at room temperature (25°C). The total chromatographic run time was 45 min. This LC system was controlled with the Chromeleon software (Dionex, Sunnyvale, CA, USA). In the autosampler system, the first rinse solvent, washing a portion of the injector, was isopropanol/water (1/1 v/v) supplemented with 0.1% formic acid, 0.014% ammonia, and 0.05% phosphoric acid, and the second one was methanol. After the injector was washed with these rinse solvents several times, the sample (typically 5 mL) was injected (see Note 3).

3.2.3.5. Mass Spectrometry Conditions

The ion spray voltage was set to −4.0 kV in the negative ion mode. The temperature of the heated capillary of the ESI

300

Nakanishi, Ogiso, and Taguchi 100

B%

80 60 40 20 0

0

10

20

30

40

50

Time (min) Fig. 4. Gradient elution programs of LC for the analysis of acidic PLs. (From Ref. 16 ).

source was raised to 300°C, to stabilize the spray of the acidic mobile phase. However, when analyzing heat-labile lipids, the temperature should be reduced. The other parameters were set according to the manufacturer’s recommendations. This MS system was controlled with the Xcalibur software. Typically, PLs were measured by a full scan within m/z 380–1,500 in the negative ion mode, using an Orbitrap FT-MS analyzer with a resolution of 60,000. The mass accuracy was always within 3 ppm. 3.2.3.6. Identification of Phospholipid Species

Each molecular species was identified by the exact mass value and the LC retention time. In addition, MS/MS measurements at a collision energy (CE) setting of 35% were performed with an LTQ-IT analyzer when needed for identifying them.

3.2.3.7. Total Phospholipid Analysis

These LC conditions successfully suppressed the peak tailings interfering with the analyses of both PA and PS. When a PL mixture extracted from mouse liver was analyzed by the method presented here, all of the detected PS species eluted with improved peak shapes (Fig. 5a). In addition, when a PL mixture extracted from mammalian cells was analyzed, some LPA and LPS species as well as the PA and PS species were detected without severe peak tailings (Fig. 5b). The molecular species of PA, PS, and their lyso-forms detected in the mammalian cells by the novel LC–MS method are listed in Table 2. Each molecular species was identified by the exact mass value and the LC retention time. In this analysis, almost all of the other PL classes, i.e., PC, SM, PE, PG, PI, LPC, LPE, and CL, were simultaneously measured.

3.2.3.8. Minor Acidic Phospholipid Analysis

A matrix effect, i.e., an ion-suppression effect, may occur when a total lipid extract from a complex biological sample is subjected to ESIMS analyses. This means that a large amount of coeluted matrix components reduces the detection



Qualitative and Quantitative Analyses of Phospholipids by LC–MS

a

301

b

TIC

TIC

18:0 LPS m/z 524.299

36:4 PS

m/z 782.498 16:0 LPA m/z 788.545

38:6 PS

m/z 806.498

38:4 PS

Relative Intensity (%)

Relative Intensity (%)

36:1 PS

m/z 409.236

18:0 LPA

34:1 PA

m/z 673.481

m/z 810.529 m/z 699.497

36:2 PA 40:6 PS

0

10

m/z 437.267

m/z 834.529 20

30

Time (min)

40

36:1 PA 0

10

20

m/z 701.528 30

40

Time (min)

Fig. 5. PA, PS, LPA, and LPS elution profiles obtained with the LC–MS conditions. (a) Several PS molecular species from mouse liver were examined by parent ion detection. (b) Several molecular species of PA, LPA, and LPS from mammalian cells were examined by parent ion detection. TIC, total ion chromatogram. (From Ref. 16 ).

capability for the analyte of interest. Furthermore, highly hydrophobic molecules may accumulate in a reverse phase column after repeated injections of complex lipid extracts, and may have harmful effects. In practice, the accumulated neutral lipids suppressed the detection of LPLs in our previous study on total lipids from liver. Relatively hydrophilic molecules, such as LPLs, are susceptible to a matrix effect. To overcome these undesired effects, a simple and rapid sample pretreatment procedure using solid-phase extraction, DEAE-cellulose, was introduced, which can separate the neutral, basic, and acidic PLs to minimize ion-suppression and other harmful effects. After the extract was divided into three fractions by the DEAE-cellulose pretreatment, the obtained acidic fraction was analyzed by the LC–MS method. As a result, detection of minor components, such as 20:1- and 22:4-LPS, 34:1-PA, and 38:1-PS in the mouse liver, was improved (data not shown) (see Note 4).

Table 2 Molecular species of PA, PS and their lyso-forms detected in biological samples

m/z

Molecular species

Retention time (min)

EL4 cells Rat liver (Total lipids) (Acidic lipids) Peak intensities (×104)

409.2361

LPA 16:0

7.9

8.9

1.2

435.2517

LPA 18:1

8.0

1.2

0.58

437.2673

LPA 18:0

9.5

19

1.5

524.2994

LPS 18:0

8.9

9.6

3.4

552.3307

LPS 20:0

9.5

n.d.

2.1

550.3151

LPS 20:1

8.1

n.d.

0.96

544.2681

LPS 20:4

6.8

n.d.

0.61

572.2994

LPS 22:4

7.2

n.d.

0.60

645.4501

PA 32:1

18.3

23

n.d.

673.4814

PA 34:1

20.9

56

3.3

675.4970

PA 34:0

24.1

38

n.d.

699.4970

PA 36:2

21.2

26

n.d.

701.5127

PA 36:1

24.0

44

n.d.

760.5134

PS 34:1

19.6

80

5.8

786.5291

PS 36:2

19.4

n.d.

7.1

788.5447

PS 36:1

22.4

640

66

806.4978

PS 38:6

16.8

32

51

810.5291

PS 38:4

19.5

390

480

816.5760

PS 38:1

25.1

37

5.2

830.4978

PS 40:8

15.4

n.d.

3.7

832.5134

PS 40:7

17.8

42

n.d.

834.5291

PS 40:6

19.3

6,500

170

836.5447

PS 40:5

21.2

820

11

838.5604

PS 40:4

22.5

190

27

840.5760

PS 40:3

23.3

n.d.

4.0

844.6073

PS 40:1

27.8

30

n.d.

870.6230

PS 42:2

27.7

39

n.d.

872.6386

PS 42:1

30.4

44

n.d.

900.6699

PS 44:1

32.8

4.3

n.d.

When the elution peak of a single charged anionic molecule was detected within 3 ppm accuracy of the theoretical m/z value, the intensity was shown. A total lipid extract from EL4 cells was subjected to LC–MS. A total lipid extract from rat liver was fractionated by DEAE-cellulose, and the obtained acidic lipid fraction was analyzed. n.d. not detected. For Ref. 16



Qualitative and Quantitative Analyses of Phospholipids by LC–MS

303

3.3. Specific Detection Method of PE-Plasmalogens Using LC–ESIMS/MS

In this study, we present a method for specific detection of PEplasmalogens using RPLC–ESIMS/MS (20–23). As a result of the collision-induced dissociation (CID) spectra of PE-plasmalogen [M + H]+, we identified the two prominent fragment ions; one that was characteristic of the sn-1 position and one that was characteristic of the sn-2 position (21). The analyses of molecular species of PE-plasmalogen were performed by precursor ion scanning using these specific fragments on a quadrupole linear ion trap mass spectrometer and the separation system was an ultra performance LC (UPLC) with a C18 (1.7 mm, 150 mm × 1.0 mm i.d.) column. Highly selective and sensitive analysis can be achieved by using this separation system for PLs.

3.3.1. Sample Procedure

1. The brains (1.3 g) from 8-weeks-old male C57BL/6J mice (n = 3) were homogenized with 15 mL of chloroform/methanol (1/2 v/v), and the total lipids were extracted using the Bligh– Dyer method (24). The extract lipids, in 10 mL of chloroform /methanol (1/2 v/v), were stored at –30°C when not in use. Before the LC–ESIMS/MS analysis, the sample was dried under a gentle stream of nitrogen gas and was redissolved in methanol.

3.3.1.1. Extraction with Bligh–Dyer Method

3.3.2. Results 3.3.2.1. HPLC and UPLC Mobile Phases 3.3.2.2. Mobile Phase A

1. Composition: Acetonitrile /methanol/water (2/2/1 v/v/v) supplemented with 0.1% formic acid, 0.028% ammonia. 2. Preparation: Mix 400 mL of acetonitrile, 400 mL of methanol, and 200 mL of water. Add 1 mL of formic acid. Add 1 mL of 28% aqueous ammonium. Degas by sonication for 5 min in a sonicator bath.

3.3.2.3. Mobile Phase B

1. Composition: Isopropanol supplemented with 0.1% formic acid and 0.028% ammonia. 2. Preparation: Mix 1 L of isopropanol, 1 mL of formic acid, and 1 mL of 28% aqueous ammonium. Degas by sonication for 5 min in a sonicator bath.

3.3.2.4. Flow Injection Experiment

The ESIMS analysis was performed using a 4000 Q-TRAP® with an UltiMate 3000 HPLC system combined with an HTC PAL autosampler. The samples (10 mL) were subjected directly to ESIMS analysis without LC separation by flow injection. The mobile phase composition was A/B = 1/1 v/v (see Subheading “HPLC and UPLC Mobile Phases”) at a flow rate of 10 mL/min.

3.3.2.5. LC Separation Experiment

The LC–ESIMS/MS analysis was performed by using a 4000 Q-TRAP ® with an ACQUITY Ultra Performance LC ®. The sample was followed by the ACQUITY UPLC™ BEH C18 (1.7 mm, 150 mm × 1.0 mm i.d.) column and then subjected directly to ESIMS/MS analysis. Samples (10 mL)

304

Nakanishi, Ogiso, and Taguchi

were simply injected by the autosampler, and the phospholipids were separated by a step gradient with mobile phases A:B ratios of 100:0 (0–5 min), 95:5 (5–20 min), 70:30 (20–21 min), 50:50 (21–45 min), 50:50 (45–84 min), 100:0 (84–85 min), and 100:0 (85–100 min) at a flow rate of 70 mL/min and a column temperature of 30°C. The specific detection was performed by precursor ion scanning. 3.3.2.6. Mass Spectrometry Conditions

MS/MS analyses were performed in positive ion mode with the following settings, ion spray voltage, 5,500 V; curtain (nitrogen), ten arbitrary units; and collision gas (nitrogen), “high”. Specific detection was performed by precursor ion scanning and neutral loss scanning with the following settings, scan range of the instrument, m/z 400–1,200; dwell times, 2 s; declustering potential, 60 V, and resolutions of Q1 and Q3, “unit.” Optimal conditions for CID were selected by individual fragments or neutral loss. Optimal conditions for the detection of appropriate precursor ions and neutral losses were obtained by MS/MS analyses of each PL preliminary class. Automatic and programmed scanning was performed for each class of PLs. The characteristic fragmentation patterns of individual PE-plasmalogen were determined by enhanced product ion (EPI) scanning. These EPI experiments were performed as continuous, not data-dependent, measurements for obtaining mass chromatograms of characteristic product ions with the following settings, CEs, 20–60 eV; scan range of the instrument, m/z 100–800; scan speed, 1,000 Th/s; Q0 trapping, “on.”; linear ion trap fill-time, 50 ms; declustering potential, 80 V; and resolution of Q1, “unit.”

3.3.2.7. MS/MS Analysis of PE-Plasmalogen

In positive ion mode with a triple quadrupole instrument, the PEs (acyl and alkenyl; plasmalogen) are detected as the [M + H]+ ions. Figure 6 shows the CID spectra of 18:1/18:1 PE and p18:0/20:4 PE obtained by collision of the corresponding [M + H]+ at m/z 744 and 752 (flow injection experiment). The CID spectra of the [M + H]+ of diacyl-PE typically results in neutral loss of the phosphoethanolamine head group (141 Da). In fact, the ion at m/z 603 was detected as the [M + H−141]+ resulting from 18:1/18:1 PE (Fig. 6a). However, the ion at m/ z 611 resulting from neutral loss of 141 Da of p18:0/20:4 PE was detected as minor peak. The CID spectra of p18:0/20:4 PE detected the m/z 392 and 361 as two prominent fragment ions (Fig. 6b). The detailed structure information of these two fragment ions was previously reported by Zemski Berry et al. (21). The m/z 392 and 361 peaks were characteristic of the fatty acid at the sn-1 and sn-2 positions, respectively. Their accurate mass values were obtained as the m/z 392.293 (C20H42O4P1N1) and 361.275 (C23H37O3) by LTQ-Orbitrap (data not shown). These characteristic fragment ions were dependent on the fatty acid pattern of the sn-1/2 positions. For example, PE-plasmalogens

p18:0/20:4 PE CID 35 eV

18:0/18:1 PE CID 30 eV

361.3

392.3

m/z

[M + H –141]+ 611.4

[M + H]+ 752.5

[M + H]+ 744.5

Qualitative and Quantitative Analyses of Phospholipids by LC–MS

Intensity (cps)

Fig. 6. Detection of characteristic fragment ions of diacyl- and 1-O-alk-1¢-enyl-2-acyl-PE by the MS/MS mode in the positive ion mode. The CID spectra of [M + H]+ of (a) 18:1/18:1 PE (m/z 744.5) was detected the m/z 603.4 [M + H−141]+ , whereas those in (b) p18:0/20:4 PE (m/z 752.5) was detected the m/z 392.3 and 361.3 as two prominent fragment ions rather than the m/z 611.4 [M + H−141]+.

b

a

[M + H –141]+ 603.4

305

306

Nakanishi, Ogiso, and Taguchi

with p16:0 and p18:1 at the sn-1 position decomposed into m/z 364 and 390, respectively, and those in 18:1 and 22:6 at the sn-2 position decomposed into m/z 339 and 385, respectively (data not shown). Hence, the precursor ion scanning of the characteristic fragment ions of the sn-1 position is more suitable for the specific detection of PE-plasmalogens than those in the sn-2 position, because it has the possibility of detecting the PLs containing the sn-2 fatty acids such as arachidonic acid and docosahexaenoic acid. This analytical method has the advantage of the fatty acid identification of the sn-2 position without MS/MS analysis. Thus, it is possible to detect and identify all of the PE-plasmalogens containing p16:0, p18:0, and p18:1 at the sn-1 position by a precursor ion scanning of m/z 364, 392, and 390, respectively (see Note 5). 3.3.2.8. Specific Detection of PE-Plasmalogen

To investigate whether the analytical method using these characteristic product ions was applicable to biological sample, we performed profile analysis of PE-plasmalogens using the mouse brain lipid extracts. In neutral scanning of 141 Da of PE specific detection method, diacyl-PE molecular species such as 16:0/22:6 (m/z 764), 18:0/20:4 (m/z 768), and 18:0/22:6 (m/z 792) were observed in main peaks, while plasmalogens were less effectively detected (Fig. 7a). However, by using precursor ion scanning of m/z 364, 392, and 390, the PE-plasmalogens molecular species were sensitively detected in mouse brain lipid extracts, these were mainly containing the 16:0, 18:1, 20:1, 20:4, 22:4, and 22:6 at the sn-2 position (Fig. 7b–d and Table 3). As a result, this analytical method was determined to be applicable to biological sample (see Note 6).

3.4. Quantitative Analysis or Profiling of Lipid Molecular Species

It is true that to obtain accurate quantitative data from mass analysis is very difficult. In the field of biological sciences, errors of less than 30%, absolute and less than 15%, relative are sufficient to recognize some specific changes in metabolic profiling. Thus, even simple profiling of lipid molecular species seems to be sufficiently useful to discover new candidates relating to the physiological phenomenon of interest. There are several factors which affect differences in ionization efficiency of PLs and glycerolipids. Ionic efficiencies mainly depend on each class of lipids and the experimental conditions. As is already well known, ionic efficiency largely depends on the structure of polar head groups within the molecules. The balance of polar groups and nonpolar groups is also an issue, especially at higher concentrations. In this chapter we typically used peak intensities of individual molecular species of phospholipids for quantitative profiling. In the new version of “Lipid Search,” quantitative data will mainly be obtained from peak areas of molecular weight-related ions. In this case, peak area data were subjected to compensation of

b

a

Pre 390

Pre 392

p16:0/16:0 676.5

Pre 364

NL 141

p18:1/16:0 702.5

p18:0/16:0 704.5

p16:0/18:1 702.5

p18:1/18:1 728.5

p18:0/18:1 730.5

18:0/18:1 746.5

p18:1/20:1 756.5

p18:1/20:4 750.5

p18:0/20:1 758.5

p18:0/20:4 752.5

m/z

18:0/20:4 768.5

18:0/22:6 792.5

p18:1/22:6 774.5

782. 6 p18:1/22:2

778. 6 p18:1/22:4

780. 6 p18:0/22:4

p18:0/22:6 776.5

752. 6 p16:0/22:4

p16:0/22:6 748. 5

p16:0/20:1 730. 5

p16:0/20:4 724. 5

16:0/18:1 718.5

16:0/20:4 740.5

16:0/22:6 764.5

22:6/22:6 836.6

Qualitative and Quantitative Analyses of Phospholipids by LC–MS

Fig. 7. Identification of individual molecular species of diacyl- and 1-O-alk-1¢-enyl-2-acyl-PE by neutral loss scanning and precursor ion scanning in the positive ion mode. The lipids extract from mouse brain was subjected to neutral loss scanning of 141 Da and precursor ion scanning of m/z 364, 392, and 390. (a) Neutral loss scanning of 141 Da is specific detection method for PE. Each precursor ion scanning of (b) m/z 364, (c) 392 and (d) 390 is corresponded to detection of PE-plasmalogens containing p16:0, p18:0 and p18:1 at the sn-1 position. NL, neutral loss scanning. Pre, precursor ion scanning.

d

c

Intensity (cps)

307

308

Nakanishi, Ogiso, and Taguchi

Table 3 Molecular species of PE identified from mouse brain by UPLC–ESIMS/MS

NL141

m/z

Molecular species

Intensity

716.5

34:2

987.6

718.5

34:1

3,942.2

720.5

34:0, p36:6

979.3

724.5

35:5, p36:4

462.6

726.5

35:4, p36:3

558.4

728.5

35:3, p36:2

1,004.3

730.5

35:2, p36:1

1,758.6

732.5

35:1, p36:0

950.1

738.5

36:5

996.0

740.5

36:4

4,992.3

742.5

36:3

1,437.7

744.5

36:2

4,996.5

746.5

36:1, p38:7

7,738.5

748.5

36:0, p38:6

2,283.6

750.5

37:6, p38:5

2,596.2

752.5

37:5, p38:4

1,208.5

754.5

37:4, p38:3

1,046.0

756.5

37:3, p38:2

645.9

758.5

37:2, p38:1

695.9

762.5

38:7

904.3

764.5

38:6

35,500.0

766.5

38:5

10,651.0

768.5

38:4

12,639.0

770.6

38:3

2,458.6

772.6

38:2, p40:8

1,066.8

774.6

38:1, p40:7

1,929.4

776.6

38:0, p40:6

1,879.4

778.6

39:6, p40:5

1,837.7

780.6

39:5, p40:4

804.3 (continued)



Qualitative and Quantitative Analyses of Phospholipids by LC–MS

309

Table 3 (continued)

Pre364

Pre392

m/z

Molecular species

Intensity

782.6

39:4, p40:3

788.6

39:1, 40:8

1,000.1

790.6

39:0, 40:7

9,555.4

792.6

40:6

31,383.0

794.6

40:5

4,892.3

796.6

40:4

3,729.6

798.6

40:3

591.7

812.6

41:3, 42:10

891.8

814.6

41:2, 42:9

1,787.7

816.6

41:1, 42:8

670.9

836.6

44:12

2,717.0

838.6

44:11

1,004.3

840.6

44:10

5,659.0

842.6

44:9

674.5

p16:0/16:1, (a16:1/16:1)

676.5

p16:0/16:0, (a16:1/16:0)

162.5

700.5

p16:0/18:2, (a16:1/18:2)

112.5

702.5

p16:0/18:1, (a16:1/18:1)

3,433.8

704.5

p16:0/18:0, (a16:1/18:0)

333.4

724.5

p16:0/20:4, (a16:1/20:4)

2,400.3

728.5

p16:0/20:2, (a16:1/20:2)

291.7

730.5

p16:0/20:1, (a16:1/20:1)

1,933.6

732.5

p16:0/20:0, (a16:1/20:0)

254.2

748.5

p16:0/22:6, (a16:1/22:6)

6,746.7

752.6

p16:0/22:4, (a16:1/22:4)

1,096.0

754.6

p16:0/22:3, (a16:1/22:3)

262.5

756.6

p16:0/22:2, (a16:1/22:2)

175.0

758.6

p16:0/22:1, (a16:1/22:1)

75.0

702.5

p18:0/16:1, (a18:1/16:1)

179.2

704.5

p18:0/16:0, (a18:1/16:0)

479.2

462.6

812.6 75.00

(continued)

310

Nakanishi, Ogiso, and Taguchi

Table 3 (continued)

Pre390

m/z

Molecular species

Intensity

728.5

p18:0/18:2, (a18:1/18:2)

187.5

730.5

p18:0/18:1, (a18:1/18:1)

5,671.5

732.5

p18:0/18:0, (a18:1/18:0)

450.1

752.5

p18:0/20:4, (a18:1/20:4)

3,904.7

756.5

p18:0/20:2, (a18:1/20:2)

354.2

758.5

p18:0/20:1, (a18:1/20:1)

2,279.5

776.5

p18:0/22:6, (a18:1/22:6)

8,059.3

780.6

p18:0/22:4, (a18:1/22:4)

2,917.0

782.6

p18:0/22:3, (a18:1/22:3)

366.7

784.6

p18:0/22:2, (a18:1/22:2)

183.4

786.6

p18:0/22:1, (a18:1/22:1)

262.5

804.6

p18:0/24:6, (a18:1/24:6)

120.8

808.6

p18:0/24:4, (a18:1/24:4)

175.0

700.5

p18:1/16:1, (a18:2/16:1)

254.2

702.5

p18:1/16:0, (a18:2/16:0)

1,075.1

726.5

p18:1/18:2, (a18:2/18:2)

95.8

728.5

p18:1/18:1, (a18:2/18:1)

5,713.2

730.5

p18:1/18:0, (a18:2/18:0)

525.1

750.5

p18:1/20:4, (a18:2/20:4)

3,158.7

754.5

p18:1/20:2, (a18:2/20:2)

416.7

756.5

p18:1/20:1, (a18:2/20:1)

2,850.4

774.5

p18:1/22:6, (a18:2/22:6)

3,579.6

776.5

p18:1/22:5, (a18:2/22:5)

470.9

778.5

p18:1/22:4, (a18:2/22:4)

1,016.8

780.6

p18:1/22:3, (a18:2/22:3)

341.7

782.6

p18:1/22:2, (a18:2/22:2)

1,366.8

784.6

p18:1/22:1, (a18:2/22:1)

358.4

p 1-O-alk-1¢-enyl aliphatic chain at the sn-1 position (plasmalogen), a 1-O-alkyl aliphatic chain at the sn-1 position, () minor molecular species



Qualitative and Quantitative Analyses of Phospholipids by LC–MS

311

the peak ratio of isotope patterns calculated by molecular elements obtained after automatic identification by “Lipid Search”. This compensation can also be automatically applied to the peak area obtained from mass spectrum analysis by precursor ion scanning and neutral loss scanning. These data will be combined with the identification data obtained from data-dependent MS/MS. Essentially for exact compensation of relative ionization efficiency in individual molecular species, use of individual standards containing stable isotopes are most accurate approach. But it is difficult both technically and economically, because more than several hundred molecules exist even in single LC–ESIMS/MS analysis for glyceroPLs or glycerolipids. Ionization efficiency is almost identical where molecules have the same polar head groups, such as those in the same class of PLs. We normally add diacyl12:0/12:0 PE as an internal standard, and separately compensate ionic efficiency in both negative and positive ion modes, by using the same amount of diacyl-16:0/16:0 or -16:0/18:1 species for each class of PLs (PC, PE, PS, PI, PG). After obtaining the target molecules, or where an exact target is to be quantified, the most accurate and commonly selected method of compensation uses a stable isotope labeled standard.

4. Notes 1. As Wakelam et al. described previously (26), use of silanized glassware was essential for handling of LPA and PA, especially their dilute solutions. 2. The standard Bligh–Dyer method is known to be somewhat ineffective for LPA extraction (27, 28). If a certain LPA molecular species must be analyzed, then an extraction procedure suitable for LPA should be chosen, such as the butanol extraction method. 3. When the PA peak tailing worsened after repetitive use, the column was detached from the MS instrument and was washed with methanol/water (4/1 v/v) containing 0.1% phosphoric acid, which allowed the peak shape to recover well. 4. Polyphosphoinositides were not detected by this method, because of unsuitable LC conditions. 5. 1-O-alkyl or 1-O-alk-1¢-enyl (plasmalogen) at the sn-1 position is indistinguishable by this method, because of producing the same characteristic product ions. 6. The characteristic product ions obtained in the CID spectra of PE-plasmalogens are valuable in identifying unique plasmalogen molecular species in biological samples.

312

Nakanishi, Ogiso, and Taguchi

Acknowlegment This study was performed with the help of Special Coordination Funds from the Ministry of Education, Culture, Sports, Science and Technology of the Japanese Government.

References 1. Kim, H. Y., Wang, T. L., and Ma, Y. C. (1994) Lipid chromatography/mass spectrometry of phospholipids using electrospray ionization. Anal. Chem. 66, 3977–3982. 2. Pulfer, M., and Murphy, R. C. (2003) Electrospray mass spectrometry of phospholipids. Mass. Spectrom. Rev. 22, 332–364. 3. Han, X., and Gross, R. W. (2005) Shotgun lipidomics: electrospray ionization mass spectrometric analysis and quantitation of cellular lipidomes directly from crude extracts of biological samples. Mass. Spectrom. Rev. 24, 367–412. 4. Schwudke, D., Liebisch, G., Herzog, R., Schmitz, G., and Shevchenko, A. (2007) Shotgun lipidomics by tandem mass spectrometry under data-dependent acquisition control. Methods Enzymol. 433, 175–191. 5. Taguchi, R., Nishijima, M., and Shimizu, T. (2007) Basic analytical systems for lipidomics by mass spectrometry in Japan. Methods Enzymol. 432, 185–211. 6. Ikeda, K., Shimizu, T., and Taguchi, R. (2008) Targeted analysis of ganglioside and sulfatide molecular species by LC/ESI-MS/MS with theoretically expanded multiple reaction monitoring. J. Lipid Res. 49, 2678–2689. 7. Houjou, T., Yamatani, K., Imagawa, M., Shimizu, T., and Taguchi, R. (2005) A shotgun tandem mass spectrometric analysis of phospholipids with normal-phase and/or reverse-phase liquid chromatography/electrospray ionization mass spectrometry. Rapid Commun. Mass Spectrom. 19, 654–666. 8. Zhao, C., Du, G., Skowronek, K., Frohman, M. A., and Bar-Sagi, D. (2007) Phospholipase D2-generated phosphatidic acid couples EGFR stimulation to Ras activation by Sos. Nat. Cell Biol. 9, 706–712. 9. Balasubramanian, K., Mirnikjoo, B., and Schroit, A. J. (2007) Regulated externalization of phosphatidylserine at the cell surface. J. Biol. Chem. 282, 18357–18364. 10. Bandoh, K., Aoki, J., Hosono, H., Kobayashi, S., Kobayashi, T., Murakami-Murofushi, K., Tsujimoto, M., Arai, H., and Inoue, K. (1999) Molecular cloning and characterization of a novel human g-protein-coupled

receptor, EDG7, for lysophosphatidic acid. J. Biol. Chem. 274, 27776–27785. 11. Parrill, A. L., Wang, D., Bautista, D. L., Van Brocklyn, J. R., Lorincz, Z., Fischer, D. J., Baker, D. L., Liliom, K., Spiegel, S., and Tigyi, G. (2000) Identification of Edg1 receptor residues that recognize sphingosine 1-phosphate. J. Biol. Chem. 275, 39379–39384. 12. Park, K. S., Lee, H., Kim, M., Shin, E. H., Jo, S. H., Kim, S. D., Im, D., and Bae, Y. (2006) Lysophosphatidylserine stimulates L2071 mouse fibroblast chemotactic migration via a process involving pertussis toxin-sensitive trimeric g-proteins. Mol. Pharmacol. 69, 1066–1073. 13. Kim, H., Wang, T. L., and Ma, Y. (1994) Lipid chromatography/mass spectrometry of phospholipids using electrospray ionization. Anal. Chem. 66, 3977–3982. 14. Taguchi, R., Houjou, T., Nakanishi, H., Yamazaki, T., Ishida, M., Imagawa, M., and Shimizu, T. (2005) Focused lipidomics by tandem mass spectrometry. J. Chromatogr. B. 823, 26–36. 15. Larsen, A., Mokastet, E., Lundanes, E., and Hvattum, E. (2002) Separation and identification of phosphatidylserine molecular species using reversed-phase high-performance liquid chromatography with evaporative light scattering and mass spectrometric detection. J. Chromatogr. B. 774, 115–120. 16. Ogiso, H., Suzuki, T., and Taguchi, R. (2008) Development of a reverse-phase liquid chromatography electrospray ionization mass spectrometry method for lipidomics, improving detection of phosphatidic acid and phosphatidylserine. Anal. Biochem. 375, 124–131. 17. Nagan, N., and Zoeller, R. A. (2001) Plasmalogens: biosynthesis and functions. Prog. Lipid Res. 40, 199–229. 18. Brites, P., Waterham, H. R., and Wanders, R. J. A. (2004) Functions and biosynthesis of plasmalogens in health and disease. Biochim. Biophys. Acta 1636, 219–231. 19. Han, X., Holtzman, D. M., and McKeel, D. W. Jr. (2001) Plasmalogen deficiency in early Alzheimer’s disease subjects and in animal models: molecular characterization using



Qualitative and Quantitative Analyses of Phospholipids by LC–MS

electrospray ionization mass spectrometry. J. Neurochem. 77, 1168–1180. 20. Kerwin, J. L., Tuininga, A. R., and Ericsson, L. H. (1994) Identification of molecular species of glycerophospholipids and sphingomyelin using electrospray mass spectrometry. J. Lipid Res. 35, 1102–14. 21. Zemski Berry, K. A., and Murphy, R. C. (2004) Electrospray ionization tandem mass spectrometry of glycerophosphoethanolamine plasmalogen phospholipids. J. Am. Soc. Mass Spectrom. 15, 1499–1508 22. Hsu, F. F., and Turk, J. (2007) Differentiation of 1-O-alk-1′-enyl-2-acyl and 1-O-alkyl-2-acylglycerophospholipidsbymultiplestage linear ion-trap mass spectrometry with electrospray ionization. J. Am. Soc. Mass Spectrom. 18,, 2065–73. 23. Yang, K., Zhao, Z., Gross, R. W., and Han, X. (2007) Shotgun lipidomics identifies a paired rule for the presence of isomeric ether phospholipid molecular species. PLoS ONE 2, e1368.

313

24. Bligh, E. G., and Dyer, W. J. (1959) A rapid method of total lipid extraction and purification. Can. J. Med. Sci. 37, 911–917. 25. Rouser, G., Kritchevsky, G., Yamamoto, A., Simon, G., Galli, C., and Bauman, A. J. (1969) Diethyaminoethyl and triethylaminoethyl cellulose column chromatographic procedures for phospholipids, glycolipids, and pigments. Methods Enzymol. 14, 273–317. 26. Wakelam, M. J. O., Pettitt, T. R., and Postle, A. D. (2007) Lipidomic analysis of signaling pathways. Methods Enzymol. 432, 233–246. 27. Xiao, Y., Chen, Y., Kennedy, A. W., Belinson, J., and Xu, Y. (2000) Evaluation of plasma lysophospholipids for diagnostic significance using electrospray ionization mass spectrometry analyses. Ann. N. Y. Acad. Sci. 905, 242–259. 28. Ishida, M., Imagawa, M., Shimizu, T., and Taguchi, R. (2005) Effective extraction and analysis for lysophosphatidic acids and their precursors in human plasma using electrospray ionization mass spectrometry. J. Mass Spectrom. Soc. Jpn. 54, 217–226.

sdfsdf

Chapter 16 Determination of Fatty Acid Profiles and TAGs in Vegetable Oils by MALDI-TOF/MS Fingerprinting Zeev Wiesman and Bishnu P. Chapagain Summary Vegetable oils are complex mixtures containing a wide range of major compounds. Triacylglycerols (TAGs; consisting of a glycerol moiety with each hydroxyl group esterified to a fatty acid) are the major components (95–98%) of vegetable oils. TAGs are an important source of energy and nutrition for humans, so their compositional analysis merits extensive interest. Analysis of TAGs has increased in recent years and the advancement has been driven by the development of analytical technologies. This chapter discusses techniques for determination of TAG and fatty acid profiles (FAPs) of vegetable oils using matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF/ MS) and gas chromatography–mass spectrometry (GC/MS). Considering the importance of TAG in its native form, rather than FAPs, special emphasis has been given to the TAG fingerprinting analyses of intact oils. MALDI-TOF/MS also enabled calculation of the main fatty acids and their compositions in a simple manner from the TAG profiles; the results are found to be very similar to the prevailing methods of derivatization using GC/MS. This study depicts the potential of MALDI-TOF/MS as an easy, fast, and reliable technique to characterize the TAG and FAPs in vegetable oils. Key words: Vegetable oils, Fatty acid profiles, Triacylglycerols, MALDI-TOF/MS, GC/MS

1. Introduction Vegetable oils (also known as plant oils), the main edible oils, are the most commonly consumed products in the food industries. Studies have shown that consumption of vegetable oils have a tremendous effect on human physiology, including lipid metabolism, development of chronic disease, and overall well being (1). Soybean, corn, and canola are the most commonly used vegetable oils, and they have their own importance, but more unique and healthier oils are always being sought. Olive oil produced from Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_16, © Humana Press, a part of Springer Science + Business Media, LLC 2009

315

316

Wiesman and Chapagain

the fruit of Olea europea, a high oleic acid content oil and one of the healthiest vegetable oils used these days, and pomegranate oil (PGO) produced from seeds of the Punica garanatum, a high linolenic acid content oil, recently became of high priority for health and cosmetic uses. These are the two vegetable oils we focus on in this chapter. There is no doubt that an effective technique for the determination of their qualities is the prerequisite for the use of any vegetable oils. Vegetable oils belong to the natural lipid group. Lipids comprise an enormous number of chemically distinct molecules arising from combinations of fatty acids with various backbone structures. Lipids are currently classified into eight categories – fatty acyls, glycerolipids, glycerophospholipids, sphingolipids, sterol lipids, prenol lipids, saccharolipids, and polyketides – containing distinct classes and subclasses of molecules (2); vegetable oils belong to the first category. The fatty acyl groups (whose main component is TAG) are primarily relevant in nutrition and for the storage of energy. 1 .1. TAGs in Vegetable Oils

TAG, which is composed of three esterified fatty acids with an attached glycerol backbone, is the main component of vegetable oils (95–98%) and characterizes the quality of the oil (3). It is well documented that TAGs are mainly affected by genetic factors delivered by the various oil crops and varieties (4); however, the environment, agrotechnology, extraction process, and other factors may also contribute to the quality of the oil, and are associated with TAGs (5). Another possible source of variability of the TAG composition in a specific vegetable oil is adulteration of the product – replacing high-cost ingredients with lower grade and cheaper constituents. For example, there is a growing tendency to adulterate virgin olive oil with different cheap seed oils, such as hazelnut and almond oils (6). Since fatty acids do not stand alone in oils, there is increasing interest in their arrangement within the oil, namely – the TAGs. It is suggested that the TAG structure, and not only its fatty acid profile (FAP), is of special importance regarding its physiological effect (7). Many studies concerning the metabolism of lipids and their effect within the human body emphasize the importance of the TAG structure–activity relationship (8–10). The molecular composition of a TAG mixture is typically very complex due to the combination of a variety of fatty acids, differing in their chain length, degree of unsaturation, and distribution between the sn-1, sn-2, and sn-3 position of glycerol backbone (11). Therefore, analytical methods enabling determination of both the FAP and TAG composition are greatly valued. Considering the importance of the TAG structure of oil in its native form, rather than only FAPs, in this chapter special emphasis has been given to the TAG fingerprinting of vegetable oils.



1 .2. Determination of TAG Profiles

Determination of Fatty Acid Profiles and TAGs

317

Different methods have been applied for the analysis of TAGs of vegetable oils. Classical chromatographic methods are the most commonly in use for this purpose, including gas–liquid chromatography (GLC), high-performance liquid chromatography (HPLC), thin-layer chromatography (TLC), and supercritical fluid chromatography (SFC) (12). Although these instruments are commonly found in many laboratories, these methods have some disadvantages, e.g., difficulties in eluting high molecular mass TAGs from HPLC columns and GLC detector response factors necessary for reliable quantification, especially for the analysis of polyunsaturated TAGs. Thus, attaining a complete separation of TAG species requires a combination of chromatographic methods, which requires a lengthy protocol (12). Furthermore, this type of analysis does not allow detection of the actual TAGs, only the total percentage of individual fatty acids can be determined. In these techniques, TAGs are transeterified to produce methyl esters prior to analysis because the esters are less polar than the corresponding fatty acids, and thus are more compatible with the various chromatographic systems (13). Enzymatic methods have been employed for quantitative analysis of TAGs; they are excellent for quantification of total concentration but give no information of the type or quantity of specific TAGs (14). Recently, soft ionization mass spectrometry (MS) techniques, such as atmospheric pressure chemical ionization (APCI), electrospray ionization (ESI), and matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF), have been applied to analyze nonvolatile TAGs (15) and have been found to require less time for chromatographic steps.

2. Materials 2.1. Equipments

1. Reflex IV- Matrix assisted laser desorption/ionization–time of flight mass spectrometry – MALDI-TOF/MS) (Bruker Daltonik GMBH, Bremen, Germany). 2. Pressurized solvent extraction (PSE) (One PSE, Applied Separations, Inc.). 3. Gas chromatography–mass spectrometry – GC–MS (Agilent 5975C, Agilent Technologies, Inc.).

2.2. Reagents

1. 2,5-dihydroxy benzoic acid (DHB). 2. Triolein. 3. Tripalmitin. 4. Methanol.

318

Wiesman and Chapagain

5. Hexane. 6. Heptane. 7. Potassium hydroxide. DHB (³99.5%), triolein (>99%), tripalmitin (~99%), and potassium hydroxide (reagent grade) all purchased from SigmaAldrich, Israel. Methanol, hexane, and heptane all were HPLC grade and obtained from Merck KGaA, Germany (methanol and hexane) and Tedia Company Inc., USA (hepatane).

3. Methods 3.1. Maldi-Tof/ms

Although MALDI-TOF/MS was originally used mainly for the analysis of proteins and peptides, synthetic polymers, and carbohydrates to some extent, recent studies have shown this tool to be applicable for the analysis of lipids (13, 16). MALDI is based on the utilization of an ultraviolet-absorbing matrix that is mixed with samples using, for example, DHB (2,5-dihydroxybenzoic acid). The matrix enables ionization of the sample when irradiated by a UV-laser. In most cases, the laser is not tunable and emits light at a fixed wavelength, generally 337 nm (N2 laser). In some cases, as in DNA analysis, infrared-absorbing matrices such as glycerol can also be used (17). Among the soft ionization MS, MALDI-TOF has become a powerful and important tool for the analysis of large biopolymers. It creates a minimal degree of analyte fragmentation and thus enables the examination of rather complex structures, rather than the most basic “building blocks” of the analyte (e.g., TAGs as opposed to fatty acids) (16). Lipids are, in fact, shown to be highly suitable for MALDI since both the sample and the most common matrices in use are soluble in organic solvents, forming a single organic phase. This creates an extremely homogeneous matrix/analyte mixture, which is an important criterion for credible MALDI results (18). There is no doubt that MALDI-TOF/MS is a fast and accurate method, not requiring any prior derivatization stage and supplying results within minutes (one sample can be analyzed in less than 1 min (16, 19). Furthermore, MALDITOF targets (sample plates), which look like common microtiter plates, allow the simultaneous (even after automation) analysis of many samples and are available from the majority of suppliers; 1 ng of oil sample is sufficient for analysis (16).

3 .2. TAG Analysis by MALDI-TOF/MS

To get proper results from TAG analysis of any vegetable oils using MALDI-TOF/MS, selection of the proper organic matrix, solvents, and conditions of the MS system is important. Previous studies have reported many matrices for TAG analyses, such as



Determination of Fatty Acid Profiles and TAGs

319

alpha-cyano cinnamic acid (CHCA), DHB, dithranol, and K4 [Fe(CN)6]/glycerol (20, 21) for lipid analyses. We have also tried many matrices using different conditions to get the best results, particularly for pomegranate and olive oil TAG analysis in our lab, and found a procedure with DHB as the matrix to be the most suitable for TAG profiling (19; Chapagain et al. 2008 manuscript in preparation), as discussed below (see Note 1). The oil sample is dissolved in hexane to 1 mg/ml. DHB is used as a matrix. MALDI matrix solution is prepared by dissolving DHB in 90% methanol to about 10 mg/ml. For MALDI analysis, small volumes of sample and matrix are mixed together in a ratio of 1:2 and 1 ml is then applied directly to a stainless steel MALDI target. Samples are analyzed on a Reflex IV MALDI-TOF mass spectrometer using 337 nm radiations from a nitrogen laser. An accelerating voltage of 20 kV is used. The spectra are recorded in reflectron mode within a mass range of 450–2,400 m/z as described by Lay et al. (22) and Kaufman and Wiesman (19). We also tried chloroform for sample dissolve but did not get good results (see Note 2). In order to get the theoretical molecular mass of the different TAG options some calculations are needed (22). For instance, to determine the TAGs of the olive oils, the four major fatty acids dominating the olive oil, i.e., palmitic acid (P, 16:0), stearic acid (S, 18:0), oleic acid (O, 18:1), and linoleic acid (L, 18:2), as previously described by Mannina et al. (23) are taken into consideration; linolenic acid (Ln, 18:3), palmitoleic acid (Po, 16:1), and other fatty acids are present to a smaller extent (usually less than 1%) and therefore not taken into account in the calculations. These fatty acid modes are also implied for the calculation of the TAGs ofPGO, which also consists of similar five major fatty acids (with the addition of Linolenic) (1, 8). The molecular mass of these fatty acids is calculated according to their chemical formulae. All the possible combinations for TAGs composing these fatty acids are put together and their molecular weight is calculated. The molecular weight of a sodium ion (Na+) is added to the TAG mass since this is the predominant ion in the DHB matrix used. It is assumed that during the ionization process this ion is added to the TAGs, in most cases one ion per TAG molecule. The calculation described above is displayed in the following equation: Mw (TAG) = Mw {Glycerol–3(OH)}  + Mw {(FA1 + FA2 + FA3) –3(H)} + Mw [Na+] (1) 3 .3. TAG Profiling of Olive Oils Determined by MALDI-TOF/MS 3.3.1. Olive Oil Sample

Olive oil produced from the fruit of Olea europae is considered one of the most common and healthiest edible vegetable oil; its quality is very unstable and depends on various conditions from variety to extraction to storage. It is estimated that olive oil represents around 15% of the total vegetable oil produced around

320

Wiesman and Chapagain

the world (24). Evidence has shown that the Mediterranean diet, in which olive oil is the main source of fat, has a beneficial effect on diseases associated with oxidative damage such as coronary heart diseases (25). One of the biological virtues of olive oil is due to its balanced fatty acid composition and antioxidant components. For these reasons the consumption of olive oil has increased in several non-Mediterranean countries, such as the US, Australia, and Japan, during last two decades (26). Since the plantation of olive trees is expanding into more arid lands, so for experimental purposes we have chosen olive samples from olive cultivars grown in the Ramat Negev Desert Agro Station located in the semiarid region of the southern Israel Negev Desert. Six of the most common olive cultivars, namely Arbequina, Barnea, Koroneiki, Leccino, Picual, and Souri, are used. The olive plants were grown in an intensive manner with medium saline irrigation water (EC 4.2 dS/m). All samples were analyzed within a month of the oil extraction and till that time samples were kept in a bottle under condition of low light and temperature in the lab (see Note 3). 3 .3.2. Possible TAG Composition

Based on the calculations as mentioned above, the possible TAG combinations and their expected molecular masses are presented in Table 1. The data in this table also show the number of carbons (carbon number, CN) in each TAG composition with the number of double bonds (DB) in the molecule and also equivalent chain numbers (ECN) which is equal to CN – (2 × DB) (3).

3 .3.3. TAG Fingerprinting by MALDI-TOF/MS

In a positive ion mode, as reported previously (20, 27), all the TAGs yielded the corresponding sodium metal adduction exclusively and no proton adducts were detected in our experiment. Small fragmented peaks of each prominent TAG have also been observed in MALDI experiments, which can be seen in the spectrum of the standard (a mix of triolein and tripalmitin; Fig. 1 used in the experiment. This type of phenomenon has also been reported in previous experiments (27). Some of the previous reports have mentioned the problem created by using a matrix, especially when using DHB while analyzing TAGs, and tried to avoid this problem by analyzing without a matrix (6, 28); however, our experiment showed better results when using a matrix (DHB) than without it (Fig. 2).

3 .3.4. Quantitative Information Olive Oil TAGs

To obtain the quantitative information, isotopic correction of the TAG peaks and background subtraction should also be considered because most large peak areas contain significant quantities of the M + 2 isotope, as shown in Fig. 1 (20) (see Note 4). Based on the calculation of molecular mass (Table 1), nine major TAG species were identified in all studied oil samples (Fig. 3, Table 2) using MALDI-TOF/MS. These identified TAGs



Determination of Fatty Acid Profiles and TAGs

321

Triolein[M+Na]+

Tripalmitin [M+Na]+

Fig. 1. Typical positive ion MALDI-TOF/MS spectrum of standard TAGs (triolein and tripalmitin mix). Two dominant peaks in the spectrum at m/z 907.6 and 829.7 are the sodiated molecular ion of the triolein and tripalmitin, respectively. The spectrum also shows two other small peaks resulting from the loss of sodium salt from the fatty acid (inset is the isotopic peak of tripalmitin).

were: PPL, PPO, LLP, POL, OOP, LLO, OOL, OOO, and OOS. In all olive cultivars the OOO (triolein) composition is found to be the most abundant, followed by the OOP (diolein-palmitoyl), OOL (dioleyol-linoleyol), and POL (pamitoyl-oleyol-linolenoyl) respectively. Both Barnea and the Koroneiki cultivars demonstrated a dominant OOO fraction (34.4% and 42.4%, respectively). The Arbequina cultivar showed quite different fractions, more distributed among the TAG compositions, both the OOO and OOP fractions being quite dominant (25.9.89% and 21.35%, respectively), followed by the OOL and POL compositions (16.23% and 14.41%, respectively). 3 .4. Major Olive Oil TAGs from MALDI-TOF/ MS and Other Conventional Methods

The determination of TAG profiles from vegetable oils is not a new approach. Many studies have previously been carried out regarding the identification of major TAGs in olive oils using various chromatographic technologies (29–31). The major TAG profiles of olive oil obtained using various methods are presented in Table 3 and compared with the TAGs obtained from our recent

Without matrix

With matrix

Fig. 2. Comparison of the MALDI-TOF/MS spectra of tripalmitin in positive ion mode with and without matrix (DHB).

Table 1 Calculated molecular mass of possible TAG compositions Massa

TAG

CN:DBb

ECNc

853.2

PLP

53:2

46

855.3

POP

53:1

48

877.3

LPL

55:5

52

879.3

POL

55:3

48

881.4

OOP

55:2

48

885.4

SPP

53:2

50

901.3

LLL

57:6

46

903.3

LOL

57:5

46

905.4

OOL

57:4

42

907.4

OOO

57:3

44

909.5

OOS

57:2

42

911.5

SOS

57:1

46

L linoleic acid, 18:2; O oleic acid, 18:1; P palmitic acid, 16:0; S stearic acid, 18:0 a  Masses are for sodium adduct ions b  CN:DB = Carbon number: number of double bonds c  ECN = Equivalent chain number which equals to CN – (2 × DB)

200000

OOO OOS

OOP

PPO

OOL

POL LLP

PPL

LLO

907 Arbequina

0 200000

909 Picual

905

0 200000

y

Intensity

0 200000

Barnea

903

0

Souri

200000 Leccino

0 200000 0 850

Koroneiki

860

870

890

880

900

910

920

m/z

Fig. 3. Positive ion mass spectra of various olive oil TAGs obtained from MALDI-TOF/MS. The upper spectrum shows the detail of one olive sample and the lower spectrum demonstrates a summarized spectrum of six different olive cultivars.

MALDI experiment. As variation in the TAG compositions among the cultivars in our experiment was observed when determined by MALDI (Table 2), an average value of all cultivars was used for comparison in this table. Variations can be seen in the TAG profiles determined by the different methods. There might be many reasons for the variation in the profile obtained from different techniques, but one major reason may be the different sample source. However, the main TAG, triolein (OOO), is found to be the highest level in all studies. In an earlier study, Jakab et al. (32) used a sophisticated mathematical model to interpret the MALDI-TOF mass spectra of different plant oils

324

Wiesman and Chapagain

Table 2 Relative TAG composition of different olive oils determined by MALDI-TOF/MS TAG composition (%) Cultivars

PPL

PPO

LLP

POL

OOP

LLO

OOL

OOO

OOS

Arbequina

2.96

5.78

4.98

14.41

21.35

7.88

16.23

25.89

0.52

Picual

3.67

4.56

4.84

13.57

24.37

4.53

11.22

30.74

2.50

Barnea

2.96

4.11

1.20

9.46

22.46

4.85

14.94

36.41

3.80

Souri

3.71

5.09

4.18

14.07

20.95

6.63

17.62

24.59

3.16

Leccino

3.87

4.83

4.42

13.17

23.22

4.97

14.68

29.13

1.70

Koroneiki

1.97

4.72

1.78

8.25

25.98

2.83

11.16

42.40

0.90

Table 3 Comparison of the major TAGs of olive oil obtained by various methods Methods Major TAGs

HPLC-RIa

HPLC-UVb

HPLC-ELSDc

GCd

MALDI–TOF/MS

OOL

15.7

6.8

11.6

2.4

12.8

PLO

8.8

5.9

2.3

9.5

10.9

OOO

31.7

43.5

60.7

36.7

27.7

POO

19.7

18.4

19.9

23.4

20.6

POP

3.7

0.7

4.9

4.6

SOO

3.7

3.3

5.7

1.9

5.1

 (30)  (29) c  (31) d  (33) a

b

and also found a similar phenomenon between the results from MALDI-TOF/MS and HPLC atmospheric pressure chemical ionization (APCI) MS. These similarities of these results indicate MALDI is a potential alternative analytical technique for analyzing the TAG profiles in vegetable oils. We should not forget that MALDI is the fastest and easiest to operate for determining intact TAGs, so it is more reliable than other methods that are generally carried out after derivatization.



Determination of Fatty Acid Profiles and TAGs

325

3 .5. TAG Fingerprinting of Pomegranate and Soybean Oil by MALDI-TOF/MS

A similar MALDI-TOF/MS procedure is employed to obtain fingerprinting of TAGs in pomegranate and soybean oils. The PGO is still not easily available in the market, so for the analysis we extracted the oil in our laboratory as described below; however the soybean oils were purchased from the local grocery store.

3.5.1. Pomegranate Oil

PGO, produced from the seeds of Punica granatum L., is a rare and novel oil. This oil is unknown to many people and has been studied to much lesser extent than other edible vegetable oils, but is known as an antioxidant and anticancer agent (34, 35). Most of the analytic studies regarding PGO present its FAP (1), which is characterized by a dominant percentage of punicic acid. Punicic acid is considered an anticancer agent, as demonstrated by its inhibition of human prostate cancer cell invasion (36). It has also been reported to reduce apolipoprotein B100 secretion, which is correlated with the incidence of coronary heart disease and atherosclerosis (37).

3 .5.2. Plant Material and Oil Extraction

Fully matured pomegranate fruits are selected and cut to release the grains from within the fruit’s compartments. The pulp is cleaned and juiced by hand, carefully washed, and then dried for 48 h in an oven heated to 70°C. The resulting dried seeds are used as raw material for the oil extraction. The PGO is extracted by PSE, using the onePSE machine (Applied Separation, Inc., Allentown, PA).

3 .5.3. TAG Profiling of Pomegranate and Soybean Oils

TAG fingerprinting of pomegranate and soybean oils are shown in Fig. 4, clearly demonstrating the different TAG fingerprints of these two oils (19). Since pomegranate and soybean oils have different FAPs and corresponding TAGs, MALDI enables differentiating fingerprinting. Furthermore, TAG fingerprinting by MALDI-TOF/MS can also easily identify adulteration of the vegetable oils. The difference in TAG on TAG fingerprinting of the pure soybean oil and a mixture with PGO (9:1 ratio) is clearly shown in Fig. 5. Even though the mixture is only one-to-ten, MALDI-TOF/MS easily shows the differences. Soybean oil does not contain linolenic acid (18:3), but PGO contains a high percentage (>80%), so the oil mixture shows the Linolenic–Linolenic–Linolenic (Ln–Ln–Ln) TAG at m/z 896.97, demonstrating an additional benefit of using MALDI in the vegetable oil industry. This not only helps to define the adulteration but also helps to upgrade the oil by creating different oil mixtures.

3 .5.4. MALDI Analysis for Fresh and Aged Oils

MALDI–-TOF/MS is not only found to be a useful tool to differentiate vegetable oils, but it is also helpful in differentiating fresh and aged oils from the same source. MALDI mass spectra

Fig. 4. TAG fingerprinting of pomegranate oil and soybean oil detected by MALDI-TOF/MS in positive ion mode. Reprinted with permission from Kaufman and Wiesman, 2007 (19).

Soybean oil

Pome-Soy oil (1:9)

Fig. 5. Comparison of the TAG region in the MALDI-TOF mass spectra from soybean oil and from pomegranate–soybean oil mixture in a ratio of 9:1. The arrow-markedpeaks in the bottom spectra clearly show the presence of TAG finger printing of Ln–Ln–Ln at m/z 896.97. Both experiments were carried out in positive ion mode.



Determination of Fatty Acid Profiles and TAGs

327

Fig. 6. MALDI mass spectra of fresh (a) and aged (b) allium seed oil. The aged oil shows significantly greater amounts of higher mass components indicating oxidation over time. Reprinted with permission from Asbury et al. (20).

of fresh (a) and aged (b) allium (Allium cepa) seed oil are shown in Fig. 6; the two spectra are found to be dramatically different. The aged seed oil shows a considerably higher mass peak than the fresh seed oil (20). If we look very clearly in the high mass peak, we can see an additional 16 mass units higher in the aged seed oil than the fresh one, indicating significant oxidation in the aging process, which is a common phenomenon in vegetable oils during storage. 3 .6. Fatty Acid Profiles of Vegetable Oils 3.6.1. Fatty Acid Profile

The FAP of vegetable oil is a measure of the proportions of individual fatty acids in the oil and is therefore an important part of the oil quality. The ratio of the different fatty acids in the vegetable oil influences the stability of the oil as well as determining the nutritional level of the oil. Some fatty acids are considered better

328

Wiesman and Chapagain

than others. In case of olive oil, oleic acid is the most desirable from the nutritional point of view. Vegetable oils that have high levels of monounsaturated oleic acid are considered of highest nutritive value (in fact oleic acid is named after olive – “olea”). Ideally the oil should also have less palmitic acid, which is the major saturated fat. The International Olive Council (IOC) has produced a list of the allowable levels for each of the fatty acids for an olive oil to be accepted as extra virgin olive oil (Table 4). Since the range is quite large, almost all olive oils will fit this guideline; however there are exceptions to this rule, especially for linolenic acid. According to this standard, linolenic acid should be less than 1%. Although linolenic acid is considered to be nutritionally beneficial, due to its polyunsaturated nature – having

Table 4 Allowable fatty acid ranges for different categories of olive oil by IOC and EC (38) Allowable range (%)

Fatty acid

Extra virgin olive oil Virgin olive oil

Olive oil and Refined olive oil

Olive–pomace oil Refined olive–pomace oil

Myristic (C14:0)

£0.05

£0.05

£0.05

Palmitic (C16:0)

7.5–20.0

7.5–20.0

7.5–25.0

Palmitoleic (C16:1)

0.3–3.5

0.3–3.5

0.3–3.5

Heptadecanoic (C17:0)

£0.3

£0.3

£0.3

Heptadecenoic (C17:1)

£0.3

£0.3

£0.3

Stearic (C18:0)

0.5–5.0

0.5–5.0

0.5–5.0

Oleic (C18:1)

55–83.0

55–83.0

55–83.0

Linoleic (C18:2)

3.5–21.0

3.5–21.0

3.5–21.0

Linolenic (C18:3)

£1.0a

£1.0

£1.0

Arachidic (C20:0)

<0.6

<0.6

<0.6

Gadoleic or eicosenoic (C20:1)

<0.4

<0.4

<0.4

Behenic (C22:0)

£0.2

£0.2

£0.3

Lignoceric (C24:0)

£0.2

£0.2

£0.2

C:181 T

0.0–0.05

0.0–0.20

0.0–0.40

C:182 T + C18:3 T

0.0–0.05

0.0–0.30

0.0–0.35

Trans fatty acids

EU-0.9

a



Determination of Fatty Acid Profiles and TAGs

329

three double bonds – it is the most reactive and is particularly unstable and susceptible to oxidation (rancidity). Oils containing greater levels of linolenic acid are likely to contribute to reduced storage stability of the oil. 3 .6.2. Determination of Fatty Acid Profiles by GC/MS

Using GC, the profile of fatty acids is measured as fatty acid methyl esters (FAME). FAME of the oil can be prepared as described by AOCS official methods (39). In the case of vegetable oils, mainly olive, our long lab experiences have shown a slight modification of AOCS methods that produced a better result, as described earlier by Kaufman and Wiesman (19). In brief, 30 ml of oil is put in an Eppendorf tube and diluted with 600 ml of heptane, adding 150 ml of 2N methanolic potassium hydroxide solution. The combined solution is shaken vigorously twice for 30 s and left to stratify until the upper solution is clear. This solution is collected and evaporated to dryness under N2 gas flow; the methyl esters are resuspended in 60 ml of heptane and left to stratify at −20°C for a few hours. A 1 ml of sample from the upper solution is then collected and injected into the GC for determination. The instrument used is a 6890N gas chromatograph (Agilent Technologies, Inc., Santa Clara, CA) equipped with a 30 m × 0.25 mm i.d., 0.25 mm, HP column. The olive oil sample was as mentioned previously for MALDI-TOF/MS with the following operating conditions: column temperature is increased from 50 to 186°C at a rate of 4°C/min, then from 186 to 210°C at 2°C/ min, and detector temperature is set to 300°C; linear velocity of the helium carrier gas is 30 cm/s. Fatty acid is identified both by retention time and mass spectra obtained with a 5973 Network Mass Selective Detector (Agilent) operating at 70 eV according to the GC–MS library. The percentage of each acid was calculated according to the following formula: % X = (area X × 100)/total area (weight percentage) as described earlier (19).

3 .6.3. Fatty Acid Profile of Pomegranate Oils

As mentioned previously, PGO obtained from the seed kernel of the Punica granatum is a unique and quite rare oil. Until now few studies have been carried out on PGO. Production of PGO is very limited as most pomegranates are used for fresh fruits. Recently we performed a comprehensive study of the fatty acid and TAG profile of the PGOs obtained, using the same procedure described previously for olive oil analysis from few Israeli commercial varieties and some recently selected new pomegranate varieties. The details of the FAPs are listed in Table 5 (19). As reported previously (40, 41), the most dominant fatty acid in all varieties of pomegranate seed oil tested was found to be linolenic acid (64–83%). Four isomers of linolenic acid (punicic acid) are clearly observed (one being larger and three smaller). The linolenic peaks obtained from the GC/MS were further analyzed by FAME and DMOX derivatives. Each sample

330

Wiesman and Chapagain

Table 5 Fatty acid profiles of various varieties of pomegranate oil determined by modified GC/MS method Oil type Fatty acids

PGO 1

PGO 2

PGO 3

PGO 4

8:00

–

–

–

0.02

14:00

0.02

0.02

0.05

0.03

15:00

0.02

0.02

0.03

0.03

15:01

–

–

–

–-

16:00

3.15

3.56

14.51

3.89

16:01

0.03

0.02

0.10

0.06

17:00

0.07

0.07

0.07

0.03

17:01

–

–

–

–

18:00

2.37

2.66

2.94

3.09

18:01

5.21

7.17

12.02

10.11

18:02

3.58

4.67

4.38

2.14

18:3, total

83.46

79.94

64.05

78.06

A

61.61

73.55

52.77

61.75

B

13.44

3.98

6.93

9.39

C

0.19

0.18

0.11

0.18

D

8.22

2.23

4.23

6.74

19:00

–

–

–

–

19:01

0.07

0.07

0.14

–

20:00

0.57

0.55

0.56

0.85

20:01

0.96

0.78

0.77

1.06

20:02

0.11

0.06

0.06

0.06

21:00

0.05

0.03

0.04

0.05

22:00

0.15

0.13

0.13

0.25

23:00

–

0.03

0.03

0.04

24:00

0.09

0.08

0.07

0.14

25:00

0.07

0.06

0.02

–

26:00

0.05

0.05

0.05

0.07

28:00

–

–

–

0.03



Determination of Fatty Acid Profiles and TAGs

331

confirmed the peak as linolenic derivates, all having a quite similar pattern and differing mainly in the intensity of the fragmentation on the peaks (Fig. 7). The MS experiment of this fatty acid showed the 2nd isomer to be conjugated. The major ion peaks in all FAME spectra were at 79, 91, 93 m/z and the molecular ion

Fig. 7. Linolenic acid isomers of pomegranate seed oil (a) MS of isomer A (Table 5) with FAME derivatization; (b) MS of the same isomer A with DMOX derivatization. The double bonds at positions 9, 11, and 13 are represented in peaks at m/z 196 (C8), m/z 208 (C9), m/z 222 (C9), m/z 234 (C11), m/z 248 (C12), and m/z 260 (C13). A difference of 12 mass units between Cn-1 and Cn implies a double bond at position Cn. Reprinted with permission from Kaufman and Wiesman (19).

332

Wiesman and Chapagain

at 292 m/z. These ion peaks are known to characterize the MS of a-eleostearic acid, and the 9-cis-, 11-trans-, 13-trans-CLNA isomers (42) 3 .7. Fatty Acid Profiles from the MALDI-TOF/ MS TAG Profiles and Comparison with GC/ MS Determination

From the TAG profiles of the MALDI-TOF/MS, profiles of each fatty acid can easily be calculated according to its relative part in each TAG and the abundance of the particular TAG; that can then be compared with the FAP determined by the GC. The FAP of the different cultivars of olive oils obtained from GC/MS and MALDI-TOT/MS, as previously described, are presented in Table 6. The results obtained from GC can be easily interpreted as the relative area of the fatty acid peaks; however, the results obtained from MALDI-TOF/MS need to be calculated. For example, the relative part of the palmitic acid (P) in the TAG ‘PPL’ is 2/3, and this TAG’s abundance in the Arbequina variety is 2.96% (Table 2). Hence, the contribution of this TAG to the P fraction in Arbequina will be 2/3 × 2.96%, which is 1.97%. Summing the contributions of the different TAGs results in the fraction of the specific fatty acid for a specific cultivar (Table 6). There are some minor differences observed in two results, namely FAPs obtained from GC/MS and MALDI-TOF/MS (Table 6). There might be a number of reasons for this. Firstly, during calculations for the possible TAG compositions only four major fatty acids of the olive oils were considered (see data processing above). This is because TAGs containing other,

Table 6 Comparison of fatty acid profiles of olive oil determined by the GC/MS and MALDI-TOF/MS methodologies subjected to different cultivars Fatty acid profile (%) Cultivars

Methods

Palmitic

Stearic

Oleic

Linoleic

Others

Arbequina

GC/MS MALDI-TOF/MS

18.16 19.41

1.35 0.17

59.28 60.64

17.92 19.78

3.29 –

Picual

GC/MS MALDI-TOF/MS

12.01 19.75

1.55 0.83

69.89 63.68

14.31 15.74

2.24 –

Barnea

GC/MS MALDI-TOF/MS

13.16 15.76

2.57 1.27

66.17 70.02

14.34 13.16

3.76 –

Souri

GC/MS MALDI-TOF/MS

16.04 18.93

2.06 1.05

62.73 61.01

16.84 19.01

2.33 –

Leccino

GC/MS MALDI-TOF/MS

16.1 19.41

3.38 0.57

64.58 63.19

13.67 16.84

2.27 –

Koroneiki

GC/MS MALDI-TOF/MS

14.99 16.46

2.2 0.30

67.08 73.03

11.73 10.20

3.91 –



Determination of Fatty Acid Profiles and TAGs

333

less abundant, fatty acids were either not identified or difficult to recognize, so there were no fatty acids defined as “others”, as in the GC results. Secondly, some MALDI peaks with a particular mass can be attributed to a few different TAG compositions. In these cases, choosing the specific composition and identifying the peak was done according to the most probable combination of fatty acids (the probability was calculated according to the abundance of each fatty acid). The composition chosen was in all cases more probable than the others by an order of at least 102. As a result of this process, some TAG compositions that might exist (it is difficult to identify whether a few peaks with the same mass have combined) at a very small level cannot be identified, and thus the fatty acids composing them will also be disregarded. It is important to mention that, in this case, these possibly “neglected” compositions were usually stearic acid (S)-containing and therefore might explain the differences for this fatty acid in particular. The differences mentioned above are known to occur when comparing GC to MALDI FAPs (43). 3.8. Conclusions

There is no doubt that the recent and rapid developments in the area of mass spectrometry and chromatography have led to significant advances within the field of lipid research. In this context, the present study clearly showed that MALDI-TOF/ MS enables rapid fingerprinting of TAG profiles in vegetable oils. Furthermore, from the TAG profiles obtained by using MALDITOF/MS, an easy calculation of the relative FAPs of the vegetable oils is possible and the results are within the agreement of the other conventional methods. Since both TAG and FAPs are very valuable in detecting oil quality and, even more importantly, detecting adulteration, this technique can be used quickly on an industrial level. Using this technique one can easily detect oil quality in a rather fast and more reliable manner. However, because of the interference of the matrix, MALDI-TOF/MS cannot be applied easily with low molecular weight compounds (<400 Da). Thus free fatty acids and monoacylglycerols that are also present in small quantities in the commercially available oils cannot be determined by this technique in its present form, further development is needed.

4. Notes 1. For MALDI-TOF–MS analysis selection of matrix is very important. In our experiment for olive and PGO, alpha-cyano cinnamic acid (CHCA), and dithranol, did not produce spectra and DHB was found the best.

334

Wiesman and Chapagain

2. The solvent used for dissolving oil for TAG analysis is also very important. We observed very poor results when we used chloroform as a dissolving solvent and hexane was found appropriate. 3. Vegetable oil is a complex compound and its composition is dependent not only on the species and extraction procedure, but also on the mode of storage and the age of oil, so these factors should always be taken under consideration. For the oil quality including TAG, oil sample should be analyzed as fresh as possible. 4. Because of the isotopic influence of the TAG peak when analyzed by MALDI-TOF–MS, for quantitative analysis it is very important to make isotopic correction of the peak.

Acknowledgments The authors would like to thank the Ramat Negev Desert Agro Station for providing the olive oil samples. Our appreciation to Dr. Mark Karpasas, Dr. Leonid Kagon, Ms Maya Kaufman, and Ms Shosh Avni for technical help, and Ms Edna Oxman in editing this manuscript.

References 1. Fadavi, A., Barzegar, M. and Azizi, M. H. (2006) Determination of fatty acids and total lipid content in oilseed of 25 pomegranates varieties grown in Iran. J Food Compos Anal 19, 676–80. 2. Fahy, E., Subramaniam, S., Brown, H. A., Glass, C. K., Merrill, A. H. Jr., Murphy, R. C., Raetz, C. R. H., Russell, D. W., Seyama, Y., Shaw, W., Shimizu, T., Spener, F., van Meer, G., VanNieuwenhze, M. S., White, S. H., Witztum, J. L. and Dennis, E. A. (2005) A comprehensive classification system for lipids. J Lipid Res 46, 839–62. 3. Kiritsakis, A. K. (1998) Olive Oil from the Tree to the Table 2nd Edition. Food & Nutrition Press Inc., Trumbull, CT. 4. Foutanzza, G., Patumi, M., Solinas, M. and Serraiocco, A. (1994) Influence of cultivars on the composition and quality of olive oil. Acta Hort 356, 358–61. 5. Wiesman, Z., Itzhak, D. and Ben-Dom, N. (2004) Optimization of saline water for sustainable Barnea olive and oil production in desert conditions. Sci Horti 100, 257–66.

6. Calvano, C. D., Palmisano, F. and Zambonin, C. G. (2005) Laser desorption/ionization time-of-flight mass spectrometry of triacylglycerols in oils. Rapid Commun Mass Spectrom 19, 1315–20. 7. Mu, H. and Porsgaard, T. (2005) The metabolism of structured triacylglycerols (Review). Prog Lipid Res 44, 430–48. 8. Suzuki, R., Noguchi, R., Ota, T., Abe, M., Miyashita, K. and Kawada, T. (2001) Cytotoxic effect of conjugated trienoic fatty acids in mouse tumor and human monocytic leukemia cells. Lipids 36, 477–82. 9. Kritchevsky, D. (1995) Fatty acids, triglyceride structure, and lipid metabolism. J Nutr Biochem 6, 172–78. 10. Bracco, U. (1994) Effect of triglyceride structure on fat absorption. Am J Clin Nutr 60, 1002S–9S. 11. Laakso, P. (2002) Mass spectrometry of triacylglycrols. Eur J Lipid Sci Technol 104, 43–49. 12. Buchgraber, M., Ulberth, F., Emons, H. and Anklam, E. (2004) Triacyglycerol profiling



Determination of Fatty Acid Profiles and TAGs

by using chromatographic techniques. Eur J Lipid Sci Technol 106, 21–48. 13. Ayorinde, F. O., Garvin, K. and Saeed, K. (2000) Determination of the fatty acid composition of saponified vegetable oils using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Rapid Commun Mass Spectrom 14, 608–15. 14. Danno, H., Jinco, Y., Budiyanto, S., Furkawa, Y. and Kimura, S. (1992) A simple enzymatic quantitative analysis of triglycerides in tissues. J Nutr Sci Viaminol 38, 517–21. 15. Stübiger, G., Pittenauer, E. and Allmaier, G. (2003) Characterisation of castor oil by online and off-line non-aqueous reverse-phase high performance liquid chromatography– mass spectrometry (APCI and UV/MALDI). Phytochem Anal 14, 337–46. 16. Schiller, J. and Arnold, K. (2000) Mass Spectrometry in Structural Biology. In: Meyers RA, ed. Encyclopedia of Analytical Chemistry. New York: Wiley, 559–85. 17. Nordhoff, E., Kirpekar, F., Karas, M., Cramer, R., Hahner, S. and Hillenkamp, F. (1994) Comparison of IR- and UV-matrix-assisted laser desorption/ionization mass spectrometry of oligodeoxynucleotides. Nucl Acids Res 22, 2460–65. 18. Schiller, J., Arnold, J., Benard, S., Müller, M., Reichl, S. and Arnold, K. (1999) Lipid analysis by matrix-assisted laser desorption and ionization mass spectrometry: A methodological approach. Anal Biochem 267, 46–56. 19. Kaufman, M. and Wiesman, Z. (2007) Pomegranate oil analysis with emphasis on MALDITOF/MS triacylaglycerol finger printing. J Agric Food Chem 55, 10405–13. 20. Asbury, G. R., Al-Saad, K., Siems, W. F., Hannan, R. M. and Hill, H. H. (1999) Analysis of triacylglycerols and whole oils by matrixassisted laser desorption/ionization time of flight mass spectrometry. J Am Soc Mass Spectrom 10, 983–91. 21. Zollner, P., Schmid, E. R. and Allmaier, G. (1996) K4[Fe(CN)6]/Glycerol – A new liquid matrix system for matrix-assisted laser desorption/ionization mass spectrometry of hydrophobic compounds. Rapid Commun Mass Spectrom 10, 1278–82. 22. Lay, J. O. Jr., Liyanage, R., Durham, B. and Brooks, J. (2006) Rapid characterization of edible oils by direct matrix assisted laser desorption/ionization time-of-flight mass spectrometry analysis using triacylglycerols. Rapid Commun Mass Spectrom 20, 952–58. 23. Mannina, L., Dugo, G., Salvo, F., Cicero, L., Ansanelli, G., Clcagni, C. and Segre, A. (2003) Study of the cultivare-compostionship

335

in siciian olive oils by GC, NMR and statistical methods. J Agric Food Chem 51, 120–27. 24. Luchetti, F. (2002) Importance and future of olive oil in the world market-an introduction to olive oil. Eur J Lipid Sci Technol 104, 559–63. 25. Renaud, S., de Lorgeril, M., Delaye, M., Guidollet, J., Jacquard, F., Mamelle, N., Martin, J. L., Monjaud, I., Salen, P. and Toubol, P. (1995) Cretan Mediterranean diet for prevention of coronary heart disease. Am J Clin Nutr 61, 1360S–67S. 26. IOOC (International Olive Oil Council). (1995) Trade standard applying to olive oil and olivepomace oil. COI/T.15/NC no. 2/Rev. 2. 27. Schiller, J. (2007) MALDI MS of Lipids. In: Hellenkamp F, Preter-Katalinic J eds. MALDI MS a Practical Guide to Instrumentation, Methods and Applications. Weinheim: WileyVCH, Verlag GmbH & Co. kGaA, 215–43. 28. Calvano, C. S., Aresta, A., Palmisano, F. and Zambonin, C. G. (2007) A laser desorption inonization time-of-flight mass spectrometry investigation into triacylglycerols oxidation during thermal stressing of edible oils. Anal Bioanal Chem 389, 2075–84. 29. Fedeli, E. (1977) Lipids of olives. Prog Chem Fats Lipids 15, 57–74. 30. Kiritsakis, A., Kanavouras, A. and Kiritsakis, K. (2002) Chemical analysis, quality control and packaging issues of olive oil. Eur J Lipid Sci Technol 104, 628–38. 31. Cunha, S. C. and Oliveira, M. B. P. (2006) Discrimination of vegetable oils by triacylglycerols evaluation of profile using HPLC/ ELSD. Food Chem 95, 518–24. 32. Jakab, A., Nagy, K., Keberger, K., Vekey, K. and Forgacs, E. (2002) Differentiation of vegetable oils by mass spectrometry combined with statistical analysis. Rapid Commun Mass Spectrom 16, 2291–97. 33. Rezanka, T., and Rezankova, H. (1999) Characterization of fatty acids and triacylglycerols in vegetable oils by gas chromatography and statistical analysis. Anal Chim Acta 398, 253–61. 34. Schubert, S. Y., Lansky, E. P. and Neeman, I. (1999) Antioxidant and eicosanoid enzyme inhibition properties of pomegranate seed oil and fermented juice flavonoids. J Ethnopharmacol 66, 7–11. 35. Kohno, H., Suzuki, R., Yasui, Y., Hosokawa, M., Miyashita, K. and Tanaka, T. (2004) Pomegranate seed oil rich in conjugated linolenic acid suppresses chemically induced colon carcinogenesis in rats. Cancer Sci 95, 481–86. 36. Lansky, E. P., Harrison, G., Froom, P. and Jiang, W. G. (2005) Pomegranate (Punica granatum)

336

Wiesman and Chapagain

pure chemicals show possible synergistic inhibition of human PC-3 prostate cancer cell invasion across Matrigel™. Invest New Drugs 23, 121–22. 37. Arao, K., Yotsumoto, H., Seo-Young, H., Koji, N. and Teruyoshi, Y. (2004) The 9-cis, 11-trans, 13-cis isomer of cojugated linolenic acid reduces apolipoprotein B100 secretion and triacylglycerol synthesis in HepG2 cells. Biosci Biotechnol Biochem 68, 2634–45. 38. IOOC (International Olive Oil Council) (2003) Trade standard applying to olive oil and olivepomace oil. COI/T.15/NC no. 3/Rev. 2. 39. AOCS Official Methods (1997) Cd 1–25, Cd 3–25, Ce 2–66. 40. Takagi, T. and Itabashi, Y. (1981) Occurrence of mixtures of geometrical isomers of conjugated octadecatrienoic acids in some seed oils:

analysis by open-tubular gas liquid chromatography and high performance liquid chromatography. Lipids 16, 546–51. 41. Ozugul-Yucel, S. (2005) Determination of conjugated linolenic acid content of selected oil seeds frown in Turkey. J Am Oil Chem Soc 82, 893–897. 42. Christie, W. W. (2007) Lipid Library http:// www.lipidlibrary.co.uk/ms/arch_me/me_ pufa/M)6779.htm 43. Guyon, F., Absalon, C., Eloy, A., Salagoity, M. H. and Esclapez, M. (2003) Comparative study of matrix-assisted laser desorption/ ionization and gas chromatography for quantitative determination of cocoa butter and cocoa butter equivalent triacylglycerol composition. Rapid Commun Mass Spectrom 17, 2317–2332.

Chapter 17 Dynamics of Adipose Tissue Development by 2H2O Labeling Etienne Pouteau, Carine Beysen, Nabil Saad, and Scott Turner Summary Adipose tissue development undergoes remodeling in terms of newly synthesized cells (hyperplasia) and newly synthesized lipids that accumulate in adipocytes (hypertrophy). Synthesis and/or breakdown rates of adipose cells and lipids follow a continuous and dynamic pattern, e.g., during obesity development. This chapter describes a unique in vivo method to measure the dynamics of adipose tissue growth using 2 H2O labeling and mass spectrometry analyses. The approach uses 2H2O as a metabolic tracer to label the adipose tissue components such as the triglycerides (TG), the fatty acids, and the genomic DNA. Deuterium from 2H2O incorporates in the C–H bonds of glycerol moiety of TG through glyceroneogenesis as well as in palmitate moiety through de novo lipogenesis (DNL). Deuterium also incorporates into DNA through the de novo nucleoside synthesis pathway. The labeled water, 2H2O, is administrated intraperitoneally and/or orally in rodents or in humans for a defined duration and biopsies are collected at the end of the labeling period. We describe the procedure to extract, isolate, and purify the adipose components (TG-glycerol, TG-palmitate, and genomic DNA) and the derivation procedure to analyze the isotopic 2H-enrichment of these components by gas chromatography/mass spectrometry. The calculation principles are described to obtain the fractional and absolute synthesis rates of TG, of DNL, and of DNA measured in the adipose tissues. The method is nonradioactive, nonhazardous, accurate, reproducible, and very sensitive. We present recent in vivo data on the ontogeny of adipose tissue growth dynamics in young and adult obese Zucker rats compared with lean Zucker rats. Key words: 2H2O labeling, Deuterium, DNA, Adipogenesis, Lipogenesis, MIDA

1. Introduction All disorders of adipose tissue mass, including obesity and lipodystrophy, have an underlying kinetic basis. Synthesis and/or breakdown rates of adipose tissue cells and lipids must be altered for changes in adipose mass to develop and be maintained. For example, the formation of new adipocytes (hyperplasia) and an increase in adipocyte size (hypertrophy) by lipid synthesis and Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_17, © Humana Press, a part of Springer Science + Business Media, LLC 2009

337

338

Pouteau et al.

accumulation occurs during obesity development. What is less well understood is the relative importance of increased synthesis and reduced breakdown of lipid and cellular components within adipose tissue. Direct measurement of the dynamics of these components has historically been problematic, largely because of the large pool size of body fat and the apparently very slow turnover rates of all of the elements (1, 2). This combination has made direct labeling techniques generally impractical, because tracer incorporation is too low to be measured reliably when most labeling protocols are used. Stable isotope-based methods can be used for indirect in vivo labeling for the simultaneous measurement of multiple parameters of adipose tissue metabolism. Specifically, rates of whole body DNL (fractional and absolute DNL), adipose tissue TG synthesis rates, and adipose cell proliferation (vascular stromal and adipocytes) can be measured concurrently from a single adipose tissue sample from animals (or humans) labeled with 2H2O (deuterated water, heavy water) (1, 3–5). In the present chapter, we describe currently used stable isotope methodologies for the in vivo measurement of adipocyte proliferation rates, TG synthesis, and DNL using 2H2O as a tracer and mass spectrometric analysis. As an application, we present the determination of the ontogeny of adipose tissue growth dynamics by using 2H2O labeling in very young and adult obese (fa/fa) Zucker rats (6). 1 .1. Measurement of Adipose TG Synthesis

The central component of all TG is the a-glycerol moiety; therefore, TG-glycerol kinetics in principle best represents the kinetics of the intact TG molecule. However, labeling of adipose tissue TG-glycerol has been problematic. Cytosolic a-glycerol phosphate is the immediate precursor for TG synthesis. Adipocytes lack significant glycerol kinase activity, and as a result labeled-free glycerol is not directly incorporated into TG in adipose tissue. [13C]- or [14C]glucose does label adipose TG but is an inefficient precursor and is expensive, especially for the long-term studies needed to achieve measurable labeling of the very large, slowturnover, whole body adipose TG pools. The use of 2H2O as a metabolic tracer has numerous advantages over other tracers. 2H2O equilibrates quickly with the total body water (TBW) pool; it is relatively inexpensive and can be administered over long periods of time. The basic principle of the 2H2O-labeling method is that TG synthesized from intracellular a-glycerol phosphate during a period of 2H2O exposure will contain labeled hydrogen atoms incorporated from tissue water. In contrast, TG that were synthesized from a-glycerol phosphate before labeled water was present will not contain covalent C–H label in their glycerol moiety, thereby allowing the proportion of newly synthesized vs. preexisting TG molecules to be calculated (7). The biochemistry of hydrogen atom incorporation from



Dynamics of Adipose Tissue Development by 2H2O Labeling 1 H 2 H CO 3 H CO 4 H CO 5H

Glycolysis

H C= O H COH H CO~ P H

H+

Glyceraldehyde 3-P

H H COH C= O H CO~ P H

NADH + H+

NAD +

Dihydroxyacetone phosphate

Triose Phosphate

339

Triacylglycerol

H H COH H COH H CO ~ P H

α-Glycerol Phosphate

Gluconeogenesis Fig. 1. Incorporation of 2H from 2H2O into TG glycerol during TG synthesis.

solvent H2O into glyceroneogenic, gluconeogenic, and glycolytic intermediates has been previously characterized in detail (1). Importantly, C–H bonds in metabolic intermediates such as a-glycerol phosphate do not exchange hydrogen with tissue water outside of enzyme-catalyzed metabolic reactions (i.e., they are not labile in solution). The same principle applies for the C–H bonds in the glycerol moiety of TG. Accordingly, any incorporation of 2H into C–H bonds of glycerol in TG requires passage of the glycerol moiety through intermediary metabolic reactions during the period of 2H2O labeling and thereby implies new assembly of the TG from a-glycerol phosphate. This is shown in Fig. 1. Quantitatively, the degree of incorporation of 2H into C–H bonds of a-glycerol phosphate is proportional to two factors: the fraction of labeled hydrogen atoms in tissue water (i.e., the enrichment of tissue in 2H2O) and the fraction of C–H bonds in glycerol that are incorporated from body water. Thus, if the body water enrichment and the number of C–H bonds in glycerol that derive from body water are known, the incorporation of 2H into H atoms in the glycerol moiety of TG reveals the fraction of newly synthesized TG molecules present, regardless of the metabolic source of the fatty acid moiety. 1 .2. Measurement of All Source De Novo Lipogenesis

Deuterated water in combination with mass isotopomer distribution analyses (MIDA) has been used extensively to measure DNL in rodents (5, 8, 9) and humans (4, 10). Fractional synthesis rates (FSR) of fatty acids (% DNL) are calculated from the incorporation of 2H into palmitate. Fractional DNL represents the

340

Pouteau et al.

percent of fatty acid synthesized via the DNL pathway during the labeling period relative to all other TG-fatty acid. This method, because of the long labeling periods, determines all source DNL, i.e., both hepatic and adipose synthesis. Shorter labeling periods or alternative labels (e.g., 13C acetate) can be utilized to measure liver DNL separately (11, 12). The present approach underestimates the contribution of DNL to newly synthesized TG, however, to the extent that pre-existing fat is present. “Non-DNL” TG could represent either pre-existing TG or newly synthesized TG from non-DNL pathways. This problem can be resolved by correcting DNL for the proportion of TG that was newly synthesized during the labeling period. The ratio of fractional DNL to fractional TG synthesis reveals the true contribution of DNL to new TG-palmitate stored (5). The absolute contribution of DNL to stored adipose lipid (e.g., mg or mmoles palmitate) can be calculated by multiplying the fractional palmitate synthesis rate by the pool size (absolute amount of palmitate in adipose TG). This value does not represent the net flux rate since any new fatty acids which are broken down are not measured; however the contribution of DNL to stored lipid over the labeling period is quantified. 1 .3. Measurement of Cell Division (DNA Replication) in Adipose Tissue

Cell proliferation is involved at several steps during the differentiation sequence of mature adipocytes (13). The most direct measurement and characteristic feature of cell division is DNA replication. Thus, the production of mature adipocytes can, in principle, be measured by the fraction of mature adipocytes with newly synthesized DNA. Deuterated water-labeling technique can also be applied to measuring DNA synthesis, and thus cell proliferation, in living organisms (3, 4, 14). To measure cell proliferation, the deoxyribose (dR) moiety of newly formed deoxyribonucleoside triphosphate (dNTPs) is labeled with deuterium, (Fig. 2), and the deuterium content of dNTPs in purified DNA is measured by mass spectrometry. The incorporation of deuterium-labeled dNTPs into replicating DNA reveals the fraction of genomic DNA that was newly synthesized during the period of exposure to label. Purine deoxyribonucleosides, such as deoxyadenosine (dA), have the advantage that the de novo nucleoside synthesis pathway contribution is high and relatively constant, whereas the nucleoside salvage pathway contribution is lower and more variable. The method described determines cell proliferation from the 2H incorporation into purine-derived deoxyribose. This stable isotope metabolic-labeling approach has several technical advantages over traditional DNA-labeling methods, such as 3H-thymidine or bromodeoxyuridine. It is more precise, not operator dependant, nontoxic, nonradioactive and the in vivo and analytical execution are easy to perform and high throughput.



Dynamics of Adipose Tissue Development by 2H2O Labeling

2H O 2

Glycogen 2H O 2

Glucose

G6P

Base Salvage

GNG 2H O 2

2H O 2

DNNS PRPP NDP dNTP (RR)

R5P

Purine and Pyrimidine bases

341

DNPS

Precursors

DNA

Deoxyribonucleoside salvage dN

3H-dT,

BrdU

Fig. 2. Incorporation of H from H2O into DNA during cell proliferation. 2

2

A number of cell types have been studied by this approach, including lymphocytes, colonocytes, and others (15, 16). Analysis of DNA from bone marrow cells, which are nearly fully turned over during few days of 2H2O labeling, serves as an internal reference value for calculating the fraction of newly divided cells in a tissue of interest. In adipose tissue this approach has been applied to the proliferation of mature adipocytes, stromal-vascular cells, and adipose macrophages.

2. Materials 2.1. Equipment 2.1.1. Lipid Analyses 2.1.1.1. Total Lipid Extraction 2.1.1.2. Separation of Lipid Classes by Thin-Layer Chromatography

1. 16 × 100-mm clear borosilicate glass tubes with Teflon-lined caps (VWR, West Chester, PA). 2. Nitrogen drying system. 3. Rotating shaker. 4. Centrifuge. 1. Thin-Layer Chromatography (TLC) Silica Gel 60A plates, 20 × 20 cm (Whatman, Kent, UK). 2. TLC tank. 3. Chromatography paper. 4. UV lamp.

342

Pouteau et al.

5. 16 × 100-mm clear borosilicate glass tubes with Teflon-lined caps. 6. Rotating shaker. 7. Centrifuge. 2 .1.1.3. Separation and Derivatization of Fatty Acids and Glycerol

1. 16 × 100-mm clear borosilicate glass tubes with Teflon-lined caps. 2. GC vial with Teflon-lined cap (VWR, West Chester, PA). 3. Nitrogen drying system. 4. Centrifuge. 5. Speed Vac.

2 .1.2. Genomic DNA Analysis

1. Microcentrifuge tubes (VWR, West Chester, PA).

2 .1.2.1. Cell and Reagent Preparation

3. 50-ml conical tubes (VWR, West Chester, PA).

2.1.2.2. DNA Hydrolysis

1. 16 mm × 100 mm clear borosilicate glass tubes with screw-caps tubes.

2 .1.2.3. Derivatization of Purine dR

1. Adjustable micropipettes (200 and 1,000-ml maximal volume).

2. 15-ml conical tubes (VWR, West Chester, PA).

2. 16 mm × 100 mm clear borosilicate glass tubes with screw-caps tubes. 3. 13 mm × 100 mm glass test tubes. 4. GC/MS vials, caps. 5. Capping/decapping devices. 6. Fume hood. 7. Speed Vac suitable for use with acids and organic solvents.

2.1.3. Body Water Analysis

1. Tubes with O-ring cap (E&K Scientific, Santa-Clara, CA).

2.1.3.1. Body Water Distillation

2. Heating block with glass beads.

2.1.3.2. Acetone Transfer

1. Microcentrifuge tubes.

2.1.3.3. Acetone Extraction

1. Microcentrifuge tubes. 2. GC/MS vials.

2.2. Reagents

All reagents, unless indicated were purchased from (VWR, West Chester, PA) or Sigma Aldrich (Sigma-Aldrich, St Louis, MO) and should be of the highest purity (ACS or better).

2.2.1. Lipid Analyses

1. Chloroform.

2 .2.1.1. Total Lipid Extraction

2. Methanol. 3. Butylated hydroxytoluene (BHT). 4. 5% Sodium chloride solution (50 g of NaCl in 1 l distilled water).

2 .2.1.2. Separation of Lipid Classes by TLC

Dynamics of Adipose Tissue Development by 2H2O Labeling

343

1. Triolein standard. Defrost the Triolein (500 mg) and dissolve in 50 ml of chloroform in a glass tube with a Teflon cap. Use 10 ml or 100 mg for the TLC plate. 2. Chloroform. 3. BHT. 4. Hexane. 5. Diethyl ether. 6. Glacial acetic acid. 7. Rhodamine 6G. Weigh out 50 mg of Rhodamine 6G and add 200 ml of acetone to the Rhodamine. Shake well to mix. Place the Rhodamine solution in a sprayer bottle.

2 .2.1.3. Separation and Derivatization of Fatty Acids and Glycerol

1. Chloroform. 2. Methanolic HCl 3N. 3. Hexane. 4. Toluene. 5. Acetic anhydride. 6. Pyridine solution. 7. Ethyl acetate.

2 .2.2. Genomic DNA Analysis

1. Cells of interest, isolated after 2H2O labeling.

2 .2.2.1. Cell and Reagent Preparation

3. Buffer ATL from DNA extraction kits (Qiagen, Valencia, CA).

2. DNA standards. 4. Ultrapure water, molecular biology grade. 5. Sodium acetate. 6. Zinc sulphate. 7. Glacial acetic acid. 8. Potato acid phosphatase, 1 kU (Calbiochem, San Diego, CA). 9. S1 nuclease.

2.2.2.2. DNA Preparation

1. DNA extraction kits (Qiagen, Valencia, CA).

2.2.2.3. DNA Hydrolysis

1. 1N NaOH, molecular biology grade. 2. 5× hydrolysis cocktail.

2 .2.2.4. Derivatization of Purine dR

1. Pentafluorobenzyl hydroxylamine (PFBHA, Sigma-Aldrich, 1 mg/ml aqueous solution; prepare fresh weekly and store at 4°C). 2. Glacial acetic acid. 3. Acetic anhydride. 4. N-Methyl imidazole. Store dry at 4°C. 5. Sodium sulfate, granular, anhydrous (Sigma-Aldrich); Hygroscopic; keep dry.

344

Pouteau et al.

6. Dichloromethane. 7. Ethyl acetate. 8. dR: deoxyribose (Sigma-Aldrich). 2.2.3. Body Water Analysis

1. 2H2O standards and samples.

2.2.3.1. Body Water Distillation

2. 10 M NaOH. 3. HPLC grade acetone.

2.2.3.2. Acetone Transfer 2.2.3.3. Acetone Extraction

1. HPLC grade hexane. 2. Sodium sulfate.

3. Methods 3 .1. Deuterated Water Labeling in Animals

Deuterium water labeling uses a stable (nonradioactive) isotope to measure kinetics in  vivo. Deuterated water is administered for days or weeks via the animals’ drinking water. The synthesis of slow-turnover molecules and cells, such as adipose TG, fatty acids, and adipocytes, are measurable through the incorporation of deuterium (2H) from deuterated water into stable C–H bonds in newly synthesized TG glycerol, palmitate, and DNA. Thus by measuring the isotopic enrichment of deuterium in these end-products by gas chromatography/mass spectrometry (GC/MS) we determine simultaneously their synthesis rates. Procedurally, the treatment the living animal receives is a single intraperitoneal (IP) injection of 99.9% deuterated water with 0.9% NaCl (35 ml/g body mass) to achieve a deuterium enrichment of 5% in body water (see Note 1). This volume is calculated by estimating 70% of body weight as water (0.7 × 0.05 = 0.035 ml/g). Subsequently, normal drinking water is replaced with 8–10% deuterated water to maintain the animal’s TBW at 5% enrichment of deuterated water. Deuterated water is maintained for as long as is requested by the study.

3 .2. Tissue and Body Water Collection

Dissect adipose tissues from the animal and immediately store on ice. Biopsies can be as little as 10–20 mg of adipose tissue (with preferably >5.0 mg of DNA although 1.0 mg of DNA is measurable). Collect bone marrow from the tibias (see Note 2). If samples cannot be analyzed immediately the samples are preferably stored at −70°C. Mince or homogenize the adipose tissues before proceeding. Collect 100 ml of plasma, serum, or urine per analysis.



Dynamics of Adipose Tissue Development by 2H2O Labeling

345

3 .3. Application Study in Growing Obese Rats

The study of fat development in humans or in animals is of great interest (1, 3, 6). The study present aimed to compare the adipose tissue development of obese homozygous (fa/fa) Zucker pups with lean heterozygous (fa/+) Zucker pups. Fifteen lean Zucker rat females were mated with obese Zucker rat males (Charles Rivers, Lyon, France). The 15 litters were administered deuterated water for 5 days prior sacrifice at different ages (at 13, 20, 27, 34, and 55 days of age) including the suckling and postweaning periods. A conventional diet (Kliba 3434, Promivi Kliba SA, Kaiseraugst, Switzerland) was given to all rats. Females were sacrificed after birth. Sixty-two male pups were utilized throughout the kinetic studies as follows: Isotope administration procedure during suckling period. Five days before the sacrifice, suckling pups at 13 and 20 days of age, and their mothers, received an IP injection of deuterated water (~35 ml/g body weight, 99% 2H2O, 0.9% NaCl). Thereafter, the litter and mother had ad libitum access to deuterium-enriched (8% 2H2O) drinking water. The milk from the rat mothers carried out deuterium-labeled water to suckling pups. Isotope administration procedure after postweaning: Five days before the sacrifice, the pups received an IP injection of deuterated water (~35 ml/g body weight, 99% 2H2O, 0.9% NaCl). Pups had ad libitum access to deuterated (8% 2H2O) drinking water. For both suckling and postweaning periods, blood, bone marrow, and adipose tissues (epididymal and retroperitoneal) were sampled and weighed the day of sacrifice. The ethical committee for experimentation in animals of the Swiss Authority (Canton de Vaud, # 1659 and 1659.1) approved the present rat study.

3.4. Lipid Analyses

1. Add 25–35 mg of minced or homogenized adipose tissue to a 16 × 100-mm clear borosilicate glass tubes with Teflonlined caps

3.4.1. Total Lipid Extraction

2. Add 5 ml of 2:1 chloroform:methanol solution containing 50 mg/P of BHT to the glass tube containing the sample 3. Shake the tube for 2 min 4. Place the tube on a rotating shaker for at least 15–20 min at room temperature 5. Add 1 ml of 1 M NaCl solution to each tube containing the sample and vortex for 10 s 6. Centrifuge the mixture at low speed (500 × g) to separate the aqueous and the solvent phases 7. Transfer the lower solvent layer to a new clean 16 × 100-mm clear borosilicate glass tubes with Teflon-lined cap.

346

Pouteau et al.

8. For adipose tissue use 0.5 ml of the lower solvent layer for further analysis (decrease volume used if TLC plate is overloaded or increase volume used if TLC plate is underloaded, see Subheading 3.4.2). The remaining of the sample can be stored at −20°C 9. Completely dry the sample under a stream of N2 3 .4.2. Separation of Lipid Classes by TLC

1. Four samples and one triolein standard can be run simultaneously on one TLC plate 2. Prepare five identical lanes on the plate by scoring lines through the silica 3. Add 30 ml of chloroform to the dried sample and vortex briefly 4. Apply one sample in one lane on the plate. Samples should be placed on the TLC plate as a line of spots that are as small as possible 5. Let the first line of spots dry on the TLC plate 6. Add another 30 ml of chloroform to tube containing the dried sample and vortex briefly 7. Apply the second part of the sample on the same lane of the TLC plate as described in step 4 8. Apply 100 mg of triolein to the middle lane on the TLC plate 9. Add 50 mg of BHT, 70 ml of hexane, 30 ml of diethylether, and 1.8 ml of glacial acetic acid to a clean TLC tank and mix the solvents. 10. Place a piece of chromatography paper down each side of the tank and put the lid on the tank 11. Leave the tank for a minimum of 30 min 12. Place the plate into the tank and leave for 40 min (or until solvent nearly reaches the top of the plate). Two plates can be run at the same time 13. Remove the plates from the tank and leave them to dry for at least 2 h 14. Spray the dry plates with Rhodamine to visualize the lipid spots 15. Place the plates under a UV lamp and mark around the triglyceride spots 16. Scrape the sample from the TLC plate and transfer to a 16 × 100-mm clear borosilicate glass tubes with Teflon-lined caps (see Note 3) 17. Add 1 ml of chloroform to the tube containing the silica 18. Place the tube with a Teflon-lined cap on a rotating shaker for 20 min



Dynamics of Adipose Tissue Development by 2H2O Labeling

347

19. Centrifuge at 1,000 g for 10 min at room temperature 20. Transfer the chloroform supernatant into new glass tube with Teflon-lined cap 3 .4.3. Separation and Derivatization of Fatty Acids and Glycerol

1. Add 1 ml of methanolic HCl 3N to the sample and vortex briefly 2. Incubate for 1 h at 55°C or incubate overnight at room temperature 3. Add 2 ml of H2O and 3 ml of hexane to the tube containing the sample (tube 1) and vortex briefly 4. Centrifuge tube 1 at 1,000 g for 10 min at room temperature to separate the top layer containing the fatty acids from the lower layer containing the glycerol

3.4.4. Fatty Acids

1. Transfer the top layer containing the fatty acids from tube 1 to a new glass tube with Teflon-lined cap (tube 2) 2. Add another 3 ml of hexane to tube 1 and vortex briefly. Repeat step 4, Subheading 3.4.3 and step 1, Subheading 3.4.4, adding the top layer to tube 2 containing the fatty acids 3. Dry the sample in tube #2 under a gentle N2 stream 4. Add 1 ml of toluene to the dried sample and transfer to a GC vial with Teflon-lined cap for GC/MS analysis

3.4.5. Glycerol

1. Transfer part of the bottom layer from tube 1 to a GC vial and dry the sample in a Speed Vac. When dry, add the remaining of the bottom layer to the same GC vial and dry. 2. Add 100 ml of freshly prepared 2:1 acetic anhydride:pyridine solution 3. Cap the GC vial and incubate at room temperature for 30 min 4. Dry the sample under a stream of N2 or in a Speed Vac 5. Add 1 ml of ethyl acetate to the sample in the GC vial for GC/MS analysis

3 .4.6. GC/MS Analysis of Palmitate

GC–MS analyses are run on an Agilent GC-Quadrupole system (6890/5973, Santa Clara, CA). The GC Column used is either a DB-17ms (J&W Scientific, Folsom, CA) or its equivalent ZB-50ms (Phenomenex, Torrance, CA), 30 m long, 250 mm I.D., 0.25 mm film thickness with Helium as the carrier gas. The Quad MS is operated in Electron Impact ionization mode (EI) with source temperature set to 230°C and quad temperature set to 150°C. The GC inlet temperature, column flow, and oven program are as follows: 1. Inlet temperature: 280°C, pulsed splitless. 2. Interface temperature: 280°C

348

Pouteau et al.

3. Column Helium flow rate: 2 ml/min 4. Oven Temperature Ramp: • Initial temperature: 100°C with 1 min hold • Ramp temperature to 230°C at 20°C/min with 1 min hold • Ramp temperature to 300°C at 20°C/min with no hold

5. Injection volume: 0.5–2.0 ml 6. SIM Parameters: m/z 270.3, 271.3, 272.3 (M0, M1, and M2) with dwell time of 15 ms and EMV set to 2,000 V

After the sequence is run, visually inspect peaks to rule out contaminating peaks. There must be a unique peak of derivatized palmitate for each sample and standard. Make sure that the sample retention time matches that of the standard and only the three isotopomers are present (m/z 270, 271, and 272).

3 .4.7. GC/MS Analysis of Glycerol

GC–MS analyses are run on an Agilent GC-Quadrupole system (6890/5973). The GC Column used is a DB-225 ms (J&W Scientific), 30 m long, 250 mm I.D., 0.25 mm film thickness with Helium as the carrier gas. The Quad MS is operated in chemical ionization mode (CI) with source temperature set to 250°C and quad temperature set to 150°C. The GC inlet temperature, column flow, and oven program are as follows: 1. Inlet temperature: 280°C, pulsed splitless. 2. Interface temperature: 280°C 3. Column Helium flow rate: 1 ml/min 4. Oven Temperature Ramp: • Initial temperature: 80°C with a 2 min hold • Ramp temperature to 160°C at 15°C/min with 2 min

hold

• Ramp temperature to 220°C at 40°C/min with 1.5 min

hold

5. Injection volume: 0.5–2.0 ml 6. SIM parameters: m/z 159, 160, 161 (M0, M1, and M2, same retention time) with dwell time of 20 ms and EMV set to 2,000 V

3 .5.1. Cell Preparation: Cells in Suspension

Isolate 2H2O-labeled cells via any suitable method (e.g., flow cytometric or immunomagnetic sorting, density gradient centrifugation, flotation). Pellet cells in a microcentrifuge tube (250–800 × g, 5–10 min, depending on cell type and whether they are fixed or not). Remove supernatant (taking care not to disturb the cell pellet, which, for small cell samples, may be invisible) and freeze cell pellets at or below −20°C until use.

3 .5.2. Cell Preparation: for Tissue Samples

Place tissue fragments directly in 180 ml Buffer ATL of the DNA extraction kit. Store at room temperature (20–25°C) or 4°C.

3 .5. Genomic DNA Analysis



Dynamics of Adipose Tissue Development by 2H2O Labeling

349

Note that a precipitate forms on cold storage but will dissolve during subsequent tissue digestion at 56°C. 3.5.3. DNA Standards

3.5.4. DNA Preparation

3.5.5. DNA Hydrolysis

Prepare multiple aliquots of 0.5 mg DNA per 200 ml water and store at or below −20°C. For preliminary work, unlabeled DNA from any source will suffice (e.g., calf thymus DNA). 2H2O-labeled DNA (1–3% EM1; e.g., splenocytes from mice labeled at »5% 2 H2O in body water for a week) is preferable to check variation of EM1 measurements (Excess M1 from GC/MS analysis) between and within experiments. 1. Five times and 1× hydrolysis buffer (sodium acetate, zinc sulfate, pH 5.0) Place 94 ml of ultrapure water into a sterile beaker. Weigh out 3.08 ± 0.02 g sodium acetate and add to water. Weigh out 21.5 ± 0.02 mg zinc sulfate and add to water. Mix until dissolved. Adjust pH to 5.0 ± 0.5 with glacial acetic acid. Bring up to 100 ml total volume. The resultant solution is 5× hydrolysis buffer. Pipette 1 ml of 5× buffer into a 15 ml conical tube and add 4 ml ultrapure water to make 1× hydrolysis buffer. 2. Acid phosphatase enzyme Resuspend entire vial of acid phosphatase in 1.0 ml ultrapure water. Keep on ice (or 4°C) until preparation of the enzyme cocktail 3. S1 Nuclease Enzyme Dilute 2.5 ml of S1 Nuclease stock (Sigma) into 2.0 ml 1× hydrolysis buffer; nuclease will be at ~0.5 units/ml. Keep on ice (or 4°C) until preparation of the enzyme cocktail 4. Five times hydrolysis cocktail Pipette 36.8 ml of 5× hydrolysis buffer into a 50 ml conical tube – label accordingly. Add 1.0 ml of acid phosphatase solution and 1.0 ml of diluted S1 nuclease solution. Vortex thoroughly. Aliquot the enzymatic hydrolysis cocktail “buffer” to microfuge tubes (approximately 520 ml per tube – enough for ten reactions). Store at −20°C for up to 12 months. Extract DNA using a Qiagen DNEasy kit. Include the cell samples, as well as duplicate extraction blanks (water only) and duplicate extraction standards (0.5 mg DNA in 200 ml water). Follow manufacturer’s instructions, except for the elution step. For fixed cells or tissues, use the tissue modification of the manufacturer’s protocol. For unfixed suspension cells, the manufacturer’s protocol for suspension cells may be used. Elute DNA in 2 × 100 ml water and pool the eluates. Do not use elution buffer, which contains EDTA and may interfere with later steps. 1. Transfer 0.5–1 mg of extracted DNA, plus extraction blanks, and extraction standards to 16 mm × 100 mm screw-capped glass hydrolysis tubes

350

Pouteau et al.

2. Add 50 ml hydrolysis cocktail to each tube and vortex 3. Incubate at 37°C for 1 h to overnight with gentle agitation. After step 3, samples can be stored at −20° until derivatization. 3 .5.6. Derivatization of Purine dR

1. Add 100 ml of aqueous PFBHA solution to each sample, as well as to each of the standards and blanks introduced during the extraction and hydrolysis steps. In addition, derivatization blanks (250 ml water) and quantitative dR standards may be added at this step. 2. Add 75 ml of glacial acetic acid to each sample. 3. Cap the tubes and place samples on heating block set at 100°C for 30 min. 4. Remove samples and allow them to cool to room temperature. 5. Add 1 ml of acetic anhydride to each sample. 6. Add 100 ml of N-methylimidazole to each tube and mix immediately. Samples may splash due to sudden overheating as the exothermic acetylation reaction proceeds. Perform this step in a fume hood, wearing protective goggles and pointing the opening of the tube away from you. N-methylimidazole can deteriorate if not properly stored. 7. Allow the reaction to proceed for 15–20 min, during which time the samples will cool down. 8. Set up and label disposable glass test tubes for each sample. Add sufficient sodium sulfate to cover the bottom of each tube. 9. Add 2 ml of water to the reactions from step 12. Vortex for 10 s. 10. Add 750 ml of dichloromethane to the tubes and vortex for 5 s. Allow phases to separate for ~1 min. 11. Wet P1000 pipette tip with dichloromethane and use it to transfer 500 ml of bottom (organic) layer into the disposable culture tube with sodium sulfate. Do not transfer any of the aqueous phases, which may introduce GC contaminants. 12. Repeat steps 16 and 17, pooling organic phases from both extractions. Vortex gently and allow sodium sulfate crystals to settle. 13. Transfer the supernatant to a clean, labeled GC vial, avoiding transfer of any sodium sulfate crystals. 14. Dry GC vials in a Speed Vac at room temperature for ~4 h to overnight. Heating the sample may cause evaporative loss of the derivative. Complete drying should be verified by inspecting the GC vial, as residual moisture or acid can damage the GC column. Drying in a stream of nitrogen is not sufficient to remove residual acetic acid. 15. Resuspend each sample in 100 ml of ethyl acetate and transfer to a GC vial insert. Wash sides and bottom of first GC vial



Dynamics of Adipose Tissue Development by 2H2O Labeling

351

with another 100-ml aliquot of ethyl acetate to remove any residual derivative. 1 6. Place insert vials in Speed Vac and dry down as above for ~1 h. 17. Resuspend each sample in 50 ml ethyl acetate and cap immediately. 3 .5.7. GC/MS Analysis of DNA

Analysis of derivatized dR is performed on a standard GC quadrupole MS instrument capable of operating in Negative Chemical Ionigation (NCI) mode, such as an Agilent 6890/5973 GC/MS system, with helium as carrier and methane as reagent gas. The GC column used is a DB-17 type column (Agilent) or similar, 30 m, 250 mm I.D., 0.25 mm film thickness. The ion source and quad temperatures are both set to 150°C. The GC inlet temperature, column flow, and oven program are as follows: 1. Inlet temperature: 280°C, splitless 2. Interface temperature: 250°C 3. Column Helium flow rate: 1 ml/min 4. Oven Temperature Ramp: • Initial temperature: 140°C with 1 min hold • Ramp temperature to 280°C at 40°C/min with 1 min hold • Ramp temperature to 300°C at 40°C/min with 1 min hold

5. Injection volume: 0.5–2.0 ml 6. SIM parameters: m/z 435 and 436 (M0 and M1) with dwell time of 15 ms and EMV set to 2,000 V The parent ion [M0] of the PFTA derivative of dR has m/z » 435; the M1 mass isotopomer has m/z » 436. The natural abundance of unenriched samples are measured and the excess M1 (EM1) abundance in the enriched samples is calculated by subtracting the M1 abundance measured in the unenriched sample from the M1 abundance in the sample. 3 .6. Body Water Analysis 3 .6.1. Body Water Distillation

1. Prepare 2H2O standards (0–10% 2H2O enrichments) 2. Pipette 100 ml of plasma or standard into the inner well of an O-ring cap 3. Hold the cap and tightly screw down an inverted tube. 4. Submerge cap in a 80°C heating block filled with glass beads 5. Incubate overnight (10–12 h) 6. Carefully remove the tube from the heating block without disturbing water droplets on the sides of the tube 7. Turn the tube horizontally without loosing any water droplets and replace original O-ring cap with new O-ring cap and seal the tube

352

Pouteau et al.

3.6.2. Acetone Transfer

1. Transfer the sample or standard in the tube with O-ring cap to a microcentrifuge tube 2. Add 1 ml of 10 M NaOH to the sample or standard 3. Add 5 ml of HPLC grade acetone to each sample or standard and vortex 4. Cap the microcentrifuge tube tightly 5. Incubate overnight (12–24 h) at room temperature

3.6.3. Acetone Extraction

1. Add 300 ml of HPLC grade hexane to each tube containing the sample or standard 2. Cap the tube immediately and vortex for 10 s. Water “droplets” will appear below hexane near bottom of the tube 3. Transfer 250 ml of hexane to a new microfuge tube containing 100 ml of sodium sulfate. Cap the tube immediately and vortex 4. Let the sodium sulfate settle and transfer 200 ml of the hexane to a GC/MS vial. Cap immediately

3 .6.4. GC/MS Analysis of Acetone

GC–MS analyses are run on an Agilent GC-Quadrupole system (6890/5973). The GC column used is a RTX-225 (Restek, Bellefonte, PA, Crossbond 50% Cyanopropyl methyl, 50% phenyl methyl polysiloxane), 30 m long, 320 mm I.D., 0.1 mm film thickness with Helium as the carrier gas. The Quad MS is operated in Electron Impact ionization mode (EI) with source temperature set to 230°C and quad temperature set to 150°C. The GC inlet temperature, column flow, and oven program are as follows: 1. Inlet temperature: 250°C, split 20:1 2. Interface temperature: 220°C 3. Column Helium flow rate: 2.6 ml/min 4. Oven Temperature Ramp: • Initial temperature: 70°C with 2 min hold • Ramp temperature to 220°C at 60°C/min with 1 min hold

5. Injection volume: 0.5–2.0 ml 6. Solvent Delay: ~1.15 min, Source Close: ~1.5 min 7. SIM Parameters: m/z 58, 59, 60 (M0, M1, and M2) with dwell time of 15 ms and EMV set to 2,000 V 8. Solvent wash vials: Toluene (A) and Ethyl Acetate (B) (No acetone wash should be used for this type of analysis). 3.7. Results 3.7.1. Calculation 3 .7.1.1. Fractional Synthesis Rate of Cells and Lipids

FSR are calculated as the ratio of 2H incorporated into the analyte of interest (TG, FA or DNA/cell) to the theoretical 2H-enrichment in that analyte if all of the sample were newly synthesized (100% new). A variety of methods are available for determining the theoretical 100% new or asymptotic enrichment (A*) value. In some cases a surrogate tissue or cell type which is completely turned over or replaced during the course of the study can be



Dynamics of Adipose Tissue Development by 2H2O Labeling

353

measured and its enrichment used as the A* value (e.g., bone marrow DNA). In other cases this is not possible, therefore A* is calculated from the measured body water enrichment in each animal and the number of sites where 2H can be incorporated (n) during synthesis (TG and FA below). Additionally the absolute synthesis rates of TG and of palmitate are calculated by considering the TG and the palmitate pools in the fat tissues. 3 .7.1.2. FSR of Adipose Tissue TG

The best-fit equation to determine A*EM1 for hydrogen atoms to enter a-glycerol phosphate in adipose tissue was: A¥ = – 6.3631*c2 + 3.3048*c + 0.0016 where c represents measured body water enrichments. The equation is generated using standard tables of mass isotopomer distribution analysis (MIDA) using (1) a value of 4 for n in adipose tissue TG and (2) using only ions M0 and M1 of glycerol. The n calculated for adipose tissue TG and liver TG is based upon plateau label incorporation after long-term 2H2O administration and combinatorial labeling patterns using MIDA (1). The value n may be affected by physiological conditions and can be tested (7). The fractional TG synthesis is calculated as TG-Glycerol EM1/calculated A* EM1 and expressed as % new TG per number of labeling days.

3.7.1.3. Fractional DNL

MIDA is used to directly calculate the fractional contribution from DNL to TG. The deuterium enrichment in body water is used as isotopic enrichment of the precursor pool and deuterium enrichment in TG-palmitate is used as the isotopic enrichment of the product. Fractional DNL was calculated based on the precursorproduct relation with the following equation: Fractional DNL = (0.99841737*(EM1 palmitate + 0.1591485) – 0.15889671)/ (((– 7199.2*c4 + 2403.2*c3 – 303.56*c2 + 14.039*c + 0.1608) – 0.15889671) + (0.99841737 – (946.16*c4 + 84.319*c3 – 60.826*c2 – 0.0202*c + 0.9972))* (EM1 palmitate + 0.1591485)), where c represents measured body water Note: While the calculation appears to be considerably more complex for DNL than TG synthesis the underlying principle is the same, in the case of FA the large number of potential incorporation sites for 2H during synthesis results in a more complex equation.

3.7.2. Lipid Synthesis Rate

H-enrichments of TG-glycerol in adipose tissues are shown in Table 1. 2H-enrichment of body water ranged from 3.8 ± 0.1 to 5.4 ± 0.1% excess (median ± standard error). The fractional and absolute synthesis rates of TG in epididymal adipose tissue of obese and lean pups are shown in Fig. 3. The fractional TG synthesis rate was higher in epididymal than in retroperitoneal fat pads (Table 1).

2

20 days

27 days

35 ± 4 46 ± 3

Retroperitoneal Epididymal

Retroperitoneal Epididymal

Retroperitoneal Epididymal

Retroperitoneal Epididymal

Retroperitoneal Epididymal

FSRDNA

EM1TG

FSRTG

EM1FA

FSRFA

21 ± 6 34 ± 7

2.6 ± 0.6 4.7 ± 0.9

45 ± 4 83 ± 3

28 ± 3 46 ± 3

5.3 ± 0.5* 4.8 ± 0.3 9.9 ± 0.4 7.2 ± 0.5

21 ± 2 32 ± 1

2.3 ± 0.2 3.6 ± 0.1

35 ± 3 47 ± 5*

5.8 ± 0.4* 7.3 ± 0.7*

20 ± 3 26 ± 5

2.5 ± 0.3 3.1 ± 0.5

5.2 ± 0.1 12.1 ± 0.2

98.5 ± 4.5

34 days

21 ± 3 45 ± 7

* P < 0.05, 2-way Anova, obese vs. lean pups

34 ± 2 54 ± 2

24 ± 6 50 ± 5*

44 ± 4* 55 ± 6*

13 days

20 days

27 days

Lean pups 34 days

55 days

27 ± 3 28 ± 1*

4.3 ± 0.4 4.4 ± 0.1*

25 ± 3* 29 ± 3*

3.0 ± 0.3* 3.5 ± 0.4*

5.3 ± 0.1* 12.1 ± 0.1*

21 ± 3 24 ± 1*

36 ± 4 83 ± 5

4.2 ± 0.5 9.7 ± 0.6

22 ± 2 35 ± 1

2.6 ± 0.2 4.0 ± 0.2

31 ± 1 41 ± 3

4.8 ± 0.1 6.4 ± 0.4

21 ± 2 35 ± 3

2.5 ± 0.3 4.0 ± 0.3

39 ± 3 64 ± 3

17 ± 3 51 ± 2

20 ± 4 31 ± 4

11.2 ± 0.9 4.6 ± 0.9 5.8 ± 1.4 19.7 ± 0.1 14.7 ± 0.5 9.1 ± 1.1

69 ± 3

8.6 ± 0.4

38 ± 2 47 ± 3

4.8 ± 0.3 5.8 ± 0.6

27 ± 3 25 ± 5

7.7 ± 0.9 7.1 ± 1.5

31 ± 4 33 ± 5

4.8 ± 0.4 5.2 ± 0.6

22 ± 2 26 ± 2

2.5 ± 0.2 3.0 ± 0.3

24 ± 4 15 ± 2

6.8 ± 1.3 4.1 ± 0.7

31 ± 2 21 ± 1

4.3 ± 0.3 2.9 ± 0.2

16 ± 2 15 ± 1

1.7 ± 0.2 1.6 ± 0.1

4.2 ± 0.1 3.8 ± 0.1 5.2 ± 0.2 5.2 ± 0.2 4.6 ± 0.1 12.5 ± 0.2 11.5 ± 0.1 11.7 ± 0.2 11.4 ± 0.2 10.8 ± 0.1

220.7 ± 7.7* 27.1 ± 1.2 41.2 ± 1.6 67.5 ± 1.9 90.5 ± 4.0 190.6 ± 12.0

55 days

10.3 ± 0.8 5.5 ± 0.8 6.9 ± 1.8 13.9 ± 1.5* 6.0 ± 0.8 17.7 ± 0.1 13.0 ± 2.3 15.8 ± 2.0* 17.9 ± 2.2* 6.9 ± 0.3*

53 ± 4

7.3 ± 0.6

4.5 ± 0.6 5.8 ± 0.4

Body water 5.4 ± 0.1 3.9 ± 0.1 5.3 ± 0.2 Bone marrow DNA 12.9 ± 0.3 11.4 ± 0.2 12.5 ± 0.3

26.3 ± 1.5 41.4 ± 1.5 70.1 ± 2.3

EM1DNA Retroperitoneal Epididymal

EM1

Body weight (g)

13 days

Obese pups

Table 1 Body weight and 2H-enrichments (in % excess M1) in body water, genomic DNA, TG-glycerol and TG-palmitate (FA), as well as fractional synthesis rates (FSR in % new molecule per 5 days, molecules are DNA, TG or FA) measured in retroperitoneal and epididymal adipose tissues in obese and lean rat pups from 13 to 55 days of age

0

2

1

20

*

40

*

*

60

0

30

60

90

120

150

0%

20%

40%

60%

80%

100%

0

4

3

20

Age (d)

*

*

*

40

*

*

60

0

5

10

15

20

25

0%

20%

40%

60%

80%

100%

0

6

5

20

*

*

* *

40

*

*

60

Fig. 3. The dynamic growth of the epididymal adipose tissue: The adipose tissue weight (1, in mg), the fractional proliferation rate of adipose tissue cells (2, FPR in % new cells/5 days), the fractional (3, FSR in % new TG/5 days) and absolute synthesis rate of triglycerides (4, in mg of TG/d) and the fractional (5, FSR in % new palmitate/5 days) and absolute synthesis rate of fatty acids (6, FA in mg of palmitate/d) measured in the epididymal adipose tissues (AT) in fatty (circle) and lean (triangle) Zucker rat pups from 13 to 55 days. Data are median ± SEMedian, * P < 0.05, 2-way Anova, obese vs. lean pups.

0%

10%

20%

30%

40%

50%

0

500

1000

1500

2000

2500

3000

FSR of TG (% new TG / 5 d) Absolute synthesis rate of TG (mg/d)

Organ weight (mg)

FPR of cells (% new cells / 5 d)

FSR of FA (% new palmitate / 5 d) Absolute DNL (mg palmitate/d)

Dynamics of Adipose Tissue Development by 2H2O Labeling 355

356

Pouteau et al.

From 27 days to adult age, the TG absolute synthesis rate was markedly elevated in obese pups compared with lean pups. 2 H-enrichments of TG-palmitate in adipose tissues are shown in Table 1. The fractional and absolute synthesis rates of palmitate (fractional and absolute DNL) in epididymal adipose TG of obese and lean pups are shown in Fig. 3. While the DNL value decreased with age in lean pups, a remarkable increase occurred in obese pups at the age of 27–34 days followed by a decrease in the 55-day-obese pups (Fig. 3). The absolute DNL increased dramatically from 20 to 34 days to adult age. H-enrichment of genomic DNA in adipose cells are shown in Table 1. DNA 2H-enrichment in bone marrow ranged from 10.8 ± 0.1 to 12.9 ± 0.3% excess. The fractional proliferation rates of epididymal adipose DNA/or cells of obese and lean pups are shown in Fig. 3 according to age. All rat pups demonstrated the most active adipose fractional cell proliferation rate during the suckling period. However, the fractional cell proliferation rate sharply decreased at the end of the suckling period (from 15 to 20 days) and remained constant throughout puberty (55 days) except in obese pups that showed a significant increase in adipose tissue cell proliferation rate at 55 days. Because of limitations in adipose tissue availability, the separate measurement of the proliferation rate in the adipocyte vs. the stromal-vascular fraction was not feasible in this study. Previously, Antelo et  al. successively separated mature adipocyte-enriched cells and observed a cell proliferation rate of about 0.4–0.7% and 1–1.5% new cells produced per day in adult rats and mice (3, 17), respectively. In sum, the fractional cell proliferation rate (hyperplasia) in adipose tissues was higher in obese than in lean rats not during the young age but only during the adult age. In the present study, we successfully considerably shortened the deuterium intake phase, which allowed determination of adipose tissue cell proliferation rate at very young age. In the present study, the 2H-enrichment reached a plateau-value of about 5% in the body water that did not match the 2H-enrichment of drinking water; that is explained from the nonlabeled water diluting the body water pool. The food intake and the endogenous metabolism produce nonlabeled water and contribute to the dilution of the body water.

3 .7.3. Fractional Cell Proliferation Rate

2

3.8. Conclusion

Deuterium water labeling uses a stable (nonradioactive) isotope to measure kinetics in  vivo. This method is without hazard, accurate, reproducible, and it is sensitive to even small changes in metabolism. The synthesis of slow-turnover molecules and cells, such as adipose triglycerides, fatty acids, and adipocytes are measurable through the incorporation of deuterium (2H) from heavy water (2H2O) into stable C–H bonds in newly synthesized



Dynamics of Adipose Tissue Development by 2H2O Labeling

357

triglyceride-glycerol, -palmitate and DNA. Thus by measuring the isotopic enrichment of deuterium in these molecules by gas chromatography/mass spectrometry, we can determine their synthesis rates. Using this method we can simultaneously measure quantitative changes in synthesis and turnover of multiple components of the adipose tissues.

4. Notes 1. Mice with weights ³30 g should be anesthetized with isoflurane before IP dosing of 2H2O. Alternately, deuterated water is administered as half the volume in morning and other half volume in afternoon. The animals experience a period of ataxia, similar to vertigo in humans, after the initial bolus dose, but it is transient, lasting less than 15 min. There are no other side effects of deuterated water administration other than this brief period. 2. The Tibia or femur bone can be used for bone marrow collection. Both ends of the bone is cut with scissors and the bone marrow is washed out by inserting a needle into one end and passing 100–200 ml of saline though the bone, collecting the eluant in an Eppendorf tube. 3. A folded piece of weigh paper can be used to transfer the scraping into the tube.

Acknowledgment We particularly thank Marc Hellerstein, Katherine Macé, Corinne Ammon-Zufferey, and Mireille Moser for their collaboration to the success of the present work. We thank the animal house staff. The present work was co financed by KineMed Inc and Nestec Ltd.

References 1. Turner, S.M., Murphy, E.J., Neese, R.A. et al., (2003) Measurement of TG synthesis and turnover in vivo by 2H2O incorporation into the glycerol moiety and application of MIDA. Am J Physiol 285, E790–E803. 2. Hirsch, J. (1965) Fatty acid patterns in human adipose tissue. In Handbook of physiology

(Renold, A.E. and Cahill, G.F., eds.), Am Physiol Soc, Washington 3. Neese, R.A., Misell, L.M., Turner, S. tet  al., (2002) Measurement in  vivo of proliferation rates of slow turnover cells by 2H2O labeling of the deoxyribose moiety of DNA. Proc Natl Acad Sci U S A 99, 15345–15350.

358

Pouteau et al.

4. Strawford, A., Antelo, F., Christiansen, M. and Hellerstein, M.K. (2004) Adipose tissue triglyceride turnover, de  novo lipogenesis, and cell proliferation in humans measured with 2H2O. Am J Physiol 286, E577–E588. 5. Turner, S.M., Roy, S., Sul, H.S. et al., (2007) Dissociation between adipose tissue fluxes and lipogenic gene expression in ob/ob mice. Am J Physiol 292, E1101–E1109. 6. Pouteau, E., Turner, S., Aprikian, O. et  al., (2008) Time course and dynamics of adipose tissue development in obese and lean Zucker rat pups. Int J Obes 32, 648–657. 7. Chen, J.L., Peacock, E., Samady, W. et  al., (2005) Physiologic and pharmacologic factors influencing glyceroneogenic contribution to triacylglyceride glycerol measured by mass isotopomer distribution analysis. J Biol Chem 280, 25396–25402. 8. Rizki, G., Arnaboldi, L., Gabrielli, B. et  al., (2006) Mice fed a lipogenic methioninecholine-deficient diet develop hypermetabolism coincident with hepatic suppression of SCD-1. J Lipid Res 47, 2280–2290. 9. Lee, W.N., Bassilian, S., Lim, S. and Boros, L.G. (2000) Loss of regulation of lipogenesis in the Zucker diabetic (ZDF) rat. Am J Physiol 279, E425–E432. 10. Leitch, C.A. and Jones, P.J. (1993) Measurement of human lipogenesis using deuterium incorporation. J Lipid Res 34, 157–163.

11. Hellerstein, M.K., Kletke, C., Kaempfer, S., Wu, K. and Shackleton, C.H. (1991) Use of mass isotopomer distributions in secreted lipids to sample lipogenic acetyl-CoA pool in  vivo in humans. Am J Physiol 261, E4 79–E486. 12. Hellerstein, M.K., Christiansen, M., Kaempfer, S. et  al., (1991) Measurement of de  novo hepatic lipogenesis in humans using stable isotopes. J Clin Invest 87, 1841–1852. 13. Smith, J., Al Amri, M., Dorairaj, P. and Sniderman, A. (2006) The adipocyte life cycle hypothesis. Clin Sci (Lond) 110, 1–9. 14. Neese, R.A., Siler, S.Q., Cesar, D. et al. (2001) Advances in the stable isotope mass spectrometric measurement of DNA synthesis and cell proliferation. Anal Biochem 298, 189–195. 15. Hellerstein, M.K., Hanley, M.B., Cesar, D. et  al., (1999) Directly measured kinetics of circulating T lymphocytes in normal and HIV-1-infected humans. Nat Med 5, 83–89. 16. Kim, J., Neese, R. and Hellerstein, M. (2000) A new method to measure proliferation rates of colon epithelial cells. FASEB J 14, A718. 17. Antelo, F., Neese, R. and Hellerstein, M. (2000) Measuring adipocyte proliferation in vivo using 2H2O incorporation into DNA. FASEB 14, A214.

Chapter 18 Analysis of Lipid Particles from Yeast Melanie Connerth, Karlheinz Grillitsch, Harald Köfeler, and Günther Daum Summary Quantitative analysis of components from different subcellular fractions is a key to the understanding of metabolic function as well as to the origin, the biogenesis, and the crosstalk of organelles. The yeast is an excellent model organism to address such questions from the biochemical, molecular biological, and cell biological viewpoints. A yeast organelle which gained much interest during the last decade is the lipid particle/droplet (LP), a storage compartment for nonpolar lipids but at the same time an organelle actively contributing to cellular metabolism. In this chapter, we describe methods and techniques that are commonly used to analyze lipids from LP at the molecular level by thin-layer chromatography, gas–liquid chromatography, and mass spectrometry. We provide an easy to follow guideline for the isolation of these organelles, the qualitative and quantitative analysis of lipid components and show results obtained with these methods. Key words: Lipid particles/droplets, Phospholipid, Triacylglycerol, Steryl ester, Yeast

1. Introduction One of the most important prerequisites to establish a well organized and preserved cellular structure is the formation of biological membranes that protect or separate organelle components from the cellular environment, and shield cells from the exterior. Major constituents of membranes are lipid molecules, especially phospholipids, sterols, sphingolipids, and glycolipids. Moreover, all types of cells contain storage lipids, which are also referred as neutral lipids or nonpolar lipids. In most cases, these neutral lipids are stored in a very specific compartment named lipid particle (LP), lipid droplet, lipid body, or oil body (1). Contrary to other cellular organelles, the LP does not contain a phospholipid Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_18, © Humana Press, a part of Springer Science + Business Media, LLC 2009

359

360

Connerth et al.

bilayer on its surface but rather a phospholipid monolayer with the polar head groups facing the hydrophilic environment and the hydrophobic part associating with the nonpolar lipids of the LP core (2, 3). The current hypothesis of LP biogenesis is based on a budding model suggesting that LP derive from the endoplasmic reticulum (ER) (4–7). It has been proposed that neutral lipids formed in the ER accumulate in certain domains, which grow and generate the LP core. Once the size of the nonpolar lipid droplet increases and exceeds accommodation in the ER phospholipid bilayer, the LP buds off forming an independent organelle. In the yeast Saccharomyces cerevisiae, steryl esters (STE) and triacylglycerols (TAG) are the major neutral lipid components of LP. Under normal growth conditions of the yeast the existence of LP is important although not essential (8). LPs are assumed to store excess amounts of fatty acids and sterols in the biological inert form of STE and TAG and thereby avoid possible lipotoxic effects of these components. At the same time, storage and mobilization of neutral lipids helps the cell to conserve energy and building blocks for membrane biogenesis for conditions of nutritional depletion or cellular stress. Yeast LPs are equipped with a very specific subset of proteins. Many of them being involved in neutral lipid metabolism, e.g., Dga1p, Tgl3p, Tgl4p, Tgl5p, Tgl1p, and Yeh2p, or in ergosterol biosynthesis (Erg1p, Erg6p, Erg7p) (2, 9). Interestingly, some of these proteins such as Erg1p or Dga1p are dually located within the cell, namely in LP and the ER. How such proteins can be localized to two different types of membrane, a monolayer in the LP and a bilayer in the ER, is not yet understood. This observation also raises the question as to the specificity of the membrane lipid composition in these two compartments, which might be important to understand the relationship between the two organelles. In this chapter, we will describe methods commonly used for the analysis of lipids from yeast LP, which can also be adapted to other cell types. These techniques are versatile and can be easily performed. As a starting point for LP analysis, the isolation procedure of this organelle and the quality control of purified organelle samples will be explained. Then, we will describe lipid extraction and quantification of individual lipid components based on thin-layer chromatography (TLC), gas–liquid chromatography (GLC), mass spectrometry (MS) and GLC/MS.

2. Material 2 .1. Equipment and Supplies

1. Microsyringe (Hamilton, Bonaduz, Switzerland) or sample applicator (CAMAG, Automatic TLC Sampler IV, Muttenz, Switzerland).



Analysis of Lipid Particles from Yeast

361

2. 12-ml Pyrex glass vials with Teflon liner caps. 3. Glass tubes with ground neck. 4. Dounce Homogenizer and pestle. 5. Silica gel 60 TLC plates (Merck, Darmstadt, Germany). 6. Ultra-Clear Centrifuge Tube (Beckman). 7. TLC chamber (Springfield Mill, UK) with saturation paper (e.g., Whatman filter paper). 8. Iodine vapor chamber. 9. Incubator (Heraeus). 10. TLC Scanner (Shimadzu chromatoscanner CS-930). 11. Table-top centrifuge (Hettich Rotina 46 R, Heraeus Fresco17). 12. Sorvall RC5 Plus or RC6 Plus, and SLC3000 or SS34 rotors. 13. Ultracentrifuge (Sorvall – combi plus) with AH629 swing out rotor. 14. GLC-MS (Hewlett-Packard 5,890 Gas-Chromatograph). 15. LC-FTICR-MS coupled to an Accela HPLC. 2.2. Reagents

1. Medium for yeast cultivation: YPD (2% glucose, 2% peptone and 1% yeast extract) SD (2% glucose, 0.67% yeast nitrogen base and amino acid mixture) YPO (0.1% yeast extract, 0.5% peptone, 0.5% KH2PO4, 0.1% glucose, 0.2% Tween 80, 0.1% oleic acid). 2. Zymolyase-20 T (Seikagaku corporation, Japan). 3. Ficoll 400 (Sigma). 4. Chemiluminescence solution: SuperSignal™ (Pierce Chemical Company, Rockford, IL, USA). 5. Solvents: chloroform and methanol, analytical grade. 6. Washing solutions 0.034% MgCl2; 2N KCl/MeOH (4:1; v/v); artificial upper phase (chloroform/methanol/water; 3:48:47; per vol.). 7. Charring solution. 0.63-g MnCl2·4 H2O, 60-ml water, 60-ml methanol, 4-ml conc. H2SO4.

3. Methods 3 .1. Isolation of Lipid Particles from the Yeast

1. LPs are isolated from 4 to 5 l of full or selective media (Subheading 2.2, item 1; see also Note 5). Cells are inoculated from a 48 h preculture to an OD600 of 0.1 and grown to stationary phase at 30°C with shaking.

362

Connerth et al.

2. After harvesting at 4,000 g for 5 min in SLC3000, cells are washed with distilled water. 3. Cells (0.5 g CWW/ml) are incubated with dithiothreitol (DTT; 0.66 mg/ml) in buffer SP-A (0.1 M Tris/SO4, pH 9.4) for 10 min at 30°C with shaking. 4. Then, cells (0.15 g CWW/ml) are washed and suspended in prewarmed SP-B (1.2 M sorbitol, 20 mM KH2PO4, pH 7.4). Cell walls are enzymatically digested with Zymolyase-20T (Seikagaku Corporation, Japan) at a concentration of 2 mg enzyme per g CWW for at least 1 h at 30°C with shaking. 5. Resulting spheroplasts are washed twice in buffer SP-B and then resuspended in buffer LP-A (12% Ficoll, 10 mM MES/Tris pH 6.9, 0.2 mM Na2EDTA·2H2O) followed by mechanical disruption in the presence of the protease inhibitor phenylmethylsulfonyl fluoride (PMSF; 1 mM final) with 30 strokes using a 30-ml Dounce Homogenizer with a loose fitting pestle. 6. The homogenate is centrifuged to remove cell debris in a Sorvall SS34 rotor at 6,000 g for 5 min. 7. Supernatants are collected, and steps 6 and 7 are repeated. 8. Combined supernatants are carefully overlaid with buffer LP-A in an Ultra-Clear Centrifuge Tube (Beckman). Ultracentrifugation at 140,000 g for 45 min using a swing out rotor AH-629 yields a white layer on top (crude LP) that can be transferred with a moistened spatula to a 15-ml Dounce Homogenizer. 9. After homogenizing with eight strokes using a loose fitting pestle in the presence of 1 mM PMSF the sample is loaded onto a new ultracentrifuge tube and carefully overlaid with buffer LP-B (8% Ficoll, 10 mM MES/Tris pH 6.9, 0.2 mM Na2EDTA·2H20). Ultracentrifugation at 140,000 g for 30 min results in a top layer containing LP (see Note 6). 10. Prior to the last ultracentrifugation step, buffer LP-D (0.25 M sorbitol, 10 mM MES/Tris pH 6.9, 0.2 mM Na2EDTA·2H2O) is filled into a fresh ultracentrifuge tube. The homogenized sample is loaded to the bottom of the tube by injection under the buffer using a syringe. 11. Ultracentrifugation at 140,000 g for 30 min leads to a top layer consisting of highly purified LP. After homogenizing the isolated LP in a 5-ml Dounce Homogenizer the samples can be stored at −80°C until required. If desired the pellet from the last centrifugation step which contains vacuoles can be collected as well.



3 .2. Lipid Particle Analysis 3.2.1. Protein Analysis

Analysis of Lipid Particles from Yeast

363

1. When applying “delipidation” to isolated LP fractions, samples are incubated with 2 volumes of diethyl ether with repeated vigorous shaking. After a high speed centrifugation step using a table-top centrifuge, the extracted non-polar lipids are withdrawn and the remaining traces of diethyl ether are removed under a stream of nitrogen. 2. Proteins are precipitated with trichloroacetic acid, and the resulting pellets are either dissolved in onefold Laemmli buffer for gel electrophoresis (10) or solubilized in 0.1% SDS/0.1 M NaOH for protein quantification. 3. Proteins are quantified using the method of Lowry et al. (11) with bovine serum albumin as a standard. Typical samples of LP fractions contain approximately 0.2–0.02 mg protein/ml depending on strain and growth conditions. 4. Samples must not be dissolved at temperatures higher than 37°C as such a treatment would result in hydrolysis of proteins or in the formation of aggregates. 5. SDS-polyacrylamide gel electrophoresis is performed as described by Laemmli (10). 6. Commassie Blue staining of gels is usually sufficient to visualize protein bands, though more sensitive staining procedures (Silver staining, inverse protein staining) may be required to detect proteins of low abundance. 7. Western blot analysis is performed according to the method of Haid and Suissa (12). 8. A set of antibodies representing typical marker proteins of various cellular organelles is used to check the quality of the isolated LP (Table 1 and related results in Fig. 1, below). 9. Peroxidase conjugated secondary antibody (Sigma) and enhanced chemiluminescent signal detection reagents (Super Signal™, Pierce Chemical Company, Rockford, IL, USA) are used to visualize immunoreactive bands. 10. When LP samples appear to be contaminated with other subcellular compartments, a further purification step can be introduced after the standard LP isolation procedure (see Note 2).

3.2.2. Lipid Extraction

1. Lipids from LP are extracted using the method of Folch et al. (13). 2. In brief, an aliquot of the LP sample (~0.1 mg protein) is added to 3 ml of CHCl3:MeOH (2:1 ; v/v) in a Pyrex glass tube. 3. Lipids are extracted to the polar organic phase by vortexing at room temperature (RT) for 1 h.

364

Connerth et al.

Table 1 Marker antibodies used for the quality control of lipid particles Marker protein

Subcellular localization

Function

Erg1p

LP, ER

Squalene epoxidase

Erg6p

LP, ER

∆ (24)-Sterol C-methyltransferase

Erg7p

LP, ER

Lanosterol synthase

Ayr1p

LP, ER

NADPH-dependent 1-acyl dihydroxyacetone phosphate reductase

Prc1p

Vac

Vacuolar carboxypeptidase Y

Por1p

Mt

Mitochondrial porin

Wbp1p

ER

b-Subunit of the oligosaccharyl transferase (OST) glycoprotein complex

Sec61p

ER

Integral ER required protein for protein import

LP lipid particle, ER endoplasmic reticulum, Mt mitochondria, Vac vacuole

Fig. 1. Quality control of lipid particles. (a) Protein pattern of homogenate (lane 1), cytosol (lane 2), lipid particles (lane 3), endoplasmic reticulum (lane 4), mitochondria (lane 5) and vacuoles (lane 6). Proteins were separated on a 12.5% SDS gel as described in Subheading 3. (b) For Western blot analysis, 15 mg protein from each fraction were separated by electrophoresis, blotted, and probed with the respective antibodies.



Analysis of Lipid Particles from Yeast

365

4. It is very important to remove all protein aggregates by discarding the upper aqueous phase as well as the protein interface layer. Alternatively, the lower polar phase can be transferred to a fresh Pyrex tube by using a glass pipette. 5. Proteins and nonpolar substances are removed by consecutive washing steps with 0.2 volumes 0.034% MgCl2, 1 ml of 2 N KCl/MeOH (4:1; v/v), and 1 ml of an artificial upper phase (CHCl3:MeOH:H2O; 3:48:47; per volume). These solutions are added to the extracts and incubated with shaking for 3 min. 6. After centrifugation for 3 min at 1,500 g in a table-top centrifuge, the aqueous phase is removed by aspiration. 7. Washing steps are repeated until no protein layer is formed any more. 8. Finally, lipids are dried under a stream of nitrogen and stored at −20°C. 9. For lipid analyses, 1D and 2D TLC can be performed to separate different lipid classes and species according to their properties (see Subheadings 3.2.3 and 3.2.4). 3.2.3. Thin-Layer Chromatographic Analysis of Phospholipids

1. Phospholipids are separated by TLC due to different properties of their headgroups. TLC plates can be loaded with a Hamilton syringe or with a sample applicator. For the analysis of PI, PC, and PE, lipids are separated by one-dimensional (1D) TLC using chloroform/methanol/25% ammonia solution (50:25:6; per vol.) as a solvent. Separations usually take 50 min/10 cm of distance on TLC plates. 2. A better separation of phospholipids can be achieved by 2D TLC. The sample is applied as single spot to a TLC plate approximately 1–1.5 cm distant from the corner. For the first dimension, chloroform/methanol/25% ammonia (65:35:5; per vol.) is used as a solvent, and for the second dimension chloroform/acetone/methanol/acetic acid/water (50:20:10:10:5; per vol.). 3. Phospholipids are visualized by staining with iodine vapor in a saturated chamber for some minutes. For destaining, TLC plates are incubated in a heating chamber for a few minutes. 4. Phospholipids can be quantified from TLC plates after removal of the iodine staining. The plate is moistened with deionized water, phospholipid spots are scrapped off and transferred to a phosphate free glass tube with ground neck. 5. The lipid phosphorus of the respective spot can be measured by subjecting the sample to hydrolysis. Therefore, 0.2 ml of conc. H2SO4/72% HClO4 (9:1; v/v) is added to each sample. Hydrolysis is performed at 180°C in a heating block for 30 min. Please note that this step has to be performed in a hood due to formation of acidic fumes!

366

Connerth et al.

6. Samples are cooled to RT, and 4.8 ml of freshly prepared 0.26% ammonium molybdate/ANSA (500:22; v/v) is added. ANSA consists of 40.0 g of K2S2O5, 0.63 g of 8-anilio-1-naphthalenesulfonic acid, and 1.25 g of Na2SO3 in 250 ml of water. Tubes are sealed with phosphate-free glass caps and after vigorous vortexing samples are heated to 100°C for 30 min in a heating chamber. 7. Finally, samples are cooled to RT and shortly centrifuged in a table-top centrifuge at 1,000 × g to sediment the silica gel. The intensity of the blue color is a measure for the lipid phosphorus. Samples are measured spectrophotometrically at a wavelength of 830 nm using a blank spot from the TLC plate without phospholipid as a control. Data are calculated from a standard curve using inorganic phosphate at known amounts. 3.2.4. Thin-Layer Chromatographic Analysis of Neutral Lipids

1. Neutral (nonpolar) lipids are extracted as described in Subheading 3.2.2. The separation of different classes of neutral lipids can be performed by TLC. For the direct densitometric quantification of lipids on the TLC plate, authentic standards are used containing defined amounts of the respective lipids. 2. Dried lipid extracts are dissolved in an appropriate volume of CHCl3/MeOH (2:1; v/v) and spotted onto a TLC plate with a Hamilton syringe or a TLC sample applicator (see above). 3. For a most efficient TLC separation of neutral lipids a twostep separation system is recommended. Initially, lipids are separated on a 10-cm TLC plate in an ascending manner using light petroleum/diethyl ether/acetic acid (70:30:2; per vol.) as a solvent until the front has reached two-thirds of the plate’s height. Then, separation of lipids is continued in the same direction using light petroleum/diethyl ether (49:1; v/v) until the solvent front reaches the top of the plate. 4. After the TLC plate has been dried, sterols can be quantified densitometrically using a TLC scanner (Shimadzu chromatoscanner CS-930) at 275 nm. Longer exposure of the TLC plate to light and oxygen causes oxidation of sterols that should be avoided. 5. Other neutral lipids are irreversibly stained by charring prior to scanning. For this purpose, the TLC plate is incubated approximately 15 s in a solution of 0.63 g MnCl2·4H2O, 60 ml of water, 60 ml of methanol, 4 ml of conc. sulfuric acid and stained in a heating chamber at 100°C for 30 min. The staining intensity depends on the incubation time. The scanning procedure should be performed directly after charring as the intensity of the spots is fading with time (also see Note 6).



Analysis of Lipid Particles from Yeast

367

6. Spots of neutral lipids visualized as described above are scanned at a wavelength of 400 nm, and spot areas are quantified relative to standards. 3.2.5. Gas–Liquid Chromatographic Analysis of Fatty Acids

1. Fatty acids are analyzed by GLC after hydrolysis and conversion to methyl esters. 2. For this purpose, lipid extracts (see Subheading 3.2.2) are treated with BF3-methanol as described in the literature (14). 3. In detail, aliquots of lipid extracts (0.1–1 ml) are transferred to glass tubes and dried under a stream of nitrogen. After addition of 1 ml of BF3-methanol, samples are heated to 95°C in a sand bath for 10 min and then cooled to RT. 0.86 ml of benzene are added and tubes are heated again in the sand bath at 95°C for 30 min. 4. After cooling to RT, 1 ml of H2Odd and 3 ml of light petroleum are added, and samples are vortexed for 30 min. Then, samples are centrifuged at 1,000 g for 2 min in a table-top centrifuge (Hettich Universal 16), and the upper phase is transferred to a new Pyrex tube. The lower phase is extracted again as described above for 30 min with vortexing using 3 ml of light petroleum. After centrifugation the organic phases are combined and dried under a stream of nitrogen. 5. Then, samples are dissolved in 100 ml light petroleum and transferred to glass tubes. Finally, fatty acid methyl esters are separated by GLC using following parameters: GLC HP

6,890

Injector mode

split

Injection volume

1 ml

Column

HP-INNOWax Polyethylene Glycol; 15 m × 0.25 mm i.d. × 0.5-mm film thickness

Carrier

Helium, 5.0

Flow

1.4-ml linear velocity 30 cm/s constant flow

Oven

160°C (5 min) with 7.5°C/min to 250°C (15 min)

Detector

FID mode: constant makeup flow (40.0 ml/ min) Air flow: 400.0 ml/min Hydrogen flow: 35.0 ml/min Makeup gas type: Helium

368

Connerth et al.

6. Fatty acids are identified by comparison to commercial fatty acid methyl ester standards (NuCheck, Inc., Elysian, MN, USA). 3.2.6. Gas–Liquid Chromatography/Mass Spectrometry of Sterols

1. Sterol analysis is performed as described previously (15). 2. A mixture of 0.6 ml of methanol (Merck) and 0.4 ml of 0.5% (w/v) pyrogallol (Fluka; 4°C freshly dissolved in methanol and 0.4 ml of 60% (w/v) aqueous KOH solution) in a Pyrex tube is prepared. Then, 5 ml of a cholesterol solution (2 mg/ml in ethanol) is added as an internal standard. 3. Samples containing approximately 0.3–0.5 mg protein are added to the mixture and tubes are heated in a water bath at 90°C for 2 h. 4. Lipids are extracted with 1 ml of n-heptane. The upper phase is transferred into a fresh tube, and the lower phase is re-extracted twice. Combined upper phases are dried under a stream of nitrogen, and lipids are dissolved in 10 ml pyridine. 5. Samples are treated immediately prior to analysis with 10 ml N¢O¢-bis (trimethylsilyl)-trifluoracetamide (SIGMA), incubated at RT for 10 min and diluted with 30 ml ethyl acetate. 6. GLC–MS analysis is carried out using the following parameters: 7. Sterols are identified according to their retention time and mass fragmentation pattern using MSD ChemStation, D.03.00.552.

GLC

HP 5,890 SeriesII Plus with Electonic Pressure Control and 6,890 automated liquid sampler (ALS)

Injector

Split/splitless 270°C, mode: splitless. Purge on: 2 min

Injector volume

1 ml

Column

HP 5-MS (Crosslinked 5% Phenyl Methyl Siloxane); 30 m × 0.25 mm i.d. ×0.25-mm film thickness

Carrier

Helium, 5.0

Flow

0.9-ml linear velocity 35.4 cm/s, constant flow

Oven

100°C (1 min) with 10°C/min to 250°C (0 min) and with 3°C/min to 310°C (0 min)

Detector

Selective Detector HP 5,972 MSD

Ionization

EI, 70 eV

Mode

Scan, scan range: 200–550 amu, 3.27 scans/s

EM

Voltage Tune Voltage

Tune

Auto Tune

3.2.7. Mass Spectrometry of Neutral Lipids and Phospholipids

Analysis of Lipid Particles from Yeast

369

1. Lipids extracts are prepared as described above and diluted 1:100 in acetonitrile/2-propanol (5:2; v/v), 1% ammonium acetate, 0.1% formic acid. 2. 5 mM TAG (species 51:0) and PC (species 24:0) are added as internal standards. 3. For chromatographic separation a thermo hypersil GOLD C18, 100 × 1 mm, 1.9 mm column is used with solvent A (water with 1% ammonium acetate, 0.1% formic acid) and solvent B (acetonitrile/2-propanol, 5:2, v/v; 1% ammonium acetate; 0.1% formic acid). 4. Gradients are established from 35 to 70% B for 4 min and then to 100% B in another step of 16 min. These conditions are held for 10 min with a flow rate of 250 ml/min. 5. Mass spectrometry is performed with the following parameters:

MS

HTQ-FT coupled to and Accela UPLC

Data acquisition

FT-MS full scan at a resolution of 100 k and < 2 ppm mass accuracy with external calibration

Spray voltage

5,000 V

Capillary voltage

35 V

Tube lens

120 V

Capillary temperature

250°C

6. Peak areas are calculated by QuanBrowser for all lipid species identified previously by exact mass (<2 ppm) and retention time. Calculated peak areas for each species are expressed as % of the sum of all peak areas in the respective lipid class. 3.3. Results

1. Figure 1a shows a typical SDS-PAGE analysis of a cell fractionation experiment using the yeast wild type strain BY4741. LP proteins occur at low abundance in the cell, but comprise a distinct set of polypeptides (16). 2. The purity of LP preparation can be tested by Western blot analysis (Fig. 1b) using marker antibodies as described in Table 1. The enrichment of LP can be verified best with antiErg1p antibody (Fig. 1b, line 1). Contamination of LP with other organelles such as mitochondria (Por1p), ER (Sec61p) or vacuole (Prc1p) can be largely excluded since signals with the respective antibodies in the LP fraction are negligible. 3. Phospholipid analysis of LP has been described previously in the literature (16). In brief, total phospholipids from LP contained approximately 36% PC, 20% PE, 32% PI, 4% DMPE, and

370

Connerth et al.

Fig. 2. Mass spectrometric analysis of phospholipids from lipid particles. Phosphatidylcholine (a) and phosphatidylethanolamine (b) species were analyzed by MS as described in Subheading 3. LP were isolated from cells grown on YPD (black bars) and YPO (white bars). Species are formed as follows: 32:1 (16:1/16:0); 32:2 (16:1/16:1); 34:1 (16:1/18:0) or (16:0/18:1); 34:2 (16:1/18:1); 36:2 (18:1/18:1); 36:3 (18:1/18:2) or (18:0/18:3).

about 3% PA. To identify species (fatty acid composition) of the major phospholipids from LP, MS analyses can be performed. Figure 2 shows an example of PC and PE species analysis from LP isolated from wild type yeast grown on YPD (glucose) or YPO (oleic acid). PC was found to contain, primarily, C16:1/ C16:1 and C16:1/C18:1 fatty acids when cells were grown on glucose as carbon source (Fig. 2a). PC from cells grown on oleate contains mainly C18:1/C18:1 with only minor amounts of C18:1/C16:1. The PE species C16:1/C18:1 and C16:0/ C18:1 or C16:1/C18:0 were predominant when cells were grown on YPD (Fig. 2b). PE with C18:1/C18:1 was the major species when cells were cultivated on oleic acid (YPO). 4. Neutral lipids of LP were analyzed by TLC and identified by comparison to standard mixtures (Fig. 3a). As cholesteryl ester was used as standard for STE the Rf-value is slightly higher (Fig. 3a, lanes 1 and 2) than for fatty acyl esters of ergosterol and precursors (lanes 3 and 4). The neutral lipid



Analysis of Lipid Particles from Yeast

371

Fig. 3. Neutral lipids of lipid particles from wild type yeast cells grown on different carbon sources. (a) Neutral lipids were separated by TLC in a two-step procedure using light petroleum/diethyl ether/acetic acid (70:30:2; per vol.) as the first solvent approximately to one half of the distance on the plate, and light petroleum/diethyl ether (49:1; v/v) as the second solvent (see Subheading 3). The arrow indicates the ascending mode of TLC. Different neutral lipid classes are indicated at the left. SQ squalene, STE steryl esters, TAG triacylglycerols, FFA free fatty acids, ERG ergosterol. 1 and 2: standard mixtures containing the different neutral lipid classes; 3: lipid pattern of LP from cells grown on glucose; 4: lipid pattern of LP from cells grown on oleic acid. (b) Mass spectrometric analysis of TAG from LP of cells grown on YPD (black bars) or YPO (white bars), respectively. The relative amount of the signal for the different species is shown. The asterisk indicates three C18:1 chains in a TAG molecule.

pattern of LP changes depending on the carbon source used for the cultivation of cells. LP from cells grown on glucose contain large amounts of STE and approximately equal amounts of TAG, whereas LP from cells grown on oleic acid as carbon source mostly accumulate this fatty acid in TAG. Other nonpolar lipids are missing in wild type LP when cells are grown under standard conditions. The amounts of the different neutral lipids were quantified by densitometric scanning,

372

Connerth et al.

and relative amounts (mg lipid/mg protein) were calculated. LP from cells grown on YPD contain approximately 47.2% STE and 52.8% TAG, whereas LP from cells grown on YPO have 0.2% STE and 99.8% TAG (our own unpublished data). 5. TAG species analysis of the neutral lipid fraction from cells grown on YPD or YPO, respectively, is shown in Fig. 3b. TAG from cells grown on YPD mainly contain C16:1/ C16:1/C18:1 (50:3 species) and C18:1/C18:1/C16:1 (52:3 species), whereas the majority of TAG from cells grown on YPO is C18:1/C18:1/C18:1 (54:3 species). In general, most of the yeast TAG contains monounsaturated fatty acids. Polyunsaturated acyl chains were only found in TAG from cells grown on YPO, e.g., C18:1/C18:1/C18:2 (54:4 species) due to the presence of polyunsaturated fatty acids as impurity in the oleic acid used a carbon source. 6. Isolated LPs were analyzed for sterol composition as described previously (17). In brief, ergosterol was found to be the major sterol with approximately 70% of total sterols followed by the precursors zymosterol (10%), fecosterol (~7%) and episterol (~7%). 7. Fatty acid analysis of yeast neutral lipids has also been reported recently (17). It was found that TAG as well as STE contain mostly C18:1 (40–50%) followed by C16:1 (~35%) and only minor amounts of C16:0 and C18:0.

4. Notes 1. Delipidation of LP samples prior to TCA precipitation of proteins can be omitted, but disturbing effects during SDSPAGE may be observed. Washing the precipitated protein pellet with cold acetone helps to circumvent this negative effect and is recommended. 2. If the purity of a LP preparation does not show the desired quality due to contamination with other cellular membranes, a further purification step can be added to the standard protocol. For this purpose, LP preparations are treated with 4.5 M urea. After 15 min of incubation at RT, the last floating centrifugation step of the standard procedure is repeated, and highly purified LP can be collected from the top of the gradient.



Analysis of Lipid Particles from Yeast

373

3. When highly concentrated LP samples are needed, one further floating centrifugation step using a table-top centrifuge at maximum speed for app. 15–30 min can be performed. The excess amount of buffer under the LP layer at the top can be removed using a syringe. 4. Induction of LP proliferation with oleic acid increases the yield of LP markedly but at the same time alters the lipid composition dramatically as shown in the previous section. 5. If a yeast strain needs to be cultured on selective minimal media the yield of LP may be extremely low. Large culture volumes may be required to obtain LP at substantial quantities. 6. Neutral lipid analysis by densitometric measurement strongly depends on the intensity of the band color. Therefore, it is necessary to routinely compare bands to a standard loaded onto the same TLC plate. STE are stained more intensively than TAG by the charring method, which has to be taken into account when using standards. Moreover, bands should not be too broad to avoid problems during the scanning process.

Acknowledgments Work on yeast nonpolar lipid metabolism in our laboratory was recently supported by the Fonds zur Förderung der wissenschaftlichen Forschung in Österreich (projects 15141, 18857 and W901-B05 to G.D.).

Abbreviations CWW ER GLC LP MS PC PE PI PS RT STE TAG TLC

Cell wet weight Endoplasmic reticulum Gas–liquid chromatography Lipid particle(s)/droplet(s) Mass spectrometry Phosphatidylcholine Phosphatidylethanolamine Phosphatidylinositol Phosphatidylserine Room temperature Steryl ester Triacylglycerol Thin-layer chromatography

374

Connerth et al.

References 1. Athenstaedt, K. and Daum, G. (2005) Tgl4p and Tgl5p, two triacylglycerol lipases of the yeast Saccharomyces cerevisiae are localized to lipid particles. J. Biol. Chem., 280, 37301–37309. 2. Athenstaedt, K., Zweytick, D., Jandrositz, A., Kohlwein, S. D., and Daum, G. (1999) Identification and characterization of major lipid particle proteins of the yeast Saccharomyces cerevisiae. J. Bacteriol., 181, 6441–6448. 3. Zweytick, D., Athenstaedt, K., and Daum, G. (2000) Intracellular lipid particles of eukaryotic cells. Biochim. Biophys. Acta, 1469, 101–120. 4. Scow, R. O., Blanchette-Mackie, E. J., and Smith, L. C. (1980) Transport of lipid across capillary endothelium. Fed. Proc., 39, 2610– 2617. 5. Blanchette-Mackie, E. J., Dwyer, N. K., Barber, T., Coxey, R. A., Takeda, T., Rondinone, C. M., Theodorakis, J. L., Greenberg, A. S., and Londos, C. (1995) Perilipin is located on the surface layer of intracellular lipid droplets in adipocytes. J. Lipid Res., 36, 1211–1226. 6. Robenek, H., Hofnagel, O., Buers, I., Robenek, M. J., Troyer, D., and Severs, N. J. (2006) Adipophilin-enriched domains in the ER membrane are sites of lipid droplet biogenesis. J. Cell Sci., 119, 4215–4224. 7. Czabany, T., Athenstaedt, K., and Daum, G. (2007) Synthesis, storage and degradation of neutral lipids in yeast. Biochim. Biophys. Acta, 1771, 299–309. 8. Sandager, L., Gustavsson, M. H., Stahl, U., Dahlqvist, A., Wiberg, E., Banas, A., Lenman, M., Ronne, H., and Stymne, S. (2002) Storage Lipid Synthesis Is Non-essential in Yeast. J. Biol. Chem., 277, 6478–6482.

  9. Rajakumari, S., Grillitsch, K., and Daum, G. (2008) Synthesis and turnover of non-polar lipids in yeast. Prog. Lipid Res., 47, 157–171. 10. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680– 685. 11. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem., 193, 265–275. 12. Haid, A. and Suissa, M. (1983) Immunochemical identification of membrane proteins after sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Methods Enzymol., 96, 192–205. 13. Folch, J., Lees, M., and Stanley, G. H. S. (1957) A simple method for the isolation and purification of total lipides from animal tissues. J. Biol. Chem., 226, 497–509. 14. Morrison, W. R. and Smith, L. M. (1964) Preparation of fatty acid methyl esters and dimethylacetals from lipids with boron fluoride– methanol. J. Lipid Res., 5, 600–608. 15. Quail, M. A. and Kelly, S. L. (1996) The extraction and analysis of sterols from yeast. Methods Mol. Biol., 53, 123–131. 16. Leber, R., Zinser, E., Zellnig, G., Paltauf, F., and Daum, G. (1994) Characterization of lipid particles of the yeast, Saccharomyces cerevisiae. Yeast, 10, 1421–1428. 17. Czabany, T., Wagner, A., Zweytick, D., Lohner, K., Leitner, E., Ingolic, E., and Daum, G. (2008) Structural and biochemical properties of lipid particles from the yeast Saccharomyces cerevisiae. J. Biol. Chem., 283, 17065–17074.

Chapter 19 Mammalian Fatty Acid Elongases Donald B. Jump Summary Very long chain fatty acids confer functional diversity on cells by variations in their chain length and degree of unsaturation. Microsomal fatty acid elongation represents the major pathway for determining the chain length of saturated, monounsaturated, and polyunsaturated fatty acids in cellular lipids. The overall reaction for fatty acid elongation involves four enzymes and utilizes malonyl CoA, NADPH, and fatty acyl CoA as substrates. While the fundamental pathway and its requirements have been known for many years, recent advances have revealed a family of enzymes involved in the first step of the reaction, i.e., the condensation reaction. Seven fatty acid elongase subtypes (Elovl #1–7) have been identified in the mouse, rat, and human genomes. These enzymes determine the rate of overall fatty acid elongation. Moreover, these enzymes also display differential substrate specificity, tissue distribution, and regulation, making them important regulators of cellular lipid composition as well as specific cellular functions. Herein, methods are described to measure elongase activity, analyze elongation products, and alter cellular elongase expression. Key words: Fatty acid elongase, Microsome, Reverse phase-high performance liquid chromatography (RP-HPLC), Recombinant adenovirus

1. Introduction Fatty acid elongation occurs in three cellular compartments: the cytosol, mitochondria, and endoplasmic reticulum (microsomes). In the cytosol, fatty acid elongation is part of de novo lipogenesis and involves acetyl-CoA carboxylase and fatty acid synthase. Fatty acid synthase utilizes acetyl CoA and malonyl CoA to elongate fatty acids by two carbons. The primary end product of de novo lipogenesis is palmitate (16:0) and to a lesser extent myristate (14:0) and stearate (18:0). Fatty acid elongation also occurs in

Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_19, © Humana Press, a part of Springer Science + Business Media, LLC 2009

375

376

Jump O C R

+

HS-CoA

OH

Fatty Acid

Acyl CoA Synthetase

ATP O

O

O

R

3-Keto Acyl CoA Synthase [Elovl]

C

+

C

C

SCoA

HO

SCoA

Complex Lipids

CH2

Malonyl CoA

Acyl CoA O

CO2 + HS-CoA O O

C C

C

CH2

SCoA

2HC

SCoA

CH2

Elongated Acyl CoA

3-Keto Acyl CoA R

R

NADPH NADP

3-Keto Acyl CoA Reductase

NADP

O

O HO

C CH

3-Hydroxy Acyl CoA

Trans 2,3 Enoyl CoA Reductase

NADPH

R

SCoA

H2O

H

C C

CH2

3-Hydroxy Acyl CoA Dehydratase

R

SCoA

CH

2,3-Trans Enoyl CoA

Fig. 1. Proposed reaction sequence for hepatic microsomal fatty acid elongation. The reaction scheme described below is a modification of the one described by Cinti et al (2), Moon and Horton (12), and Denic and Weissman (10).

mitochondria and utilizes an enoyl-CoA reductase plus acetyl CoA and fatty acyl CoA as substrates. This site appears to be a minor pathway for fatty acid elongation in eukaryotic cells (1). Microsomal fatty acid elongation is considered the predominant pathway for elongating fatty acids 12-carbons and longer (2–10). This pathway utilizes fatty acids derived from endogenous pathways, like de  novo lipogenesis, as well as exogenous fatty acids derived from the diet. The overall pathway for microsomal fatty acid elongation involves four enzymes and fatty acyl CoA, malonyl CoA, and NADPH as substrates (Fig. 1). In step 1, a 3-keto acyl-CoA synthase catalyzes the condensation of malonyl CoA with a fatty acyl-CoA precursor. In step 2, a 3-keto acyl-CoA reductase reduces the resulting 3-keto intermediate. Step 3 involves the dehydration of the 3-hydroxy species; this reaction is catalyzed by a 3-hydroxy acyl-CoA dehydratase. In step 4, reduction of the step 3 product is catalyzed by trans-2,3-enoyl-CoA reductase



Mammalian Fatty Acid Elongases

377

(2, 9). Cloning and genomic analysis have identified seven distinct 3-keto acyl-CoA synthases, i.e., fatty acid condensing enzymes, in mouse, rat, and human genomes (http://www.ensemble.org) (3, 4). These enzymes are called elongases and have been given the designation Elovl for elongation of very long fatty acids (Elovl 1–7). The elongase subtypes, their mouse accession # and alias are: Elovl-1: NM_019422, SSC1; Elovl-2: AF170908, SSC2; Elovl-3: BC016468, CIG30; Elovl-4: NM_148941, Elo4; Elovl5: NM_134255, FAE1; Elovl-6: AY053453, LCE or FACE; Elovl-7: BAB31310, 9130013K24RIK. The two reductases, 3-keto acyl-CoA reductase and trans-2,3-enoyl-CoA reductase, have been cloned from human, mouse, and yeast (9, 11). The dehydratase, however, has not been cloned. Fatty acyl-CoA substrate specificity and rate of fatty acid elongation is determined by the first step in the pathway, i.e., the activity of the condensing enzyme and not the reductases or dehydratase (1, 9). While Elovl-1, Elovl-3, and Elovl-6 elongate saturated and monounsaturated fatty acids (3, 4, 12). Elovl-2, Elovl-4, and Elovl-5 elongate polyunsaturated fatty acids. Elovl-5 also elongates some monounsaturated fatty acids, like palmitoleic acid (8). Elovl-5 specifically elongates g-linolenoyl-CoA (18:3,n-6 CoA), while Elovl-2 specifically elongates 22-carbon PUFA (3, 4, 8). Substrates for Elovl-7 have not been described. The elongases (Elovl) are expressed differentially in mammalian tissues (3, 4). For example, five elongases are expressed in rat and mouse liver, including Elovl-1, -2, -3, -5, -6 (6, 7, 13). In contrast, the heart expresses Elovl-1, -5, and -6, but not Elovl-2 (14). These enzymes are also regulated by diet, hormones, during development, and in chronic disease (3, 7–9, 12). As such, changes in their activity impact cellular lipid composition. Fatty acid elongases function with fatty acid desaturases to generate many of the long chain mono and polyunsaturated fatty acid assimilated into cellular lipids. Elovl-6 and D9-desaturase generate 18-carbon monounsaturated fatty acids like oleic acid (8, 12), while Elovl-2 and Elovl-5 along with D5 and D6 desaturases generate the end products of n-6 and n-3 synthesis, i.e., arachidonic and docosahexaenoic acid (3, 4, 8, 15). This chapter describes several methods to study elongase expression and function in cells and tissues. First is a detailed description of methods to measure fatty acid elongase activity in rat and mouse liver microsomal preparations. Second is an approach to measure fatty acid elongation in cultured cells, like rat primary hepatocytes. Third is an approach to modify the expression of fatty acid elongases in cultured cells and mouse liver.

378

Jump

2. Materials 2.1. Equipment

1. Wheaton Potter–Elvehjem tissue grinder (15 ml) with teflon pestel and a Wheaton overhead homogenizer (Fisher) 2. Nalgene 50 ml polypropylene centrifuge tubes (cat. #3110–0500) 3. 12-ml pyrex tubes with caps (#9825, Fisher) 4. 15-ml Kimble HS centrifugation tube (45500–15, Fisher) 5. 12 × 35-mm lipid storage vials with Teflon cap liners (cat.# 03–338AA, Fisher). 6. JA20.1 high-speed centrifuge rotor (Beckman Instruments) 7. SW55 TI ultracentrifuge rotor (Beckman Instruments) 8. Avanti high speed centrifuge (Beckman Instruments) 9. Optima L-90 ultracentrifuge (Beckman Instruments). 10. 25-mm cell scraper (Sarstedt, cat.# 83.1830) 11. Nitrogen evaporator: N-Evap 112 (Organomation Associates, Inc.) 12. Microcentrifuge evaporator: CentriVap Concentrator plus cold trap and organic solvent trap (Labconco) 13. Waters 600 controller (Waters Corporation) 14. Waters 2487 UV/Vis absorbance monitor 15. Waters 2420 evaporative light scatter detector 16. YMC J’Sphere ODS-H80, (4 mm, 8 nm; 250 × 4.6 mm I.D.) (Waters Corporation) 17. Flow-through b-scintillation counter, b-RAM, (IN/US Systems, Inc., Tampa, Fl) 18. Winflo software, (IN/US Systems, Inc., Tampa, Fl)

2.2. Reagents and Supplies

1. Protease inhibitors: Complete Mini, EDTA-free (Cat. # 11836170001, Roche Diagnostics, Indianapolis , IN) 2. Phosphatase inhibitors: b-glycerol phosphate, sodium pyrophosphate, Na3VO4 (Sigma-Aldrich) 3. Rotenone, malonyl CoA, NADPH, coenzyme A, butylated hydroxytoluene, porcine insulin, (Sigma-Aldrich) 4.

C-malonyl CoA (40–60 Ci/mol) (Perkin Elmer)

14

5. Ultima Gold scintillation solution (Perkin Elmer) 6. Fatty acyl CoA (Avanti Polar Lipids, Alabaster, Al) 7. Probumin, fatty acid- and endotoxin-free bovine serum albumin (Celliance, Cat# 820473; Kankakee, Ill).



Mammalian Fatty Acid Elongases

379

8. Non-esterified fatty acids, (Nu-Chek Prep, Elysian, MN) 9. HPLC-grade solvents: chloroform, methanol, acetonitrile, hexane, diethyl ether (Fisher) 10. Bio-Rad Protein assay (Quick start/Bradford, cat. # 500–0205). 11. BioCoat (type 1 collagen) cell culture plates (Beckon Dickinson, Belford, MA.) 12. Williams E medium, fetal bovine serum, glucose [dextrose], phosphate-buffered saline (Invitrogen).

3. Methods 3.1. Measuring Fatty Acid Elongase Activity in Microsomes 3.1.1. Preparation of Mouse or Rat Liver Microsomes

1. Homogenize 0.2–0.5 g liver (freshly isolated liver or snap frozen liver) in 5 ml of Buffer A [250 mM sucrose + 10 mM Tris–Cl, pH 7.4, containing protease (Complete Mini, EDTA-free, Cat. # 11836170001, Roche Diagnostics, Indianapolis, IN) and phosphatase inhibitors (1-mM b-glycerol phosphate, 2.5-mM sodium pyrophosphate, 1-mM Na3VO4 (Sigma)] using a 15-ml Wheaton Potter–Elvehjem tissue grinder with a teflon pestle attached to a Wheaton overhead stirrer (Fisher). Homogenize at a low speed (setting of 3–4) for ~2–3 min while keeping the homogenizing flask on ice. Homogenize the tissue thoroughly. 2. Transfer the homogenate to 50 ml of Nalgene polypropylene centrifuge tubes (cat.# 3110–0500) and centrifuge at 12,000 rpm (18,547 x g) for 10 min using a JA20.1 rotor and a Beckman Avanti refrigerated centrifuge (4°C). 3. Recover supernatant and centrifuge the supernatant at 40,000 rpm (194,432 x g) using a Beckman SW55 TI rotor for 30 min in a Beckman Optima L-90 ultracentrifuge at 4°C. 4. Discard the supernatant and resuspend the pellet in Buffer A [0.25 M sucrose + 10 mM Tris–Cl, pH 7.4, plus protease and phosphatase inhibitors] at ~1 ml/g liver. The pellet can be difficult to resuspend and may need to be homogenized gently with a glass rod or Teflon pestle. 5. Determine protein concentration using the Bio-Rad Protein assay (Quick start/Bradford, cat. # 500–0205). 6. Aliquot in labeled tubes and store frozen at −80°C. Microsomal fatty acid elongase activity is stable at least 1 month when stored at –80°C.

380

Jump

3.1.2. Enzymatic Assay for Fatty Acid Elongation Activity in Rat Liver Microsomes

The fatty acid elongase reaction is carried out in 1.5–1.8-ml screw-capped microfuge tubes (12). Final Concentration

Volume/amount added

100 mM KH2PO4, pH 6.5

50 mM

50 ml

1 mM rotenone (Sigma)

5 mM

0.5 ml

10 mM malonyl CoA (Sigma)

60 mM

0.6 ml

10 mCi/ml 14C-malonyl CoA (Perkin Elmer)

15 nCi

1.5 ml

10 mM Fatty acyl CoA (in 50 mM KH2PO4, pH 5.8) (Avanti Polar Lipids)

40 mM

0.5 ml

22 mM NADPH (Sigma) (made fresh each week in 50mM KH2PO4, pH 5.8)

1 mM

5 ml

400-mM bovine serum albumin (fatty acid and endotoxin free in HOH, Probumin from Celliance, Cat# 820473; Kankakee, Ill).

20 mM

5 ml

Microsomal protein

50 mg/0.1 ml

50 mg

Components

Adjust final volume to 100 ml with water. Mix and incubate 37°C. for 20 min. 3.1.3. Extraction and Analysis of Fatty Acid Elongation Reaction Products.

Stop the reaction with the addition of 100 ml of 5 M KOH-10% methanol. Incubate for 1 h at 65°C. This step will hydrolyze (saponify) the ester bond between the fatty acid and coenzyme A. Stop the saponification step by adding 100 ml of 5 M HCl and 100 ml of ethanol to each reaction. Extract fatty acids by adding 750 ml of hexane–acetic acid mix (98% hexane–2% glacial acetic acid), vortex vigorously for ~1 min. Separate the aqueous (lower) and organic (upper) phases by centrifugation in a microfuge (5 min at maximum speed). Transfer upper phase (hexane phase) to a scintillation vial. Repeat hexane extraction and combine the hexane supernatants. Add 5 ml of scintillation cocktail (UltimaGold, Perkin Elmer) and count radioactivity.

3.1.4. Control Studies for Fatty Acid Elongation Assays.

Several important controls should be run for each study. First, NADPH is required for microsomal fatty acid elongation. In a control reaction, add all the reagents except NADPH. Radioactivity recovered in the hexane–acetic acid phase from this reaction represents background 14C radioactivity. Second, add 1.5 ml of 14C-malonyl CoA directly to the scintillation vial. Once



Mammalian Fatty Acid Elongases

381

background subtractions are made, this will represent 100% of 14 C. Since the reaction contains 6 nmol of malonyl CoA, a simple % calculation of total radioactivity appearing in the hexane–acetic acid phase will indicate the nmoles of malonyl CoA transferred to a fatty acid. 3.1.5. Data Presentation for Fatty Acid Elongation Assays

Express the results as nmoles/mg microsomal protein. When using 18:3(n-6) CoA as substrate and rat liver microsomes, expect ~3–5 nmol of substrate (malonyl CoA) converted to product (fatty acid)/mg protein/20 min reaction.

3.1.6. Follow-Up Studies.

With few exceptions, the product of the fatty acid elongase reaction is the result of the action of two or more fatty acid elongases. For example, 16:0-CoA is a substrate for Elovl1, Elovl-3, and Elovl-6, while 20:4(n6)-CoA is a substrate for Elovl-2 and Elovl-5. It is possible to measure Elovl-2 and Elovl-5 specifically using 22:5(n3)-CoA and 18:3(n6)-CoA as substrates, respectively. As such, an important step in elongase analysis is to establish the elongase subtype expression in cells under study. This can be accomplished using quantitative reverse transcriptase-polymerase chain reaction. Primers for several mouse, rat, and human elongases have been reported elsewhere (8).

3.1.7. Alternative Substrates for Fatty Acid Elongation.

Because the commercial availability of fatty acyl CoAs is limited, nonesterified fatty acids can be substituted in the reaction. For details of this modification of the elongase assay, see Note 1.

3.2. Assessment of Fatty Acid Elongation in Cultured Cells

Addition of nonesterified fatty acids to cells leads to their rapid uptake, metabolism, and assimilation into complex lipids. Many fatty acids undergo elongation and desaturation. The products of these reactions can be monitored using a combination of 14C-labeled fatty acids, cell culture, total lipid extraction, and analysis of saponified fatty acids by reverse phase-high performance liquid chromatography (RP-HPLC) coupled to a flowthrough b-scintillation counter (8, 15, 16).

3.2.1. Studies with Rat Primary Hepatocytes.

Rat primary hepatocytes are treated with 14C-fatty acids to assess fatty acid elongation. Primary hepatocytes are prepared from Teklad chow-fed (ad lib) male Spraque–Dawley rats and cultured on 6-well BioCoat (type 1 collagen) plates (Beckon Dickinson, Belford, MA.) (16). Cells are incubated in Williams E medium (Invitrogen) containing 25 mM glucose, 1 microM insulin (Invitrogen, Carlsbad, CA); 25 mM bovine serum albumin (endotoxin and fatty acid free albumin from Celliance, Kankakee, Ill), and 100 mM 14C-fatty acid [1.6 Ci/mol] (16). See Note 2 regarding fatty acid treatment of other cell types.

382

Jump

3.2.2. Lipid Extraction and Analysis of Fatty Acid Metabolites.

Total lipid extraction and analysis of fatty acids is carried out after treating cells with 14C-fatty acids. At harvest, remove the media from cells and wash cells once with cold phosphate-buffered saline (PBS) containing 200 mM BSA followed by a wash with PBS alone. After removal of the PBS, resuspend cells in 0.5 ml of 40% methanol + 0.1N HCl; use a 25-mm cell scraper (Sardstedt) to remove cells from the plastic culture dish. Transfer the resuspended cells to 15-ml glass centrifuge (Kimble HS No. 45500.15) tube (see Note 3). If a protein assay is required, then 2 ml of the cell suspension can be taken for a protein assay (BioRad, Hercules, CA). Add 2 ml of chloroform:methanol (2:1) plus 1 mM butylated hydroxytoluene (BHT) (Sigma-Aldrich) to the cell suspension and vortex carefully, avoid loss of sample. After thorough mixing, the sample is centrifuged in a JA20.1 rotor at 5,000 rpm (3220 x g) using a Beckman Avanti refrigerated (4°C) centrifuge. The organic (lower) phase is recovered and transferred to a 12-ml screw-cap tube (Fisher Pyrex #9825). The aqueous phase and protein interface is re-extracted with 2-ml chloroform as before. After centrifugation, the organic phase is transferred to the same 12-ml screw-capped tube. The organic solvents are removed by a constant stream of nitrogen using a nitrogen evaporator (Organomation Associates, Inc). All the moisture is removed using a microcentrifuge evaporator (CentriVap + cold and organic solvent traps, Labconco) overnight. The next day, dissolve lipids in 500 ml of chloroform containing 1 mM BHT. Lipids are transferred to storage vials (12 × 35 mm, Fisher cat # 03338AA), labeled, and stored at –80°C. At this point, total lipid extracts can be fractionated by thin-layer chromatography to assess the distribution of 14C in complex lipids or saponified for analysis of fatty acid by RP-HPLC (15).

3.2.3. Saponification of Lipids

1. Transfer ~100 ml or >30,000 CPM of the 14C-label chloroform lipid extract to a clean 12-ml Pyrex tube with a teflonlined screw cap. Remove the organic solvent with a stream of nitrogen. 2. Resuspend the dried lipid in 500-ml saponification buffer (0.4N NaOH in 80% methanol). Screw on the cap and heat the tube to 65°C for ~1 h. 3. Let the tubes cool and add 500 ml of 0.45 M HCl to neutralize. Add 2 ml of hexane containing acetic acid at 2%; cap and vortex vigorously. 4. Let the tube sit for ~3 min or until the aqueous and organic phases have separated. Recover the organic (upper) phase and transfer to a clean 12-ml pyrex tube. 5. Repeat the hexane–acetic acid extraction as before. Vortex, let sit for phase separation, and recover the upper phase; pool the organic phases.



Mammalian Fatty Acid Elongases

383

6. Remove the organic solvent under a stream of nitrogen; dry the sample overnight in as above in a microcentrifuge evaporator. 7. Resuspend the saponified lipids in 100 ml of methanol containing 100 mM BHT. Transfer methanol soluble lipids to storage vials for –80°C storage or use for RP-HPLC. Quantify lipid recovery by b-scintillation counting. 3.2.3. Alternative Methods to Examine Elongation Products..

An alternative method for analysis of elongation products involves treating cells with saponification buffer (0.4N NaOH in 80% methanol). In this method, cells grown in 6-well plates are resuspended in 500 ml of saponification buffer per well of a 6-well plate. Transfer the resuspended cells to a 12-ml pyrex tube, capped, and heated as described above (Subheading 3.2.3, step 2). Follow steps 2–7 to prepare fatty acids for RP-HPLC.

3.2.5. High Performance Liquid Chromatography.

The equipment used for RP-HPLC includes a Waters 600 controller (Milford, MA), Waters UV/Vis absorbance monitor [model 2487], Waters evaporative light scatter detector [model 2420], and a b-RAM flow through b-scintillation counter (IN/ US Systems, Inc., Tampa, Fl). The reverse phase column is a YMC J’Sphere (ODS-H80, 4 mm, 8 nm; 250 × 4.6 mm I.D.) (Waters Corporation). Eluate from the column passes through the UV detector, exits and is split 1:1 for flow into the evaporative light scatter detector and the flow through b-scintillation counter. The scintillation cocktail for the IN/US unit is Inflow 2:1 (IN/US Systems, Inc.). The ratio of eluant to scintillation cocktail is 2:1. The software for data collection and analysis is Winflo (IN/US Systems, Inc.). Samples (20 microl) are injected and fatty acids are eluted from the column using a mobile phase consisting of a linear gradient starting at 77.5% acetonitrile + 22.4% water + 0.1% acetic acid and ending at 99.9% acetonitrile + 0.1% acetic acid over 90 min with a flow rate of 1.0 ml/min (see Note 4). Buffers are sparged with helium prior to and during HPLC runs. Fatty acids are detected using UV absorbance and evaporative light scatter (see Note 5). Waters 2487 dual absorbance monitor parameters are: 192 nm and a gain of 0.5. The parameters for the Waters 2420 light scatter detector are: gain = 100; PSI = 5 psi, ultrapure nitrogen; nebulizer tube temperature = 60% or 36°C; drift tube temperature = 45°C. While UV absorbance detects only unsaturated fatty acids, evaporative light scatter detects saturated and unsaturated fatty acids. The inline b-scintillation counter detects 14C-labeled fatty acids (8, 16). In some cases, nonradioactive fatty acids can be used to treat cells (see Note 6). Additional verification of fatty acid elongation products was established by gas chromatography–mass spectrometry at the mass spectrometry facility at Michigan State University

384

Jump

(http://genomics.msu.edu/mass_spec/index.html). The equipment used in these studies was sufficient to establish the chain length and the number of double bonds (16). In addition, the chromatograms of 14C-label fatty acids were compared to the chromatograms of unlabeled fatty acid standards. Fatty acid standards used in our studies include: 14:0; 16:0; 18:0; 16:1,n7; 18:1,n-9; 19:1,n- (recovery standard); 18:2,n-6; 18:3,n-3, 18:3,n-6; 20:3,n-6, 20:3,n-9, 20:4,n-6, 20:5,n-3; 22:5,n-3; 22:6,n-3. Each fatty acid is at 100 mM in methanol. All fatty acids are obtained from Nu-Chek Prep (Elysian, MN). 3. 3. Assessment of Fatty Acid Elongase Function in Vivo 3.3.1. Loss-of- and Gainof-Function Studies..

Mouse models of ablated expression of Elovl-3 and Elovl-6 genes have been reported (6, 17). The phenotype of these animals indicates that Elovl-3 and Elovl-6 play important roles in skin lipid composition and whole body insulin action, respectively. Elovl-2 has been overexpressed in mouse adipocyte cell lines and appears to influence triglyceride metabolism (18). Elovl-5 has been overexpressed in mouse liver (15); elevated Elovl-5 expression affects multiple pathways involved in hepatic lipid and carbohydrate composition (15). Such studies indicated that fatty acid elongases have effects on cellular functions that go beyond effects on cellular lipid composition, per se.

3.3.2. Adenoviral-Mediated Over Expression of Fatty Acid Elongases.

A key requirement for assessing elongase functions in cells and in vivo is to alter the expression of the enzyme in nearly all cells in culture or in a tissue. Our approach to alter elongase expression in cells and liver has used recombinant adenoviruses expressing specific elongases (condensing enzyme). This method allows for robust expression of the elongase in nearly every cell in culture and uniform expression in hepatic parenchyma cells in liver (15). Under these conditions, it is possible to assess the impact of these enzymes on cellular lipid composition and function. Methods to prepare recombinant adenoviruses expressing elongases and their use in cells or in  vivo have been described elsewhere (8, 15). Changes in cell/tissue lipid composition are assessed by the methods described above for monitoring elongase activity and fatty acid composition.

3.3.3. Methods to Generate Recombinant Adenovirus

Replication defective recombinant adenoviruses are a biohazard. As such, the use of these reagents require a biosafety level-2 (BSL2) for containment of both cell culture and animal infection studies. Adenoviruses generated in our laboratory are derivatives of the Ad-Easy recombinant adenovirus (Stratagene). Standard cloning protocols are followed to generate these viruses. Typically, coding regions of specific enzymes are generated by reverse transcriptase-polymerase chain reaction and inserted into the pShuttle-CMV vector (Stratagene). Cloning of cDNA for Elovl-2, Elovl-5, and Elovl-6 was described previously (7).



Mammalian Fatty Acid Elongases

385

The coding region for each transcript was ligated into Ad-Easy XL adenoviral vector system (Stratagene), recombined in BJ5183 cells, and propagated in XL10 Gold ultracompetent cells. AdDNA was packaged into adenoviral particles in Ad-293 cells. The resultant adenovirus was amplified in Ad-293 cells. Viral lysates are cesium purified twice and titered using the Adeno-X Rapid Titer Kit (Clontech). 3.3.4. Infection of Primary Hepatocytes with Recombinant Adenovirus.

Rat primary hepatocytes are infected with recombinant adenovirus using the following protocol. Rat primary hepatocytes are prepared as described (Subheading 3.2.1). Confluent primary hepatocytes are infected with virus at 5–10 plaque-forming units[PFU]/cell). When an adenovirus expressing green fluorescent protein (GFP) is used as a control for infection, greater than 80% of primary hepatocytes express functional protein at the 5–10 PFU/cell level. Under these infection conditions, fatty acid elongase activity increases twofold to sixfold (8, 15). Typically, cells are infected overnight; the next day cells are ready to assess the effects of altered elongase expression on fatty acid elongase activity (Subheading 3.1) or fatty acid elongation in cells (Subheading 3.2).

3.3.5. Infection of Mice with Recombinant Adenovirus.

C57BL/6 mice are infected with recombinant adenovirus using the following protocol. Male C57BL/6 (Charles River Laboratories or Jackson Labs) mice (20–22 g; ~2 months of age) are used for studies with recombinant adenovirus in vivo. Mice are maintained on a Harlan–Teklad (#8640) diet ad lib throughout all studies. Mice are injected with recombinant adenovirus by a retro-orbital route at a dose of 2 × 1010 viral particles (cesium-purified)/mouse while under isoflurane anesthesia. A BD ultrafine II insulin syringe (short needle) is used for the injection. Mice injected with phosphate-buffered saline + 1.0% glucose served as a “Saline” control. Control virus infection uses a recombinant adenovirus expressing firefly luciferase (Luc) (15). In our experience, the Ad-Luc virus serves as an excellent control for infection. In contrast, Ad-GFP has been found to promote hepatotoxicity as reflected by elevated levels of plasma alanine aminotransferase activity. Body weight, food, and water intake is monitored daily. Typically animals displayed no adverse effects from the saline or adenovirus injection and gained weight equally for the duration of the study (5 days). Mice are euthanized on day 5 after infection for the collection of tissues and blood. The advantage of in vivo studies is that it allows for a thorough examination of effects of altered hepatic elongase expression on liver and plasma lipid composition as well as multiple hepatic functions, like gene expression, cell signaling and glycogen, triglyceride, and cholesterol content. Such studies have provided novel insight into the impact altered elongase expression has

14C-Counts/sec

386

Jump cps 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0 0:00

20:4,n-6

22:4,n-6

10:00

20:00

30:00

Retention Time (mins)

40:00

50:00 min

Fig. 2. RP-HPLC analysis of fatty acid elongation products. Rat primary hepatocytes were treated with 100 mM 20:4,n-6 (1.6 Ci/mol) for 24 h. Lipids were extracted and saponified. Nonesterified fatty acids were separated by RP-HPLC. The distribution of 14C-fatty acids was determined by an inline b-scintillation counter (IN/US). Results are expressed as 14 C-counts/sec (Y-axis) vs. retention time in minutes (x-axis). The figure illustrates the separation of arachidonic acid (20:4,n-6) and its elongation product, adrenic acid (22:4,n-6). Nonradioactive fatty acids are not shown in this figure.

on multiple pathways involved in lipid and carbohydrate metabolism (15). A limitation of this in vivo approach, however, is that recombinant adenoviruses infect hepatic cells, including Kupffer and parenchymal cells. Following infection of mice with recombinant adenovirus, we have found ample expression of the elongase in liver, but no change in elongase expression in nonhepatic tissues, like heart, kidney, and spleen (15). A second limitation is that expression of the transgene from the infecting adenovirus persists for only a week or so. Other methods for longer expression are available using other viral vectors, e.g., adeno-associated virus or helper-dependent adenovirus. 3.4. Results

An example of the elongation of a fatty acid in rat primary hepatocytes is illustrated in Fig. 2. In this study, primary hepatocytes were treated with 14C-20:4,n-6 as described in Subheading 3.2. After treating cells with the fatty acid, cells were harvested and lipids extracted. Saponified lipids were fractionated by RPHPLC. The chromatogram in Fig. 2 illustrates the separation of 14C-fatty acids. Arachidonic acid (14C-20:4,n-6) was added to the cells and both 14C-20:4,n-6 and its elongation product, adrenic acid (14C-22;4,n-6) are recovered following lipid extraction and saponification. In this study, ~30% of the total 14C-fatty acid recovered from the cells was adrenic acid (22:4,n-6) indicating that the substrate, (20:4,n-6) was elongated in rat primary hepatocytes.



Mammalian Fatty Acid Elongases

387

In an in vivo study, Elovl-5 was overexpressed in C57BL/6 mouse liver by using the recombinant adenovirus approach described in Subheading 3.3.5. Hepatic Elovl-5 enzyme activity was elevated threefold and a product of Elovl-5 action, i.e., di-homo-g-linolenic acid (20:3,n-6), was elevated greater than two-fold in both liver and plasma (15). In addition, the expression of several genes targeted by the fatty acid regulated transcription factor, PPARa, was suppressed by elevated Elovl-5 expression. The outcome of these studies indicates that changes in hepatic Elovl-5 activity affects both hepatic and plasma lipids as well as hepatic gene expression (15).

Components

Final concentration

Amount/volume added

50 mM CoASH

100 mM

0.5 ml

42 mM ATP

1 mM

2.4 ml

1 M MgCl2

1 mM

0.1 ml

100 mM fatty acid (in Methanol) (Nu-Chek-Prep, Elysian, MN)

20 mM

0.25 ml

100 mM NaOH

20 mM

0.25 ml

4. Notes 1. Non-esterified fatty acids can replace fatty acyl CoAs as substrates. Delete fatty acyl CoA from the reaction mix and add the following components to the reaction mix described in Subheading 3.2.1. 2. When treating cells with fatty acids, it is important to establish the tolerance of cells for specific fatty acids. While hepatocytes and adipocytes can tolerate fatty acid levels as high as 1 mM, many cell types, like endothelial cells, cannot tolerate such high levels. As such, a titration experiment is required to establish conditions to minimize toxicity. It is also important to include bovine serum albumin (BSA) in the culture media. BSA has several binding sites for fatty acids (19), the mole ratio of fatty acid to BSA should not exceed 4 for in vitro studies. It is also important to establish the time course of fatty acid uptake by cells. In rat primary hepatocytes, >80% of the added 14C-fatty acid is assimilated into cellular lipids within 6 h (16).

388

Jump

3. Use glass for all lipid extractions and lipid storage. The exception is when pipetting small volumes; use plastic tips and the appropriate pipetting devices. 4. Use only HPLC-grade reagents for lipid extraction and analyses; including HPLC-grade water (Fisher) for preparing buffers. 5. Separation of fatty acids by RP-HPLC does not require trans methylation. The inclusion of acetic acid at 0.1% in the mobile phase is sufficient to protonate fatty acids, so separation and elution from the YMC J’sphere column is based on chain length and degree of unsaturation. 6. 14C-fatty acids are not always required to monitor cellular fatty acid elongation. Addition on nonradioactive 18:3,n6, 20:5,n-3 or 22:5,n-3 results in the formation of 20:3,n6, 22:5,n-3, and 24:5,n-3 in rat primary hepatocytes. These fatty acids are readily detected using either UV absorbance or evaporative light scatter.

Acknowledgments This research was supported by the National Institutes of Health, DK43220. I would like to thank Dr. Moises Torres-Gonzalez and Sasmita Tripathy for critical reading of this manuscript. References 1. Cook, H.W. 1985. Fatty acid desaturation and chain elongation in eucaryotes. In: Biochemistry of Lipids and Membranes, Eds. Vance, D.E. and Vance, J.E.: Ch. 6, pp. 181–212. 2. Cinti, D.L., Cook, L., Nagi, M.N. and Suneja, S.K. 1992. The fatty acid chain elongation system of mammalian endoplasmic reticulum. Prog Lipid Res 31: 1–51. 3. Jakobsson, A., Westerberg, R. and Jacobsson, A. 2006. Fatty acid elongases in mammals: Their regulation and role in metabolism. Prog Lipid Res 45: 237–249. 4. Leonard, A.E., Pereira, S.L., Sprecher, H. and Huang, Y.S. 2004. Elongation of long-chain fatty acids. Prog Lipid Res 43: 36–54. 5. Nagi, M.N., Cook, L., Gherquier, D. and Cinti, D.L. 1986. Site of inhibition of rat liver microsomal fatty acid chain elongation system by Dec-2-ynoyl coenzyme A. J Biol Chem 261: 13598–13605.

  6. Westerberg, R., Tvrdik, P., Unden, A-B., Mansson, J-E., Norlen, L., Jakobsson, A., Holleran, W.H., Elias, P.M., Asadi, A., Flodby, P., Toftgard, R., Capecchi, M.R. and Jacobsson, A. 2004. Role for ELOVL3 and fatty acid chain length in development of hair and skin function. J Biol Chem 279: 5621–5629.   7. Wang, Y., Botolin, D., Christian, B., Busik, C., Xu, J. and Jump, D.B. 2005. Tissue-specific, nutritional and developmental regulation of rat fatty acid elongases. J Lipid Res 46: 706–715.   8. Wang, Y., Botolin, D., Xu, J., Christian, B., Mitchell, E., Jayaprakasam, B., Nair, M., Peters, J.M., Busik, J., Olson, L.K. and Jump, D.B. 2006. Regulation of hepatic fatty acid elongase and desaturase expression in diabetes and obesity. J Lipid Res 47: 2028–2041.   9. Moon, Y.A. and Horton, J.D. 2003. Identification of two mammalian reductases involved

in the two-carbon fatty acyl elongation cascade. J Biol Chem 278: 7335–7343. 10. Denic, V. and Weissman, J.S. 2007. A molecular caliper mechanism for determining very longchain fatty acid length. Cell 130: 663–677. 11. Beaudoin, F., Gable, K., Sayanova, O., Dunn, T. and Napier, J.A. 2002. A Saccharomyces cerevisiae gene required for heterologous fatty acid elongase activity encodes a microsomal beta-keto-reductase. J Biol Chem 277: 11481–11488. 12. Moon, Y.A., Shah, N.A., Mohapatra, S., Warrington, J.A. and Horton, J.D.. 2001. Identification of a mammalian long chain fatty acyl elongase regulated by sterol regulatory element-binding proteins. J Biol Chem 276: 45358–45366. 13. Brolinson, A., Fourcade, S., Jakobsson, A., Pujol, A. and Jacobsson A. 2008. Steroid hormones control circadian Elovl 3 expression in mouse liver. Endocrinology 149: 3158–3166. 14. Igarashi, M., Ma, K., Chang, L., Bell, J.M. and Rapoport, S.I. 2008. Rat heart cannot synthesize docosahexaenoic acid from circulating a-linolenic acid because it lacks elongase-2. J Lipid Res 49: 1735–1745. 15. Wang, Y., Botolin, D., Christian, B. and Jump, D.B. 2008. Elevated hepatic expression of

Mammalian Fatty Acid Elongases

389

fatty acid elongase-5 activity affects multiple pathways controlling hepatic lipid and carbohydrate composition. J Lipid Res 49: 1538– 1552. 16. Pawar, A., and Jump, D.B. 2003. Unsaturated fatty acid regulation of peroxisome proliferator activated receptor-a activity in primary rat hepatocytes. J Biol Chem 278: 35931–35939. 17. Matsuzaka, T., Shimano, H., Yahagi, N., Kato, T., Atsumi, A., Yamamoto, T., Inoue, N., Ishikawa, M., Okada, S., Ishigaki, N., Iwasaki, H., Iwasaki, Y., Karasawa, T., Kumadaki, S., Matsui, T., Sekiya, M., Ohashi, K., Hasty, A.H., Nakagawa, Y., Takahashi, A., Suzuki, H., Yatoh, S., Sone, H., Toyoshima, H., Osuga, J. and Yamada, N. 2007. Crucial role of a long-chain fatty acid elongase, Elovl6, in obesity-induced insulin resistance. Nat Med 13: 1193–1202. 18. Kobayashi, T., Zadravec, D. and Jacobsson, A. 2007. ELOVL2 overexpression enhances triacylglycerol synthesis in 3T3-L1 and F442A cells. FEBS Lett 581: 3157–3163. 19. Hamilton, J.A., Era, S., Bhamidipati, S.P. and Reed, R.G. 1991. Locations of the three primary binding sites for long-chain fatty acids on bovine serum albumin. Proc Natl Acad Sci U S A 88: 2051–2054.

sdfsdf

Chapter 20 Membrane Lipidomics and the Geometry of Unsaturated Fatty Acids : From Biomimetic Models to Biological Consequences Carla Ferreri and Chryssostomos Chatgilialoglu Summary In the last decades, free radical processes delineated an interdisciplinary field linking chemistry to biology and medicine. Free radical mechanisms became of importance as molecular basis of physiological and pathological conditions. Lipids, in particular, unsaturated fatty acids, are susceptible of free radical attack. The reactivity of the double bond toward free radicals is well known, in particular the reversible addition of radical species to this functionality determines the cis–trans double bond isomerization. Since the prevalent geometry displayed by unsaturated fatty acids in eukaryotes is cis, the occurrence of the cis– trans isomerization by free radicals corresponds to the loss of an important structural information linked to biological activity. The formation of trans isomers can have important meaning and consequences connected to radical stress. Free radical isomerization of membrane fatty acids has been the subject of research coupling the top-down approach by model studies, such as biomimetic chemistry in liposomes, with the bottom-up approach dealing with the examination of cell membrane lipidome in living systems under several physiopathological conditions. Methodologies and molecular libraries have been settled, for both liposome experiments and the examination of the radical stress in biological membranes. This chapter will give an overview of the current procedures used for liposome models and the cis–trans isomerization experiments, in order to build-up a library of trans geometrical fatty acid isomers. Key words: Membrane lipidomics, Geometrical trans fatty acid, Radical stress, Liposome, Biomimetic chemistry, Free radical isomerization

1. Introduction Lipids are molecules with different structures having in common the feature of water insolubility. These molecules have very important biological roles, with structural, functional, and Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, DOI 10.1007/978-1-60761-322-0_20, © Humana Press, a part of Springer Science + Business Media, LLC 2009

391

392

Ferreri and Chatgilialoglu

s­ ignaling activities that have nowadays renewed attention of life science research. The novelty brought about in the last years by the approach of lipidomics consisted of a dynamic vision of lipids existing in living organisms (1, 2). In fact, lipidomics considers not only the structural and functional roles played by lipids, but also their in vivo transformations and modifications by metabolic or degradation pathways, as well as their biological consequences. In the last decades, free radical processes became of importance as molecular basis to physiological and pathological conditions. Lipids, in particular unsaturated fatty acids, are susceptible of free radical attack. In Fig. 1 two examples of mono- and polyunsaturated fatty acids are shown, and the reactive sites are indicated in the general structure: the bis-allylic position present only in the polyunsaturated fatty acids, and the double bonds present in both structures. The bis-allylic position is involved in the well-known process of lipid peroxidation, being the hydrogen atom abstraction recognized as the starting step of the reaction mechanism. Innumerable papers, books, and reviews deal with lipid peroxidation, also for its connection with biological and medical consequences.

Fig. 1. Reactive positions in the general formula of fatty acids and two examples of mono- and polyunsaturated fatty acids.



Membrane Lipidomics and the Geometry of Unsaturated Fatty Acids

393

Fig. 2. Mechanistic scheme of the radical-catalysed cis–trans isomerization of double bonds.

At the same time, the reactivity of the double bond toward free radicals is well known, being several atoms and radical species able to add reversibly to this functionality. The general scheme is shown in Fig. 2, where the radical species X• is adding to the cis or trans double bonds, thus forming a radical adduct which ejects the radical and reconstitutes the double bond (3–5). During this process, the geometry of the double bond is involved. Two configurations are possible, the cis and the trans geometrical isomers, the two isomers having a different thermodynamical stability in favor of the most stable trans isomer. During the process of Fig. 1, the isomeric purity is lost since the unsaturated substrate becomes a mixture reaching about an 80:20 trans:cis ratio. This process, called free radical isomerization, is used in chemistry and organic synthesis (6). When the isomerization of the double bond and formation of trans isomers are considered in a biological context, this result can have a “dramatic” meaning. Starting from the fact that the cis geometry of unsaturated fatty acids in eukaryotes is strictly controlled by desaturase enzymes in a regio- and stereoselective manner (7), the occurrence of the cis–trans isomerization by free radicals corresponds to the loss of an important structural information linked to biological activity. Some years ago, this consideration motivated a research interest in the role of geometry of fatty acids and the effects of free radical species, that was soon directed toward one of the most representative cell compartments for the fatty acid composition, i.e., the cellular membrane. The hydrophobic part of phospholipids is made of fatty acids and the unsaturated moieties having the cis isomeric configuration have the role of keeping membrane parameters in the optimal condition, not only for structural but also for functional reasons. It is worth noting that the discovery of the radical-based isomerization process gave a different meaning to the presence of trans fatty acid isomers in humans, previously thought to derive only from exogenous sources. In fact, in the past years, a series of panels made in several countries determined that the increase

394

Ferreri and Chatgilialoglu

of trans fatty acid isomers in humans derives from the consumption of deodorized and partial hydrogenated fats and oils that can be present in foods (8, 9). These panels arose public concern due to the harmful effects of trans fatty acids on health, involving risk factors of heart attack and coronary artery disease, impairment of fetal and infant growth and inhibition of lipid metabolic pathways (10–13). As a consequence, the Food and Drug Administration in USA decided to introduce a new regulation by 2006 for the mandatory indication in the nutritional facts of the trans fatty acid content when it is >0.5% of the total fats (14). 1 .1. A Premise for the Trans Isomer Detection

Some considerations on the trans isomer recognition are summarized below.

1.1.1. Analytical Methods

The analytical method has to allow trans and cis-fatty acid isomers to be separated and recognized. The total trans content can be identified by IR spectroscopy through the absorbance of the trans isolated double bond at a wave number of 966 cm−1 (15, 16). However, for a lipidomic study the identification of each isomer is crucially important, the most effective methodology to apply being gas chromatography (GC) under appropriate conditions. Examples have been reported in case of omega-6 and omega-3 derivatives (17, 18). Sophisticated equipments such as mass spectrometry even with soft techniques, are not yet so efficient for the specific geometrical isomer distinction. Indeed, up to now small differences are reported, which cannot be useful for complex mixtures, consisting of intensity differences of the mass peaks in the FAB-MS or CI-MS spectra (19). Trans isomers identification and separation still rely on “classical” techniques, such as GC (coupled with flame ionization or mass spectrometry detection) (8), and chromatography by silver nitrate impregnated upon silica gel (SNIS) (20), respectively.

1 .1.2. Trans Lipids from the Medium

The experimental set-up of biological systems can influence the presence of geometrical isomers. For a cell biology experiment with cell cultures, the lipid diet has to be carefully controlled. Often, fetal calf serum (FCS) is used to provide lipids and, since it is reported that biohydrogenation is occurring in the ruminant stomach with formation of trans isomers (8), the serum added to cell cultures has to be analyzed first. In this source, the most present geometrical isomer is trans-vaccenic acid (11transC18:1) (9), that in vivo can be desaturated to rumenic acid (9cis, 11trans-C18:2), a conjugated linoleic acid with some potential health benefits (21). Monounsaturated fatty acid isomers falls in the previously mentioned problem of analytical resolution, because in this fatty acid family it is important to distinguish between positional and geometrical isomers. In fact, the presence or formation of the geometrical isomer 9-trans-C18:1 as the



Membrane Lipidomics and the Geometry of Unsaturated Fatty Acids

395

result of the isomerization of 9cis-C18:1 cannot be confused with the presence of the positional isomer 11trans-C18:1, that can instead occur from dietary sources. The isomerism of monounsaturated fatty acids is important because oleic acid (9cis-C18:1) is the most represented unsaturated component in the membrane phospholipids of eukaryotes, therefore it is very likely to be involved in an isomerization process. The additives used in the cell culture medium are not limited to the lipid source, but can include vitamins and other supplements, such as 2-mercaptoethanol, thioctic acid (known also as lipoic acid), or other thiol compounds. As demonstrated by cultivation in the presence of thiol compounds, the medium composition can contain compounds able to cause lipid isomerization (22). 1 .1.3. The Membrane Compartment

What is the biological compartment involved in the study of lipid isomerization? Fatty acids are present in several biological structures, and lipid isomerization can be studied in different cellular compartments. However, as it will be detailed further, it is important to focus attention on the compartment that can have the best significance for the trans fatty acids as markers of geometrical isomerization. From the chemical work done in models of free radical stress, the importance of the cell membrane as a focal point of observation was obtained. Indeed, in this compartment, fatty acid chains have a supramolecular arrangement and the double bonds along the fatty acid chains are disposed in a way that renders them not as equivalent as they are in solution. Some double bonds are located deeper in the hydrophobic core, others are closer to the polar heads, therefore they are in different locations that influence their accessibility by diffusing reactive species. In order to investigate the disposition of the double bonds in the lipid arrangement and the reactivity of this supramolecular assembly, studies can be performed in liposomes as membrane models.

1 .2. Liposomes as Membrane Models in Biomimetic Chemistry

Liposomes are the universally accepted models for cell membranes as they can closely simulate the arrangement of the fatty acid tails in forming the hydrophobic part of the bilayer. The biotechnological applications of liposomes span from membrane modelling to drug delivery, and several books and reviews have addressed these subjects (23–25). Here we focus on the role of liposomes in biomimetic chemistry that means the reactivity of this heterogeneous system made of aqueous and lipid compartments, in terms of behaviour of compounds with different partition coefficients and the selectivity features associated with the supramolecular organization of the bilayer. This system has been applied to the study of free radical processes (3, 5) and it was of crucial importance for evidencing a tandem damage involving water soluble sulfurcontaining compounds and lipid compartment (26, 27).

396

Ferreri and Chatgilialoglu

Fig. 3. Representative phospholipid fatty acid structures in the bilayer arrangement of a liposome.

Liposomes can be represented as shown in Fig. 3, i.e., a double layer formed by spontaneous organization of the phospholipid components in water, delimiting an aqueous cavity. The fatty acid tails can be saturated or unsaturated, and the disposition of the double bonds in the vesicle depends on the supramolecular arrangement of the bilayer. Multilayer vesicles (MLV), having an onion-like structure obtained from dry lipids added with an aqueous medium and vortexed (24), have been largely used in studies of lipid peroxidation (28, 29). However, this type of vesicle are not the best membrane models, since the observation of the diffusion phenomenon through several layers cannot be directly extrapolated to the passage across a single bilayer, like it occurs in natural membranes. Monolamellar vesicles are the closest model to membranes, and they can be formed by different techniques, such as the extrusion (30) and the injection methodologies (31, 32) described below in Subheading 3. These membrane models are successfully applied to study reactivity in heterogeneous system, taking into account several factors: (a) the partition coefficients determine the compartment where compounds are mainly dissolved. Hydrophilic compounds are not able to diffuse into the lipid bilayer, whereas hydrophobic substrates can do it, and their concentration has to be carefully evaluated, in order to have a satisfactory interaction avoiding selfaggregation. Amphiphilic compounds are dissolved in aqueous and lipid compartments without any selective localization, and their diffusion occurs without restriction. Several molecules have mixed properties and it is useful to establish their molecular interactions before starting the reactivity study; (b) the compartment of free radical initiation, that determines where the first radical species is formed and what is the reactive site in its proximity. When a vesicle system is used to model free radical processes, it is important to establish what the method of radical generation is, because it influences the reactivity. For example, radical



Membrane Lipidomics and the Geometry of Unsaturated Fatty Acids

397

initiation can be obtained by diazocompounds of general formula R–N=N–R, that are homolytically cleaved at the appropriate temperature thus giving R• with the evolution of N2 (33). These azoinitiators can be either hydrophilic, the generation of radicals occurring in the aqueous compartment, or hydrophobic, generating radicals within the lipid compartment. On the other hand, if generation of radicals is obtained by irradiation, the radiation source is important. In fact, continuous radiolysis (using gamma irradiation, such as a 60Co source) is one of the best methods to study reactive oxygen species, ROS, also for biological purposes. In the case of aqueous liposome suspensions, the initiation occurs in the water compartment because the interaction of gamma rays with water generates primary water radicals (HO•, H•) and solvated electrons (eaq−) (34). Obviously, these species are able to react with several compounds in the aqueous phase, for example with RSH generating RS•. A third method for irradiation is photolysis, and experimental conditions are important for the generation of radical species. In the case of photolysis of aliphatic thiol compounds RSH to generate the corresponding RS•, the irradiation around 253 nm falls in the first absorption maximum of these compounds, and the predominant process is the S–H bond dissociation, although the weakest bond is the C–S bond (35); (c) after initiation, radical species can diffuse according to their characteristics, as previously discussed, and find a reaction partner. The diffusion of a radical species depends on its molecular weight, whereas reactivity depends on the rate constants of its reactions with possible partners, and this aspect involves competition among different partners. In particular, in a liposome model simulating an in vivo reactivity, the aspects of diffusion and competition have to be carefully addressed in order to establish the necessary close relationship with the chemical processes that occur in the complex biological scenario. Therefore, a multidisciplinary approach linking organic chemistry, physical chemistry, free radical chemistry, biological chemistry, is required together with the expertise and techniques to work in heterogeneous systems. Among the lipid molecules used for liposome experiments, glycerophospholipids are relevant that account for approximately 60 mol% of total lipids in the organism, and are made of the glycerol backbone having a polar head and two hydrophobic fatty acid residues. Synthetic phospholipids can have both fatty acid chains as monounsaturated residues (oleic acid, 9cis-C18:1), or alternatively, one unsaturated and the other saturated fatty acid chains, the latter not participating to the free radical transformation, but having the role of internal standard for the quantitative analysis of the reaction outcome. Synthetic phospholipids to be used in these studies are 1,2-dioleoylphosphatidylcholine (DOPC),1-palmitoyl-2-oleoylphosphatidylcholine (POPC) and 1-stearoyl -2-arachidonoyl phosphatidylcholine (SAPC). The expected products of isomerization are DEPC and PEPC, ­having

398

Ferreri and Chatgilialoglu

elaidic acid (9trans-C18:1) as the geometrical trans-fatty acid residue, whereas for SAPC, containing arachidonic acid, the number of expected trans isomers is 16, derived from the relationship: number of isomers = 2n, with n = number of double bonds. The radicalbased transformation of these cis phospholipids represents a convenient synthetic approach to prepare a large variety of trans isomers in good yield (36, 37). On the other hand, phosphatidylcholines of natural origins can be used, such as soybean or egg lecithins that contain the fatty acid chains as mixtures of saturated, monounsaturated, and polyunsaturated residues. For example, in egg lecithin the mean fatty acid composition is: palmitic acid (C16:0) 32%, stearic acid (C18:0) 14.1%, oleic acid (9cis-C18:1), vaccenic acid (11cisC18:1) 1.2%, linoleic acid (9cis,12cis-C18:2) 20%, arachidonic acid (5cis,8cis,11cis,14cis-C20:4) 4.8%. Lecithins can simulate much closer the various types of fatty acids present in the natural membranes. The study of the isomerization was carried out using liposome suspensions added with various thiol compounds. In Subheading 3 the experiment of large unilamellar vesicles by extrusion technique (LUVET) made of egg lecithin in the presence of thiols under irradiation is described. The irradiation can be interrupted at defined time intervals, and the work-up of the reaction consists of lipid extraction and derivatization to the corresponding fatty acid methyl esters (FAME), as described in Subheading 3. The FAME mixture can be analyzed by GC or separated by SNIS, as described in Subheading 3. 1 .3. Regioselectivity in the Cis–Trans Isomerization in LUVET

An important feature of the liposome model is the supramolecular arrangement of the fatty acid tails in the hydrophobic compartment (see Fig. 3). This can prime the regioselectivity of the isomerization reaction as appreciated in the egg lecithin liposomes. Figure 4 shows the time course of this transformation in egg lecithin liposomes by photolysis in the presence of thiols, with the cis content normalized to 100% for the three types of unsaturated fatty acid residues involved (arachidonic, linoleic, and oleic acids). The formation of trans lipids efficiently occurs, therefore indicating the ability of thiyl radicals to reach the bilayer and interact with the double bonds. In Fig. 4 it is also shown that the unsaturated fatty acids are not equivalent in reactivity, and arachidonic acid (5cis,8cis,11cis,14cisC20:4, triangle) is transformed more rapidly than linoleic acid (9cis,12cis-C18:2, square), whereas the isomerization of oleic residues (9cis-C18:2, open square) occurs slower. The presence of four double bonds in arachidonic acid and two double bonds in linoleic acid makes them more reactive toward the radical attack, however the result was not only due to the relative double bond concentration. The additional effect was due to the



Membrane Lipidomics and the Geometry of Unsaturated Fatty Acids

399

Fig. 4. Time course of the photolysis of egg lecithin LUVET in the presence of 2-mercaptoethanol (triangle, arachidonic acid; square, linoleic acid; open square, oleic acid). In the inset, the formation of linoleic acid monotrans isomers during the reaction.

relative position of the double bonds in the bilayer arrangement that renders some of them more reachable than others. In fact, linoleic acid has two double bonds in the positions 9 and 12, and the transformation into their corresponding trans configuration was not identical. In the inset of Fig. 4 it is possible to appreciate that the 9trans, 12cis isomer prevailed, thus indicating that the diffusion of the thiyl radical in the hydrophobic core favors in this case the position closer to the carboxylic end, i.e., to the aqueous phase. The examination of arachidonic acid residues was far more indicative of such regioselectivity, and the monotrans isomers in the positions 5 and 8 are the only formed. The arachidonate monotrans isomers have been characterized using gas chromatography/mass spectrometry (GC/MS) and 13C nuclear magnetic resonance (NMR), opening the possibility of the identification of radical-based isomerization in  vivo (37). Further work is in progress to define other PUFA markers of endogenous isomerization, and relate them to physiopathological conditions. 1.4. Tandem Protein– Lipid Damage

As previously explained, thiol compounds have been identified as responsible for the radical-based isomerization of the cis lipid double bonds. Since many proteins can be identified as thiol-rich substrates, they can react under free radical stress (for example, caused by HO• radicals) by donating hydrogen atoms, therefore giving thiyl radicals. Actually, the hydrogen donation is largely intended as a protective mechanism against radical degradation of cellular components. Some proteins, like albumin (38), are thought to have such protective role in living systems. It is worth underlining that, due to the high molecular weight, protein-

400

Ferreri and Chatgilialoglu

Fig. 5. Mechanistic scheme of the hydrogen atom attack to methionine with formation of a-aminobutyric acid and methane thiol.

bound thiyl radicals can have some restriction in the diffusion, therefore the ability to reach the double bond in the membrane bilayer and cause isomerization should be evaluated in each case. On the other hand, the generation of thiyl radicals from the various sulfur-containing amino acid residues can occur also from thioethers and disulfides. These functionalities are present in the amino acid residues methionine and cysteine, respectively, therefore the isomerization process was also studied in liposome suspensions containing such substrates. Free radical reactivity regarding amino acid residues was a research subject in radiation chemistry going back to the early sixties. In particular, H• atoms were studied to selectively attack the thioether function of methionine, as shown in Fig. 5 with the formation of a sulfuranyl-type radical 1 that gives a rapid desulfurization. This reaction was not connected to a biological meaning until a few years ago, when a tandem protein–lipid damage was envisaged: from one side, the desulfurization of methionine corresponds to a chemical mutation of this natural amino acid to another naturally occurring, but not genetically coded, amino acid, a-aminobutyric acid (compound 2 in Fig. 5). On the other side, the desulfurization produces methanethiol (CH3SH, compounds 3 in Fig. 5), a diffusible species that under radical stress condition exists as the corresponding thiyl radical CH3S•. This species is able to reach the lipid bilayer of a membrane model and cause double bond isomerization. Model liposomes added with methionine or related di-peptides, as well as with sequences containing methionine or cysteine, such as RNase A (26, 39), Met-enkephalin (27), and amyloid b-peptide (40) were useful to demonstrate that tandem protein–lipid damage occurs under radical stress. Molecular libraries of trans lipids and desulfurated peptides and proteins have been obtained and their application to the in vivo identification of such transformations is under development.



1 .5. Fatty Acid Isomerization in Cell Membranes

Membrane Lipidomics and the Geometry of Unsaturated Fatty Acids

401

The radical-based lipid isomerization was studied also in biological models, using trans-free systems, i.e., living systems where the trans fatty acid contribution (diet, additives) is carefully checked before the experiments. In Subheading 3, the examination of cultured cells incubated in the presence of mM levels of thiol compounds is described, followed by work-up for lipid isolation and derivatization to FAME, that are analyzed by GC. The first report described THP-1 cell lines analyzed by the membrane fatty acid content under different conditions. A basal content of trans lipids was found in controls (2.7 ± 0.5, indicated as relative percentage on the total fatty acid residues detected by gas chromatography) under normal incubation condition (22). The addition of the amphiphilic thiol 2-mercaptoethanol caused a slight increase of trans isomers (3.8 ± 0.9). Glutathione was not able to induce any increase of trans isomers during normal incubation. When these thiols are used under radical stress produced artificially in the cell cultures by g-irradiation, trans lipid formation reached up to 15.5% of the total fatty acid content (see Fig. 8 in the Results). According to the biomimetic chemistry assayed in liposomes, lipid isomerization can likely occur by diffusible thiyl radical species, either produced under normal metabolism or under radical stress, entering the membrane compartment from the aqueous phase and attacking the double bonds. The FAME analysis carried out under appropriate conditions (see Subheading 3) gave the indication that the most involved residue in this transformation was arachidonic acid, confirming the indications derived from liposome experiments. In fact, the supramolecular membrane organization renders the double bonds of different fatty acids not equivalent, therefore favoring the reactivity of arachidonic acid and in particular the formation of the mono trans isomers of this fatty acid in the positions 5 and 8 (22). The reactivity of arachidonic acid can be successfully applied to distinguish the trans isomer contribution of foods from the trans isomer formation by free radical. Taking into account that omega-6 and omega-3 are essential fatty acids, in the former series, the intake of linoleic acid provides the progenitor for arachidonic acid. The two double bonds of linoleic acid (9,12-C18:2) taken from the diet can be cis or trans, depending whether the ingestion of partially hydrogenated oils occurs. On the other hand, tracing the biosynthetic transformation of linoleic acid to arachidonic acid, other two double bonds are enzymatically formed in the positions 5 and 8 by desaturase enzymes that can be only cis, unless subsequently modified endogenously by free radical attack. The identification of the 5-monotrans and 8-monotrans isomers of arachidonic acid was the start of the library of trans lipid isomers (22, 37, 41), to be used as markers of endogenous isomerization in biological samples (42, 43).

402

Ferreri and Chatgilialoglu

The pathway of free radical isomerization of membrane fatty acids was discovered by an approach combining the top-down investigation of model studies, such as liposomes, with the bottom-up approach of the analyses of cell membrane lipidome in living systems under several conditions. This chapter will describe some of the methodologies used in these approaches that can be useful to perform experiments in the field of chemical biology of free radical processes.

2. Materials 2.1. Equipments

1. Small extruder equipment equipped with two 1-mL Hamilton syringes and polycarbonate filters with the desired pore size (typically 100 nm) 2. Shaker or vortexer (maximum 2,500 rpm) 3. The gamma irradiation is done in the continuous radiolysis apparatus. For the photolysis experiment, the reactor is shown in Fig. 6. 4. Filters (pore size 0.2 mm) 5. Glass test tubes (100 mm × 16 mm) with glass joint and connector to the vacuum line 6. Vials 4 mL with screw-top caps or 4-mL vials equipped with an open-top cap and PTFE-faced silicone septum

Fig. 6. Scheme of a photolysis reactor.



Membrane Lipidomics and the Geometry of Unsaturated Fatty Acids

403

7. 10–50-mL Hamilton syringes 8. Connection to the gas line for degassing technique, by means of a tube and a one-way spring-clip stopcock. The connection with the vials is made through a syringe needle 19-gauge, 6 in. with Luer hub and deflecting tip 9. Rotary evaporator 10. Separating funnel 11. Buchner funnel 12. Gas chromatographer with flame ionization detector (FID) and a column for the fatty acid separation (see Note 1) 13. Benchtop refrigerated centrifuge for small volumes (speed 13,000 g). 2 .2. Reagents and Supplies

1. Synthetic or natural phospholipids (40–60 mg per mL) dissolved in an organic solvent (methanol) or in the form of LUVET stock suspension (53–79 mM stock suspension using 40–60 mg of lecithins and considering an average mw = 760) 2. Acqueous phase. Distilled water, phosphate buffer, or phosphate buffer saline at the desired pH 3. Organic solvents of HPLC purity grade. Chloroform, methanol, n-hexane; for lipid extraction usually chloroform:methanol 2:1 mixture is necessary for phospholipids whereas n-hexane is used for fatty acid methyl esters (FAME). The chromatographic system used for separation of cis and trans isomers of FAME by the SNIS technique is hexane:diethyl ether 7:3 4. Stock solution in water of the thiol RSH or sulfur-containing substrate (see Note 2) 5. Brine (saturated aqueous solution of NaCl) 6. Methanol solution of KOH 0.5 M is the reagent for the transesterification 7. 5% AgNO3 (w/w) solution in acetonitrile is used for the silica gel plates in the SNIS technique. 8. Silica gel plates (see Note 3).

3. Methods 3 .1. Preparation of Large Unilamellar Vesicles by Extrusion Technique

1. Dissolve 40–60 mg of the phospholipids in chloroform (2 mL) in the test tube, and evaporate the solvent under a stream of nitrogen in order to form a thin film of the lipid on the test tube walls.

404

Ferreri and Chatgilialoglu

2. Connect the test tube to the vacuum for 30 min, in order to evacuate all the organic solvent (see Note 4). Then put a stopper and go to the vortex. 3. Add 1 mL of the aqueous phase of the experiment and vortex at 1,800 rpm for several minutes. Make sure that the lipid film is all removed during vortexing, and a uniform milky suspension is formed, without solid particles. 4. The extruder has to be prepared with the desired polycarbonate filter (see Note 5) and washed with fresh aqueous medium before charging the lipid suspension. 5. Charge one Hamilton syringe with the suspension to start the extrusion procedure. Generally, 19 times is the standard number of passages through the extruder to get a uniform size distribution. 6. The suspension is transferred to a vials equipped with a screw cap and can be stored in the refrigerator at 4–6°C for several weeks (the less is the liposome size, the longer is the storage period). For liposomes of egg lecithin or other substrates with high polyunsaturated content, it is advisable to store after filling the vials with an inert atmosphere (nitrogen). 3 .2. Preparation of Unilamellar Vesicles by the Injection Methodology

1. A solution of synthetic or natural phospholipids solution in methanol is prepared (generally 80 mM solution corresponding to 61 mg of phospholipids in 1 mL) and it is the stock solution for the injection methodology. This stock solution is further diluted to obtain a final 1 mM solution. 2. The aqueous medium is passed through a 0.2 mm filter and then transferred to a 4-mL vials, using a magnetic stirrer to keep a continuous mixing (see Note 6). 3. Put the desired quantity of 1 mM lipid solution in a Hamilton syringe (for example, 25 mL of 1 mM lipid solution in a 25-mL Hamilton) and inject it in the aqueous phase (for example, 2.5 mL of phosphate buffer, to get a final 10 mM suspension) keeping the needle just on the surface of the water, not immersed.

3 .3. LUVET Isomerization Under Irradiation Conditions

1. Prepare a LUVET stock suspension by the extrusion methodology (stock suspension 52–72 mM using 40–55 mg of lecithins and considering an average mw = 760) using 100-nm polycarbonate filters. 2. Dilute the stock suspension by adding the desired quantity of the aqueous medium in a 4-mL vial equipped with an opentop cap and PTFE-faced silicone septum. 3. Prepare the stock solution of the desired S-containing compound. If the compound is not soluble in water, the calculation of



Membrane Lipidomics and the Geometry of Unsaturated Fatty Acids

405

the initial concentration should be done in order that the addition will not overcome 1–3% of the total volume of the final suspension. Since in a standard experiment the total volume is 1,000 mL, the stock solution of the S-containing compound cannot be more than 10–30 mL. Calculate the appropriate dilution of the stock suspension and of the stock solution S-containing substrate. 4. The degassing procedure (or the appropriate gas saturation) is carried out by connecting the gas line by means of a tube and a one-way spring-clip stopcock equipped with a syringe needle that is inserted in the septum of the vial. The 4-mL vial is also connected to another small needle that provide the gas exit. In the reaction mixture of 1-mL total volume, the bubbling proceeds for 10 min (see Note 7). For the photolysis, degassing is carried out in the apparatus shown in Fig. 6. For the photolysis the LUVET suspension prepared following points a and b, is transferred to the apparatus, added with the thiol compounds and connected to the gas line through the appropriate inlet. 5. After degassing, irradiation can start (see Note 7). 6. After irradiation, or at defined time intervals, the reaction mixture can be processed in order to examine the lipid isomerization, as described in the next paragraph. 3.4. Work-Up of the Vesicle Experiment, Derivatization of Lipids a Fatty Acid Methyl Esters and Separation by Gas Chromatography

 1. The aqueous suspension (1-mL phospholipids vesicle sample) is added with 2:1 chloroform:methanol (2 mL) and after mixing, the organic layer (bottom of the vial) is removed with the help of a pipette. If necessary another 1 mL of organic phase can be used for a second extraction. The organic phase containing phospholipids is transferred to a separatory funnel, washed with brine (1 mL), collected over anhydrous sodium sulfate, filtered, and evaporated to dryness by rotary evaporator. 2. The phospholipids residue is added with 0.5 M KOH/MeOH (1 mL) in a 4-mL vial equipped with a screw cap and filled with nitrogen. The vials are shacked for 10 min at room temperature and then the fatty acid methyl esters (FAME) are extracted by adding 2 mL of n-hexane. The organic phase is separated, washed with fresh brine, and then evaporated to dryness, to afford the FAME to analyze. 3. The methyl esters are dissolved in the minimum quantity of n-hexane, and injected in the GC equipped with the appropriate column. Oven conditions have to be studied in order to get the best performance as reported in the literature (see Note 1) (15, 37).

406

Ferreri and Chatgilialoglu

3 .5. Separation of the Trans Fatty Acid Methyl Ester Isomers by Thin-Layer Chromatography with Argentation Technique

1. Silica gel plates are immersed in a 5% AgNO3 acetonitrile bath kept in the dark for a variable time (see Note 8). 2. The plate is dried and activated in the oven at 110°C for 2 h. The plates can be used immediately after preparation, or can be stored for 1–2 days in the dark. 3. The lipid mixture is added to the plate like a narrow band by means of a Hamilton syringe or a pipette (approximately 1 mg/cm). The quantity depends on the plate thickness, taking into account that for a 1-mm plate the maximum is 50 mg of the lipid mixture. 4. Several chromatographic systems for fatty acid methyl esters can be used as reported in several references (44, 45, 46), for example 7:3 hexane:diethyl ether is suggested for arachidonic acid isomers (37, 41). A general order of elution is: saturated>trans>cis, however the number of double bonds is also important to define the retention index. For a better separation, it is also advisable to stop a first elution at the half of the plate, dry, and then restart elution. 5. After elution, the separation can be evidenced by spraying the plate with cerium ammonium sulfate/ammonium molybdate reagent or 50% H2SO4. For preparative reason, it is advisable to use a small portion of the plate for the development of a band that is sprayed, and the rest of the plate to scrape off the bands, correspondently. 6. Silica gel is scraped off the plate, at each band level, and is transferred to a Buchner funnel equipped with a paper filter. Silica gel is washed with 2:1 chloroform:methanol or 95% ethanol, the filtrate is collected and dried by rotary evaporator. The residue contains the Ag-complex of the FAME that has to be freed of the silver ion. Therefore the residue is treated with a few milliliters of 0.1% ammonia solution, and the free FAME are extracted by adding n-hexane, after complexation of the Ag ions in the aqueous phase. The hexane phase is then dried affording the pure FAME isomers.

3 .6. Cell Membrane Isolation from Cultures

1. Cells are cultured at 5 × 105 cells/mL. RPMI medium and FCS and fatty acids have to be checked for the absence of trans isomers. 2. The cells are washed from the medium and suspended in PBS (1 mL) (Note 9). 0.75 mL of sample is transferred to a 1.5mL centrifuge microtube, added with pure water (0.75 mL), shaken and centrifuged at 8,500 g for 40 min at 4°C. 3. Lipid extraction is executed on the pellet adding 2:1 chloroform:methanol and following the steps a–c described in Subheading 3. The membrane phospholipid isolation and purity can be checked by TLC using appropriate elution systems (47).



Membrane Lipidomics and the Geometry of Unsaturated Fatty Acids

407

Fig. 7. Micrography of the unilamellar vesicles obtained by the injection methodology.

3.7. Results

Here, below some examples are reported of the results obtained applying the above reported methods. 1. Figure 7 shows a micrography of vesicles made of POPC obtained by injection methodology with a 127 ± 8.9 nm diameter. A crucial role is played by the alcohol concentration in the medium, limited to <1% of the total volume in order to leave the vesicles intact. It is worth noting that the corresponding vesicles prepared with a mixture of 17/83 POPC/PEPC (cis/trans) had the diameter of with a 107 ± 7 nm. This confirms the dramatic effect given by the trans double bond configuration. 2. An example of C18 fatty acid separation from cell membrane of THP-1 cells (leukemia cell line) is given in Fig. 8. It is worth noting at this point that lipidomics have been facilitated in recent times by the development of analytical tools such as electrospray ionization mass spectrometry (ES/MS) and the tandem mass spectrometry (MS/MS) (48). Since research in lipidomics examines the pattern of lipid constituents and individuate alterations or transformation due to a particular condition, there is need of rapid and informative screening to get a large number of information, then elaborated by computer-assisted procedures. Under this type of screening methodologies is the approach known as shotgun lipidomics (49), the name resembling the shotgun cloning used in the genome analysis, which is a fast approach for the analysis of complex mixtures of lipid extracts able to give molecular mass-based fingerprints related to metabolic conditions. This powerful tool is yet in its development phase analyzing thousands of molecular species at once, evidencing structural modifications and allowing for some quantitation,

408

Ferreri and Chatgilialoglu

Fig. 8. Gas chromatographic separation of cis and trans isomers of 18:1 and 18:2 fatty acids. Trace A shows the elution of standard references: (1) 9trans-C18:1; (2) 11trans-C18:1; (3) 9cis-C18:1; (4) 11cis-C18:1; (5) 9trans,12trans-C18:2; (6) 9cis,12trans-C18:2; (7) 9trans,12cis-C18:2; (8) 9cis,12cis-C18:2. Trace B: C18 fatty acid residues of membrane phospholipids in THP-1 cells. Trace C: C18 fatty acid residues of membrane phospholipids in THP-1 cells after gamma irradiation (1,000 Gy) in the presence of 10 mM 2-mercaptoethanol.

for example comparing normal and pathological conditions. However, as previously noted, the cis/trans isomerism of unsaturated lipids especially in biological samples is not yet successfully detected by these methodologies.

4. Notes 1. The column type and conditions are essential for the cis–trans isomer separation. The stationary phase is bis(cyanopropyl)methylpolysiloxane, from 50 to 100% of the composition, the column length goes from 60 to 100 m (with internal diameter and thickness of variable sizes). Helium or hydrogen is used as carrier gas and a constant pressure can be applied. Several conditions are reported in the literature (15) and are constantly updated with new findings. 2. The stock solution can be also prepared in ethanol as the solvent for different types of thiols, taking care that the ethanol quantity added to the suspension does not overcome 1–3% of the total volume, i.e., for 1,000 mL suspension volume, the addition of the stock solution must be not more than 10–30 mL.



Membrane Lipidomics and the Geometry of Unsaturated Fatty Acids

409

3. Silica gel plates can be of 0.25-mm thickness for analytical purposes, or 1–2-mm thickness for preparative use. 4. The evacuation of the solvent must be complete, because solvent traces can compromise the result of the radical reaction. 5. Two polycarbonate filters are used in the extruder. The pore size of 100 nm is necessary if liposomes are used for the photolysis experiments, in order to ensure the transparency of the suspension. 6. If necessary, the vials can be warmed up in a thermostat bath at ca. 30–35°C. This can be necessary when lipids tend to be solid at ambient temperature. 7. If the S-containing compound is volatile, it would be preferable to add it at the end of the degassing procedure. In this case, it would be advisable to prepare the stock solution of the thiol with degassed solvent. 8. The period depends from the thickness of the plates, from 20 min for a 0.25-mm plate to a maximum of 60 min. 9. It is advisable to count the number of cells of each sample to be processed in order to have at least 2 × 106 cells per mL.

Acknowledgments The authors wish to thank all collaborators that helped to develop methodologies in an interdisciplinary context. Support and sponsorship of the COST Action CM0603 “Free Radicals in Chemical Biology” is gratefully acknowledged. References 1. Watson AD. 2006. Lipidomics: a global approach to lipid analysis in biological systems, J Lipid Res 47:2101–2111. 2. Wenk MR. 2005. The emerging field of lipidomics. Nat Rev Drug Discov 4:594–610. 3. Ferreri C, Chatgilialoglu C. 2005. Geometrical trans lipid isomers: a new target for lipidomics. ChemBioChem 6:1722– 1734. 4. Chatgilialoglu C, Ferreri C. 2005. Trans lipids: the free radical path. Chem Rev 31:441–448. 5. Ferreri C, Panagiotaki M, Chatgilialoglu C. 2007. Trans fatty acids in membranes: the free radical path. Mol Biotechnol 37:19–25. 6. Dugave C, Demange L. 2003. Cis–trans isomerization of organic molecules and bio-

molecules: implications and applications. Chem Rev 103:2475–2532. 7. Vance DE, Vance JE, eds. 2002. Biochemistry of lipids, lipoproteins and membranes, 4th ed., Elsevier: Amsterdam. 8. Sébédio JL, Christie WW, eds. 1998. Trans fatty acids in human nutrition. The Oily Press: Dundee. 9. Chatgilialoglu C. 2007. Fats of life… explaining the villainous properties of trans fatty acids. Optimum Nutr 40–43. 10. Larquè E, Zamora S, Gil A. 2001. Dietary trans fatty acids in early life: a review. Early Human Develop. 65:S31–S41. 11. Hu FB, Stampfer MJ, Manson JE, Rimm E, Colditz GA, Rosner BA, Hennekens CH, Willett WC. 1997. Dietary fat intake and the risk

410

Ferreri and Chatgilialoglu

of coronary artery disease in women. N Engl J Med 337:1491–1499. 12. Koletzko B, Decsi T. 1997. Metabolic aspects of trans fatty acids. Clin Nutr 16:229–237. 13. Sébédio JL, Vermunt SHF, Chardigny JM, Beaufrére B, Mensink RP, Armstrong RA, Christie WW, Niemela J, Hènon G, Riemersma RA. 2005. The effect of dietary a-linolenic acid on plasma lipids and platelet fatty acid composition: the transLinE study. Eur J Clin Nutr 54:104–113. 14. Bender Brandt M, LeGault LA. 2003. What’s new on nutrition labelling at the United States Food and Drug Administration? J Food Comp Anal 16:383–393. 15. Dieffenbacher A, Dysseler P. 1997. The determination of trans unsaturated fatty acids in edible oils and fats by capillary gas–liquid chromarography. Pure Appl Chem 69:1829– 1837. 16. Firestone D ed. 1998. Official Methods and Recommended Practices, Recommended Practice Cd 14d-96, 5th ed, American Oil Chemists’ Society: Champaign. 17. Samadi A, Andreu I, Ferreri C, Dellonte S, Chatgilialoglu C. 2004. Thiyl radical–catalysed isomerization of oils: an entry to the trans lipid library. J Am Oil Chem Soc 81:753–758. 18. Mjøs SA. 2005. Properties of trans isomers of eicosapentaenoic acid and docosahexaenoic acid methyl esters on cyanopropyl stationary phases. J Chromat A 1100:185–192. 19. Ji H, Voinov VG, Deinzer ML, Barofsky DF. 2007. Distinguishing between cis/trans isomers of monounsaturated fatty Acids by FAB MS. Anal Chem 79:1519–1522. 20. Williams CM, Mander LN. 2001. Chromatography with silver nitrate. Tetrahedron 57:425–477. 21. Lock AL, Corl BA, Barbano DM, Bauman DE, Ip C. 2004. The anticarcinogenic effect of trans-11 18:1 is dependent on its conversion to cis-9, trans-11 CLA by delta9-desaturase in rats. J Nutr 134:2698–2704. 22. Ferreri C, Kratzsch S, Brede O, Marciniak B, Chatgilialoglu C. 2005. Trans lipid formation induced by thiols in human monocytic leukemia cells. Free Radical Biol Med 38:1180–1187. 23. Cevc G, ed. 1993. Phospholipid Handbook, Marcel Dekker: New York. 24. New RRC. 1990. Liposomes: A Practical Approach. IRL Press: Oxford. 25. Lasic DD. 1993. Liposomes. From Physics to Applications. Elsevier: Amsterdam. 26. Ferreri C, Manco I, Faraone-Mennella MR, Torreggiani A, Tamba M, Manara S, Chatgilialoglu C. 2006. The reaction of hydrogen

atoms with methionine residues: a model of reductive radical stress causing tandem protein– lipid damage. ChemBioChem 7:1738–1744. 27. Mozziconacci O, Bobrowski K, Ferreri C, Chatgilialoglu C. 2007. Reaction of hydrogen atom with Met-enkephalin and related peptides. Chem Eur J 13:2029–2033 28. Niki E, Kawakami A, Yamamoto Y, Kamiya Y. 1985. Synergistic inhibition of oxidation of soybean phosphatidycholine liposomes in aqueous dispersion by vitamin E and vitamin C. Bull Chem Soc Jpn 58:1971–1975. 29. Jacob RF, Mason RP. 2005. Lipid peroxidation induces cholesterol domain formation in model membranes. J Biol Chem 280:39380– 39387. 30. MacDonald RC, MacDonald RI, Menco BP, Takeshita K, Subbarao NK, Hu LR. 1991. Small-volume extrusion apparatus for preparation of large, unilamellar vesicles. Biochim Biophys Acta 1061:297–303. 31. Batzri S, Korn ED. 1973. Single bilayer liposomes prepared without sonication. Biochim Biophys Acta 298:1015–1019. 32. Domazou AS, Luisi PL. 2002. Size distribution of spontaneously formed liposomes by the alcohol injection method. J Liposome Res 12:205–220. 33. Hanlon MC, Seybert DW. 1997. The pH dependence of lipid peroxidation using watersoluble azoinitiators. Free Radical Biol Med 23:712–719. 34. von Sonntag C. 1987. The Chemical Basis of Radiation Biology. Taylor & Francis: New York. 35. Alfassi ZB. 1999. S-Centered Radicals. Wiley: Chichester. 36. Ferreri C, Pierotti S, Barbieri A, Zambonin L, Landi L, Rasi S, Luisi PL, Barigelletti F, Chatgilialoglu C. 2006. Comparison of phosphatidylcholine vesicle properties related to geometrical isomerism. Photochem Photobiol 82:274–280. 37. Ferreri C, Samadi A, Sassatelli F, Landi L, Chatgilialoglu C. 2004. Regioselective cistrans isomerization of arachidonic double bonds by thiyl radicals: the influence of phospholipid supramolecular organization. J Am Chem Soc 126:1063–1072. 38. Ihara H, Hashizuma N, Hemmi H, Yoshida M. 2000. Antioxidant ability of bilirubin, vitamin E, vitamin C and albumin in the peroxyl-radical-induced hemolysis of human erythrocytes. J Anal Biosci 23:425–430. 39. Ferreri C, Chatgilialoglu C, Torreggiani A, Salzano AM, Renzone G, Scaloni A. 2008. The reductive desulfurization of Met and Cys residues in bovine



Membrane Lipidomics and the Geometry of Unsaturated Fatty Acids

RNase A is associated with trans lipids formation in a mimetic model of biological membranes. J Prot Res 7:2007–2015. 40. Kadlcik V, Sicard-Roselli C, Houée-Levin C, Ferreri C, Chatgilialoglu C. 2006. Reductive modification of methionine residue in amyloid b-peptide. Angew Chem Int Ed 45:2595–2598. 41. Ferreri C, Faraone Mennella MR, Formisano C, Landi L, Chatgilialoglu C. 2002. Arachidonate geometrical isomers generated by thiyl radicals: the relationship with trans lipids detected in biological samples. Free Radical Biol Med 33:1516–1526. 42. Ferreri C, Angelini F, Chatgilialoglu C, Dellonte S, Moschese V, Rossi P, Chini L. 2005. Trans fatty acids and atopic eczema/dermatitis syndrome: the relationship with a free radical cis.trans isomerization of membrane lipids. Lipids 40:661–667. 43. Puca AA, Novelli V, Viviani C, Andrew P, Somalvico F, Cirillo NA, Chatgilialoglu C, Ferreri C. 2008. Lipid profile of erythrocyte membranes as possible biomarker of longevity. Rejuven Res 11:63–72.

411

44. Wilson R, Sargent JR. 2001. Chain separation of monounsaturated fatty acid methyl esters by argentation thin-layer chromatography. J Chromat A 905:251–257. 45. Dobson G, Christie WW, Nikolova-Damyanova B. 1995. Silver ion chromatography of lipids and fatty acids. J Chromat B 671:197–222. 46. Morris LJ. 1966. Separation of lipids by silver ion chromatography. J Lipid Res 7:717–732. 47. Rouser G, Fkeischer S, Yamamoto A. 1970. Two dimensional thin layer chromatographic separation of polar lipids and determination phospholipids by phosphorus analysis of spots. Lipids 5: 494–496. 48. Griffiths WJ. 2003. Tandem mass spectrometry in the study of fatty acids, bile acids, and steroids. Mass Spectrom Rev 22:81–152. 49. Han X, Gross RW. 2003. Global analysis of cellular lipidomes directly from crude extracts of biological samples by ESI mass spectrometry. A bridge to lipidomics. J Lipid Res 44:1071–1079.

sdfsdf

Chapter 21 OnLine Ozonolysis Methods for the Determination of Double Bond Position in Unsaturated Lipids Michael C. Thomas, Todd W. Mitchell, and Stephen J. Blanksby Summary Modern lipidomics relies heavily on mass spectrometry for the structural characterization and quantification of lipids of biological origins. Structural information is gained by tandem mass spectrometry (MS/MS) whereby lipid ions are fragmented to elucidate lipid class, fatty acid chain length, and degree of unsaturation. Unfortunately, however, in most cases double bond position cannot be assigned based on MS/MS data alone and thus significant structural diversity is hidden from such analyses. For this reason, we have developed two online methods for determining double bond position within unsaturated lipids; ozone electrospray ionization mass spectrometry (OzESI–MS) and ozone-induced dissociation (OzID). Both techniques utilize ozone to cleave C–C double bonds that result in chemically induced fragment ions that locate the position(s) of unsaturation. Key words: Ozonolysis, Double bond position, Structural characterization, Unsaturated lipids, Mass spectrometry, OzESI–MS, OzID

1. Introduction Modern, high throughput lipid analysis techniques are increasingly reliant on mass spectrometry for the structural identification and quantification of individual lipids within complex mixtures (1). With this approach, structural characterization is based solely on the collision induced dissociation (CID) behavior of the lipid ions. This is particularly effective in phospholipid analysis, where CID spectra are easily interpreted to yield information on lipid class and the total number of carbons and double bonds of any esterified fatty acids (2). As powerful as CID may be, significant limitations remain in assigning fatty acid sn-position – especially in mixtures of regioisomers – and determining the position Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_21, © Humana Press, a part of Springer Science + Business Media, LLC 2009

413

414

Thomas, Mitchell, and Blanksby

and geometry of C–C double bonds. While difficulties remain in characterizing the relative percentages of two regioisomers differing only in fatty acid sn-position, the most abundant isomer can usually be identified from CID spectra (2). Conversely, low energy negative ion CID spectra, as acquired from triple quadrupole, QToF, or ion trap instruments, do not form ions informative of double bond position. Furthermore, the relative ratios of fragment ions in the CID spectra of the [M – H]− phospholipid regioisomers differing only in double bond position are indistinguishable (3). Since no information is gained on double position, most analyses assign the position of unsaturation purely on the basis of fatty acid natural abundances as determined from gas chromatography (GC). For example, the m/z 281 18:1 carboxylate ion commonly observed in CID spectra of phospholipid anions would usually be assumed to be 9Z-18:1 (oleic acid). While oleic acid is a highly abundant fatty acid, other regioisomers such as 6Z-18:1 (petroselinic acid) and 11Z-18:1 (vaccenic acid) are also found in biological systems (4). Traditional lipid analysis has revealed twelve commonly occurring isomeric fatty acids (Table 1) (5). The failure of CID to identify double bond position is thus a significant short-coming since isomeric lipids differing only in double bond position can have distinct biochemical and biophysical properties (6). Several methods are currently available to identify double bond position, either within fatty acid methyl esters or intact lipids. Traditional gas-chromatography based methods are the most commonly applied and can distinguish regioisomeric fatty acids (7). This approach, however, requires that all fatty acid ester bonds are hydrolyzed and converted to their methyl esters. As a consequence, the assignment of each fatty acid identified to its parent lipid(s) within a complex mixture is impossible without

Table 1 A list of common isomeric fatty acids differing only in double bond position Fatty acid (no. carbons: no. double bonds)

Fatty acid class

16:1

n-7, n-9

18:1

n-7, n-9, n-12

18:3

n-3, n-6

20:3

n-3, n-6, n-9

22:5

n-3, n-6

Other naturally occurring isomeric fatty acids are also known.

OnLine Ozonolysis Methods for the Determination of Double Bond Position

415

extensive, multidimensional fractionation. To gain greater structural detail it is therefore advantageous to be able to determine double bond position within intact lipids. Several mass spectrometric methods have been developed that enable this to be done. Conceptually, the simplest of these methods is the CID of the carboxylate anions formed upon fragmentation of the parent phospholipid anion in an MS3 experiment. Comparison of the resultant MS3 spectrum with the MS/MS spectrum of the deprotonated free fatty acid can, in some instances, elucidate the double bond position in the bound fatty acid (8). To promote fragmentation at the site of unsaturation, chemical derivatization of the double bond prior to ESI–MS/MS analysis has also been demonstrated to be useful in identifying double bond position. In two notable examples, (a) Moe et al. pretreated phospholipids and free fatty acids with osmium tetroxide prior to ESI–MS/MS analysis (3, 9, 10). The dihydroxylated lipids formed by the derivatization method were shown to initiate characteristic cleavages upon CID thus identifying the position of the initial double bond (3, 9, 10). (b) Harrison and Murphy have shown that ozonolysis of glycerophospholipids in a thin film can produce near quantitative conversion of olefinic bonds to ozonides. ESI–MS/MS of these ozonides in either positive or negative ion mode leads to dissociation of the ozonide moiety and thus yields fragment ions uniquely identifying the double bond position (11). While chemical derivatization of lipids can be used for locating double bond position, it has the undesirable requirement of additional sample preparation prior to analysis. Ideally, online methods for double bond position elucidation are preferred. Such a method has been developed by Brenna and co-workers and has been applied to fatty acid methyl esters (12, 13) and triacylglycerols (14). In this method chemical ionization of unsaturated lipids in the presence of acetonitrile results in covalent adduct ions that upon collision induced dissociation yield fragment ions indicative of the position of the double bond(s) within the molecule. While this methodology offers much promise, complementary methods that are applicable to more complex lipids, such as phospholipids, are needed. Our efforts towards determining double bond position have centred on the use of ozone chemistry in conjunction with mass spectrometry. We have developed two techniques whereby ozonolysis is either performed in the ion-source or the ion-trapping region of a mass spectrometer. The simplest of these techniques is ozone electrospray ionization–mass spectrometry (OzESI–MS). In this technique, ozonolysis is initiated in the electrospray ionization source of a commercial mass spectrometer and provides two chemically induced fragment ions for each double bond that can readily identify double bond position (15, 16). This technique is quite similar to using ozonolysis in

416

Thomas, Mitchell, and Blanksby

conjunction with field-induced droplet ionization but can be achieved without specialized instrumentation (17). OzESI–MS is easy to use and highly effective for individual lipids or simple lipid mixtures, however, interpretation of OzESI–MS spectra of complex mixtures is exceedingly difficult. To overcome the limitations of OzESI–MS we developed OzID where mass selected lipid ions are isolated and trapped in the presence of ozone vapor within an ion trap mass spectrometer (18). In this chapter, the materials, modifications, and methods to utilize these two techniques are discussed.

2. Materials 2.1. Equipment

1. ThermoFinnigan LTQ® ion-trap mass spectrometer (San Jose, CA, USA) fitted with an Ionmax® source.

2.1.1. OzESI–MS (In Situ Corona Discharge)

2. 1/4-in. teflon tubing (approximately 2-m length).

2.1.2. OzESI–MS (External Ozone Generation)

1. Micromass QuattroMicro (Manchester, UK) triple quadrupole mass spectrometer fitted with a z-spray® ion source. 2. Ozone generator capable of generating ozone at 0.3% (v/v) in oxygen (in our experiments we have used an ozone generator from a NOx analyzer, Model 8840, Monitor Labs, Englewood, CO, USA). Ozone generators use either UV photolysis or corona discharge for ozone production. Corona discharge is preferred since the maximum ozone concentration that can be produced from UV photolysis ozone generators is lower (maximum of 0.2% by weight). 3. 1/4-in. teflon tubing (approximately 2-m length). 4. Tubing is required to connect the ozone generator to the ESI source. Silicone tubing is recommended and rubber tubing should not be used. There are a range of websites that provide useful information on ozone compatibility (19, 20). 5. Soft plastic adaptor to join the teflon tubing and silicone tubing together. 6. Flow-metering valve. 7. Flow meter (the NOx analyzer used in our experiments contained an in-built flow meter). 8. Solvex unlined nitrile, 0.28-mm thick, 33-cm length, slip on gloves, 37–145 (Ansell, Richmond, Vic, Australia).

2.1.3. OzID

1. ThermoFinnigan LTQ ion-trap® mass spectrometer (San Jose, CA, USA) fitted with an Ionmax source®. 2. HC-30 ozone generator (Ozone Solutions, Sioux Center, IA, USA).

OnLine Ozonolysis Methods for the Determination of Double Bond Position

417

3. Pump 11 Pico Plus syringe pump (Harvard Apparatus, Holliston, MA, USA). 4. Model 203 variable leak valve (Granville-Phillips, Boulder, CO, USA). 5. Stainless steel 1-piece 40G series 3-way ball valve (Swagelok, Salon, OH, USA). 6. Stainless steel 1-piece 40G series ball valve (Swagelok, Salon, OH, USA). 7. Stainless steel union tee (Swagelok, Salon, OH, USA). 8. 1/8-in. stainless steel tubing. 9. PEEKsil tubing restrictor (100-mm L × 1/16-in. OD × 0.025-mm ID, SGE Analytical Science, Ringwood, Vic, Australia). 10. Appropriate Swagelok fittings for connections (Swagelok, Salon, OH, USA) 11. 10-mL Norm-ject luer slip syringe (Henke Sass Wolf, Tuttlingen, Germany). 12. Boston PVC 6-mm ID tubing (Fix-a-tap, Asquith, NSW, Australia), soft plasting adaptor and plastic 7-mm OD T connector (U 099223162, Livingstone, Rosebery, NSW, Australia) for ozone collection. 13. Teflon tap. 14. Glass Dreschel bottle (or “bubbler”, 1 L). 15. ADM2000 Intelligent flowmeter (J & W Scientific, Folsom, CA, USA). 16. Solvex unlined nitrile, 0.28-mm thick, 33-cm length, slip on gloves, 37–145 (Ansell, Richmond, Vic, Australia). 2 .2. Reagents and Supplies

1. HPLC grade methanol and AR grade chloroform (APS Chemicals, Sydney, NSW, Australia) for sample preparation for all methods. AR grade sodium acetate (APS Chemicals, Sydney, NSW, Australia) may be used to form sodium adducts for OzESI–MS (external ozone generation) and OzID analysis. 2. Sodium thiosulfate, potassium iodide and Vitex indicator are required to titrate ozone. Sodium thiosulfate and potassium iodide are also required if high concentration ozone is not destroyed using a catalytic ozone destruct unit during production for OzID. 3. Ultra high purity helium (99.999%) (BOC gases, Cringila, NSW, Australia) for operation of the LTQ ion trap mass spectrometer. 4. Nitrogen (BOC gases, Cringila, NSW, Australia) for operation of the LTQ ion trap mass spectrometer when performing OzID experiments.

418

Thomas, Mitchell, and Blanksby

5. Industrial grade compressed oxygen (purity 99.5%) (BOC gases, Cringila, NSW, Australia) for ozone production in all methods. An oxygen regulator must be used when working with oxygen.

3. Methods 3 .1. OzESI–MS (In Situ Corona Discharge)

1. Close nitrogen supply and release the pressurized nitrogen in the nitrogen line using the sheath and auxiliary gas controls within the instrument control software. Remove the tubing from the nitrogen inlet push fitting located on the back panel of the LTQ ion-trap mass spectrometer (see Fig. 1). 2. Attach the oxygen cylinder to the nitrogen inlet via 1/4-in. tubing and set the pressure to 100 ± 20 psi using the oxygen compatible regulator. 3. Prepare lipid sample in methanol with a concentration of 1–10 mM for an isolated lipid. Use higher concentrations (~40 mM) for a lipid mixture. 4. Infuse the lipid sample into the ESI source with a flow rate of 3–10 mL/min. Using higher flow rates may reduce the relative yields of ozonolysis products. Furthermore, using higher flow rates may result in the ignition of methanol in the ion source (see safety precautions, Subheading 3.6). 5. Set the ionization polarity to negative. 6. Tune to maximize the abundance of the lipid ion of interest using a source voltage of −3 to −4 kV. Tuning can be done by using the automatic tune function to optimize the ion optics. The gas flow rates can be optimized using the semiautomatic

Nitrogen line attached to the push fitting

Fig. 1. The nitrogen input push fitting. For in situ corona discharge OzESI–MS the nitrogen line is replaced by connecting an oxygen cylinder via 1/4-in. teflon tubing.

OnLine Ozonolysis Methods for the Determination of Double Bond Position

419

Fig. 2. A photograph of a corona discharge on the ESI capillary caused by increasing the source voltage in negative ion mode with oxygen (instead of nitrogen) as the desolvation gas.

tune function. The sheath and auxiliary gases should not be set greater than 15 (arbitrary units) as this may be detrimental to the concentration of ozone produced. 7. Acquire a conventional ESI–MS spectrum using these optimized conditions. 8. Increase the source voltage until an ozone producing corona discharge is observed (typically −4.5 to −6 kV). A corona discharge is observed as a blue glow at the tip of the ESI capillary (see Fig. 2). If necessary the lights in the laboratory can be turned off to allow easier visualization of the corona discharge. 9. Acquire the resulting OzESI–MS (in situ corona discharge) spectrum. 3 .2. OzESI-Ms (External Ozone Gernation) - Ozone Generator Selection and Measuring Output Ozone Concentration

An ozone generator capable of producing ozone to approximately 0.3% (v/v) in oxygen is required for OzESI–MS. The use of higher ozone concentrations may be a significant hazard and therefore the output ozone concentration should be measured prior to connecting to the mass spectrometer. Ozone concentration can be measured either by titration or using a specialized ozone analyzer. To measure ozone concentration by titration, a setup designed to collect ozone in a plastic syringe needs to be made (refer to Subheading 3.5). For this setup, however, a flow-metering valve is required upstream of the ozone generator

420

Thomas, Mitchell, and Blanksby

and the teflon tap is not needed. After the ozone collection setup is prepared with the ozone generator and oxygen supply attached, perform a titration as follows: 1. Open the oxygen cylinder tap and set the pressure to 5 psi. 2. Using the flow-metering valve set the oxygen flow rate and turn on the ozone generator. 3. Wait for 5 min for ozone production to stabilize. 4. Collect a known volume of the ozone generator output in a gas tight plastic syringe. 5. Bubble collected ozone through acidified aqueous potassium iodide. 6. Back titrate against sodium thiosulfate in the presence of Vitex indicator up to the end-point (color disappears as iodine is reduced to iodide). Specialized ozone analyzers are commercially available and would also be suitable for this application. 3 .3. OzESI-MS (External Ozone Generation) – Modification of Ion Source

For a schematic of the instrument modification for OzESI–MS see Fig. 3. 1. Connect the oxygen cylinder to a metering-flow valve with 1/4-in. teflon tubing. 2. Connect the metering-flow valve to the ozone generator with an appropriate length of tubing. 3. Connect an appropriate length of 1/4-in. teflon tubing to the ozone generator and using a soft plastic fitting connect a small length of silicone tubing. 4. Ensure that the desolvation gas supply is turned off on the computer interface. Detach the desolvation gas line from the press fitting at front of the QuattroMicro triple quadrupole mass spectrometer.

Fig. 3. A schematic representation of the modifications to the Quattro triple quadrupole mass spectrometer for OzESI–MS.

OnLine Ozonolysis Methods for the Determination of Double Bond Position

421

Fig. 4. A photograph of the desolvation gas line connected to silicone tubing. This involved detaching the desolvation gas line from the push fitting.

5. Connect the silicone tubing to the desolvation gas line. Since silicone tubing is soft, this can be done by inserting the desolvation gas line into the silicone tubing (see Fig. 4 for a photograph of the ESI source modified for OzESI–MS). Check that the ion source is sealed and ensure it is vented external to the laboratory. A strip of filter paper soaked with a concentrated aqueous solution of potassium iodide is useful for the detection of leaks. The appearance of a brown coloration indicates ozone is present. 3 .4. Performing OzESI-MS (External Ozone Gernation

1. Open the tap of the oxygen cylinder and set the pressure to approximately 5 psi using the oxygen regulator. 2. Using the flow-metering valve, set the oxygen flow rate to the desired flow rate (160 mL/min was used for the ozone generator from the NOx analyzer). 3. Infuse the sample (10–40 mM in methanol) into the ESI source at a flow rate of 3–10 mL/min. Enter the required voltages for maximum ion abundance. Typical settings for negative ion mode are; capillary voltage of 3.5 kV, cone voltage 50 V with a source temperature of 80°C. For positive ion mode similar settings are used, however, a lower cone voltage is applied, typically either 30 or 40 V. 4. Acquire a conventional ESI–MS spectrum using oxygen as the desolvation gas. Ensure that the desolvation gas is not turned on in the instrument interface to avoid wasting nitrogen.

422

Thomas, Mitchell, and Blanksby

5. Turn on the ozone generator and wait for several minutes for the ozone concentration to stabilize. 6. Acquire the OzESI–MS spectrum. 7. Turn off the ozone generator and purge the ozone generator and ESI source with oxygen to remove the ozone before disassembling the unit. In a typical MS spectrum there are often ions resulting from impurities in the sample, furthermore, lipid mixtures can generate a large number of lipid ions. Both impurities and other lipid ions may interfere with identification of ozonolysis product ions in the OzESI–MS spectrum. This is why both conventional ESI–MS and OzESI–MS spectra are acquired. Comparison of the spectra should allow the assignment of ozonolysis products in the OzESI–MS spectrum. Since the Quattro is a triple quadruple instrument it is also possible to use neutral loss and precursor ion scans when performing ozonolysis. These scans can be extremely useful in simplifying OzESI spectra. In lipid extracts, these scan functions allow the OzESI spectrum to be simplified by only detecting a particular class of lipid. For example, all protonated cholinecontaining phospholipids and their aldehyde and a -methoxyhydroperoxide ozonolysis product ions fragment to form an ion of m/z 184.1 (16). Therefore, when analyzing phosphatidylcholines or sphingomyelins by OzESI, using a m/z 184.1 precursor ion scan to detect their protonated ions is recommended. To do this select the precursor ion scan function and enter the product mass as m/z 184.1. Turn the collision gas on and set the collision gas pressure to approximately 3 × 10−3 Torr and use a collision energy of 35 eV. All other steps remain the same. 3.5. OzID

With the modified LTQ we set the helium pressure to 5 psi using the regulator attached to the helium cylinder. The flow rate of helium into the mass spectrometer is controlled by the variable leak valve. The Granville-Phillip model 203 variable leak valve used in our modified instrument is generally set to around 10 (arbitrary units) to achieve an ion gauge pressure of approximately 0.8 mTorr. For OzID, high concentration ozone needs to be produced (see safety precautions, Subheading 3.6) and collected in the fumehood before being introduced into the mass spectrometer. A setup for ozone production and collection therefore needs to be prepared as shown in Fig. 5 and outlined below: 1. Connect the oxygen cylinder to the ozone generator inlet using 1/4-in. teflon tubing with Swagelok fittings. 2. Connect an appropriate length of teflon tubing to the ozone generator output port.

OnLine Ozonolysis Methods for the Determination of Double Bond Position

423

Fig. 5. A schematic diagram of the setup used to collect high concentration ozone. The Dreschel bottle (“bubbler”) contains a solution of sodium thiosulfate and potassium iodide and is used to destroy ozone before being released into a fumehood. High concentration ozone is collected by attaching a syringe to the T connector.

3. Connect a piece of PVC tubing (6-mm I.D. Boston) to the teflon tubing using a soft plastic adaptor. 4. With plastic tubing connect a teflon tap and plastic T connector in series (see Note 1). 5. On the plastic T-connector connect a small piece of PVC tubing and insert a syringe tip into the PVC tubing to act as a connector for the syringe. 6. Connect the PVC tubing to a Dreschel bottle containing a 600 mL of aqueous solution of sodium thiosulfate (70 g) and potassium iodide (5 g). This acts to destroy ozone as it leaves the system. The solution will turn brown when the thiosulfate is totally consumed. This acts as a warning that ozone is not being destroyed and that a fresh solution needs to be prepared. 7. On the outlet of the Dreschel bottle connect an appropriate adaptor so that a flow meter may be attached. To generate high concentration ozone: 1. Close the teflon tap and set the oxygen pressure to 20 psi using the oxygen regulator. 2. Connect the flow meter to the outlet of the Dreschel bottle and set the oxygen flow rate to 400–500 mL/min using the teflon tap. Once the flow rate is set, remove the flow meter. 3. Turn the ozone generator on and set power output to 68 (arbitrary units). 4. After 20–30 min, attach the flow meter to the Dreschel bottle outlet and reduce the oxygen flow rate to 30–40 mL/min. 5. Wait approximately 5 min before collecting ozone in a 10-mL disposable plastic syringe. Quickly attach a square tip needle to the syringe. Wear gloves when collecting ozone, however, be aware that gloves only provide limited protection from ozone as high concentration ozone can create cracks in the gloves.

424

Thomas, Mitchell, and Blanksby

6. Turn off the ozone generator, remove the flow meter and leave the flow rate set at 30–40 mL/min to purge ozone from the system for 10 min. 7. Close the oxygen cylinder tap and depressurize the system by opening the teflon tap to increase the oxygen flow rate. Ensure that the pressure does not increase high above 50 psi (the recommended maximum pressure of the HC-30 ozone generator). 8. Close the teflon tap so that water vapor cannot diffuse back to the ozone generator. Introducing ozone into the ion trap and setting up for OzID: 1. Connect the syringe containing ozone to the PEEKsil restriction (see Fig. 6 for a photograph of an ozone containing syringe attached to a PEEKsil restriction). 2. Place the syringe in the syringe pump and set it to 20–25 mL/ min. 3. Infuse the lipid sample into the ESI source with a flow rate of 3–10 mL/min (higher flow rates may be used if needed). Typically lipid samples are prepared in either methanol or methanol:chloroform (2:1) with a concentration of 1–10 mM

Fig. 6. A photograph of an ozone containing syringe attached to the PEEKsil restriction. Through the PEEKsil restriction the ozone is introduced into a flow of ultrahigh purity helium. The flow of helium is controlled using a variable valve which on our instrument is inside the steel box on the right hand side of the image.

OnLine Ozonolysis Methods for the Determination of Double Bond Position

425

for an isolated lipid and approximately 40 mM for a lipid mixture. Ensure that the sample volume is sufficient for at least 10–15 min of spectral accumulation time. 4. Set the ion gauge pressure to 0.80–0.85 mTorr by adjusting the helium flow with the variable leak valve if necessary (this may not be an accurate pressure reading since the ion gauge is calibrated for helium). Using the correct pressure is critical for acquiring OzID spectra. At this pressure the resolution will be poor and the m/z of the lipid ions in the MS spectrum will shift to approximately 0.5 Da lower. This is due to the presence of oxygen and ozone molecules in the ion trap. Going to higher pressures will result in potentially large mass shifts and a reduction in resolution. Conversely, going to lower pressures will cause problems with ion isolation. The gas pressure can be optimized empirically by observing the effect of helium flow on peak width and mass accuracy. 5. Using the “Define Scan” control panel, isolate the ion of interest using an isolation width of 3 Th. To isolate the correct ion a value up to 1 Da higher than the actual m/z may be required. It is important to examine the peak shape of the isolated ion, check for shoulders on either side of the peak. Adjust the entered m/z and isolation width as required. If the lipid is part of a complex mixture then an isolation width of 2 Th may be needed so that neighboring lipid ions are not also isolated. Using an isolation width of 1 Th is not possible due to the presence of oxygen and ozone in the ion trap. 6. Using the “Define Scan” control panel, set the activation energy to 0, set the number of microscans to 1 and change the activation time to 10,000 ms. In OzID the activation time is more appropriately referred to as a trapping time since no activation energy is used. 7. Acquire OzID spectra. Typically 50 scans are collected for abundant ions. For ions of low abundance it may be necessary to collect more than 100 scans to obtain a satisfactory signalto-noise ratio. 8. When finished, remove the syringe and dispose of the ozone into a fumehood. It is advisable to pull the plunger on the syringe back to ensure that there is no back pressure in the syringe. This minimizes the chance of ozone being released into the laboratory. Ozone decomposes with time (see Note 2) and therefore ozone should be prepared directly before use. When conducting continuous OzID experiments, high concentration ozone should be prepared approximately every 2 h. Ozone concentration is especially critical when performing OzID on ions of low abundance, prepare fresh ozone if required.

426

Thomas, Mitchell, and Blanksby

Collision-induced fragmentation of the parent ion during isolation can occur due to the presence of heavy molecules in the ion trap. This is observed with the sodium adducts of phosphatidylcholines and sphingomyelins where the neutral loss of trimethylamine (59 Da) is observed. CID fragment ions formed during isolation can be removed by adding a second isolation step. To do this: 1. Enter in the mass of the lipid ion and isolate the lipid ion with an isolation width of 3 Th (or less if needed) and leave the activation time set to 30 ms. 2. Enter the mass of the lipid ion again (as for an MS3 experiment) and isolate the ion using an isolation width of 10 Th. Set the activation time to 10,000 ms and ensure the number of microscans is set to 1. 3. Acquire the OzID spectrum.

It is also possible to perform CID prior to ozonolysis. An example of where this is potentially useful is phospholipids with two unsaturated fatty acids of different chain lengths but the same number of double bonds. In such cases it may be difficult to assign the position of the double bonds to a particular fatty acid (see CID/OzID results, Subheading 3.7.7). It is possible, however, to remove a fatty acid by CID and perform OzID on the resulting fragment ion. This leaves no ambiguity as to the positions of the double bonds on each fatty acid. To perform CID/ OzID on a fragment ion: 1. Enter in the mass of the lipid ion and isolate it with an isolation width of 3 Th (or less if needed) and leave the activation time set to 30 ms. Enter in a collision energy that gives a maximum abundance of the fragment ion of interest. 2. Enter the mass of the fragment ion (as in an MS3 experiment) and isolate the ion using an isolation width of 3–5 Th. Set the activation time to 10,000 ms and ensure the number of microscans is set to 1. 3. Acquire the OzID spectrum. 3 .5.1. Instrument Modification for OzID

Modification of the LTQ ion-trap mass spectrometer involves by-passing the helium splitter to make a direct connection between the helium supply and the ion trap with the helium flow rate controlled using a metering flow valve. A PEEKsil tubing restrictor (100-mm L × 1/16-in. OD × 0.025-mm ID, SGE) is connected to the helium supply line via a shut-off ball valve and T-junction downstream of the metering flow valve (see Fig. 7). Full details of this modification have been provided by Harman and Blanksby (21).

3 .6. Safety Precautions When Using Ozone

1. There is a potential explosion risk when using methanol and oxygen with the presence of a corona discharge as an ignition

OnLine Ozonolysis Methods for the Determination of Double Bond Position

427

Fig. 7. A schematic diagram of the modification to an LTQ ion trap mass spectrometer for OzID.

source. Under normal operation with sample flow rates of 3–10 mL/min the fuel:oxygen ratio is too low for ignition to occur. With increased sample flow rates there may be a danger of ignitions since the fuel:oxygen ratio is increased. Particular care needs to be taken when washing out a sample with a high flow rate of methanol since methanol vapor may build up in the ESI source. 2. Ozone is a powerful oxidizing agent and care must be taken when using ozone, especially at high concentrations. Ozone is capable of reacting explosively with certain compounds. Furthermore, ozonides and hydroperoxides can be produced from the reaction of ozone with unsaturated organic compounds. Ozonides and hydroperoxides formed from ozonolysis can decompose explosively. The inhalation of ozone causes irritation and possible damage to the respiratory tract. Ozone also causes irritation to the eyes. The odor of ozone is distinctive and detectable at low concentrations (0.02–0.05 ppm). The permissible exposure limit for ozone is 0.1 ppm (22). The Immediately Dangerous to Life and Health (IDLH) level is 5 ppm (23). 3. It is recommended that an ozone monitor is acquired if working with ozone to measure the ambient concentration of ozone. There are many suitable ozone monitors on the market including the EZ-1X EcoZone ozone monitor (Eco Sensors, Santa Fe, USA) and Series-200 ozone monitor (Aeroqual Limited, Auckland, New Zealand). A cheap method for leak detection is using a strip of filter paper wet with a solution of potassium iodide. The presence of ozone will result in a brown discolo­ ration of the filter paper. 4. When producing large quantities of ozone it is also suggested that the ozone is destroyed before being released to the atmosphere. There is a wide range of commercially available catalytic

428

Thomas, Mitchell, and Blanksby

ozone destruct unit such as the ODS-1P ozone destruct unit (Ozone Solutions, Sioux Center, Iowa, USA). Ozone can also be destroyed using a less elegant method of bubbling ozone though an aqueous solution of sodium thiosulfate and potassium iodide (dissolve 70 g of sodium thiosulfate and 5 g of potassium iodide in approximately 600 mL of water). Ozone is destroyed by reacting with thiosulfate and when the thiosulfate is totally consumed iodide is oxidized to iodine and the solution turns a dark brown/purple. When this occurs, turn off the ozone generator immediately and purge the system with oxygen before preparing a fresh solution of sodium thiosulfate with sodium iodide. 5. It is advisable to read the MSDS for ozone before working with this gas (24). 3.7. Results

This section describes how to interpret OzESI–MS and OzID spectra using the look-up tables provided and demonstrates the approach using a series of worked examples.

3 .7.1. OzESI–MS of a GPA Standard (In Situ Corona Discharge)

Ozonolysis may be initiated by corona discharge in the ESI source of a mass spectrometer when using oxygen as the drying gas (see Subheading 3.1) (15). Shown in Fig. 8a is the negative ion

Fig. 8. (a) The ESI–MS spectrum of GPA(16:0/9Z-18:1) acquired in negative ion mode with a source voltage of −3.5 kV using oxygen instead of nitrogen as the drying gas. (b) The OzESI–MS spectrum of the sample acquired using a source voltage of −4.7 kV with oxygen as the drying gas. In this spectrum ozonolysis products are observed at m/z 563 and 611 (labeled filled square and filled circle) representing the neutral losses of 110 and 62 Da. The ion at m/z 593 results from the loss of water from the m/z 611 ion. The ion at m/z 718 is a formate adduct formed during ionization and is observed in both spectra.

OnLine Ozonolysis Methods for the Determination of Double Bond Position

429

ESI–MS spectrum of a methanolic solution of a monounsaturated phosphatidic acid, GPA(16:0/9Z-18:1), acquired using oxygen as the drying gas. This spectrum was acquired with a source voltage of −3.5 kV and a corona discharge was not observed at the ESI capillary. In this spectrum the [GPA(16:0/9Z-18:1) – H]− anion is observed at m/z 673. Shown in Fig. 8b is the ESI–MS spectrum when the source voltage was increased to −4.7 kV. Under these conditions a corona discharge was observed at the ESI capillary. In this spectrum, ozonolysis product ions at m/z 563, 593, and 611 are observed. The m/z 563 and 611 ions are consistent with aldehyde and a-methoxyhydroperoxide product ions from the ozonolysis of the n-9 double bond (i.e., the 9th carbon–carbon bond from the methyl end of the fatty acyl chain). The m/z 593 ion is believed to result from the neutral loss of water from the m/z 611 ion. Interestingly, the a-methoxyhydroperoxide ozonolysis product results from a reaction of an ozonolysis intermediate with methanol, which was used as the electrospray solvent in this experiment. The participation of the ESI solvent in the ozonolysis reaction has been demonstrated using D4-methanol and ethanol as the ESI solvent where the m/z 611 ion shifted by 4 and 14 Da, respectively (15). 3 .7.2. OzESI–MS (External Ozone Generation)

In our experiments it was found that the total ion current (TIC) was adversely affected when using the high voltages needed to initiate a corona discharge. Furthermore, an ozone producing corona discharge could only be established in negative ion mode. Conversely, when using an ozone generator TIC did not decrease under OzESI–MS conditions and OzESI–MS could be preformed in both negative and positive ion mode (16). OzESI–MS experiments were performed on a triple quadrupole mass spectrometer. With this configuration it was found that m/z 184 precursor ion scanning was very useful in detecting choline containing, protonated phospholipids and their ozonolysis products and therefore was used routinely for the analysis of phosphatidylcholines (GPCho) and sphingomyelins (SM).

3 .7.3. OzESI–MS of GPCho Isomeric Standards (External Ozone Generation)

The OzESI–MS spectra of the regioisomers GPCho(9Z-18:1/ 9Z-18:1) and GPCho(6Z-18:1/6Z-18:1) acquired using a m/z 184 precursor ion scan are shown in Fig. 9. For the n-9 lipid GPCho(9Z-18:1/9Z-18:1), two ozonolysis product ions were observed at m/z 676 and 724. These ions are 110 and 62 Da lower than the GPCho(9Z-18:1/9Z-18:1) [M + H]+ ion and correspond to the aldehyde and a-methoxyhydroperoxide product ions, respectively. The neutral losses of 110 and 62 Da were also observed in the OzESI–MS spectrum of GPA(16:0/9Z-18:1) (Fig. 8) and are indicative of an n-9 double bond (16). In the OzESI–MS spectrum of GPCho(6Z-18:1/6Z-18:1) ozonolysis products were observed m/z 634 and 682 corresponding to the

430

Thomas, Mitchell, and Blanksby

Fig. 9. The OzESI–MS spectrum of a 1 mM of methanolic solution of (a) GPCho

(9Z-18:1/9Z-18:1) and (b) GPCho(6Z-18:1/6Z-18:1). Both spectra are recorded as m/z 184 precursor ion scans and thus show only the [M + H]+ molecular ion and corresponding chemically induced fragment ions. The symbols filled square and filled circle identify the ozonolysis product ions as aldehydes and a-methoxyhydroperoxides, respectively. The mass difference between the ozonolysis products and the precursor ion is also given.

neutral losses of 152 and 104 Da. These ozonolysis products are shifted down 42 Da from the ozonolysis products of GPCho (9Z-18:1/9Z-18:1). This 42-Da mass shift corresponds to three methylene groups and thus demonstrate that GPCho(6Z-18:1/ 6Z-18:1) is an n-12 lipid. Clearly, using the mass difference between the parent ion and the aldehyde and a-methoxyhydroperoxide product ions can be used to determine the double bond position. Table 2 provides a list of expected neutral losses from the OzESI of a range of monounsaturated fatty acids.

OnLine Ozonolysis Methods for the Determination of Double Bond Position

431

Table 2 A list of expected neutral losses or gains for the aldehyde and a-methoxyhydroperoxide product ions from OzESI–MS of lipids containing monounsaturated fatty acids with double bond (DB) positions of n-1 to n-15 DB-position n−

3 .7.4. OzESI–MS of a Human Lens Extract (External Ozone Generation)

Neutral loss Aldehyde

Methoxyhydroperoxide

1

−2

−50

2

12

−36

3

26

−22

4

40

−8

5

54

6

6

68

20

7

82

34

8

96

48

9

110

62

10

124

76

11

138

90

12

152

104

13

166

118

14

180

132

15

194

146

OzESI–MS may also be applied to simple lipid mixtures to determine double bond position within abundant unsaturated lipids. The positive ion m/z 184 precursor ion scan of a lipid extract of a 50-year old, cataractous human lens is shown in Fig. 10a. This spectrum shows the protonated ions of the choline-containing phospholipids, sphingomyelin, dihydrosphingomyelin, and phosphatidylcholine. The two most abundant lipid ions at m/z 705 and m/z 815 correspond to the saturated and monounsaturated dihydrosphingomyelins, SM(d18:0/16:0) and SM(d18:0/24:1), respectively. Other lipids are of relatively low abundance. The OzESI–MS spectrum of the same sample also acquired as an m/z 184 precursor ion scan is shown in Fig. 10b. In the spectrum, ions of m/z 753 and 781 are observed, which are not present in the standard ESI spectrum (Fig. 10a). Performing CID on the m/z 753 and 781 ions (data not shown) revealed

432

Thomas, Mitchell, and Blanksby

Fig. 10. (a) The positive ion ESI–MS spectrum of a lipid extract obtained from a 50-year-old, cataractous human lens recorded as a m/z 184 precursor ion scan and thus showing only the [M + H]+ molecular ions of sphingomyelin, dihydrosphingomyelin, and phosphatidylcholine lipids present within the extract. (b) The OzESI–MS spectrum of the same extract recorded under the same conditions except for the presence of ozone in the desolvation gas. The symbol filled circle indicates the a-methoxyhydroperoxide ozonolysis product ions arising from the ozonolysis of SM(d18:0/24:1). These ozonolysis products correspond to neutral losses of 62, 34, and 6 Da indicating the possible presence of three isomeric forms of this SM(d18:0/24:1) with double bonds in the n-9, n-7, and n-5 positions, respectively.

the neutral losses of 34 and 49 Da from the precursor ions, which are characteristic of protonated a-methoxyhydroperoxides (16). The a-methoxyhydroperoxide ions of m/z 753 and 781 correspond to neutral losses of 62 and 34 Da. Using the list of expected neutral losses for a-methoxyhydroperoxide ions shown in Table 2 it can be seen that the neutral losses of 62 and 34 Da are indicative of n-9 and n-7 double bond positions, respectively. Since there is only one site of unsaturation, SM(d18:0/24:1) was assigned as two isomers, SM(d18:0/15Z-24:1) and SM(d18:0/17Z-24:1), where the stereochemistry about the double bond is assumed to be Z. Unfortunately, the expected aldehyde product ions from the OzESI–MS of SM(d18:0/15Z-24:1)

OnLine Ozonolysis Methods for the Determination of Double Bond Position

433

and SM(d18:0/17Z-24:1) are predicted at m/z 705 and 733 and are thus isobaric with phospholipids already present in the extract. Interestingly, another ozonolysis product ion at m/z 809 was observed corresponding to the a-methoxyhydroperoxide ion of an n-5 double bond. This suggests that there may be a minor abundance of SM(d18:0/19Z-24:1). The abundance of the m/z 809 ion, however, was too low to allow confirmation of the a-methoxyhydroperoxide moiety by MS/MS and thus the presence of an n-5 isomer remains uncertain.”. 3 .7.5. OzID of GPCho Isomeric Standards

OzESI–MS is a useful technique for double bond identification within isolated standards and simple lipid mixtures as demonstrated with the human lens extract (Fig. 10). There are however, several problems that can occur when performing OzESI–MS on lipid mixtures. These include (a) ozonolysis products having the same m/z as other lipids ions in the OzESI–MS spectrum and (b) different molecular lipids giving ozonolysis products of the same m/z. To overcome these problems we developed OzID whereby ions are mass selected and trapped in the presence of ozone before spectrum acquisition. This allows even lipids present within complex lipid mixtures to be analyzed without the need for chromatographic separation. The OzID spectra of the [M + Na]+ ions of GPCho(9Z-18:1/ 9Z-18:1) and GPCho(6Z-18:1/6Z-18:1) are shown in Fig. 11. Both phospholipid standards were made to a concentration of 1 mM in methanol with 200- mM sodium acetate to aid sodium adduct formation during ESI. Sodium adducts were found to ionize easily and react rapidly with ozone to yield OzID spectra with a good signal-to-noise ratio. To acquire these OzID spectra two isolation steps were performed. The first isolation was made at an isolation width of 3 Th to ensure no neighboring ions were also selected. This isolation step resulted in minor fragmentation of the precursor ion and therefore a second isolation step using an isolation width of 10 Th was used to remove the CID fragments from the OzID spectra. After the second isolation step, ions were trapped for 10 s to react with ozone. For the OzID spectrum of the GPCho(9Z-18:1/9Z-18:1) sodium adduct, two abundant ozonolysis products were observed at m/z 698 and 714. These correspond to the neutral losses of 110 and 94 Da. The neutral loss of 110 Da was previously observed in the OzESI–MS spectrum of protonated GPCho(9Z-18:1/9Z-18:1) sodium adduct, ozonolysis products were observed at m/z 656 and 672 corresponding the neutral losses of 152 and 136 Da. As in OzESI–MS, these ozonolysis product ions are shifted 42 Da and indicate the n-12 double bond position. From this data it is can be seen that aldehyde and Criegee ions are the two main products from OzID. Table 3 shows the expected neutral losses/gains for the OzID of a range of monounsaturated-fatty-acid-containing lipids.

434

Thomas, Mitchell, and Blanksby

Fig. 11. The OzID spectra of the [M + Na]+ ions of (a) GPCho(9Z-18:1/9Z-18:1) and (b) GPCho(6Z-18:1/6Z-18:1). Both phospholipids were made to a concentration of 1 mM in methanol with 200 m M of sodium acetate to aid sodium adduct formation during ESI. The pair of ions resulting from ozonolysis of the double bond are labeled with the neutral loss they represent and the filled square and filled circle symbols indicating aldehyde and Criegee ions, respectively. 3 .7.6. OzID of Cow Kidney Phospholipids: Positive and Negative Ion Examples

The negative ion ESI–MS spectrum of a cow kidney lipid extract is shown in Fig. 12a. This lipid extract is significantly more difficult to analyze than the human cataractous lens extract due to the larger number of lipids and the increased levels of unsaturation. The most abundant ion within this spectrum is the m/z 885 ion which was identified by CID to be GPIns(18:0/20:4) (18). An ion 2 Da greater in mass was observed at m/z 887 and was identified as GPIns(18:0/20:3) by CID. Significantly, three isomers of 20:3 have been well documented (Table 1), these are skip-conjugated n-9, n-6, and n-3. From the CID spectrum alone,

OnLine Ozonolysis Methods for the Determination of Double Bond Position

435

Table 3 A list of expected neutral losses or gains for the aldehyde and Criegee product ions from OzID of lipids containing monounsaturated fatty acids with double bond (DB) positions of n-1 to n-15 DB-position n–

Neutral loss Aldehyde

Criegee

1

−2

−18

2

12

−4

3

26

10

4

40

24

5

54

38

6

68

52

7

82

66

8

96

80

9

110

94

10

124

108

11

138

122

12

152

136

13

166

150

14

180

164

15

194

178

the isomer(s) that were present in cow kidney GPIns(18:0/20:3) could not be ascertained. The OzID spectrum of the m/z 887 ion is shown in Fig. 12b. In this spectrum, three pairs of ozonolysis products were observed from the cleavage of each of the double bonds of the 20:3 fatty acid. (Fig. 12b). It is most important to determine the position of the double bond closest to the methyl end since this determines the fatty acid class. In the OzID spectrum of cow kidney GPIns(18:0/20:3) the terminal neutral losses of 68 and 52 Da were observed indicative of an n-6 fatty acid. The three pairs of ozonolysis product ions are also separated by 40 Da indicative of skip-conjugated double bonds (Fig. 12b). No ozonolysis products of either n-9 or n-3 20:3 were observed. Therefore, using a combination of CID and OzID the structure of this phospholipid can be assigned as GPIns(18:0/8Z,11Z, 14Z-20:3), where the stereochemistry of the double bonds is assumed to be Z. From a knowledge of the OzID behavior of

Fig. 12. (a) The negative ion ESI–MS spectrum of a cow kidney lipid extract (ca. 40 mM in 2:1 methanol–chloroform).

The label is bold is the m/z of the ion selected in the OzID experiment. (b) The OzID spectra of phospholipid anion identified by CID as GPIns(18:0/20:3). (c) The positive ion ESI–MS spectrum of the same cow kidney extract with 200 mM of NaOAc to produce sodium adducts of the GPCho lipids. The label in bold is the m/z of the ion selected in the OzID experiment. (d) The OzID spectrum of the m/z 780 ion from a cow kidney lipid extract. This has been identified by MS/MS to be the sodium adduct of GPCho(16:0/18:2) with a minor abundance of GPCho(16:1/18:1). In both OzID spectra the pairs of ions resulting from ozonolysis of each double bond are labeled with filled square and filled circle indicating aldehyde and Criegee ions, respectively. The neutral losses for both the aldehyde and Criegee ions are indicated in spectrum (b), however only the neural losses of the aldehyde ions are indicated in spectrum d.

OnLine Ozonolysis Methods for the Determination of Double Bond Position

437

lipids containing polyunsaturated fatty acids, a list of expected neutral losses from OzID was produced and is shown in Table 4. The positive ion ESI–MS spectrum of a cow kidney lipid extract is shown in Fig. 12c where 200 mM sodium acetate was added to aid the formation of sodium adducts. The most abundant ion in this ESI–MS spectrum was observed at m/z 780 and was identified by CID as predominately the sodium adduct of GPCho(16:0/18:2) with a small contribution from sodiated GPCho(16:1/18:1). The OzID spectrum of the m/z 780 ion is shown in Fig. 12d. Interestingly, in this spectrum three sets of ions are observed despite the fact that the 18:2 fatty acid has only two double bonds. The first pair of ozonolysis products is observed at 68 and 52 Da below the precursor ion indicative of an n-6 double bond (Table 4). Another pair of ozonolysis products is observed with 40 Da lower indicative of a double bond in the 9th position from the methyl terminus within a skip-conjugated polyunsaturated fatty acid. This information allows GPCho(16:0/18:2) to be assigned as GPCho(16:0/9Z,12Z-18:2) where the double bond geometry is assumed to be Z. Interestingly, a pair of ozonolysis product ions are observed at m/z 698 and 714 corresponding to the neutral losses of 82 and 66 Da. By consulting Table 3 it can be

Table 4 A list of expected neutral losses for the aldehyde and Criegee product ions from OzID of lipids containing n-3, n-6, and n-9 skip-conjugated polyunsaturated fatty acids Neutral loss Class

DB-position

Aldehyde

Criegee

n-3

3 6 9 12 15 18

26 66 106 146 186 226

10 50 90 130 170 210

n-6

6 9 12 15 18

68 108 148 188 228

52 92 132 172 212

n-9

9 12 15

110 150 190

94 134 174

The fatty acid class indicates the position of the first double bond counted from the methyl-end of the acyl chain.

438

Thomas, Mitchell, and Blanksby

seen that these neutral losses are indicative of n-7 monounsaturated fatty acids. These ozonolysis products therefore are likely arising from the isobaric lipid, GPCho(16:1/18:1). 3 .7.7. CID/OzID: Anionic Ether Phospholipid from Human Lens

The more challenging lipids for structural assignment by OzID are lipids containing two or more unsaturated fatty acids. In such cases it is possible that a particular double bond position may not be able to be assigned to an individual fatty acid. Examples of such lipids have previously been found when analyzing human lens lipids by mass spectrometry within our laboratory. The structural characterization of a GPEtn ether lipid by OzID and CID/OzID has previously been published (18). In this section the double bond assignment within another abundant human lens phospho­lipid will be demonstrated. This phospholipid was observed at m/z 772 in the negative ion MS spectrum (Fig. 13a). The CID fragmentation of this lipid is consistent with a phosphatidylserine (GPSer) with an 18:1 fatty acid and a monounsaturated ether chain. The OzID spectrum of the m/z 772 ion is shown in Fig. 13b. In this spectrum the neutral losses of 110, 94, 82, and 66 are observed indicating the presence of both n-9 and n-7 double bonds. From these data alone, however, it is impossible to assign the n-9 double bond position to either the 18:1 fatty acid or the ether chain. To gain this information, the m/z 772 ion was mass selected and subjected to CID and the m/z 403 ­fragment ion then isolated and trapped in a CID/OzID experiment (Fig. 13c). The m/z 403 ion results from the neutral losses of the GPSer headgroup and 18:1 fatty acid and therefore contains only the unsaturated ether chain. In this CID/OzID spectrum of the unsaturated ether containing fragment ion, ozonolysis products indicative of both n-9 and n-7 are observed (Fig. 13c). Interestingly, however, within the CID/OzID spectrum the abundance of the n-7 ozonolysis products is greater than the n-9 ozonolysis products unlike in the standard OzID spectrum (Fig. 13b). This suggests that the n-7 double bond may be predominately located on the ether chain. Therefore, the OzID and CID/OzID data seem to suggest that the m/z 728 GPSer ether maybe a mixture of four isomers; GPSer(11Z-18:1e/9Z-18:1), GPSer(9Z-18:1e/11Z-18:1), GPSer(9Z-18:1e/9Z-18:1), and GPSer(11Z-18:1e/11Z-18:1) where the GPSer(11Z-18:1e/9Z-18:1) isomer is likely to be the most abundant. Double bond geometry was assumed to be Z.

3.8. Conclusion

For both OzESI–MS and OzID, two main product ions are observed from double bond cleavage by ozone. Using the mass difference between the precursor lipid ion and the ozonolysis product ion, double bond position can, in most instances, be unambiguously assigned. Of these two techniques, OzID is the most promising tool for further probing molecular structure by mass spectrometry. The significant advantage being that chromatographic separation of complex lipid mixtures is not required. The technique

OnLine Ozonolysis Methods for the Determination of Double Bond Position

439

Fig. 13. (a) The ESI–MS spectrum of a human lens extract acquired on the LTQ. (b) The OzID spectrum of a GPSer ether anion of m/z 772 acquired using a trapping time of 10 s. (c) The CID/OzID of the m/z 403 fragment ion. In this experiment m/z 772 is mass isolated and subjected CID, m/z 403 is then mass-selected and allowed to react with ozone. The m/z 403 ion results from the neutral losses of the GPSer headgroup and 18:1 fatty acid. The pairs of ions resulting from ozonolysis of double bonds are labeled with filled square and filled circle indicating aldehyde and Criegee ions, respectively. The neutral losses of the ozonolysis products represent are also labeled.

440

Thomas, Mitchell, and Blanksby

does, however, require a major instrument modification. Conversely, OzESI–MS only requires a minimal instrument modification and is suitable for isolated lipids or simple lipid mixtures.

4. Notes 1. When connecting PVC tubing to parts such as the Dreschel bottle of plastic T-connector first heat the tubing with a heat gun. This softens the tubing making it easier to work with. Do not use lubricants as these may explosively react with oxygen and/or ozone. 2. In a study performed in 2002, the half-life of ozone gas in a 1 m3 PVC reaction chamber at 21°C with 73% relative humidity was found to be 1 h (25). In our experience, after 2 h of use the yield of ozonolysis products decreases significantly. The authors also acknowledge Ms Jane Deeley for providing the human lens spectra and Miss Jessica Nealon for the preparation of the cow kidney sample.

Acknowledgments S.J.B. and T.W.M. acknowledge the financial support of the University of Wollongong and the Australian Research Council (Grants DP0452849 and LP0455472). M.C.T. acknowledges the support of an Australian Postgraduate Award. The authors acknowledge Dr. David Harman for providing details of the LTQ modification. The authors also acknowledge MS Jane Deeley for Providing the human lens spectra and Miss Jessica Nealon for the Preparation of the cow kidney sample. References 1. Han X, Gross RW. (2005) Shotgun lipidomics: multidimensional MS analysis of cellular lipidomes. Expert Rev Proteomic. 2(2):253–64. 2. Pulfer M, Murphy RC. (2003) Electrospray mass spectrometry of phospholipids. Mass Spectrom Rev. 22(5):332–64. 3. Moe MK, Anderssen T, Strom MB, Jensen E. (2004) Vicinal hydroxylation of unsaturated fatty acids for structural characterization of intact neutral phospholipids by negative electrospray ionization tandem quadrupole mass spectrometry. Rapid Comm Mass Spectrom. 18(18):2121–30.

4. Cook HW. (1996) Fatty acid desaturation and chain elongation in eukaryotes. In: Vance DE, Vance J, eds. Biochemistry of lipids, lipoproteins and membranes. Amsterdam: Elsevier:129–52. 5. Fatty Acids and Eicosanoids. (Accessed at http:// www.lipidlibrary.co.uk/Lipids/fa_eic.html.) 6. Pariza MW, Park Y, Cook ME. (2001) The biologically active isomers of conjugated linoleic acid. Prog Lipid Res. 40(4):283–98. 7. Ackman RG. (2002) The gas chromatograpy in practical analyses of common and uncommon

OnLine Ozonolysis Methods for the Determination of Double Bond Position fatty acids for the 21st century. Anal Chim Acta. 465:175–92. 8. Bryant DK, Orlando RC, Fenselau C, Sowder RC, Henderson LE. (1991) 4-Sector tandem mass-spectrometric analysis of complex-mixtures of phosphatidylcholines present in a humanImmunodeficiency-virus preparation. Anal Chem. 63(11):1110–4. 9. Moe MK, Anderssen T, Strom MB, Jensen E. (2005) Total structure characterization of unsaturated acidic phospholipids provided by vicinal di-hydroxylation of fatty acid double bonds and negative electrospray ionization mass spectrometry. J Am Soc Mass Spectrom. 16(1):46–59. 10. Moe MK, Strom MB, Jensen E, Claeys M. (2004) Negative electrospray ionization lowenergy tandem mass spectrometry of hydroxylated fatty acids: a mechanistic study. Rapid Commun Mass Spectrom. 18(15):1731–40. 11. Harrison KA, Murphy RC. (1996) Direct mass spectrometric analysis of ozonides: Application to unsaturated glycerophosphocholine lipids. Anal Chem. 68(18):3224–30. 12. Van Pelt CK, Carpenter BK, Brenna JT. (1999) Studies of structure and mechanism in acetonitrile chemical ionization tandem mass spectrometry of polyunsaturated fatty acid methyl esters. J Am Soc Mass Spectrom. 10(12):1253–62. 13. Van Pelt CK, Brenna JT. (1999) Acetonitrile chemical ionization tandem mass spectrometry to locate double bonds in polyunsaturated fatty acid methyl esters. Anal Chem. 71(10):1981–9. 14. Xu Y, Brenna JT. (2007) Atmospheric pressure covalent adduct chemical ionization tandem mass spectrometry for double bond localization in monoene- and diene-containing triacylglycerols. Anal Chem. 79(6):2525–36. 15. Thomas MC, Mitchell TW, Blanksby SJ. (2006) Ozonolysis of phospholipid double bonds during electrospray ionization: a new

441

tool for structure determination. J Am Chem Soc. 128(1):58–9. 16. Thomas MC, Mitchell TW, Harman DG, Deeley JM, Murphy RC, Blanksby SJ. (2007) Elucidation of double bond position in unsaturated lipids by ozone electrospray ionization mass spectrometry. Anal Chem. 79(13):5013–22. 17. Grimm RL, Hodyss R, Beauchamp JL. (2006) Probing interfacial chemistry of single droplets with field-induced droplet ionization mass spectrometry: Physical adsorption of polycyclic aromatic hydrocarbons and ozonolysis of oleic acid and related compounds. Anal Chem. 75:3800–6. 18. Thomas MC, Mitchell TW, Harman DG, Deeley JM, Nealon JR, Blanksby SJ. (2008) Ozone-induced dissociation: elucidation of double bond position within mass-selected lipid ions. Anal Chem. 80(1):303–11. 19. Ozone compatible materials – Ozone solutions website (Accessed at http://www.ozoneapplications.com/info/ozone_compatible_materials.htm) 20. Ozone compatible materials – Ozone services website. (Accessed at http://www.ozoneservices. com/articles/003.htm.) 21. Harman DG, Blanksby SJ. (2007) Investigation of the gas phase reactivity of the 1-adamantyl radical using a distonic radical anion approach. Org Biomol Chem. 5:3495–503. 22. Limits for air contaminants – Occupational Safety and Health Administration. (Accessed at http://www.osha.gov/pls/oshaweb/owadisp. show_document?p_table = STANDARDS&p_ id = 9992.) 23. IDLH documentation – Centres for Disease Control and Prevention. (Accessed at http:// www.cdc.gov/Niosh/idlh/10028156.html.)s 24. MSDS for ozone. (Accessed at http://www. ozoneapplications.com/info/ozone_msds.htm.) 25. Baba S, Satoh S, Yamabe C. (2002) Development of measurement equipment of half life of ozone. Vacuum. 65:489–95.

sdfsdf

Chapter 22 Comprehensive Quantitative Analysis of Bioactive Sphingolipids by High-Performance Liquid Chromatography–Tandem Mass Spectrometry Jacek Bielawski, Jason S. Pierce, Justin Snider, Barbara Rembiesa, Zdzislaw M. Szulc, and Alicja Bielawska Summary There has been a recent explosion in research concerning novel bioactive sphingolipids (SPLs) such as ceramide (Cer), sphingosine (Sph), and sphingosine 1-phosphate (Sph-1P) and this has necessitated the development of accurate and user-friendly methodology for analyzing and quantitating the endogenous levels of these molecules. ESI/MS/MS methodology provides a universal tool used for detecting and monitoring changes in SPL levels and composition from biological materials. Simultaneous ESI/ MS/MS analysis of sphingoid bases (SBs), sphingoid base 1-phosphates (SB-1Ps), ceramides (Cers), ceramide 1-phosphates (Cer-1P), glucosyl/galactosyl-ceramides (Glu-Cers), and sphingomyelins (SMs) is performed on a Thermo Fisher Scientific triple quadrupole mass spectrometer operating in a multiple reaction monitoring (MRM) positive ionization mode. Biological materials (cells, tissues, or physiological fluids) are fortified with internal standards (ISs), extracted into a one-phase neutral organic solvent system, and analyzed by a LC/MS/MS system. Qualitative analysis (identification) of SPLs is performed by a Parent Ion scan of a common fragment ion characteristic for a particular class of SPLs. Quantitative analysis is based on calibration curves generated by spiking an artificial matrix with known amounts of target analyte, synthetic standards, and an equal amount of IS. The calibration curves are constructed by plotting the peak area ratios of analyte to the respective IS against concentration, using a linear regression model. This robust analytical procedure can determine the composition of endogenous sphingolipids (ESPLs) in varied biological materials and achieve a detection limit of subpicomole level. This methodology constitutes a “Lipidomic” approach to study the SPLs metabolism, defining a function of distinct subspecies of individual bioactive SPL classes. Key words: Endogenous sphingolipid, Ceramide, Dihydroceramide, Sphingoid base, Sphingoid base phosphate, Sphingosine, Sphingosine 1-phosphate, Sphingomyelin, HPLC–MS analysis, ESI/MS/MS analysis

Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, DOI 10.1007/978-1-60761-322-0_22, © Humana Press, a part of Springer Science + Business Media, LLC 2009

443

444

Bielawski et al.

1. Introduction 1.1. Overview and Background

Prevalent complex SPLs, phosphosphingolipids (PSPL), and glycosphingolipids (GSPL) are found in all eukaryotes as well as some prokaryotes and viruses. They are the main components of the plasma membrane and related organelles or membranes. SPLs constitute about 30% of the total plasma membrane’s lipids, and they also constitute the second major category of polar lipids. For example, in mammalian brains, they represent approximately 5–10% of the total lipid mass (1–3). Abnormal SPL metabolism could lead to their accumulation and deposition in multiple tissues, especially neural tissues, which results in potentially severe clinical manifestations known as sphingolipidoses (1).

1.2. Sphingolipids Structural Diversity and Nomenclature

SPLs constitute one of the most structurally diversified classes of amphiphilic lipids abundant in all living organisms. Sphingosine (Sph), dihydrosphingosine (dhSph), and/or phytosphingosine (phytoSph) are the core structural moieties with variations in the nature of the head groups attached to the primary hydroxyl group (carbohydrates, phosphorylcholine, phosphate, or phosphoinositol), N-acyl groups, and the sphingoid-base backbone resulting in a great number of chemically distinct SPLs. Thousands of natural complex SPLs have been isolated based on almost 60 distinct species of sphingoid bases (SB). However, most of them are very minor components. Sphingoid bases, the backbone of all SPLs, encompass a wide array of (2S,3R,4E)-2-amino-1,3-dihydroxyalkenes (Sph), (2S,3R)-2amino-1,3-dihydroxyalkanes (dhSph), and (2S,3S,4R)-2-amino1,3,4-trihydroxyalkanes (PhytoSph) with alkyl chain lengths from 14° to 22° carbon atoms. There are also variations in the number and position of the double bonds, hydroxyl groups, and branching methyl groups. Mammalian SPLs, the main focus of this chapter, are predominantly composed of 2-amino-1,3dihydroxyoctadecene (18C-Sph, abbreviated here as Sph) and 2-amino-1,3-dihydroxyoctadecane (18C-dhSph, abbreviated here as dhSph). Yeast and plant shingoid bases are mainly composed of 2-amino-1,3,4-trihydroxyoctadecane (18C-PhytoSph), 18C-dhSph, and their eicosa-homologs (20C-PhytoSph and 20C-dhSph). Additionally, some yeast and plant SPLs contain sphingoid bases with a double bond in position 8 or they can have double bonds in positions 4 and 8. Ceramides are N-acyl-derivatives of sphingoid bases. Combinations of different sphingoid bases with different fatty acids (including their hydroxy-analogs) generate a huge variety of Cer, dhCer, and PhytoCer. These basic SPLs are modified at the



Comprehensive Quantitative Analysis of Bioactive Sphingolipids

445

1-hydroxyl group to (a) phosphates (e.g., Sph-1P and Cer-1P), (b) phosphocholine-analogs (e.g., sphingomyelin, SM, and lysosphingomyelin, lyso-SM), and (c) glucosyl and galactosyl-analogs (e.g., glucosylceramide and galactosylceramide, known as cerebrosides, and their lyso-form psychosine). Members of the latter group also serve as precursors to hundreds of different species of complex GSPL, with lactosylceramide (LacCer) being the simplest containing only two sugar residues. 1.3. Mass Spectrometry Approach to Sphingolipid Analysis

The structural diversity of SPLs, and their implication in diseases, dictates that every step of analysis for these natural products must be carefully evaluated. To understand how SPL biosynthesis and turnover regulates cell behavior under normal and abnormal conditions we must be able to establish metabolomic profiles of SPLs. Mass spectrometry (MS) methodology offers an efficient tool to monitor changes in the composition of all these bioactive species under different environments. A variety of sample preparation, ionization modes, and instrumental designs have been developed to analyze particular SPL classes by MS technology (4). Design for this methodology has been based on the fact that different SPL subclasses dissociate into structurally distinctive patterns corresponding to their sphingoid bases, N-acyl chains, and polar headgroups (5–11). Recent advances in electrospray ionization (ESI) have provided a new approach to successfully examine total SPL components in crude lipid extracts (8–11). ESI methodology allows for the generation of intact molecular ions of molecules directly from solution. These molecular ions are derived by direct infusion or by coupling a high-performance liquid chromatography (HPLC) column directly to the mass spectrometer. Further improvements in instrumentation, such as triple quadrupoles with robust ion sources, fast scanning mass analyzers, and reduced chemical noise (mainly in MS/MS technique) allow for the identification and quantitation of SPLs with great sensitivity (subpicomole detection limits) in a highly reproducible manner. SPL identification is accomplished by tandem mass spectrometry (MS/MS) with precursor ion scans (PIs) to distinguish various molecular species in crude lipid extracts. This is accomplished by taking advantage of the unique molecular decomposition pattern (8, 11) for each SPL class (Fig. 1). SPL quantitation is performed by using positive ionization and multiple reaction monitoring (MRM) in conjunction with HPLC separation (11). Liquid chromatography/ tandem mass spectrometry (LC–MS/MS) is the only technology available that provides structural specificity, quantitative precision, and relatively high throughput for analysis of complex SPLs in small samples.

446

Bielawski et al. TYPICAL MS/MS FRAGMENTATION PATTERNS OF SPH, S1P, CERS & SM 1

a

M/Z=282 1

[M+H]-H2O

2 [M+H]-2H2O HO

3

1

18

HO

Sph

M/Z=264

[M+H]-H2O

2

d

NH3

O N

HO

Sph-1P

1 [M+H]-[P-Choline]

OH O

P O

HN O

O

M/Z=184 1

O

2

P O

M/Z=282 +Acyl 1

1

[M+H]-[2HO+Acyl]

M/Z=264 2

OH

O

5 4

NH2

c

M/Z=282 1

2 [M+H]-[O-P(O)(OH)2]-[H2O]

[M+H]-[O-P(O)(OH)2]

OH 1 2

b

M/Z=264 2

OH O HN O

Cer

SM

The specific fragment ion M/Z=264 for Cers and M/Z=184 for SM are used in the Parent Ion Scan experiments to determine molecular species prior to quantitative analysis with the MRM experiments

Fig. 1. MS/MS fragmentation patterns of the initial positive molecular ions, generated upon ESI of: Sph (a); Sph-1P (b); Cer (c); and SM (d). The specific common fragment ion of m/z = 264.2 (c) for Cers, and m/z = 184.1 (d) for SM are used in the parent ion scan experiments for determination of molecular species composition prior to quantitative analysis with the MRM experiments.

2. Materials 2.1. Equipment

1. Tandem mass spectrometry (MS/MS).

Thermo Finnigan TSQ 7000 (Thermo Fisher Scientific) or equivalent triple quadrupole mass spectrometer equipped with an Electrospray Ion Source (ESI) and Syringe pump.

2. HPLC

Agilent-1100 (Agilent Technologies) or equivalent, quaternary, or binary HPLC pump. Agilent-1100 (Agilent Technologies), or equivalent autosampler. Column: Spectra C8SR (Peeke Scientific, Redwood City, CA) 150 × 3.0 mm; 3-mm particle size.

2.2. Reagents

1. 1 M Ammonium formate

Prepare a solution of 1 M Ammonium formate using a 1.000 ml beaker by diluting 46 g (1 mol, 45 ml) of formic acid into about 500 ml of DI water. Accurately adjust the pH to 5.7 with aqueous ammonium hydroxide and then bring to volume of 1,000 ml.



Comprehensive Quantitative Analysis of Bioactive Sphingolipids

447

2. Positive Ion MS Spray Solution

Prepare a solution of methanol:water:formic acid (70:29.8:0.2) (by vol) containing 2 mM of ammonium formate.

3. Cell or Tissue Extraction Mixture

Prepare a solution of iso-propanol:water:ethyl acetate (30:10:60) (by vol)

4. Media (liquid) Extraction Mixture

Prepare a solution of iso-propanol:ethyl acetate (15:85) (by vol)

5. Yeast Extraction Mixture

Prepare a solution of iso-propanol:water:ethyl-ether:pyridine: 25% aqueous ammonium hydroxide (50:15:10:2:0.4) (by vol)

6. HPLC Mobile Phase “A”

Prepare 2 mM of ammonium formate in water containing 0.2% formic acid

7. HPLC Mobile Phase “B”

Prepare 1 mM of ammonium formate in methanol containing 0.2% formic acid

2.3. Analytical Standards 2.3.1. Sphingolipid Internal Standards

Internal standards are prepared at a concentration of 1,000 pmol/ ml each in methanol for particular SPL analyses. The sources of available standards are from MUSC’s Lipidomics Core or from a commercially available source such as Avanti Polar Lipids Inc. (Alabaster, AL) and/or Matreya LLC (Pleasant Gap, PA). 1. The table below indicates specific molecular species and their associated internal standards. Concentrations for these species are calculated by plotting peak area ratios using linear regression models. Sph, dhSph, Sph-1P, dhSph-1P, Cer, a-HO-Cer, and dhCer molecular species: 17CSph, 17CSph-1P, 13C16-Cer, 17C16-Cer, 17C24:1-Cer, 18C17-Cer, and 18C17-dhCer 2. Phyto-Sph, Phyto-Sph-1P, Phyto-Cer, and a-HO-PhytoCer molecular species: 17CSph, 17CSph-1P, 18C6-Phyto-Cer, and 18C17-Phyto-Cer 3. GluSph and GluCer molecular species: 18C8-GluCer and 18C12GluCer 4. LactSph and LactCer molecular species:18C8-LactCer and 18C12-LactCer 5. SM molecular species: 18C8-SM, 18C12-SM, and 18C17-SM

2.3.2. Endogenous Sphingolipid Calibration Standards

Calibration mixtures are prepared at a concentration of 250 pmol/ml (Calibration Mix “A”) and 1,000 pmol/ml (Calibration Mix “B”) for each target analyte from the above analytical groups in methanol. For SM molecular species mixtures are prepared at 1,000 pmol/ml (Calibration Mix “A”) and 4,000 pmole/ml (Calibration Mix “B”).

448

Bielawski et al.

2.4. Supplies

Ethyl acetate (EtOAc), methanol (MeOH), iso-propanol (i-PrOH), ethyl ether (Et2O), pyridine (Pyr), formic acid (HCOOH), ammonium hydroxide (NH4OH) (25% aqueous solution) commercial, (e. g. Honeywell Burdick & Jackson). Analytical grade reagents do not require further purification.

3. Methods 3.1. Sample Preparation

The extraction process is one of the most important steps in treatment of solid (cell pellets, tissue) and liquid (plasma, serum, whole blood, biological fluids) samples. Chloroform–methanol (2:1 by vol) extraction, known as the “Bligh & Dyer” (B&D) method, developed in 1956 for fish tissue (12), and further improved in 1959 by Bligh and Dyer (13), has become the golden standard procedure. This method is still commonly used for lipid extraction for all biological matrices. This virtually unchanged procedure is commonly applied to most SPL sample preparation; regardless of the analytical procedure subsequently used e.g., TLC, HPLC, or MS. However, the extraction efficiency is limited particularly for the most polar SPL components such as Sph-1P or lyso-SM. We have developed (14) a single-phase extraction, using ethyl acetate:iso-propanol:water system at 60:30:10; (by vol) for solid (tissue and cell pellets) and 85:15:0 (by vol) for aqueous samples. This protocol describes lipid extraction under a safe neutral condition to avoid destruction of the parent SPLs containing O-acyl groups (Fig. 2). This is done while assuring efficient and quantitative extraction of the SB-1P from biological material since they are notoriously difficult to recover quantitatively (15, 16).

3.1.1. Extraction of Sphingolipids from Cells or Tissues

1. Fortify cell pellets or tissue homogenate with 50 ml each of appropriate internal standard solution(s). 2. Extract sample with 2.0 ml of i–PrOH–Water–EtOAc (30:10:60, Subheading 2.2), sonicate for 30 s, vortex, and centrifuge for 5 min at 3.000 × g 3. Transfer organic upper phase to a new vial. 4. Re-extract sample with an additional 2.0 ml of i–PrOH–Water–EtOAc (30:10:60, by vol), sonicate for 30 s, vortex and centrifuge for 5 min at 3,000 × g 5. Combine supernatants from the extracts. 6. Evaporate organic extract under Nitrogen to dryness. 7. Reconstitute the dry residue in 150 ml of the Mobile Phase “B” (Subheading 2.2).



Comprehensive Quantitative Analysis of Bioactive Sphingolipids

Cell# 1-10X10*6 Biol. fluids ~1 mg protein Tissue homogenate:equivalent of ~1mg protein

449

Protein Determination (mg)

- add IS - add 2 ml of iso-PrOH-Water-EtOAc (30:10:60, v/v) - vortex, sonicate for 30 sec, centrifuge for 5 min at 3000g - remove supernatant - reextract pellet following the above procedure - collect supernatants

Lipid Extract (~4 ml)

1ml aliquots

dry down

Pi Determination (nmole)

- dry down - reconstitute in 150µl of the Mobile Phase "B" - centrifuge for 5 min at 3000 rpm - transfer supernatant to the HPLC autosampler with 200µl insert - inject 10µl on the HPLC system SM Further extraction

LC-MS / MS analysis

Sph1Ps, Sphs, Ceramide and GSL Components (pmole)

FINAL RESULTS Lipid/ Pi (pmole/nmole) Lipid/Protein (pmole/mg)

Fig. 2. A scheme of sample preparation for the LC/MS/MS analyses, employed in our laboratory.

8. Centrifuge for 5 min at 3,000 × g. 9. Transfer supernatant to an auto-sampler HPLC vial with 200 ml of insert. 10. Inject 10 ml on the HPLC system. 3.1.2. Extraction of Sphingolipids from Culture Media (See Note 1)

1. Fortify 2 ml of media with 50 ml each of appropriate internal standard solution(s). 2. Extract sample with 2.0 ml of 15% iso-PrOH in EtOAc (15:85, Subheading 2.2), vortex and centrifuge for 5 min at 3,000 × g. 3. Transfer organic upper phase to a new vial. 4. Acidify the aqueous phase with 100 ml of formic acid. 5. Repeat extraction with 2.0 ml of 15% i-PrOH in EtOAc (15:85 by vol), vortex and centrifuge for 5 min at 3,000 × g. 6. Combine supernatants from the extracts. 7. Evaporate organic extract under Nitrogen to dryness. 8. Reconstitute the dry residue in 150 ml of the Mobile Phase “B” (Subheading 2.2). 9. Centrifuge for 5 min at 3,000 × g.

450

Bielawski et al.

10. Transfer supernatant to an auto-sampler HPLC vial with 200 ml of insert. 11. Inject 10 ml on the HPLC system. 3.2. Analysis of Intact Sphingolipids by Electrospray Ionization/ Mass Spectrometry (ESI/MS)

Mass spectrometry is a powerful detection technique that enables separation and characterization of compounds according to their mass-to-charge ratio (m/z). Its essential components include a sample inlet, ion source, mass analyzer, detector, and data handling system. The combination of sensitivity, selectivity, speed, and ability to provide valuable structural information makes MS an ideal method for the analysis of intact lipid molecular species. Sample introduction can be either by direct infusion or through preceding separation devices such as liquid chromatography (LC). The soft ionization MS techniques such as fast atom bombardment (FAB) (17, 18), field desorption (FD) (19), thermospray (TS) (20), matrix-assisted laser desorption ionization (MALDI) (21–24), and particularly ESI (7, 11, 14, 25–27) can ionize lipid molecules without causing extensive fragmentation. Currently ESI is the preferred and widely used method to ionize biologically important lipid molecules.

3.2.1. MS Scan Modes

A number of mass analyzers are available, e.g., quadrupole, ion trap, time of flight, ion cyclotron resonance, or sector instruments, which separate charged molecules in vacuum depending on their (m/z) ratio. In the so-called full scan (FS) mode, a spectrum of primary, mostly molecular ions are identified. This is the least specific mode with low sensitivity. However, interferences from other compounds present can either suppress ionization or cause such a high chemical noise making detection of SPLs virtually impossible. Moreover, the mass analyzers can also be used for fragmentation. This is done predominantly in triple quadrupole instruments using a collision cell for controlled fragmentation, called collision-induced dissociation (CID). The degree of fragmentation is affected by the choice of collision gas, its pressure, and primarily the applied collision energy. Initial “soft” ionization of a biological extract results in numerous SPL molecular ions either positive (M + H)+ or negative (M−H)− When the precursor ion fragments, it generates a secondary (daughter) distinctive pattern of ions related to the headgroups, sphingoid bases, and fatty acids. This provides a wealth of structural information, enabling identification of SPLs in particular biological material. In addition to structural information, tandem MS provides a higher sensitivity and specificity with greatly reduced chemical background noise due to very selective modes of monitored masses.

3.2.2. Precursor Ion Scan (PI)

In a PI scan mode Q1 scans across (m/z) ranges transmitting only those primary ions which decompose to the specified product ion of interest. This highly specific scan mode eliminates or at least greatly



Comprehensive Quantitative Analysis of Bioactive Sphingolipids

451

reduces chemical noise. It constitutes a very useful identification tool since each class of SPLs yields at least one common product ion. Thus setting Q3 to this specific daughter ion and scanning Q1 over expected parent ion mass ranges a spectrum of molecular species for an unknown biological sample can be identified. In a MRM experiment, the Q1 is set to pass specific m/z precursor ions and Q3 passes only specific m/z daughter ions. This makes the MRM the most selective and sensitive MS–MS experiment, thus allowing for analysis of even very minor components of a complex mixture. Such measurement practically eliminates chemical noise, thus making it an ideal tool for quantitative analyses, particularly if coupled with HPLC physical separation. Multiple mass transitions, specific for particular compounds, can be monitored sequentially. This allows for large numbers of compounds to be analyzed together. Optimization of CID parameters for each compound of interest results in the best sensitivity and specificity. Figure 3 illustrates the power of MS, presenting examples of chromatograms, performed from the same cell extract recorded

3.2.3. Multiple Reaction Monitoring

a RT: 0.00 - 31.02

RT: 5.97

100

Full Scan Mass Range: Sph RT 3.38 min. Sph1P RT 6.68 min. C16-Cer RT 17.84 min. dhC16-Cer RT 18.41 min. C24:1-Cer RT 24.12 min.

95 90 85 80

NL: 8.34E7 TIC MS SphCer_FU LL

75 Relative Abundance

70 RT: 3.75 RT: 4.96

65 60

RT: 16.79

RT: 1.99

55 50

RT: 11.88

45 40

↓C16-Cer / dhC16-Cer

35

RT: 15.46

30

RT: 9.47

25 20

12.44 13.08

RT: 17.97 RT: 21.53 18.27 RT: 22.39

RT: 24.06

15

↓ C24-Cer RT: 26.11

10 5 0

↑Sph 0

5

27.35

↑Sph1P 10

15 Time (min)

20

25

30

Fig. 3. An example of sphingosine, sphingosone-1-phosphate, and ceramide molecular species LC/MS chromatogram recorded for the same cellular extract applying: full scan (FS) mode (panel a), selective ion monitoring (SIM) (panel b), and multiple reaction monitoring (MRM) technique (panel c). Please note that the most abundant peaks in the FS experiment do not belong to any of the sphingolipids. As a result, peaks can be easily misidentified as false detections. Only the MRM experiment (panel c) is composed exclusively of the target sphingolipid, due to highest possible specificity.

b RT: 0.00 - 31.01 100

RT: 3.78 Selective Ion Monitoring (SIM): Sph RT 3.38 min. Sph1P RT 6.68 min. C16-Cer RT 17.84 min. dhC16-Cer RT 18.41 min. C24:1-Cer RT 24.12 min.

95 90 85 80

NL: 3.62E6 TIC MS ICIS SphCer_SIM

75 RT: 5.04

70 Relative Abundance

65 60 55 50

RT: 22.65

45

RT: 3.40

40

RT: 9.83

35 30 25

RT: 16.22

20

RT: 24.06

RT: 18.41 RT: 21.33

RT: 12.48

15 RT: 24.81

10

RT: 9.09

5

RT: 28.68

0 0

5

10

15 Time (min)

20

25

30

c RT: 0.00 - 31.02 RT: 3.38 100

Multiple Reaction Monitoring (MRM): Sph RT 3.38 min. Sph1P RT 6.68 min. C16-Cer RT 17.84 min. dhC16-Cer RT 18.41 min. C24:1-Cer RT 24.12 min.

95 90 85 80

NL: 1.03E6 TIC MS ICIS SphCer_MR M

75 70 Relative Abundance

65 RT: 16.20

60 55

RT: 16.13

50 45

RT: 18.41 RT: 18.49

RT: 3.78

40 35 30 RT: 3.92

25

RT: 13.92

20

RT: 18.59

15 RT: 4.04

RT: 9.67 RT: 4.18 RT: 8.85

10 5 0 0

Fig. 3. (continued)

5

10

RT: 22.67

15 Time (min)

20

25

30



Comprehensive Quantitative Analysis of Bioactive Sphingolipids

453

with different MS scan modes. These modes are from the least sensitive full scan, through selected ion monitoring (SIM) to the most sensitive MRM mode. 3.3. Sphingolipid Identification

The complexity of SPLs that usually constitutes minor components of crude lipid extracts. The identification of individual molecular species is necessary before attempting any quantitative determination. This task can only be achieved with application of MS, particularly with precursor ion scan (PI) experiments. Qualitative analysis of SPLs from crude extracts is best accomplished by analysis of their unique molecular decomposition products. This is accomplished by using a PI scan of common fragment ions characteristic of a particular class of SPLs (Fig. 1). For mammalian SPLs, the focus of this chapter, m/z 264 and 266 are the common fragment ions used for identification of Sph and its saturated counterpart dhSph derivatives, respectively. Composition of Cer, Cer-1P, and GlcCer (C18-SB) are established by parent ion scans. These scans are performed for the common product ion m/z 264.2 and 266.1 for Sph and dhSph derivatives, respectively, at high collision energy of 35–55 eV. This is done by operating in positive ionization mode (Fig. 1). A representative sample extract is infused directly into the ESI source, and then scanned for molecular ions of those potential SPLs. Further confirmation of identity is achieved through MRM analysis with “soft” fragmentation (15–30 eV). Running samples through the HPLC system also confirms a reasonable expected retention time. Only SPLs that satisfy identification criteria in both analyses should be considered truly present in the sample. SM and dhSM composition (18C-SB) are also established in a similar manner. Their identification is performed by employing the common product ion m/z 183.9 at collision energy of 40 eV as shown in Fig. 1 (see Note 2). The use of HPLC coupled with ESI/MS is one of most powerful technologies for analysis of intact polar lipid molecules. The HPLC’s power of physical separation, to either various lipid classes or individual molecular species within the class, coupled with the MS’s highly selective detection makes possible simultaneous determination of protonated/deprotonated molecules. It also provides invaluable structural information (28, 29).

3.4. Quantitation

Quantitation of SPLs in biological extracts has been the least developed segment of the HPLC/MS/MS analysis due to structural diversity, wide range of lipid concentrations, and a very limited supply of commercially available individual standards. Only laboratories that have access to custom made synthetic standards (14) have been able to set up reliable comprehensive quantitative protocols for various SPL classes. However, Avanti Polar Lipids Inc. (Alabaster, AL) and Matreya LLC (Pleasant Gap, PA) significantly

454

Bielawski et al.

expanded their offering of synthetic standards, so those major difficulties have been partially overcome. To achieve reliable quantitation of all molecular species, calibration curves should be generated for as many representative components of SPLs as possible due to diversified MS responses. Figure 4 represents the diversified MS responses as reflected by calibration curve slopes. Selection of a representative set of internal standards, which serve as a reference for both identification and quantitation, is critical for the analysis of a complex mixture of SPLs. Internal standards should be as similar as possible to the target analytes. They should also provide similar MS fragmentation patterns as well as physico-chemical properties reflected by similar solubility, extraction efficiency, and

3.4.1. Selection of Internal Standards

Analyte / IS Peak Area Ratio

a

10

C14-

9

C16-

8

C16-

7

C18:1-

6

C18-

5

C20-

4

C24:1-

3

C24-

2 1 0

0

10

5

15

20

25

30

35

40

pmoles

b

11

Sp

Analyte / IS Peak Area Ratio

10

dhSp

9

Sph-

8

dhSph-

7 6 5 4 3 2 1 0 0

20

40

60

80

100

120

140

160

180

200

pmoles

Fig. 4. The MS response varies for molecular species even within the particular SPL class, as indicated by the calibration curve slops; therefore, individual calibrations should be generated for as many target analytes as possible. A linear instrument response (R 2 value of 0.99) is obtained for the typical calibration ranges: 1.0–200.0 pmol for SBs and SB-1Ps (panel b), as well as for all ceramide molecular species (panel a).



Comprehensive Quantitative Analysis of Bioactive Sphingolipids

455

mobile–stationary phase relationship during the HPLC separation. The best internal standards are chemically identical to the target analytes labeled with a stable isotope. These isotopes are usually 2H or 13C. However, considering the large number of SPL molecular species such an approach is impractical. Therefore, some compromise must be applied to overcome the limited number of the 2H or 13 C compounds. Often one internal standard per SPL class is used. Mostly it is a SPL with an unnatural, usually lower, number of carbon atoms in the FA moiety (7, 11, 30–32). We (14) have selected derivatives of 17C-Sph and 17C-dhSph (sphingoid bases with unnatural 17 carbon atom backbones) for use as internal standards. These derivatives are used for the quantitation of the naturally derived SPLs 18C-Sph. Additionally these internal standards have those physico-chemical properties such as the elution order and mass fragmentation pattern that accurately reflect natural SPLs, but are not present in the mammalian samples. Moreover, since the internal standards are introduced into the sample prior to extraction incomplete extraction efficiencies are corrected. This correction renders quantitation of the target SPLs more precise. 3.4.2. Calibration

Generating a calibration mechanism for each target SPL in a class greatly improves the quality of the quantitative results. However, due to the limited availability of authentic standards, as mentioned, this is not always possible. Therefore, calibration curves should be generated for as many representative standards as is practical. This way calibrations derived from the synthetic standards can be used for several structurally closely related analytes. We have adopted the following quantitative approach to SPL analysis (14). In this approach, quantitative analyses of SPLs are based on the eight-point calibration curves generated for each target analyte. The synthetic standards along with a set of internal standards are spiked into an artificial matrix. Then they are subjected to an identical extraction procedure as the “biological” samples. These extracted standards are then analyzed by the HPLC/ MS/MS system operating in positive MRM mode employing a gradient elution. Peaks for the target analytes and internal standards are recorded and processed using the instrument’s software system. Plotting the analyte/internal standard peak area ratios against analyte concentrations generates the analyte specific calibration curves. Any SPL for which no standards are available are quantitated using the calibration curve of its closest counterpart. The detailed information on HPLC and MS parameters are presented in the experimental section of this chapter.

3.5. Data Handling

Results from the MS analysis represent the mass levels of particular SPLs in picomoles per total sample used for lipid extract preparation and quantitative analysis. Usually treatment with exogenous

456

Bielawski et al.

agents will cause changes in SPL levels and compositions. For the final data presentation, MS results should be normalized to some stable parameters that are not considered to be affected by that particular treatment. Such preferred stable parameters used for normalization are total protein (mg), or phospholipid phosphate (nmol) present in the Bligh & Dyer extract (13). These parameters correspond to the amount of biological material used for MS analysis (33–35). The use of total cell number, wet or dry weight for tissue, or volume for liquid samples can also be used for normalization (36). Final results may also be expressed as changes in the relation to the control (% control). Each of the end users should develop their own data normalization strategy that will be consequently carried out over the entire course of the experimental program to preserve data integrity. 3.6. Results

Instrument: Agilant 1100 (HPLC)/TSQ 7000(MS/MS) system.

3.6.1. HPLC Conditions

Column: Spectra C8SR (Peeke Scientific) 150 × 3.0 mm; 3-mm particle size. Solvent A: 2.0 mM NH4OCOH aqueous (0.2% HCOOH) (Subheading 2.2) Solvent B: 1.0 mM NH4OCOH in MeOH (0.2% HCOOH) (Subheading 2.2) Flow rate: 0.5 ml/min Variables: Gradient elution, time (min) – % A/B, depending on particular analysis: 1. Sph; Sph-1P and Cer molecular species: 0–3 min 18/82 3–4 min 18/82 – 10/90 4–18 min 10/90 – 1/99 18–25 min 1/99 25–27 min 1/99 – 18/82 27–30 min 18/82

2. dhSph; dhSph-1P and dhCer molecular species: 0–5 min 15/85 5–10 min 15/85 – 10/90 10–15 min 10/90 – 5/95 15–20 min 5/95 – 1/99 20–26 min 1/99 26–30 min 1/99 – 15/95 30–32 min 15/85



Comprehensive Quantitative Analysis of Bioactive Sphingolipids

3. a-HO-Cer molecular species: 0–6 min 10/90 6–7 min 10/90 – 5/95 7–11 min 5/95 11–12 min 5/95 – 1/99 12–25 min 1/99 25–27 min 1/99 – 10/90 27–31 min 10/90

4. Cer-1P molecular species: 0–8 min 15/85 8–15 min 15/85 – 5/95 15–25 min 5/95 25–28 min 5/95 – 15/85 28–30 min 15/85

5. Phyto-Sph; Phyto-Sph-1P and Phyto-Cer molecular species: 0–2 min 20/80 2–3 min 20/80 – 10/90 3–17 min 10/90 – 1/99 17–24 min 1/99 24–28 min 1/99 28–30 min 1/99 – 20/80 30–32 min 20/80

6. a-HO-PhytoCer molecular species: 0–4 min 10/82 4–19 min 11/90 – 05/98 18–25 min 1/99 25–27 min 1/99 – 10/99 27–31 min 10/90

457

458

Bielawski et al.

7. GluSph and GluCer molecular species (see Note 3): 0–3 min 12/88 3–12 min 12/88 – 1/99 12–25 min 1/99 25–28 min 1/99 – 12/88 28–30 min 12/88

8. LactSph and LactCer molecular species: 0–4 min 15/85 4–15 min 15/85 – 15/95 15–25 min 5/95 25–27 min 5/95 – 15/85 27–30 min 15/85

9. Sphingomyelin (SM): 0–12 min 10/90 – 1/99 12–25 min 1/99 25–28 min 1/99 – 10/90 28–30 min 10/90 3.6.2. Sphingolipid Quantitation by HPLC–MS/ MS Using Multiple Reaction Monitoring

Typically an (2.5–400 pmol) eight-point calibration curve is prepared by spiking BSA (for cells) or serum free media (for media) with relevant volumes of Calibration Mix “A” or “B” and 50 ml of the internal standard mix. The curves are then extracted by the above procedures. Calibration standards and analytical samples are analyzed by the HPLC/MS/MS system operating in positive MRM mode employing the HPLC conditions described above. Analyte specific parent–daughter mass transitions, collision energy and the retention times are summarized below (see Note 4). 1. Sph, Sph-1P, and Cer Compound

Mass Trans

Coll E (eV)

RT (min)

17C-Sph-1P*

366.1–250.1

25

5.38

Sph-1P

380.4–264.2

25

6.85

17C-Sph*

286.1–268.0

17

2.80

Sph

300.4–282.2

18

3.28



Comprehensive Quantitative Analysis of Bioactive Sphingolipids

459

Compound

Mass Trans

Coll E (eV)

RT (min)

13C/C16-Cer*

468.6–450.5

15

14.02

C14-Cer

510.7–492.6

15

16.39

C16-Cer

538.8–520.9

15

17.88

C17-Cer*

552.7–534.6

15

18.68

C18:1-Cer

564.5–546.4

15

18.37

C18-Cer

566.4–548.3

15

19.62

C20:1-Cer

592.6–574.5

15

19.94

C20:4-Cer

566.6–568.5

15

20.85

C20-Cer

594.4–576.5

15

21.15

C22:1-Cer

620.7–602.6

17

21.67

C22-Cer

622.7–604.7

17

22.42

17C24:1-Cer*

634.6–616.5

17

22.09

C24:1-Cer

648.8–630.7

18

22.73

C24-Cer

650.8–632.7

18

24.22

C26:1-Cer

676.8–658.7

18

24.85

C26-Cer

678.8–660.7

18

25.96

2. dhSph, dhSph-1P, and dhCer Compound

Mass Trans

Coll E (eV)

RT (min)

17C-Sph-1P*

366.1–250.1

25

  5.38

dhSph-1P

382.3–284.3

18

  7.54

17C-dhSph*

288.2–270.0

20

  3.11

Sph

302.3–284.2

21

  3.68

13C/C16-Cer*

468.6–450.5

15

14.02

C14-dhCer

512.5–494.2

27

16.79

17C16-Cer*

524.6–506.5

15

17.06

C16-dhCer

540.6–522.5

22

18.38

C17-dhCer*

554.6–536.5

27

19.11

C18:1-dhCer

566.5–548.2

27

19.52

C18-dhCer

568.6–550.3

27

19.96

C20:1-dhCer

594.6–576.5

27

21.00 (continued)

460

Bielawski et al.

(continued) Compound

Mass Trans

Coll E (eV)

RT (min)

C20-dhCer

596.6–578.5

27

21.51

C22:1-dhCer

622.6–604.5

27

22.44

C22-dhCer

624.7–606.6

27

22.88

17C24:1-Cer*

634.6–616.5

17

22.09

C24:1-dhCer

650.8–632.6

27

23.47

C24-dhCer

652.7–634.6

27

24.56

C26:1-dhCer

678.8–660.5

27

26.73

C26-dhCer

680.6–662.5

27

27.80

3. a-HO-Cer Compound

Mass Trans

Coll E (eV)

RT (min)

13C/C16-Cer*

468.6–450.5

15

14.02

a-HO-C14-Cer

526.6–508.5

15

16.39

17C16-Cer*

524.6–506.5

15

17.06

a-HO-C16-Cer

554.5–536.5

15

17.88

C17-Cer*

552.7–534.6

15

18.68

a-HO-C18:1-Cer

580.6–562.5

15

18.37

a-HO-C18-Cer

582.6–564.5

15

19.62

a-HOC20:1Cer

608.6–590.5

15

19.90

a-HO-C20-Cer

610.6–592.5

15

21.15

a-HOC22:1-Cer

636.5–618.6

17

21.67

a-HOC22- Cer

638.5–620.5

12

22.42

17C24:1-Cer*

634.6–616.5

17

22.09

a-HOC24:1-Cer

664.8–646.6

15

22.73

a-HOC24-Cer

666.8–648.6

15

24.22

a-HOC26:1-Cer

692.6–876.5

17

24.85

a-HOC26-Cer

694.6–676.5

17

25.96

4. Cer-1P Compound

Mass Trans

Coll E (eV)

RT (min)

C8-Cer-1P*

506.6–264.1

25

11.46

C14-Cer-1P

590.6–264.2

30

16.12

C16-Cer-1P

618.7–264.2

30

17.82



Comprehensive Quantitative Analysis of Bioactive Sphingolipids

461

Compound

Mass Trans

Coll E (eV)

RT (min)

C18:1-Cer-1P

644.7–264.2

30

18.26

C18-Cer-1P

646.6–264.2

30

18.64

C20:1-Cer-1P

672.5– 264.2

30

19.02

C20-Cer-1P

674.5– 264.2

30

19.58

C22:1-Cer-1P

700.6– 264.2

32

20.06

C22-Cer-1P

702.6– 264.2

32

20.38

C24:1-Cer-1P

728.5– 264.2

32

20.65

C24:Cer-1P

730.6– 264.2

32

21.05

C26:1-Cer-1P

756.5– 264.2

32

21.24

C26-Cer-1P

758.5–264.2

32

21.78

5. Phyto-Sph, Phyto-Sph-1P, and Phyto-Cer Compound

Mass Trans

Coll E (eV)

RT (min)

17C-Sph-1P*

366.1–250.1

25

  5.24

Phyto-Sph-1P

398.4–282.1

15

  6.26

17C-Sph*

286.1–268.0

17

  2.74

Phyto-Sph

318.5–282.1

15

  3.11

C6Phyto-Cer*

416.2–282.1

12

  8.80

C14Phyto-Cer

528.6–282.1

24

14.02

C16Phyto-Cer

556.6–282.1

24

16.01

C17Phyto-Cer*

570.5–282.1

25

16.68

C18:1Phyto-Cer

582.6–282.1

25

16.27

C18Phyto-Cer

584.7–282.1

25

17.23

C20:1PhytoCer

610.6–282.1

25

17.79

C20Phyto-Cer

612.7–282.1

25

18.85

C22:1Phyto-Cer

638.6–282.1

25

22.35

C22Phyto-Cer

640.6–281.2

25

20.08

C24:1Phyto-Cer

666.8–282.1

28

20.79

C24Phyto-Cer

668.8–282.1

28

21.87

C26:1PhytoCer

694.6–282.1

28

23.94

C26Phyto-Cer

696.7–282.1

28

24.64

C28:1Phyto-Cer

722.6–282.1

28

25.30

C28Phyto-Cer

724.7–282.1

28

26.65 (continued)

462

Bielawski et al.

6. a-HO-Phyto-Cer Compound

Mass Trans

Coll E (eV)

RT (min)

C6-PhytoCer*

416.2–282.1

20

  9.23

a-HO-C14Phyto-Cer

544.6–282.2

20

13.57

a-HO-C16Phyto-Cer

572.6–282.0

20

15.81

C17Phyto-Cer*

570.5–282.0

20

16.18

a-HO-C18:1Phyto-Cer

598.4–282.0

20

16.76

a-HO-C18Phyto-Cer

600.4–282.0

20

17.01

a-HO-C20:1Phyto-Cer

626.5–282.0

26

18.43

a-HO-C20Phyto-Cer

628.5–282.0

26

19.01

a-HO-C22:1Phyto-Cer

654.4–282.0

26

20.31

a-HO-C22Phyto-Cer

656.4–282.0

26

20.76

a-HO-C24:1Phyto-Cer

682.5–282.0

26

21.06

a-HO-C24Phyto-Cer

684.5–282.0

26

21.97

a-HO-C26:1Phyto-Cer

710.5–282.0

29

22.07

a-HO-C26Phyto-Cer

712.5–282.0

29

23.48

a-HO-C28:1Phyto-Cer

738.4–282.0

29

23.85

a-HO-C28Phyto-Cer

740.5–282.0

29

24.12

7. Gal/Glu-Cer Compound

Mass Trans

Coll E (eV)

RT (min)

Glu-Sph

462.1–282.2

24

  1.85

C8-Glu-Cer*

588.3–264.3

35

  8.00

C12-Glu-Cer*

644.4–264.2

35

11.68

C14-Glu-Cer

672.4–264.2

35

13.02

C16-Glu-Cer

700.5–264.2

35

14.35

C18:1-Glu-Cer

726.6–264.2

35

14.60

C18-Glu-Cer

728.6–264.2

35

15.50

C20:1-Glu-Cer

754.7–264.2

35

16.12

C20-Glu-Cer

756.7–264.2

35

16.45

C22:1-Glu-Cer

782.7–264.3

35

16.85

C22-Glu-Cer

784.6–264.3

35

17.39

C24:1-Glu-Cer

810.6–264.2

35

17.92 (continued)



Comprehensive Quantitative Analysis of Bioactive Sphingolipids

463

Compound

Mass Trans

Coll E (eV)

RT (min)

C24-Glu-Cer

812.7–264.2

35

18.65

C26:1-Glu-Cer

838.6–264.2

35

19.15

C26-Glu-Cer

840.7–264.2

35

20.28

Compound

Mass Trans

Coll E (eV)

C8-Lact-Cer*

750.5–264.3

40

7.38

Lact-Sph

624.5–282.1

32

1.72

C8-Lact-Cer*

750.5–264.3

40

7.38

C12-Lact-Cer*

806.5–264.2

40

10.68

C14-Lact-Cer

834.5–264.2

40

12.66

C16-Lact-Cer

862.5–264.2

40

13.92

C18:1-Lact-Cer

888.6–264.2

40

14.25

C18-Lact-Cer

890.5–264.2

40

15.03

C20:1-Lact-Cer

916.6–264.2

40

15.35

C20-Lact-Cer

918.5–264.2

40

16.14

C22:1-Lact-Cer

944.5–264.3

40

16.45

C22-Lact-Cer

946.6–264.3

40

17.08

C24:1-Lact-Cer

972.5–264.2

45

17.36

C24-Lact-Cer

974.7–264.2

45

18.21

C26:1-Lact-Cer

1,000.6–264.2

45

18.94

C26-Lact-Cer

1,002.7–264.2

45

19.64

8. Lact-Cer RT (min)

9. SM Compound

Mass Trans

Coll E (eV)

RT (min)

Lyso-SM

465.2–183.7

32

  2.05

C6-SM*

563.3–183.7

35

  4.80

C12-SM*

647.4–183.8

35

  8.86

C14-SM

675.4–183.8

35

10.62

C16-SM

703.4–183.7

35

12.21

C17-SM*

717.4–183.7

35

13.05

C18:1-SM

729.4–183.8

35

12.86 (continued)

464

Bielawski et al.

Compound

Mass Trans

Coll E (eV)

RT (min)

C18-SM

731.4–183.7

35

13.86

C20:1-SM

757.4–183.8

35

14.38

C20-SM

759.4–183.8

35

15.37

C22:1-SM

785.4–183.8

35

16.04

C22-SM

787.4–183.7

35

16.76

C24:1-SM

813.4–183.8

35

17.03

C24-SM

815.4–183.7

35

18.47

C26:1-SM

841.4–183.8

35

18.95

C26-SM

843.4–183.8

35

20.18

4. Notes 1. Other liquid samples like whole blood, serum, plasma, or urine can be extracted the same way. An aliquot, usually 100–200 ml, are diluted to 2.0 ml with serum-free media, then subjected to the above extraction protocol. In case of difficulties with aqueous/ organic phase separation, saturate the aqueous solution with sufficient amount of solid sodium chloride. 2. It is important to optimize the ionization conditions for each class of SPLs and collision energy for each individual molecular subspecies to be applied for quantitative MRM analysis. 3. The GluSph and GluCer actually represent Glu/GalSph and Glu/ GalCer. Glucosyl sphingosine (GluSph) and glucosyl-ceramide (GluCer) cannot be separated with these or any other HPLC conditions. We have tried so far their isobaring and isomeric galactosyl sphingosine (GalSph) and galactosyl-ceramide (GalCer), respectively. 4. * Indicates internal standards, MRM mass transitions (Mass Trans), collision energy (Coll E), and retention time (RT).

Final General Notes

1. Each target analyte is identified by its specific parent–daughter ion mass transition and retention time. 2. Typical analytical batches consist of an instrument blank followed by eight-point calibration standards, assay samples, and quality control samples (QC). The QC standards are interspaced after



Comprehensive Quantitative Analysis of Bioactive Sphingolipids

465

every 10–20 assay samples to verify performance. The sequences are then finished with QC standards to further verify the instrument performance. Both the calibration and QC standards undergo identical preparation procedure as the assay samples. 3. All data are acquired to the computers’ local hard drive for data processing. The collected data for each of the target analytes and internal standards are then processed using the instrument’s supplied data handling software package. 4. Calibration curves are constructed by plotting peak area ratios of the target analyte to its respective internal standard against analyte concentrations, using linear regression models. Any analytes for which no standards are available will be quantitated using the calibration curve of its closest counterpart.

Acknowledgments Financial support was provided by NCI Grant No. IPO1CA097132 and NIH/NCRR SC COBRE grant No. P20 RR017677. Special acknowledgement is for NCRR Grant No. CO6RR018823 providing laboratory space for Lipidomics Shared Resource in the CRI building of MUSC. References 1. Sastry, P.S., (1985) Lipids of nervous tissue: composition and metabolism. Prog Lipid Res 24(2), 69–176 2. Vos, J.P., Lopes-Cardozo, M., Gadella, B.M., (1994) Metabolic and functional aspects of sulfogalactolipids. Biochim. Biophys. Acta 1211, 125–149 3. Hannun, Y.A., Obeid, L.M., (2002) The ceramide-centric universe of lipid-mediated cell regulation: stress encounters of the lipid kind. J. Biol. Chem. 277, 25847–25850. 4. Adams, J., Ann, Q., (1993) Structure determination of sphingolipids by mass spectrometry. Mass Spectrom. Rev. 12, 51–85. 5. Ann, Q., Adams, J., (1993) Structure-specific collision-induced fragmentation of ceramides cationized with alkali-metal ions. Anal. Chem. 22, 7–13. 6. Ann, Q., Adams, J., (1993) Collision-induced decomposition of sphingomyelins for structural elucidation. Biol. Mass Spectrom. 22, 285–294. 7. Sullards, M.C., (2000) Sphingolipid metabolism and cell signaling. Methods Enzymol 312, 32–45.

8. Gu, M., Kerwin, J.L., Watts, J.D., Aebersold, R., (1997) Ceramide profiling of complex lipid mixtures by electrospray ionization mass spectrometry. Anal. Biochem. 244, 347–356. 9. Mano, N., Oda, Y., Yamada, K., Asakawa, N., Katayama, K., (1997) Simultaneous quantitative determination method for sphingolipid metabolites by liquid chromatography/ionspray ionization tandem mass spectrometry. Anal. Biochem. 244, 291–300. 10. Liebisch, G., Derobnik, W., Reil, M., Trumbach, B.R., Arnecke, R., Olgemoller, B., Roscher, A., Schmitz, G.J., (1999) Quantitative measurement of different ceramide spiecies from crude cellular lipid extracts by electrospray ionization tandem mass spectrometry (ESI-MS/MS). Lipid Res. 40, 1539–1546. 11. Sullard, M.C., Merrill, A.H., (2001) Analysis of sphingosine-1-phosphate, ceramides and other bioactive sphingolipids by highperformance liquid chromatography-tandem mass spectrometry. Science’s stke. 67, 1–11. 12. Folch, J., Lees, M., Sloane, H.S., (1956) A simple method for the isolation and purifica-

466

Bielawski et al.

tion of total lipids from animal tissues. J. Biol. Chem. 196, 497–509. 13. Bligh, E.G., Dyer, W.J., (1959) A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 911–917. 14. Bielawski, J., Szulc, Z.M., Hannun, Y.A., Bielawska, A., (2006) Simultaneous quantitative analysis of bioactive sphingolipids by highperformance liquid chromatography-tandem mass spectrometry. Methods 39, 82–91. 15. Bodennec, J., Brichon, G., Zwingelstein, G., Portoukalian, J., (2000) Purification of sphingoid classes by solid-phase extraction with aminopropyl and weak cation exchange cartridges. Methods Enzymol.. 312, 101–114. 16. Kralik, S.F., Du, X., Patel, C., Walsh, J.P., (2001) A method for quantitative extraction of sphingosine 1-phosphate into organic solvent. Anal. Biochem. 294, 190–193. 17. Ali, M.T., David, B.D., Peter, S.H., Valerie Boote Oral Microbiology Laboratory, (1998) Eukaryotic cell signaling and transcriptional activation induced by bacterial porins. FEMS Immunol. Medical Microbiol. 21(1), 57–64 18. Korachi, M., Blinkhorn, A.S., Drucker, D.B. (2002) Analysis of phospholipid molecular species distributions by fast atom bombardment mass spectrometry (FAB-MS). Eur. J. Lipid Sci. Technol. 104, 50–56 19. Murray, K.E., Schulten, H.R., (1981) Field desorption mass spectrometry of lipids. I. The application of field desorption mass spectrometry to the investigation of natural waxes. Chem. Phys. Lipids 29, 11–21. 20. Ma, Y.-C., and Kim, H.-Y. (1995) Development of the on-line HPLC/thermospray MS method for the analysis of phospholipid molecular species in brain. Anal. Biochem. 226, 293–301. 21. Edmond de, H., (1996) Tandem mass spectrometry: a primer. J. Mass Spectrom. 31, 129–137 22. Jackson, S.N., Wang, H.Y.J., Woods, A.S. (2005) Direct tissue analysis of phospholipids in rat brain using MALDI-TOFMS and MALDI-ion mobility-TOFMS. J. Am. Soc. Mass Spectrom. 16, 133–138 23. Suzuki, Y., Suzuki, M., Ito, E., Goto-Inoue, N., Miseki, K., Iida, J., Yamazaki, Y., Yamada, M., Suzuki, A., (2006) Convenient structural analysis of glycosphingolipids using MALDI-QIT-TOF mass spectrometry with increased lase power and collision gas flow. J. Biochem. 139, 771–777. 24. Ugarov, M., Egan, T., Koomen, J., Gillig, K.J., Fuhrer, K., Gonin, M., Schultz, J.A., (2004) Lipid/peptide/nucleotide separation with MALDI-Ion Momility-TOF MS. Anal. Chem. 76, 2187–2195. 25. Adams, J., Ann, Q., (1992) Structure determination of ceramides and neutral glycosphingol-

ipids by collisional activation of (M + Li)+ ions. J. Am. Soc. Mass Spectrom. 3, 260–263. 26. Vieu, C., Chevy, F., Rolland, C., Barbaras, R., Chap, H., Wolf, C., Perret, B., Collet, X., (2002) Coupled assay of sphingomyelin and ceramide molecular species by gas liquid chromatography. J. Lipid Res. 43, 510–522 27. Isaac, G., Bylund, D., Masson, J.E., Markides, K.E., Bergquist, J., (2003) Analysis of phosphatidylcholine and sphingomyelin molecular species from brain extracts using capillary liquid chromatography electrospray ionization mass spectrometry. J. Neurosci. Method 128, 111–119. 28. Karlsson, A.A., Michelsen, P., and Odham, G., (1998) Molecular species of sphingomyelin: Determination by high-performance liquid chromatography mass spectrometry with electrospray and high-performance liquid chromatography tandem mass spectrometry with atmospheric pressure chemical ionization. J. Mass Spectrom. 33, 1192–1198. 29. Monick, M.M., Mallampalli, R.K., Bradford, M., McCoy, D., Gross, T.J., Flaherty, D.M., Powers, L.S., Cameron, K., Kelly, S., Merrill, A.H., Hunninghake, G.W., (2004) Cooperative prosurvival activity by ERK and Akt in human alveolar macrophages is dependent on high level of acid ceramidase activity. J. Immunol. 173, 123–135. 30. Adams, J.M., Pratipanawatr, T., Berria, R., Wang, E., DeFronzo, R.A., Sullard, M.C., Mandarino, L., (2004) Ceramide content is increased in skeletal muscle from obese insulinresistant humans. Diabetes 53, 25–31. 31. Zheng, W., Kollmeyer, J., Symolon, H., Momin, A., Munter, E., Wang, E., Kelly, S., Allegood, J.C., Liu, Y., Peng, Q., Ramaraju, H., Sullard, M.C., Cobot, M., Merrill, A.H., (2006) Ceramides and other bioactive sphingolipids backbone in health and disease: Lipidomic analysis, metabolism and roles in memebrane structure, dynamic, signalic and autopaphy. Biochim. Biophys. Acta. 1758 (12), 1864–1884. 32. Maceyka, M., Sankala, H., Hait, N.C., LeStunff, H., Liu, H., Toman, R., Collier, C., Zhang, M., Satin, L.S., Merrill, A.H., Milstien, S., Spiegel, S., (2005) SphK1 and SphK2, sphingosine kinase isoenzymes with opposition functions in sphingolipid metabolism. J. Biol. Chem. 280, 37118–37129. 33. Bose, R., Verheij, R., Haimovitz-Friedman, A., Scotto, K., Fuks, Z., Kolesnick, R.N., (1995) Ceramide synthase mediates daunorubicininduced apoptosis; an alternative mechanism for generating death signals. Cells 82, 405–414. 34. Luberto, C., Hannun, Y.A., (1998) Sphingomyelin synthase, a potential regulator of intra-



Comprehensive Quantitative Analysis of Bioactive Sphingolipids

cellular levels of ceramide and diacylglycerol during SV40 transformation – Does sphingomyelin synthase account for the putative, phosphatidylcholine-specific phospholipase C? J. Biol. Chem. 273, 14550–14559. 35. Bielawska, A., Perry, D.K., Hannun, Y.A., (2001) Determination of ceramides and dig-

467

lycerides by the diglyceride kinase assay. Anal. Biochem. 298, 141–150. 36. Sullard, M.C., Wang, E., Peng, Q., Merrill, Jr. A.H., (2003) Metabolomic profiling of sphingolipids in human glioma Cell lines by liquid chromatography tandem mass spectrometry. Cell. Mol. Biol. 49, 789–797.

sdfsdf

Chapter 23 The Use of Charged Aerosol Detection with HPLC for the Measurement of Lipids Marc Plante, Bruce Bailey, and Ian Acworth Summary A gradient HPLC-charged aerosol detection method was applied to the measurement of different lipids including: free fatty acids, fatty alcohols, glycerides, steroids, phospholipids, and fat-soluble vitamins. An algal oil sample is used as an example. Charged aerosol detection offers a new approach to the routine analysis of any nonvolatile lipid. It shows low ng (on column) sensitivity, a dynamic range of over four orders of magnitude, good reproducibility, gradient compatibility, and similar analyte response for nonvolatile species, independent of chemical structure. Furthermore, charged aerosol detection uses mobile phases that are fully compatible with LC–MS – enabling both detectors to be used in a “lipidomics platform.” Key words: Lipids, Fatty acids, Fatty alcohols, Fat-soluble vitamins, Biofuels, Algal oils, Charged aerosol detection, Corona, HPLC

1. Introduction Lipids are a structurally diverse group of naturally occurring, water-insoluble compounds that can, for convenience, be divided into the following eight categories (see Note 1): fatty acyls (e.g., fatty acids), glycerolipids (e.g., monoacylglycerides, diacylglycerides, triacylglycerides), glycerophospholipids (e.g., phosphatidyl choline, phosphatidyl serine), sphingolipids, sterol lipids (e.g., cholesterol, bile acids, vitamin D), prenol lipids (e.g., vitamins E and K), saccharolipids, and polyketides (e.g., aflatoxin B1) (1). Then within a particular category, there can be great structural

Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, DOI 10.1007/978-1-60761-322-0_23 © Humana Press, a part of Springer Science + Business Media, LLC 2009

469

470

Plante, Bailey, and Acworth

complexity. For example, although triglycerides are composed of one glycerol molecule and three fatty acid molecules, differences in fatty acid chain length, degree of unsaturation, position of unsaturation, and position on the glycerol backbone, results in numerous triglyceride molecules. Lipids also participate in a variety of biochemically and physiologically important roles such as acting as structural components of cell membranes, a metabolic fuel, energy stores, vitamins, antioxidants, as signal transduction molecules, lubricants, and waxes (2, 3). Conversely, some lipid oxidation products (e.g., malondialdehyde, 4-hydroxynonenal) are cytotoxic, and the disruption of lipid biochemical pathways are associated with diabetes, cancer, neurodegeneration, and infectious diseases (4, 5). It has been estimated that the biological system lipidome is comprised of ~200,000 lipids (6), while the cellular lipidome contains ~1,000 distinct lipids (7), differing in their chemical structure, location, and level, making their global measurement extremely challenging. Therefore, lipid fingerprinting (i.e., the measurement of all lipids simultaneously) is particularly difficult and considered to be beyond one single analytical approach. Rather, a number of analytical techniques(8) are used to cover the lipidome – a brief synopsis of their advantages and disadvantages are presented in Table 1. A relatively new approach, HPLC with Corona® charged aerosol detection (CAD®), is proving to be useful for the measurement of nonvolatile lipids (9) and lipid peroxidation products (10). Lipids, typically lacking chromophores, do not exhibit any significant response to UV absorption, unless they are derivatized. Lipids are also typically nonvolatile, making them ideal for detection by evaporative light scattering detection (ELSD) or by charged aerosol detection. These two methods, though similar in their nebulization stage, use different means of detection: ELSD uses the amount of light scattered to correlate with analyte concentration, and CAD uses a measurement of charge that is carried by the analyte to an electrometer for the correlation (11). This method of measurement avoids the complex mixture of light scattering mechanisms present in ELSD, which creates a limited dynamic range and nonuniformity of response. The CAD, can provide a dynamic range of four orders of magnitude, allowing for the determination of a wide range of component concentrations, and is typically tenfold more sensitive than the ELSD. The correlation is also more consistent than that provided by ELSD, either fit with a linear correlation or with a quadratic curve over the entire range of concentrations. The principles of the Corona CAD have been reviewed elsewhere (12, 13). Two major categories of HPLC include normal phase and reverse phase chromatography: Normal phase consists of a relatively nonpolar mobile phase and a polar stationary phase

All

Phospholipids

P NMR

Poor sensitivity; poor resolution; lipid detection by chemical means (dyes; iodine vapor) or by radiolabel

Ionization suppression; matrix backgrounds

Ionization suppression; absolute quantification requires external standards; expensive

Nonvolatile and some semivolatile analytes only; mobile phase must be volatile; destructive

The Use of Charged Aerosol Detection with HPLC for the Measurement of Lipids

Well established; easy to perform, but can be labor intensive; reproducibility

Most lipid categories

TLC

Direct; easy to automate; wide selection of chromatographic chemistries; sensitive; wide dynamic range; analytes do not have to possess a chromophore; little variation in response between analytes

Good sensitivity; direct measurement (m/z)

Nonvolatile and some semivolatile

HPLC-CAD

Mobile phase limitations; narrow dynamic range; poor sensitivity; not gradient compatible

Relatively few lipids have active chromophores (e.g., carotenoids, retinoids). Low wavelength UV can measure unsaturated lipids but lack sensitivity and selectivity; wide variation in response (dependent upon chemical structure)

Requires derivatization. Auto-oxidation of polyunsaturated fatty acids may be an issue. Thermal degradation of some lipids

Low sensitivity

Complex spectra; low sensitivity; abundant lipids may dominate the spectrum

Disadvantages

Direct; easy to automate; wide selection of chromatoNonvolatile and some semivolatile analytes only; mobile graphic chemistries; analytes do not have to possess a phase must be volatile; moderate sensitivity; moderate chromophore dynamic range; destructive; wide variation in response between analytes

Many

Nonvolatile and some semivolatile

HPLC–ELSD

Direct; quantitative; nondestructive; simple; easy to automate

MALDI

All

HPLC–RI

Direct; quantitative; nondestructive; simple; easy to automate; wide selection of chromatographic chemistries

High sensitivity; high resolution; direct measurement (m/z); ease of automation; numerous LC methods

Lipids containing chromophores

HPLC–UV

Excellent resolution for neutral lipids in simple mixtures

Direct; quantitative; nondestructive

Direct; quantitative; nondestructive; structural information

Advantages

LC–ESI–MS/ Polar (ESI) and LC–APCI– more apolar MS (APCI)

Many

GC, GC–MS

31

1

H NMR

Lipids measured

Table 1 Comparison of different approaches used for measurement of the lipidome (based, in part, on (8) )

471

472

Plante, Bailey, and Acworth

(column packing), and reverse phase consists of a relatively polar mobile phase and a nonpolar stationary phase. These differences yield different separation results. Normal phase chromatography generally separates analytes based on adsorption of analytes on to the solid phase, and reverse phase chromatography separates analytes according to their relative hydrophobicity and the related partitioning between the mobile phase and stationary phase (14) In this chapter we explore the use of a reversed-phase-CAD approach for the measurement of numerous lipid species. Modification of the gradient to increase selectivity of different lipid classes will be presented. The application of these methods for the determination of free fatty acids, fatty alcohols, fat-soluble vitamins, and natural oils will be discussed.

2. Materials 2.1. Equipment

1. The HPLC system consisted of a Model 1100 HPLC with a quarternary pump, degasser, and autosampler, Agilent Technologies, Santa Clara, CA. 2. A Corona CAD Plus detector, ESA Biosciences, Inc., Chelmsford, MA. 3. Nitrogen gas supplied by Model N2–4000-L1252 Nitrogen Generator, ESA Biosciences, Inc., Chelmsford, MA. 4. HPLC Column heater, Model: Hot Pocket, Keystone Scientific, Charlotte, North Carolina. 5. HPLC column. Halo C8, 2.7 mm, 4.6 × 150 mm, Mac-Mod Analytical, Chadds Ford, PA.

2.2. Reagents and Supplies

1. In-house, deionized water. 2. Methanol, CMOS Grade, J.T. Baker #9073–05. 3. Alcohol, Denatured, Anhydrous, J.T. Baker #9229–03. 4. Acetonitrile, EMD Omnisolv, #AX0142–1. 5. Tetrahydrofuran, uninhibited, EMD Omnisolv, #TX0279–6. 6. Glacial acetic acid, J.T. Baker #9508–03. 7. Chloroform, Sigma-Aldrich #650498 or #288306 (see Note 2). 8. Butylated hydroxyanisole (BHA), Sigma-Aldrich #B87806. 9. Lauric Acid, Sigma-Aldrich #L-4250. 10. Myristic Acid, Sigma-Aldrich #M-3128. 11. Palmitic Acid, Sigma-Aldrich #P-0500. 12. Oleic Acid, Sigma-Aldrich #O-1008. 13. Linoleic Acid, Sigma-Aldrich #L-1376.



The Use of Charged Aerosol Detection with HPLC for the Measurement of Lipids

473

14. Linolenic Acid, Sigma-Aldrich #L-2376. 15. Steric Acid, Sigma-Aldrich #S-4751. 16. Tetradecanol, Sigma-Aldrich #185388. 17. Hexadecanol, Sigma-Aldrich #258741. 18. Octadecanol, Sigma-Aldrich #258768. 19. Eicosanol, Sigma-Aldrich #234494. 20. 20 Docosanol, Sigma-Aldrich #169102. 21. DPPC, Sigma-Aldrich #P-4329. 22. Retinol (Vitamin A), Sigma #R-7632. 23. Cholecalciferol (Vitamin D3), Sigma #C-9756. 24. alpha-Tocopherol (Vitamin E), Sigma #T-3251. 25. delta-Tocopherol (Vitamin E), Sigma #T-2028. 26. gamma-Tocopherol (Vitamin E), Fluka #89560. 27. Vitamin K1, Sigma #V-3501. 28. Luteine, Sigma #X-6250. 29. Algae Oil Extract, Sapphire Energy, San Diego, CA. 2.3. HPLC Mobile Phases

1. In a 1-L glass bottle, add 750 mL of methanol.

2.3.1. Mobile Phase A Preparation

3. Add 4 mL of glacial acetic acid.

2.3.2. Mobile Phase B Preparation

1. To a 1-L glass bottle, add 500 mL of acetonitrile.

2. Add 250 mL of deionized water. 4. Mix and degas (do not filter) the mobile phase under vacuum and sonicate for 10 min, or by other suitable means.

2. Add 375 mL of methanol. 3. Add 125 mL of tetrahydrofuran. 4. Add 4 mL of glacial acetic acid. 5. Mix and degas (do not filter) the mobile phase under vacuum and sonicate for 10 min, or by other suitable means.

3. Methods 3.1. HPLC System Parameters

1. Column temperature: 40°C 2. Sample temperature: 10°C 3. Flow rate: 0.8 mL/min 4. Injection volume: 10.0 mL 5. Run time: 72.0 min 6. General HPLC gradient ( Table 2)

474

Plante, Bailey, and Acworth

Table 2 General HPLC gradient

3.2. Corona CAD Parameters

Time (min)

%A

%B

0.00

100.0

0.0

40.00

30.0

70.0

60.00

10.0

90.0

65.00

10.0

90.0

65.10

100.0

0.0

72.00

100.0

0.0

1. Range: 500 pA for analog 2. Filter: NONE 3. Nebulizer Heater: ON (30°C) 4. Nitrogen Pressure: 35.0 + 0.2 psi. Nitrogen or compressed air supplied directly to the CAD must be free from particles and oils, and be regulated to maintain a uniform pressure of 35.0 ± 0.2 psi.

3.3. Sample Preparations

1. Sample Diluent, methanol:chloroform (1:1), was prepared by mixing 10 mL of methanol and 10 mL of chloroform and mixing. The diluent was stored in a glass vial and capped. For more hydrophobic analytes, the diluent can be adjusted to methanol:chloroform (1:3) without significantly affecting the chromatography. 2. For very hydrophobic materials, dissolving the analyte first in 3 parts chloroform and then adding 1 part methanol is helpful in the preparation. The greater the amount of methanol in the sample solution, the sharper the peaks will be. 3. All individual sample solutions were prepared by measuring 1.3 ± 0.3 mg of analyte into an HPLC vial and dissolving the analyte in Sample Diluent. Either vortex mixing or sonication was used to complete dissolution, if necessary. 4. The free fatty acids and fatty alcohols were dissolved in sample diluent, at a concentration of 1 mg/mL. 5. The vitamin samples (retinol, cholecalciferol, tocopherols, luteine, and Vitamin K1) were premade in ethanol/BHA (10 mg/L) at a concentration of 100–1,000 mg/mL and subsequently diluted through the synthesis of the combined analyte solution. 6. Once an appropriate sample diluent is determined, dissolve 10.0 mg of sample in 1.00 mL of diluent.



The Use of Charged Aerosol Detection with HPLC for the Measurement of Lipids

475

7. Dilute the sample 100 mL + 900 mL of diluent, forming a solution of 1.0 mg/mL in concentration. 3.4. Standard Preparations

1. For the full dynamic ranges presented here, standards can generally be prepared by dissolving 10.0 mg of each of the analytes of interest, in separate vials, in 1.0 mL of sample diluent. 2. If the retention times of the analytes need to be confirmed, 1.0 mL of each of these solutions can be injected. 3. A combined standard can be prepared by diluting 100 mL of each stock solution. Additional sample diluent may be used to obtain a final volume of 1.00 mL. 4. The final concentration of each standard analyte is calculated using Analyte Conc.(mg/mL) = Stock Conc.(mg/mL) ×

0.1(mL) . Final volume(mL)

5. Make serial dilutions according to Table 3. 3.5. Analysis

1. Inject a blank, consisting of the blank diluent, two times. This conditions the system as well as confirms the purity of the sample diluent. 2. Inject each of the prepared standards, 1–3 times, depending on system precision requirements, collecting peak areas for each of the analytes. 3. Plot peak area on the y-axis and column load or concentration on the x-axis. Fit the data either with a linear fit, or if that is unacceptable, a second-order polynomial fit (quadratic), y = Ax2 + Bx + C.. Using a smaller range of concentrations may allow for accurate linear fits. With a quadratic fit, concentrations or load can be calculated by using the quadratic formula, shown below.

Concentration or load =

−B + B 2 − 4 A (C − A rea) . 2A

Table 3 Serial dilution Standard

Aliquot (mL)

Diluent (mL)

Approximately Conc. (mg/mL)

A

(Stock)

(Stock)

1,000

B

400 of A

600

400

C

400 of B

600

160

D

400 of C

600

64

E

400 of D

600

25.6

476

Plante, Bailey, and Acworth

4. Accuracy can be measured by comparing the calculated concentration (use the peak area obtained by a standard injection in the equation above) against the actual value. If the values are within acceptable amounts (typically within 5 or 10% of the actual value), then the fit can be considered acceptable. 5. If poor fits or accuracy are observed, plot the standard concentrations (mg/mL) or standard load on-column (ng) on the y-axis and analyte peak area on the x-axis, and fit to a quadratic curve (inverse quadratic). With this fit, concentrations or load can be calculated by using the second-order polynomial, shown below. Concentration or load = A(Area)2 + B(Area) + C 3.6. Results

1. Free fatty acids

A series of runs for free fatty acids was performed on the general method gradient using seven compounds (lauric, myristic, palmitic, linolenic, linoleic, oleic, and stearic acids) at concentrations between 10 and 1,100 mg/mL in methanol/ chloroform (1:1). An overlay of the chromatograms is shown in Fig. 1. The retention times were reproducible, between 9.5 and 28 min for the acids tested. The change in response across the different acids is a result of difference in vapor pressure (volatility) of the different acids: as the vapor pressure increases, less of it is available to produce particles, and the amount of signal decreases. A standardization curve, using peak area on the y-axis against column load on the x-axis provided the quadratic fits, shown in Fig. 2. Correlation coefficients were all >0.997, but the accuracy of the fit was poor for the stearic acid, with points off of the curve. The equation relating concentration to peak area is found in the quadratic formula:

Fig. 1. HPLC-CAD chromatograms, overlay of free fatty acids, in the range of 10–1,100 mg/mL.

The Use of Charged Aerosol Detection with HPLC for the Measurement of Lipids

477

3000 Stearic

2500 Palmitic

2000

Peak Area



Oleic

1500

Linoleic

1000

Linolenic Myristic

500 Lauric

0

0

2000

4000

6000

8000

Column Load (ng)

10000

12000

Fig. 2. Standardization curves for free fatty acids, 250–11,000 ng on column using a quadratic fit.

Fig. 3. Standardization curves for free fatty acids, in the range of 250–11,000 ng on column using an “inverted” quadratic fit.

Improved accuracy can be obtained with smaller dynamic ranges or by changing the way in which the data are fitted. Better accuracy was achieved for the stearic acid when the standardization curves were created by plotting column load (ng) on the y-axis against peak area on the x-axis and fitting to quadratic curves (12). Plots are provided in Fig. 3.

478

Plante, Bailey, and Acworth

All correlation coefficients were slightly lower, at > 0.996, but the accuracy for stearic acid, with the higher degree of curvature, was improved. The accuracy of the other components was equivalent to those obtained with the quadratic fit, shown in Fig. 3. 2. Fatty Alcohols Analyses of fatty alcohols was performed using the general gradient method. Five alcohols (tetradecanol, hexdecanol, octadecanol, eicosanol, and docosanol) in concentrations ranging from 10 to 1,200 mg/mL in sample diluent, were analyzed. An overlay of the chromatograms is shown in Fig. 4. As with the free fatty acids, above, there is a decrease in response with increasing volatility, with tetradecanol being the smallest alcohol detectable with this method. Standardization curves, fit to a second-order polynomial, are shown in Fig. 5. All correlation coefficients were > 0.992. 3. Fat-Soluble Vitamins Vitamins play an important nutritional need, in terms of calcium incorporation in bones, radical suppression, etc. This method, as described here, resolves many of the fat-soluble vitamins, including Vitamin A (retinol, including derivatives), Vitamin D (D2 and D3 are not resolved), Vitamin E (all variations), Vitamin K1, and possibly others. Fat-soluble vitamin solutions were prepared in Ethanol/ BHA (10 mg/L) at 100 and 1,000 ppm concentrations. A mixed standard solution was prepared to maintain concentrations near 30 ppm and injected using the method described above. An HPLC–CAD chromatogram is shown in Fig. 6, and standardization curves are provided in Fig. 7. 4. Biofuels: Algal Oil A relatively new source of raw material for biofuels is being found in algae, which is simple to grow and can produce high

Fig. 4. HPLC-CAD chromatograms, overlay of Fatty Alcohols, in the range of 100–12,000 ng on column.



The Use of Charged Aerosol Detection with HPLC for the Measurement of Lipids

479

Fig. 5. Standardization curves for fatty alcohols, in the range of 200–12,000 ng on column using a quadratic fit.

Vitamin E delta Vitamin E gamma

Vitamin K1

Vitamin E alpha Vitamin D Vitamin A

Lutein

Fig. 6. HPLC-CAD chromatogram of fat-soluble vitamins in the range with 38 ng on column.

v­ olumes of oils per unit area of water. The oil from algae contains a complex mixture of compounds, containing a wide range of compounds and lipids of great variety.(15) For biofuel use, the hydrophobic compounds are the raw materials needed for transformation into a useful fuel. HPLC can be used to monitor both the raw material processes, as well as the quality of the finished product. A sample of algal oil, resulting from a hexane extraction of an algal sample in water, was dried and redissolved in sample diluent and injected under the conditions indicated above. A chromatogram of this sample, complete with labels of other classes of lipids, is provided in Fig. 8. These labels are the result of other experiments. As can be seen from this chromatogram, a wide variety of lipids are resolved. With a longer gradient, it may be possible

480

Plante, Bailey, and Acworth

Fig. 7. Standardization curves for fat-soluble vitamins, in the range of 9–300 ng on column using a quadratic fit. a-Toc alpha-tocopherol, d-Toc delta-tocopherol, g-Toc gamma-tocopherol.

Fig. 8. HPLC-CAD chromatogram of 1 mg/mL of Algal oil in sample diluent.

to further resolve other species not seen here. By adjusting the gradient to change over a longer period of time, further resolution may be possible. The hydrophilic compounds can be extracted from the oils or run through a silica column, before or after the chemical transesterification, to convert the material into a crude biofuel. This HPLC method resolves a broad range of lipids, from the relatively hydrophilic steroids to the hydrophobic triacylglycerides. The use of the Corona CAD allows for quantitation over a wide dynamic range, covering at least four orders of magnitude,



The Use of Charged Aerosol Detection with HPLC for the Measurement of Lipids

481

with coverage in the lower nanogram range when the concentration range is lowered. Compared to the ELSD, the Corona CAD provides for a uniform response curve over a wider range of concentrations, and typically with lower limits of detection. Free fatty acids are all resolved for all acids larger than lauric acid up to at least stearic acid, including the unsaturated forms (oleic, linoleic, and linolenic acids). The vapor pressure of these materials does vary, and this was found to contribute to lower response factors for the smaller acids. The same dependence on vapor pressure was seen with the fatty alcohols, with the larger alcohols providing more response per unit mass than the smaller, more volatile alcohols. Tetradecanol was able to be seen, but only at higher amounts injected on the column. Fat-soluble vitamins were examined, including vitamins A, D, E, and K1, as well as luteine. Other fat-soluble vitamins will also resolve, including for example, derivatives of vitamin A (data not shown). An example of the Corona CAD providing a “lipid fingerprint” was shown with the algal oil sample. A complex array of compounds were resolved. This method can provide a useful means of separating many compounds sufficiently well for further characterization studies. The Corona CAD is compatible with most mobile phases used with LC–MS. As nebulizer-based detectors are destructive to the sample, it is common to use the Corona CAD in parallel with the MS – the eluent from the column being split between the two detectors (16). Interestingly, Miroslav used HPLC–CAD and LC–MS to quantify triacylglycerols from several plant oils (17). As an analyte’s response (response factor, RF) on the Corona CAD is independent of its chemical structure, they concluded that “Compared to previously published quantitative methods based on the knowledge of RFs, the current method is cheaper (no TG standards are needed), faster (no calibration curves are required for the determination of relative concentrations), and the obtained precision is acceptable for most analytical purposes.”

4. Notes 1. See http://www.lipidmaps.org for lipid classifications. 2. The HPLC solvents used with the Corona CAD should contain the lowest values for particulates, or residue after evaporation, as possible. These solvents can typically be found in LC–MS grade solvents. Another characteristic to be aware of is the use of preservatives, including BHT in tetrahydrofuran. The

482

Plante, Bailey, and Acworth

chloroform used as a sample solvent in these experiments was stabilized with amylenes, which would appear as a retained “solvent” peak in the Corona CAD chromatograms when injected. Ethanol is also used as a preservative in chloroform (Sigma-Aldrich #288306), and this would not show any peak as an injected solvent. If mobile phase solvents contain nonvolatile preservatives, their use will result in high backgrounds, and possibly lower sensitivity for the detector.

References 1. Fahy, E., Subramaniam, S., Brown, A., Glass, C., Merrill, A., Murphy, R., Raetz, C., Russell, D., Seyama, Y., Shaw, W., Shimizu, T., Spencer, F., van Meer, G., VanNieuwenhze, M., White, S., Witztum, J., and Dennis, E. (2005). A comprehensive classification system for lipids. J. Lipid Res., 46, 839–861. 2. Watson, A. (2006). Lipidomics: A global approach to lipid analysis in biological systems. J. Lipid Res., 47, 2101–2110. 3. Stryer, L. (2006). Biochemistry, 6th Edition, W.H. Freeman and Co., New York. 4. Acworth, I. N. (2003). The Handbook of Redox Biochemistry. CD-ROM. ESA Biosciences. Part number: 70–6090. 5. Wenk, M. (2005). The emerging field of lipidomics. Nat. Rev. Drug Discov., 4, 594–610. 6. Seppanen-Laakso, T., and Oresic, M. (2008). How to study lipidomes. J. Mol. Endocrinol. In press. 7. van Meer, G., (2005). Cellular lipidomics. EMBO J., 24, 3159–3165. 8. Wenk, M.R. (2005) The emerging field of lipidomics. Nature, 4, 594–610. 9. Morreau, R. (2006). The analysis of lipids via HPLC with a charged aerosol detector. Lipids, 41, 727–734. 10. Cascone, A., Eerola, S., Ritieni, A., and Rizzo, A. (2006). Development of analytical procedures to study changes in the composition of meat phospholipids caused by induced oxidation. J. Chromatogr. A, 1120, 211–220. 11. Ramosa, R.G., Libonga, D., Rakotomangab, M., Gaudina, K., Loiseaub, P.M., Chaminade, P. (2008). Comparison between charged aerosol

detection and light scattering detection for the analysis of Leishmania membrane phospholipids. J. Chromatogr. A, 1209, 88–94. 12. Liu, X-K., Fang, J.B., Cauchon, N., Zhou, P.. (2008). Direct stability-indicating method development and validation for analysis of etidronate disodium using mixed-mode column and charged aerosol detector. J. Pharm. Biomed. Anal., 46, 639–644 13. Nair, L.M., Werling, J.O. (2009). Aerosol based detectors for the investigation of phospholipid hydrolysis in a pharmaceutical suspension formulation. J. Pharm. Biomed. Anal., 49(1), 95–99. 14. Snyder, L.R., Kirkland, J.J., Glajch, J.L. (1997). Practical HPLC Method Development, Second Edition, Wiley, New Jersey. 15. Cardozo, K.H.M., Guaratini, T., Barros, M.P., Falcao, V.R., Tonon, A.P., Lopes, N.P., Campos, S., Torres, M.A., Souza, A.O., Colepicolo, P., Pint, E. (2007). Metabolites from algae with economical impact. Comp. Biochem. Physiol. Part C: Toxicol. Pharmacol. 146(1,2), 60–78. 16. Crafts, C., Bailey, B., Plante, M., Acworth, I. (2009). Evaluation of method for the simultaneous analysis of cations and anions using HPLC with charged aerosol detection and a zwitterionic stationary phase. J. Chromatogr. Sci. In Print. 17. Miroslav, L., Lynen, F, Holˇcapek, M., and Sandra, P. (2007). Quantitation of triacylglycerols from plant oils using charged aerosol detection with gradient compensation. J. Chromatogr. A, 1176, 135–142.

Part III Lipid Maps

sdfsdf

1

Chapter 24 Non-invasive Mapping of Lipids in Plant Tissue Using Magnetic Resonance Imaging Thomas Neuberger, Hardy Rolletschek, Andrew Webb, and Ljudmilla Borisjuk

2 3 4 5

Summary

6

Plant oil has become an important component in the search for a replacement for non-renewable energy sources, as well as being used for a wide range of industrial purposes, all in addition to its vital importance for human diet. Detailed knowledge of the lipid distribution in plants is fundamental for the understanding of local regulatory networks covering storage metabolism, and for the development of new approaches for plant breeding and transgenic research. We here review a measurement protocol or “tool” based on magnetic resonance imaging (MRI), which allows the non-invasive detection and quantitative visualization of lipid in living plant tissue. The method provides quantitative lipid maps with a resolution close to the cellular level and can be used on a wide range of plants and is applicable at the level of individual tissues, organs, or entire plants during ontogeny. Lipid imaging is designed for both biotechnology and basic science and can be combined with histological, biochemical, and gene expression analysis. Here we present the method as practiced in our group, and discuss unique advantages and limitations of the lipid-imaging tool. Seeds of barley and rapeseed were used as a model for visualization of local oil accumulation at the organ- and tissue-specific scale.

7 8 9 10 11 12 13 14 15 16 17 18 19

Key words: Magnetic resonance imaging, Plant lipid, Oil storage, Seed development

1. Introduction

20

21

Visualization has become a very popular topic in plant science, exemplified by the article “Seeing is understanding” (1). Knowledge concerning lipid accumulation in vivo, i.e., in the living plant, is necessary in order to understand and manipulate such accumulation. Despite a domestication history of tens of thousands of years, the quantitative spatial distribution of lipid storage in crop plants has only been visually assessable recently. The acquisition of Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_24, © Humana Press, a part of Springer Science + Business Media, LLC 2009

485

22 23 24 25 26 27 28

486

Neuberger et al.

quantitative data on the lipid distribution within plant organs, and especially seeds, has traditionally relied largely on assays carried out after tissue dissection and extraction. Currently available non-destructive techniques typically allow either quantification or visualization of lipids (2–6), but not both. Recent progress has been based on techniques such as mass spectrometry (7), non-invasive magic-angle spinning 13C nuclear magnetic resonance (NMR) spectroscopy (8) and 1H NMR spectroscopy (9). All enable a detailed quantitative description of lipid composition, but are not appropriate for imaging. In similar vein, chemical imaging and other current lipid visualization methods are not reliable in terms of quantification in living seeds. For example, techniques such as fluorescence correlation spectroscopy imaging and multi-photon excitation have the potential for accurate in vivo measurements of lipids and numerous other compounds. However, they suffer from interference from photodamage, light scattering, and other characteristics of living plant tissues. Third-harmonic generation microscopy, for example, allows the visualization (10) and measurement of lipids in biological liquids (11). Unfortunately, the poor penetration of optical signals into and out of internal tissue layers represents a major problem, especially with respect to developing crop seeds. Non-destructive magnetic resonance (MR)-based approaches potentially enable both structural and quantitative investigations (8, 12, 13). In this article, a non-destructive magnetic resonance imaging (MRI) method is described for spatially resolved quantitative analysis of lipid storage in living plant seeds. Two models were chosen. The first model was developing grains of barley (Hordeum vulgare, monocot species). Its sophisticated lipid compartmentalization and extremely low lipid content (about 2% of dry weight) present a particular technical challenge. The second model was the developing rapeseed (Brassica napus, dicot species). Rapeseed accumulates about 45% oil and represents one of the most important oil plants.

29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75

1 .1. Advantages and Disadvantages of MRI

MR imaging and spectroscopy are routinely used as powerful techniques in medicine and animal research, and have recently become popular in experimental botany as well (14–16). Selective imaging of lipid distribution is relatively simple, relying on the slightly different resonance frequencies of water and lipid, and is performed using frequency-selective radiofrequency pulses. MRI allows the study of plants in a non-destructive and quantitative mode which opens up new perspectives for in  vivo analysis. The major advantages are as follows: 1. The non-invasive nature of the method, i.e. no requirement for tissue preparation. 2. The spatial resolution of lipid mapping is sufficiently high to distinguish local lipid content at the tissue level. Even with short



Non-invasive Mapping of Lipids in Plant Tissue

487

measurement times of about 1 h (necessary since the size and form of a growing plant organs/tissues can change rapidly) reasonable spatial resolutions can be achieved. Moreover, mapping can be performed using moderate magnetic field strengths which are available in many laboratories.

76

3. The lack of any RF penetration or scattering effects (unlike optical techniques) means that variable sizes and pigmentations of seed samples cause no problems

81 82

4. The simultaneous monitoring of anatomy and lipid deposition offers the possibility to relate lipid accumulation to the developmental changes of individual seeds. Following lipid mapping, seeds can be treated by any desired method (biochemical, histological, molecular analysis) or can be allowed to grow further.

84 85

5. Lipid mapping can be performed on different types of seeds independent of anatomical differences or lipid content. A careful calibration procedure enables quantitative information to be obtained. Lipid mapping can be applied to remarkably different types of plant tissues (seed, stem, root, and others), which differ widely in lipid content. The major limitation of the method is that detailed information on the chemical composition of the lipids is not easily obtained using MRI or magnetic resonance spectroscopy (8, 12). This is due to the inherent line width of the lipid signal in living seeds (depending on the magnetic field strength and the seed composition the line width could be of hundreds of Hertz) and the low concentrations of different single fatty acids. However, efforts are ongoing to try to mitigate this effect, and it may well be possible in the future to achieve the necessary spectral resolution for specific detection of single fatty acids.

90

2. Materials 2.1. Equipment

77 78 79 80

83

86 87 88 89 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105

106

1. All experiments were conducted on a Bruker 500-MHz Avance system (Bruker Biospin, Rheinstetten, Germany) or a Bruker 750-MHz Avance system. Both systems have a magnet with a clear bore of 89-mm diameter.

107

2. The gradient sets had a maximum strength of 660 mT/m in the case of the Avance 500 and 1,000 mT/m in the case of the Avance 750. The inner diameter, and therefore the usable space, was 40 mm in both cases.

111 112

3. Various commercially available and custom-built radiofrequency resonators were used in this study. Examples include a commercial 5-mm birdcage resonator (Bruker) used for

115

108 109 110

113 114 116 117

488

Neuberger et al.

barley imaging and a custom-built four turn solenoid coil for rape seed imaging.

118 119

4. Chromatograph Agilent 6890 (Waldbronn, Germany) fitted with a capillary DB-23 column (30 m × 0.25 mm; 0.25-mm coating thickness; J&W Scientific, Agilent, USA).

120 121 122 124

5. PALM Laser Microbeam instrument (Bernried, Germany) for dissection of very small samples if needed.

125

6. Basic lab equipment (centrifuge, pipettes, etc.).

123

126 127

2.2. Reagents

128 129

1. Fatty acid mixture (Sigma, Germany) containing methyl esters of arachidate, linoleate, linolenate, oleate, palmitate, and stearate; used for identification and calibration of individual fatty acids. 2. Chemicals used for fatty acid extraction: hexane, isopropanol, butylated hydroxytoluene, triheptadecanoate (internal standard), potassium sulphate, methanol, sodium methoxide, NaCl, HCl, acetonitrile, distilled water.

130 131 132 133

134

3. Methods

135

3 .1. Sample Preparation and Processing

MRI measurements were conducted on intact seeds, part of plants or entire plants at distinct developmental stages without any tissue preparation. If used for conventional metabolite analysis and calibration, samples were snap-frozen in liquid nitrogen immediately after the MRI experiment and stored at −80°C.

3 .2. Determination of Total Esterified Fatty Acids

Total fatty acids were analysed as their corresponding fatty acid methyl esters (FAMEs) by gas chromatography (13). FAMEs were prepared by direct transmethylation with methanol containing 2% (v/v) dimethoxypropane and 2.75% (v/v) sulphuric acid. After 1 h at 80°C, 0.2 ml of 5 M NaCl was added and FAMEs were extracted with 2 ml of hexane. The organic phase was dried with Na2SO4 and evaporated to dryness under nitrogen. FAMEs were dissolved in acetonitrile. The analysis of the FAMEs was performed with a gas chromatograph fitted with a capillary column (DB-23). Helium was used as carrier gas (1 ml/min). The temperature gradient was 150°C for 1 min, 150–200°C at 8 K/min, 200–250°C at 25 K/min and 250°C for 6 min. The total amount of lipids was calculated as sum of all detected FAMEs.

3 .3. Lipid Determination by MRI

As an example on how the lipid concentration in seeds can be determined with MRI, the work conducted on barley spikes is described in the following sections.

136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157



3.4. Data Acquisition

Non-invasive Mapping of Lipids in Plant Tissue

489

A single-slice spin-echo image was acquired using either a short 90° sinc pulse (which excites protons in both water and lipid) or a longer 8-ms sinc pulse (which, as a result of its bandwidth of ~780 Hz, excites only one of the resonances). In both cases, the slice select gradient was present only during the 180° refocusing pulse. With offset frequencies taken from a global spectrum, selective images of the water resonance, and the main lipid resonance were obtained. To validate the pulse sequences, experiments on a phantom containing a smaller vegetable oil filled tube inside a larger water filled tube were conducted. Figure 1a

Fig. 1. Application of lipid imaging for the in vivo investigation of developing seeds. (a) Selective frequency imaging using a phantom containing water (outer annulus) and vegetable oil (inner tube). Upper panel: non-selective image showing both water and oil in a cross-section through the tube. Water and oil are recognizable as bright signals, and glass is invisible/black. Midpanel : water selective image through the same tube. Bottom panel: lipid selective image. (b) Barley spike used for the NMR measurements. The red star indicates the individual grain visualized non-invasively in (c), (d), and (e). (c) Non-selective NMR image of a slice through an intact barley spike. (d) Water distribution in the grains. (e) Lipid distribution in the grains. 1 endosperm, 1* surface of endosperm (aleurone), 2 embryo, 3 pericarp, 4 stem/rachis, 5 husk, bar 3 mm. Figure modified with permission from ref. 13.

158 159 160 161 162 163 164 165 166 167

490

Neuberger et al.

shows a cross-section through the tubes: a non-selective image (upper panel) and two selective images (mid and bottom panel). The non-selective image visualized both water and oil. The selective images distinguish between water and oil, producing separate images from each. The same protocol was applied to living barley grains attached to the spike (Fig. 1b). The structural representation of the tissue pattern (non-selective image; Fig. 1c) and the distribution of water (Fig. 1d) and lipid (Fig. 1e) were visualized within the longitudinal section through the spike. When the lipid-specific pulse sequence was used, the lipid signal was detected only in the embryo and the surface of the endosperm (Fig. 1e). By contrast, the highest water signal was found in the stem/rachis, the husk, and the crease region of the grain (Fig. 1d).

168 169 170 171 172 173 174 175 176 177 178 179 180 181

185

To convert the signal intensity of the lipid-specific MRI image into an actual lipid content, a calibration procedure must be performed. The signal intensity at each image pixel for a spin-echo sequence is given by:

186

S ∝ M0 (1 – e –TR/T1) e–TE/T2,

187 188

where T1 is the spin–lattice relaxation time, T2 the spin–spin relaxation time, M0 the thermal equilibrium magnetization, TR the repetition time, and TE the echo time. Therefore, if there are substantial differences in the T1 or T2 values of the different organs, then these differences have to be accounted for in order to provide quantitative lipid distributions. Suitable choice of data acquisition parameters can simplify the quantitative process considerably. For example, if the value of TR is much greater than the longest T1 value in the sample, then differences in T1 values of different organs are not important. Similarly, if the TE value is much less than the shortest T2 value, then differences are again not important. In practice, the first criterion is much easier to achieve than the second. In order to address the potential issue of different T2 values, a multiple-echo spin-echo imaging sequence can be used to determine the T2 value on a pixel-by-pixel basis, followed by application of the appropriate correction factor. It should also be noted that if the RF field is spatially inhomogeneous (using, for example, a surface coil) a B1 map can be produced and used for correction. In all of our experiments, we used an RF coil with a homogeneous B1 field and so correction was not necessary. After acquiring MRI images from individual barley seeds of different developmental stages (early-, mid-, and late-storage stages) under identical experimental conditions the specific organs of the seeds were segmented from the images using a 3D visualization software package (AMIRA), and the average pixel intensity for each organ was determined using:

182 183 184

189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212

3 .5. Calibration Procedure for MR Imaging



Non-invasive Mapping of Lipids in Plant Tissue

avgSig organ =

491 213

∑ pixelSignal organ pixel

(∑ pixelorgan ) × avgNoise × VoxelVolume ×

NA

.

To take into account possible slight differences in data acquisition parameters, receiver gain settings, coil loading etc. in different samples, the signal is divided by the average noise, calculated from an area that does not contain any signal. To take into account the different number of signal averages which might be acquired in different scans, the sum of the pixel signals has to be divided by the square root of the number of averages (NA). Immediately after imaging, specific organs of the particular seeds were subsequently dissected using a stereomicroscope. The samples (dissected parts of the embryo, pericarp, branch endosperm, chalazal region) were freeze-dried and the total lipid content was measured using gas chromatography. The measured MRI signals of the various areas were plotted vs. the known total lipid content (gas chromatography data) of the respective dissected tissue samples. The data are shown as a calibration curve, where the ordinate shows the normalized MRI signal intensity and the abscissa shows the total lipid values in the corresponding tissues (Fig. 2). Each data point on the graph represents the average and standard deviation of two to three MR imaging measurements and gas chromatography readings. The calibration curve, represented by a least-squares linear fit, indicates that the choice of MRI data acquisition parameters enables a good correlation to be established between the lipid content measured by MR imaging and that by gas chromatography.

Fig. 2. Calibration curve for MRI-based lipid quantification. The MRI signal was plotted against a set of standard lipid samples. FW fresh weight.

214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237

492

Neuberger et al.

Finally, it should be noted that for some samples, with a very short T2 value, care has to be taken in calculating the actual spatial resolution, i.e., pixel size of the image. The resolution in the phase encoding direction is given by the image field-of-view divided by the number of data points acquired. In the frequency encoding direction, there is an additional contribution to the spatial resolution (∆xT2) given by:

238 239 240 241 242 243 244 245

∆x T2 =

where G is the value of the frequency encoding gradient. It is important, therefore, to use strong gradients in order to minimize the additional broadening factor.

246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281

1 , γ G pT2

3.6. Results

The fitted calibration curve in Fig. 2 was used to quantify the lipid content in various parts of the barley grain, providing quantitative images of lipid distribution. The resulting quantitative image of lipid distribution represents a colour coded “lipid map” (Fig. 3). The superposition of the lipid distribution derived from chemical analysis with high-resolution MR images enabled a detailed analysis to be made of lipid gradients within any tissue and/or developmental stage of the seed. A picture of intact barley seed harvested at the mid-developmental stage (25 days after flowering [DAF]) is shown in Fig. 3a. The seed was dissected by scalpel to demonstrate its anatomy. A non-selective MR image allows visualization of the longitudinal view of the grain, where all seed organs are recognizable (Fig. 3b left panel). A colour-coded lipid map within the same region shows the lipid distribution (Fig. 3b, right panel). The maps have an in-plane pixel resolution of 31 × 31 mm, and thus identify lipid distribution close to the cellular level. At 25 DAF, the strongest signal (given in red) was localized in the scutellum and nodular region of the embryo. A clear decrease in lipid content occurred in the direction of the coleorhizae and coleoptile. Particularly steep gradients were observed in the endosperm, where the highest lipid levels were present within the aleurone layer, with very low levels in the basal endosperm. The pericarp did not show lipid deposition. Overall, lipid mapping visualizes the tissue-specific deposition of lipids in a quantitative and non-invasive manner, with high spatial resolution. Lipid mapping reveals steep gradients within the developing barley grain. Lipid deposition in plant cells occurs mainly in lipid bodies, which possess a matrix of TAGs surrounded by a lipid monolayer that harbours oleosins as structural proteins (17). An embryo of barley grain was used for more detailed analysis (Fig. 3c). Transmission electron microscopy (TEM) was applied to relate



Non-invasive Mapping of Lipids in Plant Tissue

493

Fig. 3. Quantitative imaging of lipid in a living barley grain. (a) Morphology of the barley grain at the main storage stage after dissection of the grain via scalpel. Dotted lines show the level of the longitudinal section through the fresh harvested grain. (b) MRI (non-invasive) visualization of grain tissue structures at mid developmental stages. The lipid content is colour coded and given in units of mmol/gram fresh weight. (c) Distribution of lipid within the basal part of caryopsis showed by Sudan staining. (d) Lipid distribution and (e) tissue structure analysed by MRI in the same region. (f) Visualization of lipid bodies by TEM within the region of low lipid signal in coleorhizae and (g) high signal within the nodule. Abbreviations: a aleurone, cl coleoptile, co coleorhiza, e endosperm, em embryo, n nucleus, p pericarp, sc scutellum. Figure modified with permission from ref. 13.

the in  vivo lipid map to the intracellular localization of lipids. It was shown that high signal intensity in the barley lipid map is associated with the accumulation of lipid bodies. According to the lipid map (Fig. 3d), distinct regions of tissues (Fig. 3e) with high or low lipid level were dissected and analysed by TEM (Fig. 3f, g). A very low lipid signal was characteristic of the pericarp, whose cells contain cuticular compounds, plastoglobuli, and intracellular membrane systems, but only a few lipid bodies (data not shown). Similarly, the central part of the starchy endosperm, which generated no specific lipid signals in the MR map, contained neither lipid bodies nor oil deposits (except for the plastoglobuli within plastids). Tissues with a slightly elevated MR lipid signal had a low number of small lipid bodies (e.g. the embryo radicle (Fig. 3f). The moderate lipid level in the parenchymal nodule tissues (linking the coleorhizae with the coleoptile) was associated with numerous, large lipid bodies. High lipid levels within the scutellar parenchyma were associated with an increase in the size of the lipid bodies (Fig. 3g). Taken together, the gradual increase in the MR signal in the lipid map was associated with a shift from cells containing very small lipid bodies (coleorhizae and coleoptile)

282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301

494

Neuberger et al.

to cells accumulating numerous, variously sized, lipid bodies (nodule and scutellum). The highest lipid signal was observed in cells consisting predominantly of lipid bodies. Thus, the lipid map indicates the accumulation of lipid bodies, i.e. oil storage, within the tissues.

302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329

3 .6.1. Smaller Samples: Rapeseed

Analysis of small plant organs, such as the seeds of Arabidopsis, oilseed rape (Brassica napus) (Fig. 4a) or samples with more intricate form (roots, fruits) faces the difficulty of tissue isolation and the determination of tissues volumes, which are required for the calibration procedure outlined earlier. In this case tissue isolation can be performed by micro-dissection or using laser instruments coupled to laser pressure catapulting and reconstruction of 3D structure using commercial software, e.g Amira 4.1 (Fig. 4b). MRI studies on smaller samples are also more challenging. In order to obtain higher spatial resolution, higher static magnetic field strengths are desirable. The application of small custom-made solenoid coils adjusted to the size of object is recommended (18). If acquisition of a 3D dataset is not feasible, a multi-slice multi-echo (MSME) spin echo sequence can be substituted. Here we give the brief description of lipid mapping in rapeseed as example for the processing of intricate samples. The thickness of slices was 105 mm and in plane resolution was 30 × 30 mm2. A long repetition time (TR = 5 s) was used to avoid the need for T1 correction. Water suppression (to produce a lipid-only image) used a 90° sinc saturation pulse and gradient spoiling before the MSME sequence. From the multi-echo data, the T2 relaxation time was calculated on a pixel-by-pixel basis as

Fig. 4. Quantitative imaging of lipid in small-sized rape seed. (a) Morphology of the rape seed at the main storage stage as appears post-dissection of the seed per scalpel. (b) Reconstruction of grain structures at mid-developmental stages using Amira 4.1. (c) T2 map and (d) M0 map. (e) Segmentation (left panel) produces a tissue mask for each organ of the seed and determines the corresponding signal intensity (right panel) (f) The final lipid distribution shown as a colour coded representation. Abbreviations: oc outer cotyledon, ic inner cotyledon, ra radicle.



Non-invasive Mapping of Lipids in Plant Tissue

495

shown in the T2 map in Fig. 4c. An average T2 of 21.3 ± 5.3 ms was calculated from the whole seed. A T2 correction was applied and an example of a resulting net magnetization (M0) map is shown in Fig. 4d. The M0 map represents the actual spin density distribution of the lipids in the seed. A B1 map showed that the homogeneity of the coil was sufficiently high so as not to require a B1 field correction. Figure 4e shows on the right the mask and on the left the overlay of the mask with the M0 map of the three segmented organs that were cut out and processed as described before to get an average lipid concentration with the gas chromatography method. Figure 4b shows the masks of the whole seed as a 3D model. Having measured the average lipid concentration in the organs the final quantitative lipid distribution in the seed could be calculated (Fig. 4f). Compared to the gas chromatography method, which only gives an average lipid concentration, the MRI-based method shows a detailed quantitative lipid distribution on a pixel-by-pixel basis throughout the seed. Lipid gradients inside specific organs are now visible.

Acknowledgments

330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348

349

We are grateful to P. Jakob and U. Wobus for support and critical discussion, G. Melkus, A. Purea, and V. C. Behr for help with the MRI. I. Feussner and C. Göbel are acknowledged for help with gas chromatography. Our special thanks to T. Rutten for electron microscopy, U. Tiemann and K. Lipfert for artwork and A. Stegmann for excellent technical assistance. We are grateful for funding by the Deutsche Forschungsgemeinschaft (Project number RO 2411/2–1/2–2 and BO 1917/2–1), and the Federal Ministry of Education and Research (BMBF; GABI SEED II grant). Andrew G. Webb and Thomas Neuberger acknowledge support of the Alexander von Humboldt Stiftung, Wolfgang Paul Preis. References 1. Breithaupt H. (2006) Seeing is understanding. EMBO; Rep. 7: 467–470. 2. Kano H, Ishida N, Kobayashi T, Koizuml M. (1990) 1H-NMR imaging analysis of changes of free water distribution in barley and soybean seeds during maturation. Jpn. J. Crop Sci.; 59: 503–509. 3. Pope JM, Jonas D, Walker RR. (1993) Applications of NMR micro-imaging to the study of water, lipid, and carbohydrate distribution in grape berries. Protoplasma; 173: 357–363.

350 351 352 353 354 355 356 357 358 359 360 361 362

4. Baeten V, Aparicio R. (2000) Edible oils and fats authentication by Fourier transform Raman spectrometry. Biotechnol. Agron. Soc. Environ.; 4: 196–203. 5. Baranska M, Schulz H, Reitzenstein S, Uhlemann U, Strehle MA, Kruger H, Quilitzsch R, Foley W, Popp J. (2005) Vibrational spectroscopic studies to acquire a quality control method o f Eucalyptus essential oils. Biopolymers; 78: 237–248.

363 364 365 366 367 368 369 370 371 372

496 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400

Neuberger et al.

6. Borisjuk L, Nguyen TH, Neuberger T, Rutten T, Tschiersch H, Claus B, Feussner I, Webb AG, Jakob P, Weber H, Wobus U. and Rolletschek H. (2005) Gradients of lipid storage, photosynthesis and plastid differentiation in developing soybean seeds. New Phytol.; 167: 761–776. 7. Adlof, R.O. (2003) Advances in Lipid Methodology – Five. Bridgwater, Somerset: The Oily Press. 8. Ratcliffe RG, Shachar-Hill Y. (2001) Probing plant metabolism with NMR. Annu. Rev. Plant Physiol. Plant Mol. Biol.; 52: 499–526. 9. Knothe G, Kenar JA. (2004) Determination of the fatty acid profile by 1H-NMR spectroscopy. Eur. J. Lipid Sci. Technol.; 106: 88–96. 10. Débarre D, Supatto W, Pena AM, Fabre A, Tordjmann T, Combettes L, Schanne-Klein MC, Beaurepaire E. (2006) Imaging lipid bodies in cells and tissues using third-harmonic generation microscopy. Nat. Methods; 3: 47–53. 11. Débarre D, Beaurepaire E. (2007) Quantitative characterization of biological liquids for third-harmonic generation microscopy. Biophys. J.; 92: 603–612. 12. Köckenberger W, DePanfilis C, Santoro D, Dahiva P, Rawsthorne S. (2004) High resolution

NMR microscopy of plants and fungi. J. Microsc.; 214: 182–189. 13. Neuberger T, Sreenivasulu N, Rokitta M, Rolletschek H, Göbel C, Rutten T, Radchuk V, Feussner I, Wobus U, Jakob P, Webb A, Borisjuk L. (2008) Quantitative imaging of oil storage in developing crop seeds. Plant Biotechnol. J.; 6: 31–45 14. Ishida N, Koizumi M, Kano H. (2000) The NMR microscope: a unique and promising tool for plant science. Ann. Bot.; 86: 259–278. 15. Van der Weerd L, Claessens MAE, Ruttink T, Vergeldt FJ, Schaafsma TJ, Van As H. (2001) Quantitative NMR microscopy of osmotic stress responses in maize and pearl millet. J. Exp. Bot.; 52: 2333–2343. 16. Van As H. (2007) Intact plant MRI for the study of water relations, membrane, permeability, cell-to-cell and long distance water transport. J. Exp. Bot.; 58: 743–756 17. Huang AHC. (1992) Oil bodies and oleosins in seeds. Annu. Rev. Plant Physiol. Mol. Biol.; 43: 177–200. 18 Webb AG. (1997) Radiofrequency microcoils in magnetic resonance. Prog. NMR Spectrosc.; 31: 1–42.

Chapter 25 Mapping the Lipolytic Proteome of Adipose Tissue Using Fluorescent Suicide Inhibitors Maximilian Schicher, Manfred Kollroser, and Albin Hermetter Summary Lipases are responsible for the hydrolysis of acylglycerols and cholesteryl esters in animals, plants, and microorganisms. In this chapter we describe a tool for the concomitant analysis of lipases in complex proteomes. For this purpose, the target enzymes are selectively and covalently labelled with fluorescent suicide inhibitors. Stable lipid–protein complexes are formed that are separated by gel electrophoresis and identified by mass spectrometry. Key words: Lipase, Esterase, Serine hydrolase, Functional proteomics, Fluorescent lipid, Lipase inhibitor

1. Introduction In mammals and humans, lipids are stored as intracellular triacylglycerols (TG) and cholesteryl esters (CE) in adipose tissue as the primary source of energy during periods of starvation and increased energy demand. Lipolytic enzymes are responsible for mobilization and degradation of these lipids and therefore are important for regulation of energy homeostasis. Hydrolysis of TG and CE leads to the release of free fatty acids (FA), which can be a risk factor for the development of insulin resistance in type-2 diabetes or related disorders, if their concentration varies (1). In adipose tissue, cholesteryl esters are hydrolysed by cholesteryl esterases, whereas the degradation of triacylglycerols is achieved by three different lipases. Adipose triglyceride lipase (ATGL) is predominantly responsible for the initial step in TG hydrolysis resulting in the formation of diacylglycerols (DG) and free fatty acids. Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_25, © Humana Press, a part of Springer Science + Business Media, LLC 2009

497

498

Schicher, Kollroser, and Hermetter

Hormone-sensitive lipase (HSL) hydrolyses diacylglycerols (and to a lesser extent triacylglycerols and monoacylglycerols) leading to monoacylglycerols (MG) and free fatty acids. The final step is catalysed by monoglyceride lipase (MGL) which hydrolyses monoacylglycerols to form glycerol and fatty acids (2, 3). The active site of lipases and esterases, which belong to the family of serine hydrolases, usually contains a catalytic triad formed by Ser–His–Asp/Glu. The serine residue, which is localized to a highly conserved GXSXG motif, acts as the nucleophile in the initial step of carboxyl ester hydrolysis (4). There are several methods available for the determination of lipase activities (5–8), which are based on different types of substrates including natural and synthetic triacylglycerols as well as radioactive or chromogenic lipids. We used the technique of activity-based proteomics (9–12) for identifying ”all” active lipolytic enzymes in a proteome. For this purpose, we designed substrate analogous inhibitors for specific fluorescence labelling of lipases and esterases in mouse adipose tissue. A typical activitybased probe contains (a) a recognition site specific for a certain enzyme class, (b) a reactive site that forms a covalent bond with the enzyme, and (c) a tag for visualization and/or purification of the covalently bound target. We have developed biotin- and fluorescently labelled alkylphosphonates of different polarities as activity-based probes which are well suited to profile serine hydrolase activity in complex proteomes. The nucleophilic serine in the active site of lipolytic enzymes forms a covalent bond with the phosphonate of the probe, whereas the p-nitrophenyl group acts as a leaving group (Fig. 1). The stable probe–enzyme complexes can be separated by gel electrophoresis and visualized on the basis of their fluorescence using a laser scanner. Fluorescent protein spots are excised from the gel, tryptically digested and

Fig. 1. Specific tagging of lipolytic enzymes by activity-based probes. The nucleophilic serine in the active site of lipolytic enzymes replaces the p-nitrophenyl group in the fluorescent suicide inhibitor. The enzyme is irreversibly inactivated due to the formation of a stable fluorescent lipid–protein complex.



Adipose Tissue Using Fluorescent Suicide Inhibitors

499

i­dentified by mass spectrometry (LC–MS/MS). Using this system we were able to detect ”all” known lipolytic enzymes as well as novel enzyme candidates. The fluorescent (NBD-labeled) activity-based probes described in this chapter include phosphonic acid esters mimicking a single-chain carboxylic acid ester (NBD-HE-HP), a triacylglycerol (NBD-TG), and a cholesterol ester (NBD-CP) (Fig. 2) (1). It has to be noted, that the fluorescent protein patterns can be extremely different depending on the inhibitor. NBD-HE-HP, for example, is the simplest but least specific inhibitor of the lipid set presented here, and therefore leads to the most complex protein pattern. NBD-CP on the other hand, is a very selective inhibitor and therefore generates a very simple protein pattern. The ”history” of the samples has also a profound impact on the apparent proteomes detected by the phosphonates. Freezing and thawing of the samples do not only lead to a loss in overall activity but also to changes in the lipolytic proteomes, because the individual enzyme components are likely to exhibit different stabilities.

Fig. 2. Phosphonate activity-based probes for lipolytic enzymes. (a) NBD-HE-HP mimicks a single-chain carboxylic acid ester, (b) NBD-TG mimicks a triacylglycerol, and (c) NBD-CP mimicks a cholesterol ester.

500

Schicher, Kollroser, and Hermetter

In this chapter, detailed protocols are provided for homogenization of mouse adipose tissue, tagging of lipolytic enzymes using specific activity labels, and the separation and identification of these enzymes (1). These protocols have also been used for the identification of the lipolytic proteomes in other tissues including mouse liver subfractions (13), heart muscle, skeletal muscle, kidney, adrenal glands, and brain (M. Morak, M. Geier, H. Susani-Etzerodt, G. Rechberger, M. Kollroser, R. BirnerGruenberger, A. Hermetter, unpublished).

2. Materials 2.1. Equipment

1. Motor-driven Teflon-glass homogenizer (Schuett Labortechnik GmbH, Göttingen, Germany). 2. Centrifuge (maximum force: ~10,000 × g). 3. Thermomixer (Eppendorf Thermomixer compact, Eppendorf, Vienna, Austria). 4. Argon 4.6 (Linde Gas, Stadl-Paura, Austria). 5. Equipment for gel casting (Spacer plates with 1.0-mm integrated spacers, short plates, Mini-PROTEAN Combs, 10-well, 1.0 mm, Mini-PROTEAN 3 Multi-Casting Chamber, PROTEAN Plus Hinged Spacer Plates 1.5 × 200 × 205 mm, PROTEAN Plus Multi-Casting Chamber, Bio-Rad, Vienna, Austria). 6. Electrophoresis system (Bio-Rad Mini PROTEAN 3 Dodeca cell or PROTEAN Plus Dodeca cell, Bio-Rad, Vienna, Austria). 7. Rehydration Tray for IPG Strips (Serva, Heidelberg, Germany). 8. Isoelectric focusing (IEF) system (Amersham Biosciences Multiphor II, GE Healthcare, Vienna, Austria). 9. Laser scanner (Bio-Rad Molecular Imager™ FX Pro Plus, Bio-Rad, Vienna, Austria). 10. Scalpel. 11. Vacuum centrifuge (Savant ISS110 SpeedVac® System, Thermo, Breda, Netherlands). 12. Nano-HPLC system (FAMOS™ autosampler, SWITCHOS™ loading system, and ULTIMATE™ 3000 dual gradient system, LC Packings, Amsterdam, Netherlands). 13. m-Pre-column™ Cartridge (LC Packings Acclaim PepMap C18, 5 mm, 100 Å, 300-mm inner diameter × 5 mm). 14. Nanocolumn (LC Packings Acclaim PepMap 100 C18, 3 mm, 100 Å, 75-mm inner diameter × 150 mm).



Adipose Tissue Using Fluorescent Suicide Inhibitors

501

15. Finnigan nano-ESI source equipped with NanoSpray tips (PicoTip™ Emitter, New Objective). 16. Thermo-Finnigan LTQ linear ion trap mass spectrometer. 17. Software Spectrum Mill Rev. 03.03.078 (Agilent). 2.2. Reagents

1. Phosphate-buffered saline (PBS) solution.

2.2.1. Fractionation of Mouse Adipose Tissue

2. Lysis buffer: 0.25 M sucrose, 1 mM EDTA, 1 mM dithioerythritol (DTT), 20 mg/ml leupeptin, 2 mg/ml antipain, and 1 mg/ml pepstatin. 3. Bio-Rad Protein Assay Dye Reagent Concentrate (Bio-Rad, Vienna, Austria).

2.2.2. Activity Tagging

1. Lysis buffer (see Subheading 2.2.1). 2. 10 mM Triton X-100 in chloroform (CHCl3). 3. Stock solution of activity tag: 1 nmol/10 ml chloroform (0.1 mM). 4. Argon 4.6 (Linde Gas, Stadl-Paura, Austria). 5. 50% (w/v) trichloroacetic acid solution. 6. Acetone p.a. 7. 1D sample buffer. 20 mM KH2PO4, 6 mM EDTA, 60 mg/ml sodium dodecyl sulphate (SDS), 100 mg/ml glycerol, 0.5 mg/ml bromophenol blue, 20 ml/ml mercaptoethanol, pH 6.8. 8. 2D sample buffer. 7 M urea, 2 M thiourea, 4% (w/v) CHAPS, 60 mM DTT, 2% (v/v) Pharmalyte pH 3–10, 0.002% (w/v) bromophenol blue.

2.2.3. Gel Electrophoresis

1. Acrylamide solution 40% (Rotiphorese® Gel 40 (37.5:1 (w/w) Acrylamide:Bisacrylamide), Lactan, Graz, Austria). 2. 1 M Tris–HCl, pH 6.8. 3. 1 M Tris–HCl, pH 8.8. 4. SDS. 5. Ammonium peroxodisulfate (APS). 6. N,N,N¢,N¢-Tetramethylethylenediamine (TEMED, electrophoresis grade). 7. 1-Butanol saturated with water. 8. Running buffer: 3 g/l Tris-(Hydroxymethyl)-Aminomethan (TRIS, buffer grade, Lactan, Austria), 334 mg/l EDTA, 14.2 g/l glycin, and 1 g/l SDS 9. Immobiline™ DryStrip pH 3–10 NL, 7 cm or 18 cm (GE Healthcare Life Sciences, Vienna, Austria). 10. Immobiline™ DryStrip Cover Fluid (GE Healthcare Life Sciences, Vienna, Austria).

502

Schicher, Kollroser, and Hermetter

11. Equilibration buffer. 50 mM Tris–HCl, pH 8.8, 6 M urea, 30% (v/v) glycerol, 2% (w/v) SDS, 0.002% (w/v) bromophenol blue. 12. DTT. 13. Iodoacetamide (IA). 14. Agarose sealing solution. 1% Agarose low melt and 0.002% (w/v) bromophenol blue in running buffer (see Step 8). 15. Fixing solution. 7.5% (v/v) acetic acid, 10% (v/v) ethanol 2.2.4. Visualization

1. SYPRO® Ruby protein gel stain (Invitrogen, Lofer, Austria).

2.2.5. Proteolytic In-Gel Digestion

1. EtOH p.a. 2. 100 mM NH4HCO3 buffer. 3. Acetonitrile p.a. 4. Reduction buffer. 10 mM DTT in 100 mM NH4HCO3. 5. Alkylation buffer. 55 mM Iodoacetamide in 100 mM NH4HCO3. 6. Trypsin (sequencing grade modified, Promega, Mannheim, Germany). 7. Trypsin solution. Dissolve 20 mg of trypsin in 200 ml of 1 mM HCl. 8. Digestion buffer. 15 ml trypsin solution, 50 ml of 100 mM NH4HCO3, 50 ml bidistilled water, 5 ml of 120 mM CaCl2 × 2H2O. 9. Incubation buffer. 65 ml bidistilled water, 50 ml of 100 mM NH4HCO3, 5 ml of 120 mM CaCl2 × 2H2O. 10. Extraction buffer. 25 mM NH4HCO3 (dilute 100 mM NH4HCO3 1:4 in H2O). 11. 5% (v/v) formic acid.

2.2.6. LC–MS/MS Analysis

1. Formic acid (HPLC grade). 2. Acetonitrile (HPLC grade).

3. Methods 3.1. Homogenization of Mouse Adipose Tissue

1. Surgically remove adipose tissue of a mouse, wash it in PBS, and put it in a glass petri dish. 2. Put the tissue on ice, add lysis buffer (2–5 ml) and cut it in little pieces. 3. Transfer the pieces to a Falcon tube and add fresh lysis buffer so that the pieces are covered with buffer (see Note 1).



Adipose Tissue Using Fluorescent Suicide Inhibitors

503

4. Homogenize the tissue pieces with the Teflon-glass homogenizer at 1,500 rpm for 5 s, then pause for 10 s (put the sample on ice). Repeat this cycle 5–8 times. 5. Transfer the homogenate back to the Falcon tube and centrifugate at 1,000 × g for 15 min at 4°C. 6. Transfer the lysate to a fresh Falcon tube and discard the pellet (see Note 2). 7. Quantify the protein in the lysate using the Bio-Rad protein assay (see Note 3). 8. The protein sample is now ready to be used, or store it at –70°C. 3.2. Activity Tagging

1. Adjust the concentration of the protein sample to 0.5 mg/ ml using lysis buffer. 2. Mix 10 ml of 10 mM Triton X-100 and 20 ml of 0.1 mM activity tag in a 1.5 ml Eppendorf tube and evaporate the organic solvent under a stream of argon (see Note 4). 3. Add 100 ml of protein sample (0.5 mg/ml) and incubate the resultant mixture on a thermomixer for 2 h at 37°C and 550 rpm under protection from daylight (see Note 5). 4. Put the Eppendorf tube on ice and add ice-cold 50% trichloroacetic acid solution to a final concentration of 10%. Leave the sample on ice for 1 h under protection from daylight. 5. Centrifuge the suspension at 4°C at 10,000 × g for 15 min. Remove the supernatant. 6. Wash the pellet once with 150 ml ice-cold acetone and centrifuge for another 5 min at 4°C at 10,000 × g. Remove the supernatant. 7. Resuspend the pellet in 1D sample buffer (20 ml per 30 mg) or 2D sample buffer (125 ml per 50 mg, for 7 cm strips or 340 ml per 500 mg, for 18 cm strips) (see Note 6). 8. The tagged protein samples can now be used for gel electrophoresis or stored at −20°C (see Note 7).

3.3. Gel Electrophoresis 3.3.1. 1D Gel Electrophoresis

1. Cast a separating gel containing 10% acrylamide. The amounts of reagents required for two gels with dimensions 1 × 73 × 83 mm are as follows: 8.6 ml of 40% acrylamide solution, 12.9 ml of 1 M Tris–HCl, pH 8.8, 0.1% SDS, and 12.4 ml of bidistilled water. Start polymerization by adding 165 ml of a freshly prepared solution of 10% (w/v) APS in water, and 34 ml of TEMED. Mix thoroughly and pour between prepared glass plates. Cover with 1 ml of 1-Butanol saturated with water and leave for approximately 45 min to polymerize (see Notes 8 and 9).

504

Schicher, Kollroser, and Hermetter

2. Thoroughly wash off the 1-Butanol with water and overlay the separating gel with a 4.5% stacking gel. The amounts of reagents for two gels are: 0.8 ml of acrylamide solution 40%, 0.8 ml of 1 M Tris–HCl, pH 6.8, 0.1% SDS, and 4.9 ml of bidistilled water. Start polymerization by adding 33 ml 10% (w/v) APS and 7 ml TEMED. Put a comb in the stacking gel and leave it for 45 min to polymerize. 3. Fill cold running buffer in the Bio-Rad Mini PROTEAN 3 Dodeca cell and keep it at 15°C throughout the entire electrophoresis run. 4. Load the protein samples in 1D sample buffer in the gel pockets (20 ml per pocket) and put the gels in the electrophoresis cell. 5. Start electrophoresis at 20 mA constant current per gel, and stop electrophoresis when the bromophenol leaves the bottom of the gel. 6. Put the gels in fixing solution and slightly shake them for at least 30 min under protection from daylight. 3.3.2. 2D Gel Electrophoresis

1. Fill the protein samples dissolved in 2D sample buffer in the rehydration tray for IPG strips and put the Immobiline™ DryStrips with the gel side down onto the solutions. Overlay the strips with 1 ml of DryStrip Cover Fluid and rehydrate at room temperature over night under protection from daylight (see Note 10). 2. Transfer the rehydrated IPG strips to the IEF system and start isoelectric focusing of the proteins at 6.5 kV h (7 cm strips) or 12 kV h (18 cm strips). After IEF, immediately use the strips for second-dimension separation or store at –70°C. 3. Prepare the agarose sealing solution by heating to 75°C for at least 30 min. Then keep it at 45°C before use (step 7), to make sure it remains fluid. 4. Place the IPG strips in individual tubes for equilibration. Add 100 mg of DTT per 10 ml of equilibration buffer and add approximately 5 ml of the DTT-containing buffer to each tube. Fix the tube horizontally and shake for 15 min (see Note 11). 5. Prepare a solution of 250 mg of iodoacetamide per 10 ml of equilibration buffer (without DTT). Empty the tubes, add approximately 5 ml of IA-containing buffer to each tube and equilibrate for another 15 min under protection from daylight (see Note 11). 6. Cast 10% separating gels according to step 1 in Subheading 3.3.1 for 7 cm strips (for 18 cm strips, the amounts of reagents required for two gels with dimensions 1.5 × 200 × 205 mm are:



Adipose Tissue Using Fluorescent Suicide Inhibitors

505

46.9 ml of 40% acrylamide solution, 70 ml of 1 M Tris–HCl, pH 8.8, 0.1% SDS, 67.5 ml of bidistilled water, 896 ml of 10% (w/v) APS, and 186 ml of TEMED). Wash off the 1-butanol (see Note 12). 7. Place an IPG strip on top of the gel with the positive end on the upper left. Cover the strip with agarose sealing solution and wait for 5 min for the agarose to polymerize. Put the gel in the electrophoresis cell and separate proteins at 20 mA constant current per gel (Mini PROTEAN 3 Dodeca cell, small gels) or at 200 V constant voltage (PROTEAN Plus Dodeca cell, large gels). 8. Put the gels in fixing solution and slightly shake them for at least 30 min under protection from daylight. 3.4. Visualization

1. Put the fixed gels on the glass plate of the laser scanner. 2. Scan the gels with the excitation and emission wavelengths appropriate to the optical properties of the fluorescence activity tag used (for NBD, excitation wavelength is 488 nm and emission wavelength is 530 nm). 3. For visualization of the whole protein pattern, stain the gels in a SYPRO Ruby solution for approximately 4 h and then put them in fixing solution for 30 min. Scan the gels at 605 nm using an excitation wavelength of 488 nm.

3.5. Proteolytic In-Gel Digestion

Always wear gloves and be very careful not to contaminate your probes.

3.5.1. Gel Fragment Preparation

1. Excise fluorescent protein bands from the gel and cut each into 1-mm pieces. 2. Transfer the pieces to a fresh Eppendorf tube. Samples can be stored in 10% aqueous EtOH at –20°C.

3.5.2. Washing

Perform all washing steps using a thermomixer at 37°C and 550 rpm. 1. Wash the gel pieces with at least 10 volumes of bidistilled water (~ 150 ml) for 5 min. Remove the supernatant and discard it. 2. Add 150 ml of freshly prepared 100 mM NH4HCO3 buffer and shake for 5 min. Discard the supernatant. 3. Add 150 ml of 50% (v/v) acetonitrile, wash again for 5 min, and discard the supernatant.

3.5.3. Destaining

Perform all destaining and dehydrating steps using a thermomixer at 37°C and 550 rpm. 1. Destain two times (or until the colour disappears) with 50 ml of 50% (v/v) acetonitrile for 15 min. Discard the supernatant (see Note 13).

506

Schicher, Kollroser, and Hermetter

2. Dehydrate the gel pieces using 50 ml of 100% acetonitrile for 15 min. Carefully remove the supernatant and discard it. 3. Dry the gel particles in the vacuum centrifuge at medium temperature settings (45°C) for 15 min. 4. The dried gel pieces can be stored at –20°C. 3.5.4. Reduction and Alkylation

Perform all steps using a thermomixer at 37°C and 550 rpm, unless otherwise stated (see Note 14). 1. Rehydrate the dried gel pieces in 60 ml of reduction buffer at 56°C for 30 min. Discard the buffer. 2. Dehydrate the gel particles with 50 ml of 100% acetonitrile for 5 min. Carefully remove the supernatant and discard it. 3. Add 60 ml of alkylation buffer and incubate for 20 min under protection from daylight. Discard the supernatant. 4. Equilibrate the gel pieces with 150 ml of 100 mM NH4HCO3 for 5 min. Remove the supernatant. 5. Dehydrate samples by adding 50 ml of 100% acetonitrile and shaking for 15 min. Carefully remove the supernatant. 6. Dry the gel particles in the vacuum centrifuge for 15 min at medium temperature settings (45°C). 7. The dried gel pieces can be stored at –20°C.

3.5.5. Tryptic Digestion

1. Put the Eppendorf tubes with the dried gel pieces on ice. 2. Freshly prepare digestion and incubation buffers. 3. Re-swell the gel pieces by addition of digestion buffer. Begin with 3–4 ml, wait for 10–20 min, and then add small amounts (1–2 ml) so that the gel fully rehydrates without leaving a huge excess of trypsin solution. Wait for 10–20 min after each addition of digestion buffer (see Note 15). 4. Cover the samples with a small volume (~ 20 ml) of incubation buffer to keep the gel pieces wet during the digestion. 5. Incubate at 37°C and 550 rpm on the thermomixer over night (see Note 16).

3.5.6. Peptide Extraction

Perform all incubation and dehydration steps using a thermomixer at 37°C and 550 rpm. 1. Spin the tubes briefly in the centrifuge to recover the condensation water. Add 15 ml of extraction buffer, vortex briefly, and incubate for 15 min. 2. Transfer the supernatant to a new Eppendorf tube. 3. Dehydrate the gel particles with 150 ml of 100% acetonitrile for 15 min. Carefully remove the supernatant and combine it with the first one.



Adipose Tissue Using Fluorescent Suicide Inhibitors

507

4. Add 40 ml of 5% (v/v) formic acid to the gel pieces and incubate for 15 min. Remove the supernatant and combine it with the others. 5. Dehydrate the gel pieces once more with 150 ml of 100% acetonitrile for 15 min. Add the supernatant to the previous ones and discard the dried gel pieces. 6. Dry the combined supernatants in the vacuum centrifuge for 2 h at medium temperature settings (45°C). 7. The dried peptide extract can now be stored at –20°C. 3.6. LC–MS/MS Analysis

1. Dissolve the dried peptide extracts in approximately 80 ml of 0.1% (v/v) formic acid. 2. Inject 70 ml of the sample on the C18 pre-column. Peptides are concentrated and desalted with 0.1% formic acid at a flow rate of 20 ml/min for 8 min. 3. After the pre-concentration and clean-up step, switch the pre-column in line with the analytical column. Use a gradient from 0.3% formic acid and 5% acetonitrile to 0.3% formic acid and 50% acetonitrile over 60 min at a flow rate of 300 nl/min. 4. Peptides are ionized in the nano-ESI source and then analysed in the linear ion trap mass spectrometer. 5. Use SpectrumMill Rev. 03.03.078 software to analyse the MS/MS data. For protein identification search the National Center for Biotechnology Information (NCBI) public database.

3.7. Results 3.7.1. Lipolytic Proteome

3.7.2. LC–MS/MS Analysis

A detailed analysis of the protein patterns shown in Fig. 3 led to the identification of several lipolytic and esterolytic proteins including the enzymes responsible for full degradation of triacylglycerols in intracellular lipid droplets. These proteins are the recently discovered ATGL, HSL, and MGL. Other promising lipase candidates are currently being subject to further investigation in order to characterize their substrate preference and physiological role in fat cells. A typical example for mass spectroscopic identification of a distinct protein spot in the fluorescently labelled protein pattern of adipose tissue homogenate is shown in Fig. 4. The enzyme candidates with the highest scores were typical triglyceride hydrolysing proteins. Fatty acid amide hydrolase may also be considered as a lipase since it degrades long-chain fatty acid derivatives. The other components listed in this table are either unrelated proteins or have so far not been screened for their capacity to hydrolyse lipid esters or amides.

508

Schicher, Kollroser, and Hermetter

Fig. 3. The lipolytic proteome of brown adipose tissue (BAT). Murine BAT was incubated with phosphonic inhibitors followed by protein separation by 2D gel electrophoresis and imaging of the fluorescently labelled enzymes. Inhibitors: (a) NBD-HE-HP, (b) NBD-TG, (c) blank, autofluorescent enzymes, and (d) SYPRO Ruby, total protein stain.

A detailed description and discussion of the data of proteomic analysis shown in Figs. 3 and 4 is provided in (1).

4. Notes 1. Smaller Eppendorf tubes can be used instead of Falcon tubes depending on the amount of tissue pieces. 2. Before labelling with activity-based probes, protein samples should always be kept on ice in order to avoid loss of activity. 3. The Bio-Rad protein assay is based on the method of Bradford (14). This is a common laboratory procedure, and therefore is not described in this chapter.



Adipose Tissue Using Fluorescent Suicide Inhibitors

509

Fig. 4. Analysis of MS/MS data of tryptic peptides obtained from a protein spot in 2D gel electrophoresis. The two main lipid hydrolases identified in this spot were triacylglycerol hydrolase and carboxylesterase ML1.

4. Fluorophores are very sensitive to light, so never expose fluorescent samples, e.g. during incubations, to daylight unless necessary. 5. The final concentrations of detergent and activity tag in the protein sample are 1 mM Triton X-100 and 20 mM inhibitor, respectively. 6. Heat the samples for 1D gel electrophoresis to 95°C for 3 min and then incubate under shaking at 37°C for at least 30 min, to make sure the pellet is fully resuspended. Protein samples for 2D electrophoresis are incubated under shaking at 37°C for 1–2 h, or until the pellet is fully resuspended. 7. After storing the samples at –20°C, make sure to rehomogenize them by shaking at 37°C for at least 30 min before they are used for gel electrophoresis. 8. Alternatively, pre-cast gels can be purchased from vendors. In this case the protocol can be started following step 3. 9. APS and TEMED should be added immediately before pouring the gels, because these compounds start polymerization of the gels. 10. Rehydration should be performed for at least 10 h, or better over night. 11. Choose the volume of equilibration buffer such that the strip is entirely covered by buffer when the tube is fixed horizontally.

510

Schicher, Kollroser, and Hermetter

12. Alternatively, pre-cast gels can be purchased from vendors. In this case the protocol can be continued at step 7. 13. Step 1 is only required for gels that are stained with SYPRO Ruby. If gels have not been stained, continue with step 2. 14. Reduction and alkylation steps only need to be performed for 1D gels. The samples for 2D gel electrophoresis are already reduced and alkylated during steps 4 and 5 in Subheading 3.3.2. In this case you can proceed to Subheading 3.5.5. 15. It is important that always small amounts of digestion buffer are added to the gel pieces which are then left on ice for at least 10 min to take up as much trypsin as possible. If enough trypsin solution was added, the gel pieces look like melting ice cubes. 16. The trypsin-containing samples should be incubated for at least 12 h, or better over night.

Acknowledgements This work was supported by the GOLD (Genomics of Lipidassociated Disorders) project in the framework of GEN-AU (GENome Research in AUstria; http://www.gen-au.at/), funded by the Austrian Federal Ministry of Science and Research.

References 1. Birner-Gruenberger R, Susani-Etzerodt H, Waldhuber M et al. (2005) The lipolytic proteome of mouse adipose tissue. Mol Cell Proteomics 4(11):1710–1717. 2. Fredrikson G, Tornqvist H, Belfrage P. (1986) Hormone-sensitive lipase and monoacylglycerol lipase are both required for complete degradation of adipocyte triacylglycerol. Biochim Biophys Acta 876(2):288–293. 3. Zimmermann R, Strauss JG, Haemmerle G et al. (2004) Fat mobilization in adipose tissue is promoted by adipose triglyceride lipase. Science 306(5700):1383–1386. 4. Ollis DL, Cheah E, Cygler M et  al. (1992) The alpha/beta hydrolase fold. Protein Eng 5(3):197–211. 5. Brockman HL. General features of lipolysis: reaction scheme, interfacial structure and experimental approaches. In: Borgstrom B, Brockman HL, editors. Lipases. Amsterdam: Elsevier; 1984 p. 3–46.

6. Duque M, Graupner M, Stutz H et al. (1996) New fluorogenic triacylglycerol analogs as substrates for the determination and chiral discrimination of lipase activities. J Lipid Res 37(4):868–876. 7. Henderson AD, Richmond W, Elkeles RS. (1993) Hepatic and lipoprotein lipases selectively assayed in postheparin plasma. Clin Chem 39:218–223. 8. Hendrickson HS. (1994) Fluorescence-based assays of lipases, phospholipases, and other lipolytic enzymes. Anal Biochem 219:1–8. 9. Drahl C, Cravatt BF, Sorensen EJ. (2005) Protein-reactive natural products. Angew Chem Int Ed Engl 44(36):5788–5809. 10. Hemelaar J, Galardy PJ, Borodovsky A, Kessler BM, Ploegh HL, Ovaa H. (2004) Chemistrybased functional proteomics: mechanismbased activity-profiling tools for ubiquitin and ubiquitin-like specific proteases. J Proteome Res 3(2):268–276.



Adipose Tissue Using Fluorescent Suicide Inhibitors

11. Pohl NL. (2005) Functional proteomics for the discovery of carbohydrate-related enzyme activities. Curr Opin Chem Biol 9(1):76–81. 12. Speers AE, Cravatt BF. (2004) Chemical strategies for activity-based proteomics. ChemBioChem 5:41–47. 13. Birner-Gruenberger R, Susani-Etzerodt H, Kollroser M, Rechberger G, Hermetter A.

511

(2008) Activity-based proteome mapping of lipases and esterases in murine liver. Proteomics 8:3645–3656. 14. Bradford MM. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254.

sdfsdf

Chapter 26 Identifying the Spatial Distribution of Vitamin E, Pulmonary Surfactant and Membrane Lipids in Cells and Tissue by Confocal Raman Microscopy J. Renwick Beattie and Bettina C. Schock Summary Every organ compromises of several different cell types. When studying the effects of a chosen compound within this organ or tissue uptake, localisation, metabolism, and the effect itself can be expected to differ between cells. Using the example of Vitamin E in pulmonary tissue we introduce confocal Raman Microscopy as a superior method to localise lipid-soluble compounds within tissues and cells. We describe the analyses of vitamin E, its oxidation products, and metabolites as well as pulmonary surfactant phospholipids in fixed lung tissue sections. Examples of main structural membrane lipids (PC, cholesterol) and an example of a lipid-signalling molecule (ceramide) are also included. Confocal Raman microscopy is a non-destructive optical method of analysing chemical and physical composition of solids, liquids, gases, gels, and solutions. The method is rich in information allowing discrimination of chemically similar molecules (including geometric isomers) and sensitive monitoring of subtle physical interactions. Additionally, Raman spectroscopy is relatively insensitive to water allowing the analysis of aqueous solutions and suspensions typical in biochemistry. In contrast, Raman spectroscopy is sensitive to non-polar molecules making it ideal for lipidomics research. Key words:  Raman Microscopy, Cell Imaging, Membrane Lipids, Vitamin E, Surfactant lipids, Subcellular lipid distribution, Lipid metabolites, Lipid signalling

1. Introduction 1.1. Localisation and Distribution of Molecules Within Tissues and Cells

Every organ compromises of different cell types with specialised function. The lung, for example, has over 30 different cell types and some of these cells, such as alveolar type-II cells are known to have a higher content/uptake of alpha-tocopherol (aT), a form of vitamin E, than, for instance, alveolar type-I cells or bronchial epithelial cells (1–3). Furthermore, tocopherols are differentially

Donald Armstrong (ed.), Lipidomics, Methods in Molecular Biology, vol. 579, doi 10.1007/978-1-60761-322-0_26, © Humana Press, a part of Springer Science + Business Media, LLC 2009

513

514

Beattie and Schock

metabolised depending on the cell type (4, 5). Differential uptake and metabolism of a compound are of particular interest when investigating its therapeutic potential. Using Raman microscopy we are able to localise different molecules in a tissue section and to determine cell-specific uptake of compounds, their localisation, and metabolism. Identifying the spatial distribution and sub-cellular localisation and metabolism provide important information additionally to the overall (e.g. anti-inflammatory, anti-proliferative, or pro-apoptotic) effect. For instance, the phospholipid structure (e.g. saturated or unsaturated PC) can have a significant impact on the drug–cell interaction as shown for nystatin, an antifungal drug (6). Conventionally, the effects of nutrients, supplements, or drugs are tested by analyses of the compound within a whole tissue, but its spatial distribution is generally not investigated. In the case of tocopherols, uptake and metabolism can be determined using specially labelled compounds (e.g. deuterated tocopherols) (7, 8). For instance, when rats with experimental acute lung injury inhaled aT, a reduction of some major pro-inflammatory cytokines (proteins and mRNA) was observed (9) confirming the overall immunomodulatory effect of aT. However, the cells were mostly affected by aT have not been investigated. Similarly, aerosolised aT in an animal burn and smoke inhalation injury model improved the pulmonary gas exchange (10), an effect that has also been observed using nebulised albuterol or anticoagulants (11, 12). Aerosol delivery is a specific and fast way to administer agents to the lungs and into the pulmonary circulation. In any case, knowledge about the cell-specific uptake, metabolism, and potential interference with co-localised compounds is of great importance to exclude adverse effects of any therapeutic agent. This will be of particular significance in the light of new developments into personalised drug therapy. In this chapter we will introduce confocal Raman microscopy to localise lipid-soluble compounds within tissue, especially lung cells and tissue. We shall be using the example of measuring vitamin E, its oxidation Products, and metabolites as well as pulmonary surfactant phospholipids in fixed lung tissue sections. Examples of main structural membrane lipids (PC, cholesterol) and an example of a lipid-signalling molecule (ceramide) are also included. 1.2. Membrane Lipids

Lipids in eukaryotic cells have several very different functions: storage of energy, formation of barriers (cellular membranes), active transport of molecules and ions, and signal transduction messengers (13). The major lipids in eukaryotic cell membranes are glycerophospholipids, sphingolipids, and cholesterol, with phosphatidyl-choline (PC) accounting for over 50% of the

Identifying the Spatial Distribution of Vitamin E

515

phospholipid fraction. The lipid composition and saturation of the acyl chain of the membrane as well as the cholesterol content determines its fluidity, which is important for membrane protein function (eg. recepters) and may also affect the membranes interaction with drugs (6). Phospholipids that occur in smaller quantities in the membrane are as important. For example, phosphatidyl-ethanolamine (PE) has a smaller sized polar head group, which when PE is included into the PC bilayers causes a curvature stress onto the membrane. This is important for budding, fission, and fusion (14), as well as major components of vital cell functions such as apoptosis, proliferation, phagocytosis, and secretion. Bioactive lipids such as eicosanoids and sphingolipids are key signalling and regulatory molecules. Ceramide is considered the “central hub” of sphingolipid metabolism and mediates many cell stress responses including differentiation, cell senescence, inflammation, and apoptosis (15). For instance, ceramide accumulation is associated with upregulation of cyclooxygenase (COX)-2 and prostaglandin E2 (PGE2) in aged mice (16). Recently Teichgräber et al. reported that ceramide accumulation in murine cystic fibrosis (CF) cells contributes to the chronic pro-inflammatory state of CF epithelial cells (17). Indeed, using Raman microscopy we show a more varied distribution in human epithelial CF cells (Fig. 5). 1.3. Pulmonary Surfactant

Pulmonary surfactant is produced by alveolar type-II cells in the alveoli. It is a complex mixture of lipids and proteins that forms a surface-active film at the air–liquid interface of alveoli and is capable of reducing the surface tension to nearly 0 mN/m. In the upper airways it may function as a lubricant between the sol and gel phase of the airway mucus. Human pulmonary surfactant contains approximately 70% saturated and unsaturated PC, 10% phosphatidyl glycerol (PG), 10% surfactant-associated proteins, and 3% cholesterol. The remaining mass is made up by other phospholipids (18). Saturated PC at room temperature is more gel like, while the addition of unsaturated PC or cholesterol makes the surfactant film more liquid or fluid. Such changes in the surfactant composition can occur rapidly in response to environmental stresses and occur physiologically in heterothermic animals (19). During acute lung injury, cholesterol and protein concentrations increase in lung lavage fluid of rats (20) and in vitro studies have shown that increasing the cholesterol concentration in pulmonary surfactant above the physiological levels (e.g. to 20%) abolishes surfactant functions (20, 21).

1.4. Vitamin E

Alpha-tocopherol (5,7,8-trimethyltocol, aT) is the predominant form of vitamin E in mammals and while it has other biological roles, its importance originates from its radical scavenging activity:

516

Beattie and Schock

the chromanol ring reacts with lipid peroxyl radicals to form an oxidised a-tocopheroxyl free radical thereby inhibiting the propagation of those radicals and stabilising polyunsaturated fatty acids in lipoproteins and cell membranes and therefore reduces the release of arachidonic acid from the membrane and its metabolism to prostaglandins (22–24). In the lung, aT concentrations are tightly regulated through active uptake and secretion by alveolar type-II cells (2, 3) and aT is concentrated within the surfactant containing lamellar bodies (1). In the extracellular space it is present at the air–liquid interface as part of the surfactant phospholipid bilayer acting as a first line of defence against inhaled reactive oxygen species. Here it may protect surfactant lipids from peroxidation (25–27) and may reduce the oxidative stress onto the underlying cells. Using Raman microscopy and HLPC–EC analyses we found that an exogenous surfactant preparation (BLESTM) normally used in newborns suffering from respiratory distress syndrome contains 0.066 ± 0.019 and 0.059 ± 0.014 nmol aT/mg PL, respectively. Work by Atkinson and co-workers indicate that tocopherols may enhance curvature stress or even counteract similar stresses generated by other lipids (28). If so, tocopherols such as aT may play a critical role in vital cell functions where budding, fission, or fusion is involved and knowledge of its concentrations and distribution in the membrane would be crucial to confirm its biological impact. However, it is unknown if aT exhibits any additional therapeutically benefit in the surfactant preparations. Similarly, very little is known about the concentration, distribution, and function of aT in lipid rafts. Cushieri et  al. showed that aTsuccinate inhibits endotoxin-mediated transport of phosphates to lipid rafts, which provides an alternative explanation of the antiinflammatory action of aT (29). 1.5. Analyses of Lipids and Lipid-Soluble Compounds

Generally, identification of lipids and lipid-soluble membrane components (such as vitamin E) in biological samples involves the extraction of such with organic solvents (30, 31) and the addition of antioxidants (e.g. butylated hydroxytoluene or ascorbic acid) to prevent oxidation of the analyte during processing. Traditional lipids were analysed using thin-layer chromatography (TLC), were a mixture of extracted lipids separated by different adsorption to the solid phase due to their different solubility in the solvent. The same principle applies to any column chromatography and the latter, especially high-pressure liquid chromatography (HPLC), has now replaced most of the TLC analyses. However, there are many advantages of using TLC (which cannot be discussed here) and in skilled hands the methods is still very useful. Similarly to

Identifying the Spatial Distribution of Vitamin E

517

membrane lipids, pulmonary surfactant lipids can be analysed using these methods. Vitamin E is most commonly analyzed using HPLC with electrochemical (EC), fluorescence, or UV detection (32, 33) or liquid chromatography–mass spectroscopy (LC–MS) (34). HPLC–EC has a sensitivity in the nmol range (32), while massspectroscopic methods are more sensitive (LC–MS, pmol range (35)). However, the concentrations of metabolites, nitration, or oxidation products together with aT add valuable information about the redox balance in biological samples, but their determination requires separate isolation and analyses (35, 36). It is important to note that the determination of the concentration of aT or membrane lipids in specific cells or cell fractions (e.g. mitochondrial membranes) requires the application of stringent laborious isolation procedures (37, 38). This approach, however stringently and precisely it is performed, has still one major disadvantage when studying biological effects on cells and tissues; namely having to extract the lipid-soluble compounds in order to quantify them will exclude vital information about its spatial distribution from the investigator. While metabolites may be analysed separately, no information can be obtained about the co-localisation with other (lipid-derived) compounds or those involved in metabolism. Furthermore, separate extractions and analyses may be rather time consuming reducing the applicability to follow kinetics of uptake and metabolism, again especially with co-localisation of other lipids of interest. Other methods involve labelling of lipid membrane-associated proteins using e.g. fluorescence antibodies. However, using confocal laser scanning microscopy, this would only “indirectly” identify functional groupings such as lipid rafts (39, 40). Antibodies against aT and cholesterol (USBiological, Antibodies-online GmbH, Germany) (41) have been developed, but their reactivity with “free” aT or steroid is poor. Recently, time-of-flight secondary ion mass spectrometry (TOF-SIMS) was used to visualise a number of lipid compounds, including vitamin E (42). While TOF-SIMS achieved similar levels of spatial resolution to Raman Microscopy (around 1 mm), it, like all other methods mentioned above, destroys the sample. 1.6. Raman Spectroscopy

Raman spectroscopy analyses the vibrations of a molecule after excitation with monochromatic (laser) light. The amount of energy removed from the incident light is directly related to the chemical structure as well as any intra- or intermolecular interactions that may affect the freedom of this vibration. The light scattered from the sample is collected and separated into different energies giving a Raman spectrum, which is rich in information on both the chemical and physical properties of the sample. Con-

Beattie and Schock

a

RamanIntensity / Arbitrary

518

b

c

d Pulmonary surfactant

500

1000

1500 2000 RamanShift / cm-1

2500

3000

Fig. 1. Raman spectra (633 nm excitation) and chemical structures of some main exemplary structural membrane lipids: (a) DPPC, (b) Ceramide (C2-ceramide) as ana example of a lipid signalling molecule, (c) Cholesterol, and (d) Pulmonary surfactant (BLESTM, lung surfactant extract).

sequently Raman spectroscopic methods have been extensively applied to the study of a wide range of lipid systems including fatty acid analysis, essential oils, and hydrophobic vitamins (1, 43–46). Figure 1 shows the Raman spectra of selected lipid compounds that play a major role in lipid membranes and physiological surfactants. The discrimination power of Raman spectroscopy is briefly illustrated in Fig. 2, which shows the Raman signals obtained from neat samples of four tocopherolic compounds. Indicated by arrows are a number of changes in the Raman spectrum that are a direct consequence of the changes in molecular structure. For example, gT is identical to aT except that it lacks a methyl group at the top of the aromatic ring (position 5), and this difference results in the disappearance of the aT mode at 485 cm–1. The 2 e– oxidation product, aTQ differs in that the chromanol ring has been destroyed and replaced by a quinine group, and this is

Identifying the Spatial Distribution of Vitamin E

519

a CH3 HO H3C

Raman Intensity / Arbitrary

b

CH3 O CH3

CH3 O

H3C CH3

O

CH3

d

CH3 CH3

CH3

CH3

OH

H3C H3C

CH3 CH3

HO

c

CH3

CH3

CH3

CH3 CH3

O

CH3 CH3

HO H3C

O CH3

CH3

COOH

e

500

750

1000

1250 1500 Raman Shift / cm-1

1750

2750

3000

Fig. 2. Raman spectra (633 nm excitation) and chemical structures of chromanol molecules (a) a-tocopherol (b) g-tocopherol (c) a-tocopherol quinone (d) a-3¢carboxyethylhydroxychromanol (a-CEHC) and (e) a-tocopherol in cell homogenate. Significant changes in the Raman signal, of a-tocopherol quinone, compared to a-tocopherol, are indicated by arrows, and these changes can be directly related to the change in the chemical structure.

reflected in the Raman signal by the loss of the doublet at 1,585 and 1,615 cm–1 and the appearance of a much stronger doublet at 1,633 and 1,655 cm–1. The 1 e– oxidation product of aT has had the phytyl chain removed and replaced by an acid group, resulting in the disappearance of the strong doublet at 2,845 and 2,865 cm–1, which is a characteristic doublet of extended branched aliphatic chains. Table 1 summarises the typical Raman shifts for a range of common lipid molecules (1, 47–51). In addition to the high information content, Raman spectroscopic methods offer many further benefits. For example, it is non-destructive and water is a very weak Raman scatterer allowing analysis in aqueous environments. Additionally the optical nature of the Raman methodology allows it to be combined with a very wide range of standard microscopic procedures and therefore allows the method to be incorporated very non-invasively into an existing experimental setup. All that is required is that the

520

Beattie and Schock

Table 1  Some Raman band assignments for common moieties in lipids (1, 47–51) Band position

Assignment

Notes

3,100–3,600

OH

Sensitive to hydrogen bonding strength

3,060

CH stretch

Sensitive to conjugation and geometric isomers

3,030

CH3 stretch

In choline headgroup

2,960

CH3 asymmetric stretch

2,930

CH2 asymmetric stretch Ratio of intensity to mode at 2,850 cm–1 is commonly used as a marker of polymethylene chain conformation

2,870

CH3 symmetric stretch

2,850

CH2 symmetric stretch

1,710–1,750

COOR

Carbonyl stretch in ester, used as molar reference band in fatty esters

1,550–1,700

COO–, C = O

Position and width dependant on strength of hydrogen bonding

1,670

Trans isolated C = C

Correlation between intensity and number of trans C = C in molecule

1,660

Cis isolated C = C

Correlation between intensity and number of cis C = C in molecule

1,630–55

Quinone, conjugated dienes

1,607–30

Aromatic, conjugated trienes

1,582–7

Aromatic ring

1,515–1600

Conjugated C = C,

Length correlated with Raman shift

1,410–1490

CH2

Decreases with increasing hydrophobicity. Sensitive to intermolecular forces, i.e. crystallinity. Correlates with polymethylene chain length.

1,400–1420

COO–

Sensitive to hydrogen bonding strength

1,380

CH3 twist

1,300

CH2 twist

Position sensitive to intramolecular forces, i.e. chain conformation. Intensity correlates with chain length

1,250–12,70

=C–H in cis

Band gives very weak intensity for trans isomers.

1,260

PO2– asymmetric stretch

1,020–1,130

C–C, N, O stretch.

Sensitive to intramolecular forces. Characteristic pattern of three bands within linear polymethylene chains. Intensity correlates with polymethylene chain length. (continued)

Identifying the Spatial Distribution of Vitamin E

521

Table 1 (continued) Band position

Assignment

Notes

1,090

PO2– symmetric stretch

880–965

C–C

Terminal HxC–CH3 and C–C adjacent to C = O

875

C–N stretch

Choline group, trans conformation

761

OPO symmetric stretch

Diphospho-ester

720/770

C–N totally symmetric stretch

Sensitive to conformation in the choline headgroup, high Raman shift for trans, low for cis

<200

Whole molecule

Related to whole molecule motion, translation, rotation, breathing, etc.

Table 2  Advantages and disadvantages of Raman spectroscopic techniques Advantages

Disadvantages

Minimal interference from water

Weak effect

Non-contact, non-invasive.

Expensive

All physical states of matter

Can be time-consuming

Minimal sample preparation

Hi-tech

Non-destructive

Fluorescence can interfere

Spatial resolution diffraction limited

Processing tools underdeveloped

Measure < 1 pg material Rich chemical and physical information content Instrumentation highly adaptable Special techniques extend: Sensitivity (e.g. Surface Enhanced RS), Specificity (e.g. Resonance RS), Spatial resolution (e.g. Tip Enhanced RS)

user is able to direct a laser onto the sample and collect the light scattered from it. Table 2 summarises the key advantages and disadvantages of Raman spectroscopic methods. These advantages have allowed Raman spectroscopy to be used to investigate a range of tissues both in vitro and in vivo (52–54).

522

Beattie and Schock

Combining Raman spectroscopy with optical microscopy offers these advantages on a micron-scale of spatial resolution, without lengthy sample processing and/or the addition of (bio) chemicals. In order to explore other lipid constituents or lipid bilayers it is not necessary to make changes to the Raman methodology, but calibration with a suitable approach for that constituent and recording its reference Raman spectra under the same conditions is required. It should be noted that sample preparation is not dictated by the needs of the Raman spectrometer, but rather by the conditions under which the sample is in a relevant state. Furthermore, as Raman microscopy can differentiate the main constituents of membrane lipids (sphingolipids, ceramide, cholesterol, PC), the method may also be applied to visualise microdomains such as cholesterol and sphingolipid-enriched lipid rafts. Using water immersion objectives will enable the visualisation of the organisation and interaction with functional proteins within such domains, providing a novel method to study the role of lipid rafts in e.g. endocytosis and membrane signalling (55). 1.7. Multivariate Statistics

Because of the complexity of the information measured by Raman spectroscopic methods, it is commonplace to employ multivariate statistics in order to simplify the data and home in on only the information that is relevant. Qualitative analysis is most commonly carried out using principal component analysis (PCA), which identifies the limited range of basic spectral signals that, in various combinations, account for all the signals within the dataset. The results of most interest from PCA are the basic spectral signals (loadings) and the scores, which reflect the relative contribution of these basic spectral signals to each sample spectrum. Partial least squares (PLS) regression is the most commonly applied quantitative analysis, which correlates the sample spectrum with a reference parameter. The regression coefficients are the spectral bands that were used to correlate with the reference value, and are an excellent way of checking the correlation based on direct spectral measurement of the target analyte. An excellent introduction to multivariate statistics is available from Umetrics AB (Umea, Sweden) (56, 57).

2. Materials 2.1. Equipment

1. Horiba Jobin-Yvon LabRam HR800 Raman confocal microscope. (see Note 1) 2. A 633-nm excitation laser (10 mW at sample, focused beam diameter <1 mm, see Note 2).

Identifying the Spatial Distribution of Vitamin E

523

3. A “window” (range of the spectrum detectable in one acquisition) of 800–3,100 cm–1 see Note 3 4. A 300 groove mm−1 diffraction grating (giving 12 cm−1 spectral resolution). 5. Microscope objectives (see Note 4, magnification 10×, 50×, and 100×; e.g. Olympus M-plan NA0.9, confocal hole 600 mm). 6. A 0.1-mm step size motorised, computer-controlled xyz stage. 7. Labspec™ (Jobin-Yvon, Villeneuve d’Ascq, France) to record and process the optical and spectral images. 8. Software capable of performing multivariate analyses (PCA, PLS), such as “The Unscrambler” (Camo, Oslo, Norway). 9. A more detailed and technical account of Raman instrumentation is available (58). 2.2. Reagents and Supplies

Pure reference samples for any lipid species to be targeted should be used. 1. dl-alpha-Tocopherol and Cholesterol (SIGMA, UK). 2. Palmitic acid methyl ester (PAME) (SIGMA, UK). 3. Carboxyethylhydroxychromanol, a-CEHC. 4. C2-Ceramide (BIOMOL International LP, UK). 5. 1,2-Dipalmitoyl-sn-glycero-3-phosphocholine, saturated DPPC (Avanti Polar Lipids (no 850355), USA). 6. Bovine lung extract surfactant, BLES™ (BLES Biochemical Inc. Canada). 7. CaF2 microscope slides (Crystrans, Poole, UK, see Note 5). 8. N2 (or other inert gas) to flush the sample to prevent oxidation during signal aquisition. 9. Standard methods were applied for cell culture and homogenising of tissue (1).

3. Method 3.1. Calibrating the Raman Signal ( See Note 6)

1. First obtain pure reference material for the substances of interest and a sample containing the reference material in the relevant matrix. Record their Raman spectrum; the exact procedure will be dependant on the instrumentation employed. 2. Because Raman spectroscopy is sensitive to physical interactions it is important to measure the reference materials under a range of conditions and within a relevant matrix (see Note 6). As an example, Fig. 2a shows the Raman spectra obtained

524

Beattie and Schock

from the pure reference material (aT) as well as the reference material in cell homogenate (Fig. 2e). Because Raman spectroscopy is based on UV, visible, or NIR radiation it is possible to measure a sample through any substance transparent to the excitation wavelength. This has the consequence that Raman experiments can easily be performed using chambers or incubators that control local environment. This opens the possibility of probing chemical and physical changes occurring in response to temperature, humidity, controlled atmospheres, and pressure. 3.2. Validation of the Spectroscopic Analyses

1. Because we are calibrating a spectroscopic method rather than directly measuring a parameter it would be ideal to prepare a sufficiently large set of samples to allow for a separate calibration and validation set (2:1 ratio of samples). If it is unrealistic to prepare sufficient samples for two sets, then cross validation can be used on a single set of data, though this may have a tendency to be slightly optimistic about model performance. The calibration should be performed over a realistic range of concentrations reflecting the range encountered in  vivo (remembering, if the target analyte is likely to be localised, that its local concentration may be greater than the average concentration within the whole tissue). 2. To calibrate the analyses for both the Raman signal within biological matrix and quantification of aT, we supplemented lung epithelial cells in culture with increasing concentrations of aT (0–50 mM, dissolved in ethanol) and determined the aT concentration of a cell homogenate by HPLC-ECD (1). The homogenate should be in the same state as the tissue to be mapped (e.g. wet or dehydrated). A portion of the homogenate (10 ml) was also placed into a 0.8-mm diameter well in CaF2 or aluminium and dried under N2. 3. Record Raman spectra from a representative grid over the homogenate and then average the spectra for each sample (see Note 7). The signal-to-noise ratio (S/N) of the band around 2,930 cm–1 from the average spectrum of each sample should be at least 50 for the calibration. Remove the background signal from the Raman spectra (see Note 8), normalise the spectra by dividing each point by the average intensity of the spectrum (see Note 9) then input the data into the statistical package, along with HPLC-measured parameters. 4. Carry out a PLS regression, incorporating wavelength selection to create a more robust model. Compare the regression coefficients with the spectra of the pure reference compounds as shown in Fig. 3 (see Note 10).

Identifying the Spatial Distribution of Vitamin E

525

Raman Intensity / Arbitrary

a

b

c

1000

1400

1800 2200 Raman Shift / cm-1

2600

3000

Fig. 3. Interpreting the regression coefficients used in a PLS regression of the Raman signal against HPLC-determined a-tocopherol concentration in A549 cells supplemented with a range of tocopherol concentrations. (a) Raman signal of pure a-tocopherol (b) regression coefficients used to transform the Raman signal into the predicted a-tocopherol concen­ tration (c) the Raman signal of an unsupplemented A549 cells multiplied by −1 for comparison with negative coefficients.

3.3. Acquiring the Raman Data of the Sample

Select a suitable area of the sample. If the target analyte is highly localised it is possible to use a low-spatial resolution (5–10 mm spacing) scan over a large area; follow the remaining procedure and then repeat in higher resolution in an area identified from the initial screen as containing the target analyte. The individual spectra used to quantitatively predict the target analyte should have an S/N of at least 10.

3.4. Qualitative Mapping

1. It is not recommended to normalise the spectral intensity of mapped data before qualitative analysis. Import the data into the statistics package and perform a PCA. Identify principal components (PCs) relating to target analytes by comparing the loadings with the reference spectra recorded initially. These loadings contain a positive and a negative constituent, which can make interpretation difficult. In order to visualise the

Beattie and Schock

spectra related to the component it is necessary to average the most extreme scores and their corresponding spectra for each loading. 2. For the highest rank PCs these spectra will match the signals of the constituents contributing to the PC, but for lower ranked PCs these signals will not necessarily be the dominant band shape within the spectrum. Add the loadings onto the extreme high score spectrum and subtract the loadings from the extreme low score spectra until just before negative bands start to appear in the spectrum. The process is illustrated in Fig. 4 and it is clear that the spectral constituents contributing to the PC are easier to identify, allowing the identification of a-tocopherol and a-tocopherol quinone signals in nitrogenflushed and unflushed tissues, respectively. The scores of the PCs can be mapped using their original coordinates in order to obtain a pseudo-image like those shown in Fig. 5a and 6 e–j.

a

PC weighting/ arbitrary

526

b

c

d

1000

1400

1800

2200

Raman Shift / cm-1

2600

3000

Fig. 4. Extracting spectral signals from PCA. Spectral loadings containing tocopherolic signals from (a) N2flushed lung section and (b) unflushed lung section. The negative contributor to the loadings makes it difficult to confidently assign the tocopherol. By adding the loadings to the average high score spectra the tocopherol signals contributing to the high score samples (c + d) are clear. The position of the arrowed band in (c) matches with that of the aromatic mode in a-tocopherol, while that of (d) matches a-tocopherol quinine, indicating that without a N2 flush, oxidation of the a-tocopherol can occur.

Identifying the Spatial Distribution of Vitamin E

HBE

a

527

CFBE

Light microscopic image (x100)

High

Raman Intensity / Arbitrary

Low

PC8 score (Ceramide) / Arbitrary

b PC8 score (ceramide) / Arbitrary

High

Low HBE

CFHBE

Fig. 5. Raman mapping of human bronchial epithelial cell lines with and without a CFTR mutation (HBE and CFBE, respectively) and PCA analyses for the lipid signalling molecule ceramide. (a) The light microscopic image, Raman map (Intensity) and Raman map after processing for PC8 scores (ceramide). (b) A summary plot of the PC8 scores for ceramide.

528

Beattie and Schock

x50

a

x40

b

c

10 µm

5µm x100

d

e

5 µm

g

f

Prot/DNA

h

αT, γT

j

i

αTQ, Heme

k

Surfactant

Heme

αT, α-CEHC

l

αT/PA

αT/Prot

Fig. 6. Raman maps for a number of biochemical constituents identified in mouse lung during PCA. (a) Bright-field image of mouse lung at ×50 magnification (10-mm section after fixation with 4% PFA and short-term storage in 5% sucrose). (b) Histological appearance of the same mouse lung section after haematoxylin/eosin (HE) staining. (c) Raman map with 2-mm spacing of the same mouse lung section (×100) showing nuclear protein (green), heme (red), and a-tocopherol (blue). (d) Insert shown in (b) at ×100 magnification after HE staining. (e–l) Raman maps of the insert shown in (a) at 0.8-mm spatial resolution: (e) protein/DNA (green); (f) heme (red); (g) a-tocopherol (cyan) and g-tocopherol (magenta); (h) a-tocopherol quinone (yellow-green due to overlap with red) and heme (red); (i) a-tocopherol (cyan) and a-CEHC (magenta); (j) pulmonary surfactant (yellow); (k) a-tocopherol/palmitic acid (turquoise); (l) a-tocopherol/protein (green). Figure 6a–d and g–j reprinted with permission from (1).

Identifying the Spatial Distribution of Vitamin E

529

3.5. Quantitative Mapping

Process the data in the same way as the calibration data was processed but exclude any signals wka signal-to-noise ratio below 10 as these are below detail in limit. Import the data into the statistics package and perform a prediction using the PLS regression model created during the calibration step. Use the original mapping coordinates to reconstruct a pseudo-colour image of the predicted analyte as shown in Fig. 6k and l.

3.6. Interpretation of Results

1. In order to interpret the Raman spectra from the sample under study, it is essential to have a comprehensive database of representative spectra available for comparison. Use of spectral database software will compare the sample spectrum with the reference spectra and report which reference spectra are most similar to the sample. In some cases it may not be possible to obtain a pure reference spectrum of the target analyte, in which case a spectral database will not be applicable. Alternatively, the sample spectrum can be compared with published spectra, tables of assignment and comprehensive vibrational analyses of related molecules. Despite its age, Tu’s book on Raman spectroscopy in biology remains a very comprehensive reference for typical spectral signals of a wide range of biochemicals (59). 2. As mentioned above, it is imperative that physical interactions and states are replicated accurately between the sample for the reference spectrum and the form that is found naturally within the sample. In Fig. 2 spectra a and e compare aT in the neat liquid and in association with cells, and it is clear that the physical interactions within the cells cause measurable changes in the Raman spectrum. Understanding the relationship between physical interactions and the Raman signal of lipids allows careful probing of lipid structure in membranes as well as chemical composition (60). 3. Because of the close link between the Raman spectrum and the chemical structure of the analyte it is possible to speculate on the origin of an unidentified signal. If a reference sample of the suspected molecule is obtained then it can be accounted for in the original data without the need to re-record the sample data (61). 4. Raman spectroscopic methods are a point of focus method and the intensity of light scattered from the sample is affected by a vast range of interferences (58). It is therefore important to understand the principal source of signal variation in the data. When mapping a small area of a sample relatively quickly as in (1), the principal source of intensity variation is likely to be due to the absolute amount of molecules in the sampling volume – but will also be sensitive to the morphology of the sample. In Fig. 5 we show the absolute Raman signal of ceramides in human bronchial epithelial cells using

530

Beattie and Schock

the widely used cell lines HBE and CFBE (D. Gruenert, San Francisco, USA). Raman microscopy shows a more varied distribution in the CF cell line, which may suggest that these cells are in a higher activation state as increased incorporation and movement of ceramides within lipid rafts is required for bacterial internalisation, induction of apoptosis, and controlled release of cytokines and membrane receptor activation. In Fig. 6 the maps c–j are constructed from the PCA scores of non-standardised spectra, and consequently they reflect the absolute number of each molecule with the sampling volume at each point. Thus the concentrations of most materials appear higher in the centre of cells due to the fact there is more material there than at the edge. However, it is rarely the case that the absolute number of moles is of interest, but rather the concentration of the analyte with respect to its matrix or to another constituent. For quantitative analysis by Raman spectroscopy it is imperative that the data is appropriately normalised (62). Figure 6 maps k and l show the distribution of aT relative to protein and surfactant, respectively. It is clear that by ratioing the analyte signal to another constituent, a completely different type of information is gained. Comparison of Fig. 6g (aT in cyan) with j and k shows that although the largest total amount of aT molecules is in the centre of the cell, when expressed relative to the surfactant the concentration appears to be higher in the lining of the alveolus, the site of highest oxidative stress.

4. Notes 1. For larger samples (e.g. 100 ml of extracted material) a macro Raman instrument is preferable, with spot sizes available from 50 to 500 mm in diameter. For discussion on the relative merits of micro- and macro-Raman instrumentation, see (60). 2. For fluorescent samples use of 785-nm excitation is typically recommended. For non-fluorescent samples 633 nm offers the benefit of enhanced Raman scattering efficiency, optical transmission, and detection efficiency as well as allowing use of extra-white glass as a substrate if cost is an important consideration. Accurate measurement of beam diameter is essential for determining spatial resolution. Measure the Raman spectrum along a line projecting over the vertical edge of a silicon wafer and measure the distance between 95 and 5% signal intensity – this distance represents the 2-standard deviation diameter of the beam intensity. 3. The size of the spectral window is inversely related to the spectral resolution of the instrument.

Identifying the Spatial Distribution of Vitamin E

531

4. For tissue mapping or low volume extracts a confocal Raman microscope is required, allowing measurement from volumes of the order of 1 mm3, <1 pg of material (see Note 2). When using a high magnification objective, a water immersion lens is required for use on hydrated samples to avoid refraction problems at the sample surface. Raman methods can be easily incorporated into a plethora of configurations, allowing the sample to be probed under almost any condition imaginable. Use of water immersion objectives or an inverted configuration immediately opens the possibility of probing living tissue and cells. Phase contrast Normaski and other contrast enhancing methods can be incorporated to help the user visualise unstained biological materials, making the selection of analysis area easier. 5. The main issues associated with choosing a suitable substrate for the sample are minimal spectral features and as low a broadband signal as possible. CaF2 slides give a very low background, with a single sharp peak around 325 cm–1, making them an ideal substrate for many applications, but are expensive. If high throughput demands a cheaper alternative extra white slides (Menzel-Gläser, Braunschweig, Germany) give a much higher background than CaF2, but when compared with quartz these have smaller spectral features but comparable intensity of broadband background. Standard glass gives broadband signal intensity around three times that of extra white glass. However, if the excitation wavelength is in the NIR (>750 nm) then glass (standard or extra white) gives a very strong emission that swamps the Raman signal so that CaF2 or quartz must be employed. 6. It is important to design the calibration to measure the analyte in relevant circumstances. Supplementation studies ensure biological relevance, but the cell lines used must have biochemical properties similar to the situation in situ. For example, two calibrations allow prediction of aT/protein and aT/ surfactant ratios. Supplemented (alveolar type II-cells A549) created a prediction of aT/protein ratio. A separate calibration of standardised mixtures of aT and palmitic acid methyl ester were used to create a prediction of aT/surfactant ratio. In this later situation the dominance of palmitic acid in lung surfactant allowed modelling using a simpler and cheaper system, but using pulmonary surfactant instead of palmitic acid would also be valid. 7. It is most useful to record multiple (minimum ten for confocal microscope systems) spectra from different regions of the sample to check for consistency. PCA will highlight any deviations such as impurities, polymorphism, or domain formation. Using this approach any detectable impurities can be eliminated from the signal post acquisition, leaving the pure signal of the target analyte.

532

Beattie and Schock

8. The raw data acquired from a Raman experiment is not a pure Raman spectrum, but contains contributions from optical ­phenomena other than Raman, such as fluorescence. For this reason it is necessary to eliminate this low information broadband signal from the data set prior to quantitative analysis. A large number of methods exist, with excellent review available in (63) and some promising new automated methods (61, 64). 9. The absolute Raman intensity is very difficult to measure reproducibly, being dependant on a very large number of factors (58, 62) and for this reason it is necessary to standardise the intensity prior to quantitative analysis. 10. The positive bands in the regression coefficients should match distinctive bands in the target analyte, while the negative bands should match distinctive bands in the cells or tissue under study. If the bands do not match those in the reference material, the model is depending on a cross correlation that may not apply in independent samples. Wavelength selection ignores Raman shifts that are unreliable for predicting the target analyte, either because the target analyte shares intensity with the cell or because there is a third constituent unrelated to the analyte.

Acknowledgement Research was supported by a grant from the BBSRC, UK (JREI 18471) and the R&D Office, Northern Ireland (SPI/2384/03). JRB is supported by a grant from the Medical Research Council (MRC), UK (G0600053). The authors thank Mr. C. Maguire for technical support. BLES™ was a gift from BLES Biochemicals Inc., Canada (H. Dick) and the tocopherol metabolite was a gift from F. Galli and F. Mazzini (University of Perugia, Italy). The authors are very grateful for these contributions.

References 1. Beattie, J.R., Maguire, C., Gilchrist, S., Barrett, L.J., Cross, C.E., Possmayer, F., Ennis, M., Elborn, J.S., Curry, W.J., McGarvey, J.J., and Schock, B.C.,(2007) The use of Raman microscopy to determine and localize vitamin E in biological samples. FASEB J. 21, p. 766–76. 2. Kolleck, I., Schlame, M., Fechner, H., Looman, A.C., Wissel, H., and Rustow, B.,(1999) HDL is the major source of vitamin

E for type II pneumocytes. Free Radic Biol Med. 27, p. 882–90. 3. Hosomi, A., Goto, K., Kondo, H., Iwatsubo, T., Yokota, T., Ogawa, M., Arita, M., Aoki, J., Arai, H., and Inoue, K.,(1998) Localization of alpha-tocopherol transfer protein in rat brain. Neurosci Lett. 256, p. 159–62. 4. You, C.S., Sontag, T.J., Swanson, J.E., and Parker, R.S.,(2005) Long-chain carboxychromanols

Identifying the Spatial Distribution of Vitamin E are the major metabolites of tocopherols and tocotrienols in A549 lung epithelial cells but not HepG2 cells. J Nutr. 135, p. 227–32. 5. Jiang, Q., Freiser, H., Wood, K.V., and Yin, X.,(2007) Identification and quantitation of novel vitamin E metabolites, sulfated longchain carboxychromanols, in human A549 cells and in rats. J Lipid Res. 48, p. 1221–30. 6. Hac-Wydro, K., Kapusta, J., Jagoda, A., Wydro, P., and Dynarowicz-Latka, P.,(2007) The influence of phospholipid structure on the interactions with nystatin, a polyene antifungal antibiotic A Langmuir monolayer study. Chem Phys Lipids. 150, p. 125–35. 7. Galli, F., Lee, R., Dunster, C., and Kelly, F.J.,(2002) Gas chromatography mass spectrometry analysis of carboxyethylhydroxychroman metabolites of alpha- and gamma-tocopherol in human plasma. Free Radic Biol Med. 32, p. 333–40. 8. Leonard, S.W., Paterson, E., Atkinson, J.K., Ramakrishnan, R., Cross, C.E., and Traber, M.G.,(2005) Studies in humans using deuterium-labeled alpha- and gamma-tocopherols demonstrate faster plasma gamma-tocopherol disappearance and greater gamma-metabolite production. Free Radic Biol Med. 38, p. 857–866. 9. Hybertson, B.M., Chung, J.H., Fini, M.A., Lee, Y.M., Allard, J.D., Hansen, B.N., Cho, O.J., Shibao, G.N., and Repine, J.E.,(2005) Aerosol-administered alpha-tocopherol attenuates lung inflammation in rats given lipopolysaccharide intratracheally. Exp Lung Res. 31, p. 283–94. 10. Morita, N., Traber, M.G., Enkhbaatar, P., Westphal, M., Murakami, K., Leonard, S.W., Cox, R.A., Hawkins, H.K., Herndon, D., Traber, L.D., and Traber, D.L.,(2006) Aerosolized alpha-tocopherol ameliorates acute lung injury following combined burn and smoke inhalation injury in sheep. Shock. 25, p. 277–82. 11. Enkhbaatar, P., Cox, R.A., Traber, L.D., Westphal, M., Aimalohi, E., Morita, N., Prough, D.S., Herndon, D.N., and Traber, D.L.,(2007) Aerosolized anticoagulants ameliorate acute lung injury in sheep after exposure to burn and smoke inhalation. Crit Care Med. 12, p. 2805–10 12. Palmieri, T.L., Enkhbaatar, P., Bayliss, R., Traber, L.D., Cox, R.A., Hawkins, H.K., Herndon, D.N., Greenhalgh, D.G., and Traber, D.L.,(2006) Continuous nebulized albuterol attenuates acute lung injury in an ovine model of combined burn and smoke inhalation. Crit Care Med. 34, p. 1719–24.

533

13. van Meer, G., Voelker, D.R., and Feigenson, G.W.,(2008) Membrane lipids: where they are and how they behave. Nat Rev Mol Cell Biol. 9, p. 112–24. 14. Marsh, D.,(2007) Lateral pressure profile, spontaneous curvature frustration, and the incorporation and conformation of proteins in membranes. Biophys J. 93, p. 3884–99. 15. Hannun, Y.A. and Obeid, L.M.,(2008) Principles of bioactive lipid signalling: lessons from sphingolipids. Nat Rev Mol Cell Biol. 9, p. 139–50. 16. Wu, D. and Meydani, S.N.,(2008) Age-associated changes in immune and inflammatory responses: impact of vitamin E intervention. J Leukoc Biol. 84, p. 900–14 17. Teichgraber, V., Ulrich, M., Endlich, N., Riethmuller, J., Wilker, B., De Oliveira-Munding, C.C., van Heeckeren, A.M., Barr, M.L., von Kurthy, G., Schmid, K.W., Weller, M., Tummler, B., Lang, F., Grassme, H., Doring, G., and Gulbins, E.,(2008) Ceramide accumulation mediates inflammation, cell death and infection susceptibility in cystic fibrosis. Nat Med. 14, p. 382–91. 18. Jobe, A.H. and Ikegami, M.,(2001) Biology of surfactant. Clin Perinatol. 28, p. 655–69, vii–viii. 19. Orgeig, S., Daniels, C.B., Johnston, S.D., and Sullivan, L.C.,(2003) The pattern of surfactant cholesterol during vertebrate evolution and development: does ontogeny recapitulate phylogeny? Reprod Fertil Dev. 15, p. 55–73. 20. Panda, A.K., Nag, K., Harbottle, R.R., Rodriguez-Capote, K., Veldhuizen, R.A., Petersen, N.O., and Possmayer, F.,(2004) Effect of acute lung injury on structure and function of pulmonary surfactant films. Am J Respir Cell Mol Biol. 30, p. 641–50. 21. Gunasekara, L., Schurch, S., Schoel, W.M., Nag, K., Leonenko, Z., Haufs, M., and Amrein, M.,(2005) Pulmonary surfactant function is abolished by an elevated proportion of cholesterol. Biochim Biophys Acta. 1737, p. 27–35. 22. Ingold, K.U., Webb, A.C., Witter, D., Burton, G.W., Metcalfe, T.A., and Muller, D.P.,(1987) Vitamin E remains the major lipid-soluble, chain-breaking antioxidant in human plasma even in individuals suffering severe vitamin E deficiency. Arch Biochem Biophys. 259, p. 224–5. 23. Traber, M.G., Frei, B., and Beckman, J.S.,(2008) Vitamin E revisited: do new data validate benefits for chronic disease prevention? Curr Opin Lipidol. 19, p. 30–8. 24. Traber, M.G.,(2005) Vitamin E, in Modern Nutrition in Health and Disease, (Shils, M.E.,

534

Beattie and Schock

Olson, J.A., Shike, M., and Ross, A.C., eds.), Lippincott Williams & Wilkins: Baltimore. p. 396–411. 25. Cross, C.E., van der Vliet, A., O’Neill, C.A., Louie, S., and Halliwell, B.,(1994) Oxidants, antioxidants, and respiratory tract lining fluids. Environ Health Perspect. 102 Suppl 10, p. 185–91. 26. Bouhafs, R.K. and Jarstrand, C.,(1999) Phagocyte-induced lipid peroxidation of lung surfactant. Pediatr Pulmonol. 27, p. 322–7. 27. Bouhafs, R.K. and Jarstrand, C.,(1999) Lipid peroxidation of lung surfactant by bacteria. Lung. 177, p. 101–10. 28. Atkinson, J., Epand, R.F., and Epand, R.M.,(2008) Tocopherols and tocotrienols in membranes: a critical review. Free Radic Biol Med. 44, p. 739–64. 29. Cuschieri, J., Bulger, E., Biligren, J., Garcia, I., and Maier, R.V.,(2007) Vitamin E inhibits endotoxin-mediated transport of phosphatases to lipid rafts. Shock. 27, p. 19–24. 30. Barbas, C., Castro, M., Bonet, B., Viana, M., and Herrera, E.,(1997) Simultaneous determination of vitamins A and E in rat tissues by high-performance liquid chromatography. J Chromatogr A. 778, p. 415–20. 31. Ito, Y., Ochiai, J., Sasaki, R., Suzuki, S., Kusuhara, Y., Morimitsu, Y., Otani, M., and Aoki, K.,(1990) Serum concentrations of carotenoids, retinol, and alpha-tocopherol in healthy persons determined by high-performance liquid chromatography. Clin Chim Acta. 194, p. 131–44. 32. Podda, M., Weber, C., Traber, M.G., and Packer, L.,(1996) Simultaneous determination of tissue tocopherols, tocotrienols, ubiquinols, and ubiquinones. J Lipid Res. 37, p. 893–901. 33. Ikenoya, S., Abe, K., Tsuda, T., Yamano, Y., Hiroshima, O., Ohmae, M., and Kawabe, K.,(1979) Electrochemical detector for highperformance liquid chromatography. II. Determination of tocopherols, ubiquinones and phylloquinone in blood. Chem Pharm Bull (Tokyo). 27, p. 1237–44. 34. Leonard, S.W., Gumpricht, E., Devereaux, M.W., Sokol, R.J., and Traber, M.G.,(2005) Quantitation of rat liver vitamin E metabolites by LC–MS during high-dose vitamin E administration. J. Lipid. Res. 46, p. 1068–75. 35. Leonard, S.W., Bruno, R.S., Paterson, E., Schock, B.C., Atkinson, J., Bray, T.M., Cross, C.E., and Traber, M.G.,(2003) 5-nitro-gamma-tocopherol increases in human plasma exposed to cigarette smoke in vitro and in vivo. Free Radic Biol Med. 35, p. 1560–7. 36. Galli, F., Lee, R., Dunster, C., and Kelly, F.J.,(2002) Gas chromatography mass

spectrometry analysis of carboxyethylhydroxychroman metabolites of alpha- and gamma-tocopherol in human plasma. Free Radic Biol Med. 32, p. 333–40. 37. Saito, Y., Yoshida, Y., Nishio, K., Hayakawa, M., and Niki, E.,(2004) Characterization of cellular uptake and distribution of vitamin E. Ann N Y Acad Sci. 1031, p. 368–75. 38. Panin, L.E., Polyakov, L.M., Kolosova, N.G., Russkikh, G.S., and Poteryaeva, O.N.,(1998) Distribution of tocopherol and apolipoprotein A-I immunoreactivity in rat liver chromatin. Membr Cell Biol. 11, p. 631–40. 39. Kiss, E., Nagy, P., Balogh, A., Szollosi, J., and Matko, J.,(2008) Cytometry of raft and caveola membrane microdomains: from flow and imaging techniques to high throughput screening assays. Cytometry A. 73, p. 599–614. 40. Gombos, I., Steinbach, G., Pomozi, I., Balogh, A., Vamosi, G., Gansen, A., Laszlo, G., Garab, G., and Matko, J.,(2008) Some new faces of membrane microdomains: a complex confocal fluorescence, differential polarization, and FCS imaging study on live immune cells. Cytometry A. 73, p. 220–9. 41. Biro, A., Cervenak, L., Balogh, A., Lorincz, A., Uray, K., Horvath, A., Romics, L., Matko, J., Fust, G., and Laszlo, G.,(2007) Novel anticholesterol monoclonal immunoglobulin G antibodies as probes and potential modulators of membrane raft-dependent immune functions. J Lipid Res. 48, p. 19–29. 42. Touboul, D., Brunelle, A., Halgand, F., De La Porte, S., and Laprevote, O.,(2005) Lipid imaging by gold cluster time-of-flight secondary ion mass spectrometry: application to Duchenne muscular dystrophy. J Lipid Res. 46, p. 1388–95. 43. Beattie, J.R., Bell, S.E.J., Borgaard, C., Fearon, A., and Moss, B.W.,(2006) Prediction of adipose tissue composition using Raman spectroscopy: Average properties and individual fatty acids. Lipids. 41, p. 287–294. 44. Beattie, J.R., Bell, S.E.J., Borgaard, C., Fearon, A.M., and Moss, B.W.,(2004) Multivariate prediction of clarified butter composition using Raman spectroscopy. Lipids. 39, p. 897–906. 45. Hancewicz, T.M. and Petty, C.,(1995) Quantitative analysis of vitamin A using Fourier transform Raman spectroscopy. Spectroc Acta Pt A-Molec Biomolec Spectr. 51, p. 2193–2198. 46. Schulz, H. and Baranska, M.,(2007) Identification and quantification of valuable plant substances by IR and Raman spectroscopy. Vib Spectrosc. 43, p. 12–25.

Identifying the Spatial Distribution of Vitamin E 47. Beattie, J.R., Bell, S.E., and Moss, B.W.,(2004) A critical evaluation of Raman spectroscopy for the analysis of lipids: fatty acid methyl esters. Lipids. 39, p. 407–19. 48. Bunow, M.R. and Levin, I.R.,(1977) Raman spectra and vibrational assignments for deuterated membrane lipids. 1,2 dipalmitoyl phoshatisylcholine-d9 and d62. Biochim Biophys Acta. 489, p. 191–206. 49. Kint, S., Wermer, P.H., and Scherer, J.R.,(1992) Raman-Spectra of hydrated phospholipid-bilayers 2. Water and headgroup interactions. J Phys Chem. 96, p. 446–452. 50. Lawson, E.E., Anigbogu, A.N.C., Williams, A.C., Barry, B.W., and Edwards, H.G.M.,(1998) Thermally induced molecular disorder in human stratum corneum lipids compared with a model phospholipid system; ft-Raman spectroscopy. Spectrochim Acta A-Mol Biomol Spectros. 54, p. 543–558. 51. Wallach, D.F.H., Verma, S.P., and Fookson, J.,(1979) Application of laser Raman and infrared spectroscopy to the analysis of membrane structure. Biochim Biophys Acta. 559, p. 153–208. 52. Beattie, J.R., Brockbank, S., McGarvey, J.J., and Curry, W.J.,(2005) Effect of excitation wavelength on the Raman spectroscopy of the porcine photoreceptor layer from the area centralis. Mol Vis. 11, p. 825–832. 53. Haka, A.S., Volynskaya, Z., Gardecki, J.A., Nazemi, J., Lyons, J., Hicks, D., Fitzmaurice, M., Dasari, R.R., Crowe, J.P., and Feld, M.S.,(2006) In vivo margin assessment during partial mastectomy breast surgery using Raman spectroscopy. Cancer Res. 66, p. 3317–3322. 54. Min, Y.K., Yamamoto, T., Kohda, E., Ito, T., and Hamaguchi, H.,(2005) 1064 nm nearinfrared multichannel Raman spectroscopy of fresh human lung tissues. J Raman Spectrosc. 36, p. 73–76. 55. Lajoie, P. and Nabi, I.R., (2007) Regulation of raft-dependent endocytosis. J Cell Mol Med. 11, p. 644–53. 56. Eriksson, L., Johansson, E., Kettaneh-Wold, N., Trygg, J., Wikström, C., and Wold, S.,

535

eds. Multi- and Megavariate Data Analysis Part I: Basic Principles and Applications, Second revised and enlarged edition 2nd ed. Vol. 1. 2006, Umetrics: Umea. p. 425. 57. Eriksson, L., Johansson, E., Kettaneh-Wold, N., Trygg, J., Wikström, C., and Wold, S., eds. Multi- and Megavariate Data Analysis Part II: Advanced Applications and Method Extensions, Second revised and enlarged edition 2nd ed. Vol. 2. 2006, Umetrics: Umea. 345. 58. McCreery, R.L.,(2000) Raman Spectroscopy for Chemical Analysis. 1 ed. in Chemical Analysis: a Series of Monographs of Analytical Chemistry and its Applications., (Winefordner, J.D., ed). Wiley, New York. 59. Tu, A.T.,(1982) Raman Spectroscopy in Biology: Principles and Applications. Wiley, New York. 60. Bell, S.B., JR; McGarvey, J.J.; Peters, K L., Sirimuthu, N.M.S., Speers, S J.,(2004) Development of sampling methods for Raman analysis of solid dosage forms of therapeutic and illicit drugs. J Raman Spectrosc. 35, p. 409–417. 61. Glenn, J.V., Beattie, J.R., Barrett, L., Frizzell, N., Thorpe, S.R., Boulton, M.E., McGarvey, J.J., and Stitt, A.W.,(2007) Confocal Raman microscopy can quantify advanced glycation end product (AGE) modifications in Bruch’s membrane leading to accurate, nondestructive prediction of ocular aging. FASEB J. 21, p. 3542–52. 62. Beattie, J.R., Glenn, J.V., Boulton, M.E., Stitt, A.W., and McGarvey, J.J.,(2009) Effect of signal intensity normalization on the multivariate analysis of spectral data in complex ‘real-world’ datasets. J Raman Spectrosc. 40, p. 429–435. 63. Jirasek, A., Schulze, G., Yu, M.M.L., Blades, W., and Turner, R.F.B.,(2004) Accuracy and precision of manual baseline determination. Appl Spectrosc. 58, p. 1488–1499. 64. Lieber, C.A. and Mahadevan-Jansen, A.,(2003) Automated method for subtraction of fluorescence from biological Raman spectra. Appl Spectrosc. 57, p. 1363–1367.

sdfsdf

Index A Abundant neutral losses.................................................... 43 Acidic phospholipids...................................................... 216 Adenine nucleotide transporter.......................................... 4 Adipose tissue genomic DNA analysis cells preparation and suspension................. 348–349 GC/MS analysis......................................... 350–351 hydrolysis.................................................... 349–350 preparation.......................................................... 349 purine dR derivatization............................. 350–351 standards............................................................. 349 2 H2O labeling adipose tissue TG............................................... 353 body water analysis......................342, 344, 351–352 cell division measurement........................... 340–341 de novo lipogenesis (DNL)........................ 339–340 deuterated water labeling, animals...................... 344 fatty acids and glycerol................................ 342, 343 fractional cell proliferation rate........................... 356 fractional DNL................................................... 353 fractional synthesis rate (FSR).................... 352–353 genomic DNA analysis............................... 342, 343 growing obese rats application............................ 345 isotope methodologies........................................ 338 lipid analyses............................................... 341, 343 lipid synthesis rate...................................... 353–356 thin-layer chromatography......................... 341–342 total lipid extraction.................................... 341–343 lipid analyses fatty acids and glycerol........................................ 347 glycerol GC/MS analysis.................................... 348 glycerol separation and derivatization................. 347 lipid classes separation, TLC...................... 346–347 palmitate GC/MS analysis......................... 347–348 total lipid extraction.................................... 345–346 lipolytic proteome mapping activity tagging........................................... 501, 503 equipment................................................... 500–501 fractionation....................................................... 501 gel electrophoresis....................................... 501–505 homogenization.......................................... 502–503 LC–MS/MS analysis.................................. 502, 507 lipolytic proteome....................................... 507–508 phosphonate activity-based probes............. 499–500 proteolytic in-gel digestion..................502, 505–507

specific tagging................................................... 498 tryptic peptides........................................... 507, 509 visualization................................................ 502, 505 metabolic pathways..................................................... 71 TG synthesis measurement.............................. 338–339 tissue and body water collection............................... 344 Ageing erythrocytes........................................................ 130 Age-related macular eye disease (AMD)................ 247, 250 Albumin........................................................................... 71 O-Alk-1′-enyl aliphatic chain......................................... 289 Alzheimer’s disease......................................................... 289 Ammonium hydroxide..................................................... 44 Anionic glycerophospholipids.................................... 25, 26 Anion, of stearic acid...................................................... 117 Antibody-based assays...................................................... 73 Apolipoprotein................................................................. 71 Apolipoprotein A1........................................................... 82 AQUA peptide................................................................. 79 Arabinogalactan.......................................................... 90, 95 Arachidonic acid..................................................33, 73, 271 Atherosclerosis............................................................... 104 Atmospheric pressure chemical ionization (APCI)..................................104, 162, 202, 223 Atmospheric pressure chemical ionization mass spectrometry (APCI/MS)............................. 166 Atmospheric pressure photo ionization (APPI)...........................................166–167, 223 Automatic gain control (AGC)...................................... 253

B Behenyl oleate (BO)............................................... 226, 229 BHT, for protecting lipids from oxidation........................ 83 Bioactive lipids................................................................. 73 Bioactive sphingolipids (SPLs) quantitative analysis ceramide 1-phosphates (Cer-1P)...................... 460–461 collision energy and retention times................. 458–464 data presentation.............................................. 445–456 ESI/MS multiple reaction monitoring...................... 451–453 precursor ion scan (PI)................................ 450–451 scan modes.......................................................... 450 glucosyl/galactosyl-ceramides (Glu-Cers)........ 462–433 HPLC conditions............................................. 456–458 identification............................................................ 453 lactosylceramide (Lac-Cer)....................................... 463 mass spectrometry approach............................. 445–446

537

ipidomics 538  LIndex



 

Bioactive sphingolipids (SPLs) quantitative analysis (Continued) materials analytical standards............................................. 447 equipment........................................................... 446 reagents....................................................... 446–447 supplies............................................................... 448 quantitation calibration curve slopes............................... 453–455 internal standards selection......................... 454–455 sample preparation and extraction.................... 448–450 sphingomyelins (SMs)...................................... 463–464 sphingosine....................................................... 446, 456 sphingosine 1-phosphate.................................. 446, 457 structural diversity and nomenclature............... 444–445 Biological tissues lipids, types......................................... 212 Biomimetic models. See also Membrane lipidomics and unsaturated fatty acids geometry fatty acid isomerization..................................... 401–402 liposomes, biomimetic chemistry membrane models bilayer arrangement............................................ 396 free radical processes........................................... 395 heterogeneous system reactivity.................. 396–397 isomerization...................................................... 398 multilayer vesicles (MLV).................................. 396 2,3-Bisphosphoglycerate mutase deficiency................... 131 BLAST, for peptides to confirm structure........................ 79 Body water analysis acetone extraction and transfer................................. 352 body water distillation.............................................. 351 GC/MS analysis, acetone......................................... 352 Brain mitochondrial isolation......................................... 5–7 Brain mitochondrial lipidome.......................................... 17 Bronchoalveolar lavage fluid (BALF)............................. 163 Bruch’s membrane.......................................................... 248 Butylated hydroxy-toluene (BHT)......................... 139, 175

C Calcium homeostasis.......................................................... 3 Cancer....................................................................... 20, 162 Cardiolipin......................................................4, 17, 21, 152 Cardiovascular disease.............................................. 20, 162 Carnitine acyltransferase.................................................... 4 Cationic lipids................................................................ 190 Caveolae cell labeling of laurdan...................................... 266, 268 GC analysis, fatty acid methyl esters................ 265–267 GP-value calculation................................................ 268 isolation procedure from cultured cells.............................. 262 solutions preparation.................................. 263–265 visualization of lipid rafts using............................................... 266

RGB images, of laurdan-labeled CD4+ T cells........................................................... 269 Cell lipidome, with 13C isotope........................................ 92 Cell membrane structure, alteration of........................... 261 Ceramide 1-phosphates (Cer-1P).......................... 460–461 Ceramides.........................35, 39, 67, 74, 133, 222, 227, 229 Cerebrospinal fluids (CSF) apolipoprotein A1, peptidemass digest of................... 82 lipid-metabolizing proteins liquid chromatography, of peptides....................... 77 mass spectrometry, of peptides.............................. 77 quantification of............................................. 77–79 Shotgun-sequencing of......................................... 76 from UniProt Knowledgebase......................... 80–81 lipids detection and analyses of................................ 75–76 extraction of.......................................................... 75 Charged aerosol detection, HPLC analysis............................................................. 475–476 approaches comparison..................................... 470–472 corona CAD parameters........................................... 474 fat-soluble vitamins.......................................... 478–482 fatty alcohols..................................................... 478–479 free fatty acids................................................... 476–478 HPLC system parameters................................ 473–474 materials equipment........................................................... 472 HPLC mobile phases......................................... 473 reagents and supplies.......................................... 472 reversed-phase-CAD approach................................ 472 sample preparations.......................................... 474–475 Chip-based nano-electrospray ionization (nESI)............. 37 Chl esters........................................................................ 222 Chloride (Cl-) ion............................................................. 42 Cholera toxin b immunostaining........................................ 9 Cholesterol (Chl)............................................................ 222 Cholesteryl esters............................................................. 35 Cholesteryl oleate........................................................... 227 Chronic obstructive pulmonary disease (COPD)........... 162 Collision energies............................................................. 43 Collision energy and retention times...................... 458–464 Collision-induced dissociation (CID)..................21, 55, 76, 141, 142, 145, 150, 157, 169, 273, 413–414 Colonocyte..................................................................... 262 Complementary precursor ion scan............................. 35, 47, 48, 54, 57, 61. See also Neutral loss scans mode analysis Confocal Raman microscopy advantages and disadvantages........................... 521–522 band assignments.............................................. 519–520 biochemical constituents................................... 528, 530 calibration......................................................... 523–524 chemical structures....................................517–518, 529 discrimination power........................................ 518–519 lipids and lipid-soluble compounds.................. 516–517

Lipidomics 539 Index   

localisation and distribution of molecules......... 513–514 materials equipment................................................... 522–523 reagents and supplies.......................................... 523 membrane lipids............................................... 514–515 multivariate statistics................................................ 522 physical interactions................................................. 529 pulmonary surfactant................................................ 515 qualitative mapping.......................................... 525–528 quantitative mapping................................................ 529 validation.......................................................... 524–525 vitamin E.......................................................... 515–516 Contaminants to avoid............................................................. 139, 217 in LC/MS chromatograms....................................... 215 Corona CAD parameters............................................... 474 Creatine kinase................................................................... 4 Cyclooxygenase (COX).....................................73, 162, 271 Cysteinyl-containing leukotrienes.................................. 168 Cytochrome P450 (CYP)............................................... 162

D 7-Dehydrocholesterol..................................................... 140 De novo lipogenesis (DNL)................................... 339–340 Depression........................................................................ 20 Design peptides, in silico using Peptidemass.................... 79 DHA-containing glycerophospholipids................19, 20, 29 quantification, mass spectrometer scan........................................................... 27–28 structural characterization.......................................... 27 Diabetes mellitus (DM)................................................. 164 Diacylglycerols........................................................ 103, 132 Di-and triesters.............................................................. 222 Dietary fat composition on membrane structure, impact of....................................................... 261 to measure biomarkers for........................................ 210 Dietary lipid, changes in................................................... 72 Dihomo-gamma linoleic (20:3, n-6) (DHGLA)............ 271 2,5-Dihydroxybenzoic acid (DHB).................106, 108, 253 Dihydroxyeicosatetraenoic acids (DHETEs)................. 168 Diisopropyl-amine (DIPA)............................................ 137 2-Dimensional liquid chromatography............................. 84 Dimethyl sulfoxide (DMSO)........................................... 93 1,2-Dimyristoyl-phosphatidylethanolamine................... 239 2D ion trap..................................................................... 168 3D ion trap..................................................................... 168 Dipalmitin...................................................................... 227 1,2-Distearoyl-phosphatidylserine (SSPS)............. 239, 240 2,6-Di-tert-butyl-p-cresol.............................................. 139 DNL. See De novo lipogenesis (DNL) 2D NMR spectroscopy..................................................... 91 Docosahexaenoic acid (DHA22:6 n-3).................19, 33, 34 Down syndrome............................................................. 289

Dry eye disease............................................................... 247 2D-TLC systems.............................................................. 90

E EDTA-containing samples............................................. 135 EFA profiling, of RBC membrane................................. 128 Eicosanoids................................................73, 161, 162, 271 analysis of................................................................. 163 biosynthesis.............................................................. 161 chromatographic separation of................................. 179 ESI/MS/MS chromatogram.................................... 180 MS2 analysis, of 5-HETE and 12-HETE............... 181 quantitation of.................................................. 171–174 selective reaction monitoring of................................ 171 Eicosapentaenoic (20:5, n-3) (EPA) acids.......... 19–22, 271 Electron capture (EC).................................................... 166 Electron capture dissociation (ECD)............................. 169 Electron transport chain..................................................... 4 Electrospray ionization (ESI)...........................34, 104, 162, 165–166, 202, 223, 272 Electrospray ionization HPLC/ ESI-MS........................ 90 Electrospray ionization/mass spectrometry (ESI/MS) multiple reaction monitoring............................ 451–453 precursor ion scan (PI)...................................... 450–451 scan modes............................................................... 450 Electrospray-ionization mass spectrometry (ESI–MS)....................................................... 20 ELISA, for detection of enzyme expression..................... 73 Enzyme-immunoassay techniques (EIA)............... 162, 272 EPO-receptor signaling system...................................... 131 Epoxyeicosatrienoic acids (EETs).................................. 161 Epoxygenases.................................................................... 73 Equipment for lipidomics.............................................................. 73 for proteomics............................................................. 74 Ergosterol 140 Essential fatty acid (EFA).............................................. 127

F Fat-soluble vitamins, HPLC-charged aerosol detection............................................... 478–482 Fatty acid (FA)........................................................... 21, 73 amides....................................................................... 222 molar ratios............................................................... 128 Fatty acid elongases cultured cells assessment extraction and analysis, metabolites.................... 382 high performance liquid chromatography................. 383–384 primary hepatocytes.................................... 381–382 saponification.............................................. 382–383 in vivo assessment adenoviral-mediated over expression.................. 384 loss-of-and gain-of-function.............................. 384

ipidomics 540  LIndex



 

Fatty acid elongases (Continued) primary hepatocytes infection..................... 384–385 recombinant adenovirus generation methods................................................ 384–385 recombinant adenovirus infection............... 385–386 microsomes control studies............................................. 380–381 enzymatic assay................................................... 380 extraction and analysis........................................ 380 preparation.......................................................... 379 Fatty acid methyl esters (FAMEs).................................. 488 Fatty acid profiles and TAGs determination, MALDI-TOF/MS fingerprinting conventional methods....................................... 321–324 determination methods............................................. 317 GC/MS determination comparison................. 332–333 materials equipments......................................................... 317 reagents....................................................... 317–318 olive oils............................................................ 319–321 composition........................................................ 320 fingerprinting......................................320–321, 324 oil extraction............................................... 319–320 quantitative information......................319–320, 322 pomegranate and soybean oil............................ 325–327 MALDI analysis......................................... 325, 327 plant material and extraction.............................. 325 TAG profiling............................................. 325–326 theoretical molecular mass........................................ 319 vegetable oils......................................318–319, 327–332 Fatty acid profiles, vegetable oils allowable fatty acid ranges........................................ 328 GC/MS determination............................................. 329 linolenic acid isomers................................................ 331 oil quality.......................................................... 327–329 pomegranate oils............................................... 329–332 Fatty alcohols, HPLC-charged aerosol detection................................... 478–479 Ficoll and sucrose gradients................................................ 4 Ficoll gradient.................................................................. 16 Fluorescence resonance energy transfer (FRET)......................................................... 268 Fluorescent suicide inhibitors, lipolytic proteome mapping activity tagging......................................................... 503 gel electrophoresis............................................. 503–505 homogenization................................................ 502–503 LC–MS/MS analysis........................................ 502, 507 lipolytic proteome............................................. 507–508 materials activity tagging................................................... 501 equipment................................................... 500–501 fractionation....................................................... 501 gel electrophoresis....................................... 501–502 proteolytic in-gel digestion................................. 502 phosphonate activity-based probes................... 499–500 proteolytic in-gel digestion............................... 505–507

specific tagging......................................................... 498 tryptic peptides................................................. 507, 509 visualization...................................................... 502, 505 Fourier transform ion cyclotron resonance mass spectrometry (FT–ICR MS).......... 90, 169–171 Fractional synthesis rate (FSR)............................... 352–353 Free cholesterol....................................................... 227, 229 Free fatty acids................................................................ 222 Free fatty acids HPLC-charged aerosol detection............................................... 476–478 FSR. See Fractional synthesis rate FTICR/MS instrument................................................. 170 Full width at half-height maximum (FWHM).............. 165

G Gangliosides................................................................... 216 Gas chromatography (GC)............................................... 34 Gas chromatography coupled to mass spectrometry (GC/MS)...................................................... 162 Gas chromatography separation, cis and trans isomers.................................................. 405, 408 Gas–liquid chromatography (GLC)................128, 367–368 Gel electrophoresis 1D gel electrophoresis...................................... 503–504 2D gel electrophoresis...................................... 504–505 Genomic DNA analysis, adipose tissue cells preparation and suspension....................... 348–349 GC/MS analysis............................................... 350–351 hydrolysis.......................................................... 349–350 preparation................................................................ 349 purine dR derivatization................................... 350–351 standards................................................................... 349 GLC. See Gas–liquid chromatography (GLC) Glial cell bodies.................................................................. 3 Global 13C-enrichement strategy...................................... 92 Glucosyl/galactosyl-ceramides (Glu-Cers)............. 462–433 Glutamate metabolism....................................................... 3 Glycerophosphatidic acid (GPA)............................... 35, 67 Glycerophosphatidylcholine (GPCho)............35, 39, 42–48 Glycerophosphatidylglycerol (GPGro)....................... 35, 67 Glycerophosphatidylinositol............................................. 35 Glycerophosphatidylserine (GPSer)..................... 35, 54–59 Glycerophospholipids (GPLs)................... 16, 19, 20, 28, 37 containing DHA, in mouse skeletal muscle................ 29 distribution of DHA, workflow for tracking.............. 23 G-protein-coupled receptors.......................................... 288

H Haematopoietic stem cells.............................................. 130 Hemoglobinopathies...................................................... 133 n-Hexane-propan-2-ol solvent....................................... 244 High performance liquid chromatography (HPLC) charged aerosol detection analysis........................................................ 475–476 approaches comparison............................... 470–472

Lipidomics 541 Index   

corona CAD parameters..................................... 474 fat-soluble vitamins.................................... 478–482 fatty alcohols............................................... 478–479 free fatty acids............................................. 476–478 HPLC system parameters........................... 473–474 materials..................................................... 472–473 reversed-phase-CAD approach.......................... 472 sample preparations.................................... 474–475 disadvantage............................................................. 104 eicosanoids quantification......................................... 162 meibum..................................................................... 222 phospholipids........................................................... 289 solvents...........................74, 75, 223, 244, 275, 279, 280 High-performance liquid chromatography–tandem mass spectrometry (HPLC–MS).................. 458–464 High-pressure liquid chromatography–mass spectrometry (HPLC–MS)................................................ 221 2 H2O labeling, adipose tissue development adipose tissue TG..................................................... 353 body water analysis............................342, 344, 351–352 cell division measurement................................. 340–341 De novo lipogenesis (DNL)............................. 339–340 deuterated water labeling, animals............................ 344 fatty acids and glycerol...................................... 342, 343 fractional cell proliferation rate................................. 356 fractional DNL......................................................... 353 fractional synthesis rate (FSR).......................... 352–353 genomic DNA analysis..................................... 342, 343 growing obese rats application.................................. 345 isotope methodologies.............................................. 338 lipid analyses..................................................... 341, 343 lipid synthesis rate............................................ 353–356 thin-layer chromatography............................... 341–342 total lipid extraction.......................................... 341–343 HPLC–MS. See High-performance liquid chromatography–tandem mass spectrometry (HPLC–MS) HPLC–MSn procedures................................................. 243 Human erythrocytes lipidomic analysis of......................................... 134–135 blood sample collection............................... 135–136 erythrocyte storage...................................... 135–136 lipid extraction............................................ 136–139 lipid extracts, storage of.............................. 139–140 RBC phospholipids analysis HPLC separation............................................... 154 by tandem hybrid MS/MS (see Tandem hybrid MS/MS) by tandem MS2.......................................... 141–145 Human meibum, analysis............................................... 225 chromatogram, of human meibum sample............... 231 extracted chromatogram, of ion m/z369................... 232 fragmentation patterns, of ions................................. 238 HPLC gradient, for polar lipid analysis on............... 239 HPLC–MS analyses

of equimolar mixture, of Chl-O and Chl............ 233 nonpolar lipids, structural elucidation of....................................................234, 235, 237 of nonpolar lipid standard........................... 227, 229 mass spectra of mixture, of standard lipids and......... 230 nonpolar lipid analyses.............................................. 226 HPLC procedures.............................................. 226 mass spectrometric procedures.................... 226–227 polar lipid analysis, using gradient NP HPLC–APCI MS chromatogram, of mixture of polar lipids........... 240 HPLC procedures...................................... 238–239 mass spectrometric procedures............................ 239 reconstructed chromatograms, of polar lipids..................................................... 241, 243 sample collection...................................................... 225 structural elucidation, of meibum compound m/z 619......................................................... 236 total ion chromatogram............................................ 228 Human red blood cell (RBC) membrane....................... 127 Hybrid analyzers............................................................ 171 Hydrocarbons................................................................. 222 Hydroxyeicosatetraenoic acids (HETEs)....................... 161 12-Hydroxyheptadecatrienoic acid (12-HHT).............. 168

I Imaging mass spectrometry (IMS)................................. 248 for analyzing flat-mounted eye segments................. 252 averaged MS/MS spectrum, of m/z 782.6................ 258 fixation conditions, verification of............................ 254 fundus and macula, images captured by.................... 257 image processing, and interpretation................ 255–256 lipid identification.................................................... 256 mounting tissue for................................................... 252 Immunoelectron microscopy.......................................... 268 Infrared multiphoton dissociation (IRMPD)................. 169 Infrared spectroscopy...................................................... 222 In situ corona discharge...................................416, 418–419 Ion trap mass analyzers....................................167–169, 223 Isomeric phospholipids..................................................... 29 Isotope corrections........................................................... 30

J JAK2 mutations.............................................................. 131

L Lactosylceramide (Lac-Cer)........................................... 463 Large unilamellar vesicles by extrusion technique (LUVET)..............................398–400, 403–405 cis–trans isomerization regioselectivity.............. 398–399 irradiation conditions, isomerization................ 404–405 preparation........................................................ 403–404 Laurdan.......................................................................... 266 Laurdan-labeled CD4+ T cells........................................ 269

ipidomics 542  LIndex



 

LC/MS/MS-based quantification method, for eicosanoid................................................ 163 LCQ ion trap mass spectrometer..................................... 77 Lecithin-cholesterol acyl transferase.............................. 128 Less polar nonwater-soluble solvent....................... 137–138 Leukotrienes (LTs)......................................................... 161 Limit of detection (LOD).............................................. 164 Lipases........................................................................ 72, 73 Lipid analyses fatty acids.................................................................. 347 fatty acids and glycerol separation and derivatization................................................ 347 glycerol..................................................................... 347 glycerol GC/MS analysis.......................................... 348 lipid classes separation, TLC............................ 346–347 palmitate GC/MS analysis............................... 347–348 total lipid extraction.......................................... 345–346 Lipid composition altering during thawing............................................ 217 of human TF and TFLL........................................... 222 Lipid extraction for determining phosphoinositides, PtdIns, PIP, and PIP2....................................................... 190 and fractionation....................................................... 193 from rat retinal tissue............................................ 36–37 requiring ratio of chloroform, methanol, and water...................................................... 217 SCX column for....................................................... 192 solvent preparation for.............................................. 191 Lipid maps database......................................................... 83 Lipid mass spectrum analysis (LIMSA)................... 37, 129 Lipid metabolic pathways................................................. 71 disruption of............................................................... 66 Lipid metabolizing proteins, changes in........................... 72 Lipid modification............................................................ 72 Lipidomic analysis.......................................................... 202 approach, lipids from................................................ 208 to investigate lipid composition effects..................................... 203 lipid metabolism role.......................................... 203 large lipids analysis........................................... 212–213 by LC/MS.......................................................... 213 by MS/MS.......................................................... 214 by LC–ESIMS/MS.......................................... 291–292 LC/MS analysis of lipids....................................................... 210–212 maintenance and system suitability testing................................................... 215–216 prostanoids assay....................................................... 273 reagents for................................................................. 74 sample preparation, for analysis................................ 207 biological fluids........................................... 209–210 cells..................................................................... 207 tissues.......................................................... 208–209

small lipids analysis........................................... 210–212 by LC/MS.......................................................... 211 sterol lipids analysis.......................................... 212–213 to study abiotic environmental factors.............................. 202 genetic diversity effects............................... 202–203 lipid metabolism in role...................................... 203 nutrients effect.................................................... 202 of patient erythrocyte membranes...................... 129 pharmacological treatments effects..................... 202 to understand biochemical mechanisms for............................... 203 physiological roles by endogenous lipids in......... 203 Lipid particles analysis, yeast cell fractionation SDS-PAGE analysis............. 364, 369 extraction and quantification.................................... 360 fatty acids GLC analysis................................... 367–368 isolation............................................................ 361–362 lipid extraction.................................................. 363–365 mass spectrometric analysis...................................... 370 material equipment and supplies.............................. 360–361 reagents............................................................... 361 neutral lipids and phospholipids mass spectrometry.............................................. 369 TLC separation.......................................... 370–372 phospholipids TLC analysis............................. 366–367 protein analysis................................................. 363, 364 sterols GLC analysis................................................. 368 Lipids biological roles.................................................. 201–202 chemical classification of.......................................... 205 rafts........................................................................... 268 Lipids non-invasive mapping. See also Magnetic resonance imaging calibration procedure........................................ 490–492 data acquisition................................................. 489–490 grains and rapeseed................................................... 486 living barley grain quantitative imaging............ 492–493 materials equipment................................................... 487–488 reagents............................................................... 488 sample preparation and processing........................... 488 small-sized rape seed........................................ 494–495 total esterified fatty acids.......................................... 488 Lipolytic proteome mapping, adipose tissue activity tagging................................................. 501, 503 equipment......................................................... 500–501 fractionation............................................................. 501 gel electrophoresis............................................. 501–505 homogenization................................................ 502–503 LC–MS/MS analysis........................................ 502, 507 lipolytic proteome............................................. 507–508 phosphonate activity-based probes................... 499–500

protein patterns........................................................ 507 proteolytic in-gel digestion........................502, 505–507 specific tagging......................................................... 498 tryptic peptides................................................. 507, 509 visualization...................................................... 502, 505 Lipoprotein lipase........................................................... 128 Lipoxins (LXs)............................................................... 162 Lipoxygenase (LOX)................................................ 73, 162 Liquid chromatography electrospray ionization mass spectrometry (LC–ESIMS)................. 288 Liquid chromatography–tandem mass spectrometry (LC–MS/MS)...................................... 162, 272 Long chain polyunsaturated fatty acids (LCPUFA)............................................... 33, 34 LTQ database of CSF peptides........................................ 79 LUVET. See Large unilamellar vesicles by extrusion technique (LUVET) LysoPC (LPC)............................................................... 289 Lysophosphatidic acid (LPA)......................................... 288 Lyso-phosphatidylcholine.............................................. 107 Lysophosphatidylcholine acyltransferase (LPCAT)....... 134 Lysophosphatidylcholines (LPCs).......................... 104, 190 Lysophosphatidylserine (LPS)....................................... 288 Lysophospholipids............................................................ 73

M Magnetic resonance imaging advantages and disadvantages........................... 486–487 calibration procedure........................................ 490–492 data acquisition................................................. 489–490 grains and rapeseed................................................... 486 living barley grain quantitative imaging............ 492–493 materials equipment................................................... 487–488 reagents............................................................... 488 sample preparation and processing........................... 488 small-sized rape seed........................................ 494–495 total esterified fatty acids.......................................... 488 MALDI-TOF. See Matrix-assisted laser desorption/ ionization time-of-flight MALDI-TOF MS. See Matrix-assisted laser desorption and ionization time-of-flight mass spectrometry Mammalian fatty acid elongases cultured cells assessment extraction and analysis, metabolites.................... 382 high performance liquid chromatography................. 383–384 primary hepatocytes.................................... 381–382 saponification.............................................. 382–383 hepatic microsomal reaction sequence.............. 375–377 in vivo assessment adenoviral-mediated over expression.................. 384 loss-of-and gain-of-function.............................. 384 primary hepatocytes infection..................... 384–385

Lipidomics 543 Index    recombinant adenovirus generation methods................................................ 384–385 recombinant adenovirus infection............... 385–386 materials equipment........................................................... 378 reagents and supplies.................................. 378–379 microsomes control studies............................................. 380–381 enzymatic assay................................................... 380 extraction and analysis........................................ 380 preparation.......................................................... 379 pathway.................................................................... 376 Marker antibodies, lipid particles........................... 363–364 Mass spectrometry (MS)........................................ 104, 222 instrumentation.................................................. 8–9, 25 with or without HPLC............................................. 222 phospholipids containing DHA, identification............................................. 25–27 process for thin brain tissue section by..................... 249 sample preparation........................................................ 8 techniques, for eicosanoid HPLC/MS/MS analysis.................................................. 164–165 Mass spectrum of crude rat retina lipid extract.............................. 38–39 of erythrocyte PE and PC........................................ 145 of phosphatidylcholines and sphingomyelins............ 147 of phosphatidylethanolamine species........................ 146 for unfractionated rat retina lipid extract.................... 38 Matrix-assisted laser desorption and ionization time-of-flight mass spectrometry.......... 104–107 brain phosphoinositides, analysis by......................... 189 MALDI-TOF mass spectra of lipid extract..................................................... 120 of organic bovine liver extract............................. 115 of 1-palmitoyl-2-oleoyl-sn-phosphatidylcholine and................................................................ 113 quantitative data analysis............................ 111–112 recording of................................................. 110–111 positive ions, detected of liver extract and........ 116–117 sample processing artificial lipid samples................................. 108–109 crude lipid, separation of............................. 109–110 liver extract, preparation..................................... 109 Matrix assisted laser desorption/ionization (MALDI)............................................... 34, 258 Matrix-assisted laser desorption/ionization mass spectrometry (MALDI–MS)........................ 248 Matrix-assisted laser desorption/ionization time-of-flight conventional methods....................................... 321–324 determination methods............................................. 317 GC/ MS determination comparison................ 332–333 materials equipments......................................................... 317 reagents....................................................... 317–318 olive oils............................................................ 319–321

ipidomics 544  LIndex



 

Matrix-assisted laser desorption/ionization time-of-flight (Continued) composition........................................................ 320 fingerprinting......................................320–321, 324 oil extraction............................................... 319–320 quantitative information......................319–320, 322 pomegranate and soybean oil............................ 325–327 MALDI analysis......................................... 325, 327 plant material and extraction.............................. 325 TAG profiling............................................. 325–326 theoretical molecular mass........................................ 319 vegetable oils......................................318–319, 327–332 Meibomian glands (MG)............................................... 221 Membrane fluidity.............................................................. 4 Membrane lipidomics and unsaturated fatty acids geometry cell membrane isolation, cultures...................... 406–407 flame ionization detector (FID)............................... 403 gas chromatography separation......................... 405, 408 isomerization, cell membranes.......................... 401–402 liposomes, biomimetic chemistry membrane models bilayer arrangement............................................ 396 free radical processes........................................... 395 heterogeneous system reactivity.................. 396–397 isomerization...................................................... 398 multilayer vesicles (MLV).................................. 396 LUVET cis–trans isomerization regioselectivity........ 398–399 irradiation conditions, isomerization.......... 404–405 preparation.................................................. 403–404 materials equipments................................................. 402–403 reagents and supplies.......................................... 403 photolysis reactor...................................................... 402 radical-catalysed cis–trans isomerization................... 393 reactive positions.............................................. 392–393 tandem protein–lipid damage........................... 399–400 TLC, argentation technique..................................... 406 trans isomer detection analytical methods.............................................. 394 membrane compartment..................................... 395 trans lipids.................................................. 394–395 unilamellar vesicles injection methodology preparation..................... 404 micrography................................................ 407–408 Membrane lipids.................................................... 514–515 functions........................................................... 514–515 lipids and lipid-soluble compounds.................. 516–517 localisation and distribution............................. 513–514 materials equipment................................................... 522–523 reagents and supplies.......................................... 523 Membrane proteins.......................................................... 20 Membrane rafts.............................................................. 261 Methylcarbonate (CH3OCO2–) anions..................... 34, 46

Microsomes. See also Mammalian fatty acid elongases control studies................................................... 380–381 enzymatic assay......................................................... 380 extraction and analysis.............................................. 380 preparation................................................................ 379 Mitochondria DNA............................................................................. 3 lipid composition........................................................ 10 of C57BL/6J mouse brain.................................... 11 purification............................................................. 7 NS mitochondria.................................................... 7 Syn mitochondria............................................... 7–8 purity...................................................................... 9 protein markers, on Western blots........................ 10 mass content, of lipids of anionic lipids.............................................. 11–13 of weak anionic lipids........................................... 14 weak polar lipids............................................. 15–16 Mitochondrial lipidome..................................................... 3 Mitochondrial membrane lipids......................................... 4 Mono acyl glycerols.................................................. 35, 222 1-Monomyristoyl glycerol.............................................. 227 MS/MS conditions........................................................... 37 MS/MS scan mode.......................................................... 35 MSn analyses.................................................................. 223 Multiple reaction monitoring (MRM)........78, 79, 273, 288 Murine brains................................................................. 191 lipid extract fractions mass spectra of.................................................... 196 negative ion mode mass spectra.......................... 197 Mycobacterial lipidome.................................................... 90 Mycobacterial lipid profiling............................................ 91 cell culture and preparation.................................. 92–93 extraction methods..................................................... 96 M. bovis BCG, HSQC spectrum 2D HSQC lipid profile........................................ 97 2D-HSQC mycolic acid profile............................ 99 polar and apolar lipid profiles............................. 100 mycolic acid pool, preparation.................................... 95 to probe gene function................................................ 98 total free lipid pool, extraction and analysis.......... 93–95 Mycobacterium tuberculosis..................................................90 Myelin membranes............................................................. 4

N Nano-electrospray ionization (nESI).......................... 37, 38 N-Cyclohexyl-N′′-dodecanoic acid urea (CUDA)....................................................... 172 Nephritis......................................................................... 162 nESI-MS.......................................................................... 37 Neutral lipids and phospholipids mass spectrometry.................................................... 369 TLC separation................................................ 370–372 Neutral loss scan (NLS)............................................. 43, 44

Lipidomics 545 Index   

Neutral loss scans mode analysis for class-specific identification of lipid molecular species................................ 40–41 for global lipid analysis............................................... 39 of rat retina DG and TG lipids.......................59, 61, 62 of rat retina GPEtn and GPIns lipids.............48, 52, 54 of rat retina GPSer lipids...................................... 54–59 of rat retina SM and GPCho lipids............................ 39 negative ion analysis................................. 39, 42–44 positive ion analysis........................................ 44–48 Neutral scanning, for PE................................................ 289 Nitrocellulose................................................................. 259 Nitrogen......................................................................... 136 NMR software................................................................. 91 NMR spectroscopy..................................................... 90, 91 Nonpolar lipids............................................................... 222 Nonpolar nonwater-soluble solvents.............................. 137 Nonsynaptic (NS) mitochondria........................................ 3 Normal phase HPLC (NP HPLC)................................ 223 Nuclear magnetic resonance spectrometry...................... 222

O Obesity............................................................................. 20 Octadecyl (C18) acyl chains........................................... 137 Oleamide................................................................ 215, 227 Olive oils, TAG profiling composition.............................................................. 320 fingerprinting............................................320–321, 324 oil extraction..................................................... 319–320 quantitative information............................319–320, 322 Omega-3 fatty acids................................................. 19, 262 Online ozonolysis methods, double bond position determination anionic ether phospholipid, human lens................... 438 collision induced dissociation (CID)................ 413–414 OzESI–MS ESI–MS spectrum, GPA............................ 428–429 external ozone generation....................416, 419–421 GPCho isomeric standards......................... 429–431 human lens extract...................................... 431–433 ion source modification.............................. 420–421 neutral losses/gains..................................... 430–431 performance................................................ 421–422 in situ corona discharge........................416, 418–419 total ion current (TIC)....................................... 429 OzID................................................................ 416–417 GPCho isomeric standards......................... 433–434 instrument modification..................................... 426 ion trap and setting..................................... 424–425 PEEKsil restriction..................................... 423–424 positive and negative ion examples............. 434–437 preparation.................................................. 422–423 trimethylamine neutral loss................................. 426 position difference.................................................... 414

precursor lipid ion vs. ozonolysis product ion......................................................... 438–439 reagents and supplies........................................ 417–418 safety precautions............................................. 426–428 Optic nerve head (ONH)............................................... 255 Oxo-fatty acids............................................................... 161 5-Oxo-7-glutathionyl-8,11,14-eicosatrienoic acid (FOG7)......................................................... 168 Oxylipin stereoisomers, analysis of................................. 164 OzESI–MS. See Ozone electrospray ionization–mass spectrometry (OzESI–MS) OzID. See Ozone-induced dissociation (OzID) Ozone electrospray ionization–mass spectrometry (OzESI–MS) ESI–MS spectrum, GPA.................................. 428–429 external ozone generation..........................416, 419–421 GPCho isomeric standards............................... 429–431 human lens extract............................................ 431–433 ion source modification.................................... 420–421 neutral losses/gains........................................... 430–431 performance...................................................... 421–422 in situ corona discharge..............................416, 418–419 total ion current (TIC)............................................. 429 Ozone-induced dissociation (OzID)................................ 27 GPCho isomeric standards............................... 433–434 instrument modification........................................... 426 ion trap and setting........................................... 424–425 PEEKsil restriction........................................... 423–424 positive and negative ion examples................... 434–437 preparation........................................................ 422–423 trimethylamine neutral loss...................................... 426

P 1-Palmitoyl-2-oleoylphosphatidic acid (POPA)................................................. 239, 240 1-Palmitoyl-2-oleoyl-phosphatidylcholine (POPC).........................................239, 240, 242 1-Palmitoyl-2-oleoyl-phosphatidylglycerol (POPG).........................................239, 240, 242 Para-nitroaniline (PNA)................................................. 106 Peak–area ratio............................................................... 282 PE-plasmalogens, detection using RPLC–ESIMS/MS........................................ 289 data-dependent scanning.................................... 292 detection, of PA and PS...................................... 297 2D-map of phospholipids................................... 293 identification of phospholipids........................... 292 mass spectrometry conditions............................. 292 molecular species of PC mixture, identified by.................................................. 295 MS/MS spectra of [M + HCO2]-ions of PCs from...................................................... 294 search engine, for Identification of................... 296–297 by ultra performance LC (UPLC)

ipidomics 546  LIndex



 

PE-plasmalogens, detection (Continued) diacyl-and 1-O-alk-1′-enyl-2-acyl-PE, identification of............................................ 307 HPLC and UPLC mobile phases....................... 303 LC separation experiment.......................... 303–304 mass spectrometry, conditions............................ 304 molecular species of PE identified.............. 308–310 MS/MS analysis......................................... 304–306 quantitative analysis.................................... 306, 311 sample procedure................................................ 303 specific detection of............................................ 306 Peptidoglycan................................................................... 90 Peroxisomal enzyme....................................................... 128 PGO. See Pomegranate oil Phagocytosis................................................................... 130 Pharmacologic assays........................................................ 73 Phenolicglycolipid (PGL)................................................ 97 Phophatidylcholine (PCho)............................250, 255, 256 Phosphatidic acid..............................................21, 103, 288 Phosphatidylcholines................. 20, 118, 128, 129, 133, 289 Phosphatidylethanolamine (PE)-plasmalogens.............. 288 Phosphatidylethanolamines.......................20, 118, 128–130 contour plot of.......................................................... 153 separation of..................................................... 155–156 Phosphatidylglycerol...................................21, 67, 112, 289 Phosphatidylinositol........................... 21, 35, 118, 128, 129, 132, 190, 192–195, 289 calibration curves...................................................... 198 Phosphatidylinositol-4,5-bisphosphate (PIP2).............. 190 Phosphatidylinositols..................................................... 128 Phosphatidylserines.................... 21, 74, 118, 128–130, 132, 141, 152, 288 Phosphocholine-containing lipids.................................... 34 Phosphocholine-derived ions........................................... 43 Phosphoinositides.................................................. 103, 216 Phospholipase A2............................................................. 72 Phospholipase C............................................................... 73 Phospholipase D.............................................................. 73 Phospholipids (PLs)....................................4, 103, 222, 250 Phospholipids qualitative and quantitative analyses approaches................................................................ 287 LC–ESIMS.......................................287–289, 291–269 lipid molecular species...................................... 306, 311 materials column................................................................ 290 reagents....................................................... 290–291 PE-plasmalogens, LC–ESIMS/MS Bligh–Dyer method extraction........................... 303 HPLC and UPLC mobile phases....................... 303 mass spectrometry conditions............................. 304 MS/MS analysis......................................... 304–306 specific detection........................................ 306–310 phosphatidic acid and phosphatidylserine Bligh–Dyer method, extraction.......................... 298 butanol extraction............................................... 298

DEAE-cellulose column, pretreatment.................... 298–299 reverse phase LC–ESIMS.................................. 297 untargeted and global lipidomics data-dependent scanning.................................... 292 2D-map...................................................... 292–293 full-mass scanning.............................................. 292 HPLC mobile phases......................................... 292 LC separation experiment.................................. 292 mass spectrometry conditions............................. 292 molecular species................................................ 295 MS/MS spectra.................................................. 294 phospholipids identification, reverse phase LC–ESIMS/MS..................292–293, 295 survey methods................................................... 297 Phosphonate activity-based probes......................... 499–500 Phosphorus assay........................................................ 24–25 Photodiode array (PDA)................................................ 168 Pigmented epithelium (PE)........................................... 250 Plasmalogen-deficiency disorders................................... 289 Plasma/RBC contamination.......................................... 135 Platelet-activating factor.................................................. 73 Polar water-soluble solvent............................................. 137 Polycythemias................................................................. 131 Polyketides........................................................................ 89 Polysaccharides................................................................. 89 Polyunsaturated fatty acids (PUFA)..........83, 139, 271, 272 Pomegranate and soybean oil, TAG fingerprinting MALDI analysis.............................................. 325, 327 plant material and extraction.................................... 325 TAG profiling................................................... 325–326 Pomegranate oil (PGO)..........................325–326, 329–330 POPE, as negative ion.................................................... 114 POPG, as negative ion................................................... 114 Postsource decay (PSD) mass spectra............................. 117 Post-translational modifications (PTMs)............................. of proteins....................................................... 79 Precursor ion scan (PI)........................................... 450–451 Precursor lipid ion vs. ozonolysis product ion........ 438–439 Primary hepatocytes infection........................................ 385 Prostacyclins................................................................... 271 Prostaglandin glyceryl esters........................................... 167 Prostaglandins (PGs)...................................................... 161 Prostanoids biological sample collection and storage................................. 276–277 preparation.................................................. 274–275 data analysis.............................................................. 282 HPLC solvents......................................................... 279 LC/ESI–MS/MS analysis................................ 280–281 sample extraction and LC/ESI–MS/MS assay for........................................................ 273 sample preparation............................................ 277–278 sample reconstitution and......................................... 279 solid phase extraction.................................275, 278–279

Lipidomics 547 Index   

solutions preparation........................................ 274–276 calibration lines................................................... 276 Prostanoid synthases....................................................... 272 Protein kinase C............................................................... 20 Proteolytic enzymes.......................................................... 83 Proteolytic in-gel digestion destaining......................................................... 505–506 gel fragment preparation........................................... 505 peptide extraction............................................. 506–507 reduction and alkylation........................................... 506 washing..................................................................... 505 Proteomics approaches, in regulating lipid metabolism................. 73 reagents for................................................................. 75 Proton gradient................................................................... 4 Pulmonary surfactant alveolar type-II cells................................................. 515 lipids and lipid-soluble compounds.................. 516–517 localisation and distribution............................. 513–514 materials equipment................................................... 522–523 reagents and supplies.......................................... 523

Retinal pigment epithelial cells......................................... 33 Retinitis pigmentosa....................................................... 289 Reversed-phase-CAD approach..................................... 472 Reverse phase (RP) LC approaches................................ 288 Reverse phase LC coupled to electrospray ionisation (ESI)–MS/MS............................. 272 Reverse phase LC–ESIMS for lipidomics.............. 303–306 HPLC mobile phases............................................... 299 LC separation experiment........................................ 299 mass spectrometry conditions........................... 299–300 molecular species, of PA, PS and lyso-forms detected......................................................... 302 for PA and PS........................................................... 297 phospholipid analysis........................................ 300–301 quantitative analysis.......................................... 306, 311 sample pretreatment......................................... 298–299 sample procedure Bligh–Dyer method............................................ 298 with butanol........................................................ 298 Rheumatoid arthritis...................................................... 104

Q

Saponification, lipids.............................................. 382–383 Saturated fatty acids....................................................... 261 SCX column................................................................... 192 Secondary ion mass spectrometry (SIMS)...................... 249 Secretory PLA2.............................................................. 128 Selective reaction monitoring (SRM)..................... 162, 288 Serum lipolytic enzymes................................................. 128 Shotgun lipidomics....................................................... 4, 29 analysis of highly purified brain mitochondria by....... 16 to evaluate lipid content and....................................... 10 identification of DHA-containing glycerophospholipids...................................... 20 for individual lipid species in unfractionated lipid extracts.................................................... 34 procedure...................................................................... 9 and quantitation of cellular lipidomes....................... 136 and tandem MS............................................................ 4 using high-throughput platform which........................ 4 Signal-to-noise ratio....................................................... 111 Single ion monitoring (SIM)......................................... 231 Sodium chloride (NaCl)....................................................... for dissociation of lipids................................ 122 Sodium (Na+) ions............................................................ 46 Solid phase extraction (SPE).......................................... 273 Sphingolipid analysis....................................................................... 35 identification............................................................ 453 Sphingomyelins (SM)............... 35, 118, 129, 140, 141, 289 Sphingosine............................................................ 446, 456 Sphingosine 1-phosphate....................................... 446, 457 Sphingosines................................................................... 222 Sterol lipids.................................................................... 212

Q-Trap analysis of diacyl species of phosphatidylethanolamines........ 149 of phosphatidylethanolamines extracted from.......... 148 Quadrupole analyzer...................................................... 167 Quadrupole ion trap time of flight (QITTOF).............. 181 Quantitative analysis, of LOX products in bovine tissues............................................................ 168 Quasimolecular ions....................................................... 105

R Radical-catalysed cis–trans isomerization....................... 393 Radio-immunoassay techniques (RIA)................... 162, 272 Rat retina, lipid analysis by MS/MS scan mode DG and TG molecular species............................. 63–65 GPEtn species............................................................ 53 GPIns species............................................................. 55 GPSer molecular species............................................ 60 SM and GPCho molecular species....................... 49–51 Rat retina sphingomyelin (SM)........................................ 35 RBC membranes for medical investigations................................. 129–130 haematopoiesis and RBC lifetime.............. 130–131 incorporation of fatty acids into.................. 132–133 membrane lipids, in hemoglobinopathies................. 133–134 phospholipid synthesis, regulation.............. 131–132 remodeling of phospholipids.............................. 133 RBC phospholipids........................................................ 128 Retina............................................................................... 33 homogenates............................................................... 36

S

ipidomics 548  LIndex



 

Sterols GLC analysis...................................................... 368 Storage lipids.................................................................. 204 Strong cation exchange (SCX) gel.................................. 190 Sucrose gradient........................................................... 9, 16 Sustained off resonance irradiation (SORI)........... 169, 170 Synaptic (Syn) mitochondria.............................................. 3 Synaptic nerve terminal...................................................... 3 Synaptosomes..................................................................... 7

T TAG analysis, MALDI-TOF/MS......................... 318–319 Tandem hybrid MS/MS RBC phospholipids, analysis of CID mode.......................................................... 145 consequences, of RBC storage on....................... 152 on enhanced mass resolution mode.................... 145 ether-and diester subclasses, separation of.......... 152 ether-, diacyl-, and oxidized RBC PE........ 155–156 loss of neutral fragment...................................... 145 phosphatidylethanolamine, mass spectrum of........................................................... 146–147 Q-Trap analysis of.............................................. 148 separation, as function of fatty acid composition.................................................. 154 Tandem mass spectrometry analyses. See Neutral loss scans mode analysis Tandem protein–lipid damage................................ 399–400 Tear film (TF)................................................................ 222 Tear film lipid layer (TFLL)........................................... 222 Thiamine pyrophosphate (TPP).................................... 175 Thin-layer chromatography (TLC).........104, 289, 366–367 Thromboxanes (TXs)............................................. 161, 271 Time of flight (TOF)..................................................... 169 TLC. See Thin-layer chromatography Total fatty acid.................................................................. 16 Total ion chromatograms (TIC)..................................... 231 Total lipids........................................................................ 22 extraction.............................................................. 22–24 Total phospholipid concentration..................................... 24 Trans fatty acid methyl ester isomers separation............. 406 Trans isomer detection, unsaturated fatty acids geometry analytical methods.................................................... 394 membrane compartment.......................................... 395 trans lipids........................................................ 394–395 Tri-and di-acylglycerols.....................................59, 132, 222 Trifluoroacetic acid (TFA)............................................. 108 Trihydroxyeicosatrienoic acids (THETAs)..................... 170 Trimethylamine (TEA).................................................. 137 Triolein........................................................................... 227 Triple quadrupole instruments....................................... 223 Triple quadrupole mass spectrometer....................8, 35, 223 Triple quadrupole (TQ) spectrometer............................ 141 Trypsin............................................................................. 83

TSQ Quantum Ultra triple quadrupole mass spectrometer................................................... 37 Two-dimensional 1H-13C heteronuclear single quantum coherence (HSQC) NMR............... 91 Type 2 diabetes................................................................. 20

U Ultra performance liquid chromatography (UPLC)................................................ 162, 303 Unfractionated lipid extracts............................................ 34 Unsaturated lipids double bond position determination anionic ether phospholipid, human lens................... 438 collision induced dissociation (CID)................ 413–414 OzESI–MS ESI–MS spectrum, GPA............................ 428–429 external ozone generation....................416, 419–421 GPCho isomeric standards......................... 429–431 human lens extract...................................... 431–433 ion source modification.............................. 420–421 neutral losses/gains..................................... 430–431 performance................................................ 421–422 in situ corona discharge........................416, 418–419 total ion current (TIC)....................................... 429 OzID 416–417 GPCho isomeric standards......................... 433–434 instrument modification..................................... 426 ion trap and setting..................................... 424–425 PEEKsil restriction..................................... 423–424 positive and negative ion examples............. 434–437 preparation.................................................. 422–423 trimethylamine neutral loss................................. 426 position difference.................................................... 414 precursor lipid ion vs. ozonolysis product ion......................................................... 438–439 reagents and supplies........................................ 417–418 safety precautions............................................. 426–428 UV absorption of olefinic residues.................................. 104 UV detector.................................................................... 104

V Variants of, APOA1_HUMAN....................................... 79 Vegetable oils. See Fatty acid profiles and TAGs determination, MALDI-TOF/MS fingerprinting Vinyl ether...................................................................... 139 Vitamin E, spatial distribution biological roles.................................................. 515–516 lipids and lipid-soluble compounds.................. 516–517 localisation and distribution............................. 513–514 materials equipment................................................... 522–523 reagents and supplies.......................................... 523 Voltage dependent, ion channel.......................................... 4

Lipidomics 549 Index   



W

Z

Warburg theory of cancer................................................. 17 Wax esters....................................................................... 222 Western analysis............................................................... 73 Wet eye disease............................................................... 248

Zellweger syndrome....................................................... 289 Zwitterionic phospholipid................................................ 30