Human Chorionic Gonadotropin (hCG)
Human Chorionic Gonadotropin (hCG) Laurence A. Cole Stephen A. Butler
AMSTERDAM •...
37 downloads
280 Views
8MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Human Chorionic Gonadotropin (hCG)
Human Chorionic Gonadotropin (hCG) Laurence A. Cole Stephen A. Butler
AMSTERDAM • BOSTON • HEIDELBERG • LONDON • NEW YORK • OXFORD • PARIS • SAN DIEGO SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO
Elsevier 32 Jamestown Road London NW1 7BY 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA First edition 2010 Copyright © 2010 Elsevier Inc. All rights reserved No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangement with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress ISBN: 978-0-12-384907-6 For information on all Elsevier publications visit our website at www.elsevierdirect.com This book has been manufactured using Print On Demand technology. Each copy is produced to order and is limited to black ink. The online version of this book will show color figures where appropriate.
Acknowledgments
This book was written by 21st century scientists in honor of the forefathers and pioneers of hCG research. Without them, we would have never been able to discover the wondrous molecules that we know today. I honor and dedicate this book to those who discovered hCG in the the 1920s: Toyoichi Hirose of the Osaka Medical College in Japan, Selamar Aschheim of the University of Berlin in Germany, and Bernhard Zondek of the Berlin-Spandau Hospital in Germany. I also wish to dedicate this book to those 25 people that I consider to be the pioneers of hCG research, without whom there would be little to write about: Mario Ascoli, PhD University of Iowa, Iowa City, IO. Om Bahl, PhD University at Buffalo, Buffalo, NY. Peter Berger, PhD Austrian Academy of Sciences, Innsbruck, Austria. Jean-Michel Bidart, PhD Institute Gustave-Roussy, Villejuif, France. Steven Birken, PhD National Institute of Health, Bethesda, MD. Irving Boime, PhD Washington University, St Louis, MO. Glenn Braunstein, MD Cedars-Sinai Medical Center, Los Angeles, CA. Robert Canfield, MD Columbia-Presbyterian Hospital, New York, NY. Timothy Chard, MD St. Bartholomew’s Hospital, London, UK. Maria Dufau MD, PhD National Institutes of Health, Bethesda MD. John Fiddes, PhD Cold Spring Harbor Laboratory, Cold Sprong Harbor, NY. Robert Hussa, PhD Medical College of Wisconsin, Milwauke WI, Sunnyvale, CA USA.
xii
Acknowledgments
Ray Iles, PhD Middlesex University, London, UK. Akira Kobata, PhD Tokyo University, Japan. Klauss Mann, MD University of Essen, Germany. Ryuichiro Nishimura, MD Hyogo Cancer Center, Akashi, Japan. Bruce Nisula, MD National Institutes of Health, Bethesda, MD. Robert Norman, MD University of Adelaide, Australia. William Odell, MD, PhD Utah Medical Center, Salt Lake City, UT. David Puett, PhD University of Georgia, Athens, GA. C.V. Rao, PhD Herbert Wertheim College of Medicine, Florida International University, Miami, FI, USA. Raymond Ruddon, MD, PhD University of Michigan, Ann Arbor, MI. Ulf Stenman, PhD Helsinki University Central Hospital, Finland. Judith Vaitukaitis, MD National Institutes of Health, Bethesda, MD. Bruce Weintraub, MD National Institutes of Health, Bethesda, MD. Stephen Butler, PhD, Department of Biomedical Sciences, Middlesex University, London, was an invaluable sub-author of this book. He wrote multiple chapters and aided in polishing the book’s content. Butler’s efforts were essential to the assembly and management of this book. Thank you to Camille Sapienz for editing the grammar, punctuation, and diction of this book. I also thank all the other authors who contributed essential chapters to this book: Robert Hussa, PhD; Ulf Stenman, PhD; Akira Kobata, PhD; C.V. Rao, PhD; Francis Byrn, MD; and Ervin Jones, MD. Once again, I say thank you to everybody involved.
About the Author
In 1971, Larry Cole began studying medicine in England. A stroke in 1975 put him in a coma for multiple months and left him with amnesia and severe brain damage. He was forced to abandon his career in medicine as a result of this incident. After his recovery, he moved to Israel in 1976, then to the United States in 1977. He managed to “program” the other side of the brain. He taught himself to learn and memorize science and medicine again and in 1978, Cole attended the Medical College of Wisconsin, Milwaukee, to work on a PhD in Biochemistry. In 1981, Larry obtained his PhD while studying cancer cell hCG with Robert Hussa, PhD. During this time, Larry also worked with Roland Pattillo, MD, on gestational trophoblastic disease cases. In the same year, he attended the first international symposium on gestational trophoblastic diseases and has attended all 15 consecutive biannual symposia since then. In 1983, Larry completed a postdoctoral fellowship with Raymond Ruddon, PhD, at the University of Michigan, Ann Arbor where he continued to specialized in hCG. His experiences working with Dr. Ruddon are what inspired him to study hCG carbohydrate structure. After his postdoctoral fellowship, Larry joined the faculty of the University of Michigan. After 3 years, he took a position at Yale University, where he spent 13 years as part of the Obstetrics and Gynecology faculty. He slowly advanced at Yale University from Assistant Professor to Associate Professor, then to Full Professor. While at Yale University, Larry’s research purely focused on hCG, investigating hCG and gestational trophoblastic disease, hCG as a tumor marker, and oligosaccharides and hCG. It was at Yale University that he studied the structure of choriocarcinoma hCG, which ultimately led to his discovery of hyperglycosylated hCG and to new diagnostic protocols in gestational trophoblastic diseases. In 1999, Larry moved to the University of New Mexico, Albuquerque, as a tenured Full Professor. While at the University of New Mexico, he started the U.S. government CLIA-endorsed program called the USA hCG Reference Service, which globally advises scientists who research hCG and physicians who treat gestational trophoblastic disease and patient with persistent low levels of hCG. With the Reference Service came his endowment by appreciative false positive hCG patients. In 2004, he became the Howard and Friedman Distinguished Professor of Obstetrics and Gynecology. Today, 29 years after receiving his PhD, Dr. Cole still specialize in hCG research. His clinical interests lie with studying and researching gestational trophoblastic
xiv
About the Author
diseases. In 1987, Larry’s PhD advisor, Robert Hussa, wrote the first book on hCG. It is with great dignity that Larry follows in the footsteps of his advisor and writes the second specialized book on hCG in 2010. Laurence A. Cole, PhD The Howard and Friedman Distinguished Professor of Obstetrics and Gynecology Director, USA hCG Reference Service Chief, Women’s Health Research Departments of Obstetrics and Gynecology, and Biochemistry and Molecular Biology
Abbreviations
ACT: ActD: Ala: Arg: Asn: ATF1: BEP: BSO: cAMP: CG: CG-H: CHO: CKGF: CNS: COH: COS: CRE: CREBP: CTP: Cys: DSA: DSD: EDTA: EGF: EMA-CO: EMA-EP: FDA: FIGO: free : free : free -CTP:
-subunit activator element actinomycin D chemotherapy alanine arginine asparagine activating transcription factor 1 chemotherapy regimen combining bleomycin, etoposide, and cisplatin bilateral salpingo-oophorectomy cyclic adenosine monophosphate chorionic gonadotropin hyperglycosylated chorionic gonadotropin Chinese hamster ovary cystine-knot growth factor central nervous system controlled ovarian hyperstimulation controlled ovarian stimulation cAMP response element cAMP response element binding protein human chorionic gonadotropin -subunit C-terminal peptide cysteine Datura stramonium agglutinin downstream domain ethylene diamine tetra-acetic acid epidermal growth factor alternating weekly administration of chemotherapy agents etoposide, methotrexate, and actinomycin D, with cyclophosphamide and onvocin (vincristine) alternating weekly administration of chemotherapy agents etoposide, methotrexate, and actinomycin D, with etoposide and cisplatin Food and Drug Administration, USA Federation Internation of Gynecologic Oncology free -subunit of human chorionic gonadotropin free -subunit of human chorionic gonadotropin free -subunit missing the -subunit C-terminal peptide
xvi
FSH: Fuc: FUT: Gal: GalNAc: GATA: GlcNAc: Glu: Gly: GnRH: GnT: GTD: GTN: hCG: hCG: hCG: hCG-CTP: hCG-H: hCG-H: hCGp: hLH: hMG: ICE: IFCC: IRP: IRR: IS: ISOBM: IU: IUI: IU/l: IVF: Leu: LIF: LH: LUF: Lys: Man: MCSF: Met: MoM: MCW: mIU: ml:
Abbreviations
follicle stimulating hormone fucose fucosyltransferase galactose N-acetylgalactosamine DNA sequence and DNA coding N-acetylglucosamine glutamic acid glycine gonadotropin releasing hormone N-acetylglucosaminyltransferase gestational trophoblastic disease gestational trophoblastic neoplasm human chorionic gonadotropin free -subunit of human chorionic gonadotropin free -subunit of human chorionic gonadotropin human chorionic gonadotropin -subunit C-terminal peptide hyperglycosylated human chorionic gonadotropin hyperglycosylated human chorionic gonadotropin free -subunit pituitary hCG human luteinizing hormone human menopausal gonadotropins chemotherapy regimen combining ifosamide, carboplatin, and etoposide International Federation of Clinical Chemistry international reference preparation international reference reagent International Standard International Association of Biological Markers international units intrauterine insemination international units per liter in-vitro fertilization leucine leukemia inhibitory factor luteinizing hormone luteinized unruptured follicle syndrome lysine mannose macrophage colony stimulating factor Methionine multiples of the median Medical College of Wisconsin milli-international units milliliter
Abbreviations
MMP: Mtx: NCBI: NeuAc: NGF: ng/ml: NIH: Ob/Gyn: OHSS: OI: Oop: OTC: PAPP-A: PCOS: PCR: PDGFB: pg/ml: PIH: PKA: PlGF: POC: Pro: rhCG: RIA: RR: RT-PCR: SAB: Ser: SP1: ST3Gal: ST6Gal: TAH: TGF: Thr: TJA-I: TJA-II: TSE: TSH: TVH: URE: USD: Val: VEGF: WHO:
xvii
metalloproteinase methotrexate chemotherapy The National Center for Biotechnology Information N-acetylneuraminic acid or sialic acid nerve growth factor nanograms per millilter National Institutes of Health obstetrics and gynecology ovarian hyperstimulation syndrome ovulation induction oophorectomy over-the-couter pregnancy-associated plasma protein-A polycystic ovary syndrome polymerase chain reaction platelet-derived growth factor B picograms per milliliter pregnancy-induced hypertension phosphokinase A placental growth factor point-of-care proline recombinant human chorionic goadotropin radioimmunoassay reference reagent reverse transcription polymerase chain reaction spontaneous abortion or miscarriage serine selective promoter factor 1 sialyltransferase-3-galactose sialyltransferase-6-galactose transabdominal hysterectomy transforming growth factor threonine Trichosanthes japonica agglutinin-I Trichosanthes japonica agglutinin-II tissue/trophoblast-specific element thyroid stimulating hormone transvaginal hysterectomy upstream regulatory element upstream domain valine vascular endothelial growth factor World Health Organization
Contributors
Stephen A. Butler, PhD Biomedical Sciences, Middlesex University, London, UK
Ervin E. Jones, PhD, MD Genetics and IVF Institute, Fairfax, Virginia, USA
Francis W. Byrn, MD Obstetrics and Gynecology, University of New Mexico, Albuquerque, NM, USA
Akira Kobata, PhD The Noguchi Institute, Tokyo, Japan
Laurence A. Cole, PhD USA hCG Reference Service, Albuquerque, NM, USA Robert O. Hussa, PhD Medical College of Wisconsin, Milwauke WI, Sunnyvale, CA, USA Ray K. Iles, PhD Middlesex University, London, UK
Carolyn Y. Muller, MD USA hCG Reference Service, Albuquerque, NM, USA C.V. Rao, PhD Herbert Wertheim College of Medicine, Florida International University, Miami, FL, USA Ulf-Håkan Stenman, PhD Helsinki University Central Hospital, Finland
1 The Expanding World of hCG Robert O. Hussa Medical College of Wisconsin, Milwauke WI, Sunnyvale, CA, USA
More than four decades ago, in 1969, the door to the world of hCG opened for me when I joined Roland Pattillo’s research team in the Department of Ob/Gyn at the Medical College of Wisconsin (MCW) in Milwaukee. Pattillo established the hCGsecreting BeWo choriocarcinoma cell line in 1966 in George Gey’s laboratory at Johns Hopkins University [1]. Pattillo literally carried the cell line in a culture flask filled with culture fluid in his shirt pocket during his move from Baltimore to Milwaukee. Pattillo, along with Eleanor Delfs and Richard Mattingly, comprised the trophoblast disease clinical team that migrated from Johns Hopkins University to Milwaukee. Mattingly was chairman of the Department of Ob/Gyn at MCW and editor-in-chief of Obstetrics & Gynecology. From my desk in the laboratory (the room adjoining Pattillo’s cell culture laboratory), I would observe Eleanor Delfs personally perform hCG extractions on serum samples from her hydatidiform mole and choriocarcinoma patients prior to injecting the samples into rats, per her uterineweight bioassay—which was published the year I was born [2]. The emphasis of the Department of Ob/Gyn at MCW on treatment of patients with trophoblast disease and the study of hCG in patients, as well as in hCG-secreting cell lines, led to the need to quantitatively measure hCG in large numbers of samples. Pattillo’s thriving cell culture laboratory soon established a second hCG-secreting trophoblast cell line, JAr [3]. In those years, the glycoprotein hormone field was filled with excitement, such as that engendered by the revelation that hCG, LH, FSH, and TSH all contained a common -subunit and a hormone-specific -subunit [4–6]. In addition, the technology of the new radioimmunoassays (RIAs) was making great strides, particularly with the -subunit-specific RIA created by Vaitukaitis and colleagues in Griff Ross’s laboratory at the National Institutes of Health [7]. Thus it was that our laboratory also evolved in the 1970s and 1980s and began to use RIAs to measure hCG and its - and -subunits in our research. Another significant body of research emanated out of labs such as that of Om P. Bahl, where the carbohydrate structure of hCG was characterized utilizing digestion with specific exoglycosidases [8]. More excitement for our lab came when Pattillo’s crackerjack tissue culture staff established the hCG-secreting CaSki and DoT cell lines from cervical carcinoma [9,10]. It needs to be emphasized that Roland Pattillo has provided a great service to science by willingly providing his trophoblastic and nontrophoblastic cell lines to all requesters over the past four decades. It was into this 1979 setting that a young biochemistry student at MCW, Laurence A. Cole, began his PhD research in my laboratory. Larry actually occupied the same Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00001-3 © 2010 Elsevier Inc. All rights reserved.
4
Human Chorionic Gonadotropin (hCG)
desk that I had used in my first years in the laboratory at MCW. Larry, a UK native, who greeted me each morning with a friendly “Good morning, sire,” was a hardworking, prodigiously productive, and insightful researcher, who occasionally found the time to bring special treats for the lab, such as fresh scones with clotted cream flown in from London, or some of his home-brewed beer. Larry’s dissertation on the characterization of the ectopic hCG secreted by the CaSki and DoT cervical carcinoma cell lines [11] led to his first publications [12–15] and established his early fascination with the oligosaccharide structure of hCG. This experience undoubtedly led to Larry Cole’s great contributions toward a better understanding of the many variant forms of hCG in pregnancy and cancer, as described in this book. It is indeed fitting that Larry has taken on the monumental task—akin to that of herding cats—of gathering the work of so many experts in the various areas of hCG (most of whom who have been contributing to the field for decades) and compiling it in this comprehensive compendium of reviews on all aspects of hCG, from its molecular and oligosaccharide structure and biosynthesis to its applications as a clinical marker and use as a therapeutic agent. A Web of Science search of this book’s areas of emphasis was conducted at the end of 2009 (Table 1.1). Despite the fact that no effort was made to cull through the articles and citations to eliminate redundant and irrelevant publications, the results give an excellent broad perspective of how the world of hCG has expanded over Table 1.1 Web of science hCG search, 1975–2009. Area of hCG research Biosynthesis Glycosylation Hyperglycosylated hCG Genes, mRNA Structure Degradation products Receptor, biological functions Clinical applications, infertility, pregnancy, nontrophoblastic disease Diet, sports, HIV Vaccines, cancer Antibodies Assays, tests Standards
Articles Articles Citations Citations Most Times 1975–1985 1986–2009 1975–1985 1986–2009 cited cited article 19 9 0
448 270 9
691 242 0
10,286 7,452 158
(23) (19) (32)
274 327 94
11 9 0 88
1,677 680 9 3,210
712 385 0 2,030
37,713 16,913 177 74,845
(33) (34) (35) (36)
551 602 99 2,975
265
6,753
4,032
99,162
(37)
738
0 56 80 36 2
154 1,475 1,014 1,195 72
0 971 1,217 370 66
2,698 23,784 20,081 19,380 1,122
(38) (39) (40) (40) (41)
310 556 363 363 78
The Expanding World of hCG
5
time. The number of publications in a given area is an indicator of the amount of activity in that area. Thus, the number of times a reference is cited by peers should correlate with the publications that experts value as being the most significant. In the years from 1975 to 1985 and 1986 to 2009, there were more publications and citations on clinical applications of hCG than in any other category; followed by receptor activity and biological functions of hCG. In contrast, there were no publications or citations prior to 1986 on hyperglycosylated hCG, degradation products of hCG, or diet/sports/HIV, and only two publications devoted to hCG standards and reference preparations. The lack of publications in these areas of emphasis underscores the timeliness and relevance of the reviews for each of these areas in this book. After the 1980 reviews [16,17], various aspects of hCG have been reviewed more recently, including three-dimensional structure [18,19], gene expression and cloning [20,21], biosynthesis [22], glycosylation [23], immunochemistry [24], clinical measurement [25–28], receptors [29,30], and infertility [31]. The recently emerging category that has received the most attention (as measured by the 2698 publications since 1985, compared to no prior publications) is that of diet, sports, and HIV. Similarly, the Varki review on the biological roles of oligosaccharides was cited in an astonishing 2975 references [36]; the next most cited reference in hCG literature (738 citations) was on placental implantation, by Cross et al. [37]; the third most cited publication (602 citations) was on the crystal structure of hCG, by Lapthorn et al. [34]. The yearly trend of publications and citations provides interesting information on the dynamics of activity in each of the areas of emphasis described in this book. In the areas of hCG biosynthesis, hCG genes/mRNA, hCG structure, and hCG receptor/ biological functions, there was a jump in research activity in 1991–1994 (Figures 1.1–1.4), with a corresponding increase in the number of citations starting in 1992 (Figures 1.5–1.8). In the area of hCG clinical applications, the publishing activity (Figure 1.9) has remained relatively constant (260–400 articles per year), whereas the number of citations has increased dramatically in linear fashion since 1991 (Figure 1.10).
Citations in each year 40 35 30 25 20 15 10 5 0
90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Years
Figure 1.1 Web of Science search for biosynthesis of hCG; publications (ordinate) by year (abscissa).
Published item in each year 110 100 90 80 70 60 50 40 30 20 10 0
86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Years
Figure 1.2 Web of Science search for hCG genes and mRNA; publications (ordinate) by year (abscissa). Published items in each year 70 60 50 40 30 20 10 0
86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Years
Figure 1.3 Web of Science search for hCG structure; publications (ordinate) by year (abscissa). Published items in each year 200 180 160 140 120 100 80 60 40 20 0
86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Years
Figure 1.4 Web of Science search for hCG receptor/biological functions; publications (ordinate) by year (abscissa).
Citations in each year 800 700 600 500 400 300 200 100 0
91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Years
Figure 1.5 Web of Science search for biosynthesis of hCG; citations (ordinate) by year (abscissa). 3600
Citations in each year
3200 2800 2400 2000 1600 1200 800 400 0
87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Years
Figure 1.6 Web of Science search for hCG genes and mRNA; citations (ordinate) by year (abscissa). Published items in each year 1500 1200 900 600 300 0
87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Years
Figure 1.7 Web of Science search for hCG structure; citations (ordinate) by year (abscissa).
8
Human Chorionic Gonadotropin (hCG) Citations in each year 6500 6000 5500 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0
87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Years
Figure 1.8 Web of Science search for hCG receptor/biological functions; citations (ordinate) by year (abscissa).
Published items in each year 400 360 320 280 240 200 160 120 80 40 0
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Years
Figure 1.9 Web of Science search for hCG and clinical applications, infertility, pregnancy, and nontrophoblastic disease; publications (ordinate) by year (abscissa).
Very similar trends were observed for the search areas of hCG assays/tests and vaccines/cancer (not shown). The newly emerging emphasis areas of diet/sports/HIV reveal sporadic publishing activity starting in 1991 (Figure 1.11) and a surge of citations beginning around 1996 (Figure 1.12). I hope the tapestry of hCG research trends from over the decades will provide an appropriate background for readers and guide them toward a favorite area of interest
The Expanding World of hCG
9 Citations in each year
10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0
86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Years
Figure 1.10 Web of Science search for hCG and clinical applications, infertility, pregnancy, and nontrophoblastic disease; citations (ordinate) by year (abscissa).
Published items in each year 13 12 11 10 9 8 7 6 5 4 3 2 1 0
87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Years
Figure 1.11 Web of Science search for hCG and diet/sports/HIV; publications (ordinate) by year (abscissa).
within the evolving world of hCG as covered in this comprehensive book. As mentioned earlier, several of these topics are being reviewed for the first time; others are continuously updated with new information emanating from the rapidly expanding field of hCG research. I am grateful and proud to have been part of this body of investigation since 1969.
10
Human Chorionic Gonadotropin (hCG) Citations in each year 250 200 150 100 50 0
91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Years
Figure 1.12 Web of Science search for hCG and diet/sports/HIV; citations (ordinate) by year (abscissa).
Acknowledgments I thank Jack Black, the senior librarian at El Camino Hospital Health Library and Resource Center, Mountain View, California, for his expert assistance with the literature searches on Web of Science, provided by Thomson Reuters.
References [1] Pattillo RA, Gey GO. The establishment of a cell line of human hormone-synthesizing trophoblastic cells in vitro. Cancer Res 1968;28:1231–6. [2] Delfs E. An assay method for human chorionic gonadotropin. Endocrinology 1941; 28:196–202. [3] Pattillo RA, Ruckert A, Hussa R, Bernstein R, Delfs E. The JAr cell line–continuous human multi-hormone production and controls. In Vitro 1971;6(Abstract):398–9. [4] Swaminathan N, Bahl OP. Dissociation and recombination of the subunits of human chorionic gonadotropin. Biochem Biophys Res Commun 1970;40:422–7. [5] Pierce JG. Eli Lilly Lecture: the subunits of pituitary thyrotropin–their relationships to other glycoprotein hormones. Endocrinol 1971;89:1331–44. [6] Bahl OP, Carlsen RB, Bellisario R, Swaminathan N. Human chorionic gonadotropin: amino acid sequence of the and subunits. Biochem Biophys Res Commun 1972; 48:416–22. [7] Vaitukaitis JL, Braunstein GD, Ross GT. A radioimmunoassay which specifically measures human chorionic gonadotropin in the presence of human luteinizing hormone. Am J Obstet Gynecol 1972;113:751–8. [8] Bahl OP, Marz L, Kessler MJ. Isolation and characterization of N- and O-glycosidic carbohydrate units of human chorionic gonadotropin. Biochem Biophys Res Commun 1978;84:667–76.
The Expanding World of hCG
11
[9] Pattillo RA, Hussa RO, Story MT, Ruckert ACF, Shalaby MR, Mattingly RF. Tumor antigen and human chorionic gonadotropin in CaSki cells: a new epidermoid cervical cancer cell line. Science 1977;196:1456–8. [10] Pattillo RA, Hussa RO, Ruckert ACF, Story MT, Mattingly RF. hCG-beta production by epidermoid carcinoma of the cervix. Endocrinol(Suppl) 1977;100(Abstract):179. [11] Cole LA. Studies with ectopic hCG: synthesis, glycosylation, isolation and characterization. PhD Dissertation, Milwaukee: Medical College of Wisconsin; 1982. [12] Cole LA, Hussa RO, Rao CV. Discordant synthesis and secretion of human chorionic gonadotropin and subunits by cervical carcinoma cells. Cancer Res 1981;41:1615–19. [13] Cole LA, Hussa RO. Use of glycosidase digested human chorionic gonadotropin and -subunit to explain the partial binding of ectopic glycoprotein hormones to Con A. Endocrinol 1981;109:2276–8. [14] Story MT, Cole LA, Hussa RO. A procedure using immobilized antibody for the isolation of the -subunit of human chorionic gonadotropin from the culture fluid of a -subunitsecreting nontrophoblastic cell line. J Clin Endocrinol Metab 1981;53:1090–5. [15] Cole LA, Birken S, Sutphen S, Hussa RO, Pattillo RA. Absence of the COOH-terminal peptide on ectopic human chorionic gonadotropin -subunit (hCG ). Endocrinol 1982; 110:2198–200. [16] Hussa RO. Biosynthesis of human chorionicgonadotropin . Endocr Rev 1980;1:268–94. [17] Hussa RO. The clinical marker hCG. New York, NY: Praeger Publishers; 1987. [18] Lustbader JW, Yarmush DL, Birken S, Puett D, Canfield RE. The application of chemical studies of human chorionic gonadotropin to visualize its three-dimensional structure. Endocr Rev 1993;14:291–311. [19] Wu H, Lustbader JW, Liu Y, Canfield RE, Hendrickson WA. Structure of human chorionic gonadotropin at 2.6 å resolution from MAD analysis of the selenomethionyl protein. Structure 1994;2:545–58. [20] Habener JF. Molecular cloning of hormone genes. Clifton, NJ: The Humana Press; 1987. [21] Jameson JL, Hollenberg AN. Regulation of chorionic gonadotropin gene expression. Endocr Rev 1993;14:203–21. [22] Merz WE. Biosynthesis of human chorionic gonadotropin: a review. Eur J Endocr 1996;135:269–84. [23] Van den Steen P, Rudd PM, Dwek RA, Opdenakker P. Concepts and principles of O-linked glycosylation. Crit Rev Biochem Mol Biol 1998;33:151–208. [24] O’Connor JF, Birken S, Lustbader JW, Krichevsky A, Chen Y, Canfield RE. Recent advances in the chemistry and immunochemistry of human chorionic gonadotropin: impact on clinical measurements. Endocr Rev 1994;15:650–83. [25] Chard T. Pregnancy tests: a review. Hum Reprod 1992;7:701–10. [26] Duffy MJ. Clinical uses of tumor markers: a critical review. Crit Rev Clin Lab Sci 2001; 38:225–62. [27] Reis FM, D’Antona D, Petraglia F. Predictive value of hormone measurements in maternal and fetal complications of pregnancy. Endocr Rev 2002;23:230–57. [28] Stenman UH, Alfthan H, Hotakainen K. Review. Human chorionic gonadotropin in cancer. Clin Biochem 2004;37:549–61. [29] Ascoli M, Fanelli F, Segaloff DL. The lutropin/choriogonadotropin receptor, a 2002 perspective. Endocr Rev 2002;23:141–74. [30] Kleinan G, Krause G. Thyrotropin and homologous glycoprotein hormone receptors: structural and functional aspects of extracellular signaling mechanisms. Endocr Rev 2009;30:133–51.
12
Human Chorionic Gonadotropin (hCG)
[31] Macklon NS, Stouffer RL, Giudice LC. Fauser BCJM. The science behind 25 years of ovarian stimulation for in vitro fertilization. Endocr Rev 2006;27:170–207. [32] Elliott MM, Kardana A, Lustbader JW, Cole LA. Carbohydrate and peptide structure of the - and -subunits of human chorionic gonadotropin from normal and aberrant pregnancy and choriocarcinoma. Endocrine 1997;7:15–32. [33] Kumar TR, Wang Y, Lu N, Matzuk MM. Follicle stimulating hormone is required for ovarian follicle maturation but not male fertility. Nature Genetics 1997;15:201–4. [34] Lapthorn AJ, Harris DC, Littlejohn A, Lustbader JW, Canfield RE, Machin KJ, et al. Crystal structure of human chorionic gonadotropin. Nature 1994;369:455–61. [35] Rotmensch S, Cole LA. False diagnosis and needless therapy of presumed malignant disease in women with false-positive human chorionic gonadotropin concentrations. Lancet 2000;355:712–15. [36] Varki A. Biological roles of oligosaccharides—all of the theories are correct. Glycobiol 1993;3:97–130. [37] Cross JC, Werb Z, Fisher SJ. Implantation and the placenta—key pieces of the development puzzle. Science 1994;266:1508–18. [38] Hansen JE, Lund O, Tolstrup N, Gooley AA, Williams KL, Brunak S. NetOglyc: prediction of mucin type O-glycosylation sites based on sequence context and surface accessibility. Glycoconj J 1998;15:115–30. [39] Mead GM, Stenning SP, Cook PInternational Germ Cell Cancer Consensus Group. International germ cell consensus classification: a prognostic factor-based staging system for metastatic germ cell cancers. J Clin Oncol 1997;15:594–603. [40] Jayasena SD. Aptamers: an emerging class of molecules that rival antibodies in diagnostics. Clin Chem 1999;45:1628–50. [41] Cole LA, Kardana A. Discordant results in human chorionic gonadotropin assays. Clin Chem 1992;38:263–70.
2 History and Introduction to Human Chorionic Gonadotropin (hCG): One Name for at least Three Independent Molecules Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
In 1919, Japanese scientist Toyoichi Hirose was the first to demonstrate the effects of implanting fragments of human placenta on the ovaries and uteruses of rabbits [1,2]. Hirose’s work showed that there was a clear hormonal link between the placenta and the uterus. These findings were confirmed by Murata and Adashi in 1927 [3]. Around this time, the name human chorionic gonadotropin (hCG) was conceived: chorion is Latin for “placenta” and the hormone is produced by the placenta, hence chorionic; gonadotropin because the hormone is tropic, acting on female gonad tissue (ovaries), promoting steroid-induced actions. In 1927, Ascheim and Zondek demonstrated that the blood and urine of pregnant women contained a gonad-stimulating substance [4]. They showed that injecting this substance subcutaneously into intact immature female mice produced follicular maturation, luteinization, and hemorrhage into the ovarian stroma. These findings were confirmed by others [5,6] and the first hCG/pregnancy test was born [4–6]. These early tests primarily used urine to promote ovulation in rabbits [4–25], and were commonly referred to as the “rabbit” or Friedman test (Table 2.1). Over the next four decades, bioassays like the rabbit test were the only practical way to detect pregnancy or measure hCG [4–10]. In 1960, we saw the first antibody-based pregnancy test. The first antibody-based tests examined hemagglutination inhibition and latex agglutination [11,12]. These were insensitive slide tests that detected hCG at a concentration of 1000 mIU/ml or greater. In 1964, the competitive hCG radioimmunoassay (RIA) was invented [13–17] and revolutionized pregnancy testing. At last a test was available that could measure hCG as low as 5 mIU/ml and measure pregnancy as early as the day of a missed period. The invention of the RIA led to readily available pregnancy testing/hCG measurement at clinical laboratories throughout the world [13–17]. The initial RIAs were problematic because they used an antibody against hCG dimer and detected both hCG and luteinizing hormone (LH). The problem was that the -subunit of hCG was identical to the -subunit of LH, and the -subunit of hCG was 80% homologous with the -subunit of LH. Hence, the early hCG dimer RIA detected both hCG and LH and could only show pregnancy hCG and exclude LH by demonstrating a continual increase in hormone levels. In 1973, Vaitukaitis et al. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00002-5 © 2010 Elsevier Inc. All rights reserved.
14
Human Chorionic Gonadotropin (hCG)
Table 2.1 History of the Laboratory Pregnancy Test (the hCG and hCG assay). Year Test Description Described
Sensitivity
1927
Follicles promote corpora lutea in mice Urine hCG promotes ovulation in rabbits Urine hCG promotion of ovulation in toad Serum hCG increases rat uterine weight Urine causes hyperemia in rat ovaries Urine hCG promotes toad sperm erection Blood hCG hemagglutination inhibition Serum hCG promoted latex agglutination Serum hCG radioimmunoassay Serum hCG radioimmunoassay Serum hCG radioimmunoassay
5000 mIU/ml 5 days
5 mIU/ml 5 mIU/ml 5 mIU/ml
4 hours 4 hours 4 hours
Serum hCG radioimmunoassay Serum hCG radioimmunoassay Serum hCG radioimmunoassay Serum enzyme immunometric assay Serum enzyme immunometric assay Serum enzyme immunometric assay Serum chemiluminescent immunometric test
5 mIU/ml 5 mIU/ml 5 mIU/ml
4 hours 4 hours 4 hours
5 mIU/ml
30 minutes Sekiya et al., [23]
5 mIU/ml
30 minutes Batzer [24]
5 mIU/ml
30 minutes Joshi [25]
1 mIU/ml
30 minutes Barnard et al., [26]
1931 1934 1941 1943 1948 1960 1962 1964 1965 1967 1967 1968 1972 1981 1980 1981 1984
Test Time Reference
400 mIU/ml 3 days
Ascheim and Zondek [4] Friedman and Lapham [6] Shapiro and Zwarenstein [7] Delfs [8]
500 mIU/ml 2 hours
Kupperman [9]
3000 mIU/ml 2 hours
Galli-Mainini [10]
5000 mIU/ml 1.5 days 3000 mIU/ml 18 hours
1000 mIU/ml 1 day
Wide and Gemzell [11] 3000 mIU/ml 2 minutes Wide [12] Paul and Odell [13] Wilde CE et al. [14] Lunenfeld and Eshkol [15] Aono et al. [16] Rushworth et al. [17] Vaitukaitis et al., [18]
introduced the hCG test [18], a RIA pregnancy test using an antibody against the -subunit of hCG. The hCG test was the first hCG-specific RIA. Unlike its predecessor, which detected both hCG and LH, the hCG test measured hCG alone and did not detect LH [18]. This was an important distinction because LH has no relationship to pregnancy and need not be measured when testing for pregnancy. The hCG test RIA became the world standard for the next 20 years. Even today, in the age of immuno metric assays, both physicians and textbooks still describe hCG tests as hCG tests. The discovery of monoclonal antibodies in 1975 was paramount to the development of modern immunometric hCG tests [19]. Modern 2- or 3-antibody immunometric hCG assays were developed in 1981. With these assays came the concept of
History and Introduction to Human Chorionic Gonadotropin (hCG)
15
antibody enzyme labeling and high-sensitivity fluorimetric and spectrometric detection [20–25]. The advent of chemiluminescent and europium labeling, automation, and sensitive detection led to the rapid high-sensitivity hCG tests that are used today. As described later in this book (Chapter 19), at least 13 automated platforms use cartridges that rapidly and accurately measure serum hCG and other molecules via chemiluminescent methods. The history of the pregnancy test and the development of variations on the original rabbit test are outlined in Table 2.1. The principle behind the modern immunometric test begins with one or two antibodies (called the capture antibodies) binding to one or more antigen sites on hCG and its free -subunit. This binding immobilizes the hCG and free -subunit. A second antibody (called a tracer antibody) is labeled with a chemiluminescent labeling agent, and binds to a distant antigen site on hCG or its free -subunit. In so doing, the immobilized complex becomes what is labeled the capture antibody–hCG-tracer antibody complex. This complex can then be quantified; the amount of tracer antibody is linearly proportional to the amount of complex and the concentration of hCG. Dualantibody immunometric technologies are the principle of most modern physicians’ office point-of-care (POC) rapid pregnancy tests and home/over-the-counter (OTC) rapid pregnancy tests [25]. These tests use one antibody immobilized in the result window on the nitrocellulose device, and one antibody (labeled with a blue or gold dye) mixed with the serum or urine. A positive result is indicated by a line formed in the plastic window by the immobilized antibody–hCG-dye antibody complex. This book starts with a three-chapter introduction to hCG: Chapter 1 by Robert Hussa; Chapters 2 and 3 by Laurence Cole. In this book, we present articles describing every aspect of hCG assays, hCG antibodies, laboratory hCG tests, POC and OTC hCG tests, false-positive hCG tests, hCG assay specificity, and hCG standards. Chapter 18, “Antibodies for intact hCG, total hCG, free subunits, glycosylation variants, and hCG fragments,” by Laurence Cole; Chapter 19, “Quantitative hCG assays,” by Laurence Cole; Chapter 20, “False-positive hCG assays,” by Laurence Cole; Chapter 21, “Specificity of different hCG assays,” by Laurence Cole; Chapter 22, “Point-of-care pregnancy tests,” by Laurence Cole; Chapter 23, “Over-the-counter pregancy tests,” by Laurence Cole, and Chapter 24, “hCG standards,” by Ulf Stenman. In this book, we examine all possible applications of hCG-related molecule assays. We explore the advantages and disadvantages of different commercial hCG assays, and the advantages and disadvantages of different hCG standards; examine the applications of hCG, hyperglycosylated hCG (hCG-H), and free tests in detecting pregnancies and monitoring the likelihood of pregnancy failures; examine the use of hCG and hCG-H in predicting Down Syndrome pregnancies and cases of maternal pre-eclampsia; and investigate the relationship of hCG-H and gestational trophoblastic disease, and of free and nontrophoblastic malignancies. “Background hCG,” by Laurence Cole; Chapter 26, “Pregnancy Testing,” by Laurence Cole; Chapter 27, “Predicting Spontaneously Aborted (SAB) Pregnancies,” by Laurence Cole; Chapter 28, “hCG, hyperglycosylated hCG, and free -subunit in predicting Down syndrome pregnancies and preeclampsia,” by Laurence Cole; Chapter 29, “hCG in Monitoring Gestational Trophoblastic Diseases,” by Laurence Cole; Chapter 30, “Use of Hyperglycosylated hCG as a Unique Marker of Gestational Trophoblastic Neoplasms,” by Laurence Cole and Carolyn Muller; Chapter 31, “Pituitary
16
Human Chorionic Gonadotropin (hCG) (A)
ion
lat
Pituitary gonadotrope cell
u Ov
hCGp
Ovarian graafian follicle (B) Human cancer cell th y ow nc Gr gna ali m
hCG-Hβ ne
ro
(C) Syncytiotrophoblast cell hCG
te es
og
Pr
Corpus luteal cell (D) Cytotrophoblast cell th ow n Gr asio ncy inv igna al m
hCG-H
Figure 2.1 Illustration of (A) endocrine action of hCGp, (B) cytokine or autocrine activities of cancer cell hCG-H, (C) endocrine action of hCG, and (D) the cytokine or autocrine actions of hCG-H.
hCG and Familial hCG,” by Laurence Cole, and Chapter 32, “hCG, free -subunit, and -core fragment as markers of malignancies,” by Laurence Cole. Hirose’s 1919 discoveries about the interaction of placental and ovarian tissues started a long chain of investigations into the molecule we now know as hCG [1,2]. Today, we look at hCG not as one hormone, but as a mixture of two hormones and two autocrines. The likely activities of hCG, pituitary hCG (hCGp), hCG-H, and hCG-H free -subunit (hCG-H) are illustrated in Figure 2.1. These discoveries have been slowly unearthed over many years of development in the laboratory. The real understanding of hCG has emerged from the combined efforts of numerous individuals and multiple research centers. Multiple variants of hCG have been discovered. A variant of hCG with doublesized sugar side-chains (hCG-H) is made by cytotrophoblast cells [16] and promotes invasion during implantation of pregnancy and malignancy as occurs in chorio carcinoma [17–19]. A free -subunit of hCG-H is made by nontrophoblastic neoplasms. This variant also has a role in cancer cell growth and malignancy [20–24]. Finally, there is hCGp, a sulfated form of hCG produced by the pituitary gland during the menstrual cycle [26]. hCGp supplements LH in promoting ovulation and
History and Introduction to Human Chorionic Gonadotropin (hCG)
17
progesterone production [26]. Thus, there are four separate forms of hCG: regular hCG, hCG-H, hCGp, and free hCG. Each is an individual molecule with a unique structure and independent functions. We examine the placental system that produces hCG and hCG-H in Chapter 3, “Introduction to pregnancy implantation, villous formation, and hemochorial placentation,” by Laurence Cole. We also examine the amino acid and peptide structures and biological functions of regular hCG, hCGp, hCG-H, and hCG-H; the molecular bio logy and genetics of hCG and the existence of multiple copies of the hCG -subunit gene in detail (see Chapters 4–10). Chapter 4, “The molecular genetics of hCG” by Stephen Butler; Chapter 5, “Structure, synthesis, secretion, and function of hCG,” by Laurence Cole and Stephen Butler; Chapter 6, “Comparison of the structures of hCG and Hyperglycosylated hCG,” by Laurence Cole; Chapter 7, “Structures of free - and -subunits,” by Laurence Cole; Chapter 8, “Glycobiology of hCG,” by Akira Kobata; Chapter 9, “Degradation products of hCG, hyperglycosylated hCG, and free -subunit,” by Laurence Cole, and Chapter 10, “Three-dimensional structure of hCG,” by Laurence Cole. In Chapters 11–14, we examine the regular hCG receptor and the clinical functions of both hCG and hCG-H: Chapter 11, “Paradigm shift on the targets of hCG actions,” by C.V. Rao; Chapter 12, “The hCG receptor,” by Laurence Cole; Chapter 13, “Biological function of hyperglycosylated hCG,” by Laurence Cole, and Chapter 14, “Biological function of the free -subunit: Expression and treatment target in cancer,” by Stephen Butler and Ray Iles. Basic purification methods are presented and the numerous cell lines that make hCG and its free -subunit are described in Chapters 33–35, “hCG and hyperglycosylated hCG purification from serum, urine, and culture fluids,” by Laurence Cole; Chapter 34, “Dissociation, desialylation, and cleavage of hCG,” by Laurence Cole, and Chapter 35, “hCG and free -subunit producing cell lines,” by Laurence A. Cole. The final three chapters of this book examine new research that brings hCG into the center of an evolution story as the source of human evolution (see Chapters 36 and 37). Chapter 36, “Evolution of hCG, evolution of humans, and evolution of human pregnancy disorders and cancer,” by Laurence Cole. The entire book is then summarized briefly: Chapter 37, “Summary: hCG a remarkable molecule,” by Laurence Cole. Finally, future experiments are considered for all areas of hCG science in Chapter 38, “hCG and the future,” by Laurence Cole. All told, every aspect of hCG, hCG-H, and free -subunit is carefully considered and reviewed in this book, including receptors, biological functions, and detections. The history of hCG has come a very long way since its beginnings in 1920 (see Figure 2.2). First, the amino acid sequence of the hCG subunits was discovered in 1972 [27], and perfected in 1975 [28]. In 1980 came the discovery that the hCG -subunit evolved from LH -subunit [29] and that there were seven back-to-back genes coding for hCG -subunit [30]. The hCG dimer pathways of dissociation were then elaborated and the pathways of degradation of hCG -subunit in serum to urine -core fragment were uncovered [31]. The sugar side-chain sequences of the N- and O-linked oligosaccharides on hCG were determined [32,33], and the structure of the human hCG/LH receptor was resolved [34]. Studies concluded that hCG had functions well beyond the short-term progesterone-promotion function observed by Hirose [1,2]. In the 1990s, it was discovered that hCG functions throughout the length of pregnancy by enhancing angiogenesis in the uterine circulation, enhancing
18 Human Chorionic Gonadotropin (hCG)
Figure 2.2 Key steps in the history of hCG, hCGp, hCG-H, and hCG-H; four variants of hCG, separately produced with independent biological functions. (We apologize to any author missing from this synopsis who feels that other key research should be cited.)
History and Introduction to Human Chorionic Gonadotropin (hCG)
19
fetal nutrition [35], promoting the fusion of mononuclear cytotrophoblasts to multinuclear syncytiotrophoblast cells [36], and performing multiple other maternal systemic functions [35,36]. Finally, research has used X-ray crystallography to show the three-dimensional structure of the subunits of the deglycosylated molecule and its noncovalent interaction [37]. Every perspective of hCG research is detailed in Chapters 15–17: Chapter 15, “Use of hCG in reproductive dysfunction,” by Francis Byrn; Chapter 16 “hCG in assisted reproduction,” by Ervin Jones; and Chapter 17, “Illicit use of hCG in dietary programs and use to promote anabolism,” by Laurence Cole. Hyperglycolsylated hCG has a unique story unto itself, which begins with how this cytokine is formed. It was discovered [32,38] that choriocarcinoma cells make two hCG molecules, one regular hCG and one hCG-H. This led to the demonstration that hCG-H is not just produced in malignant trophoblastic disease, but also is the principal hCG form produced during implantation of pregnancy [39]. As shown in Figure 2.2, hCG-H is made by primitive cytotrophoblast cells, most notably the extravillous cytotrophoblast cells that drive invasion at implantation [40,41]. hCG-H is the invasion signal and invasion stimulus during both implantation and malignant trophoblastic disease. Without hCG-H, invasion cannot occur [42]. New research shows that hCG-H is critical to successful pregnancy implantation and that its deficiency apparently causes miscarriages [43]. Recent studies show that the evolutionary refinement of hCG and hCG-H were critical in the development of human beings [44]. All of this and very much more is presented in this book. The various forms of hCG seem to be involved in many important pathways. Modern research is showing that hCG-H is critical in cancer malignancy (Figure 2.2). Finding hCG-H in serum [45] and discovering the -subunit urine degradation product (-core fragment) were both critical to the cancer story [31]. hCG-H is a serum cancer tumor marker; -core fragment is a more potent tumor marker found exclusively in urine [46]. hCG-H is produced by most cancers and functions by promoting cancer cell growth and malignancy [47,48]. Finally, we deal with the structure of hCG-H and its hyperglycosylation [49]. The trophoblast is not the only endocrine gland producing hCG-related molecules. As shown in Figure 2.2, the gonadotrope cells of the pituitary gland also produce elevated levels of an hCG variant [50]. As noted, this hCG has a different biological activity than regular hCG and is sulfated in structure [51]. Very low levels of hCGp are seemingly produced prior to ovulation during every woman’s menstrual period [52]. hCGp is appropriately covered in this book. Twenty-five years later came the publication of the amino acid sequence, carbohydrate structure, and molecular biology of hCG subunits, and the story appeared close to being complete. At that time, more than 20 research groups in the United States focused solely on hCG. Since then, new research areas have emerged, and we now know that hCG is not just one molecule with the single function of promoting corpus luteal progesterone production, but a group of four independent molecules(hCG, hCGp, hCG-H, and hCG-H), each with its own separate structure and independent functions. Today, only three research groups in the United States focus solely on hCG. The field is, however, growing stronger and new knowledge continues to emerge, such as the role of hCG in human evolution, the use of hCG antibodies to treat cancer, and
20
Human Chorionic Gonadotropin (hCG)
the involvement of hCG-H in pregnancy disorders. As detailed in this book, the field seems more energized than ever. The hCG field has its oddments, such as -subunit being produced in excess of -subunit and hCG dimer; the -subunit is common to all the glycoprotein hormones, yet has independent biological functions. It certainly has a structure and biology of its own, and that is not forgotten in this book. We also examine the entire hCG field, hCG, hCGp, and hCG-H and how in many ways this team of molecules works together. We suggest that you read this entire book to foresee the exciting future that is developing regarding hCG.
References [1] Hirose T. Experimentalle histologische studie zur genese corpus luteum. Mitt Med Fakultd Univ ZU (Tokyo) 1919;23:63. [2] Hirose T. Exogenous stimulation of corpus luteum formation in the rabbit: influence of extracts of human placenta, decidua, fetus, hydatid mole, and corpus luteum on the rabbit gonad. J Jpn Gynecol Soc 1920;16:1055. [3] Murata M, Adachi K. Uder die künstliche erzeugung des corpus luteum durch injection der plazenta substanz aus frühen schwangerschaft-smonaten. Altscher Gentsch Gynak 1927;92:45. [4] Ascheim S, Zondek B. Hypophysenvorderlappen hormone und ovarial hormone im Harn von Schwangeren. Klin Wochenschr 1927;6:13–21. [5] Zondek B, Aschheim S. The zondek-ascheim pregnancy test. Can Med Assoc J 1930; 22:251–3. [6] Friedman MH, Lapham ME. A simple, rapid procedure for the laboratory diagnosis of early pregnancies. Am J Obstet Gynec 1931;21:405–10. [7] Shapiro HA, Zwarenstein H. Metabolic changes associated with endocrine activity and the reproductive cycle in Xenopus laevis: IV. The effects of injection of ovarian and pituitary extracts on the serum calcium in normal, ovariectomised and hypophysectomised toads. J Exp Biol 1934;11:267–72. [8] Delfs E. An assay method for human chorionic gonadotropin. Endocrinol 1941;28: 196–202. [9] Kupperman HS, Greenblatt RB, Noback CR. A two and six-hour pregnancy test. J Clin Endocrinol 1943;3:548–50. [10] Galli-Mainini C. Pregnancy test using the male batrachia. J Am Med Assoc 1948;138: 121–5. [11] Wide L, Gemzell CA. An immunological pregnancy test. Acta Endocrinol 1960;35(2): 261–7. [12] Wide L. An immunological method for the assay of human chorionic gonadotropin. Acta Endocrinol (Kobenhavn) 1962;4:1–111. [13] Paul WE, Odell WD. Radiation inactivation of the immunological and biological activities of human chorionic gonadotropin. Nature 1964;203:979–80. [14] Wilde CE, Orr A, Bagshaw K. A radioimmunoassay for human chorionic gonadotropin. Nature 1965;205:191–2. [15] Lunenfeld B, Eshkol A. Immunology of human chorionic gonadotropin (HCG). Vit Hor 1967;25:137–90.
History and Introduction to Human Chorionic Gonadotropin (hCG)
21
[16] Aono T, Goldstein DP, Taymor ML, Dolch K. A radioimmunoassay method for human pituitary luteinizing hormone (LH) and human chorionic gonadotropin (HCG) using 125-I-labeled LH. Am Journal Obstet Gynecol 1967;98:996–1001. [17] Rushworth AG, Orr AH, Bagshawe KD. The concentration of HCG in the plasma and spinal fluid of patients with trophoblastic tumours in the central nervous system. Br J Cancer 1968;22:253–7. [18] Vaitukaitis JL, Braunstein GD, Ross GT. A radioimmunoassay which specifically measures human chorionic gonadotropin in the presence of human luteinizing hormone. Am J Obstet Gynecol 1972;113:751–8. [19] Kohler G, Milstein C. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 1975;256:495–7. [20] Maggio ET, Nakamura RM. Biomedical advances: 1. Clinical assays employing enzymes in innovative ways. Ligand Rev 1981;3:16–24. [21] Ishikawa E, Kawai T, Miyai K. Enzyme immunoassay. Tokyo: Igaku-Shoin; 1981:1–280. [22] Hussa RO, Hudson En. A two-site immunometric assay in evaluation of low levels of serum hCG. Am Clin Prod Rev 1984;3:12–17. [23] Sekiya T, Furuhashi Y, Goto S, Kaseki S, Tomoda Y, Kato K. Sandwich-type enzyme immunoassay for human chorionic gonadotropin. J Endocrinol Invest 1981;4:275–9. [24] Batzer FR. Hormonal evaluation of early pregnancy. Fertil Steril 1980;34:1–13. [25] Joshi UM, Roy R, Sheth AR, Shah HP. A simple and sensitive color test for the detection of human chorionic gonadotropin. Obstet Gynecol 1981;57:252–4. [26] Barnard GJ, Kim JB, Brockelbank JL, Collins WP, Gaier B, Kohen F. Management of choriogonadotropin by chemiluminescense immunoassay and immunochemiluminometric assay: 1. Use of isoluminol derivatives. Clin Chem 1984;130:538–541. [27] Bahl OP, Carlsen RB, Bellisario R, Swaminathan N. Human chorionic gonadotropin: amino acid sequence of the alpha and beta subunits. Biochem Biophys Res Commun 1972;48:416–22. [28] Morgan FJ, Birken S, Canfield RE. The amino acid sequence of human chorionic gonadotropin. The alpha subunit and beta subunit. J Biol Chem 1975;250:5247–58. [29] Fiddes JC, Goodman HM. The cDNA for the beta-subunit of human chorionic gonadotropin suggests evolution of a gene by readthrough into the 3’-untranslated region. Nature 1980;286:684–7. [30] Talmadge K, Boorstein WR, Vamvakopoulos NC, Gething M-J, Fiddes JC. Only three of the seven human chorionic gonadotropin beta subunit genes can be expressed in the placenta. Nuc Acid Res 1984;12:8415–36. [31] Wehmann RE, Amr S, Rosa C, Nisula BC. Metabolism, distribution and excretion of purified human chorionic gonadotropin and its subunits in man. Annales de’Endocrinologie 1984;45:291–5. [32] Cole LA, Birken S, Perini F. The structures of the serine-linked sugar chains on human chorionic gonadotropin. Biochem Biophys Res Comm 1985;126:333–9. [33] Endo T, Nishimura R, Kaxano T, Mochizuki M, Kobata AJ. Structural differences found in the asparagine-linked sugar chains of human chorionic gonadotropins purified from the urine of patients with invasive mole and with choriocarcinoma. Cancer Res 1987;47:5242–5. [34] Minegishi T, Nakamura K, Takakura Y, Miyamoto K, Hasegawa Y, Ibuki Y, et al. Cloning and sequencing of human LH/hCG receptor cDNA. Biochem Biophys Res Commun 1990;172:1049–54. [35] Lei ZM, Reshef E, Rao CV. The expression of human chorionic gonadotropin/luteinizing hormone receptors in human endometrial and myometrial blood vessels. J Clin Endo crinol Metab 1992;75:651–9.
22
Human Chorionic Gonadotropin (hCG)
[36] Shi QJ, Lei ZM, Rao CV, Lin J. Novel role of human chorionic gonadotropin in differentiation of human cytotrophoblasts. Endocrinol 1993;132:1387–95. [37] Wu H, Lustbader JW, Liu Y, Canfield RE, Hendrickson WA. Structure of human chorionic gonadotropin at 2.6 Å resolution from MAD analysis of the selenomethionyl protein. Structure 1994;2:545–58. [38] Kobata A, Takeuchi M. Structure pathology and function of the N-linked sugar chains of hCG. Biochim Biophys Acta 1999;1455:315–26. [39] O’Connor JF, Ellish N, Kakuma T, Schlatterer J, Kovalevskaya G. Differential urinary gonadotropin profiles in early pregnancy and early pregnancy loss. Prenat Diagn 1998;18:1232–40. [40] Kovalevskaya G, Genbacev O, Fisher SJ, Cacere E, O’Connor JF. Trophoblast origin of hCG isoforms: cytotrophoblasts are the primary source of choriocarcinoma-like hCG. Mol Cellular Endocrinol 2002;194:147–55. [41] Handshuh K, Guibourdenche J, Tsatsaris V, Guesnon M, Laurendeau I, Evain-Brion D, et al. Human chorionic gonadotropin produced by the invasive trophoblast but not the villous trophoblast promotes cell invasion and is down-regulated by peroxisome proliferator-activated receptor-a. Endocrinol 2007;148:5011–19. [42] Cole LA, Dai D, Leslie KK, Butler SA, Kohorn EI. Gestational trophoblastic diseases: 1. Pathophysiology of hyperglycosylated hCG-regulated neoplasia. Gynecol Oncol 2006;102:144–9. [43] Sasaki Y, Ladner DG, Cole LA. Hyperglycosylated hCG: the source of pregnancy failures. Fertil Steril 2008;89:1871–86. [44] Cole LA, Khanlian SA, Kohorn EI. Evolution of the human brain, chorionic gonadotropin and hemochorial implantation of the placenta: insights into origins of pregnancy failures, preeclampsia and choriocarcinoma. J Reprod Med 2008;53:449–557. [45] Cole LA, Kroll TG, Ruddon RW, Hussa RO. Differential occurrence of free and free subunits of human chorionic gonadotropin (hCG). J Clin Endocrinol Metab 1984;58: 1200–2. [46] Cole LA, Wong Y, Latif M, Chambers JT, Chambers SK, Schwartz PE. Urinary human chorionic gonadotropin free -subunit and -fragment: new markers of gynecologic cancers. Cancer Res 1988;48:1356–60. [47] Gillott DJ, Iles RK, Chard T. The effects of hCG on the in vitro growth of bladder cancer cells. Br J Cancer 1996;73:323–6. [48] Butler SA, Ikram MS, Mathieu S, Iles RK. The increase in bladder carcinoma cell population induced by the free beta subunit of hCG is a result of an anti-apoptosis effect and not cell proliferation. Brit J Cancer 2000;82:1553–6. [49] Valmu L, Alfthan H, Hotakainen K, Birken S, Stenman UH. Site-specific glycan analysis of human chorionic gonadotropin -subunit from malignancies and pregnancy by liquid chromatography–electrospray mass spectrometry. Glycobiol 2006;16:1207–18. [50] Matsuura S, Ohashi M, Chen HC, Shownkeen RC, Hartree AS, Reichert Jr LE, et al. Physicochemical and immunological characterization of an hCG-like substance from human pituitary glands. Nature 1980;286:740–1. [51] Birken S, Maydelman T, Gawinowicz MA, Pound A, Liu Y, Hartree AS. Isolation and characterization of human pituitary chorionic gonadotropin. Endocrinol 1996;137:1402–11. [52] Cole LA, Gutierrez JM. Production of hCG during the menstrual cycle. J Reprod Med 2009;54:245–50.
3 Introduction to Pregnancy
Implantation, Villous Formation, and Hemochorial Placentation Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
As noted in several publications [1,2], hCG is principally produced during pregnancy by villous syncytiotrophoblast cells (the nutrition transfer cells of the placenta), and hyperglycosylated hCG (hCG-H) is produced by extravillous cytotrophoblast cells (the invasive component of placenta). Figure 3.1 illustrates hCG and hCG-H production sites. It shows hCG production by fused multinuclear syncytiotrophoblast cells, which are the result of fusion of cytotrophoblast cells. It also shows hCG-H production by extravillous cytotrophoblast cells. This figure illustrates a major function of hCG: direct promotion of fusion of cytotrophoblast cells to syncytiotrophoblast cells [3]. It also shows terminal differentiation of syncytiotrophoblast cells as occurs in formation of a placental syncytium at 10–40 weeks of gestation. Throughout pregnancy, hCG has numerous functions. As published in 1920 [4], hCG promotes progesterone production in corpus luteal cells. This endocrine action is limited, however, to the time of pregnancy implantation and the 3 weeks that follow (3–6 weeks of gestation). The progesterone produced during this period is needed to maintain the decidua and prevent menstrual bleeding. It also promotes angiogenesis in the maternal uterine vasculature so that uterine spiral arteries optimally meet with the implanting placenta. hCG suppresses the maternal immune system rejection of fetal and placental tissues, and inhibits myometrial contractions through the term of pregnancy [5–8]. As the fetus grows, hCG promotes growth of the uterus, modifies myometrial smooth muscle tissues, and prepares decidual and myometrial tissues for reception of trophoblast cells [9–13]. It also promotes the differentiation of mononuclear cytotrophoblast cells to polynuclear syncytiotrophoblast cells, which are the functional cells of the maternal–fetal circulation barrier [3,9]. As described in Chapter 11, hCG also has functions during pregnancy in growth and differentiation of fetal organs. Interestingly, this hCG receptor disappears in the adult. The hCG-H made by cytotrophoblast cells promotes growth and invasion at implantation of pregnancy and during invasion by choriocarcinoma cells [14,15]. As has been shown, hCG-H blocks apoptosis in cytotrophoblast cells, thus promoting growth [16]. Metalloproteinases (MMPs) are the pathway of invasion during implantation of pregnancy and in choriocarcinoma [17,18]. As such, hCG-H might signal invasion through MMPs. Although hCG-H promotes growth of root cytotrophoblast Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00003-7 © 2010 Elsevier Inc. All rights reserved.
24
Human Chorionic Gonadotropin (hCG) Extravillous invasive cytotrophoblast cells
hCG-H
hCG
Stem cytotrophoblast cells
hCG-promoted fusion Syncytiotrophoblast cells from fusion of cytotrophoblast cells n sio fu ed ot om r p CG
Villous cytotrophoblast cells
hCG h
Terminal syncytiotrophoblast cells
Figure 3.1 Trophoblast differentiation showing the extravillous cytotrophoblast cells that produce hCG-H, and the villous syncytiotrophoblast cells that secrete hCG.
cells throughout pregnancy, hCG also promotes differentiation of these cells into floating villi (the active site for maternal-fetal nutrient exchange). The combination of hCG-H and hCG may lead to cytotrophoblast growth and differentiation, or together to placenta growth and trophoblastic villi. Armed with the knowledge of this hCG and hCG-H production and function information, we examine the overall processes of implantation and placentation in humans. Figure 3.2 illustrates a hatched blastocyst immediately prior to implantation, at 3 weeks of gestation (3 weeks from the start of the last menstrual period). Any hCG or hCG-H produced at this time will be spilled into the uterine cavity and not enter the maternal circulation. At this pre-implantation time, the syncytiotrophoblast cells form microvilli for their initial adhesion with the uterus, using integrin and EGF as adhesion signals [19]. Figure 3.3 illustrates the engulfment of the blastocyst by the decidua at 3.5 weeks of gestation. It is cytotrophoblast hCG-H that drives this initial invasion into
Mic r cytio ovilli on trop hob last
syn
ng
ali
En
ion
ria
et
m
do
es
GF
h ad
n sig
nd
la
lg
dE
ri
eg
Int
n na
Inner
Cytotrophoblast
ass
cell m
Blastocyst cavity
Figure 3.2 Hatched blastocyst attachment to the uterus. Image shows tiny microvilli formed by syncytiotrophoblast on attachment lip of blastocyst. Integrin and EGF are produced by blastocysts to communicate with an attachment site. Inner cell mass (amnion) in the blastocyst is the component that becomes the fetus. Attachment occurs between 6 and 7 days after ovulation or at 3 weeks since last menstrual period (3 weeks gestation). Cytotrophoblast cells are shown in white, syncytiotrophoblast cells in light grey, and uterine epithelium and inner cell mass (amnion) in dark grey.
Introduction to Pregnancy Implantation, Villous Formation, and Hemochorial Placentation
Uterine m epitheliu
Decidua
25
hC pr G p o a ge rom an nd ste ot gio ute ron es ge rin e ne e sis
26
n s io P as MM v In -H, G hC
hC Inv G- as H, ion M Invasion M Ps hCG H, MMPs
t
las st ob obla roph h p o t tr Cyto ytio nc Sy
Decidua
Endometr
Em
A ca mn vit iot y ic
br
yo n
ic
sk
Blastocyst cavity
e m rin liu te e U pith e
Figure 3.3 Complete implantation of blastocyst. Approximately 3–4 days after attachment, or at 3.5 weeks of gestation, the blastocyst becomes wholly implanted into the decidua. The blastocyst forms notable features, including a cytotrophoblast extension or extravillous cytotrophoblast. This produces hCG-H, which inhibits cytotrophoblast apoptosis, promoting growth and MMPs that drive invasion through the decidua. The syncytiotrophoblast cells immediately start making hCG, which maintains progesterone production by the corpus luteum and promotes growth of myometrial and decidual spiral arteries. The inner cell mass or amnion forms an amniotic cavity. Cytotrophoblast cells are shown in white, syncytiotrophoblast cells in light grey, and uterine epithelium and the inner cell mass (amnion) in dark grey.
Human Chorionic Gonadotropin (hCG)
ial gland
di
hC pr G p o a g ro an nd este mo gi ut ro tes og er n en ine e es is
Introduction to Pregnancy Implantation, Villous Formation, and Hemochorial Placentation
Invasion hCG-H, MMPs
27
yto
sc
lou Vil
tro
Chorionic cavity
Cy
Endometr
to tro
ph
ob la
st
Yolk sac
o ph
Invasion hCG-H, MMPs
Amniotic cavity
t
s bla
tio
cy
n Sy
t
s bla
ho
p tro
Decidua
ial gland
e in um er eli Ut pith e
Figure 3.4 At 4 weeks gestation, amnion differentiation occurs with formation of a clear yolk sac. Multiple cytotrophoblast extensions become the roots of villous structures. hCGH-driven invasion continues through the decidua and into the myometrium. hCG promotes corpus luteal progesterone production and myometrial and decidual spiral artery angiogenesis. Cytotrophoblast cells are shown in white, syncytiotrophoblast cells in light grey, and uterine epithelium and amnion cells in dark grey.
the decidua by the blastocyst. The blastocyst takes on notable features at this time, including a cytotrophoblast column [20] (Figure 3.3). This produces hCG-H, which in turn inhibits cytotrophoblast apoptosis, promotes growth, and seemingly promotes continual invasion through MMPs. The early syncytiotrophoblast cells make hCG, which enters the circulation and maintains progesterone production by the corpus luteum. hCG also promotes growth of uterine spiral arteries, as well as the fusion of cytotrophoblast cells to syncytiotrophoblast cells, and inhibits immune rejection of the foreign cells invading the uterus. Figure 3.4 illustrates the amnion and chorion at 4 weeks gestation. Amnion differentiation occurs with formation of a clear yolk sac. Multiple cytotrophoblast extensions or columns form. These are the roots of villous trophoblast structures that form in the following weeks. Figure 3.5 illustrates the amnion and chorion at 5 weeks gestation. The cytotrophoblast columns (promoted by hCG-H) combine via differentiation to syncytiotrophoblast cells (promoted by hCG) and lead to the initiation of villus formation. At 6 weeks gestation, complete villus structures are observed (Figure 3.6). Villus structures are primarily composed of cytotrophoblast cells overlaid by syncytiotrophoblasts.
28
hCG-H promotes cytotrophoblast and villus growth
Maternal blood
hCG promotes villus differentiation hCG promotes uterine angiogenesis
Amniotic cavity
Amnion
Yolk sac
Embryonic disk
Figure 3.5 Trophoblast cells and amnion cells at 5 weeks gestation. Cytotrophoblast columns grow because of hCG-H promotion; with hCG differentiation, they form primitive extensions or villus structures. Amnion grows and the yolk sac expands.
Human Chorionic Gonadotropin (hCG)
Body stalk
e
In va d
Flo
ati
ng
vil
lus
Decidua
Anchor villu
hCG-promoted angiogenesis
s
Extravillous cytotrophoblast
Fl
oa
tin
g
vil
lu
s
Villous cytotrophoblast Villous syncytiotrophoblast Fetal circulation
29
Figure 3.6 A fully formed villus structure at 6 weeks gestation. This structure is formed by the growing cytotrophoblast column (promoted by hCG-H) and differentiated by hCG. Two types of villi are formed: floating villi and an anchor villus. The anchor villus is terminated by a mass of extravillous cytotrophoblast cells. hCG-H (produced by the extravillous cytotrophoblast cells) drives invasion to as deep as one-third the thickness of the myometrium. hCG (produced by syncytiotrophoblast cells) continuously promotes angiogenesis in decidual and spiral arteries. At this point, syncytiotrophoblast cells now produce progesterone themselves, and take over hCG production from corpus luteal cells. As illustrated, an active fetal circulation begins; the syncytiotrophoblast cells of floating villi extract nutrition from surrounding maternal blood and filter it into the fetal circulation. Cytotrophoblast cells are shown in white and syncytiotrophoblast cells in light grey.
Introduction to Pregnancy Implantation, Villous Formation, and Hemochorial Placentation
Inv ade
e
de
Inva
d va In
hCG-H, MMPs
Spiral artery
30
Human Chorionic Gonadotropin (hCG)
Endometrial vein
Spiral artery
Decidua parietalis
Umbilical artery
Myometrium
Umbilical vein Umbilical cord
Figure 3.7 Active hemochorial placentation at 8–10 weeks gestation. This figure shows the complete fetal nutrition system, with villous structures implanted into the decidua parientalis within the myometrium.
Two types of villi are formed within each structure: floating villi, the active villi for maternal/fetal nutrient exchange; and the anchor villi for invasion and anchoring of villi (Figure 3.6). The anchor villi are terminated by a mass of extravillous cytotrophoblast cells. hCG-H produced by the extravillous cytotrophoblast cells drives invasion of the villous structures to as deep as one-third the thickness of the myometrium. hCG produced by syncytiotrophoblast cells continuously promotes angiogenesis in decidual and myometrial vasculature so that vasculature can meet the deeply invading placental villi. By the sixth week of gestation, hCG-promoted corpus luteal progesterone production comes to an end. At this time, the placental syncytiotrophoblast cells take over progesterone production. hCG continues to immune-suppress the invading trophoblast cells, promote uterine growth, and relax contraction of the myometrial muscle cells. As illustrated in Figure 3.6, a fetal circulation begins at this stage, feeding into a primitive umbilical cord. The villus structure is now soaking in maternal blood; the floating villus is transferring oxygen, glucose, and other nutrients from the maternal blood into the circulation of the developing fetus. By 8–10 weeks of pregnancy, the villus structures have attached to the decidua parientalis and hemochorial placentation is now fully active (Figures 3.6 and 3.7). Villous structures have maximally invaded the myometrium and floating villi are now optimally filtering surrounding maternal blood into the fetal circulation. Invasion by the extravillous cytotrophoblast cells and hCG-H continues. Figure 3.7 shows hemochorial placentation in place and active for the remainder of the pregnancy. All told, the hemochorial placentation absorbs maternal nutrients and passes them into the fetal circulation very efficiently.
Introduction to Pregnancy Implantation, Villous Formation, and Hemochorial Placentation
31
Figure 3.8 Villous structure at sixth week of gestation. Right side of slide shows fused villous syncytiotrophoblast cells (with three and four nuclei) and underlying villous cytotrophoblast cells (extreme right side of slide).
Figure 3.9 Term pregnancy villous structure showing terminally differentiated syncytiotrophoblast cells.
Figure 3.8 shows a slide of human placental tissue at the sixth week of pregnancy. The slide illustrates the preponderance of multinucleated villous syncytiotrophoblast cells (the production site of hCG) and the underlying villous cytotrophoblast cells. Figure 3.9 is a slide of term placenta tissue from a term pregnancy. It shows the presence of terminal syncytiotrophoblast cells with an abundance of nuclei. The excess of
32
Human Chorionic Gonadotropin (hCG)
hCG subunit mRNA generated by the 20–50 nuclei of these cells causes little hCG to be made, because of the shortage of endoplasmic reticulum in the nucleus-filled cells.
References [1] Kovalevskaya G, Genbacev O, Fisher SJ, Cacere E, O’Connor JF. Trophoblast origin of hCG isoforms: cytotrophoblasts are the primary source of choriocarcinoma-like hCG. Mol Cell Endocrinol 2002;194:147–55. [2] Handshuh K, Guibourdenche J, Tsatsaris V, Guesnon M, Laurendeau I, Evain-Brion D, et al. Human chorionic gonadotropin produced by the invasive trophoblast but not the villous trophoblast promotes cell invasion and is down-regulated by peroxisome proliferator-activated receptor-a. Endocrinol 2007;148:5011–19. [3] Shi QJ, Lei ZM, Rao CV, Lin J. Novel role of human chorionic gonadotropin in differentiation of human cytotrophoblasts. Endocrinol 1993;132:1387–95. [4] Hirose T. Exogenous stimulation of corpus luteum formation in the rabbit: influence of extracts of human placenta, decidua, fetus, hydatid mole, and corpus luteum on the rabbit gonad. J Jpn Gynecol Soc 1920;16:1055. [5] Lei ZM, Reshef E, Rao CV. The expression of human chorionic gonadotropin/luteinizing hormone receptors in human endometrial and myometrial blood vessels. J Clin Endo crinol Metab 1992;75:651–9. [6] Herr F, Baal N, Reisinger K, Lorenz A, McKinnon T, Preissner KT, et al. HCG in the regu lation of placental angiogenesis. Results of an in vitro study. Placenta 2007;28(Suppl A): S85–93. [7] Zygmunt M, Herr F, Munstedt K, Lang U, Liang OD. Angiogenesis and vasculogenesis in pregnancy. Euro J Obstet Gynecol Reprod Biol 2003;110(Suppl 1):S10–18. [8] Toth P, Lukacs H, Gimes G, Sebestyen A, Pasztor N, Paulin F, et al. Clinical importance of vascular LH/hCG receptors: a review. Reprod Biol 2001;1:5–11. [9] Rao CV. Physiological and pathological relevance of human uterine LH/hCG receptors. J Soc Gynecol Invest 2006;13:77–78. [10] Ticconi C, Zicari A, Belmonte M, Rao CV, Piccone E. Pregnancy-promoting actions of hCG in human myometrium and fetal membranes. Placenta 2007;28:S137–43. [11] Rao CV, Leu ZM. The past, present and future of nongonadal LH/hCG actions in reproductive biology and medicine. Mol Cell Endocrinol 2007;269:2–8. [12] Keay SK, Vatish M, Karteris E, Hillhouse EW, Randeva HS. The role of hCG in reproductive medicine. Br J Obstet Gynecol 2004;111:1218–28. [13] Kornyei JL, Lei ZM, Rao CV. Human myometrial smooth muscle cells are novel targets of direct regulation by human chorionic gonadotropin. Biol Reprod 1993;49:1149–57. [14] Cole LA, Dai D, Leslie KK, Butler SA, Kohorn EI. Gestational trophoblastic diseases: 1. Pathophysiology of hyperglycosylated hCG-regulated neoplasia. Gynecol Oncol 2006;102:144–9. [15] Cole LA, Khanlian SA, Riley JM, Butler SA. Hyperglycosylated hCG (hCG-H) in gestational implantation, and in choriocarcinoma and testicular germ cell malignancy tumorigenesis. J Reprod Med 2006;51:919–29. [16] Hamade AL, Nakabayashi K, Sato A, Kiyoshi K, Takamatsu Y, Laoag-Fernandez JB, et al. Transfection of antisense chorionic gonadotropin gene into choriocarcinoma cells suppresses the cell proliferation and induces apoptosis. J Clin Endocrinol Metab 2005;90:4873–9.
Introduction to Pregnancy Implantation, Villous Formation, and Hemochorial Placentation
33
[17] Qinglei L, Hongmei W, Yunge Z, Haiyan L, Qingxiang AS, Cheng Z. Identification and specific expression of matrix metalloproteinase-26 in rhesus monkey endometrium during early pregnancy. Mol Hum Reprod 2002;8:934–40. [18] Feng H, Cheung A, Xue W, Wang Y, Wang X, Fu S, et al. Down-regulation and promoter methylation of tissue inhibitor of metalloproteinase 3 in choriocarcinoma. Gynecol Oncol 2004;94:375–82. [19] Cross JC, Werb Z, Fisher SJ. Implantation and the placenta: key pieces of the development puzzle. Science 1994;266:1508–18. [20] Norwitz ER, Schust DJ, Fisher SJ. Implantation and the survival of early pregnancy. New Engl J Med 2001;345:1400–8.
4 The Molecular Genetics of hCG Stephen A. Butler Biomedical Sciences, Middlesex University, London, UK
We know that hCG is a complicated molecule. It is not surprising, then, that the transcription and regulation of hCG gene expression are also complicated. In the first complete text on hCG [1], a short paragraph was dedicated to the complex gene cluster of hCG, which was first fully described only 5 years earlier [2] and mapped only a year earlier [3]. Since then, the advances in molecular biology have thrown the field wide open, but despite these advances, many questions posed in the early 1980s still remain unanswered more than a quarter of a century later: most importantly, exactly how is hCG gene expression controlled?
4.1 The LH/hCG Gene Cluster Chorionic gonadotropin -subunit (hCG) gene expression is fairly straightforward. Common to all of the glycoprotein hormones (Luteinizing hormone (LH), follicle stimulating hormone (FSH), thyroid stimulating hormone (TSH), and hCG), the -subunit amino acid sequence (see Figure 4.3; also detailed in Chapter 5) is encoded by a single copy gene on chromosome 6 (6q14-q21) [4], the genomic location of which is found at 6: 87300021–88300020 bp. The -subunit of hCG, however, is encoded in a region on chromosome 19 at 19q13.32 by a cluster of six CG paralogs originally designated CG1, CG2, CG3, CG5, CG7, and CG8; they are adjacent to the CG4 gene encoding the hLH -subunit [2,5–8]. CG5 is found at genomic location 19: 49047815 bp, approximately in the center of the region coding the six CG genes. CG5 is often used, perhaps rather erroneously, as an example of a “typical” hCG gene (Figure 4.1). Sequencing CGβ7
CGβ8
CGβ1
CGβ (CGβ3 )
LHβ 3′
5′
6/7
8
5
1
2
3′
3/9
4 5′
CGβ5
CGβ2
Figure 4.1 A diagrammatic representation of the arrangement of genes in the LH/hCG gene cluster on chromosome 19q13.32. The figure indicates the original assignments, shown inside the strands, which are based on earlier studies [2,3,9]. The latest terminology is annotated externally with the transcription orientation arrows. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00004-9 © 2010 Elsevier Inc. All rights reserved.
38
Human Chorionic Gonadotropin (hCG)
studies have since shown that genes CG7 and CG3 also have allelic variants, CG6 and CG9, respectively [9,10]. CG3 is now simply reported as CG and was probably the first true CG gene in the cluster; it arises adjacent to the LH gene (CG4). This unique cluster was initially thought to have been a result of an evolutionary change following a duplication of the -subunit gene [11,12], based on the high degree of structural homology in peptide sequences, which indicates that hCG and hCG were orthologs and shared a common ancestral gene. This was soon dismissed in part, however, following the development of full sequence data, which showed large discrepancies between intronic and exonic regions in terms of size, number, and position between and sequences and their distant location on different chromosomes [1]. Although not a recent duplication, newer data suggest that the similarities extend much further back to very early evolutionary divergence, as regions involving cystine knot formation appear to be highly conserved in hCG subunits as well as other cystine knot growth factors [13]. It is, however, accepted that the first CG gene arose as a result of a duplication event from LH and that this occurred at some point in early primate divergence [14]. With this event, a single point mutation occurred in the duplicated gene toward the carboxyl end (corresponding to aa 114) of the DNA coding sequence. This point mutation resulted in both a frameshift (Gln Arg) and readthrough of the termination codon into the untranslated region adjacent to the terminal LH exon [11]. This effectively gave rise to what we refer to as the hCG carboxy terminal peptide (CTP) and serves as the largest sequence difference between hCG and LH in the mature proteins. The pattern of increased number of genes within the cluster, along with the evolution of primate species, is compelling to read [15] and is expressed in terms of clinical significance by Cole in Chapter 36. The current published human sequences for each of the CG genes are now readily available at the National Center for Biotechnology Information (NCBI) websites but are, of course, dependent on the source material sequenced (the current sequences are shown in Figures 4.2, 4.6, 4.7, and 4.8). The groups of Fiddes and coworkers and Boime and coworkers together described the genetics of both hCG subunits in a series of publications in the late 1970s to the mid-1980s, revealing the loci and cluster maps, and establishing the foundations for the evolution of the genes which we are again now exploring [3–5,11,12,16–20].
4.2 Control of hCG Gene Expression: hCG hCG -subunit gene expression was reported in 1979 [16] and was shown to be expressed in translatable levels in placental and tumor tissue [17,18]. The sequence was originally published in 1981 [4]; an adapted version can be seen in Figure 4.2. Online versions of the sequences can now be readily obtained from websites such as that of NCBI (Figure 4.3). -Subunit mRNA is 10 times more abundant in trophoblast tissue than the -subunit. Thus, the regulatory mechanisms resulting in - and -subunit gene expression are likely to be different. The pathways involved in the up-regulation of hCG genes are not fully understood; however, cAMP has clearly been shown to
TCATTGGACGGAATTTCCTGTTGATCCCAGGGCTTAGATGCAGGTGGAAACACTCTGCTGGTATAAAAGCAGGTGAGGACTTCATTAACTGCAGTTACTG 101 AGAACTCATAAGACGAAGCTAAAATCCCTCTTCGGATCCACAGTCAACCGCCCTGAACACATCCTGCAAAAAGCCCAGAGAAAGGTAATATGAATGAAAT 201 AATTTTGGGGGACTTTAATTGAGGAGTAAGATATTTGAGAATA// 6.4 Kb 234 TTTTTTTTTTTTTTTTTTGCCATGTCTGTCTGCAGGAGCGCCATGGATTACTACAGAAAATATGCAGCTATCTTTCTGGTCACATTGTCGGTGTTTCTGC
The Molecular Genetics of hCG
1
334 ATGTTCTCCATTCCGCTCCTGATGTGCAGGGTGCGTGACCAAATTTGTGGTTCAAGTAATAAGGACAACACACATTT// 1.7 Kb 411 TTCTTTTTGAGTCTTTTTTGGATATTTTACTCTGCCTTTTTTTTTCCCTGATAGATTGCCCAGAATGCACGCTACAGGAAAACCCATTCTTCTCCCAGCC 511 GGGTGCCCCAATACTTCAGTGCATGGGCTGCTGCTTCTCTAGAGCATATCCCACTCCACTAAGGTCCAAGAAGACGATGTTGGTCCAAAAGAACGTCACC 611 TCAGAGTCCACTTGCTGTGTAGCTAAATCATATAACAGGGTAAGAACCTCAAGATCCCCAGAAGCTTT// 1.4 Kb 679 ATAATATGTTTTTTTTTCCTTCCCCTTTAGGTCACAGTAATGGGGGGTTTCAAAGTGGAGAACCACACGGCGTGCCACTGCAGTACTTGTTATTATCACA 779 AATCTTAAATGTTTTACCAAGTGCTGTCTTGATGACTGCTGATTTTCTGGAATGGAAAATTAAGTTGTTTAGTGTTTATGGCTTTGTGAGATAAAACTCT 879 CCTTTTCCTTACCATACCACTTTGACACGCTTCAAGGATATACTGCAGCTTTACTGCCTTCCTCCTTATCCTACAGTACAATCAGCAGTCTAGTTCTTTT 979 CATTTGGAATGAATACAGCATTAAGCTTGTTCCACTGCAAATAAAGCCTTTTAAATCATCATTCAATCACTGAATTATCATTTTTCTTCAAAGTAAG
39
Figure 4.2 Original published gene sequence of CG indicating introns and exons (in bold black). The amino acids coded for by the italicized (bold) regions are removed during posttranslational modifications; remaining regions therefore code only for the mature protein. Large portions of intronic regions were omitted from the published sequence, indicated here by //. Adapted from [4].
40
Human Chorionic Gonadotropin (hCG)
Nucleotide sequence ATGGATTACTACAGAAAATATGCAGCTATCTTTCTGGTCACATTGTCGGTGTTTCTGCATGTTCTCCATTCCG CTCCTGATGTGCAGGATTGCCCAGAATGCACGCTACAGGAAAACCCATTCTTCTCCCAGCCGGGTGCCCCAAT ACTTCAGTGCATGGGCTGCTGCTTCTCTAGAGCATATCCCACTCCACTAAGGTCCAAGAAGACGATGTTGGTC CAAAAGAACGTCACCTCAGAGTCCACTTGCTGTGTAGCTAAATCATATAACAGGGTCACAGTAATGGGGGGTT TCAAAGTGGAGAACCACACGGCGTGCCACTGCAGTACTTGTTATTATCACAAATCTTAA Translated sequence MDYYRKYAAIFLVTLSVFLHVLHSAPDVQDCPECTLQENPFFSQPGAPILQCMGCCFSRAYPTPLRSKKTMLV QKNVTSESTCCVAKSYNRVTVMGGFKVENHTACHCSTCYYHKS
Figure 4.3 Gene sequence and amino acid translation of CG (GP), adapted from the sequence sourced from the NCBI database. The sequence shows exons only, with splice junctions indicated by vertical lines. The translated sequence undergoes posttranslational modifications and is trimmed to the sequence shown between the vertical dashed lines corresponding to 111 amino acids.
increase both - and -subunit gene expression in placental and choriocarcinoma cells [21–23]. All of the glycoprotein hormone genes contain 5 cAMP response element (CRE) enhancer sequences that act together with specific placental or pituitary promoters [24]. The promoter region for the -subunit gene contains two identical neighboring CREs found between 146 and 111, which are responsible for the binding of CRE binding proteins (CREBPs) and other members of the B-Zip transcription factor family [25–30]. At position 180 to 151, an upstream regulatory element (URE) is located and overlapped by several additional binding protein regions [26,28,29,31–34]. Initially designated URE-1 and URE-2 [28], they were later identified as -subunit activator element (ACT) and tissue/trophoblast-specific element (TSE) [33]. Although these are essentially the same regions, the mapping experiments conducted showed differences in the precise location of each region. URE-1/ACT is located at 161 to 142 and binds one of the members of the ubiquitous (a family of transcriptional factors) GATA family (A family of transcriptional factors) of DNA binding proteins. URE-2/TSE is located at 182 to 159 and binds the TSE-binding protein (TSEBP). In addition, there are two further regulatory sequences: the downstream domain (DSD) located at 172 to 151, and the upstream domain (USD) at 177 to 156 [28]. These regions overlap each other while at the same time overlapping the URE-1/ACT sites. This could indicate that these regions are activated by at least two different binding proteins that may be specifically expressed by either pituitary or placental tissue. Although distinct base-pair regions have been established in murine and equine models, human gene expression studies have as yet failed to confirm this. These observations are summarized in Figure 4.4. In studies on human villous trophoblasts, it has been established that the entire region to 365 bp upstream is sufficient to control all cell-specific transcription of the hCG gene. It was further confirmed that the CREs are required for transcription following luciferese reporter assays [35]. Interestingly, it was also shown that activating transcription factor 1 (ATF1) was strongly implicated in CRE binding and, to a
The Molecular Genetics of hCG
41 365-bp Region responsible for transcription
–182
–159 –161
–142
URE2/TSE URE1/αACT URE CRE –180
–151 –145/6 DSD USD
CRE –111
+1
–172 –151 –177 –156
Figure 4.4 A diagrammatic representation of the CG (GH) promoter sequence based on Knofler et al. [35], with significant additions from references cited in text [21–34]. In addition to the CREs, the URE is overlapped by several additional binding protein regions; such an overlap could indicate that these regions are activated by at least two different tissue-specific binding proteins.
lesser extent, CREB-1; however, ATF2, ATF3, ATF4, and CREB-2 do not appear to be involved [35]. According to an additional finding in human studies, CREBP binding appears to be dependent on URE binding [34].
4.3 Control of hCG Gene Expression: hCG The control of hCG gene expression is independent from that of hCG; where hCG continues to be produced throughout gestation, hCG diminishes and therefore appears to regulate the overall hCG concentration in the blood [36]. In light of these early observations, it follows that expression of hCG must also be under very tight control and that there are factors influencing hCG expression which are likely to be distinct from those regulating expression of hCG. The expression of hCG can be detected in cultures of villous cytotrophoblasts [35,37], and the production increases over time. It has been noted, however, that there is a 12-plus-hour delay in hCG expression, which appears to be consistent with cytotrophoblast fusion to syncytial tissue (whereupon maximum hCG production is initiated) [35]. Again, this highlights a distinct role for the expression of hCG as governor of overall hCG production even during very early gestation. Early studies on transfected murine cells in vitro showed that CG5, CG3 (now CG), and CG8 were transcriptionally active, in a descending order of magnitude [38]. PCR analysis of placental tissue mRNA showed that genes CG7 and either CG1 or CG2 were also active, but to a lesser extent. Bo and Boime concluded that at least five of the six CG genes are expressed in vivo [9]. More recently, Laan’s group examined placental tissue by RT-PCR from multiple normal and abnormal pregnancies [39]. In the majority of samples, they confirmed that CG7 was practically inactive, as were CG1 and CG2. This was in contrast to earlier work. Later,
42
Human Chorionic Gonadotropin (hCG)
this study concluded that expression of CG8 was, in fact, dominant; CG5 and CG followed, with approximately equal expressions. This was noted to change slightly in the third trimester, when CG8 and CG5 are co-dominantly expressed and CG is expressed to a slightly lesser extent [39]. It is clear that the order of magnitude of expression, even in placental tissue, is still open to further study. CG1 and CG2 have long been regarded as pseudogenes [2], and there is some disagreement as to the functional nature of the proteins produced by these genes. The message arising from either gene (1 or 2) was noted to be shorter than those from the other genes (a result of alternative splicing) [9]. Transcription of CG1 and CG2 results in three splice variants, only one of which represents a product that may translate into a functional hCG protein [9,40]. The differences in the translated products can be seen by comparing Figures 4.5, 4.6, and 4.7. In more recent studies, CG1 splice variants were again detected, but not those of CG2; these genes (CG1 and CG2) appear to be up-regulated to some extent in the first trimester of pregnancy, suggesting a potential role in implantation [39]. We discuss this further in Chapter 14, where we explore the potential implications of these genes in cancer. The variations in reports about CG paralog expression indicate that tissuespecific promoters are at play in regulating the gene expression within the CG cluster from different cell lineages. Unlike the hCG gene, identifying promoter sequence regions in hCG genes has been extremely difficult because of the absence of known consensus sequences (like the CAAT or the TATA box) within 200 bp upstream of the cap site. Initial studies indicated that expression was higher using transfected constructs (including upstream sequences of 279 bp), and basal promoter regions were identified between 78 and 40 [38,41]. A region rich in CG (187 to 38) is also believed to be involved in basal transcription. Subsequently, a basal transcription factor binding region between 37 and 104 was identified [42]. Later studies indicated that there was a large cAMP-responsive region upstream between 311 and 200 bp on the CG gene; this is required in its entirety for gene expression [43]. Sequences within this region (311 to 202) appear to be required for basal transcription and, according to Steiger and coworkers, lie between 305 and 279 [33]. Within this region, there are multiple binding sites, including two CREBP binding domains (as seen in the -gene promoter region). Some binding sites within this region appear to be more important than others. The tandem CRE 1 (299 to 289) and CRE 2 (240 to 219) regions both bind c-Jun (a down-regulator of hCG expression) [44]; 305 to 249 binds Oct-3/4, which has been shown to silence hCG expression by more than 90% [45]. A TSE site has also been identified. This site contains a sequence similar to the -subunit gene promoter at 301 to 275. It also has an additional two sequences within this essential region that have also been shown to bind TSEBPs, suggesting coordination between hCG and expression [33]. Sites 311 to 274 and 250 to 200 (which do not bind CREBPs) are involved in binding activating protein 2 (AP2), an up-regulator of hCG expression [46,47]. These observations are summarized in Figure 4.8. We now know that this complex region spanning 118 bp actually has multiple AP2 and selective promoter factor 1 (SP1) binding regions [47] in addition to c-Jun and Oct-3/4 transcription factor binding regions. A complex model proposes
The Molecular Genetics of hCG
43
Nucleotide sequences ATGGAGATGTTCCAGGGGCTGCTGCTGTTGCTGCTGCTGAGCATGGGCGGGACATGGGCATCC ATGGAGATGTTCCAGGGGCTGCTGCTGTTGCTGCTGCTGAGCATGGGCGGGACATGGGCATCC ATGGAGATGTTCCAGGGGCTGCTGCTGTTGCTGCTGCTGAGCATGGGCGGGACATGGGCATCC ATGGAGATGTTCCAGGGGCTGCTGCTGTTGCTGCTGCTGAGCATGGGCGGGACATGGGCATCC AAGGAGCCGCTTCGGCCACGGTGCCGCCCCATCAATGCCACCCTGGCTGTGGAGAAGGAGGG AAGGAGCCGCTTCGGCCACGGTGCCGCCCCATCAATGCCACCCTGGCTGTGGAGAAGGAGGG AGGGAGATGCTTCGGCCACGGTGCCGCCCCATCAATGCCACCCTGGCTGTGGAGAAGGAGGG AAGGAGCCGCTTCGGCCACGGTGCCGCCCCATCAATGCCACCCTGGCTGTGGAGAAGGAGGG * ** CTGCCCCGTGTGCATCACCGTCAACACCACCATCTGTGCCGGCTACTGCCCCACCATGACCCGC CTGCCCCGTGTGCATCACCGTCAACACCACCATCTGTGCCGGCTACTGCCCCACCATGACCCGC CTGCCCCGTGTGCATCACCGTCAACACCACCATCTGTGCCGGCTACTGCCCCACCATGACCCGC CTGCCCCGTGTGCATCACCGTCAACACCACCATCTGTGCCGGCTACTGCCCCACCATGACCCGC GTGCTGCAGGGGGTCCTGCCGGCCCTGCCTCAGGTGGTGTGCAACTACCGCGATGTGCGCTTC GTGCTGCAGGGGGTCCTGCCGGCCCTGCCTCAGGTGGTGTGCAACTACCGCGATGTGCGCTTC GTGCTGCAGGGGGTCCTGCCGGCCCTGCCTCAGGTGGTGTGCAACTACCGCGATGTGCGCTTC GTGCTGCAGGGGGTCCTGCCGGCCCTGCCTCAGGTGGTGTGCAACTACCGCGATGTGCGCTTC GAGTCCATCCGGCTCCCTGGCTGCCCGCGCGGCGTGAACCCCGTGGTCTCCTACGCCGTGGCT GAGTCCATCCGGCTCCCTGGCTGCCCGCGCGGCGTGAACCCCGTGGTCTCCTACGCCGTGGCT GAGTCCATCCGGCTCCCTGGCTGCCCGCGCGGCGTGAACCCCGTGGTCTCCTACGCCGTGGCT GAGTCCATCCGGCTCCCTGGCTGCCCGCGCGGCGTGAACCCCGTGGTCTCCTACGCCGTGGCT CTCAGCTGTCAATGTGCACTCTGCCGCCGCAGCACCACTGACTGCGGGGGTCCCAAGGACCAC CTCAGCTGTCAATGTGCACTCTGCCGCCGCAGCACCACTGACTGCGGGGGTCCCAAGGACCAC CTCAGCTGTCAATGTGCACTCTGCCGCCGCAGCACCACTGACTGCGGGGGTCCCAAGGACCAC CTCAGCTGTCAATGTGCACTCTGCCGCCGCAGCACCACTGACTGCGGGGGTCCCAAGGACCAC CCCTTGACCTGTGATGACCCCCGCTTCCAGGACTCCTCTTCCTCAAAGGCCCCTCCCCCCAGCC CCCTTGACCTGTGATGACCCCCGCTTCCAGGACTCCTCTTCCTCAAAGGCCCCTCCCCCCAGCC CCCTTGACCTGTGATGACCCCCGCTTCCAGGCCTCCTCTTCCTCAAAGGCCCCTCCCCCCAGCC CCCTTGACCTGTGATGACCCCCGCTTCCAGGACTCCTCTTCCTCAAAGGCCCCTCCCCCCAGCC * TTCCAAGCCCATCCCGACTCCCGGGGCCCTCGGACACCCCGATCCTCCCACAATAA CGβ TTCCAAGTCCATCCCGACTCCCGGGGCCCTCGGACACCCCGATCCTCCCACAATAA CGβ5 TTCCAAGTCCATCCCGACTCCCGGGGCCCTCAGACACCCCGATCCTCCCACAATAA CGβ7 TTCCAAGTCCATCCCGACTCCCGGGGCCCTCGGACACCCCGATCCTCCCACAATAA CGβ8 * * Translated sequence MEMFQGLLLLLLLSMGGTWASKEPLRPRCRPINATLAVEKEGCPVCITVNTTICAGYCPTMTRVLQGV MEMFQGLLLLLLLSMGGTWASREMLRPRCRPINATLAVEKEGCPVCITVNTTICAGYCPTMTRVLQGV * * LPALPQVVCNYRDVRFESIRLPGCPRGVNPVVSYAVALSCQCALCRRSTTDCGGPKDHPLTCDDPR LPALPQVVCNYRDVRFESIRLPGCPRGVNPVVSYAVALSCQCALCRRSTTDCGGPKDHPLTCDDPF QDSSSSKAPPPSLPSPSRLPGPSDTPILPQ QASSSSKAPPPSLPSPSRLPGPSDTPILPQ *
CGβ, CGβ5, CGβ8 CGβ7
Figure 4.5 Gene sequence of CG genes (CG, CG5, CG7, and CG8) showing homology alignment as adapted from the sequence sourced from the NCBI database. The sequence is 948 nucleotides long and shows exons only (with splice junctions indicated by vertical lines). The asterisks indicate regions where sequence differences can be seen. Translated sequences (165 amino acids) are shown below; the only differences seen in the mature protein occur in the product of CG7, where there are three amino acid substitutions (indicated by asterisks).
44
Human Chorionic Gonadotropin (hCG)
Nucleotide sequence ATGTCAAAGAGGCTGCTGCTGTTGCTGCTGCTGAGCATGGGCGGGACATGGGCATCCAAGGA GCCGCTTCGGCCACGGTGCCGCCCCATCAATGCCACCCTGGCTGTGGAGAAGGAGGGCTGC CCCGTGTGCATCACCGTCAACACCACCATCTGTGCCGGCTACTGCCCCACCATGACCCGCGT GCTGCAGGGGGTCCTGCCGGCCCTGCCTCAGGTGGTGTGCAACTACCGCGATGTGCGCTTC GAGTCCATCCGGCTCCCTGGCTGCCCGCGCGGCGTGAACCCCGTGGTCTCCTACGCCGTGG CTCTCAGCTGTCAATGTGCACTCTGCCGCCGCAGCACCACTGACTGCGGGGGTCCCAAGGAC CACCCCTTGACCTGTGATGACCCCCGCTTCCAGGACTCCTCTTCCTCAAAGGCCCCTCCCCC CAGCCTTCCAAGTCCATCCCGTCTCCCGGGGCCCTAG Translated sequence MSKRLLLLLLLSMGGTWASKEPLRPRCRPINATLAVEKEGCPVCITVNTTICAGYCPTMTRVLQGV LPALPQVVCNYRDVRFESIRLPGCPRGVNPVVSYAVALSCQCALCRRSTTDCGGPKDHPLTCDDP RFQDSSSSKAPPPSLPSPSRLPGP
Figure 4.6 Gene sequence and amino acid translation of CG1 adapted from the sequence sourced from the NCBI database. The sequence is 468 nucleotides long and shows exons only (splice junctions indicated by vertical lines). The translated sequence (155 amino acids), shown under the gene sequence, is generally believed to be a nonfunctional protein with no biological significance.
Nucleotide sequence ATGTCAAAGGGGCTGCTGCTGTTGCTGCTGCTGAGCATGGGCGGGACATGGGCATCCAAGG AGCCGCTTCGGCCACGGTGCCGCCCCATCAATGCCACCCTGGCTGTGGAGAAGGAGGGCTG CCCCGTGTGCATCACCGTCAACACCACCATCTGTGCCGGCTACTGCCCCACCATGACCCGCG TGCTGCAGGGGGTCCTGCCGGCCCTGCCTCAGGTGGTGTGCAACTACCGCGATGTGCGCTT CGAGTCCATCCGGCTCCCTGGCTGCCCGCGCGGCGTGAACCCCGTGGTCTCCTACGCCGTG GCTCTCAGCTGTCAATGTGCACTCTGCCGCCGCAGCACCACTGACTGCGGGGGTCCCAAGGA CCACCCCTTGACCTGTGATGACCCCCGCTTCCAGGCCTCCTCTTCCTCAAAGGCCCCTCCCC CCAGCCTTCCAAGCCCATCCCGACTCCCGGGGCCCTCAGACACCCCGATCCTCCCACAATAA Translated sequence MSKGLLLLLLLSMGGTWASKEPLRPRCRPINATLAVEKEGCPVCITVNTTICAGYCPTMTRVLQGV LPALPQVVCNYRDVRFESIRLPGCPRGVNPVVSYAVALSCQCALCRRSTTDCGGPKDHPLTCDDP RFQASSSSKAPPPSLPSPSRLPGPSDTPILPQRFQDSSSSKAPPPSLPSPSRLPGP
Figure 4.7 Gene sequence and amino acid translation of CG2, adapted from the sequence sourced from the NCBI database. The sequence is 492 nucleotides long and shows exons only (splice junctions indicated by vertical lines). The translated sequence (163 amino acids), shown under the gene sequence, is generally believed to be a nonfunctional protein with no biological significance.
that CG transcription is essentially activated through SP1 and AP2 binding, which can be enhanced by cAMP. SP3 suppresses basal transcription through inhibition of SP1 [48,49]. c-Jun or Oct-3/4 can potentially silence CG transcription activation almost completely [44,45,48]. The most recent studies suggest that the model may be even more complex, with a fine interplay between SP1/3 and AP2() [50]. The model appears to confirm a role for SP3 in suppressing SP1 basal transcription in
The Molecular Genetics of hCG
45
–311 to –200 bp region required in entirety for transcription including –305 to –279 which is required for transcription
–187 to +104 200 bp required in entirety for transcription including –37 to +104 which is required basal for transcription factor binding
–275 –301 –249 –305 TSE CRE1
Oct 3/4
–299 –289
CRE2
–240
AP2 –311
AP2 –274 –250
+1
–219
+104
SP1 –200 –188
Figure 4.8 Diagrammatic representation of the promoter sequence of CG5. Essentially, transcription is controlled by AP2 and SP1 binding multiple sites between 311 and 188, and this is affected positively by cAMP. However, effective silencing of the promoter can be brought about following binding of Oct-3/4 and or c-Jun in the same region. Schematic drawn using information from [41–50].
cytotrophoblasts, while itself being suppressed in syncytial tissue, thus allowing for the known elevation in CG transcription at this juncture in gestation. In conclusion, we have come far in hCG molecular genetics, but definitive answers still elude us. This is certainly due, in no small part, to the similarity in the sequences of each CG paralog in the LH/hCG gene cluster. However, because of more recent work, the holes in the molecular puzzle surrounding hCG gene expression are starting to be filled in once again. Clues from the primate evolution of CG genes and insights into the control of gene expression are certain to reveal the importance and significance of each gene in the cluster. Subsequently, the effect of any resulting functional protein, following translation and posttranslational modification, can then be explained.
References [1] Hussa RO. The clinical marker hCG. New York: Praeger Publishers; 1987. [2] Boorstein WR, Vamvakopoulos NS, Fiddes JC. hCG -subunit is encoded by at least eight genes arranged in tandem and inverted in pairs. Nature 1982;300:419–22. [3] Policastro PF, Daniels-McQueen S, Carle G, Boime I. A map of the hCG-LH gene cluster. J Biol Chem 1986;261:5907–16. [4] Fiddes JC, Goodman HM. The gene encoding the common alpha subunit of the four glycoprotein hormones. J Mol Appl Genet 1981;1:3–18. [5] Policastro P, Ovitt CE, Hoshina M, Fukuoka H, Boothby MR, Boime I. The beta subunit of human chorionic gonadotropin is encoded by multiple genes. J Biol Chem 1983;258: 11492–9. [6] Julier C, Weil D, Couillin P, Cote JC, Nguyen VC, Foubert C, et al. The beta chorionic gonadotropin beta luteinising gene cluster maps to human chromosome 19. Hum Genet 1984;67:174–7.
46
Human Chorionic Gonadotropin (hCG)
[7] Graham MY, Otani T, Boime I, Olsen MV, Carle GF, Chaplin DD. Cosmid mapping of the human chorionic gonadotropin beta subunit genes by field inversion electrophoresis. Nucleic Acids Res 1987;15:4437–48. [8] Jameson JL, Lindell CM, Habener JM. Gonadotropin and thyrotropin and subunit gene expression in normal and neoplastic tissues characterised by using specific ribonucleic acid hybridisation probes. J Clin Endocrinol Metab 1987;64:319–26. [9] Bo M, Boime I. Identification of the transcriptionally active genes of the chorionic gonadotropin gene in vivo. J Biol Chem 1992;267:3179–84. [10] Bidart JM, Baudin E, Troalen F, Bellet D, Schlumberger M. Eutopic and ectopic production of glycoprotein hormones alpha and beta subunits. Ann Endocrinol 1997;58:125–8. [11] Fiddes JC, Goodman HM. The cDNA for the -subunit of hCG suggests evolution of a gene by readthrough into the 3 untranslated region. Nature 1980;286:684–7. [12] Fiddes JC, Talmage K. Structure, expression and evolution of the genes for the human glycoprotein hormones. Recent Prog Hormone Res 1984;40:43–74. [13] Vitt UA, Hsu SY, Hsueh AJ. Evolution and classification of cystine knot-containing hormones and related extracellular signaling molecules. Mol Endocrinol 2001;15:681–94. [14] Maston GA, Ruvolo M. Chorionic gonadotropin has a recent origin within primates and an evolutionary history of selection. Mol Biol Evol 2002;19:320–35. [15] Hallast P, Saarela J, Palotie A, Laan M. High divergence in primate-specific duplicated regions: human and chimpanzee chorionic gonadotropin beta genes. BMC Evol Biol 2008;8:195. [16] Fiddes JC, Goodman HM. Isolation, cloning and sequence analysis of the cDNA for the alpha-subunit of human chorionic gonadotropin. Nature 1979;281:351–6. [17] Boothby M, Kukowska J, Boime I. Imbalanced synthesis of human chorionic gonadotropin alpha and beta subunits reflects the steady state levels of the corresponding mRNAs. J Biol Chem 1983;258:9250–3. [18] Boothby M, Ruddon RW, Anderson C, McWilliams D, Boime I. A single gonadotropin alpha-subunit gene in normal tissue and tumor-derived cell llines. J Biol Chem 1981;256:5121–7. [19] Talmadge K, Boorstein WR, Fiddes JC. The human genome contains seven genes for the beta subunit of chorionic gonadotropin but only one gene for the beta subunit of LH. DNA 1983;2:281–9. [20] Talmadge K, Vamvakopoulos NC, Fiddes JC. Evolution of the genes for the beta subunits of human chorionic gonadotropin and luteinizing hormone. Nature 1984;307:37–40. [21] Burnside J, Nagelberg SB, Lippman SS, Weintraub BD. Differential regulation of hCG alpha and beta subunit mRNAs in JEG-3 choriocarcinoma cells by 8-bromo-cAMP. J Biol Chem 1985;260:12705–9. [22] Chin WW, Gharib SD. Organization and expression of gonadotropin genes. Adv Exp Med Biol 1986;205:245–65. [23] Ringler GE, Kao LC, Miller WL, Strauss JF 3rd. Effects of 8-bromo-cAMP on expression of endocrine functions by cultured human trophoblast cells: regulation of specific mRNAs. Mol Cell Endocrinol 1989;61:13–21. [24] Balfour NJ, Franklyn JA, Gurr JA, Sheappard MC. Multiple DNA elements determine basal and thyroid hormone regulated expression of the human glycoprotein subunit gene in pituitary cells. J Mol Endocrinol 1990;4:187–90. [25] Silver BJ, Bokar JA, Virgin JB, Vallen EA, Milsted A, Nilson JH. Cyclic AMP regulation of the human glycoprotein hormone alpha-subunit gene is mediated by an 18-basepair element. Proc Nat Acad Sci U S A 1987;84:2198–202. [26] Delegeane AM, Ferland LH, Mellon PL. Tissue-specific enhancer of the human glycoprotein hormone alpha-subunit gene: dependence on cyclic AMP-inducible elements. Mol Cell Biol 1987;7:3994–4002.
The Molecular Genetics of hCG
47
[27] Hoeffler JP, Meyer TE, Yun Y, Jameson JL, Habener JF. Cyclic AMP-responsive DNA-binding protein: structure based on a cloned placental cDNA. Science 1988; 242:1430–3. [28] Jameson JL, Albanese C, Habener JF. Distinct adjacent protein-binding domains in the glycoprotein hormone alpha gene interact independently with a cAMP-responsive enhancer. J Biol Chem 1989;264:16190–6. [29] Jameson JL, Powers AC, Gallagher GD, Habener JF. Enhancer and promoter element interactions dictate cyclic adenosine monophosphate mediated and cell-specific expression of the glycoprotein hormone alpha-gene. Mol Endocrinol 1989;3:763–72. [30] Habener JF. Cyclic AMP response element binding proteins: a cornucopia of transcription factors. Mol Endocrinol 1990;4:1087–94. [31] Bokar JA, Keri RA, Farmerie TA, Fenstermaker RA, Andersen B, Hamernik DL, et al. Expression of the glycoprotein hormone alpha-subunit gene in the placenta requires a functional cyclic AMP response element, whereas a different cis-acting element mediates pituitary-specific expression. Mol Cell Biol 1989;9:5113–22. [32] Fenstermaker RA, Farmerie TA, Clay CM, Hamernik DL, Nilson JH. Different combinations of regulatory elements may account for expression of the glycoprotein hormone alpha-subunit gene in primate and horse placenta. Mol Endocrinol 1990;4:1480–7. [33] Steiger DJ, Buscher M, Hecht JH, Mellon PL. Coordinate control of the alpha- and betasubunit genes of human chorionic gonadotropin by trophoblast-specific element-binding protein. Mol Endocrinol 1993;7:1579–88. [34] Pittman RH, Clay CM, Farmerie TA, Nilson JH. Functional analysis of the placentaspecific enhancer of the human glycoprotein hormone alpha subunit gene: emergence of a new element. J Biol Chem 1994;269:19360–8. [35] Knofler M, Saleh L, Strohmer H, Husslein P, Wolschek MF. Cyclic AMP and differentiation-dependent regulation of the proximal HCG gene promoter in term villous trophoblasts. Mol Hum Reprod 1999;5:573–80. [36] Braunstein GD, Rasor JL, Engvall E, Wade ME. Interrelationships of human chorionic gonadotropin, human placental lactogen, and pregnancy-specific 1-glycoprotein throughout normal human gestation. Am J Obstet Gynecol 1980;138:1205–13. [37] Kato Y, Braunstein GD. Discordant secretion of placental protein hormones in differentiating trophoblasts in vitro. J Clin Endocrinol Metab 1989;68:814–20. [38] Otani T, Otani F, Bo M, Chaplin D, Boime I. Promoter sequences in the chorionic gonadotropin subunit gene. In: Mochizuki M, Hussa R, editors. Placental protein hormones. Amsterdam: Elsevier; 1988. [39] Rull K, Laan M. Expression of -subunit of human chorionic gonadotropin genes during the normal and failed pregnancy. Hum Reprod 2005;20:3360–8. [40] Dirnhofer P, Hermann M, Hittmair A, Hoermann R, Kapelari K, Berger P. Expression of the human chorionic gonadotropin - gene cluster in human pituitaries and alternative use of exon 1. J Clin Endocrinol Metab 1996;81:4212–17. [41] Otani T, Boime I, Mochizuki M. Studies on the transcriptional regulatory elements of the hCG beta gene cluster using a choriocarcinoma cell line which expresses hCG beta eutopically. Nippon Sanka Fujinka Gakkai Zasshi-Acta Obstet Gynaecol Jpn 1989;41:1885–90. [42] Hollenberg AN, Pestell RG, Albanese C, Boers ME, Jameson JL. Multiple promoter elements in the human chorionic gonadotropin beta subunit genes distinguish their expression from the luteinizing hormone beta gene. Mol Cell Endocrinol 1994;106:111–19. [43] Albanese C, Kay TW, Troccoli NM, Jameson JL. Novel cyclic adenosine 3’,5’-monophosphate response element in the human chorionic gonadotropin beta-subunit gene. Mol Endocrinol 1991;5:693–702.
48
Human Chorionic Gonadotropin (hCG)
[44] Pestell RG, Hollenberg AN, Albanese C, Jameson JL. c-Jun represses transcription of the human chorionic gonadotropin alpha and beta genes through distinct types of CREs. J Biol Chem 1994;269:31090–6. [45] Liu L, Roberts RM. Silencing of the gene for the subunit of human chorionic gonadotropin by the embryonic transcription factor Oct-3/4. J Biol Chem 1996;271:16683–9. [46] Johnson W, Albanese C, Handwerger S, Williams T, Pestell RG, Jameson L. Regulation of the human chorionic gonadotropin - and -subunit promoters by AP-2. J Biol Chem 1997;272:15405–12. [47] LiCalsi C, Christophe S, Steger DJ, Buescher M, Fischer W, Mellon PL. AP-2 family members regulate basal and cAMP-induced expression of human chorionic gonadotropin. Nucleic Acids Res 2000;28:1036–43. [48] Johnson W, Jameson JL. AP-2 (activating protein 2) and Sp1 (selective promoter factor 1) regulatory elements play distinct roles in the control of basal activity and cyclic adenosine 3,5-monophosphate responsiveness of the human chorionic gonadotropin- promoter. Mol Endocrinol 1999;13:1963–75. [49] Krikun G, Schatz F, Mackman N, Guller S, Demopoulos R, Lockwood CJ. Regulation of tissue factor gene expression in human endometrium by transcription factors Sp1 and Sp3. Mol Endocrinol 2000;14:393–400. [50] Knofler M, Saleh L, Bauer S, Galos B, Rotheneder H, Husslein P, et al. Transcriptional regulation of the human chorionic gonadotropin gene during villous trophoblast differentiation. Endocrinol 2004;145:1685–94.
5 Structure, Synthesis, Secretion, and Function of hCG
Laurence A. Cole1 and Stephen A. Butler2 1
USA hCG Reference Service, Albuquerque, NM, USA Biomedical Sciences, Middlesex University, London, UK
2
hCG is a hormone composed of an - and a -subunits that are joined by noncovalent hydrophobic and ionic interactions. The molecular weight of hCG is 36,700. It is an unusual hormone in that 25–41% of the molecular weight (25–30% in hCG, 25–41% in hyperglycosylated hCG) is derived from the sugar side-chains. The entire story of hCG has been one of ever-changing understanding. The amino acid sequence of hCG subunits, for instance, was first proposed by Carlsen, Bahl, and Swaminathan; and by Bellisario, Carlsen, and Bahl in 1973 [1,2], then corrected by Morgan, Birken, and Canfield in 1975 [3]. Similarly, the structure of the N-linked sugar side-chains was first proposed by Bahl in 1969 [4], modified by Endo and colleagues in 1979 [5], and finalized by Elliott and colleagues in 1997 [6]. The structure of the O-linked oligosaccharides was proposed by Kessler and colleagues in 1979 [7], modified by Cole and colleagues in 1985 [8], and finalized by Valmu and colleagues in 2006 [9]. The position of the disulfide bonds on hCG was proposed by Giudice and Pierce in 1979 [10], modified by Mise and Bahl in 1981 [11], and finalized by Lapthorn et al. in 1994 [12]. In 1919, the hormonal function of hCG was shown to be progesterone promotion [13,14]. Today, the function of hCG is still noted in textbooks as being progesterone promotion, but we now know that hCG has many other important functions in addition to progesterone promotion. From the time of implantation, hCG takes over promotion of progesterone production from luteinizing hormone (LH), acting on a LH/hCG receptor in ovarian corpus luteal cells. This continues for approximately 3 weeks. After that time, the syncytiotrophoblast cells of the placenta take over progesterone production. We now know that hCG plays key roles in promoting the fusion and differentiation of cytotrophoblast to syncytiotrophoblast cells [15], as well as in promoting angiogenesis in the maternal vasculature [16–19], but it also has other roles in placental–fetal development [20,21]. The full biological function of hCG is still being slowly resolved (see Chapter 11). In this chapter, we examine the structure and function of the hormone hCG in detail. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00005-0 © 2010 Elsevier Inc. All rights reserved.
50
Human Chorionic Gonadotropin (hCG)
5.1 Amino Acid Sequence of hCG The amino acid sequence of hCG was firmly established by Morgan and colleagues in 1975 [3]. They discovered that hCG has an -subunit consisting of 92 individual amino acids and a -subunit consisting of 145 amino acids. The hCG -subunit sequence is identical to the sequences of the -subunits of LH, follicle stimulating hormone (FSH), and thyroid stimulating hormone (TSH)—a family of molecules called glycoprotein hormones. In contrast, the -subunit sequence is unique to hCG. Morgan et al. [3] found significant problems with Bahl and colleagues’ original amino acid sequence published in 1973 [1,2], including three erroneous amino acids on the -subunit and others on the -subunit. The corrected amino acid sequence (see Figure 5.1) shows two N-linked oligosaccharides on the hCG -subunit at residues 52 and 78, two N-linked oligosaccharides on the hCG -subunit at residues 13 and 30, and four O-linked oligosaccharides on the C-terminal peptide of the hCG -subunit at residues 121, 127, 132, and 138 [3]. The hCG peptide sequence has 18 acidic and 24 basic amino acids overall. As a result, the balance of amino acids is basic in structure. The basic isoelectric point, however, is made acidic by the addition of eight oligosaccharides, each terminating with one or two sialic acid residues. The sialic acid makes hCG an acidic molecule with an isoelectric point of pI 3.5 [22]. The acidity repels hCG from binding the glomerular basement membrane. This gives hCG a long (36-hour) circulating half-life in blood, compared to the 0.43-hour circulating half-life of its nonacidic sister hormone, LH [23]. hCG completely lacks the aromatic amino acid called tryptophan. The -subunit has eight aromatic amino acids and the -subunit has just five, making hCG and its subunits deficient in aromatic amino acids and difficult to detect by absorbance at 280 nm (extinction coefficient 1.41 104 M1 cm1) [24]. The sequence of the C-terminal extension on the -subunit of hCG (the site of the four O-linked oligosaccharides) is like a proteoglycan sequence (a polymer of Ser and Pro). The C-terminal sequence formed between residues 118 and 145 consists of 27 amino acids, including 8 Ser and 8 Pro residues. The high Pro content and the four O-linked oligosaccharides on the C-terminal peptide prevent this region from significant folding. At this point, it is worth noting the amino acid sequence of hCG (Figure 5.1). Both hCG and hyperglycosylated hCG (hCG-H), two independent molecules, share a common amino acid sequence [6]. They are, however, separate molecules with completely different oligosaccharide structures [6].
5.2 Carbohydrate Structure of hCG The structure of the O- and N-linked oligosaccharides attached to hCG is also a story of refinement. Originally, the N- and O-linked structures were determined using classic biochemical methods [5–8]. In recent years, oligosaccharides at specific attachment sites and their effect on molecular weight has been examined by mass spectrometry [9]. Biochemical studies of hCG, or pregnancy hCG, show that the average pregnancy sample -subunit contains 86% trisaccharide/tetrasaccharide-type O-linked oligosaccharides and 15% hexasaccharide/pentasaccharide-type structures
(1) ala-pro-asp-val-gln-asp-cys-pro-glu-cys-thr-leu-gln-glu-asp-pro-phe-phe-ser-gln-pro-gly-ala-pro-ile (26) leu-gln-cys-met-gly-cys-cys-phe-ser-arg-ala-tyr-pro-thr-pro-leu-arg-ser-lys-lys-thr-met-leu-val-gln N (51) lys-asn-val-thr-ser-glu-ser-thr-cys-cys-val-ala-lys-ser-tyr-asn-arg-val-thr-val-met-gly-gly-phe-lys (76) val-glu-asn-his-thr-ala-cys-his-cys-ser-thr-cys-tyr-tyr-his-lys-ser (92) N β-subunit of regular hCG N N (1) ser-lys-glu-pro-leu-arg-pro-arg-cys-arg-pro-ile-asn-ala-thr-leu-ala-val-glu-lys-glu-gly-cys-pro-val-cys-ile-thr-val-asn-
Structure, Synthesis, Secretion, and Function of hCG
-subunit of hCG
(31) thr-thr-ile-cys-ala-gly-tyr-cys-pro-thr-met-thr-arg-val-leu-gln-gly-val-leu-pro-ala-leu-pro-gln-val-val-cys-asn-tyr-arg (61) asp-val-arg-phe-glu-ser-ile-arg-leu-pro-gly-cys-pro-arg-gly-val-asn-pro-val-val-ser-tyr-ala-val-ala-leu-ser-cys-gln-cys(91) ala-leu-cys-arg-arg-ser-thr-thr-asp-cys-gly-gly-pro-lys-asp-his-pro-leu-thr-cys-asp-asp-pro-arg-phe-gln-asp-ser-ser-ser (121) ser-lys-ala-pro-pro-pro-ser-leu-pro-ser-pro-ser-arg-leu-pro-gly-pro-ser-asp-thr-pro-ile-leu-pro-gln (145) O O O O
Figure 5.1 The corrected amino acid sequence of hCG as shown by Morgan et al. [3]. The symbols N and O mark the sites of the N- and O-linked oligosaccharides on hCG subunits.
51
52
Human Chorionic Gonadotropin (hCG)
Figure 5.2 O-linked oligosaccharides on hCG and hCG-H. (A) Simple; (B) complex.
(Figure 5.2) [5,6]. Sialic acid studies show that the average O-linked oligosaccharides contain 1.25 pmol/pmol sialic acid, which explains the trisaccharide/tetrasaccharide and hexasaccharide/pentasaccharide balance. Recent mass spectrometry studies show that all hCG molecules contain one hexasaccharide/pentasaccharide structure at residue 121, regardless of the source of hCG [9]; however, if this were true, the average pregnancy hCG would contain 25% hexasaccharide/pentasaccharide, rather than the 15% indicated by biochemical studies [6]. When we examine the N-linked oligosaccharides on the -subunit, we find that the average pregnancy hCG contains 49% structure SM, 37% structure SS, 7.3% structure SSF, 4.5% structure SSSF, and 2.4% structure SSS (Figure 5.3) [6]. Examination of the N-linked oligosaccharides attached to the -subunit shows that the average pregnancy hCG contains 47% structure SSF, 31% structure SS, 12.6% structure SSSF, 5.4% structure SM, and 3.8% structure SSS. As such, the pregnancy -subunit contains 6.9% triantennary and the -subunit contains 16% triantennary oligosaccharides.
5.3 hCG Primary Structure The primary structure of hCG consists of a single linear amino acid sequence -subunit of 145 residues [13]. This is larger than the amino acid sequence of LH
Structure, Synthesis, Secretion, and Function of hCG
53
Figure 5.3 N-linked oligosaccharides on hCG and hCG-H. (A) Biantennary; (B) triantennary.
by 30 amino acids, as a result of the presence of a unique, serine-rich, 30-amino-acid carboxy-terminal extension on the hCG peptide chain. This is a result of readthrough into the 3’ untranslated region of the LH gene [14]. Within the primary sequence, 12 cysteine residues provide the thiol groups for the six disulfide bridges that form during initial folding. The serines within the carboxy-terminal extension also provide the anchors for the four O-linked glycosylations (Figure 5.1).
54
Human Chorionic Gonadotropin (hCG)
A high degree of sequence homology has been maintained between the first 115 amino acids of hCG and the other three -subunits of the glycoprotein hormone family (LH, 82%; TSH, 46%; FSH, 36%) (Figure 5.1). In each case, the most conserved regions occur around the common 12 cysteine residues [15].
5.4 hCG Secondary Structure The exact nature of the molecular folding patterns within hCG remained unresolved for 20 years after the initial sequence was revealed. Many attempts at crystallization were impeded by the high degree of complex glycosylations found on the molecule. Although two studies successfully deglycosylated hCG and obtained viable crystals, X-ray diffraction was not performed to an adequate resolution [16,17]. Initial studies on the secondary folding patterns of molecular hCG were reviewed by Pierce and Parsons [15]. It was estimated that, in the intact holohormone, only 5–8% of the molecule consisted of -helices, and that a much higher percentage (25–40%) was accounted for by -pleated regions. Using methods similar to those of Harris and coworkers, Lapthorn et al. grew hCG crystals from ammonium sulfate solutions, following deglycosylation with anhydrous hydrofluoric acid. Electron density maps gave rise to tracings of the - and -subunits, including amino acids 5–89 of hCG and 2–111 of hCG. The remaining 34 amino acids of the hCG carboxy-terminal extension were not visible and were therefore believed to adopt a random conformation [12]. It was later revealed that there are three distinct areas of -pleated structures in each subunit: two -sheet bulges (62–65 and 79–82 in the -subunit; 59–62 and 85–88 in the -subunit) and a loop of doublestranded -sheet-like structure which, in hCG, is punctuated with two helix turns involving residues 38–46.
5.5 hCG Tertiary Structure Many attempts at cysteine pairing have been made since hCG was first sequenced. Although there was an agreement on three of the disulfide pairs within hCG (Cys 23–72, Cys 26–110, and Cys 93–100) and three disulfide pairs within hCG (Cys 11–35, Cys 32–64, and Cys 63–91) [18,19], other possible hCG and hCG disulfide bonds remained contested. Reduction and alkylation studies [18,20], receptor binding studies [21,22], antigenic mapping studies [23], and cysteine mutagenic studies [24] all failed to adequately confirm the exact disulfide pairing. It was only when the crystal structure was established in 1994 that the complete disulfide linkages were revealed [12]. Lapthorn found that the Cys–Cys assignments of Mise and Bahl [18], in which four of the assigned disulfides in the -subunit were correct, were the most accurate. The six disulfide pairings for hCG were established as Cys 9–57, Cys 23–72, Cys 26–110, Cys 34–88, Cys 38–90, and Cys 93–100; the five for hCG were established as Cys 7–31, Cys 10–60, Cys 28–82, Cys 32–84, and Cys 59–87 [12]. Three of the
Structure, Synthesis, Secretion, and Function of hCG
55
Figure 5.4 The cystine knot structures on the (A) hCG -subunit and (B) hCG -subunit (right).
cystine bonds formed produce a central cluster that has been termed a “cystine knot” structure [12,25]. This structure provides the molecule with its characteristic folding pattern and stabilizes the formation of a single, long, loop-like structure set opposite to two similarly shaped hairpin loops. The positions of the three cystines in the cystine knot is almost identical in both subunits. Two disulfide bridges (linking residues 34–88 and 38–90 on the -subunit and 28–82 and 32–84 on the -subunit) join the antiparallel strands of the peptide chain, forming a central loop through which the third disulfide passes (linking residues 9–57 on the -subunit and 10–60 on the -subunit) (Figure 5.4). The long loop of the -subunit maintains its shape as a result of the main chain hydrogen bonds of the -sheet and -helix-like structures mentioned previously. Studies by Lustbader and colleagues [17] show that three disulfide linkages are quickly added to the -subunit before synthesis is complete in the endoplasmic reticulum. These disulfide linkages are 9–57, 34–88, and 38–90. Three more disulfide linkages are later added to the -subunit at 23–72, 93–100, and 26–110. Within the -subunit, however, the corresponding structure is stabilized with side-to-main chain hydrogen bonding from Arg 43 to Pro 50 and leucine (Leu) 52, and from Gln 54 to Met 41. In the -subunit, the remaining two cystines keep the C- and N-terminal sequences away from the central body of the molecule. In the -subunit, another cystine (Cys 23–72) maintains a tight association between the opposing hydrophobic and hydrophilic regions of the hairpin loops. The -subunit forms another unique structure, which has been termed the “seat belt,” at residues 70–105. This region bends away from the central knot as a result of the presence of another disulfide (Cys 93–100) which is then attached to the first hairpin loop by the final crosslink (Cys 26–110) (Figure 5.5).
5.6 hCG Quaternary Structure With regard to subunit interactions and receptor binding, the three-dimensional structural patterns of hCG have been extensively reviewed [26,27], confirmed, and updated [12].
56
Human Chorionic Gonadotropin (hCG)
Figure 5.5 The hCG -subunit seat belt. Image shows the crystal structure of deglycosylated hCG missing the -subunit C-terminal segment as shown by Lapthorn and colleagues [12]. Black line is -subunit and gray line is -subunit.
The “cystine knot” structure and the binding loop regions are stabilized by it, which accounts for both structural and functional interactions (Figure 5.5). The subunits come together in a “top-to-tail” orientation, and although the - and -subunits dimerize, they are not joined by covalent bonds. Instead, the -subunit sits in a pocket of the -subunit and is held in place by the “seat belt” (Figure 5.5). Site-directed mutagenic studies of the conserved cysteine residues have shown that interruption of any -subunit disulfide bond prevents the formation of an intact hCG heterodimer [24,26]. The explanation for the necessity of the central cystines became apparent following the publication of the crystal structure [12]. It was revealed that the -sheet segments, held close to the cystine knot in each subunit, came together in the hetero dimer to form a short, seven-stranded -barrel, at the center of which lies 34–36 [12], a sequence considered essential for dimer formation [27] (Figure 5.5). Although they are not covalently linked, the subunits are held in very close proximity to each other, burying a total surface area of 4525 Å2 [12] or 3860 Å2 [28]. This tight association is maintained almost entirely through hydrogen bonds, the majority of which are located between the hairpin/long loop interfaces of each subunit. The most important feature in maintaining heterodimer integrity, however, is probably the seat belt (Figure 5.5). A short strand of -sheet between the inner side of the seat belt and the -subunit form part of the central barrel, and the Cys 26–110 bridge is believed to “click” into place, thus “hugging” the -subunit and finalizing – dimerization. Chemical modification studies of both subunits indicated that Tyr 41 and Tyr/Phe 69 on the -subunit, and tyrosines or phenylalanines on the -subunit, are conserved in -subunit regions (e.g., 37, 61, and 84). These are located in areas of subunit interaction and hence implicate binding. These tyrosines are protected from modification in the intact hCG molecule [15]. Huang and Puett [29] investigated Gln 54,
Structure, Synthesis, Secretion, and Function of hCG
57
which is conserved in all -subunits, and noted that lysine (Lys) 54 mutants did not associate with the -subunit and that Glu 54 only gave rise to a slight reduction in hormonal activity.
5.7 Combination of hCG Subunits Saccuzzo et al. [30] examined the combination of hCG subunits. Their work showed that the rate-limiting step for combination of - and -subunits is the rate of disulfide linkage formation in the -subunit. Approximately 70% of the free -subunit made in a trophoblast cell is degraded and not used for combination. Only 50% of the nondegraded free -subunit is bound. One explanation for incomplete dimer formation appears to be biochemical differences between the free and combined -subunits that limit combination of free -subunit [30]. As explained in Chapter 7 of this book, free -subunit and free -subunit are both hyperglycosylated, which may limit combination.
5.8 Synthesis and Secretion of hCG hCG is produced in syncytiotrophoblast cells, whereas the variant hCG-H is made in cytotrophoblast cells [31]. Both have the same amino acid sequence [6], suggesting that both originate from the same - and -subunit genes. hCG -subunit is encoded by one gene on chromosome 6q21,1-23 [32]. There is a single gene coding for the common hCG, LH, FSH, and TSH -subunit. In contrast, hCG -subunit is coded for by as many as eight genes on chromosome 19q13.3 [33]. As detailed in Chapter 4, genes 5, 3, and 8 are natural promoters with high transcriptional activity. The 5’ flanking region of the five other genes contains gaps and deletion, making them less commonly transcribed [33]. It is assumed that the -subunit amino acid sequence (shown in Figure 5.1) derives from genes 5, 3, and 8. Cis and trans gene activators of hCG transcription include activating protein-2, cAMP response element binding, upstream regulatory element, and cAMP response element [34]. Promoters of hCG expression include glucocorticoids, cAMP, dehydroepiandrosterone, and gonadotropin stimulating hormone [34]. As hCG subunits are translated, dolichol adds a standard N-linked oligosaccharide structure to two asparagine residues on the -subunit (residue 52 and 78) and -subunit (residue 13 and 30), the GlcNAc 2, Man 9, Glc 3 structure [35]. These are further processed after translation of the subunit is completed in the endoplasmic reticulum. N-linked oligosaccharide processing continues in the Golgi apparatus prior to the rapid secretion of hCG in secretory granules [35]. Disulfide bridge formation in hCG subunits also starts in the endoplasmic reticulum before translation of the subunits is complete. The disulfide bridge formation continues after translation [30]. Addition of all disulfide bridges on the - and -subunits is thought to limit complete folding of the structure [30]. Finally, O-linked oligosaccharides are added to the C-terminal peptide of the -subunit of hCG, thus completing hCG synthesis in the Golgi apparatus [30,35].
58
Human Chorionic Gonadotropin (hCG)
hCG is rapidly secreted in unique syncytiotrophoblast secretory granules formed in the Golgi apparatus [36,37]. hCG is not accumulated in cells. It is only rapidly secreted [37]. In fact, it is rapidly secreted in a pulsatile manner [38]. Figure 5.6 shows the hCG concentrations and total hCG concentrations in serum and urine through the length of pregnancy, as observed by the USA hCG Reference Service [39]. As shown, hCG production reaches a peak in both media at 10 weeks of gestation (since last menstrual bleeding). In both serum and urine, hCG-H is first detected at 3 weeks. During this time, hCG-H concentration is much higher than hCG concentration. This is due to the predominant production of hCG-H in the first 3 weeks of gestation [39]. Similarly, total hCG is approximately twice as high as hCG in urine during the second and third trimesters of pregnancy. This is due to detection of -core fragment, which can predominate in urine at this time. Urine hCG results are generally lower than serum hCG results, particularly during the second and third trimesters of pregnancy.
Figure 5.6 Solid circles are concentration of total hCG (hCG plus hCG-H and free -subunit); open circles are hCG only throughout the length of pregnancy (determined by subtracting hCG-H and free -subunit). (A) Concentration in 345 serum samples; (B) concentration in 789 urine samples.
Structure, Synthesis, Secretion, and Function of hCG
59
hCG production during pregnancy is generally regulated by differentiation of cytotrophoblast cells and hCG-promoted differentiation [45]. In early pregnancy, hCG concentration is initially proportional to the growing mass of cytotrophoblast and syncytiotrophoblast cells (Figure 5.5). Due to hCG-promoted differentiation and the accumulating mass of syncytiotrophoblast cells (the hCG-producing cells), hCG levels rise rapidly during 6–9 weeks of gestation (Figure 5.5). At 10 weeks, hCG reaches a peak. This peak is somewhat artificial or a cell differentiation artifact, because it is due to the increasing production of hCG with increasing syncytiotrophoblast mass, combined with the growing differentiation of syncytiotrophoblast cells with 3–4 nuclei into terminally differentiated cells or a syncytium with 30–50 nuclei. These terminal cells are poor producers of hCG because they are all nuclei and cytoplasm limited. This drives down overall hCG production, starting at the 10-week peak and continuing until term (Figure 5.5). The 10-week peak represents a combination of rising hCG from syncytiotrophoblast cells and declining hCG from those supernucleated cells forming a syncytium.
5.9 Functions of hCG hCG evolved from LH and has LH-like functions, such as binding a common LH and hCG receptor. As shown in 1919, hCG takes over promotion of progesterone production from LH by corpus luteal cells very early in pregnancy, preventing both progesterone failure and menstrual bleeding [43,44]. This function continues for 3 weeks, and begins at the time of blastocyst implantation (first hCG production). At approximately 6 weeks of gestation (weeks following start of the last menstrual bleeding), enough syncytiotrophoblast cells are present in the placenta to directly make sufficient progesterone and take over progesterone production. hCG has a clear function in enhancing differentiation of cytotrophoblast cells to multinucleated syncytiotrophoblast cells with 3–5 nuclei during the course of pregnancy [45]. In the final stages of differentiation promoted by hCG, a syncytium is formed from syncytiotrophoblast cells with 30–50 nuclei. Syncytiotrophoblast cells with 3–5 nuclei are the optimal hCG producers. hCG also plays a key role throughout the length of pregnancy. It promotes angiogenesis in the maternal vasculature [46–49] so that maximal maternal blood supply can reach the invading placenta. This leads to efficient nutrient exchange with the fetus. The presence of hCG/LH receptor in fetal organs suggest that hCG also has roles in fetal development [50,51] (see Chapter 11). Similarly, hCG receptors have been found on decidual tissue, suggesting a role in blastocyst–decidua interaction [50]. More specifically, hCG receptors have been found on the endometrial epithelium, suggesting roles in either pregnancy initialization or endometrial–blastocyst interaction. Other studies suggest that hCG plays a role in decidualization of endometrial stromal cells and immune suppression at the blastocyst–endometrial implantation site [50]. Other roles for hCG include uterine enlargement to accommodate fetal growth and suppression of myometrial muscles, blocking contractions during the course of pregnancy. The full biological function of hCG is still being resolved (see Chapter 11) [50].
60
Human Chorionic Gonadotropin (hCG)
hCG-H has functions independent of hCG. hCG-H promotes invasion of the decidua and myometrium by cytotrophoblast cells [40–42]. It also inhibits apoptosis and promotes the growth of cytotrophoblast columns to form villi. Cytotrophoblast cells in villi are differentiated by hCG into active cells in maternal–fetal nutrition uptake (syncytiotrophoblast cells). It could be said that hCG-H and hCG function together to promote the growth of trophoblast villi. The two molecules (hCG and hCG-H) seemingly work side by side in maximizing maternal–fetal interaction and placental function during pregnancy (see Chapter 3).
References [1] Carlsen RB, Bahl OP, Swaminathan N. Human chorionic gonadotropin: linear amino acid sequence of the -subunit. J Biol Chem 1973;248:6810–27. [2] Bellisario R, Carlsen RB, Bahl OP. Human chorionic gonadotropin: linear amino acid sequence of the -subunit. J Biol Chem 1973;248:6796–809. [3] Morgan FJ, Birken S, Canfield RE. The amino acid sequence of human chorionic gonado tropin. The alpha subunit and beta subunit. J Biol Chem 1975;250:5247–58. [4] Bahl OP. Human chorionic gonadotropin: II. Nature of the carbohydrate units. J Biol Chem 1969;244:575–83. [5] Endo Y, Yamashita K, Tachibana Y, Tojo S, Kobata A. Structures of the asparagine-linked sugar chains of human chorionic gonadotropin. J Biochem 1979;85:669–79. [6] Elliott MM, Kardana A, Lustbader JW, Cole LA. Carbohydrate and peptide structure of the - and -subunits of human chorionic gonadotropin from normal and aberrant pregnancy and choriocarcinoma. Endocrine 1997;7:15–32. [7] Kessler MJH, Mise T, Ghai RD, Bahl OP. Structure and location of the O-glycosidic carbohydrate units of human chorionic gonadotropin. J Biol Chem 1979;254:7909–14. [8] Cole LA, Birken S, Perini F. The structures of the serine-linked sugar chains on human chorionic gonadotropin. Biochem Biophys Res Comm 1985;126:333–9. [9] Valmu L, Alfthan H, Hotakainen K, Birken S, Stenman UH. Site-specific glycan analysis of human chorionic gonadotropin -subunit from malignancies and pregnancy by liquid chromatography–electrospray mass spectrometry. Glycobiol 2006;16:1207–18. [10] Giudice LC, Pierce JG. Studies on the disulfide bonds of glycoprotein hormones. J Biol Chem 1979;254:1164–9. [11] Mise T, Bahl OP. Assignment of disulfide bonds in the -subunit of human chorionic gonadotropin. J Biol Chem 1981;256:6587–92. [12] Lapthorn AJ, Harris DC, Littlejohn A, Lustbader JW, Canfield RE, Machin KJ, et al. Crystal structure of human chorionic gonadotrophin. Nature 1994;369:455–61. [13] Carlsen RB, Bahl OP, Swaminathan N. hCG: linear amino acid sequence of the -subunit. J Biol Chem 1973;248:6810–27. [14] Fiddes JC, Talmage K. Structure, expression and evolution of the genes for the human glycoprotein hormones. Recent Prog Hormone Res 1984;40:43–74. [15] Pierce JG, Parsons TF. Glycoprotein hormones: structure and function. Ann Rev Biochem 1981;50:465–95. [16] Harris DC, Machin KJ, Evin GM, Morgan FJ, Isaacs NW. Preliminary X-ray diffraction analysis of human chorionic gonadotropin. J Biol Chem 1989;264:6705–9.
Structure, Synthesis, Secretion, and Function of hCG
61
[17] Lustbader JW, Birken S, Pileggi NF, Kolks MA, Pollak S, Cuff ME, et al. Crystallization and characterization of human chorionic gonadotropin in chemically deglycosylated and enzymatically desialylated states. Biochem 1989;28:9239–43. [18] Mise T, Bahl OP. Assignments of disulfide bonds in the subunit of hCG. J Biol Chem 1981;256:6587–92. [19] Willey KP, Leidenberger F. Functionally distinct agonist and receptor binding regions in hCG. J Biol Chem 1989;264:19716–29. [20] Cornell JS, Pierce JG. Studies on the disulfide bonds of glycoprotein hormones. Locations in the alpha chain based on partial reductions and formation of 14C-labeled S-carboxymethyl derivatives. J Biol Chem 1974;249:4166–74. [21] Keutmann HT, Charlesworth MC, Mason KA, Ostrea T, Johnson L, Ryan RJ. A receptorbinding region in human choriogonadotropin/lutropin beta subunit. Proc Natl Acad Sci 1987;84:2038–42. [22] Ryan RJ, Charlesworth MC, McCormick DJ, Milius RP, Keutmann HT. The glycoprotein hormones: recent studies of structure–function relationships. Fed Am Soc Exp Biol 1988;2:2661–9. [23] Moyle WR, Matzuk MM, Campbell RK, Cogliani E, Dean-Emig DM, Krichevsky A, et al. Localization of residues that confer antibody binding specificity using human chorionic gonadotropin/luteinizing hormone beta subunit chimeras and mutants. J Biol Chem 1990;265:8511–18. [24] Suganuma N, Matzuk MM, Boime I. Elimination of disulfide bonds affects assembly and secretion of the hCG subunit. J Biol Chem 1989;264:19302–7. [25] Murray-Rust J, McDonald NQ, Blundell TL, Hosang M, Oefner C, Winkler F, et al. Topological similarities in TGF-P2, PDGF-BB and NGF define a superfamily of polypeptide growth factors. Structure 1993;1:53–159. [26] Ryan RJ, Keutmann HT, Charlesworth MC, McCormick DJ, Milius RP, Calvo FO, et al. Structure–function relationships of gonadotrophins. Recent Prog Hormone Res 1987;43:383–417. [27] Chen F, Puett D. A single amino acid residue replacement in the beta subunit of human chorionic gonadotrophin results in the loss of biological activity. J Mol Endocrinol 1992;8: 87–9. [28] Wu H, Lustbader JW, Liu Y, Canfield RE, Hendrickson WA. Structure of human chorio nic gonadotropin at 2.6 Å resolution from MAD analysis of the selenomethionyl protein. Structure 1994;2:545–58. [29] Huang J, Puett D. On the role of the invariant glutamine at position 54 in the human choriogonadotropin beta subunit. Mol Cell Biochem 1994;136:183–6. [30] Saccuzzo B, Huth JR, Ruddon RW. Combination of the chorionic gonadotropin free -subunit with . Endocrinol 1990;126:384–91. [31] Xing Y, Williams C, Campbell RK, Cook S, Knoppers M, Addona T, et al. Threading of a glycosylated protein loop through a protein hole: implications for combination of human chorionic gonadotropin subunits. Protein Sci 2001;10:226–35. [32] Fiddes JC, Goodman HC. The gene encoding the common alpha subunit of four human glycoprotein hormones. J Mol Appl Genet 1981;1:3–10. [33] Boorstein WR, Vamvakopoulos NC, Fiddes JC. Human chorionic gonadotropin betasubunit is encoded by at least eight genes arranged in tandem and inverted pairs. Nature 1982;300:419–22. [34] Becker KL, Bilezikian JP. Endocrinology of trophoblastic tissue. In: Principle and practice of endocrinology. Ed: Becker KL. Lipcott W. and Wilkins. 2004:1096–100.
62
Human Chorionic Gonadotropin (hCG)
[35] Hannover JA, Elting J, Mintz GR, Lennarz WJ. Temporal aspects of the N- and O-glycosylation of human chorionic gonadotropin. J Biol Chem 1982;257:10172–7. [36] Morrish DW, Marusyk H, Olivia S. Demonstration of specific secretory granules for human chorionic gonadotropin in placenta. J Histochem Cytochem 1987;35:93–101. [37] Ruddon RW, Hanson CA, Addison NJ. Synthesis and processing of human chorionic gonadotropin subunits in cultured choriocarcinoma cell. Proc Natl Acad Sci 1979;76:5143–7. [38] Odell WD, Griffin J. Pulsatile secretion of human chorionic gonadotropin in normal adults. N Engl J Med 1987;317:1688–91. [39] Cole LA. hCG Tests. Expert Rev Mol Diag 2009;9:721–49. [40] Cole LA, Dai D, Leslie KK, Butler SA, Kohorn EI. Gestational trophoblastic diseases: 1. Pathophysiology of hyperglycosylated hCG-regulated neoplasia. Gynecol Oncol 2006;102:144–9. [41] Cole LA, Khanlian SA, Riley JM, Butler SA. Hyperglycosylated hCG (hCG-H) in gestational implantation, and in choriocarcinoma and testicular germ cell malignancy tumorigenesis. J Reprod Med 2006;51:919–29. [42] Handshuh K, Guibourdenche J, Tsatsaris V, Guesnon M, Laurendeau I, Evain-Brion D, et al. Human chorionic gonadotropin produced by the invasive trophoblast but not the villous trophoblast promotes cell invasion and is down-regulated by peroxisome proliferator-activated receptor-a. Endocrinol 2007;148:5011–19. [43] Hirose T. Experimentalle histologische studie zur genese corpus luteum. Mitt Med Fakultd Univ ZU (Tokyo) 1919;23:63. [44] Hirose T. Exogenous stimulation of corpus luteum formation in the rabbit: influence of extracts of human placenta, decidua, fetus, hydatid mole, and corpus luteum on the rabbit gonad. J Jpn Gynecol Soc 1920;16:1055. [45] Shi QJ, Lei ZM, Rao CV, Lin J. Novel role of human chorionic gonadotropin in differentiation of human cytotrophoblasts. Endocrinol 1993;132:1387–95. [46] Lei ZM, Reshef E, Rao CV. The expression of human chorionic gonadotropin/luteinizing hormone receptors in human endometrial and myometrial blood vessels. J Clin Endocrinol Metab 1992;75:651–9. [47] Herr F, Baal N, Reisinger K, Lorenz A, McKinnon T, Preissner KT, et al. HCG in the regulation of placental angiogenesis. Results of an in vitro study. Placenta 2007;28(Suppl A):S85–93. [48] Zygmunt M, Herr F, Munstedt K, Lang U, Liang OD. Angiogenesis and vasculogenesis in pregnancy. Euro J Obstet Gynecol Reprod Biol 2003;110(Suppl 1):S10–S18. [49] Toth P, Lukacs H, Gimes G, Sebestyen A, Pasztor N, Paulin F, et al. Clinical importance of vascular LH/hCG receptors: a review. Reprod Biol 2001;1:5–11. [50] Rao CV. Physiological and pathological relevance of human uterine LH/hCG receptors. J Soc Gynecol Invest 2006;13:77–8. [51] Lei ZM, Rao CV, Kornyei J, Licht P, Hiatt ES. Novel expression of human chorio nic gonadotropin/luteinizing hormone receptor gene in brain. Endocrinol 1993;132: 2262–70.
6 Comparison of the Structures of hCG and Hyperglycosylated hCG Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
It has taken 23 years to fully resolve the structure of hyperglycosylated hCG (hCG-H). The search began in 1983, when Mizuochi et al. used basic biochemistry methods to show a difference between the N-linked oligosaccharides on hCG in urine from pregnancies and urine from choriocarcinoma cases [1]. In 1987, a second chapter was added to the story when Cole used similar basic biochemistry methods to demonstrate that the O-linked oligosaccharides were the major carbohydrate difference between the pregnancy and choriocarcinoma molecules [2]. Recently, the story neared completion as Valmu et al. conducted mass spectrometry studies with hCG peptides which showed the site specificity of oligosaccharide structures [3]. hCG-H is a form of hCG dimer in which larger oligosaccharides predominate, accounting for up to 41% of the molecular weight of hCG (25%–30% of regular hCG, and 35–41% of hyperglycosylated hCG); triantennary N-linked oligosaccharides and hexasaccharide/pentasaccharide (Figure 6.1) O-linked oligosaccharides predominate in early pregnancy and choriocarcinoma cases [1–4]. In 1997, Elliott et al. conducted the most thorough hCG oligosaccharides study of its time, examining 18 individual urine samples from pregnancy or from choriocarcinoma [4]. The significantly large number of cases used in this study confirmed all previous findings regarding Nand O-linked oligosaccharides (Figures 6.1 and 6.2). It also showed that the - and -subunit peptide sequences on choriocarcinoma hCG were exactly the same as those on pregnancy hCG, but that choriocarcinoma-hCG molecules were more extensively cleaved at sites on the - and -subunit peptide. hCG-H is the principal form of hCG made in gestational trophoblastic neoplasms and during implantation of pregnancy [4–8]. As described elsewhere in this book (Chapter 13), hCG-H has biological functions separate from those of hCG. hCG-H promotes implantation in pregnancy and invasion in gestational trophoblastic neoplasms. In this chapter, we examine the peptide and carbohydrate structures of hCG-H.
6.1 Peptide Structure of hCG-H Research indicates that the peptide sequence of hCG-H (choriocarcinoma hCG) is the same as that of regular hCG [4,9]. In 1988 [10], Nishimura and colleagues showed that Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00006-2 © 2010 Elsevier Inc. All rights reserved.
64
Human Chorionic Gonadotropin (hCG)
Figure 6.1 O-linked oligosaccharides on hCG and hCG-H. (A) Simple; (B) complex.
only choriocarcinoma hCG (hCG-H) was cleaved/nicked in the center of the -subunit. This was confirmed by others [4,11,12] and shown to occur primarily at 47–48 and, to a lesser extent, at 43–44 and 44–45 [12] (Table 6.1). Other studies showed an -subunit cleavage on hCG-H (choriocarcinoma hCG) starting at residue 3 or 4 [12]. Elliott et al. [4] thoroughly investigated these - and -subunit cleavages in two standards, eight individual pregnancies, four individual hydatidiform moles, and six individual choriocarcinoma-hCG preparations isolated from urine. As shown in Table 6.1, choriocarcinoma hCG is cleaved an average of 17% on the -subunit, whereas pregnancy hCG is cleaved only 4.2% on the -subunit. A significant difference (P 0.0048) was shown between pregnancy- and choriocarcinoma-hCG molecules. Similarly, the -subunit is 64% nicked in choriocarcinoma hCG, but only 13% nicked in pregnancy hCG (Table 6.1). A significant difference (P 0.0037) was shown between pregnancy- and choriocarcinoma-hCG molecules. Nicking/cleavage of hCG at these sites can be attributed to macrophage or leukocyte elastase (see Chapter 9) [12]. As shown, macrophages are more abundant around choriocarcinoma or other malignancies and lead to greater cleavage of released hCG. Interestingly, individual pregnancy-hCG samples are only nicked/cleaved at 47–48, whereas choriocarcinomahCG molecules are cleaved at 42–43, 43–44, and 44–45 (see Table 6.1). It is suggested that choriocarcinoma cells produce hCG-H. The large sugar sidechain on hCG-H seems to open the molecule, possibly making it more susceptible to protease cleavage. This might expose some sites not normally cleaved on hCG.
Comparison of the Structures of hCG and Hyperglycosylated hCG
65
Figure 6.2 N-linked oligosaccharides on hCG and hCG-H. (A) Biantennary; (B) triantennary.
For example, the average 64% nicking on choriocarcinoma-hCG molecules (hCG-H) changes the molecule’s three-dimensional shape. A loop on the -subunit is formed by Cys residues 38 [38–90] and 57 [9–57]. Arg residue 43 charges one side of this loop. If hCG-H is cleaved at 43–44, 44–45, or 47–48, one arm becomes charged and the second arm becomes neutral or hydrophobic. The arm with Arg residue 43 will then move toward the surface of the molecule, whereas the neutral hydrophobic arm will
Table 6.1 Distribution of Cleavages in the Amino Acid Sequence of - and -Subunits of hCG and hCG-H in Two Standards and 18 Individual Cases with Pregnancy, Hydatidiform Mole, and Choriocarcinoma [4]. Proportion of Molecules Cleaved is Indicated as Percentage. -Subunit N-Terminal Sequence Analysis (%)
Sample Code Properties 3
4
-Subunit N-Terminal Sequence Analysis (%)
% Cleaved
% Cleaved
43
71
3, 4, 43, 71
44
45
48
76
44, 45, 48, 76
0 4
0 0
12 14
0 0
7 4
12 0
0 0
19 4
0 0 0 0 0 0 0 0 0.0
0 0 0 0 0 0 0 0 0.0
0.0 0.0 7.4 0.0 0.0 18 6.5 5.6 4.2
0 0 0 0 0 0 0 0 0.0
0 0 0 0 0 0 0 0 0.0
0 17 29 29 27 0 0 0 19
0 0 0 0 0 0 0 0 0.0
0.0 17 29 29 27 0.0 0.0 0.0 13
A. Pregnancy-hCG Standards CR127 CR129
2 0
12 11
B. Pregnancy, Weeks of Gestationa P1, 8 weeks P2, 6 weeks P3, 9 weeks P5, 7 weeks P6, 7 weeks P7, 9 weeks P8, 9 weeks P9, 8 weeks Mean
0 0 0 0 0 22 0 0 2.4
0 0 8 0 0 0 7 6 2.3
C. Hydatidiform Mole, Complete or Partial M1, complete M2, complete M3, partial M4, complete Mean
6 5 3 0 3.5
0 7 5 8 5.0
0 0 0 0 0.0
0 0 0 4 1.0
5.6 11 7.4 11 8.8
9 4 0 0 3.3
0 0 0 0 0.0
9 3 39 98 37
6 8 0 0 3.5
24 15 39 98 44
0 19 0 0 0 0.10 4.8
0 0 0 0 0 0 0.0
15 32 8.3 13 14 17 17
23 51 0 0 0 0 12
0 0 0 0 0 0.03 0.5
39 45 24 100 100 0 51
0 0 0 0 0 0 0.0
62 96 24 100 100 3.0 64
D. Choriocarcinoma, Metastasesa C1, lung C2, lung, brain C3, lung, brain C4, lung, brain C5, lung, brain C7, lung, brain Mean a
6 0 9 5 3 5 4.7
12 15 0 10 13 5 9.2
Significant difference observed in cleavage of -subunit at residue 4 (P 0.012), in the proportion -subunit cleavage (%) (P 0.0048), and in the proportion -subunit cleavage (%) at 48 (P 0.033), and total -subunit cleavage (P 0.0037).
68
Human Chorionic Gonadotropin (hCG)
bury itself in the hydrophobic region of the molecule, thus reshaping the molecule. As noted in publications, nicked molecules have no biological activity at the LH/hCG receptor [13], which is consistent with reshaping of the hCG or hCG-H molecule by the nicking process. In summary, the peptide sequence on choriocarcinoma-hCG (hCG-H) molecules is the same as on pregnancy-hCG molecules. The original assumption that nicks or cleavages on the -subunit marked choriocarcinoma hCG was partly correct, in that nicks or cleavages at 43–44 and 44–45 are unique to choriocarcinoma; however, molecules nicked or cleaved at the principal 47–48 site were found on both pregnancy and choriocarcinoma molecules, or both hCG and hCG-H.
6.2 N- and O-Linked Oligosaccharide Structures of hCG-H As mentioned in the introduction to this chapter, Mizuochi et al. used basic biochemistry technologies during hydrazinolysis to release oligosaccharides, and used gel filtration with glycosidases and methylation analyses to show that the structures of the N-linked oligosaccharides on choriocarcinoma hCG were different from those seen in pregnancy hCG [1]. In 1987, Cole [2] first examined O-linked oligosaccharides in choriocarcinoma using -elimination to release oligosaccharides; he used gel filtration with glycosidase to compare structures. A much bigger difference was observed in O-linked oligosaccharide structures between choriocarcinoma and pregnancy hCG (Figure 6.1). Choriocarcinoma molecules are hCG-H, so both studies examined the difference between pregnancy hCG and hCG-H. Others have since confirmed Mizuochi’s and Cole’s findings using different technologies. In 1987, for example, Elliott et al. [4] confirmed these findings in 18-patient samples using N- and O-glycanases to enzymatically release the oligosaccharides with Dionex HPLC glycosidases; oligosaccharide standards were used to analyze the structure. The results of Elliott et al. [4] almost exactly matched the earlier data of Mizuochi et al. and Cole et al. More recently, Valmu et al. [3] used mass spectrometry of hCG tryptic fractions to further confirm the N- and O-linked structural findings. Mass spectrometry permitted them to show site specificities of the oligosaccharides [3]. Their studies showed that the O-linked structures attached to Ser residue 121 were always the pentasaccharide/hexasaccharide structure, which explained the constant pregnancy 12–19% pentasaccharide/hexasaccharide content (Figure 6.1, Table 6.1). Valmu’s studies also showed that structures at Ser residue 138 were always the trisaccharide/ tetrasaccharide structure (Figure 6.1). Their results showed that only the sugars at Ser residues 127 and 132 were changed between pregnancy and choriocarcinoma urine preparations. Biantennary and trisaccharide/tetrasaccharide oligosaccharides are characteristic of hCG and triantennary and hexasaccharide/pentasaccharide oligosaccharides are characteristic of hCG-H. The structures of O-linked oligosaccharides are detailed in Figure 6.1; the N-linked oligosaccharide structures are shown in Figure 6.2. As shown in Table 6.2, the hCG -subunit undergoes the least change between pregnancy, hydatidiform mole, and choriocarcinoma. Pregnancy hCG -subunit has 6.9% triantennary structures, whereas benign hydatidiform mole has 7.7% triantennary
Comparison of the Structures of hCG and Hyperglycosylated hCG
69
Table 6.2 Distribution of N-Linked Oligosaccharides on hCG and hCG-H -Subunits from Pregnancy, Hydatidiform Mole, and Choriocarcinoma [4]. Oligosaccharides SM, SSF, SSM, SS, SSSF, and SSS Are Those Used in Figure 6.2. Sample Code
SM (%)
SSF (%)
SSM (%)
SS (%)
SSSF (%)
SSS (%)
Triantennary (%)
0 0
38 38
0 0
4.8 9.4
4.8 9.4
0 0 0 0 0
36 37 41 29 37
2.6 14 10 0 4.5
0 0 0 0 2.4
2.6 14 10 0.0 6.9
0 0 0 0
32 37 55 41.3
3.6 0 0.5 1.4
3.9 5.9 9.2 6.3
7.5 5.9 9.7 7.7
3.6 3.7 2.2 2.3 5.1 3.4
19 5.7 4.9 3.5 6.8 8.0
6.2 0 0 0 8.0 2.8
17 9.8 2.9 1.2 4.3 7.0
27 14 5.1 3.5 17 13
A. Pregnancy-hCG Standards CR127 CR129
48 50
8.9 3.1
B. Pregnancy, Weeks of Gestationa P3, 9 weeks P7, 9 weeks P8, 9 weeks P9, 8 weeks Mean
54 41 42 62 49
7.3 7.9 7.3 9.4 7.3
C. Hydatidiform Mole, Complete or Partial M1, complete M2, complete M4, complete Mean
43 53 36 44
17 4.4 0 7.0
D. Choriocarcinoma, Metastasesa C1, lung C2, lung, brain C3, lung, brain C5, lung, brain C7, lung, brain Mean
39 57 71 75 57 60
16 24 19 18 20 20
a
A significant difference was observed between pregnancy and choriocarcinoma proportions of SSF (P 0.000003), SSM (P 0.000006), and SS (P 0.000006).
structures and malignant choriocarcinoma has 13% triantennary structures (1.9 times the pregnancy value) [4]. It is noted that the triantennary structure SSM, as shown in Figure 6.2, is found only on the -subunit of choriocarcinoma molecules (Table 6.2). No significant difference in SSM content was observed between pregnancy and choriocarcinoma hCG. In contrast, the hCG -subunit had a much more significant difference in the percentage of triantennary structures (Table 6.3). Pregnancy hCG b-subunit has 16% triantennary structures, benign hydatidiform mole 28% triantennary structures, and malignant choriocarcinoma 53% triantennary structures (3.3 times the pregnancy value) [4]. As again shown, the triantennary structure SSM (Figure 6.2) is found only on the -subunit of choriocarcinoma molecules (Table 6.3). A very significant
70
Human Chorionic Gonadotropin (hCG)
Table 6.3 Distribution of N-Linked Oligosaccharide on hCG and hCG-H -Subunits from Pregnancy, Hydatidiform Mole, and Choriocarcinoma [4]. Oligosaccharides SM, SSF, SSM, SS, SSSF, and SSS Are Those Used in Figure 6.2. Sample Code
SM (%)
SSF (%)
SSM (%)
SS (%)
SSSF (%)
SSS (%)
Triantennary (%)
0 0
35 28
5.4 11
0 0
5.4 11
0 0 0 0
39 23 25 36 31
8.3 13 17 12 12.6
2.1 4.5 4.2 4.2 3.8
10 17 21 16 16
0 0 0
19 23 39 27
17 21 19 19
2.4 11 12 8.5
19 32 31 28
0 0 0 0 9 1.8
15 23 12 4.4 8.0 12
52 38 33 43 35 40
11 10 15 5.0 13 11
63 48 48 48 57 53
A. Pregnancy-hCG Standards CR127 CR129
3.6 2.2
56 59
B. Pregnancy, Weeks of Gestationa P3, 9 weeks P7, 9 weeks P8, 9 weeks P9, 8 weeks Mean
6.5 4.4 1.9 8.8 5.4
44 55 52 38 47
C. Hydatidiform Mole, Complete or Partial M1, complete M2, complete M4, complete Mean
5.5 6.9 8.1 6.8
57 38 23 39
D. Choriocarcinoma, Metastasesa C1, lung C2, lung, brain C3, lung, brain C5, lung, brain C7, lung, brain Mean
6.8 5.9 16 16 11 11
15 23 24 32 24 24
a
A significant difference was observed between pregnancy and choriocarcinoma proportions of SSF (P 0.0012), SS (P 0.0083), SSF (P 0.00029), SSS (P 0.0090), and triantennary proportion (P 0.0000040).
difference (P 0.0000040) was observed between the triantennary structures in pregnancy and choriocarcinoma molecules. The monoantennary structure SM (Figure 6.2) is the principal N-linked oligosaccharide on the choriocarcinoma hCG-H -subunit (mean 60% of structures). The fucosylated biantennary structure SSF (Figure 6.2) accounts for 47% of structures on pregnancy hCG and 39% of structures on hydatidiform mole hCG, but accounts for only 24% of structures on choriocarcinoma-hCG molecules (Table 6.2 and 6.3). In contrast, the fucosylated triantennary structure SSSF (Figure 6.2) balances these percentages out, showing 12.6% of structures in pregnancy, 19% in hydatidiform mole, and 40% in choriocarcinoma.
Comparison of the Structures of hCG and Hyperglycosylated hCG
71
Table 6.4 Distribution of O-Linked Oligosaccharide on hCG and hCG-H -Subunits from Pregnancy, Hydatidiform Mole, and Choriocarcinoma [4]. Oligosaccharides Are Those Shown in Figure 6.2. The Proportions of Trisaccharide to Tetrasaccharide and Pentasaccharide to Hexasaccharide Can Be Calculated from the Sialic Acid Content. For Instance, CR127 Has a Sialic Acid Content of 66%, Suggesting 14% of Structures with Two Sialic Acids (Hexasaccharide and Tetrasaccharide) and 86% of Structures with One Sialic Acid (Pentasaccharide and Trisaccharide). Sample Code
Trisaccharide/ Tetrasaccharide (%)
Hexasaccharide/ Pentasaccharide (%)
Sialic Acid (pmol/pmol)
(%)
19 16
1.32 1.25
66 63
12 19 13 15 15
1.62 1.02 1.46 0.88 1.25
81 51 73 44 62
11 38 20 23
0.45 0.80 0.98 0.74
23 40 49 37
67 48 88 103 68 75
0.52 1.14 0.97 1.15 1.05 0.97
26 57 49 58 53 49
A. Pregnancy-hCG Standards CR127 CR129
82 85
B. Pregnancy, Weeks of Gestationa P3, 9 weeks P7, 9 weeks P8, 9 weeks P9, 8 weeks Mean
88 81 88 85 86
C. Hydatidiform Mole, Complete or Partial M1, complete M2, complete M4, complete Mean
89 62 80 77
D. Choriocarcinoma, Metastasesa C1, lung C2, lung, brain C3, lung, brain C5, lung, brain C7, lung, brain Mean
33 52 12 0 32 26
a
A significant difference was observed between pregnancy and choriocarcinoma proportions of trisaccharide/ tetrasaccharide (P 0.00069) and hexasaccharide/pentasaccharide (P 0.00086).
The principal difference between pregnancy- and choriocarcinoma-hCG sugars is clearly in O-glycosylation (the four O-linked oligosaccharides on the hCG -subunit). As shown in Table 6.4, pregnancy hCG contains 15% of hexasaccharide/ pentasaccharide structures, whereas choriocarcinoma hCG has 75% of these structures (5 times the pregnancy values). A very significant difference of P 0.00069 is noted between pregnancy and choriocarcinoma hCG [4].
72
Human Chorionic Gonadotropin (hCG)
When compared to serum samples, all the N- and O-linked oligosaccharide structures in urine are deficient in sialic acid [14]. This leads to the conclusion that sialic acid-containing structures are removed through the liver. As the published literature shows [14], both serum molecules and the molecules secreted by trophoblast cells are fully sialylated. Table 6.4 lists trisaccharide or tetrasaccharide as the simple glycosylation profile (Figure 6.1), and hexasaccharide or pentasaccharide as the complex glycosylation profile (Figure 6.1). Upon secretion and in serum, it appears that tetrasaccharide and hexasaccharide structures predominate. What is hCG-H? As the name implies, it is hyperglycosylated hCG. Structurally, hCG-H is a regular hCG molecule with oversized oligosaccharides. It is produced during the processes of choriocarcinoma development and pregnancy implantation. hCG-H is also choriocarcinoma hCG, with two N-linked oligosaccharides on the -subunit having 13% triantennary oligosaccharide structure, two N-linked oligosaccharides on the -subunit having 53% triantennary oligosaccharide structure, and four O-linked oligosaccharides on the -subunit having 75% pentasaccharide/ hexasaccharide structure. Overall, 54% of choriocarcinoma hCG oligosaccharides are of the larger, hyperglycosylated form. Table 6.2 and 6.4 show five choriocarcinoma samples investigated (C1 to C7); respectively, 56%, 40%, 57%, 64%, and 52% are of the larger oligosaccharide structure. In comparison, the four pregnancy samples (P3 to P9) have 9.2%, 17%, 14%, and 12% of the larger structure, respectively (Table 6.2 and 6.4). Thus, we arbitrarily claim that hCG-H is an hCG molecule with not only more than 35% of the larger O- and N-linked oligosaccharide structures, but also an increasing cleaved - and -subunit peptides. Hyperglycosylation seems to affect hCG folding. The -subunit, for example, is more exposed on hCG-H than on regular hCG. This makes hCG-H more easily recognized by some free -subunit-specific antibodies [12]. hCG-H is also more readily nicked than regular hCG. It is cleaved at sites 42–43, 43–44, and 44–45, which are not touched on regular hCG [12,13]. This suggests that the - and -subunits are bound together more loosely in hCG-H than in regular hCG. This might expose some receptor-binding sites and lead to the differing biological activities of hCG and hCG-H [15].
References [1] Mizuochi T, Nishimura R, Derappe C, Taniguchi T, Hamamoto T, Mochizuki M, et al. Structures of the asparagine-linked sugar chains of human chorionic gonadotropin produced in choriocarcinoma: appearance of triantennary sugar chains and unique biantennary sugar chains. J Biol Chem 1983;258:14126–9. [2] Cole LA. The O-linked oligosaccharides are strikingly different on pregnancy and choriocarcinoma hCG. J Clin Endocrinol Metab 1987;65:811–13. [3] Valmu L, Alfthan H, Hotakainen K, Birken S, Stenman UH. Site-specific glycan analysis of human chorionic gonadotropin -subunit from malignancies and pregnancy by liquid chromatography–electrospray mass spectrometry. Glycobiol 2006;16:1207–18.
Comparison of the Structures of hCG and Hyperglycosylated hCG
73
[4] Elliott MM, Kardana A, Lustbader JW, Cole LA. Carbohydrate and peptide structure of the - and -subunits of human chorionic gonadotropin from normal and aberrant pregnancy and choriocarcinoma. Endocrine 1997;7:15–32. [5] Kobata A, Takeuchi M. Structure, pathology and function of the N-linked sugar chains of human chorionic gonadotropin. Biochim Biophy Acta 1990;1455:315–26. [6] Cole LA, Butler SA, Khanlian SA, Giddings A, Muller CY, Seckl MJ, et al. Gestational trophoblastic diseases: 2. Hyperglycosylated hCG as a reliable marker of active neoplasia. Gynecol Oncol 2006;102:150–8. [7] O’Connor JF, Ellish N, Kakuma T, Schlatterer J, Kovalevskaya G. Differential urinary gonadotropin profiles in early pregnancy and early pregnancy loss. Prenat Diagn 1998; 18:1232–40. [8] Sasaki Y, Ladner DG, Cole LA. Hyperglycosylated hCG: the source of pregnancy failures. Fertil Steril 2008;89:1781–6. [9] Shen QX, Bahl OP. cDNA-derived amino acid sequences of choriocarcinoma alpha- and beta-subunits of human choriogonadotropin. Mol Cell Endocrinol 1990;72:167–73. [10] Nishimura R, Ide K, Utsunomiya T, Kitajima T, Yuki Y, Mochizuki M. Fragmentation of the beta-subunit of human chorionic gonadotropin produced by choriocarcinoma. Endocrinol 1988;123:420–5. [11] Bidart JM, Puisieux A, Troalen F, Foglietti MJ, Bohuon C, Bellet D. Characterization of a cleavage product in the human choriogonadotropin -subunit. Biochem Biophys Res Comm 1988;154:626–32. [12] Kardana A, Elliott ME, Gawinowicz MA, Birken S, Cole LA. The heterogeneity of hCG: I. Characterization of peptide variations in 13 individual preparations of hCG. Endocrinol 1991;129:1541–50. [13] Cole LA, Kardana A, Andrade-Gordon P, Gawinowicz MA, Morris JC, Bergert ER, et al. The heterogeneity of hCG: III. The occurrence, biological and immunological activities of nicked hCG. Endocrinol 1991;129:1559–67. [14] Nishimura R, Kitajima T, Hasegawa K, Takeuchi K, Mochizuki M. Molecular forms of human chorionic gonadotropin in choriocarcinoma serum and urine. Jpn J Cancer Res 1989;80:968–74. [15] Cole LA. New discoveries on the biology and detection of human chorionic gonadotropin. Reprod Biol Endocrinol 2009;7:1–37.
7 Structures of Free - and -Subunits Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
hCG is a dimer composed of an -subunit and a b-subunit linked together by hydrophobic and ionic interactions in a noncovalent manner. The -subunit is common to hCG and other glycoprotein hormones, including luteinizing hormone (LH), follicle stimulating hormone (FSH), and thyroid stimulating hormone (TSH). The -subunit of hCG, in contrast, is unique and a separate independent molecule. During pregnancy, the placenta produces limited amounts of -subunit and excess amount of -subunit produced in pregnancy lead. The excess amounts of -subunit lead to both hCG dimer production and the secretion of a free -subunit [1,2]. This excess -subunit is then converted into a free -subunit with an alternative structure. Free -subunit is a glycoprotein hormone-processing intermediate with no known independent functions. It cannot combine with -subunit [1,2]. In pregnancy serum, free -subunit is primarily derived from the dissociation of hCG-H and regular hCG. The free -subunit is detected in serum and urine samples. In most cases, excess amounts of -subunit are made in hydatidiform moles or choriocarcinoma cells. This is secreted as free -subunit. Cancer cells retrodifferentiate or are otherwise transformed and express the -subunit of hCG. The -subunit expression is limited. This might lead to free -subunit production by cancer cells [3]. Free -subunit has been shown to have a biological function separate from that of hCG. In cancer cells, the free -subunit produced acts as an autocrine on the same cancer cells, promoting cell growth, invasion, and metastases [3–7]. In this chapter, we describe free - and -subunits structure, synthesis, and basic properties in detail.
7.1 Free -Subunit In the normal synthesis of all glycoprotein hormones (i.e., LH, FSH, TSH, and hCG), the -subunit is limited and controls hormone synthesis; this results in an excess -subunit [8,9]. Free -subunit is made by the gonadotrope and thyrotrope cells of the pituitary gland during the menstrual cycle, and by the trophoblast cells of the placenta during pregnancy. All these cells produce a free -subunit in addition to the hormone. As a result of larger oligosaccharides, the free -subunit is larger than the -subunit in the hormone dimer. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00007-4 © 2010 Elsevier Inc. All rights reserved.
76
Human Chorionic Gonadotropin (hCG)
During pregnancy, the placenta produces limited -subunit and excess -subunit, leading to hCG dimer production and production of a free -subunit [1,2]. In pregnancy serum and urine, the presence of free -subunit is usually the result of the dissociation of both hCG-H and hCG. In cancer cells, hydatidiform mole cells, choriocarcinoma, and testicular germ-cell malignancies, excess -subunit can be produced. Thus, the -subunit can limit dimer formation and a free -subunit rather than a free -subunit will be produced by these cells [10]. The excess -subunit made during pregnancy is converted into a free -subunit. The free -subunit is larger than the -subunit of hCG, giving the appearance of two molecules; however, only one gene codes for all -subunits. The large size is apparently due to larger or more complex oligosaccharides. As shown by numerous researchers, the larger size prevents free -subunit from being able to combine with -subunit to form an hCG dimer [1,2,11,12]. Multiple studies have examined the oligosaccharide structures of free -subunit, but all these studies have been limited to lectin and endoglycosidase specificitybased techniques rather than mass spectrometry or carbohydrate sequencing methods. In purifying free -subunit and isolating the pure N-linked oligosaccharides, it was demonstrated that, after complete removal of oligosaccharides with N-glycanase, the ability to combine with -subunit was recovered [12]. This confirms that oligosaccharides block the combination of subunits [12]. Lens culinaris agglutinin (lectin) binds oligosaccharides—except when a N-acetylglucosamine (GlcNAc) is 1,4 linked to the 1,3 mannose (Man) in the core of N-linked oligosaccharides. Datura stramonium agglutinin differentiates a GlcNAc that is 1–6 linked to the 1,6 mannose in the core of N-linked oligosaccharides from other linkages. Using these lectins [13,14], it has been shown that the 1,6 GlcNac is found on 1,6 mannose and 1,4 GlcNAc is found on the 1,3 mannose of hCG free -subunit (Figure 7.1). This unusual 1,4 and 1,6 GlcNAc linkage differentiates the structure of free -subunit from the structures of both hCG dimer -subunit and hCG-H -subunit (see Chapters 5 and 6). Although monoantennary and biantennary sugar structures are located on the -subunit of hCG dimer, multiple larger oligosaccharides are found on free -subunit (15,16; see Chapter 6). Monoantennary (SM) oligosaccharides account for 12% of total free -subunit structure; biantennary (SS and SSF) oligosaccharides for 18%; and biantennary terminal mannose structures (SSM and SSMF) for 21%; four types of triantennary structure (SSS 1,6; SSSF 1,6; SSS 1,3; and SSSF 1,3) account for 23%, and tetraantennary (SSSS and SSSF) oligosaccharide structures account for 8% of total free -subunit structure (Figure 7.1). Each of these structures can be fucosylated. In free -subunit, it is estimated that oligosaccharides are 50% fucosylated. Thus, an equal mixture of fucosylated and nonfucosylated structures are found on free -subunit [13,14]. The oligosaccharides on the average sample of free -subunit are mostly a mixture of biantennary, triantennary, and tetraantennary containing an average of 2.7 antennae. As pregnancy advances, there is a clear shift in the structure of hCG free -subunit from 60.4% SSS with the GlcNAc 1,4 Man 1,3 structure down to only 24.8% [13]. The reverse occurs with the GlcNAc 1,6 mannose 1,6 structure: the proportion
Structures of Free - and -Subunits
77
Figure 7.1 The oligosaccharide structures attached to hCG free -subunit [13, 14]. Structure names ending with the letter F (i.e., SSF, SSMF, SSSF 1,6, SSSF 1,3 and SSSF) contain 1,6 fucose side arms.
surges in the latter part of pregnancy (late second vs. early second trimester). As pregnancy advances, the degree of fucosylation of free -subunit also rises. What these changes mean is uncertain. Free -subunit concentration in blood and urine starts low in the first trimester of pregnancy and continuously rises as pregnancy advances. In the third trimester of pregnancy, total hCG concentration equals
78
Human Chorionic Gonadotropin (hCG)
hCG hCG-H free -subunit [17]. As free -subunit production rises, it appears that its structure changes. Free -subunit has no known biological function, so the structural changes have no known biological significance. Crystals have been made from the deglycosylated free -subunit. The folding of these crystals has been studied using NMR spectroscopy [18]. These studies have shown that the free -subunit folding abnormalities are distinct and not found on the -subunit of hCG dimer. A structural disorder occurs at residues 33–57, and changes in the hairpin loops at residues 20–23 and 70–74 are reported [18]. Once again, we note that free -subunit has no known biological function. Hence, this difference in folding has no known biological significance. Free -subunit can be ectopic, meaning it is produced by nontrophoblastic cancer cells. Ectopic free -subunit was isolated from a cancer patient’s urine [19] and shown not to combine with -subunit. The amino acid sequence was then investigated and glutamic acid (Glu) 56 was replaced with alanine (Ala) 56; in other words, an acidic amino acid was replaced with a hydrophobic residue. This probably affected the three-dimensional structure of free -subunit. In this case, it is assumed that either a mutation in the single -subunit gene or an error in amino acid translation occurred. The codons GAA and GAC code for Glu, whereas GCA and GCC code for Ala. It is easy to visualize how a single point mutation might explain a Glu–Ala switch. Although free -subunit is considered to be a primarily N-glycosylated molecule, it can be O-glycosylated [20,21]. For example, an O-linked oligosaccharide has been located on the free -subunit produced in choriocarcinoma and on that made by Jar choriocarcinoma cells. The O-linked oligosaccharides detected on hCG have a structure of NeuAc 2,3 Gal 1,3 GalNAc [20,21]. The free -subunit with O-linked oligosaccharide is similar in structure. It is concluded that in addition to two N-linked sugar units, free -subunit contains an O-linked oligosaccharide located on tryptic peptide 36–42. The tryptic peptide contains a Thr residue at amino acid 39, the likely attachment point of the O-linked oligosaccharide.
7.2 Free -Subunit A free -subunit is made primarily during pregnancy and is the artifact of excess -subunit and limiting -subunit production [17,22]. Once we move from pregnancy to complete hydatidiform mole and choriocarcinoma, -subunit can be made in excess and lead to the detection of a naturally secreted free -subunit in serum and urine [23]. As discussed later (Chapter 14), cancer cells ectopically express the free -subunit of hCG, allowing it to be detected in serum samples [24–29]. Acevedo and colleagues found that free -subunit is present in all cell membranes and in every cancer known to humans [28,29]. This is a controversial finding that others have not been able to confirm. The problem is that only 30% of cancers produce high enough levels of free -subunit to be detected in serum. In most cases, the free -subunit is present in urine as -core fragment (the terminal degradation product of
Structures of Free - and -Subunits
79
free -subunit), which is rapidly removed from the serum. -core fragment comprises -subunit residues 6–40 linked to residues 55–92 by disulfides [30]. -core fragment has degraded oligosaccharides and a mannose core structure, Man1,3 (Man1,6) Man1,4GcNAc1,4GlNAc [30]. While serum free -subunit reveals 30% of malignancies, tests using urine -core fragment can detect 48% of malignancies (3; see also Chapter 32). As recently shown [31–34], malignancies produce free -subunit that acts on those same malignancies to inhibit apoptosis, promoting both cancer cell growth
Figure 7.2 The (A) N- and (B) O-linked oligosaccharides attached to free -subunit [33].
80
Human Chorionic Gonadotropin (hCG)
and metastasis. Free -subunit is a biologically active cytokine-like molecule. For 20 years, the free -subunit on SDS polyacrylamide gels has been observed migrating as a slightly larger molecule than the hCG dimer -subunit [23,32]. It has been suggested that the molecular weight of free -subunit is 24,000, compared with 22,200 for the -subunit of hCG [32]. In recent years, mass spectrometry has confirmed the molecular weight of free -subunit. Recently, the structure of free -subunit has been investigated using the urine of a patient with complete hydatidiform mole [33]. Complete hydatidiform mole normally produces pregnancy-like hCG dimer. The free -subunit studied, however, had hCG-H-like oligosaccharide structures [33]. Like hCG-H -subunit, the free -subunit contained 37% triantennary N-linked structures at one glycosylation site (Asn 13), and 30% triantennary structures at the second N-linked site (Asn 30) (Figure 7.2). The O-linked oligosaccharides were hexasaccharide/pentasaccharide in structure; 85% at Ser 121, 86% at Ser 127 and 132, and 79% at Ser 138 [33]. Like choriocarcinoma hCG -subunit, it is inferred that free -subunit is hyperglycosylated. This is probably the case regardless of whether free -subunit is produced by complete hydatidiform mole or cancer cells, and is different from the dissociated free -subunit observed in pregnancy. Valmu et al. [33] speculate that the hyperglycosylation of -subunit might block combination with hCG -subunit, just as hyperglycosylation blocks free -subunit combination. Interestingly, Butler and colleagues [34] and Iles and colleagues [31] have both shown that the isolated -subunit of pregnancy hCG works as well as the free -subunit produced by cancer cells to promote cancer cell growth. Hence, the hyperglycosylation is not an intrinsic part of the biological activity of free -subunit; a receptor binding site is obviously exposed in both molecules. Interestingly, free -subunit has been shown to homodimerize, making a – dimer [34]. The dimer is made in multiple cancers. The dimer, however, can confuse hCG immunometric assay results when using antibodies to two sites on the -subunit.
References [1] Posillico EG, Handwerger S, Tyrey L. Human chorionic gonadotropin -subunit of normal placenta: characterization of synthesis and association with -subunit. Biol Reprod 1985;32:1101–8. [2] Beebe JS, Kresinski RF, Norton SE, Perini F, Peters BP, Ruddon RR. Identification and characterization of subpopulations of free -subunit that vary in their ability to combine with chorionic gonadotropin-. Endocrinol 1989;124:1613–24. [3] Muller C, Cole LA. The quagmire of hCG and hCG testing in gynecologic oncology. Gynecol Oncol 2009;112:663–72. [4] Butler SA, Ikram MS, Mathieu S, Iles RK. The increase in bladder carcinoma cell population induced by the free beta subunit of hCG is a result of an anti-apoptosis effect and not cell proliferation. Br J Cancer 2000;82:1553–6. [5] Iles RK. Ectopic hCG expression by epithelial cancer: malignant behavior metastasis and inhibition of tumor cell apoptosis. Mol Cell Endocrinol 2007;260:264–70.
Structures of Free - and -Subunits
81
[6] Carter WB, Sekharem M, Coppola D. Human chorionic gonadotropin induces apoptosis in breast cancer. Breast Cancer Res Treat 2006;100:S243–4. [7] Iles RK. hCG expression by cervical squamous carcinoma: in vivo histological association with tumour invasion and apoptosis. Histopathol 2008;53:147–55. [8] Kourides IA, Hoffman BJ, Landon MD. Difference in glycosylation between secreted and pituitary free alpha-subunit of the glycoprotein hormones. J Clin Endocrinol Metab 1980;51:1372–7. [9] Blackman M, Gershengorn A, Weintraub BD. Excess production of free alpha subunits by mouse pituitary thyrotropic tumor cells in vitro. Endocrinol 1978;102:499–508. [10] Ozturk M, Berkowitz R, Goldstein D, Bellet D, Wands JR. Differential production of human chorionic gonadotropin and free subunit in gestational trophoblastic disease. Am J Obstet Gynecol 1988;158:192–8. [11] Prosillo EG, Hadwerger S, Tyrey L. Demonstration of intracellular and secreted forms of large human chorionic gonadotropin alpha subunit in cultures of normal placental tissue. Placenta 1983;4:439–48. [12] Blithe DL, Iles RK. The role of glycosylation in regulating glycoprotein hormone free -subunit and free -subunit combination in the extraembryonic coelomic fluid of early pregnancy. Endocrinol 1995;136:903–10. [13] Nemansky M, Thotokura NR, Lyons CD, Ya S, Reinhold BB, Reinhold VN, et al. Developmental changes in the glycosylation of glycoprotein hormone free -subunit during pregnancy. J Biol Chem 1998;273:12068–76. [14] Blithe DL. Carbohydrate composition of the -subunit of human chorionic gonadotropin (hCG) and the free molecules produced in pregnancy: most free and some combined hCG molecules are fucosylated. Endocrinol 1990;126:2788–99. [15] Elliott MM, Kardana A, Lustbader JW, Cole LA. Carbohydrate and peptide structure of the - and -subunits of human chorionic gonadotropin from normal and aberrant pregnancy and choriocarcinoma. Endocrine 1997;7:15–32. [16] Kobata A, Takeuchi M. Structure, pathology and function of the N-linked sugar chains of human chorionic gonadotropin. Biochim Biophy Acta 1990;1455:315–26. [17] Cole LA. Immunoassay of hCG, its free subunits and metabolites. Clin Chem 1997; 43:2233–43. [18] Erbel PJ, Karimi Nejad Y, De Beer T, Boelens R, Kamerling JP, Vliegenthart JF. Solution structure of the alpha-subunit of human chorionic gonadotropin. Eur J Biochem 1999;260:490–8. [19] Nishimura R, Shin J, Ji I, Russell Middaugh C, Kruggel W, Lewis RV, et al. A single amino acid substitution in an ectopic -subunit of a human choriogonadotropin. J Biol Chem 1986;261:10475–7. [20] Cole LA, Perini F, Birken S, Ruddon RW. An oligosaccharide of the O-linked type distinguishes the free from the combined form of hCG -subunit. Biochem Biophys Res Comm 1984;122:1260–7. [21] Cole LA. Distribution of O-linked sugar units on hCG and its free -subunit. Mol Cell Endocrinol 1987;50:45–57. [22] Cole LA. Human chorionic gonadotropin (hCG), free (free ), free (free ) and -core fragment (-core). Diagn Endocrinol Metab 1997;15:199–220. [23] Cole LA, Hartle RJ, Laferla JJ, Ruddon RW. Detection of the free -subunit of human chorionic gonadotropin (hCG) in cultures of normal and malignant trophoblast cells, pregnancy serum, and sera of patients with choriocarcinoma. Endocrinol 1983;113:1176–8. [24] Cole LA, Wang Y, Elliott M, Latif M, Chambers JT, Chambers SK, et al. Urinary human chorionic gonadotropin free beta-subunit and beta-core fragment: a new marker for gynecological cancers. Cancer Res 1988;48:1356–60.
82
Human Chorionic Gonadotropin (hCG)
[25] Nagelberg SB, Cole LA, Rosen SW. A novel form of ectopic human chorionic gonadotrophin -subunit in the serum of a woman with epidermoid cancer. J Endocrinol 1985;107:403–8. [26] Hussa RO, Fein HG, Tattillo RA, Nagelberg SB, Rosen SW, Weintraub BD, et al. A distinctive form of human chorionic gonadotropin -subunit-like material produced by cervical cancer cells. Cancer Res 1986;46:1948–54. [27] Cole LA, Nam J-H, Chambers JT, Schwarz PE. Urinary gonadotropin fragment, a new tumor marker. Gynecol Oncol 1990;36:391–4. [28] Acevedo HF, Tong JY, Hartsock RJ. Human chorionic gonadotropin-beta subunit gene statement in cultured human fetal and cancer cells of different types and origins. Cancer 1995;76:1467–75. [29] Acevedo HF, Hartstock RJ. Metastatic phenotype correlates with high expression of membrane-associated complete -human chorionic gonadotropin in vivo. Cancer 1996;78:2388–99. [30] McChesney R, Wilcox AJ, O’Connor JF, Weinberg CR, Baird DB, Schlatterer JP, et al. Intact HCG, free HCG subunit and HCG core fragment: longitudinal patterns in urine during early pregnancy. Hum Reprod 2005;20:928–35. [31] Crawford RA, Iles RK, Carter PG, Caldwell CJ, Shepherd JH, Chard T. The prognostic significance of beta human chorionic gonadotropin and its metabolites in women with cervical carcinoma. J Clin Pathol 1998;51:685–8. [32] Ruddon RW, Hartle RJ, Peters BP, Anderson C, Huot RI, Stromberg K. The biosynthesis and secretion of chorionic gonadotropin subunits by organ cultures of first trimester human placenta. J Biol Chem 1981;256:11389–92. [33] Valmu L, Alfhan H, Hotakainen K, Birken S, Stenman U-H. Site-specific glycan analysis of human chorionic gonadotropin 11-subunit from malignancies and pregnancy by liquid chromatography–electrospray mass spectrometry. Glycbiol 2006;16:1207–18. [34] Butler SA, Laidler P, Porter JR, Keiman AT, Chard T, Cowan DA, et al. The -subunit of hCG exists as a homodimer. J Mol Endocrinol 1999;22:182–92.
8 Glycobiology of hCG Akira Kobata The Noguchi Institute, Tokyo, Japan
8.1 C haracteristic Features of the Sugar Chains of Glycoproteins Glycoproteins are the proteins to which various sugar chains are covalently linked. Two major types of sugar chains (N- and O-linked) are found in glycoproteins. N-linked sugar chains contain an N-acetylglucosamine (GlcNAc) residue at their reducing termini, which is linked to the amide group of an asparagine (Asn) residue of a polypeptide. O-linked sugar chains contain an N-acetylgalactosamine (GalNAc) residue at their reducing termini, which is linked to a serine (Ser) or threonine (Thr) residue of a polypeptide backbone. Accumulation of the structural data of various glycoproteins revealed that N-linked sugar chains include more structural rules than O-linked sugar chains. All N-linked sugar chains contain the 3 mannose (Man), 2 GlcNAc pentasaccharide: Man1–6(Man1–3)Man1–4GlcNAc1–4GlcNAc as a common core, which will be called trimannosyl core in this review. Based on the structures and locations of the extra sugar residues added to the trimannosyl core, N-linked sugar chains are further classified into three subgroups: complex type, high-mannose type, and hybrid type (Figure 8.1). Sugar chains classified as the complex type contain no mannosyl residues other than the trimannosyl core. Outer chains with a GlcNAc residue at their reducing terminal are linked to two Man residues of the trimannosyl core. The highmannose sugar chains contain only -mannosyl residues in addition to the trimannosyl core. A heptasaccharide with two branching structures, as enclosed by a dotted line in Figure 8.1, is commonly included in this type of sugar chain. Variations are formed by the locations and numbers of up to four Man1–2 residues linked to the three nonreducing terminal -mannosyl residues of the common heptasaccharide. Hybrid-type sugar chains were so named because the oligosaccharides have the structural characteristics of both high-mannose-type and complex-type sugar chains. One or two -mannosyl residues are linked to the Man1–6 arm of the trimannosyl core, as in the case of the high-mannose type, and the outer chains found in the complextype sugar chains are linked to the Man1–3 arm of the trimannosyl core. Presence or absence of an -fucosyl residue linked to the C-6 position of the proximal GlcNAc residue and a GlcNAc residue linked to the C-4 position of the -mannosyl residue of the trimannosyl core (called a bisecting GlcNAc) contributes to the structural variation of the complex-type and hybrid-type sugar chains [1]. Among the three subgroups of N-linked sugar chains, the complex type has the largest structural variation. This variation is formed mainly by two structural factors. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00008-6 © 2010 Elsevier Inc. All rights reserved.
84
Human Chorionic Gonadotropin (hCG)
Figure 8.1 Three subgroups of N-linked sugar chains. Structures within the solid line are the pentasaccharide core common to all N-linked sugar chains. The bars in front of the nonreducing terminal monosaccharides indicate that the sugars can be further extended by adding sugars. Modified from figure 6 in [1].
Between one and five outer chains are linked to the trimannosyl core by different linkages (Figure 8.1), resulting in formation of mono-, bi-, tri-, tetra-, and pentaantennary sugar chains (Figure 8.2). Two isomeric triantennary sugar chains containing either the GlcNAc1–4 (GlcNAc1–2)Man1–3 group or the GlcNAc1–6(GlcNAc1–2)Man1–6 group can be found. These isomeric sugar chains are called 2,4- and 2,6-branched triantennary sugar chains, respectively. Starting from the -GlcNAc residues located at the nonreducing termini of the oligosaccharides in Figure 8.2, various outer chains are formed. Combinations of the antennary with various outer chains can form a large number of different complex-type sugar chains. In contrast to N-linked sugar chains, O-linked sugar chains have fewer structural rules. These sugar chains can be categorized into at least four groups according to their core structures (Figure 8.3). In addition, O-linked sugar chains with the GlcNAc1– 6GalNAc core and the GalNAc1–3GalNAc core are also found in a limited number of glycoproteins.
8.2 B iosynthetic Pathways of the Sugar Chains of Glycoproteins to Form Their Characteristic Features O-linked sugar chains are formed by stepwise addition of monosaccharides to the Ser and Thr residues of polypeptides from nucleotide sugars. In contrast, N-linked sugar
Glycobiology of hCG
Figure 8.2 Branching of complex-type sugar chains.
Figure 8.3 Four major core structures found in O-linked sugar chains.
85
86
Human Chorionic Gonadotropin (hCG)
chains are formed by a series of complex pathways including lipid-linked intermediates [2]. First, Glc3-Man9-GlcNAc2-P-P-Dol is formed by a complicated process starting from dolichol phosphate (Dol-P) as an acceptor [3]. The tetradecasaccharide moiety of the lipid derivative is then transferred en bloc to the Asn residue of the polypeptide chain, which is being translated in the rough endoplasmic reticulum by the catalytic action of an oligosaccharyltransferase complex residing in the endoplasmic membrane [4]. Only the Asn residue in the sequence of Asn-Xaa-Ser/Thr (where Xaa can be any amino acid other than proline) is glycosylated. Accordingly, these tripeptide sequences are called potential glycosylation sites. Asn-Xaa-Cys can be glycosylated equally well as Asn-Xaa-Ser/Thr [5]; however, very few of the tripeptide sequences in natural glycoproteins are glycosylated [6], probably because the SH group of the Cys residue quickly forms an S–S linkage with another Cys residue, and cannot contribute to the ring formation as in the case of an Asn-Xaa-Ser/Thr group. The completely translated polypeptide with the tetradecasaccharide is then transported to the Golgi apparatus. During this transport, three -glucosyl residues and at least one Man1–2 residue are removed by the action of two -glucosidases and an -mannosidase residing in the membrane of the endoplasmic reticulum (Figure 8.4). After being translocated to the cis-Golgi, the N-linked sugar chain of the polypeptide is converted to Man5-GlcNAc2 by the action of Golgi -mannosidase I, which removes all Man1–2 residues from the sugar chain (Figure 8.4). A series of high-mannose-type sugar chains is considered to be the intermediary product of this trimming process. When the glycoprotein is translocated to the medial-Golgi, GlcNAc residue is added at the C-2 position of the Man1–3 arm of the trimannosyl core by the action of N-acetylglucosaminyltransferase-I (GnT-I) [7]. Addition of this GlcNAc residue probably changes the steric arrangement of the two -mannosyl residues linked to the Man1–6 arm so that they can be removed by Golgi -mannosidase II [8]. These are the entire features of the processing pathway that forms the prototype of monoantennary complex-type sugar chains. Starting from a monoantennary complex-type sugar chain, a series of prototypes of the complex-type sugar chains is formed by the action of various GnTs (Figure 8.5). Each -N-acetylglucosamine residue is further elongated by the action of various glycosyltransferases and sulfotransferases. Glycosyltransferases that catalyze these reactions have strict specificities for donor nucleotide sugars and acceptor sugar-chain structures.
8.3 T he hCG Sugar Chains from Urine of Pregnant Women and Placenta hCG is a heterodimer composed of - and -subunits. Both subunits contain two N-linked sugar chains [9,10], and the -subunit contains four O-linked sugar chains in addition [10]. Endo et al. investigated several hCG samples purified from pooled urine of healthy pregnant women, and elucidated the complete structures of the N-linked sugar chains of hCG [11]. As shown in Figure 8.6A, five sialylated oligosaccharides were found to occur in all samples. Occurrence of A-3 and A-5 in hCG
Glycobiology of hCG
Figure 8.4 Processing pathway in the biosynthesis of N-linked sugar chains.
87
88 Human Chorionic Gonadotropin (hCG)
Figure 8.5 Formation of branching structures of complex-type sugar chains. R and R’ represent the GlcNAc1–4GlcNAc and the GlcNAc1– 4(Fuc1–6)GlcNAc groups, respectively.
Glycobiology of hCG
Figure 8.6 Structures of the N-linked sugar chains of hCG purified from the urine of pregnant women (A), and their desialylated forms (B).
89
90
Human Chorionic Gonadotropin (hCG)
was also independently reported by Kessler et al. [12]. Upon sialidase digestion, the five acidic oligosaccharides were converted to the three neutral oligosaccharides shown in Figure 8.6B. Studies of the hCG sample (purified from human placenta) revealed that it contains both sialylated and nonsialylated N-linked sugar chains; however, it gave oligosaccharides A, C, and D shown in Figure 8.6B in the molar ratio of approximately 1:2:1 after desialylation, as in the case of urinary hCG [13]. hCG can be dissociated into - and -subunits by treating with 8 M urea [14]. These subunits are termed hCG and hCG in this chapter. A comparative study of the sugars released from the hCG and the hCG of placental hCG by hydrazinolysis revealed that the oligosaccharides shown in Figure 8.6B are not evenly distributed in the subunits [13]. hCG contained nearly equal amounts of oligosaccharides A and C, but did not contain D. In contrast, hCG contained nearly equal amounts of oligosaccharides C and D, but did not contain A. These results indicated sitespecific N-glycosylation in hCG; the two N-linked sugar chains of -subunit are never fucosylated, and one of them remains at the monoantennary stage. In contrast, both N-linked sugar chains of -subunit are converted to biantennary complex-type sugars, and one of them is fucosylated.
8.4 C haracteristic Features of the Sugar Chains of Free -Subunit A small amount of free-floating -subunit occurs in the urine of pregnant women, and was aptly named free -subunit. Interestingly, in contrast to the hCG dissociated from hCG, this free -subunit cannot bind to hCG. A study of the sugar chains of free -subunit revealed that it contains only one N-linked sugar chain [15]. A structural study of the oligosaccharides (released from free -subunit by hydrazinolysis) indicated that they were sialylated oligosaccharides C and D (Figure 8.6B) in a molar ratio of 91:9 [15]. Based on these findings, we hypothesized the biosynthetic mechanism of the N-linked sugar chains of hCG as shown in Figure 8.7. Both - and -subunits of hCG (produced in the rough endoplasmic reticulum of the trophoblasts of placenta) have two tetradecasaccharides (Glc3-Man9-GlcNAc2) at their potential N-glycosylation sites, as a result of the catalysis of oligosaccharyltransferase complex. These nascent sugar chains are converted to Man8~9-GlcNAc2 by the action of -glucosidases and ER--mannosidase in the endoplasmic reticulum (Figure 8.4). During this early processing, a small portion of -subunit fails to accept Glc3-Man9-GlcNAc2 at its one potential N-glycosylation site. This failure probably occurs because folding of the polypeptide moiety of -subunit hides one of the two potential N-glycosylation sites of the -subunit, as found in the case of ovalbumin. Ovalbumin contains two potential N-glycosylation sites Asn292Leu-Thr and Asn311-Leu-Ser in its polypeptide [16]; however, Asn311 is never glycosylated. Denatured ovalbumin can be further N-glycosylated by an in vitro system, indicating that Asn311 residue might be buried within the folded peptide, before accepting the tetradecasaccharide from the dolichol derivative.
Glycobiology of hCG 91
Figure 8.7 Maturation of the N-linked sugar chains of hCG and of free -subunit. S is sialic acid, G is galactose, M is mannose, F is fucose, GN is N-acetylglucosamine. Taken from [1].
92
Human Chorionic Gonadotropin (hCG)
An -subunit with two Man8~9-GlcNAc2 will associate with a -subunit, which also has two Man8~9-GlcNAc2. The four Man8~9-GlcNAc2 of the heterodimer will be processed to Man1–6 (GlcNAc1–2Man1–3)Man1–4GlcNAc1–4GlcNAc when the heterodimer reaches the medial-Golgi. Maturation of the four hCG N-linked sugar chains to complex-type sugar chains will then be controlled by the steric effect of the polypeptide moieties of the two subunits. By this mechanism, one N-linked sugar chain of the -subunit remains in the monoantennary state, while the other is converted to a biantennary complex-type sugar chain. In contrast, the control effect will allow the two N-linked sugar chains of the -subunit to become a biantennary complex-type sugar chain, but will inhibit one sugar chain from being fucosylated. The -subunit with one Man8~9-GlcNAc2 cannot combine with a -subunit, and therefore reaches the Golgi as a free -subunit. Because the maturation of its N-linked sugar chain is not controlled by the steric effect of the polypeptide moiety of the -subunit, it will grow to an oligosaccharide D (Figure 8.6B) by the complete action of the glycosylation machinery of trophoblasts. The work of Matzuk and Boime turned this hypothesis into a reality [17]. They prepared mutants of the -subunit gene, in which one of the two N-glycosylation sites was eliminated by site-directed mutagenesis. Expression of each mutant gene in CHO cells, together with the gene of wild-type -subunit of hCG, revealed that removal of the Asn52 N-glycosylation site significantly decreased the capacity of the -subunit to bind with -subunit; removal of the Asn78 site did not eliminate this binding capacity with -subunit, but rather caused the mutant subunit to be degraded quickly in the absence of -subunit. Therefore, free -subunit might lack the sugar chain at its Asn52.
8.5 C omparative Studies of the N-Linked Sugar Chains of hCG Samples Purified from the Urine of Patients with Various Trophoblastic Diseases Trophoblastic diseases are histologically classified as hydatidiform mole, invasive mole, and choriocarcinoma. The prognosis of these diseases differs greatly. Hydatidiform mole is considered a benign lesion, but some of the moles apparently show more malignant characteristics, such as invasion into surrounding tissues and metastasis. Because of these somewhat malignant characteristics, they are distinguished from typical moles by the name “invasive mole.” Choriocarcinoma shows the characteristics of a malignant tumor. Study of the oligosaccharides released from an hCG sample purified from the urine of a patient with choriocarcinoma revealed that none of the oligosaccharides was sialylated, although four moles of oligosaccharides were released from one mole of the hCG sample, as in the case of urinary hCG samples from healthy pregnant women [18]. Further study of the N-linked sugar chains of hCG samples purified from the urine of three additional patients with choriocarcinoma revealed that deletion of sialic acid is not a common phenomenon to choriocarcinoma hCG [19]; however, it was found that the eight oligosaccharides listed in Table 8.1 were included in all choriocarcinoma hCG samples as the neutral portions of the oligosaccharides [18,19].
Glycobiology of hCG
93
Table 8.1 Structures of the Desialylated N-Linked Sugar Chains of Urinary hCG from Normal Pregnant Women and from Patients with Trophoblastic Disease. Sugar-chain structures
Pregnant women
Hydatidiform mole
Invasive mole
Choriocarcinoma
Study of the N-linked sugar chains of urinary hCG samples purified from patients with hydatidiform mole revealed that these samples contain exactly the same N-linked sugar chains as urinary hCG samples obtained from healthy pregnant women [19]. The sugar chains were highly sialylated, and only oligosaccharides A, C, and D in Table 8.1 were detected in the molar ratio of approximately 1:2:1 after desialylation. Study of the N-linked sugar chains of urinary hCG samples obtained from patients with invasive mole revealed that these chains contain the six sialylated oligosaccharides shown in Table 8.1 [20]. The alteration of the N-linked sugar chains of choriocarcinoma hCG can be induced by the increase and expression of two enzymes. The molar ratios of total fucosylated oligosaccharides of the four choriocarcinoma hCG samples reach almost 50% of the total oligosaccharides, which are twice that of normal and hydatidiform mole hCG. This evidence indicates that the level of fucosyltransferase-8 (FUT-8) [21], which forms the Fuc1–6GlcNAc group, is increased in choriocarcinoma. Apparently, oligosaccharides E, F, G, and H are formed by adding the Gal1– 4GlcNAc1–4 group to oligosaccharides A, B, C, and D in Table 8.1, respectively. Therefore, GnT-IV [22], which is responsible for formation of the GlcNAc1– 4Man1–3 group, should be ectopically expressed in choriocarcinoma. Although the enzyme is detected in many cells other than trophoblasts, oligosaccharides E and F have not been found in the glycoproteins produced by normal tissues (Figure 8.2). This
94
Human Chorionic Gonadotropin (hCG)
Figure 8.8 Biosynthesis of the abnormal biantennary sugar chains found in choriocarcinoma cells. GnT is N-acetylglucosaminyl transferase.
indicates that GnT-IV in normal cells cannot catalyze the pathway 2 in Figure 8.8. This evidence indicates that the ectopically expressed GnT-IV in choriocarcinoma acquired the new characteristic to widen its substrate specificity. Because hCG purified from the urine of invasive mole patients contains oligosaccharides G and H (Table 8.1), ectopic expression of GnT-IV also occurs in this lesion. However, the absence of oligosaccharides E and F in the hCG indicates that the newly expressed GnT-IV did not acquire wider substrate specificity. Based on the sum of this evidence, oligosaccharides E and F were named abnormal biantennary complex-type sugar chains. Occurrences of these sugar chains were later found in some glycoproteins produced by other tumors, such as -glutamyltranspeptidase purified from human hepatoma [23], and carcinoembryonic antigen obtained from liver metastases of primary colon cancers [24].
8.6 A lteration Induced in the O-Linked Sugar Chains of hCG by Malignant Transformation of Trophoblasts As already described, the -subunit of hCG contains four O-linked sugar chains. Structures of the O-linked sugar chains of hCG were elucidated as sialylated core 1 (shown in Figure 8.3) by Kessler et al. [25]. Cole et al. [26] later found that small amounts of sialylated core 2 (shown in Figure 8.3) are included as the O-linked sugar chains of hCG. Amano et al. [27] analyzed the O-linked sugar chains of hCG purified from urine of patients with a variety of trophoblastic diseases, and found that tumorrelated structural alterations were induced in the O-linked sugar chains as well; however, the alteration is not qualitative as in the case of N-linked sugar chains, but
Glycobiology of hCG
95
Figure 8.9 Percent molar ratio of the O-linked sugar chains with core 1 and core 2 (shown in white and black, respectively) in various urinary hCG samples. Taken from [27].
quantitative. The proportion of oligosaccharides with core 2 prominently increased in choriocarcinoma hCG (Figure 8.9). A moderate but significant increase was also observed in invasive mole hCG, but not in hydatidiform mole hCG.
8.7 Altered Expression of GnT-IV in Choriocarcinoma Cells Molecular biological studies of GnT-IV in placenta and choriocarcinoma cell lines have revealed some of the enzymatic background of the altered N-glycosylation of choriocarcinoma hCG. GnT-IV is expressed in many animal species [22,28–31]; however, GnT-IV activities in these cells are lower than other GnTs that are responsible for the branching of complex-type N-linked sugar chains (shown in Figure 8.5). Oguri et al. successfully purified GnT-IV from bovine small intestine [32]. As reported by Gleeson and Schachter [22], the purified enzyme required the presence of the GlcNAc1–2Man1–3 group in the acceptor oligosaccharides. Cloning of the gene encoding this enzyme revealed that the enzyme has a type II membrane-bound protein structure, but has no homology with other cloned GnTs [33]. In order to investigate the enzymatic basis of the altered N-glycosylation of the hCG produced in choriocarcinoma cells, human cDNA for GnT-IV was obtained from a human liver cDNA library. Unexpectedly, two active GnT-IV genes with 91% and 64% homology to the bovine GnT-IV gene were obtained [34]. The translation products of these genes were named GnT-IVa and GnT-IVb, respectively. Expressions of the two genes in various organs were strikingly different. The GnT-IVb gene was expressed in all human organs at almost the same level, whereas the GnT-IVa gene was highly expressed in specific organs such as pancreas, thymus, small intestine, and leukocytes [35]. When activities of the glycosyltransferases related to the formation of the abnormal biantennary sugar chains were comparatively investigated in normal placenta
96
Human Chorionic Gonadotropin (hCG)
and several choriocarcinoma cell lines, the GnT-IV activity was strikingly increased in the cancer cells [36]. GnT-III activity was also increased, although GnT-I, GnT-II, 1,4-galactosyltransferase and -mannosidase II activities were not increased significantly from those of normal placenta. In addition, GnT-II activities in normal placenta and choriocarcinoma cells were found to be lower than the activities in other cells that do not produce monoantennary sugar chains [36]. Northern blot analysis revealed that the GnT-IVa gene was extraordinarily overexpressed in the cancer cells, whereas the GnT-IVb gene was expressed at the same level as in normal placenta [36]. So far, no difference in the substrate specificities of GnT-IVa and GnT-IVb has been found. It was also found that both enzymes produce the abnormal biantennary structures in vitro. The data described thus far indicated that low GnT-II activity and extremely high GnT-IV activity induced by overexpression of the GnT-IVa gene are the enzymatic basis of the formation of the abnormal biantennary sugar chains in choriocarcinoma cells. Although pathways 2 and 3 in Figure 8.8 were strongly enhanced, the monoantennary sugar chain will be mainly converted to the abnormal biantennary sugar chain because pathway 1 was very weak. The abnormal biantennary sugar chain is then derived to the triantennary sugar chain through pathway 4, which should be weak as well; however, these speculations cannot explain the subtle differences detected in the N-linked sugar chains of choriocarcinoma hCG and invasive mole hCG, as already discussed. Therefore, more detailed studies of the kinetic parameters of the enzymes and an investigation of their topology in the Golgi membrane of the cells of the two diseases must be performed in the future.
8.8 G lycosylated hCG as a Diagnostic Marker of Trophoblastic Diseases Differential diagnosis of trophoblastic diseases is clinically important for the proper treatment of patients. Because oligosaccharides G and H (Table 8.1) were detected in both invasive mole hCG and choriocarcinoma hCG, any method that specifically detects the hCG containing the Gal1–4GlcNAc1–4(Gal1–4GlcNAc1–2)Man group in the sugar moieties could be useful in discriminating these patients from pregnant women or patients with hydatidiform mole. By investigating the behavior of various complex-type oligosaccharides on several immobilized lectin columns, Yamashita et al. [37] found that these oligosaccharides can be separated into three groups by passing through a Datura stramonium agglutinin (DSA)–Sepharose column. The oligosaccharides, which are weakly bound to the column and eluted with buffer in the retarded fraction, all contain the nonsubstituted Gal1–4GlcNAc1–4(Gal1–4GlcNAc1–2)Man group. The oligosaccharides, which are strongly bound to the column and eluted from the column with the buffer containing a 1% mixture of GlcNAc1–4 oligomers, have either the nonsubstituted Gal1–4GlcNAc1–6(Gal1–4GlcNAc1–2)Man group or the nonsubstituted Gal1– 4GlcNAc1–3Gal1–4GlcNAc1 group as their partial structures. The oligosaccharides that contain none of these groups pass through the column without any interaction.
Glycobiology of hCG
97
Figure 8.10 Percent molar ratio of urinary hCG bound to a DSA-Sepharose column before () and after () sialidase digestion. (A) Urine samples from normal pregnant women, (B) those from patients with hydatidiform mole, (C) those from patients with invasive mole, (D) those from patients with choriocarcinoma. Taken from [38].
The binding specificity of the DSA-Sepharose column was expected to be useful for discriminating hCG with or without oligosaccharides G and H (Table 8.1), so the behavior of urinary hCG from various trophoblastic diseases on this column was investigated [38]. Almost all hCG in the urine of a pregnant woman passed through the column without interaction. The elution pattern did not change even after the urine was pretreated by sialidase digestion. In contrast, only a portion of hCG in the urine of a patient with choriocarcinoma passed through the column. The remainder was not eluted even with the buffer containing a 1% mixture of GlcNAc1–4 oligomers, but was completely recovered by elution with 0.1N acetic acid. This unexpectedly strong binding might have occurred because the hCG molecule contains at least two oligosaccharides with the Gal1–4GlcNAc1–4(Gal1–4GlcNAc1–2) Man group. Therefore, the elution step with the buffer containing a 1% mixture of GlcNAc1–4 oligomers was omitted, and the amounts of hCG in the two fractions obtained by elution with simple buffer and 0.1N acetic acid were measured to determine the percentage of hCG with the Gal1–4GlcNAc1–4(Gal1–4GlcNAc1–2) Man group in the sugar chains. As shown in Figure 8.10, the values for normal pregnant women and patients with hydatidiform mole were less than 15%, and the values did not increase after sialidase digestion. The values for patients with invasive mole were also small, but sialidasetreated samples afforded much higher values. The data for patients with choriocarcinoma varied more than others. Some of them behaved very similarly to invasive mole hCG. However, one of the choriocarcinoma hCG samples bound completely to
98
Human Chorionic Gonadotropin (hCG)
the column without sialidase treatment, indicating that some choriocarcinoma hCG lacked sialic acid residues. Therefore, affinity-column chromatography with use of a DSA-Sepharose can be used effectively to discriminate malignant hCG from nonmalignant hCG in urine samples. By assigning the cutoff value as 20%, desialylated hCG samples of patients with invasive mole and choriocarcinoma can be completely discriminated from those of patients with hydatidiform mole and pregnant women. Birken et al. produced a monoclonal antibody B152 by using choriocarcinoma hCG as an antigen [42]. Although the exact epitope recognized by the antibody had not been clearly elucidated except for its possible occurrence in the COOH-terminal peptide region of the -subunit [47], it was recently estimated that the antibody recognizes core 2 O-linked sugar chains on Ser132 of -subunit [40]. Usefulness of this antibody for discriminating various trophoblastic diseases [44,45] and for the diagnosis of several abnormal pregnancies has been described [41,43,46–50].
8.9 Functional Role of the hCG Sialic Acid Residues Many studies revealed that modification of the hCG N-linked sugar chains alters the hormonal activity of hCG [51–55]. These reports indicated that complete removal of sialic acid residues from hCG enhances its binding to the target cells, but reduces its hormonal activity to approximately 50%. Removal of the whole N-linked sugar chains from hCG further increased the binding of hCG to its target cells, but almost completely eliminated its hormonal activity. Deglycosylated hCG behaves as an antagonist to native hCG [51]. Calvo and Ryan [56] reported that the glycopeptides mixture (obtained from hCG and hCG by exhaustive pronase digestion) blocks hCG signal transduction; thus, they suggested that a lectin-like membrane component in addition to hCG-receptor might be involved in the signal transduction. Thotakura et al. reported that glycopeptides and oligosaccharides (obtained from other glycoproteins) can also prevent the hormonal action of hCG, and suggested that a lectin-like polypeptide portion might be included in the hCG receptor itself [57]. Following the study of the functional role of the N-linked sugar chains of -subunit (as introduced in a previous section of this chapter) [17], Matzuk et al. [58,59] further indicated that all four N-linked sugar chains of hCG are important for constructing the correct conformation of the two subunits by comparatively investigating the bioactivities of hCG that lacked one or more of the four N-linked sugar chains by site-directed mutagenesis. The role of N-linked sugar chains in intracellular folding of hCG was also investigated by Feng et al. [60]. To elucidate the mechanism of suppression of the hormonal activity of hCG by desialylation, Amano et al. [61] investigated the functional role of the sialic acid residues of hCG. It was found that all sialic acid residues of hCG occur as the Neu5Ac2–3Gal group [11]. To find out whether this particular disaccharide group is important in expressing the biological function of hCG, the following experiment was performed. MA-10 cells, a mouse Leydig tumor cell line established by Ascoli [62], produce cyclic AMP (cAMP) in response to the addition of hCG to their culture
Glycobiology of hCG
99
medium. When hCG was desialylated, its hormonal activity was reduced to approximately 50%, as in the case of the other target cell lines. When the desialylated hCG was resialylated by incubation with CMP-NeuAc and Gal1–4GlcNAc:2–6sialyltransferase, the isomeric hCG containing the Neu5Ac2–6Gal group thus obtained gave almost the same dose–response curve of cAMP production as the natural hCG [49]. These results indicated that the sialic acid residues of hCG are important for the full expression of their hormonal activity in vitro, but the effect is independent of their linkage to the galactose residues. Nemansky et al. [63] confirmed this finding and provided additional important evidence. They found that the decrease of hormonal activity of hCG caused by desialylation was restored only by the addition of a Sia2–6 residue, but not a Gal1–3 residue, to the galactose moiety of the Gal1–4GlcNAc1–2Man1–3Man arm of the N-linked sugar chains. Further 6-sialylation of the galactose residue of the Gal1–4GlcNAc1–2Man1–6Man arm reduced the hormonal activity of hCG, indicating that sialylation of the outer chain on the Man1–3Man arm, rather than the Man1–6Man arm, of the N-linked sugar chains of hCG plays an essential role in signal transduction. They also indicated that sialylation of the O-linked sugar chains is not important in LH/hCG receptor signaling. To elucidate the role of sialic acid residues in the signal transduction of hCG, an N-acetylneuraminic acid hexamer, obtained from partial degradation of colominic acid, was added to the reaction mixture of hCG and MA-10 cells. The oligosaccharide did not inhibit the binding of hCG to the surface of the target cells, but cAMP production was reduced to 50% when 2 mM solution of the hexasaccharide was added [64]. These results indicated that the hexasaccharide can only inhibit the interaction of the sialic acid residues of hCG with the specific binding site on the cell surface, but does not influence the binding of the peptide portion of hCG to the receptor. The possibility that the action of the sialic acid hexamer might be due to a nonspecific anionic polymer effect was ruled out because the addition of fucoidin did not show any inhibition of the [3H]hCG binding to the cell surface receptor or cAMP production by hCG. Presence of the sialic acid binding site on the MA-10 cells was confirmed by using 3’-sialyllactose-conjugated BSA as a probe [64]. Based on the data indicating that sialic acid residues bind directly to the cell surface, a model of the hCG-receptor complex was constructed, as shown in Figure 8.11. A dual interaction of the peptide portion and the sialylated N-linked sugar chain of hCG with respective binding sites is essential for signal transduction. However, it is still not clear whether the lectin-like membrane component is a part of the hormone receptor or a part of a different molecule, as shown in Figure 8.11. In connection with this, it is interesting that a region homologous to the soybean lectin was detected in the human hCG-receptor [65]. Amano and Kobata [66] attempted to assign the N-linked sugar chains of hCG, which interact with the lectin-like component on MA-10 cells. By investigating which sugar chains are resistant to sialidase digestion in the presence of MA-10 cells, they found that the sialic acid residues linked to A and D (Figure 8.6B) became resistant to sialidase digestion by binding to the target cells. These results indicated that one of each sialylated N-linked sugar chains of hCG and hCG are covered
100
Human Chorionic Gonadotropin (hCG)
Figure 8.11 Schematic presentation of hCG-receptor complex. G is Gs protein, L is lectin, R is hCG receptor. Modified from [64].
when the hormone binds to the target cells. Therefore, one or both of the sugar chains at these two sites should play a critical role in signal transduction in the hormonal action of hCG.
8.10 Future Prospects Several reports conflict with the data described so far in this review. Weisshaar et al. [67] reconfirmed the finding that site-specific distribution of the N-linked sugar chains occurs at the four N-glycosylation sites of hCG. They also found less strict distribution than that supposed by Mizuochi and Kobata 30 years ago (based on data of the sugar pattern analyses of commercial hCG and hCG, which were guaranteed more than 99% pure) [13]. Weisshaar et al. also reported that small amounts of sialylated hybrid-type sugar chains Man1–3Man1–6(Gal1–4GlcNAc1– 2Man1–3)Man1–4GlcNAc1–4GlcNAc and Man1–6(Man1–3)Man1–6 (Gal1–4GlcNAc1–2Man1–3)Man1–4GlcNAc1–4GlcNAc are included as the sugar that is linked at Asn52 of hCG [67].These sugar chains could be included in the minor peaks of gel-permeation chromatography, which Mizuochi and Kobata neglected to analyze. Several papers have reported the occurrence of triantennary sugar chains in normal hCG samples. These samples, however, were purified from urine samples collected from many pregnant women. Because triantennary sugar chains occur in large amounts in urinary hCG from invasive mole and choriocarcinoma patients, it is possible that the larger sugar chains might have originated from trophoblastic disease hCG and contaminated the pooled urine. Our analyses of hCG samples purified from the urine of several healthy pregnant women never revealed the occurrence of triantennary sugar chains. Furthermore, the behaviors of desialylated hCG samples from urine of healthy pregnant women in a DSA-Sepharose column (see Figure 8.10) clearly indicated that no triantennary sugar chains occur in these samples.
Glycobiology of hCG
101
Valmu et al. [40] analyzed the oligosaccharide profiles of the two N-glycosylation sites and four O-glycosylation sites of -subunits dissociated from hCG samples purified from the urine of a patient with choriocarcinoma, a patient with invasive mole, and three pregnant women at different stages of fertilization by using liquid chromatography–electrospray mass spectrometry. The reported data gave much useful information. Most of the N-linked sugar chains linked to Asn13 were not fucosylated, whereas those linked to Asn30 were fucosylated. These results extended the information reported by Mizuochi and Kobata [13]; however, Valmu et al. also found triantennary and even small amounts of tetraantennary sugar chains in both sites. As to O-linked sugar chains, they reported the occurrence of site-specific distribution of sialylated core 1 and core 2 sugar chains. According to their data, approximately 25% of the O-linked sugar chains of pregnancy hCG contain core 2; this amount is much larger than the data shown in Figure 8.9 and by Elliott et al. [39]. It must be pointed out, however, that mass-spectrometric analysis cannot yield quantitative data like the hydrazinolysis-radioactive labeling used in our studies. A paper published by Elliott et al. cannot be overlooked when addressing this problem [39]. By analyzing the carbohydrate structures of hCG samples purified from the urine of normal pregnant individuals, they also confirmed the occurrence of subunit-specific N-glycosylation, indicating that hCG contains oligosaccharides A and C (Figure 8.6B) in the percent molar ratio of 36.7and 49.3; hCG contains C and D as the major sugar chains. As described earlier, Elliott et al. also reported that hCG contains small amounts of oligosaccharides D and H (Table 8.1), and hCG contains small amount of mono- and triantennary N-linked sugar chains. Because these researchers analyzed individual samples, the problem of trophoblastic-disease hCG contamination in pooled pregnancy urine sample could not be ruled out. By using lectin affinity-column chromatography, Skarulis et al. [68] found that the glycosylation pattern of hCG changes as gestation progresses. Therefore, more detailed structural analysis of the N-linked sugar chains of hCG produced at different stages of gestation must be performed to confirm this evidence. The crystal structure of hCG was reported by Lapthorn et al. [69]. They indicated that only the sugar chains at Asn52 of hCG are present at the interface of - and -subunits. The other three N-linked sugar chains are located on the outer face of the heterodimer molecule. Purohit et al. [70] purified hCG samples from the culture media of insect cells transfected with the hCG gene, which lacked the potential N-glycosylation site at Asn52 or Asn78, together with the intact hCG gene. Their studies of the circular dichroism measurements and dissociation rates of the two mutant hCGs concluded that the absence of sugar chain at Asn52, but not at Asn78, resulted in conformational changes in the mutant. Based on this evidence, Purohit et al. considered that the loss of hormonal activity of hCG (lacking the sugar chain at Asn52) was probably due to a conformational change in the heterodimer rather than to the loss of interaction of the Asn52 sugar chain with the lectin on the target cells. A conformational study of the sugar chains of hCG in solution by NMR, however, revealed that the sugar chains at Asn52 appear to extend into solution [71,72]. This data supports the dual-receptor theory described in the previous section of this chapter. Thijssen-van Zuylen et al.
102
Human Chorionic Gonadotropin (hCG)
[73] later reported that the Asn52 linked sugar chain of the hCG is not susceptible to digestion with peptide N-glycosidase F, in contrast to that of free hCG. Therefore, more information is required for discussing the actual conformation of the sugar chains of hCG in solution. In 1998, Fukushima et al. [74] found that a part of choriocarcinoma patient serum hCG and hCG purified from the culture media of choriocarcinoma cell lines JEG-3 and BeWo binds to an immobilized Trichosanthes japonica agglutinin-I (TJA-I) column and an immobilized Trichosanthes japonica agglutinin-II (TJA-II) column; serum hCG from pregnant women completely passes through both columns without interaction. The TJA-I column specifically retains the oligosaccharides containing the Sia2–6Gal1–4GlcNAc group [75]; the TJA-II column retains the oligosaccharides containing the Fuc1–2Gal1 group [76]. Based on these finding, Fukushima et al. thoroughly investigated the activities of glycosyltransferases in human placenta, JEG-3 cells, and BeWo cells. The 2–3sialyltransferase activities in the microsomal fractions of placenta and from the two choriocarcinoma cell lines were almost the same; however, the 2–6sialyltransferase activities of the two choriocarcinoma cell lines were much higher than that of placenta. The 1–2 fucosyltransferase activities in the microsomal fractions of placenta, JEG-3, and BeWo were 0.14, 0.60, and 0.29 nmol/h/mg of microsomal protein, respectively. These results indicated that the activities of 2–6sialyltransferase and of 1–2 fucosyltransferase are increased significantly as a result of malignant transformation of trophoblasts. Measurement of the level of enzyme transcripts by competitive PCR revealed that ST6Gal-I [77] transcripts in JEG-3 and BeWo cells are approximately 40 times higher than that in placenta; no difference was detected in the level of ST3Gal-III [78] or ST3Gal-IV [79] transcripts among the three samples. By using the same technique, it was confirmed that the levels of the fucosyltransferase I and II transcripts [80], which are responsible for formation of the Fuc1–2Gal group in JEG-3 cells and BeWo cells, were more than 20 times higher than in placenta. Accordingly, the Sia2–6Gal1–4GlcNAc group and the Fuc1–2Gal1–4GlcNAc group are expressed on the N-linked sugar chains of hCG by malignant transformation of trophoblasts, and these alterations could be effectively used for the diagnosis of choriocarcinoma. More recently, Takamatsu et al. [81] reported additional interesting evidence regarding the N-linked sugar chains of hCG produced by malignant cells. As described already in Section 8.8, both GnT-III and GnT-IV increased tremendously in choriocarcinoma cell lines [36]. Because bisected complex-type sugar chains have never been found in the N-linked sugar chains of hCG to date, Takamatsu et al. chose JEG-3 cells, which produce the largest amount of hCG among the choriocarcinoma cell lines, and investigated the structures of the N-linked sugar chains of hCG produced by this cell line. As reported by Mizuochi et al. [18,19], seven oligosaccharides (except for B in Table 8.1) were detected; however, they also found bisected C, D, G, and H as major components, in accordance with the enhanced expression of GnT-III in JEG-3 cells. Accordingly, altered expression of glycosyltransferases in trophoblasts by malignant transformation could be more complicated than discussed in this review.
Glycobiology of hCG
103
References [1] Kobata A. Structures and functions of the sugar chains of glycoproteins. Eur J Biochem 1992;209:483–501. [2] Kornfeld R, Kornfeld S. Assembly of asparagine-linked oligosaccharides. Annu Rev Biochem 1985;54:631–64. [3] Hemming FW. The coenzymic role of phosphodolichols. In: Montreuil J, Vliegenthart JFG, Schachter H, editors. Glycoproteins. Amsterdam: Elsevier; 1995. pp. 127–44. [4] Bause E, Hardt B. Oligosaccharyltransferase complex, Ribophorin-I, Ribophorin-II, OST48, and DAD1. In: Taniguchi N, Honke K, Fukuda M, editors. Handbook of glycosyltransferases and related genes. Tokyo: Springer; 2002. pp. 577–86. [5] Ronin C, Granier C, Caseti C, Bouchilloux S, Van Rietschaten J. Synthetic substrates for thyroid oligosaccharide transferase. Effects of peptide chain length and modifications in the Asn-Xaa-Thr-region. Eur J Biochem 1981;118:159–64. [6] Stenflo J, Fernlund P. Amino acid sequence of the heavy chain of bovine protein C. J Biol Chem 1982;257:12180–90. [7] Stanley P. N-acetylglucosaminyltransferase-I. In: Taniguchi N, Honke K, Fukuda M, editors. Handbook of glycosyltransferases and related genes. Tokyo: Springer; 2002. pp. 61–9. [8] Moremen KW. -Mannosidase-II. In: Taniguch N, Honke K, Fukuda M, editors. Handbook of glycosyltransferases and related genes. Tokyo: Springer; 2002. pp. 600–6. [9] Bellisario R, Carlsen RB, Bahl OP. Human chorionic gonadotropin: linear amino acid sequence of the subunit. J Biol Chem 1973;248:6797–807. [10] Carlsen RB, Bahl OP, Swaminathan N. Human chorionic gonadotropin: linear amino acid sequence of the subunit. J Biol Chem 1973;248:6810–25. [11] Endo Y, Yamashita K, Tachibana Y, Tojo S, Kobata A. Structures of the asparagine-linked sugar chains of human chorionic gonadotropin. J Biochem 1979;85:669–79. [12] Kessler MJ, Reddy MS, Shah RH, Bahl OP. Structures of N-glycosidic carbohydrate units of human chorionic gonadotropin. J Biol Chem 1979;254:7901–8. [13] Mizuochi T, Kobata A. Different asparagine-linked sugar chains on the two polypeptide chains of human chorionic gonadotropin. Biochem Biophys Res Commun 1980;97:772–8. [14] Swaminathan N, Bahl OP. Dissociation and recombination of the subunits of human chorionic gonadotropin. Biochem Biophys Res Commun 1970;40:422–7. [15] Kawano T, Endo T, Nishimura R, Mizuochi T, Mochizuki M, Kochibe N, et al. Structural differences found in the sugar chains of eutopic and ectopic free -subunits of human glycoprotein hormone. Arch Biochem Biophys 1988;267:787–96. [16] McReynolds L, O’Malley BW, Nisbet AD, Fothergill JE, Givol D, Fields S, et al. Sequence of chicken ovalbumin mRNA. Nature 1978;273:723–8. [17] Matzuk MM, Boime I. The role of the asparagine-linked oligosaccharides of the subunit in the secretion and assembly of human chorionic gonadotropin. J Cell Biol 1988;106:1049–59. [18] Mizuochi T, Nishimura R, Derappe C, Taniguchi T, Hamamoto T, Mochizuki M, et al. Structures of the asparagine-linked sugar chains of human chorionic gonadotropin produced in choriocarcinoma: appearance of triantennary sugar chains and unique biantennary sugar chains. J Biol Chem 1983;258:14126–9. [19] Mizuochi T, Nishimura R, Taniguchi T, Utsunomiya T, Mochizuki M, Derappe C, et al. Comparison of carbohydrate structure between human chorionic gonadotropin present in urine of patients with trophoblastic diseases and healthy individuals. Jpn J Cancer Res 1985;76:752–9.
104
Human Chorionic Gonadotropin (hCG)
[20] Endo T, Nishimura R, Kawano T, Mochizuki M, Kobata A. Structural differences found in the asparagine-linked sugar chains of human chorionic gonadotropins purified from the urine of patients with invasive mole and with choriocarcinoma. Cancer Res 1987;47:5242–5. [21] Miyoshi E, Taniguchi N. 6-Fucosyltransferase (FUT8). In: Taniguchi N, Honke K, Fukuda M, editors. Handbook of glycosyltransferases and related genes. Tokyo: Springer; 2002. pp. 259–63. [22] Gleeson PA, Schachter H. Control of glycoprotein synthesis. J Biol Chem 1983;258: 6162–73. [23] Yamashita K, Totani K, Kuroki M, Matsuoka Y, Ueda I, Kobata A. Structural studies of the carbohydrate moieties of carcinoembryonic antigens. Cancer Res 1987;47:3451–9. [24] Yamashita K, Totani K, Iwaki Y, Takamizawa I, Tateishi N, Higashi T, et al. Comparative study of the sugar chains of -glutamyltranspeptidases purified from human hepatocellular carcinoma and from human liver. J Biochem 1989;10:728–35. [25] Kessler MJ, Mise T, Ghai RD, Bahl OP. Structure and location of the O-glycosidic carbohydrate units of human chorionic gonadotropin. J Biol Chem 1979;254:7909–14. [26] Cole LA, Birken S, Perini F. The structures of the serine-linked sugar chains on human chorionic gonadotropin. Biochem Biophys Res Commun 1985;126:333–9. [27] Amano J, Nishimura R, Mochizuki M, Kobata A. Comparative study of the mucin-type sugar chains of human chorionic gonadotropin present in the urine of patients with trophoblastic diseases. J Biol Chem 1988;263:1157–65. [28] Yamashita K, Tachibana T, Ohkura T, Kobata A. Enzymatic basis for the structural changes of asparagine-linked sugar chains of membrane glycoproteins of baby hamster kidney cells induced by polyoma transformation. J Biol Chem 1985;260:3963–9. [29] Koenderman AH, Wijermans PW, van den Eijnden DH. Changes in the expression of N-acetylglucosaminyltransferase III, IV, V associated with the differentiation of HL-60 cells. FEBS Lett 1987;222:42–6. [30] Nishikawa A, Gu J, Fujii S, Taniguchi N. Determination of N-acetylglucosaminyltransferases III, IV and V in normal and hepatoma tissues of rats. Biochim Biophys Acta 1990;1035:313–18. [31] Ogier-Denis E, Bauvy C, Moutsia R, Aubery M, Codogno P. Increased UDP-GlcNAc: alpha-mannoside beta (1–4) N-acetylglucosaminyltransferase activity during chick embryo development. Biochim Biophys Acta 1990;1054:149–53. [32] Oguri S, Minowa MT, Ihara Y, Taniguchi N, Ikenaga H, Takeuchi M. Purification and characterization of UDP-N-acetylglucosamine:alpha1,3-d-mannoside beta1,4-N-acetylglucosaminyltransferase (N-acetylglucosaminyltransferase-IV) from bovine small intestine. J Biol Chem 1997;272:22721–7. [33] Minowa MT, Oguri S, Yoshida A, Hara T, Iwamatsu A, Ikenaga H, et al. cDNA cloning and expression of bovine UDP-N-acetylglucosamine: alpha1,3-d-mannoside beta1,4-Nacetylglucosaminyltransferase IV. J Biol Chem 1998;273:11556–62. [34] Yoshida A, Minowa MT, Takamatsu S, Hara T, Ikenaga H, Takeuchi M. A novel second isoenzyme of the human UDP-N-acetylglucosamine:alpha1,3-d-mannoside beta1,4N-acetylglucosaminyltransferase family: cDNA cloning, expression, and chromosomal assignment. Glycoconjugate J 1998;15:1115–23. [35] Yoshida A, Minowa MT, Takamatsu S, Hara T, Oguri S, Ikenaga H, et al. Tissue-specific expression and chromosomal mapping of a human UDP-N-acetylglucosamine: alpha1,3d-mannoside beta1,4-N-acetylglucosaminyltransferase. Glycobiology 1999;9:303–10. [36] Takamatsu S, Oguri S, Yoshida A, Minowa MT, Nakamura K, Takeuchi M, et al. Unusually high expression of N-acetylglucosaminyltransferase-IVa in human choriocarcinoma cell lines: a possible enzymatic basis of the formation of abnormal biantennary sugar chain. Cancer Res 1999;59:3949–53.
Glycobiology of hCG
105
[37] Yamashita K, Totani K, Ohkura T, Takasaki S, Goldstein IJ, Kobata A. Carbohydrate binding properties of complex-type oligosaccharides on immobilized Datura stramonium lectin. J Biol Chem 1987;262:1602–7. [38] Endo T, Iino K, Nozawa S, Iizuka A, Kobata A. Immobilized Datura stramonium agglutinin column chromatography, a novel method to discriminate the urinary hCG of patients with invasive mole and choriocarcinoma from those of normal pregnant women and patients with hydatidiform mole. Jpn J Cancer Res 1988;79:160–4. [39] Elliott MM, Kardana A, Lustbader JW, Cole LA. Carbohydrate and peptide structure of alpha- and beta-subunits of human chorionic gonadotropin from normal and aberrant pregnancy and choriocarcinoma. Endocrine 1997;7:15–32. [40] Valmu L, Alfthan H, Hotakainen K, Birken S, Stenman UH. Site-specific glycan analysis of human chorionic gonadotropin -subunit from malignancies and pregnancy by liquid chromatography–electrospray mass spectrometry. Glycobiology 2006;16:1207–18. [41] Cole LA, Khanlian SA, Sutton JM, Davies S, Stephens N. Hyperglycosylated hCG (invasive trophoblast antigen, ITA), a key antigen for early pregnancy detection. Clin Biochem 2003;36:647–55. [42] Birken S, Krichevsky A, O’Connor J, Schlatterer J, Cole LA, Kardana A, et al. Development and characterization of antibodies to a nicked and hyperglycosylated form of hCG from a choriocarcinoma patient. Endocrine 1999;10:137–44. [43] Cole LA, Shahabi S, Oz UA, Bahado-Singh RO, Mahoney MJ. Hyperglycosylated hCG (invasive trophoblast antigen) immunoassay: a new basis for gestational Down syndrome screening. Clin Chem 1999;45:2109–19. [44] Cole LA, Butler SA, Khanlian SA, Giddings A, Muller CY, Seckl MJ, et al. Gestational trophoblastic diseases: 2. Hyperglycosylated hCG as a reliable marker of active neoplasia. Gyn Oncol 2006;102:150–8. [45] Cole LA, Muller Y. hCG in the management of quiescent and chemorefractory gestational trophoblastic diseases. Gyn Oncol 2010;116:3–9. [46] O’Connor JF, Ellish N, Kakuma T, Schlatterer J, Kovalevskaya G. Differential urinary gonadotropin profiles in early pregnancy and early pregnancy loss. Prenat Diagn 1998;18: 1232–40. [47] Kovalevskaya G, Birken S, Kakuma T, Ozaki N, Sauer M, Lindheim S, et al. Differential expression of human chorionic gonadotropin (hCG) glycosylation isoforms in failing and continuing pregnancies: preliminary characterization of the hyperglycosylated hCG epitope. J Endocrinol 2002;172:497–506. [48] Sasaki Y, Ladner DG, Cole LA. Hyperglycosylated hCG: the source of pregnancy failures. Fertil Steril 2008;89:1871–86. [49] Pandian R, Cole LA, Palomaki GE. Second-trimester maternal serum invasive trophoblast antigen: a marker for Down syndrome screening. Clin Chem 2004;50:1433–5. [50] Bahado-Singh RO, Oz AU, Kingston JM, Shahabi S, Hsu CD, Cole LA. The role of hyperglycosylated hCG in trophoblast invasion and the prediction of subsequent preeclampsia. Prenat Dagn 2002;22:478–81. [51] Moyle WR, Bahl OP, Maerz L. Role of carbohydrate of human chorionic gonadotropin in the mechanism of hormone action. J Biol Chem 1975;250:9163–9. [52] Chen H-C, Shimohigashi Y, Dufau ML, Catt KJ. Characterization and biological properties of chemically deglycosylated human chorionic gonadotropin: role of carbohydrate moieties in adenylate cyclase activation. J Biol Chem 1982;257:14446–52. [53] Manjunath P, Sairam MR. Biochemical, biological, and immunological properties of chemically deglycosylated human choriogonadotropin. J Biol Chem 1982;257: 7109–15.
106
Human Chorionic Gonadotropin (hCG)
[54] Amano J, Nishimura R, Sato S, Kobata A. Altered glycosylation of human chorionic gonadotropin decreases its hormonal activity as determined by cyclic-adenosine 3’,5’monophosphate production in MA-10 cells. Glycobiology 1990;1:45–50. [55] Goverman JM, Parson THF, Pierce JG. Enzymatic degradation of the subunits of chorionic gonadotropin: effects on formation of tertiary structure and biological activity. J Biol Chem 1982;257:15059–64. [56] Calvo FO, Ryan RJ. Inhibition of adenylyl cyclase activity in rat corpora luteal tissue by glycopeptides of human chorionic gonadotropin and the alpha-subunit of hCG. Biochemistry 1985;24:1953–9. [57] Thotakura NR, Weintraub BD, Bahl OP. The role of carbohydrate in human choriogonadotropin (hCG) action: effects of N-linked carbohydrate chains from hCG and other glycoproteins on hormonal activity. Mol Cell Endocrinol 1990;70:263–72. [58] Matzuk MM, Boime I. Site-specific mutagenesis defines the intracellular role of the asparagine-linked oligosaccharides of chorionic gonadotropin subunit. J Biol Chem 1988; 263:17106–11. [59] Matzuk MM, Keene JL, Boime I. Site specificity of the chorionic gonadotropin N-linked oligosaccharides in signal transduction. J Biol Chem 1989;264:2409–14. [60] Feng W, Matzuk MM, Mountjoy K, Bedows E, Ruddon RW, Boime I. The asparaginelinked oligosaccharides of the human chorionic gonadotropin beta subunit facilitate correct disulfide bond pairing. J Biol Chem 1995;270:11851–9. [61] Amano J, Sato S, Nishimura R, Mochizuki M, Kobata A. Sialic acids, but not their linkage to galactose residues, are required for the full expression of the biological activity of human chorionic gonadotropin. J Biochem 1989;105:339–40. [62] Ascoli M. Characterization of several clonal lines of cultured leydig tumor cells: gonadotropin receptors and steroidogenic responses. Endocrinology 1981;108:88–95. [63] Nemansky M, DeLeeuw R, Wijnands RA, van den Eijnden DH. Enzymic remodeling of the N- and O-linked carbohydrate chains of human chorionic gonadotropin: effects on biological activity and receptor binding. Eur J Biochem 1995;227:880–8. [64] Amano J, Kobata A. Direct interaction of the sialic acid residue of human lutropin and chorionic gonadotropin with target cell is necessary for the full expression of their hormonal action. Arch Biochem Biophys 1993;305:618–21. [65] McFarland KC, Sprengel R, Phillips HS, Kohler M, Rosenblit N, Nikolics K, et al. Lutropin-choriogonadotropin receptor: an unusual member of the G protein-coupled receptor family. Science 1989;245:494–9. [66] Amano J, Kobata A. Structures and function of the N-linked sugar chains of glycohormones. In: biopolymers and bioproducts: structures, function and application. Bangkok: Samakkhisan Public C; 1995. pp. 210–7. [67] Weisshaar G, Hiyama J. Renwick AGC. Site-specific N-glycosylation of human chorionic gonadotropin-structural analysis of glycopeptides by one- and two-dimensional 1H NMR spectroscopy. Glycobiology 1991;1:393–404. [68] Skarulis MC, Wehmann RE, Nisula BC, Blithe DL. Glycosylation changes in human chorionic gonadotropin and free alpha subunit as gestation progresses. J Clin Endocrinol Metab 1992;75:91–6. [69] Lapthorn AJ, Harris DC, Littlejohn A, Lustbader JW, Canfield RE, Machin KJ, et al. Crystal structure of human chorionic gonadotropin. Nature 1994;369:455–61. [70] Purohit S, Shao K, Balasubramanian SV, Bahl OP. Mutant of human choriogonadotropin lacking N-glycosyl chains in the alpha-subunit. 1. Mechanism for the differential action of the N-linked carbohydrates. Biochemistry 1997;36:12355–63.
Glycobiology of hCG
107
[71] Weller CT, Lustbader J, Seshadri k, Brown JM, Chadwick CA, Kolthoff CE, et al. Structural and conformational analysis of glycan moieties in situ on isotopically 13C, 15N-enriched recombinant human chorionic gonadotropin. Biochemistry 1996;35:15–23. [72] DeBeer T, van Zuylen CW, Leeflang BR, Hard K, Boelens R, Kaptein R, et al. NMR studies of the free alpha subunit of human chorionic gonadotropin: structural influences of N-glycosylation and the beta subunit on the conformation of the alpha subunit. Eur J Biochem 1996;241:229–42. [73] CWEM Thijssen-van Zuylen, de Beer T, Leeflang BR, Boelens R, Kaptein R, Kamerling JP, et al. Mobilities of the inner three core residues and the man(alpha 1–6) branch of the glycan at Asn78 of the alpha-subunit of human chorionic gonadotropin are restricted by the protein. Biochemistry 1998;37:1933–40. [74] Fukushima K, Hara-Kuge S, Seko A, Ikehara Y, Yamashita K. Elevation of 2–6 sialyltransferase and 1–2 fucosyltransferase activities in human choriocarcinoma. Cancer Res 1998;58:4301–6. [75] Yamashita K, Umetsu K, Suzuki T, Ohkura T. Purification and characterization of a Neu5Ac2–6Gal1–4GlcNAc and HSO3-6Gal1–4GlcNAc specific lectin in tuberous roots of Trichosanthes japonica. Biochemistry 1992;31:11647–50. [76] Yamashita K, Ohkura T, Umetsu K, Suzuki T. Purification and characterization of a Fuc1–2Gal1- and GalNAc1-specific lectin in root tubers of Trichosanthes japonica. J Biol Chem 1992;267:25414–22. [77] Hamamoto T, Tsuji S. ST3Gal-I. In: Taniguchi N, Honke K, Fukuda M, editors. Handbook of glycosyltransferases and related genes. Tokyo: Springer; 2002. p. 295–300. [78] Kitazume-Kawaguchi S, Tsuji S. ST3Gal-III. In: Taniguchi N, Honke K, Fukuda M, editors. Handbook of glycosyltransferases and related genes. Tokyo: Springer; 2002. pp. 279–83. [79] Kitazume-Kawaguchi S, Tsuji S. ST3Gal-IV. In: Taniguchi N, Honke K, Fukuda M, editors. Handbook of glycosyltransferases and related genes. Tokyo: Springer; 2002. pp. 284–8. [80] Oriol R, Mollicone R. 2-Fucosyltransferases (FUT1, FUT2, and Sec1). In: Taniguchi N, Honke K, Fukuda M, editors. Handbook of glycosyltransferases and related genes. Tokyo: Springer; 2002. pp. 205–17. [81] Takamatsu S, Katsumata T, Inoue N, Watanabe T, Fujibayashi Y, Takeuchi M. Abnormal biantennary sugar chains are expressed in human choriocarcinoma cell line JEG-3. Glycoconj J 2004;20:473–81.
9 Degradation Products of hCG,
Hyperglycosylated hCG, and Free -Subunit Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
Trophoblast cells make excess amounts of hCG and hyperglycosylated hCG (hCG-H). The amounts of hCG and hCG-H promote progesterone production, promote invasion as part of implantation, and promote uterine spiral artery angiogenesis, uterine growth, trophoblast differentiation, myometrial muscle relaxation, and immune suppression hCG [1]. Humans have intricate mechanisms working to degrade and remove hCG from the body’s circulation. These mechanisms are particularly useful during the third trimester of pregnancy, when hCG no longer plays a significant role in placental function. Four competing pathways work to clear both hCG and hCG-H. Each one abates hCG biological activity [1]. First, hCG is cleared from the circulation by the liver (clears 78% total hCG); the kidneys only clear 22% of total hCG through the urine [2]. hCG is an acidic glycoprotein (pI 3.5) and has an overall circulating half-life of 36 h [3,4]. In contrast, the neutral glycoprotein luteinizing hormone (LH) (pI 8.0) has an overall circulating half-time of just 26 min [5]. When hCG is desialylated, the acidic sugar is removed. This reduces the clearance rate of hCG (how rapidly hCG is removed) more than 300-fold, from 36 h to minutes [3]. hCG is synthesized with neither the maximal component of sialic acid residues nor exposed galactose sugar residues. Over time, sialic acid residues are lost by hCG and cleaved by macrophage neuraminidases; this results in ultimate liver clearance [6–8]. With advancement of pregnancy, hCG is generally less sialylated (less acidic) and as a result clears more rapidly in the latter trimesters of pregnancy. hCG, and particularly hCG-H, undergoes continuous dissociation (Table 9.1). The remaining three pathways involve urine clearance. First, the resulting free -subunit and free -subunit components are cleared from the circulation much more rapidly than hCG [9,10]. Second, hCG is continuously nicked/cleaved upon secretion at 44–45 or 47–48. The nicking opens the structure of hCG and leads to fivefold more rapid dissociation (see Table 9.1) [11,12]. Nicking is the first cleavage step in the synthesis of -core fragment (the end product of hCG degradation found only in urine) (Figure 9.1). Third, enzymes can cleave and remove the C-terminal region of the hCG -subunit. By removing the O-linked oligosaccharides, hCG acidity is Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00009-8 © 2010 Elsevier Inc. All rights reserved.
110
Human Chorionic Gonadotropin (hCG)
Table 9.1 The Clearance of hCG and Its Degradation Products in Circulation [3–12]. 1. Time of Dissociation of Dimer in Serum in the Laboratory at 37 °C Third I.S. hCG hCG-H Nicked hCG Nicked hCG-H
Dissociation half-time 240 h Dissociation half-time 120 h Dissociation half-time 44 h Dissociation half-time 22 h
2. Effect of Leukocyte Elastase at 37 °C Third I.S. hCG, 5 nmol, 1 h incubation CR129 -subunit, 5 nmol, 1 h incubation
30% nicked at 44–35 100% nicked at 44–45
3. Time for Injected Molecules in Humans to Leave the Circulation Third I.S. hCG -Subunit -Subunit Asialo hCG -Core fragment Third I.S. hCG -Subunit -Subunit Asialo hCG -Core fragment
Fast phase half-time 5.97 h Fast phase half-time 0.68 h Fast phase half-time 0.22 h Fast phase half-time 0.060 h Fast phase half-time 0.058 h Slow phase half-time 36 h Slow phase half-time 3.9 h Slow phase half-time 1.3 h Slow phase half-time 0.096 h Slow phase half-time 0.38 h
4. Time for Natural Molecules to Leave the Circulation at Parturition hCG post-parturition Free -subunit post-parturition Free -subunit post-parturition
Disappearance half-time 38–64 h Disappearance half-time 103–462 h Disappearance half-time 15–126 h
Four pathways are examined: 1. Time of dissociation of dimer in serum in the laboratory at 37 °C; 2. Effect of leukocyte elastase at 37 °C; 3. Time for injected molecules in humans to leave the circulation; 4. Time for natural molecules to leave the circulation at parturition.
significantly reduced and circulating half-life is diminished [11,12]. The C-terminus cleavage is particularly evident in the serums of both gestational trophoblastic disease patients and cancer patients [11–13]. C-terminus cleavage is also part of the cleavage process that leads to -core fragment formation (Figure 9.1). All these pathways are discussed and compared in detail later in this chapter.
9.1 Pure hCG Preparations As described earlier, the hCG degradation and clearance processes are always at work. All natural hCG preparations and standards contain a heterogeneous mixture of intact, dissociated, nicked, and cleaved molecules. The only exception is recombinant
Degradation Products of hCG, Hyperglycosylated hCG, and Free -Subunit
111
Figure 9.1 Cleavage sites and proteases in the degradation of hCG, hCG-H, and free -subunit.
hCG synthesized in Chinese hamster ovary cells. In one publication [14], 13 purified hCG and hCG-H preparations were examined: 11 of the 13 preparations were nicked/cleaved between 44–45 or 47–48; 2 of 13 preparations were missing the -subunit C-terminus (cleaved at 92–93); and 7 of 13 preparations were cleaved at the N-terminal section of -subunit. In a later study [15], which investigated the structure of individual hCG and hCG-H preparations as well as the CR series preparation (hCG standards for World Health Organization), 12 of 14 individual hCG preparations and all the CR series standards were nicked to some extent; 3 of 14 were missing the -subunit C-terminal peptide; and 13 of 14 (which included all CR series standards) were cleaved at the N-terminus of the -subunit [15]. As a result of
112
Human Chorionic Gonadotropin (hCG)
these two studies, it is now acceptable to conclude that the amino acid sequences of all hCG and hCG-H preparations (except recombinant hCG) contain molecules that are cleaved to some extent. Thus, hCG preparations are really always a mixture of intact hCG, nicked hCG, hCG missing the C-terminal peptide, and -subunit cleaved N-terminus hCG.
9.2 Nicking and Enzyme Cleavage As illustrated in Figure 9.2, trophoblastic villi contain monocytes and placental macrophages (Hofbauer cells) and are associated with classic macrophage structures [10,15]. Both of these are white blood cell types that produce a leukocyte elastase that cleaves proteins at Leu, Gly, and Val residues. Leukocyte elastase acts on newly synthesized cytotrophoblast hCG-H and syncytiotrophoblast hCG molecules [11,12,16]. Figure 9.1 shows cleavage sites on the - and -subunits of hCG. These include the -subunit N-terminal cleavage sites at 2,3 and 3,4, the -subunit 38–57 loop cleavage site, the -subunit 5–6 N-terminal cleavage site and the -subunit 92–93 C-terminal cleavage site. Many of these sites are cleaved by leukocyte elastase. It has been shown [11,12] that incubating hCG or -subunit with serum leads to nicking at 44–45 and 47–48. Longer incubation periods lead to cleavage of the C-terminus at 92–93, which also appears to be due to leukocyte elastase. These actions were repeated using purified human leukocyte elastase, and yielded similar results [11,16]. -core fragment is the terminal urinary degradation product of hCG. It is comprised of -subunit residues 6–40 that are linked by disulfides to the -subunit residue 55–92. -core fragment appears to be generated by cleavage of the C-terminus at 92–93 and nicking at 44–45 or 47–58 by leukocyte elastase. Carboxypeptidases
Figure 9.2 Placenta floating villus stained with antibody B204 (binds -subunit, nicked -subunit, -subunit missing C-terminus, and -core fragment). The villi contain numerous monocytes stained from associated hCG cleavage products [1] as well as a placental macrophage (Hofbauer cell complex) [2]. An associated macrophage is shown attached to syncytiotrophoblast cells [3].
Degradation Products of hCG, Hyperglycosylated hCG, and Free -Subunit
113
and aminopeptidases then remove 40–47 and 48–55, taking cleavage from the nicking site to as far as two amino acids away from a disulfide bridge on either side (Figure 9.1). How the amino terminus of -subunit is cleaved has not been established. It could be the result of an aminopeptidase action or of further leukocyte elastase cleavage at 5–6. In terms of hCG clearance, the cleavage process is exponential. Once hCG is nicked by leukocyte elastase, it dissociates 5 times more rapidly and clears from the circulation 7 times more rapidly (Figure 9.1). It should be noted that if the nicked and dissociated molecules are cleared through the kidney, they will become -core fragment as an end product. Nicked -subunit and nicked -subunit missing the C-terminus are both commonly detected in serum [13], but -core fragment is only detected in urine [17]. It is thought that the aminopeptidase and carboxypeptidase stages of -core fragment synthesis are confined strictly to the kidneys. It has been shown that administration of pure recombinant hCG to humans leads to 13% of the administered amount being degraded to urine -core fragment. It is assumed that circulating leukocyte elastase and dissociation start the degradation process [17]. Certain cleavage processes are unique to hCG produced in choriocarcinoma and other cancers. For example, both nicking at an Arg residue 43–44 and cleavage of the C-terminus are processes unique to choriocarcinoma and other cancer cases [15]. Many hCG tests incorporate an antibody to the -subunit C-terminal peptide to specifically bind the hormone. Tests that use these antibodies should be avoided when monitoring cancer cases if a C-terminal peptide is absent [11] (see Chapter 21).
9.3 Dissociation Dissociation is the central event in the degradation and clearance processes of hCG. The dissociation half-life of intact hCG is extremely slow, 240 h in serum at 37 °C (Table 9.1). hCG-H dissociates approximately twice as fast as regular hCG, but nicked hCG and nicked hCG-H dissociate 5 times faster (Table 9.1), which leads to rapid clearance of nicked hCG - and -subunits. Each event in the clearance process appears to be exponential. If hCG is nicked, for example, it dissociates 5 times faster and is cleared from the circulation 7 times faster. If hCG or hCG-H is dissociated, the free -subunit is subsequently nicked much more rapidly (Table 9.1). It appears that nicking and dissociation work together, each speeding up the clearance processes of the other’s product.
9.4 Liver Clearance Most hCG is cleared from the circulation through the liver (78%), but some is cleared by the kidneys (22%) [2]. hCG is an acidic glycoprotein (pI 3.5). When it is desialylated (removing all the acidic sugar), it speeds up clearance to more than 300 times the base value [4]. hCG can be synthesized either without the complete component
114
Human Chorionic Gonadotropin (hCG)
of sialic acid residues or with exposed galactose residues. Over time, sialic acid residues are lost and are cleaved by macrophage neuraminidases. This leads to rapid liver clearance and utilization of the liver galactose receptor [6–8].
9.5 Degradation with Pregnancy Advancement A general parallelism is observed between increasing concentration of degradation products during advancing pregnancy (Figures 9.3 and 9.4). During 4–40 weeks of pregnancy, a continuous rise in free -subunit and -core fragment in urine is noted. This rise is apparently due to the dissociation and degradation of hCG. A similar rise is observed in nicked hCG and the free -subunit in serum [11]. During the first 2 months of pregnancy, 9% of hCG is nicked in serum, 13% in the next 2 months, 17% in the next 2 months, and 21% in the final 3 months of pregnancy [11]. In urine, -core fragment concentration is higher than that of hCG (nmol/l) throughout most of pregnancy. In the third trimester, the median -core fragment concentration is approximately 10-fold greater than the median hCG concentration.
Figure 9.3 Serum concentrations of hCG, free -subunit, free -subunit, and nicked hCG during the course of pregnancy [10].
Degradation Products of hCG, Hyperglycosylated hCG, and Free -Subunit
115
As pregnancy advances, increased proportions of hCG are degraded through nicking, dissociation, cleavage of -subunit C-terminus, and the processes described in this chapter. All the clearance pathways are summarized in Figure 9.5. All lead to the rapid clearance of hCG from the circulation. As shown in Table 9.1, following parturition of pregnancy, hCG disappears from the circulation more rapidly than free - and -subunits [18]. This is a direct result of the degradation protocol, when free - and -subunits are being continuously generated. In the weeks following parturition, the degradation protocol encompasses nicking, dissociation, cleavage of -subunit C-terminal peptide, and aminopeptidase and carboxypeptidase action, as indicated in Figure 9.5. It is estimated that following a first-trimester termination of pregnancy, hCG takes an average of 30 days to clear the body [19,20]; following an ectopic pregnancy, it can take on average 24 days to clear [21]; and following a term pregnancy, it takes on average 28 days to clear [18]. The more acidic hCG-H can take as long as 65 days to clear the body [22]. Longer clearance times have been observed with hydatidiform mole, choriocarcinoma, and testicular cancer.
Figure 9.4 Urine concentrations of hCG, free -subunit, free -subunit, and nicked hCG during the course of pregnancy [10].
116 Human Chorionic Gonadotropin (hCG)
Figure 9.5 Pathways in the degradation, urinary, and liver clearance of hCG, hCG-H, and the free -subunit. Nicking is cleavage at 44–45 and 47–48. Aminopeptidases and carboxypeptidases cleave nicked molecules to -subunit 40–55 as found in -core fragment. Cleavage CTP is cleavage of the -subunit C-terminal region at 92–93 [3–12].
Degradation Products of hCG, Hyperglycosylated hCG, and Free -Subunit
117
References [1] Cole LA. Human chorionic gonadotropin and associated molecules. Expert Rev Mol Diag 2009;9:51–73. [2] Nisula BC, Blithe DL, Akar A, Lefort G, Wehmann RE. Metabolic fate of human chorionic gonadotropin. J Steroid Biochem 1989;33:733–7. [3] Strickland TW, Puett D. The kinetic and equilibrium parameters of subunit association and dissociation. J Biol Chem 1982;257:2954–60. [4] Rosa C, Amr S, Birken S, Wehmann R, Nisula B. Effects of desialylation on human chorionic gonadotropin on its metabolic clearance rate in humans. J Clin Endocrinol Metab 1984;59:1215–19. [5] Veldhuis JD, Fraioli F, Rogol AD, Dufau ML. Metabolic clearance of biologically active luteinizing hormone in man. J Clin Invest 1986;77:1122–8. [6] Lefort GP, Stolk JM, Nisula BC. Evidence that desialylation and uptake by hepatic receptors for galactose-terminated glycoproteins are immaterial to the metabolism of human choriogonadotropin in the rat. Endocrinol 1984;115:1551–7. [7] Van Hall EV, Vaitukaitis JL, Ross GT, Hickman JW, Ashwell G. Effects of progressive desialylation on the rate of disappearance of immunoreactive hCG from plasma in rats. Endocrinol 1971;89:11–15. [8] Wide L, Hobson B. Some qualitative differences of hCG in serum from early and late pregnancies and trophoblastic diseases. Acta Endocrinol 1987;116:465–72. [9] Wehmann RE, Nisula BC. Metabolic clearance rates of the subunits of human chorionic gonadotropin. J Clin Endocrinol Metab 1979;48:753–9. [10] Wemann RE, Nisula BC. Metabolic and renal clearance rates of purified human chorionic gonadotropin. J Clin Invest 1981;68:184–94. [11] Cole LA, Karadana A, Park S-Y, Braunstein GD. The deactivation of hCG by nicking and dissociation. J Clin Endocrinol Metab 1993;76:704–15. [12] Karadana A, Cole LA. Human chorionic gonadotropin -subunit nicking enzymes in pregnancy and cancer patient serum. J Clin Endocrinol Metab 1994;79:761–7. [13] Cole LA, Shahabi S, Butler S, Mitchell H, Newlands ES, Behrman HR, et al. Utility of commonly used commercial hCG immunoassays in the diagnosis and management of trophoblastic diseases. Clin Chem 2001;47:308–915. [14] Kardana A, Elliott ME, Gawinowicz MA, Birken S, Cole LA. The heterogeneity of hCG: I. Characterization of peptide variations in 13 individual preparations of hCG. Endocrinol 1991;129:1541–50. [15] Elliott MM, Kardana A, Lustbader JW, Cole LA. Carbohydrate and peptide structure of the - and -subunits of human chorionic gonadotropin from normal and aberrant pregnancy and choriocarcinoma. Endocrine 1997;7:15–32. [16] Cole LA, Kardana A, Andrade-Gordon P, Gawinowicz MA, Morris JC, Bergert ER, et al. The heterogeneity of hCG: III. The occurrence, biological and immunological activities of nicked hCG. Endocrinol 1991;129:1559–67. [17] Norman RJ, Buchholtz MW, Somogyl AA, Amato F. hCG- core fragment is a metabolite of hCG: evidence from the infusion of recombinant hCG. J Endocrinol 2000;164:299–305. [18] Korhonen J, Alfhan H, Ylostalo P, Veldhuis J, Stenmam U-H. Disappearance of human chorionic gonadotropin and its - and -subunits after term pregnancy. Clin Chem 1997;43:2155–63. [19] Aral K, Gunrkan Zorlu C, Gokman O. Plasma human chorionic gonadotropin levels after induced abortion. Adv Contracept 1996;12:11–14.
118
Human Chorionic Gonadotropin (hCG)
[20] Lajos L, Pali K. Choriongonadotropin clearance tests. Endokrinologie 1951;28:129–34. [21] Kamrava MM, Taymor ML, Berger MJ, Thompson IE, Seibel MM. Disappearance of human chorionic gonadotropin following removal of ectopic pregnancy. Obstet Gynecol 1983;62:486–8. [22] Cassels JW, Jr, Mann K, Blithe DL, Nisula BC, Wehmann RE. Reduced metabolic clearance of acidic variants of human choriogonadotropin from patients with testicular cancer. Cancer 1989;64:2313–18.
10 Three-Dimensional Structure of hCG
Laurence A. Cole USA hCG Reference Service, Albuquerque NM USA
For years, Lapthorn, Lustbader, Wu, and their colleagues worked meticulously to generate hCG crystals for X-ray crystallography in hopes of determining the fine, three-dimensional (3D) structure of hCG. Unfortunately, the heterogeneity of the eight carbohydrate side-chains prevented any crystallization. Even desialylization, or removal of the principal charged carbohydrate component, did not allow fine crystal formation [1]. Using an anhydrous hydrofluoric acid treatment, they carefully removed the N-linked oligosaccharides from hCG, leaving the deglycosylated protein intact. To remove the four O-linked oligosaccharides, they cleaved the 34-amino-acid C-terminal peptide at 111. It was assumed that the C-terminal peptide was a random, nonfolded component, similar to a semi-polymer of proline and serine. Finally, after successfully preparing hCG crystals, X-ray crystallographic analysis was finally performed [1–5]. The fine 3D structure of deglycosylated hCG was determined and disulfide bridges, charge interactions, hydrophobic interactions, and hydrogen bonding of the – dimer were all investigated [2,5]. Lapthorn, Lustbader, and Wu surely tried their best to resolve this impossible dilemma, but it is clear that X-ray crystallographic methods will probably never reveal the actual structure of the real hormone; glycosylated hCG with sugars greatly affecting structure; hCG-H, the invasion-promoting autocrine; pituitary hCG, the sulfated form of hCG; and hyperglycosylated free -subunit as produced by cancer cells. They are four separate molecules. They bind three or four different receptors and have separate functions, but they all share the same deglycosylated structure. Is this deglycosylated structure representative of hCG, hCG-H, pituitary hCG, or free -subunit? What do the folding and linkages of deglycosylated hCG represent? This is the mysterious root molecule, which never exists in the cell. What does deglycosylated hCG tell us about hCG, hCG-H, pituitary hCG, and free -subunit? In some ways, this chapter has a misleading title, because “The ThreeDimensional Structure of hCG” is actually based on pure speculation. We subjectively predict what hCG, hCG-H, pituitary hCG, and free -subunit might really look like. Nonetheless, we will carefully examine the X-ray crystallography of a hypothetical deglycosylated root molecule. X-ray crystallography has shown us sequences common with the evolution of TGF [6] and unique-to-TGF-like cystine knots. Each of these will be discussed in the following sections. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00010-4 © 2010 Elsevier Inc. All rights reserved.
120
Human Chorionic Gonadotropin (hCG)
10.1 hCG The deglycosylated form of hCG binds to the luteinizing hormone (LH)/hCG receptor but with greatly reduced affinity [2,4]. Because of its weak binding to the LH/ hCG receptor, it might be considered a form of regular hCG. Adding oligosaccharides increases the molecular weight of hCG by 25–30%, but does this really change its shape? We know that three disulfide bonds are added quickly to -subunit at linkages at 9–57, 34–88, and 38–90. These are the -subunit cystine-link inherited disulfide bonds that come from hCG’s common evolutionary root with TGF [6]. As has now been shown, these linkages form the core structure of the hCG -subunit. They are essential to the formation of – dimer structures and to LH/hCG receptor binding [3,7]. We ask ourselves again, is regular hCG different from the deglycosylated structure (Figure 10.1)? Adding charge from sialic acid lowers the isoelectric point of hCG from pI 8.0 to pI 3.7 and extends the circulating half-life from 0.060 to 5.97 h [8], but does it do anything more? We speculate that the structure of glycosylated hCG is different from the structures proposed by Wu et al. [4,5] and Lapthorn et al. and Lustbader et al. [1–3]. In Figure 10.2, light gray ellipses show the region that might be affected by the charged oligosaccharides in 3D folding. We know that regular hCG is cleaved by a leukocyte elastase at 47–48 [9,10]. Could an N-linked oligosaccharide at 78 be repelling this sequence from -subunit 43–47, allowing the elastase to cleave the -subunit (as indicated in Figure 10.3)? Could the major charge on the C-terminal peptide of the -subunit (Figure 10.3, arrow C) repel this structure outward from the hCG core? These are all possibilities, but very subjective thoughts.
Figure 10.1 X-ray crystallography structure of hCG dimer as published by Elliott et al. [3]. The -subunit C-terminal peptide is an assumed random 3D projection, a semi-polymer of proline and serine.
Three-Dimensional Structure of hCG
121
Figure 10.2 X-ray crystallography structure of hCG dimer showing assumed -subunit C-terminal extension. Sites and potential effects of hCG-H (2–3 sialic acid charge units) shown as light gray ellipse. Region of nicking at 47–48 with charged Arg residue at 43 shown as a separate gray ellipse.
Figure 10.3 X-ray crystallography structure of hCG showing assumed -subunit C-terminal extension. Effects of hCG-H charged oligosaccharides sites on 3D folding shown as gray ellipses. Gray ellipses include region of nicking at 47–48 with charged Arg residue at 43. Lettered arrows show inferred effects of charged carbohydrate units on 3D peptide structure, indicating linkages that might not form with hCG-H and free -subunit. Arrow A indicates effect of hyperglycosylation at -Asn 13. B indicates effect of hyperglycosylation at -Asn 30. C shows possible effect of hyperglycosylation at the -subunit C-terminal peptide. D and E show possible effects of hyperglycosylation at -Asn 52. F shows effect at -Asn 78. G and H show possible effects of nicking on hCG-H.
122
Human Chorionic Gonadotropin (hCG)
10.2 Hyperglycosylated hCG We know that hCG-H dissociates more rapidly than regular hCG and this suggests a more open molecule (Chapter 9). We know that most hCG-H preparations are nicked by leukocyte elastase at 47–48 (Chapter 9), which also suggests a more open structure. We also know that antibodies against free -subunit bind more readily to hCG-H than to regular hCG [9,10]. We infer that that the double-sized sugar structures have a major effect on opening the – subunit structure. The large -subunit N-linked oligosaccharides at residues 13 and 30, for example, seemingly pull the base sequence away from the body of hCG -subunit (Figure 10.3, arrows A and B). Similarly, a large oligosaccharide at 53 probably projects this peptide sequence away from the core (Figure 10.3, arrow B); the associated -subunit loop, sequence 90–100 (Figure 10.3, arrow E), projects in a different direction. Together, these sequences seem to completely open the cystine knot structure at 9–57, 34–88, and 38–90, which could make binding to cystine knot growth factor receptors possible. As discussed later in this book (Chapter 13), we speculate that hCG-H might be a TGF antagonist. During both implantation of pregnancy and the invasion process of gestational trophoblastic disease, hCG-H appears to be involved in the blockage of TGF-modulated functions. It blocks apoptosis, which is normally controlled in trophoblast cells by TGF [11,12], and functions through promoting TGF-controlled metaloproteinases and collagenases [13]. We look at the effects that large carbohydrates might have on hCG structure (Figure 10.3). Both the large, charged C-terminal peptide and the -subunit sequence surrounding 78 oligosaccharides (Figure 10.3, arrow F) might be repelled from the body of the hCG molecule (Figure 10.3, arrow C). hCG-H is predominantly nicked [3]. With nicking, the initial sequence 40–47, charged by an Arg residue at 43, probably springs to the surface of the molecule, while the continuing sequence (48–55), which is hydrophobic, springs to the center of the molecule, potentially opening the cystine knot. Considering the extension of the C-terminal peptide and the -subunit 90–100 loop as indicated by arrows C and E, it is possible that the three slow-forming disulfide bridges on hCG 23–72, 93–100, and 96–100 might not form on hCG-H. This might explain the TGF antagonist-like biological activity of hCG-H.
10.3 Free -Subunit Free -subunit is a hyperglycosylated molecule produced by nontrophoblastic cancer cells [14]. It is assumed that free -subunit undergoes structural changes similar to those described for hCG-H. Like hCG-H, it seems to act through antagonism of the TGF receptor on cancer cells [15]. Interestingly, the structure is so different from regular hCG that it forms – dimers instead of – dimers (16). This leads to the speculation that the folding is completely different and possible with different linkages formed by slow-forming disulfide bridges 23–72, 93–100, and 96–100.
Three-Dimensional Structure of hCG
123
References [1] Lustbader JW, Birken S, Pileggi NF, Kolks MAG, Pollak S, Cuff ME, et al. Crystallization and characterization of human chorionic gonadotropin in chemically deglycosylated and enzymatically desialylated states. Biochem 1989;28:9239–43. [2] Lapthorn AJ, Harris DC, Littlejohn JW, Lustbacer JW, Canfield RE, Mackin KJ, et al. Crystal structure of human chorionic gonadotropin. Nature 1994;389:488–91. [3] Elliott MM, Kardana A, Lustbader JW, Cole LA. Carbohydrate and peptide structure of the - and -subunits of human chorionic gonadotropin from normal and aberrant pregnancy and choriocarcinoma. Endocrine 1997;7:15–32. [4] Lustbader JW, Wang C, Zhang X, Birken S, Wu H, Brown JM, et al. Human chorionic gonadotropin: progress in determining its tertiary structure. In: Glycoprotein hormones. New York: Springer-Verlag; 1994; pp 81–102. [5] Wu H, Lustbader JW, Liu Y, Canfield RE, Hendrickson WA. Structure of human chorionic gonadotropin at 2.6 Å resolution from MAD analysis of the selenomethionyl protein. Structure 1994;2:545–58. [6] Vitt UA, Hsu SY, Hsush AJW. Evolution and classification of cytosine kny-containing hormones and related extracellular signaling molecules. Mol Endicrinol 2001;15:681–94. [7] Mishra AK, Makal SD, Iyer KS. Disulfide bonds Cys9–Cys57, Cys34–Cys88 and Cys34–Cys90 of the -subunit of human chorionic gonadotropin are crucial for heterodimer formation with -subunit: experimental evidence for the conclusions from the crystal structure of hCG. Biochem Biophys Acta 2003;1645:49–55. [8] Rosa C, Amr S, Birken S, Wehmann R, Nisula B. Effects of desialylation on human chorionic gonadotropin on its metabolic clearance rate in humans. J Clin Endocrinol Metab 1984;59:1215–19. [9] Cole LA, Karadana A, Park S-Y, Braunstein GD. The deactivation of hCG by nicking and dissociation. J Clin Endocrinol Metab 1993;76:704–15. [10] Karadana A, Cole LA. Human chorionic gonadotropin -subunit nicking enzymes in pregnancy and cancer patient serum. J Clin Endocrinol Metab 1994;79:761–7. [11] Schuster N, Krieglstein K. Mechanisms of TGF--mediated apoptosis. Cell Tissue Res 2002;307:1–14. [12] Hamade AL, Nakabayashi K, Sato A, Kiyoshi K, Takamatsu Y, Laoag-Fernandez JB, et al. Transfection of antisense chorionic gonadotropin gene into choriocarcinoma cells suppresses the cell proliferation and induces apoptosis. J Clin Endocrinol Metab 2005;90:4873–9. [13] Knittel T, Mehde M, Kobold D, Saile B, Dinter C, Ramadori G. Expression patterns of matrix metalloproteinases and their inhibitors in parenchymal and non-parenchymal cells of rat liver regulation by TNF-alpha and TGF-beta1. J Hepatol 1999;30:48–60. [14] Valmu L, Alfthan H, Hotakainen K, Birken S, Stenman UH. Site-specific glycan analysis of human chorionic gonadotropin -subunit from malignancies and pregnancy by liquid chromatography–electrospray mass spectrometry. Glycobiology 2006;16:1207–18. [15] Butler SA, Ikram MS, Mathieu S, Iles RK. The increase in bladder carcinoma cell population induced by the free beta subunit of hCG is a result of an anti-apoptosis effect and not cell proliferation. Br J Cancer 2000;82:1553–6. [16] Butler SA, Laidler P, Porter JR, Kicman AT, Chard T, Cowan DA, Iles RK. The beta subunit of human chorionic gonadotropin exists as a homodimer. J Molec Endocrinol 1999;22:185–192.
11 Paradigm Shift on the Targets of hCG Actions C.V. Rao Herbert Wertheim College of Medicine, Florida International University, Miami, FL, USA
hCG is a heterodimeric glycoprotein hormone that belongs to the cystine-knot growth-factor family [1,2]. In addition, it has properties of cytokines and chemokines [3]. Its -subunit is encoded by a single gene and its -subunit is encoded by a gene cluster [1,3]. Although human placental syncytiotrophoblasts produce large quantities, various normal and cancerous tissues can secrete small amounts of hCG, which usually fall below the detection limits of conventional hCG assays [3]. hCG binds to a receptor that is shared with LH [1]. For a long time, this so-called hCG/LH receptor was thought to be present only in gonads [1, 3]. However, the data published from around the world in the past 20 years from various laboratories have demonstrated that many nongonadal reproductive and nonreproductive organs and cells also contain them [4–6]. The activation of these receptors results in functional consequences that vary with the tissue, cell type, and physiological state [4–6]. The receptors are encoded by a single copy gene [7,8]. The receptor binds hCG with somewhat higher affinity than LH and only dimeric hormones bind [9]. Isolated hCG subunits have occasionally been reported to bind, but this binding can usually be attributed either to contaminating dimer hCG or to homodimerization of -subunits, which mimics the conformation of native hormone [1,3]. There is no convincing evidence to support hCG binding to any other type of receptor. Some of the earlier studies, conducted before the paradigm shift, indicated that hCG receptors might be present in nongonadal tissues [6]. These were thought to be where hCG binds to human placenta and regulates its endocrine functions, and why there is lower uterine uptake after radioiodinated hCG is injected into superovulated rodents [6]. These findings suggested the receptor presence, but the possibility was not further investigated due to overwhelmingly accepted dogma at the time. The paradigm shift was further fueled by the new technologies that emerged in the 1980s and 1990s, which allowed receptor detection at DNA, mRNA, and protein levels [4–6]. Table 11.1 shows a list of hCG-receptor-positive female and male nongonadal reproductive and nonreproductive organs and cells [4–6]. These receptors were detected in various laboratories using multiple techniques, which minimized the possibility of artifacts [4–6]. The receptor levels are lower in nongonadal tissues than in gonads [4–6]. However, the total number of receptors, in a case like Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00011-6 © 2010 Elsevier Inc. All rights reserved.
128
Human Chorionic Gonadotropin (hCG)
Table 11.1 hCG-Receptor-Positive Nongonadal Reproductive and Nonreproductive Tissues and Cells. Fallopian tubes Uterus Cervix Oocyte/early embryo/blastocyst Placenta Fetal membranes Decidua Umbilical cord Brain Pineal gland Spinal cord Cavernours sinus–carotid rete vascular complex
Neural retina Breast Skin Bone Urinary bladder Adrenals-zona reticularis Blood vessels in target tissues Supressor and regulatory T-cells/monocytes/ macrophages/dentritic cells Prostate Epididymis Seminal vesicles Sperm
uterine tissue, far exceeds the number in the gonads [4–6]. Although it has not been determined for every tissue, the nongonadal receptor affinity appears to be similar to gonadal tissue [4–6]. Receptor detection by traditional ligand binding assay had been difficult because of the lower receptor numbers in nongonadal tissues. Many of the hCG-receptor-positive nongonadal tissues are of human origin [4–6]. However, as exemplified by uterus, the nongonadal receptors appear to be well conserved across species [4–6]. hCG receptors have also been identified in lower organisms [10]. It is likely that the molecular nature of these receptors is different from those present in vertebrates and they could have evolved to serve different functions in vertebrates. Thus, hCG receptors are probably very ancient molecules that were active during the evolutionary process. Lungs, liver, kidneys, spleen, and small and large intestines, which are hCG-receptornegative in the adult, are hCG-receptor-positive in the human fetus [11]. This finding suggests that hCG might play growth- and differentiation-promoting roles in fetal organs; however, once these actions are completed, hCG receptors might disappear. hCG is one of the earliest signals that a developing embryo can sense, and the levels of hCG rapidly increase when cells of the inner cell mass differentiate into tissues of all three germ layers [3]. Because of its unique properties [12], it should not be surprising that hCG might have a role in the differentiation of cells of the inner cell mass, along with the other known factors. Trophoectoderm, which develops into placenta, is likely to contain hCG receptors, a fact suggesting that the hCG regulation of trophoblast functions might go back to the earliest stages of placental formation [3]. The hCG actions in the mother vary with the tissue and stage of pregnancy [4–6]. hCG receptors are present in gametes and could be involved in their maturation and fertilization [4–6]. The receptors in the early embryo might promote its growth and development [4–6]. The receptors in blastocysts might promote
Paradigm Shift on the Targets of hCG Actions
129
implantation, independent of estradiol and progesterone [4–6]. The process of implantation involves bidirectional communication between the implanting blastocyst and the receptive endometrium, and hCG derived from both these sources might play a role in this communication [4–6,13–17]. The human fetus might produce its own hCG from the kidneys and liver [18,19]. The levels in fetal circulation, however, are much lower than in the mother, suggesting that placental hCG secretion is almost always directed toward the maternal circulation and is prevented from entering into fetal circulation [3,20]. There could be a good reason for this. For instance, if maternal hCG floods fetal circulation, especially at certain critical times, it could possibly interfere with growth and differentiation of fetal tissues. The hCG actions in the reproductive tract are dependent on the region, reproductive phase, and cell type. For example, hCG might regulate the secretion and motility of Fallopian tubes that is conducive to gamete maturation, fertilization, early embryonic growth and development, and timely transport of the embryo into the uterus for implantation [4–6]. hCG in the uterus promotes blastocyst implantation by increasing the blastocyst’s ability to attach and invade the endometrium and superficial myometrium. This increases blood supply to the uterus through a combination of new vessel formation and vasodilation of the existing vessels, inhibits myometrial activity, and prevents rejection of the newly implanted blastocysts by the maternal immune system [4–6,13–15,21–25]. The functions in the fetoplacental unit, which consists of placenta, fetal membranes, decidua, and umbilical cord, contribute to pregnancy maintenance, fetal well-being, and the successful initiation and progression of labor at the end of pregnancy [3–6]. In placenta, hCG regulates its own synthesis, promotes the differentiation of cytotrophoblasts into syncytiotrophoblasts, and promotes the invasion of extravillous trophoblasts. It also regulates the secretion of various cytokines and up-regulates indoleamine 2,3-dioxygenase, which catalyzes the breakdown of tryptophan, a potent activator of T-cells, thereby preventing T-cell attack on the fetal tissues [3–6,24,26–30]. All the changes that are now attributed to hCG are also induced by other regulatory agents, which raises the question of which comes first. This question has been answered for the differentiation of cytotrophoblasts. Although many other regulatory agents promote the differentiation, they are ineffective when the hCG actions are blocked, indicating that hCG actions are central to the formation of syncytiotrophoblasts from cytotrophoblasts [3]. hCG influences the eicosanoid metabolism in fetal membranes [4–6]. It is possible that this action contributes to the progressive weakening of fetal membranes during the progression of labor. Moreover, fetal membranes that extend into the cervical canal during active labor might contribute eicosanoids for cervical softening, which is a prerequisite for successful delivery. The presence of hCG receptors in umbilical cord cells and blood vessels indicates that hCG might play a role in maintaining plasticity of the cord and regulating blood flow [4–6]. If it does indeed perform such regulation, hCG could be an important determinant of nutrient delivery to the fetus and removal of metabolic waste from the fetus. Several cells of the immune system, such as suppressor and regulatory T-cells, dendritic cells, monocytes, and macrophages contain hCG receptors
130
Human Chorionic Gonadotropin (hCG)
[4–6,22,26,27,29]. The functions that hCG regulates in these cells include influencing cytokine secretion, chemoattraction, and up-regulating indoleamine dioxygenase [4,24,26–30]. These actions broadly fall into categories of establishing pregnancy, inducing apoptosis, engulfing the dead cells by activated macrophages along the path of trophoblast invasion in the uterus, and protecting the fetus from immunorejection. hCG receptors are present in several areas of the central nervous system (CNS), such as the hippocampus, hypothalamus, brainstem, and anterior pituitary [4–6]. Consistent with this broad receptor distribution, hCG might have diverse CNS functions. For example, nausea and vomiting during pregnancy could be hCG driven, because the brainstem, which contains the centers for nausea and vomiting reflexes, has no blood–brain barrier, allowing free in-and-out diffusion of hCG. Although this finding formalized the positive relationship between hCG levels and nausea and vomiting, several previous reports suggested this possibility. For example, the frequency and intensity of nausea and vomiting are higher in twin pregnancies and gestational trophoblastic diseases than in singleton pregnancies, and higher in the first trimester than in the second and third trimesters in singleton pregnancies. These findings, along with the facts that not every pregnant woman has these episodes and that these episodes can extend into the second and third trimesters, could mean that the threshold levels that trigger these episodes might vary from pregnant individual to pregnant individual. In preparation for lactation after the pregnancy, breast tissue grows and differentiates, and these changes can be induced in nonpregnant rodent models by administering hCG [4–6]. These actions do not seem to be mediated by increasing ovarian estradiol and/or progesterone secretions alone. Moreover, breast epithelial cells contain hCG receptors, and their activation results in numerous changes, including the nonreversible differentiation of proliferative- to secretory-type epithelial cells [4–6]. The secretorytype epithelial cells are relatively more resistant to carcinogenic transformation. As a result, women who complete a full-term pregnancy before 20 years of age have a significantly decreased risk of breast cancer in later life. This pregnancy-induced protection gradually decreases with increasing maternal age and disappears by about 36 years of age. There are also many other hCG-driven changes, including adrenal steroid secretion and urinary incontinence during pregnancy [4–6,31]. Table 11.2 summarizes the potential therapeutic possibilities of the newly discovered hCG actions. There is a clear rationale for these possibilities, and preclinical data exist for some of them [4–6]. hCG is inexpensive, nontoxic, and has a few or no physiologic side effects. Moreover, if hCG does not work in a given patient, for whatever reason, one can fall back on conventional treatments, which are generally expensive, toxic, have side effects, and are not well tolerated by all patients. Even if these conventional treatments must be used, perhaps hCG can be used in a combination therapy to reduce the dose of conventional therapeutic agents to decrease their expense and toxicity and to increase tolerability. If hCG is proven useful in novel therapeutic applications, then there is a possibility of developing orally active hCG analogs—analogs that survive in the circulation much longer than hCG itself—and nanoparticle technology that might deliver hCG for weeks at a time after a single injection.
Paradigm Shift on the Targets of hCG Actions
131
Table 11.2 Therapeutic Potential from the Newly Discovered hCG Actions. Increasing pregnancy rates in assisted reproductive technologies Decreasing miscarriages Preventing preterm births Breast cancer treatment and prevention Treatment of gynecologic infections and HIV/AIDS Treatment of certain autoimmune diseases, including rheumatoid arthritis Treatment of spinal cord injuries Treatment of urinary incontinence
11.1 Summary and Perspectives hCG should no longer be considered just a hormone with a limited role of maintaining corpus luteum secretion of progesterone during the first weeks of pregnancy. It is a multifaceted molecule with properties of cystine-knot growth factors and cytokines. It plays roles from very early pregnancy to the end of pregnancy, which stands to reason because hCG is present all through pregnancy. Logically, then, there should be no pregnancy in the absence of hCG. It is unethical to manipulate hCG levels during pregnancy. Simply by default, pregnancy complications under the conditions of low or high hCG levels have been ascribed to other hormones and regulatory agents. It is possible, however, that these hormones and regulatory agents might have a modifying role on the actions of hCG. Obviously, much further research is needed to investigate these possibilities.
References [1] Pierce JG, Parsons TF. Glycoprotein hormones: structure and function. Annu Rev Biochem 1981;50:465–95. [2] Lapthorn AJ, Harris DC, Littlejohn A, Lustbader JW, Canfield RE, Machin KJ, et al. Crystal structure of human chorionic gonadotropin. Nature 1994;369:455–61. [3] Lei ZM, Rao CV. Endocrinology of the trophoblast tissue. In: Becker KL, editor. Principles and practice of endocrinology and metabolism. Philadelphia, PA: Lippincott Williams & Wilkins; 2001. p. 1096–102. [4] Rao CV. Nongonadal actions of LH and hCG in reproductive biology and medicine. Sem Reprod Med 2001;19:1–119. [5] Rao CV. An overview of the past, present and future of nongonadal LH/hCG actions in reproductive biology and medicine. Sem Reprod Endocrinol 2001;19:7–17. [6] Rao CV, Lei ZM. The past, present and future of nongonadal LH/hCG actions in reproductive biology and medicine. Mol Cell Endocrinol 2007;269:2 8. [7] McFarland KC, Sprengel R, Phillips HS, Kohler M, Rosemblit N, Nikolics K, et al. Lutropin-choriogonadotropin receptor: an unusual member of the G-protein-coupled receptor family. Science 1989;245:494 9.
132
Human Chorionic Gonadotropin (hCG)
[8] Loosfelt H, Misrahi M, Atger M, Salesse R, Vu Hai-Luu Thi MT, Jolivet A, et al. Cloning and sequencing of porcine LH/hCG receptor cDNA: variants lacking transmembrane domain. Science 1989;245:525–8. [9] Rao CV. Differential properties of human chorionic gonadotropin and human luteinizing hormone binding to plasma membranes of bovine corpora lutea. Acta Endocrinol 1979;90:696–710. [10] Castellanos Sanchez VO, Gomez-Conde E, Rocha-Gracia RD, Pimentel A, Aluja AS, Hernandez-Jauregui P, et al. Chorionic gonadotropin hormone receptors on Taenia solium and Taenia crassiceps cysticerci in culture. J Parasitol 2009;5:1. [11] Abdallah MA, Lei ZM, Li X, Greenwold N, Nakajima ST, Jauniaux E, et al. Human fetal nongonadal tissues contain human chorionic gonadotropin/luteinizing hormone receptors. J Clin Endocrinol Metab 2004;89:952–6. [12] Baal N, Reisinger K, Jahr H, Bohle RM, Liang O, Munstedt K, et al. Expression of transcriptional factor oct-4 and other embryonic genes in CD133 positive cells from human umbilical cord blood. Throm Haemost 2004;92:767–75. [13] d’Hauterive SP, Berndt S, Tsampalas M, Charlet-Renard C, Dubois M, Bourgain C, et al. Dialogues between blastocyst hCG and endometrial LH/hCG receptor: which role in implantation? Gynecol Obstet Invest 2007;64:156–60. [14] Fluhr H, Bischof-Islami D, Krenzer S, Licht P, Bischof P, Zygmunt M. Human chorionic gonadotropin stimulates matrix metalloproteinases-2 and -9 in cytotrophoblastic cells and decreases tissue inhibitor of metalloproteinases-1, -2, and -3 in decidualized endometrial stromal cells. Fertil Steril 2008;90:1390–5. [15] Fluhr H, Carli S, Deperschmidt M, Wallwiener D, Zygmunt M, Licht P. Differential effects of human chorionic gonadotropin and decidualization on insulin-like growth factors-I and -II in human endometrial stromal cells. Fertil Steril 2008;90:1384–9. [16] Zimmerman G, Baier D, Majer J, Alexander H. Expression of beta hCG and alpha CG mRNA and hCG hormone in human decidual tissue in patients during tubal pregnancy. Mol Hum Reprod 2003;9:81–9. [17] Zimmermann G, Ackermann W, Alexander H. Epithelial chorionic gonadotropin is expressed and produced in human secretory endometrium during normal menstrual cycle. Biol Reprod 2009;80:1053–65. [18] McGregor WG, Raymoure WJ, Kuhn RW, Jaffe RB. Fetal tissues can synthesize a placental hormone: evidence for chorionic gonadotropin -subunit synthesis by human fetal kidney. J Clin Invest 1981;68:306–9. [19] Goldsmith PC, McGregor WG, Raymoure WJ, Kuhn RW, Jaffe RB. Cellular localization of chorionic gonadotropin in human fetal kidney and liver. J Clin Endocrinol Metab 1983;57:654–61. [20] Katabuchi H, Ohba T. Human chorionic villous macrophages as a fetal biological shield from maternal chorionic gonadotropin. Dev Growth Differ 2008;50:299–306. [21] Zygmunt M, Herr F, Keller-Schoenwetter S, Kunzi-Rapp K, Munsteadt K, Rao CV, et al. Characterization of human chorionic gonadotropin as a novel angiogenic factor. J Clin Endocrinol Metab 2002;87:5290–6. [22] Filicori M, Fazleabas AT, Huhtaniemi IT, Licht P, Rao ChV, Tesarik K, et al. Novel concepts of human chorionic gonadotropin reproductive system interactions and potential in the management of infertility. Fertil Steril 2005;84:275–84. [23] Berndt S, d’Hauterive SP, Blacher S, Pequeux C, Lorquet S, Munaut C, et al. Angiogenic activity of human chorionic gonadotropin through LH receptor activation on endothelial and epithelial cells of the endometrium. FASB J 2006;20:2189–98.
Paradigm Shift on the Targets of hCG Actions
133
[24] Lei ZM, Yang M, Li X, Takikawa O, Rao CV. Upregulation of placental indoleamine 2,3dioxygenase by human chorionic gonadotropin. Biol Reprod 2007;76:639–44. [25] Berndt S, Blacher S, d’Hauterive SP, Thiry M, Tsampalas M, Cruz A, et al. Chorionic gonadotropin stimulation of angiogenesis and pericyte recruitment. J Clin Endocrinol Metab 2009;94:4567–74. [26] Wan H, Versnel MA, Leijten LME, van Helden-Meeuwsen CG, Fekkes D, Leenen PJM, et al. Chorionic gonadotropin induces dendritic cells to express a tolerogenic phenotype. J Leukocyte Biol 2008;83:894–901. [27] Wan H, Versnel MA, Cheung WY, Leenen PJM, Khan NA, Benner R. Chorionic gonadotropin can enhance innate immunity by stimulating macrophage function. J Leukocyte Biol 2007;82:1–9. [28] Kayisli UA, Selam B, Guzeloglu-Kayisli O, Demir R, Arici A. Human chorionic gonadotropin contributes to maternal immunotolerance and endometrial apoptosis by regulating Fas-Fas ligand system. J Immunol 2003;171:2305–13. [29] Schumacher A, Brachwitz N, Sohr S, Engeland K, Langwisch S, Dolaptchieva M, et al. Human chorionic gonadotropin attracts regulatory T cells into the fetal–maternal interface during early human pregnancy. J Immunol 2009;182:5488–97. [30] Mightmo JL, Perez AP, Sanchez-Margalet V, Duenas JL, Calvo JC, Varone CL. Upregulation of placental leptin by human chorionic gonadotropin. Endocrinol 2009; 150:304–13. [31] Rizk DEE, Osman NA, Shafiullah MM, Nagelkerke NJD, Fahim MA. Effect of human chorionic gonadotropin on in vitro contractions of stimulated detrusor muscle strips of female rats. J Obstet Gynaecol Res 2009;35:835–41.
12 The hCG Receptor Laurence A. Cole1 and Stephen A. Butler2 1
USA hCG Reference Service, Albuquerque, NM, USA Biomedical Sciences, Middlesex University, London, UK
2
hCG binds a common receptor with luteinizing hormone (LH). This receptor is located on ovarian corpus luteal cells for promotion of progesterone production, on myometrial blood vessels for promotion of angiogenesis, on cytotrophoblast and syncytiotrophoblast cells for enhancement of trophoblast cell differentiation, on myometrial muscle cells for suppression of contractions, on trophoblast cells for immuno-suppression and macrophage-suppression for foreign invading cells, on multiple fetal cell tissues for enhancement of growth, and on other tissues as described in Chapter 11. The hCG/ LH receptor, as it has become known, binds both regular hCG and hyperglycosylated hCG (hCG-H) [1]. Although the receptor is not activated by hCG free -subunit or free -subunit, there is clear evidence showing that the -subunit has a role in receptor binding and that -subunit has a function in receptor specificity [2,3]. The human hCG/LH receptor is a protein composed of 675 amino acids [4,5]. The rat receptor is similar (90% homology), and is composed of 674 amino acids [6]. The hCG/LH receptor is coded for by 70,000 bases in length on human chromosome 2p21, with 11 exons and 10 introns [7–9] as a single transcript; exons 1–10 and a portion of exon 11 encode the extracellular domain [10]. When hTSH, hFSH, and hLH receptors are compared, it is noted that, within the primary sequence, there is conservation of nine cysteines that appear in their extracellular domains, and nine that appear in their transmembrane/cytoplasmic domains. This suggests a common folding pattern through the disulfide pairing that exists amongst all glycoprotein hormone receptors [11]. LH/hCG receptor cloning studies indicate that this cell surface protein belongs to a large family of guanine nucleotide binding protein (G-protein) coupled membrane receptors [12,13]. It is an asymmetrical glycoprotein consisting of one polypeptide chain and six candidate N-linked glycoprotein sites. The large, 340-amino-acid, leucinerich extracellular hydrophilic N-terminus is distinct among this class of receptor, and is responsible for high-affinity hCG and LH binding and for binding specificity [10,14,15]. The domain is believed to have evolved in this specific way in order to ensure high-affinity binding to both hCG and LH [16]. The transmembrane domain has seven -helical membrane-spanning segments linked by three extracellular loops (exoloops) and three intracellular loops (cytoloops) (Figure 12.1). These transmembrane helix structures play an important role in receptor activity. Transmembrane helices 3, 6, and 7 are critical to receptor activation [17,18]. Together with the cytoloops, the intracellular hydrophobic C-terminus shares numerous motifs with Ser, Thr, and Tyr residues. This suggests a potential for modulation of receptor function Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00012-8 © 2010 Elsevier Inc. All rights reserved.
136
Human Chorionic Gonadotropin (hCG)
Figure 12.1 Diagrammatic representation of the hCH/LH receptor.
by phosphorylation, serine-threonine protein kinases, and tyrosine kinases [19]. The receptor is palmitoylated on two Cys residues at the C-terminal tail [20,21], and this region also undergoes phosphorylation on 4 Ser residues [7,22]. The role of the phosphorylation is uncertain, although one group has shown that the receptor dissociates from adenylate cyclase following phosphorylation [23]. Only the intact heterodimer of hCG or LH can stimulate the LH/hCG receptor; however, it has been demonstrated that the free subunits will also bind, albeit with a much lower affinity [24]. As mentioned, studies have shown that upon receptor binding, the -subunit of hCG is internalized and is associated with stimulation of adenylate cyclase activity. This implies that the binding model is not simple, and that hormone/receptor binding involves a topographical determinant extending across both ligand subunits [25]. The 40% structural homology between -subunits of the glycoprotein hormones [24] is thought to relate to the common property of hetero dimer formation. Therefore, the heterologous regions are likely to relate to the specificity of hormone receptor binding conferred by the -subunit [26]. Receptor binding and signaling appear largely concomitant, with both stages being dependent on the same regions of both the hormone and the receptor molecules. Receptor stimulation also appears to involve multiple steps, where signal transduction occurs as a result of a low-affinity site interaction following the initial highaffinity binding of the ligand. Amino acid Asp 397 of the receptor has been implicated in signal transduction, but not necessarily binding; it could, therefore, be involved in the low-affinity coupling [27]. Asp 383 is found within the second transmembrane domain and has also been implicated in ligand binding and cAMP second-messenger responses [28]. The hormonal glycosylations were initially thought to be essential for signal transduction [29]; however, although it is agreed that the sugar moieties affect ligand
The hCG Receptor
137
Figure 12.2 Activation of hCG/LH receptor, G-protein and cAMP, protein kinase expression, and production of LHRBP. Synthesis of LHRBP activates exo- and endonucleases that destroy receptor mRNA, limiting expression and down-regulating the receptor.
binding, more recent studies have shown that deglycosylated hCG can still stimulate the LH/hCG receptor [30]. Similarly, hCG-H has been shown to bind the receptor as well, albeit with a diminished activity [1]. Both of these observations indicate that optimal receptor stimulation is dependent on very specific glycosylation of hCG. Upon stimulation by hCG (or, indeed, LH), activation of the heterotrimeric (-, and -subunits) Gs-protein [19] results in stimulation of the membrane-bound adenylate cyclase by the -subunit. The sixth transmembrane segment and/or cytoloop-3 are believed to be responsible for Gs–protein interaction [31,32]. Activation of adenylate cyclase catalyses the conversion of ATP to cAMP, thus elevating intracellular levels. Following up-regulation of cAMP, activation of phosphokinase A (PKA) ensues, resulting in phosphorylation and activation of the CRE binding proteins (CREBPs) and subsequent cAMP-responsive elements (CREs) [5]. Although cAMP is considered the major second messenger involved in the LH/ hCG-R cascade, there is an additional stimulation of phospholipase C; hence elevated inositol phosphates, diacylglycerol, and intracellular calcium [33]. In both cases, the culmination of these events leads to enhanced steroid production by catalysis of the conversion of cholesterol to pregnenolone by 20-hydroxylase side-chain cleavage enzyme, and then leads to progesterone production by 3-hydroxysteroid dehydrogenase-isomerase via 17-hydroxyprogesterone [34]. It is thought that the activation of protein kinase activates mitogen protein kinase pathways and a Janus-kinase signaling pathway [17]. Clearly, all hormonal function involves DNA transcription or generation of mRNA. Promotion of progesterone production in corpus luteal cells involves the synthesis of cholesterol side-chain cleavage enzyme. Fetal tissue growth involves protein synthesis. A parallel mechanism promotes the synthesis of an LH/hCG receptor binding protein (LHRBP). This, in turn, activates exo- and endonuclease and leads to the destruction of receptor mRNA (Figure 12.2). This mechanism limits receptor expression, effectively down-regulating the receptor [18]. Transmembrane helix structures play an important role in receptor activity. Transmembrane helices 3, 6, and 7 are critical to receptor activation [19,20].
138
Human Chorionic Gonadotropin (hCG)
References [1] Cole LA, Kardana A, Andrade-Gordon P, Gawinowicz MA, Morris JC, Bergert ER, et al. The heterogeneity of hCG: III. The occurrence, biological and immunological activities of nicked hCG. Endocrinology 1991;129:1559–67. [2] Rao CV. Differential properties of human chorionic gonadotropin and human luteinizing hormone binding to plasma membranes of bovine corpora lutea. Acta Endocrinol 1979;90:696–710. [3] Dufau ML, Kusuda S. Purification and characterization of ovarian LH/hCG and prolactin receptors. J Recept Res 1987;7:167–93. [4] Jia XC, Oikawa M, Bo M, Tanaka T, Ny T, Boime I, et al. Expression of human luteinizing hormone (LH) receptor: interaction with LH and chorionic gonadotropin from human, but not equine, rat, and ovine species. Mol Endocrinol 1991;5:759–68. [5] Segaloff DL, Ascoli M. The lutropin/choriogonadotropin receptor—4 years later. Endocr Rev 1993;14:324–47. [6] McFarland RC, Sprengel R, Phillips HS, Kohler M, Resemblit N, Nikolics R, et al. Lutropin-choriogonadotropin receptor: an unusual member of the G protein-coupled receptor family. Science 1989;245:494–9. [7] Ascoli M, Fanelli F, Segaloff DL. The lutropin/choriogonadotropin receptor, a 2002 perspective. Endocr Rev 2002;23:141–74. [8] Fanelli F, Puett D. Structural aspects of luteinizing hormone receptor. Endocrine 2002; 18:285–93. [9] Fanelli F, Themman APN, Puett D. Lutropin receptor function: insights from natural, engineered, and computer-simulated mutations. Intl Union Biochem Mol Bio Life 2001; 51:149–55. [10] Puett D, Li Y, Agelova K, DeMars G, Meehan TP, Ganelli F, et al. Structure–function relationships of the luteinizing hormone receptor. Ann N Y Acad Sci 2005;1061:41–54. [11] Dias JA. Recent progress in structure–function and molecular analyses of the pituitary/ placental glycoprotein hormone receptors. Biophys Acta 1982;1135:278–94. [12] McFarland KC, Sprengel R, Phillips HS, Kohler M, Rosemblit N, Nikolies K, et al. Lutropin-choriogonadotrophin receptor: an unusual member of the G-protein coupled receptor family. Science 1989;245:494–9. [13] Loosfelt H, Salesse R, Jallal B, Garnier J, Milgrom E. Cloning and sequencing of porcine LH-hCG receptor cDNA: variants lacking transmembrane domain. Science 1989;245:525–8. [14] Fredriksson R, Lagerstrom MC, Lundin LG, Schioth HB. The G-protein-coupled receptors in the human genome form five main families. Phylogenetic analysis, paralogon groups, and fingerprints. Mol Pharmacol 2003;63:1256–72. [15] Puett D, Li Y, DeMars G, Angelova K, Fanelli F. A functional transmembrane complex: the luteinizing hormone receptor with bound ligand and G-protein. Mol Cell Endocrinol 2007;2:126–36. [16] Combarnous Y. Molecular basis of the specificity of binding of glycoprotein hormones to their receptors. Endocr Rev 1992;13:671–91. [17] Angelova K, Fanelli F, Puett D. A model for constitutive lutropin receptor activation based on molecular simulation and engineered mutations in transmembrane helices 6 and 7. J Biol Chem 2002;277:32202–13. [18] Angelova K, Narayan P, Simon JP, Puett D. Functional role of transmembrane helix 7 in the activation of the heptahelical lutropin receptor. Mol Endocrinol 2000;14:459–71.
The hCG Receptor
139
[19] Davis JS. Mechanisms of hormone action: luteinizing hormone receptors and secondmessenger pathways. Curr Opin Obstet Gynaecol 1994;6:254–61. [20] Zhu H, Wang H, Ascoli M. The lutropin/choriogonadotropin receptor is palmitoylated at intracellular cysteine residues. Mol Endocrinol 1995;9:141–50. [21] Kewate N, Menon KMJ. Palmitoylation of luteinizing hormone/human chorionic gonadotropin receptors in transfected cells. Abolition of palmitoylation by mutation of Cys621 and Cys-622 residues in the cytoplasmic tail increases ligand-induced internalization of the receptor. J Biol Chem 1994;269:30651–8. [22] Hipkin RW, Wang Z, Ascoli M. Human chorionic gonadotropin- and phorbol ester stimulated phosphorylation of the luteinizing hormone/CG receptor maps to serines 635, 639, 649 and 652 in the C-terminal cytoplasmic tail. Mol Endocrinol 1995;9:151–8. [23] Wang Z, Liu X, Ascoli M. Phosphorylation of the lutropin/choriogonadotropin receptor facilitates uncoupling of the receptor from adenylyl cyclase and endocytosis of the bound hormone. Mol Endocrinol 1997;11:183–92. [24] Pierce JG, Parsons TF. Glycoprotein hormones: structure and function. Ann Rev Biochem 1981;50:465–95. [25] Ryan RJ, Keutmann HT, Charlesworth MC, McCormick DJ, Milius RP, Calvo FO, et al. Structure–function relationships of gonadotropins. Recent Prog Horm Res 1987; 43:383–417. [26] Sairam MR, Manjunath P. Hormonal antagonistic properties of chemically deglycosylated hCG. J Biol Chem 1983;258:445–9. [27] Ji I, Ji TH. Receptor activation is distinct from hormone binding in intact lutropinchoriogonadotropin receptors and Asp397 is important for receptor activation. J Biol Chem 1993;268:20851–4. [28] Ji I, Ji TH. Asp383 in the second transmembrane domain of the lutropin receptor is important for high-affinity hormone binding and cAMP production. J Biol Chem 1991; 266:14953–7. [29] Keene JL, Matzuk MM, Boime I. Expression of recombinant human chorionic gonadotropin in chinese hamster ovary glycosylation mutants. Mol Endocrinol 1989;3:2011–17. [30] van Loenen HJ, van Gelderen-Boele S, Flinterman JF, Merz WE, Rommerts FFG. The relative importance of the oligosaccharide units in hCG for LH/CG receptor activation in rat leydig cells and mouse leydig cells. J Endocrinol 1995;147:367–75. [31] Lefkowitz RJ, Cotecchia S, Samama P, Costa T. Constitutive activity of receptors coupled to guanine nucleotide regulatory proteins. Trends Pharmacol Sci 1993;14:303–7. [32] Abell AN, McCormick DJ, Segaloff DL. Certain activating mutations within helix 6 of the human luteinizing hormone receptor might be explained by alterations that allow transmembrane regions to activate Gs. Mol Endocrinol 1998;12:1857–69. [33] Iles RK, Chard T. Molecular insights into structure and function of human chorionic gonadotropin. J Mol Endocrinol 1993;10:217–34. [34] Marton I, Oakey RE. 3 Beta-hydroxysteroid dehydrogenase-isomerase activity in placentae from pregnancies complicated by steroid sulphatase deficiency. J Steroid Biochem 1980;13:475–9.
13 Biological Function of
Hyperglycosylated hCG Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
Hyperglycosylated hCG (hCG-H) is a form of hCG with large oligosaccharides constituting as much as 35–41% of the molecular weight of the molecule (Chapter 6). As shown in various papers by Cole et al., hCG-H, rather than hCG, promotes growth and invasion of the uterus during pregnancy implantation. Cytotrophoblast cells also promote growth, invasion, and malignancy throughout the body during choriocarcinoma [1–3]. hCG-H produced by either choriocarcinoma or normal placenta cytotrophoblast cells (Table 13.1) promotes invasion of Matrigel membranes (basement membrane on gel) [1,2]. hCG-H promotes choriocarcinoma growth in pregnancy and testicular germ cancer cells [3]. It also promotes JEG-3 choriocarcinoma tissue growth and malignancy in nude mice. Antibodies specific to hCG-H completely stop the growth and spread of choriocarcinoma in a nude mouse, blocking all disease. Thus, it is inferred that hCG-H is essential for choriocarcinoma escalation and malignancy. In this chapter, we carefully examine the biochemistry of trophoblast cell invasion and implantation and consider the roles of hCG-H in these processes. We logically connect these findings and propose a model for hCG-H-regulated invasion and implantation (Figure 13.3). Handshuh et al. [4] have confirmed some these invasive roles for hCG-H. They repeated our Matrigel invasion experiment using first-trimester placental tissue, and confirmed that hCG-H drives invasion. Studies conducted by Hamade et al. [5] and by Lei et al. [6] have confirmed that the form of hCG produced by choriocarcinoma cells (hCG-H [1–3]) drives invasion. They showed blockage of cancer cell growth and invasion in cells. Lei and Hamade repeated the Matrigel and nude mice experiments described earlier using a JAR line of choriocarcinoma tissue. Because they could not obtain an hCG-H assay, they called the molecule choriocarcinoma hCG. Their tests confirmed the role of choriocarcinoma hCG or hCG-H in cell growth and invasion. Just as we blocked hCG-H with a specific antibody, they blocked production using antisense DNA for the -subunit [5] and for the -subunit [6]. As reflected in these publications, the JEG-3 and JAR lines of choriocarcinoma cells produce only hCG-H and free -subunit [7,8]. As shown by Kovalevskaya et al. [9] and by Handshuh et al. [4], hCG is made in villous syncytiotrophoblast cells, whereas hCG-H is made by extravillous cytotrophoblast cells. Figures 13.1 and 13.2 illustrate the functions of hCG and hCG-H at Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00013-X © 2010 Elsevier Inc. All rights reserved.
142
Human Chorionic Gonadotropin (hCG)
Table 13.1 Action of hCG-H on Cell Invasion of Matrigel Membranes [3]. % Penetration of Matrigel Membranes (mean SD) Pregnancy Cytotrophoblast Cells Control cultures hCG-H, 10 ng/ml Regular hCG, 10 ng/ml
40 10% 66 13%a 34 9%b
JEG-3 Choriocarcinoma Cells Control cultures hCG-H, 10 ng/ml Regular hCG, 10 ng/ml
48 11% 88 6%a 38 3%b
Cytotrophoblast cells were prepared from term placenta. Pregnancy cytotrophoblast cells and JEG-3 choriocarcinoma cells (5000 total) were cultured for 24 h on Matrigel basement membranes and control inserts in triplicate. Concentrations of hCG and hCG-H used to promote invasion were 4 times the concentration (ng/ml per 1000 cells) normally produced by the cells: 10 ng/ml for term pregnancy cytotrophoblast and 100 ng/ml for JEG-3 choriocarcinoma cells. The lower side of Matrigel basement membranes containing penetrated or invaded cells was stained and counted. Cell penetration was compared with that of control inserts. The percentage penetration or invasion was calculated using the formula described by the manufacturer. a A significant difference was observed by t-test in % penetration between control cultures and those with added hCG-H in both cell sources P 0.05 (cytotrophoblast cells) and P 0.005 (JEG-3 choriocarcinoma cells). b No significant difference was observed by t-test in % penetration between control cultures and those with added regular hCG in either cell source (P > 0.05).
Figure 13.1 Invasive blastocyst at 4 weeks gestation, illustrating implantation of pregnancy.
Biological Function of Hyperglycosylated hCG
143
Figure 13.2 Invasive trophoblast villous structure at 6 weeks’ gestation.
4 and 6 weeks gestation. Regular hCG promotes progesterone production and angiogenesis in decidual and myometrial spiral arteries. hCG-H is known to be the signal of invasion for extravillous cytotrophoblast cells [1–6]. As such, hCG-H is an autocrine rather than an endocrine or hormone. As shown by O’Connor et al. [10], hCG-H is not only the principal form of hCG produced in choriocarcinoma, but is also the principal form of hCG produced in early pregnancy, as part of implantation. In a study conducted by Sasaki et al. [11], 56 pregnancies were tested for hCG and hCG-H on the day of implantation—the earliest point of gestation. Sasaki found that successful pregnancies occur only when hCG-H accounts for more than 50% of the total hCG on the day of implantation. Pregnancies that contained less than 50% hCG-H were found to be biochemical pregnancies or first-trimester miscarriages. It was concluded that hCG-H is critical for pregnancy implantation. At Yale University, the team of Cole et al. discovered that hCG-H is an independent form of hCG. At Columbia University, Birken, Krichevsky, and Canfield generated antibody B152 and assays specific to hCG-H using Cole’s samples. In order to generate a Down syndrome screening assay, Yale and Columbia exclusively licensed these discoveries. As a result, research laboratories around the world have a very limited supply of hCG-H, hCG-H antibodies, and hCG-H assays.
144
Human Chorionic Gonadotropin (hCG)
Implantation of pregnancy is driven by proteases, both metalloproteinases and collagenases [12–19]. These degrade adjacent tissue, invading the uterine space needed by the trophoblast villous structures (Figures 13.1 and 13.2). The same enzymes promote invasion and malignancy in choriocarcinoma [19–24]. TGF receptors are found on both choriocarcinoma and extravillous cytotrophoblast cells [25–37]. TGF receptors “turn on” apoptosis and “turn off ” collagenase and metalloproteinase production. During implantation and in choriocarcinoma, the TGF receptor is antagonized or turned off. As a result, apoptosis is turned off in pregnancy, whereas collagenases and metalloproteinases are very much turned on [25–37]. Interestingly, hCG-H fits in this picture. It promotes invasion/invasion enzymes and specifically inhibits apoptosis [6]. Once we piece these findings together, they strongly suggests that hCG-H, a promoter of invasion and blocker of apoptosis [1–6], is possibly the antagonist of the TGF receptor. The three-dimensional structure of hCG has been determined by X-ray crystallography [38]. TGF and hCG share a common evolutionary origin and interrelated sequences [39–41]. As shown on the -subunit, buried in the heart of the hCG molecule is an extremely rare cystine-knot structure, which is formed by the overlap and linkage of four short peptide sequences [38]. The cystine-knot structure is found in hCG, TGF, and other cytokines. Thus, it seems logical for hCG-H to be a TGF antagonist. The large oligosaccharides on hCG-H seemingly hold the molecule open, preventing tight folding. As demonstrated, hCG-H is recognized by certain antibodies against the free -subunit of hCG, possibly because the opening of the molecule is caused by hyperglycosylation [42]. This might expose the cystine knot and permit it to be an antagonist at the TGF receptor. Thus, apoptosis is inhibited [6] and TGF-blocked production of degradative proteases seems to be permitted [25–37]. It appears to be more than a coincidence that this molecule regulates a process that works through TGF antagonism. The studies of Khoo et al. suggest that a molecule the size of hCG-H binds the TGF receptor in choriocarcinoma, promoting invasion [43]. Beyond this, further confirmation of hCG-H action through the TGF receptor is required. Interestingly, the free -subunit of hCG produced in nontrophoblastic cancers is very similar to hCG-H. It is hyperglycosylated [4], promotes invasion by nontrophoblastic cancers [44–46], and blocks apoptosis like hCG-H [44,47,48]. As shown by the published literature [44,48], free -subunit made by bladder cancer acts by antagonizing the TGF receptor on bladder cancer cells. It is strongly inferred that hCG-H and free -subunit are analogous molecules; that hCG-H is like a free -subunit with an exposed cystine knot made by cytotrophoblast cells. It is also strongly inferred that the hCG-H-like free -subunit antagonizes the TGF receptor on cytotrophoblast cells. Figure 13.3 is a possible model of hCG-H action. According to this model, hCG-H antagonizes TGF, which in turn blocks apoptosis. The antagonism leads to collagenase and metalloproteinase production, as well as uterine invasion by the extravillous cytotrophoblast cell. These same cells produce hCG-H, which then enters the circulation and returns to antagonize these cells in an autocrine manner. This model looks at blastocyst invasion (Figure 13.1) and villous invasion (Figure 13.2) and envisages hCG-H in action. A recent article by Guibourdenche et al. [49] confirms the production of hCG-H by extravillous cytotrophoblast cells and its role as the signal for invasion during the course of pregnancy.
Biological Function of Hyperglycosylated hCG
145
Figure 13.3 Proposed mechanism of hCG-H action. hCG-H antagonizes TGF receptor, leading to blockage of apoptosis and production of invasive metalloproteinases and collagenases.
Figure 13.4 The X-ray crystal structure of hCG [38].
Figure 13.4 shows the X-ray crystallographic structure of hCG. The cystine-knot structure within the -subunit of hCG involves the linkages 9–57, 34–88, and 38–90. It has always been hard to envision what effect the large oligosaccharides had on the molecule’s structure and how they could expose the TGF common structural elements.
146
Human Chorionic Gonadotropin (hCG)
In all, six papers and four independent groups across the world have used varying methods to confirm that hCG-H does in fact have functions different from those of regular hCG, promoting growth, invasion, and malignancy of pregnancy and choriocarcinoma cells, in vivo and in vitro. In conclusion, hCG-H acts on choriocarcinoma and implanting blastocyst/cytotrophoblast cells to promote growth and invasion. This process seemingly occurs as a result of antagonism of a TGF receptor. Both the fine biochemistry and the TGF-receptor interaction urgently require confirmation.
References [1] Cole LA, Dai D, Butler SA, Leslie KK, Kohorn EI. Gestational trophoblastic diseases: 1. Pathophysiology of hyperglycosylated hCG-regulated neoplasia. Gynecol Oncol 2006;102:144–9. [2] Cole LA, Khanlian SA, Riley JM, Butler SA. Hyperglycosylated hCG in gestational implantation and in choriocarcinoma and testicular germ cell malignancy tumorigenesis. J Reprod Med 2006;51:919–29. [3] Cole LA, Butler SA. Hyperglycosylated hCG and its free -subunit, tumor markers and tumor promoters: a review. J Reprod Med 2008;53:499–510. [4] Handshuh K, Guibourdenche J, Tsatsaris V, Guesnon M, Laurendeau I, Evain-Brion D, et al. Human chorionic gonadotropin produced by the invasive trophoblast but not the villous trophoblast promotes cell invasion and is down-regulated by peroxisome proliferator-activated receptor-a. Endocrinol 2007;148:5011–19. [5] Hamade AL, Nakabayashi K, Sato A, Kiyoshi K, Takamatsu Y, Laoag-Fernandez JB, et al. Transfection of antisense chorionic gonadotropin gene into choriocarcinoma cells suppresses the cell proliferation and induces apoptosis. J Clin Endocrinol Metab 2005;90:4873–9. [6] Lei ZM, Taylor DD, Gercel-Taylor C, Rao CV. Human chorionic gonadotropin promotes tumorigenesis of choriocarcinoma JAR cells. Troph Res 1999;13:147–59. [7] Cole LA, Butler SA, Khanlian SA, Giddings A, Muller CY, Seckl MJ, et al. Gestational trophoblastic diseases: 2. Hyperglycosylated hCG as a reliable marker of active neoplasia. Gynecol Oncol 2006;102:151–9. [8] Cole LA, Khanlian SA, Sutton JM, Davies S, Stephens N. Hyperglycosylated hCG (invasive trophoblast antigen, ITA): a key antigen for early pregnancy detection. Clin Biochem 2003;36:647–55. [9] Kovalevskaya G, Genbacev O, Fisher SJ, Caceres E, O’Connor JF. Trophoblast origin of hCG isoforms: cytotrophoblasts are the primary source of choriocarcinoma hCG. Mol Cell Endocrinol 2002;194:147–55. [10] O’Connor JF, Ellish N, Kakuma T, Schlatterer J, Kovalevskaya G. Differential urinary gonadotropin profiles in early pregnancy and early pregnancy loss. Prenat Diagn 1998;18:1232–40. [11] Sasaki Y, Ladner DG, Cole LA. Hyperglycosylated hCG: the source of pregnancy failures. Fertil Steril 2008;89:1781–6. [12] Rechtman MP, Zhang J, Salamonsen LA. Effect of inhibition of matrix metalloproteinases on endometrial decidualization and implantation in mated rats. J Reprod Fertil 1999;117:169–77. [13] Sharkey ME, Adler RR, Nieder L, Brenner CA. Matrix metalloproteinase expression during mouse peri-implantation development. Am J Reprod Immunol 1996;36:72–80.
Biological Function of Hyperglycosylated hCG
147
[14] Nardo LG, Nikas G, Makrigiannakis A. Molecules in blastocyst implantation: role of matrix metalloproteinases, cytokines and growth factors. J Reprod Med 2003;48:137–47. [15] Qinglei L, Hongmei W, Yunge Z, Haiyan L, Qingxiang SA, Zhu C. Identification and specific expression of matrix metalloproteinase-26 in rhesus monkey endometrium during early pregnancy. Mol Hum Reprod 2002;8:934–40. [16] Bischof P, Campana A. A model for implantation of the human blastocyst and early placentation. Hum Reprod Update 1996;2:262–70. [17] Librach CL, Werb Z, Fitzgerald ML, Chiu K, Corwin NW, Esteves RA, et al. 92-kD type IV collagenase mediates invasion of human cytotrophoblasts. J Cell Biol 1991; 113:437–49. [18] Norwitz ER, Schust DJ, Fisher SJ. Implantation and the survival of early pregnancy. N Engl J Med 2001;345:1400–8. [19] Lewis MP, Sullivan MH, Elder MG. Regulation by interleukin-1 beta of growth and collagenase production by choriocarcinoma cells. Placenta 1994;15:13–20. [20] Sekiya S, Oosaki T, Suzuki N, Takamizawa H. Invasion potential of human choriocarcinoma cell lines and the role of lytic enzymes. Gynecol Oncol 1985;22:324–33. [21] Chen Q, Stone P, McCowan L, Chamley L. Interaction of Jar choriocarcinoma cells with endothelial cell monolayers. Placenta 2005;26:617–25. [22] Maquoi E, Noël A, Foidart J. Matrix metalloproteinases in choriocarcinoma cell lines: a potential regulatory role of extracellular matrix components. Placenta 1997;18:123–42. [23] Okamoto T, Niu R, Yamada S, Osawa M. Reduced expression of tissue inhibitor of metalloproteinase (TIMP)-2 in gestational trophoblastic disease. Mol Human Reprod 2002;8:392–8. [24] Vegh GL, Tuncer ZS, Fulop V, Genest DR, Mok SC, Berkowitz RS. Matrix metalloproteinases and their inhibitors in gestational trophoblastic diseases and normal placenta. Gynecol Oncol 1999;75:248–53. [25] Schuster N, Krieglstein K. Mechanisms of TGF--mediated apoptosis. Cell Tissue Res 2002;307:1–14. [26] Kamijo T, Rajabi MR, Mizunuma H, Ibuki Y. Biochemical evidence for autocrine/paracrine regulation of apoptosis in cultured uterine epithelial cells during mouse embryo implantation in vitro. Mol Human Reprod 1998;4:990–8. [27] Pampferf S. Apoptosis in rodent peri-implantation embryos: differential susceptibility of inner cell mass and trophectoderm cell lineages—A review. Placenta 2000;21:S3–S10. [28] Shooner C, Caron PC, Fréchette-Frigon G, Leblanc V, Déry M-C, Asselin E. TGF-beta expression during rat pregnancy and activity on decidual cell survival. Reprod Biol Endocrinol 2005;3:20. [29] Liu Y-X, Gao F, Wei P, Chen X-L, Gao H-J, Zou R-Z, et al. Involvement of molecules related to angiogenesis, proteolysis and apoptosis in implantation in rhesus monkey and mouse. Contraception 2005;71:249–62. [30] Knittel T, Mehde M, Kobold D, Saile B, Dinter C, Ramadori G. Expression patterns of matrix metalloproteinases and their inhibitors in parenchymal and non-parenchymal cells of rat liver regulation by TNF-alpha and TGF-beta1. J Hepatol 1999;30:48–60. [31] Murphy G, Reynolds JJ, Whitham SE, Docherty AJ, Angel P, Heath JK. Transforming growth factor beta modulates the expression of collagenase and metalloproteinase inhibitor. EMBOJ 1987;6:1899–904. [32] Qureshi HY, Sylvester J, El Mabrouk M, Zafarullah M. TGF-beta-induced expression of tissue inhibitor of metalloproteinases-3 gene in chondrocytes is mediated by extracellular signal-regulated kinase pathway and Sp1 transcription factor. J Cell Physiol 2005;203:345–52.
148
Human Chorionic Gonadotropin (hCG)
[33] Stetler-Stevenson WG, Brown PD, Onisto M, Levy AT, Liotta LA. Tissue inhibitor of metalloproteinases-2 (TIMP-2) mRNA expression in tumor cell lines and human tumor tissues. J Biol Chem 1990;265:13933–88. [34] Pringle K, Roberts C. New light on early post-implantation pregnancy in the mouse: roles for insulin-like growth factor-II (IGF-II)? Placenta 2007;28:286–97. [35] Hwang JH, Koh SH, Han DI, Chung SR, Park MI, Hwang YY, et al. Expression of transforming growth factor-1, 2 in the decidua of the early pregnancy: comparison of decidua basalis and decidua parietalis. Korean J Obstet Gynecol 2001;44:1145–49. [36] Kingsley-Kallesen M, Johnson L, Scholtz B, Kelly D, Rizzino A. Transcriptional regulation of the TGF-beta 2 gene in choriocarcinoma cells and breast carcinoma cells: differential utilization of cis-regulatory elements. In vitro Cell Dev Bio Anim 1997;33:294–301. [37] Filla MS, Kaul KL. Relative expression of epidermal growth factor receptor in placental cytotrophoblasts and choriocarcinoma cell lines. Placenta 1997;18:17–27. [38] Lapthorn AJ, Harris DC, Littlejohn A, Lustbader JW, Canfield RE, Machin KJ, et al. Crystal structure of human chorionic gonadotropin. Nature 1994;369:455–61. [39] Laub M, Jennissen HP. Identification of the antihelix motif in the TGF- superfamily by molecular 3D-rapid prototyping. Materialwissenschaft und Werkstofftechnik 2003;34:1113–19. [40] Lehnert SA, Akhurst RA. Embryonic expression pattern of TGF beta type-1 RNA suggests both paracrine and autocrine mechanisms of action. Development 1988;104:263–73. [41] Wu H, Lustbader JW, Liu Y, Canfield RE, Hendrickson WA. Structure of human chorionic gonadotropin at 2.6 Å resolution from MAD analysis of the selenomethionyl protein. Structure 1994;2:545–8. [42] Isaacs NW. Cystine knots. Curr Opin Struct Biol 1995;5:391–5. [43] Khoo NK, Bechberger JF, Shepherd T, Bond SL, McCrae KR, Hamilton GS, et al. SV40 Tag transformation of the normal invasive trophoblast results in a premalignant phenotype I. Mechanisms responsible for hyperinvasiveness and resistance to anti-invasive action of TGF. Intl J Cancer 1998;77:429–39. [44] Butler SA, Ikram MS, Mathieu S, Iles RK. The increase in bladder carcinoma cell population induced by the free beta subunit of hCG is a result of an anti-apoptosis effect and not cell proliferation. Br J Cancer 2000;82:1553–6. [45] Delves PJ, Iles RK, Roitt IM, Lund T. Designing a new generation of anti-hCG vaccines for cancer therapy. Mol Cell Endocrinol 2007;260:276–81. [46] Carter WB, Sekharem M, Coppola D. Human chorionic gonadotropin induces apoptosis in breast cancer. Breast Cancer Res Treat 2006;100:S243–4. [47] Butler SA, Iles RK. Ectopic human chorionic gonadotropin secretion by epithelial tumors and human chorionic gonadotropin -induced apoptosis in Karposi’s sarcoma: is there a connection? Clin Cancer Res 2003;9:4666–73. [48] Iles RK. Ectopic hCG expression by epithelial cancer: malignant behavior metastasis and inhibition of tumor cell apoptosis. Mol Cell Endorcinol 2007;260:264–70. [49] Guibourdenche J, Handschuh K, Tstsaris V, Gerbaud P, Muller F, Evain Brion D, et al. Hyperglycosylated hCG is a marker of early human trophoblast invasion. J Clin Endocrinol Metab 2010; in press.
14 Biological Function of the Free -Subunit: Expression and Treatment Target in Cancer Stephen A. Butler and Ray K. Iles Biomedical Sciences, Middlesex University, London, UK
Reports of ectopic hCG molecules expressed in vivo by nongestational tumors were noted as early as 1904 [1]. A rare case of a bladder tumor expressing biologically active gonadotropin (called chorioepithelioma) contained syncytiotrophoblast elements and had widely metastasized. The tumor occurred in a postmenopausal woman whose ovaries were atrophic, despite a hyperplastic endometrium. Therefore, 23 years before the discovery of hCG by Aschheim and Zondeck [2], it was correctly concluded that these changes were due to a gonadotrophic hormone produced by the tumor. Ectopic production of biologically active hCG produced by non-germ-cell tumors was next reported in 1946 [3]. It is not uncommon for ectopic hCG production to be explained by dedifferentiation (trophoblastic differentiation), where it is assumed that the tissue has reverted to pluripotence, taking on the characteristics of the syncytiotrophoblast and thus expressing hCG. In almost all cases, however, the sole criterion for the trophoblastic differentiation claim is the detection of hCG; this detection is often a result of a misinterpreted hCG-positive assay. Because common epithelial tumors will express hCG [4], most of these claims are the result of false dogma. It is, in fact, quite distinct, as germ-cell tumors will express both hCG and hCG. Resulting in the production of the gonadotropic holo-hormone hCG, ectopic expression by common epithelial tumors consists almost exclusively of the free -subunit. Only rarely is the holo-hormone found in advanced-stage carcinomatosis [5], and it has been only sporadically noted in liver and lung cancers [6]. Thus, such de-differentiation is a much rarer event than has been claimed, and has more to do with confusion over assay specificity and hCG/hCG terminology than de-differentiation or carcinomatosis. Although the holo-hormone hCG is produced by placental and germ-cell tumors, the free -subunit (hCG) is produced by epithelial tumors and is (more often than not) independent of glycoprotein hormone alpha gene expression [7,8]. Ectopic production of free hCG by bladder carcinoma is well described, and the majority of our work has been concentrated in this field [9]; however, expression of hCG is not exclusive to bladder carcinoma. It has also been shown in cervical and endometrial carcinoma, as well as many other non-germ-cell tumors of the breast, colon, lung, ovary, oral/facial tissue, prostate, pancreas, vulva/vagina, kidney, and neuroendocrine tissue (Table 14.1). Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00014-1 © 2010 Elsevier Inc. All rights reserved.
150
Human Chorionic Gonadotropin (hCG)
Table 14.1 Summary of the Frequencies of hCG Expression by Non-Germ-Cell Epithelial Cancers Reported over the Past 18 Years. Publication Year
Tissue of Origin
Serum hCG
Urine hCG/cf
Immunohisto- Prognostic chemistry hCG
[10] 1998 [11] 1996 [12] 1996 [13] 1996 [14] 1996 [15] 1995 [16] 2005 [17] 1998 [18] 1997 [19] 1996 [20] 1992 [21] 2008 [22] 1998 [18] 1997 [23] 1995 [24] 2002 [25] 2000 [26] 1996 [27] 1995 [18] 1997 [23] 1995 [28] 2004 [29] 2002 [30] 2000 [31] 1999 [32] 1998 [33] 1997 [34] 1995 [35] 2008 [36] 1999 [37] 1999 [38] 2008 [39] 1997 [23] 1995 [40] 1996 [41] 2006 [42] 2004 [43] 2001 [44] 1998 [45] 1997 [18] 1997 [46] 1995
Bladder Bladder Bladder Bladder Bladder Bladder Breast Breast Breast Breast Breast Cervical Cervical Cervical Cervical Colorectal Colorectal Colorectal Colorectal Endometrial Endometrial Kidney Kidney Lung Lung Lung Lung Lung Neuroendocrine Oral/Facial Oral/Facial Ovarian Ovarian Ovarian Prostate Prostate Pancreas Pancreas Pancreas Vulval/Vaginal Vulval/Vaginal Vulval/Vaginal
– 76% (n 33) – 30% (n 76) – – – – – – – – – – 35% (n 40) 16% (n 204) 17% (n 232) – – – 30% (n 39) – 23% (n 177) – 14% (n 85) 22% (n 92) – – 12% (n 360) 20% (n 59) – 33% (n 173) 36% (n 73) 41% (n 27) – 4% (n 104) 50% (n 160) – 42% (n 36) 10% (n 118) – 38% (n 50)
– – 73% (n 237) – 35% (n 52) 60% (n 63) – – 19% (n 72) – – – 33% (n 46) 29% (n 49) – –
35% (n 104) – – – – – 13% (n 334) 49% (n 99) – 80% (n 32) 56% (n 27) 78% (n 89) 26% (n 46) – – – 22% (n 232) 37% (n 123) 54% (n 377) – – 0% (n 55) – 9% (n 90) 28% (n 85) – 64% (n 14) 93% (n 44) – – 64% (n 45) – – – 15% (n 80) – – 56% (n 107) – – – –
– – 32% (n 25) – – – – – – – – – – – – – – 10% (n 104) – – – – 17% (n 12) –
– – – Yes Yes – – No – – – – Yes – – Yes Yes Yes Yes – – No Yes – Yes – – – Yes Yes Yes Yes Yes – Yes No Yes Yes Yes Yes – Yes
A total of 42 studies revealed production of elevated levels of hCG or hCG -core fragment in serum, urine, and by immunohistochemistry. Of the 23 studies that investigated prognosis, 87% indicated that hCG expression was a good marker for poor prognosis. (–) indicates that this specific element was not studied in this case.
Biological Function of the Free -Subunit: Expression and Treatment Target in Cancer
151
14.1 hCG Gene Expression in Cancer In 1987, Hussa proposed several hypotheses for nontrophoblastic production of hCG. These included what he referred to as gene depression, blocked ontogeny, incomplete suppression of the hCG genome, natural DNA recombination, and vestigial autocrineisocrine secretion [47]. As discussed in Chapter 4, the control of hCG gene expression has not been completely elucidated; it is still unclear which CG genes are truly active in terms of the protein they ultimately translate and the effects that the protein produced might have, whether eutopically or ectopically expressed. Considering some of the hypotheses emerging in the field and alluded to or detailed in this book, it is tempting to suggest that the proteins arising from our CG paralogs may indeed have evolved new functions; however, the true hierarchy of expression within the CG gene cluster has yet to be established, as do the mechanisms that control them. What we do know is that the LH-CG gene cluster is not amplified or rearranged in any way, certainly not in bladder cancer [48]. These findings strongly indicate that the regulation of transcription and/or translation is likely to hold the key to ectopic hCG production. Following the complete cloning of the LH-CG gene cluster, two groups of hCG producers (termed types I and II) were discovered [49]. Type I refers to the gene product of CG7 [6], which appears to lead to the synthesis of an hCG protein with an alanine residue at position 117; this gene appears to be largely inactive in pregnancy [50,51]. Type II refers to the gene products of CG3 (CG9 or CG; see Chapter 4), CG5, and CG8; these encode an aspartic acid residue at position 117 [49]. The difference in which CG type was expressed was explained through malignant transformation, suggesting that normal urothelium expresses CG7 and bladder carcinoma expresses CG, CG5, and CG8 [49]. More recently, other studies on breast, lung, and renal carcinoma all indicate that CG7 is expressed at a comparatively low level [52–54]. From our own studies, however, it appears that CG7 is always expressed at a lower level, and that pseudogene status should perhaps be conferred on this paralog rather than genes CG1 and CG2. These genes are often overlooked because they had been largely regarded (incorrectly) as nonfunction pseudogenes. It is, of course, dangerous to start comparing choriocarcinoma and placental data with that derived from epithelial cancer studies, as their respective origins are so distinct; however, new data now appear to implicate CG1 products in the invasion process of trophoblast tissues [51]. Although this has been shown in first-trimester pregnancies only, it is tempting to start drawing parallels between placental implantation and tumor invasion.
14.2 hCG Expression in Epithelial Cancer Well-documented evidence for ectopic production of hCG/hCG by bladder tumors was not published until 1973 [55]. Subsequently, more case reports were reviewed in 1978 [56] and again in 1983 [57], whereupon ectopic hCG expression became a generally accepted phenomenon in cancer. Two years later, one article stated that the production of an hCG-like material by bladder carcinoma was “probably not rare”
152
Human Chorionic Gonadotropin (hCG)
[58]; however, it was not until in vitro studies in 1987 [59] and 1989 [60] that the free -subunit of hCG was identified as a potential bladder tumor marker; at this time, the hCG-like product was determined to be hCG and not hCG [7]. Bladder cancer is not unique in expressing hCG, it has merely been our focus. Over the past three decades, there have been many reports highlighting the ectopic production of hCG from non-germ-cell or nonplacental origins [57,59,61–70]; this has culminated in the latest wave of publications, linking hCG expression to prognosis (see Table 14.1). Since 1992, 42 significant studies (omitting case reports) on the detection of hCG in various epithelial cancers have been published (see Table 14.1). Each study investigated the expression of hCG either by immunohistochemistry, serum analysis, or urinalysis (of either hCG or hCG -core fragment). Positive detection ranges from 0% in renal cell cancer to 93% in small cell (SC) lung cancer. About 30% of epithelial cancer patients will have elevated serum levels and 37% will have elevated urinary levels; 48% of tumors will stain hCG positive. Twenty-three of the studies specifically investigated cancer prognosis, and 20 of these studies (87%) report a strong association between hCG detection and poor patient prognosis, which would correlate with metastatic spread. Other studies have investigated mRNA levels, such as in prostate cancer [71,72] and renal carcinoma [73], but many such studies have been compromised by high-level CG mRNA detection in normal tissues. There are problems associated with the use of PCR-based detection methods for CG gene expression. For example, expression does not seem to correlate well with detection of hCG protein and we are now of the opinion that the clinical association with poor prognosis can be seen as significant only when using serum or urine hCG assay as the detection methodology. In bladder cancer, the relative incidence of expression has been disputed in many studies [9]; however, from our own studies, only around 35% (see Table 14.1) of bladder cancer patients have elevations in immunoreactive hCG material (intact hCG, hCG, or hCG -core fragment). There is a good correlation of hCG expression by such tumors with grade and stage; metastatic cases often tend to be hCG positive [74], and in those cases, up to 75% of all serum and urine samples have elevated levels of immunoreactive hCG [75]. With a focus on early detection, however, hCG appears to be of little use diagnostically. According to Table 14.1, optimistic assessment could conclude that 93% of SC lung carcinomas [34], 80% of breast tumors [19], or 78% of cervical carcinomas [21] express hCG; however, the real picture is probably quite different, as frequencies from the original studies range from 0% to 93% (Table 14.1). Following our meta-analysis, we calculated the overall incidence of hCG expression in epithelial cancer to be 32% (Table 14.2), and as a rule (if one can apply rules), this appears to be a good generalization in most cases. We could conclude, therefore, that about a third of all epithelial cancers express hCG. This, however, would clearly be a gross generalization. We can see that in renal, prostate, vulval/vaginal, and neuroendocrine cases, the frequency is somewhat lower (20%); in bladder, cervical, and pancreatic carcinoma, the frequency is a little higher, at around 50% of cases. The remaining cancers studied to date (breast, colorectal, endometrial, lung, oral/facial, and ovarian) all appear to indicate a frequency of approximately 30% (see Table 14.2).
Biological Function of the Free -Subunit: Expression and Treatment Target in Cancer
153
Table 14.2 Simple Meta-Analysis of the Percentage Frequency of hCG Detection from Cases of Carcinoma Originating in 13 Different Epithelial Tissues. Epithelial Cancer Tissue of Origin Bladder Breast Cervical Colorectal Endometrial Kidney Lung Neuroendocrine Oral/Facial Ovarian Prostate Pancreas Vulval/Vaginal Overall for carcinoma
hCG Expression (Overall % Frequency) 55.1 29.5 46.2 32.2 31.8 11.6 27.8 12.0 33.3 34.6 9.2 51.2 18.2 31.9
Total Number of Samples (n)
Number of Studies
565 564 270 1168 62 232 410 360 104 273 288 303 180 4779
6 5 5 5 2 2 6 1 2 3 2 3 3 45
Original data and references can be seen in Table 14.1. In this table, frequency is weighted by n-values. Overall incidence is also shown for discussion purposes.
With a sensitivity of 50% at best, there is a question of clinical significance that has maintained interest in ectopic hCG expression: why would cancer produce a pregnancy hormone subunit? Only since the advent of specific immunoassays [76] has it been possible to identify free -subunit as the hCG-related antigen and distinguish it from the holo-hormone hCG and free hCG. Only the intact – heterodimer is an active gonadotropic hormone, and for some researchers this strengthens the argument that hCG production is not an epiphenomenon. The prognostic significance of hCG expression has not gone unnoticed and, as highlighted in Table 14.1, appears almost always to indicate poor prognosis. The exceptions here were prostate cancer, renal carcinoma, and breast cancer. Prostate and kidney cancers do not appear to be significantly associated with the expression of hCG at all. From our analysis, prostate and kidney cancers are indicated by the lowest expression frequencies of around 10% (Table 14.2). This might suggest that a different mechanism is at play, or it could be an artifact associated with the small number of studies with low n-values. Breast cancer and hCG expression, however, is a complex and well-investigated topic—and one that is often at odds with our own studies. Much of the conflicting evidence could simply come down to hCG/hCG confusion, as the molecule here is more likely to be exogenous hCG having a differentiation effect rather than ectopic hCG, which has an autocrine tumor-promoting effect. It is with this hypothesis in mind that we recently published a review on the topic [77]. It is not uncommon for authors to report very aggressive tumors where hCG has been detected, and find lower response rates to radiotherapy and chemotherapy in
154
Human Chorionic Gonadotropin (hCG)
100 90 80 % survival
70 60 50 40 30 20 10 0
0
2
4
6
8
10
12
14
16
18
Months posttreatment
Figure 14.1 Survival curves of hCG-positive and hCG-negative patients with T2-T4 bladder tumors. The x-axis indicates percentage survival of patients with normal (- - - -) and elevated (-------) hCG as a function of months posttreatment (the y-axis). Adapted from [9]; original data from [14].
patients where hCG is elevated. In our own studies, survival analysis indicated a very strong association between early death and hCG expression (Figure 14.1), suggesting an autocrine control mechanism which we then went on to explore further [78].
14.3 The Biological Action of hCG on Epithelial Tumors Historically, the expression of hCG by epithelial cancers has been dismissed as an epiphenomenon, with no oncogenic significance. Thus, hCG (and by association, hCG) expression is often solely regarded as a marker of germ cells, as mentioned earlier. In 1996, we published our first study reporting that bladder cancer cell numbers increased following incubation with hCG in a dose-dependent manner [78]. No effect could be seen following treatment with intact hCG, growth-promoting hormone alpha (GPH), or hCG -core fragment. Neither effect was observed when hCG was co-incubated with polyclonal anti-hCG antiserum. Furthermore, when this antiserum was added to bladder cancer cell cultures known to produce hCG, their growth was significantly inhibited in proportion to the quantity of hCG produced from each cell line (i.e., high hCG expressers were more adversely affected by the antisera than low to nonexpressors). This suggested autocrine stimulation [78]. Anti-hCG antibodies’ inhibitory effects on the growth of tumor cells has also been demonstrated in vivo in athymic (nude) mice [79] and in cell lines incubated with murine antisera from mice immunized with anti-hCG-CTP vaccines [80]. More recently, antisense oligonucleotides directed to the hCG -chain have also been shown to have antiproliferative effects on tumor cell lines in vitro [81,82].
Biological Function of the Free -Subunit: Expression and Treatment Target in Cancer
155
Our later studies showed that the increase in the cancer cell population in response to hCG was brought about by inhibition of apoptosis rather than by stimulation of cell replication [83], and that the recombinant hCG was at least as potent, if not more potent, than the CR129 preparation used before [78]. Furthermore, given that the LH/hCG receptor is not expressed by these tissues, and only intact hCG can stimulate the LH/hCG receptor (as described in Chapter 15), we have to assume that any activity observed must be mediated by an as-yet-unidentified pathway. The X-ray crystallographic structure of hCG was revealed [84,85] after we finalized and published our studies on the prognostic significance of hCG in bladder cancer [14] and the autocrine function of hCG on bladder cancer cells in vitro [78]. The most striking feature noted was the cystine knot, which we described in Chapter 10. This conserved structure was identified in a number of cytokines designated as the “cystineknot growth factor” (CKGF) family. This family includes transforming growth factor (TGF), platelet-derived growth factor B (PDGFB), and nerve growth factor (NGF). Since the identification of the structure of hCG in 1994, the family of growth factors containing a cystine-knot motif has expanded, and now includes the vascular endothelial growth factors (VEGFs) and placental growth factor (PlGF), among others. Each of the growth factors exists as a functional homodimer, and in 1999 we published our findings suggesting that hCG also formed naturally occurring homodimers [86]. We later indicated that the – homodimer of hCG was no more bioactive in our cell line model than the monomeric hCG [87]. We now suggest that the ability to homodimerize is indicative of structural variation within bioactive hCG rather than a requirement for ectopic bioactivity. The latter additions to the CKGF family, and the ability of hCG to homodimerize, further increased the intriguing possibilities that cross-talk may occur between both multiple growth regulatory systems and that ectopic expression of molecules containing a cystine knot may give rise to oncogenic molecules (whose disruption of these regulatory systems favor cancer development). Homology modeling (using current data on the structure of hCG as obtained from the published crystal structure) [84] was used in conjunction with data from the Protein Data Bank (www.pdb.org) and Clustalw2 (http://www.ebi.ac.uk/Tools/clustalw2/index. html). By mapping hCG to either TGF or VEGF crystal structures, it is possible to generate potentially naturally occurring molecular models of hCG– dimers (Figure 14.2), where each model of hCG dimerizing is energetically favorable. Homology model monomers alone have 51% and 69% alignment with TGF and VEGF, respectively, without any changes to amino acid sequences. If hCG folds in one of these conformations, we propose that the molecule is likely to interact with the respective CKGF receptor. Exploring these methods further, hCG can be demonstrated to be very likely to bind the VEGF receptor kinase insert domain receptor (KDR). This claim is substantiated by VEGF homology-modeled hCG having four of the eight KDR receptor binding hot spots (Ile 47, 53, 87; and Pro 90) in correct alignment for binding and stimulation of the receptor (unpublished data). This is two more than is required by VEGF-C to bind and stimulate the same receptor [88]. Similarly, amino acid regions in the receptor binding loop of TGF 3, which are essential for binding TGF-RII, have been identified in similar positions in hCG (unpublished data). These latest findings add to the hypothesis conceived 15 years ago: that ectopic hCG cross-reacted
156
Human Chorionic Gonadotropin (hCG)
Figure 14.2 CKGF dimers. The figure shows images of the TGF dimer, hCG homodimer, and VEGF dimer following bioinformatics studies into hCG folding and receptor interaction. Note the striking resemblances, especially those of hCG and VEGF.
in CKGF receptors. The advances and availability of bioinformatics and molecular biology tools mean that teasing-out of the molecular mechanism of the ectopic function of hCG is tantalizingly close. If one examines the roles of VEGF and TGF during both placentation and oncogenesis, our hypotheses extend further than structural similarities alone. Recent studies have suggested roles for VEGF and hCG in placental vasculature [89,90]. hCG has been proposed as being a novel angiogenic factor in its own right [91], and VEGF is strongly implicated in the development of blood vessels during placentation as a distinct process involving both VEGF receptor expression and stimulation [92]. VEGF is involved in placental vascularization and neovascularization in oncogenesis, and we have now linked immunohistochemically detected hCG expression by cervical cancer with the extent of tumor vascularization [21]; serum hCG levels
Biological Function of the Free -Subunit: Expression and Treatment Target in Cancer
157
have recently been linked to neovascularization in nonseminomatous testicular germcell tumors [93]. In bladder cancer, TGF and its receptors are expressed in high numbers, yet tumors appear to evade apoptosis [94–96]. We now believe hCG may antagonize these pathways [83]. It was also noted that the proximity of the TGF gene is close to the hCG gene cluster on chromosome 19 [97], suggesting possible early evolutionary duplication from TGF. Indeed, it has been shown that there are highly conserved regions between CKGFs with regard to key cystine residues involved in cystine-knot formation [98]. Whatever complex molecular mechanisms are revealed, the most fundamental observation to date is that the immunodepletion of hCG in closed in vitro cultures increases apoptosis of hCG-expressing cancer cells. It is clear that focus should turn to the prognostic significance of hCG detection and utilize what we now know to identify patients who are at greater risk of a more rapid decline in health. From this and from the data we have on hCG immunodepletion, a patient-selection process for studies in which hCG-targeted therapeutics are being investigated is a must if we want to see any significant improvements in the potential for new adjuvant therapies.
14.4 hCG Cancer Vaccines Investigation into the use of immunotherapy as an alternative approach to cancer treatment is still in its infancy, and any efforts to employ vaccination for the treatment of cancer have met with only minimal success. Different approaches either modify tumor cells with immunomodulatory molecules (such as cytokines) [99] or co-stimulatory molecules [100], which enhance either their immunogenicity or their ability to immunize with well-defined specific antigens [101]. Many such antigens relate specifically to studies on melanoma [101,102]; however, p53, p21ras, Her2/neu, EGFRVIII, and MUC1 have been described in studies on cancers of the breast, colon, ovaries, and prostate [101,103–106]. The Her2 example is now perhaps the best understood and, in many ways, suggests a model to which hCG expression by cancer can be compared, but even here the immunogenicity remains a problem [107]. Many of these approaches try to stimulate tumor-specific cytotoxic T-cell-mediated responses. Unfortunately, specifically targeting this arm of the adaptive immune system subjugates the stimulation of B lymphocytes and, therefore, the production of good antibody titers are affected [108]. Immunogens alone are often insufficient to stimulate an adequate immunological response. Carriers such as the diptheria toxoid (DT) or keyhole limpet hemocyanin (KLH) must be employed along with administrations of toxic adjuvants such as Bacillus Calmette–Guerin (BCG). Cowpea mosaic virus (CPMV) was also employed as a successful, and safer, alternative to these [109]. The focus on tumor therapy appears to shift between cell-mediated cytotoxicity and immunodepletive therapy specifically targeting tumor-associated antigens in both the hCG [108] and Her2 arenas [107]. Immunodepletion targets peptides produced by tumors that have been identified to elicit specific tumor effects. Therefore, antibodies to these antigens, in turn, might play a role in preventing metastatic disease. In fact,
158
Human Chorionic Gonadotropin (hCG)
the greatest logical potential for successful therapy is as an adjuvant treatment of cancer. Purposefully targeting metastasis through their production of cancer-associated bioantigens (i.e., those with a demonstrable biological function), rather than the primary tumor itself, would effectively stave off tumor progression, have a greater effect on overall survival, and at the same time allow current therapies to perhaps be more effective. The concept of immunodepletion exploits this and relies on the stimulation of a B-cell antibody response, which results in the reduction of circulating autocrine/ paracrine factors produced and required by the tumor. Therefore, hCG (and in particular hCG) is an ideal target for this form of cancer immunotherapy, because of the control hCG appears to have over the proliferation of carcinoma cells in vitro. Interestingly, vaccines targeting hCG were developed as a potential contraceptive, not as a cancer therapy. The first report was written in 1976 [110], and that same research group dominated the field for the remainder of the decade. Not surprisingly, the majority of studies focused on toxicity, immunogenicity, and epitope selection, preparing for phase I trials. Historically, hCG-CTP has been the epitope antigen of choice in developing anti-hCG fertility vaccines, as any resulting antibody would fail to recognize LH. Although fundamentally important in vaccines for use as infertility treatments, the cross-reaction with LH is not the primary concern in cancer immunotherapy; however, the change in direction from contraception to cancer was a logical and rapid one. Subsequently, the same epitopes selected as targets in fertility control could again be used for specific responses to the hCG (hCG) produced by cancer. There are obvious parallels between the use of anti-hCG vaccines developed as contraception and the potential use in immunodepletion of ectopic hCG expression by tumors, particularly when considering the tolerance of such a vaccine in healthy females where no serious side effects had been reported [111,112]. In each case, the isolated hCG was the target antigen, with the unique carboxy terminal region being the target epitope. In early studies, CTP was coupled to the powerful adjuvant/carrier diptheria toxin to generate higher titers of antibody [113]. It was suggested that it might, therefore, be possible to further increase the effects by utilizing a more immunogenic receptorsignificant hCG epitope for vaccine development. It has been suggested that it might be possible to further increase immunogenicity using vaccines displaying peptides involved in hCG/LH receptor binding, such as those within the long loop region at residues 38– 57 [75]. It was shown that immunization with this peptide, which contains a conformational epitope, conjugated to DT-generated antibodies that have only low cross-reactivity with LH [114]; a hangover obsession from fertility vaccines. When these DT conjugates (including both CTP and hCG 38–57) were investigated, however, both a higher immunological potency and more marked inhibition of hCG bioactivity were observed [115]. An alternative epitope found within the hCG -core fragment and distinct from CTP entirely has also achieved some success in fertility trials in India [116]. We have tested numerous hCG antisera, and monoclonal antibodies directed against the hCG-CTP or other regions clearly bring about reductions in cell number when introduced to hCG-expressing bladder carcinoma cells. In one of our own studies, we investigated the antiserum from mice vaccinated with hCG constructs expressed on CPMV. The expressed constructs were a 37-mer peptide of amino acids 109–145
Biological Function of the Free -Subunit: Expression and Treatment Target in Cancer
159
of the hCG C-terminus (TCDDPRFQDSSSSKAPPPSLPSPSRLPGPSDTPILPQ) (long CTP), and a 24-mer peptide comprising amino acids 109–118 and 132–145 from hCG C-terminus (TCDDPRFQDSSRLPGPSDTPILPQ) (short CTP) identified as significant epitopes (8 and 9) within the CTP [117,118]. Pooled sera from mice immunized with long CTP vaccine were shown to markedly reduce the in vitro growth of the hCG-expressing bladder cancer cell lines used [80]; cell number was reduced by almost 50% in 48 h. In contrast, these sera had no effect on the growth of cells that did not express hCG. Interestingly, sera from mice immunized with short CTP vaccine did not significantly inhibit the growth of any cells tested [80]. The reason for this difference is unclear, but the results suggest a difference in affinity for hCG between the two antibody populations generated, or may indicate that certain regions of CTP are involved in the ectopic biological activity of hCG. With this information, one might wonder: is CTP the best antigen? CTP plays no known functional role and is assumed to be a peripheral entity. CTP-specific antibodies still react with hCG even when it is bound to the hCG/LH receptor [117]. The epitope is not particularly immunogenic, and yields low-affinity antibodies when compared to other hCG epitopes [119]. Its selection as a target was, in many ways, only to distinguish hCG from LH, and there are at least five other unique epitopes on hCG [118]—named 1, 6, and 7—in addition to 8 and 9 on CTP. In more recent studies, these epitopes were used to generate high-affinity antibodies using mutant hCG with disrupted LH common epitopes [108]. One mutant with a Arg Glu substitution at amino acid 68 generated high-affinity antibodies in rabbits [120]. The vaccines are generally effective at generating titers. Fertility vaccine antibodies have been observed to remain at detectable levels in humans for up to 10 months [15]; this would provide ideal postsurgical protection against any metastatic spread. In addition, a method of administering single-dose CTP-DT vaccine in biodegradable microspheres gave measurable hCG antibodies for more than a year [121]. Phase I clinical trials using anti-hCG vaccines have been carried out in patients with advanced-stage epithelial tumors. The initial results have indicated that the diphtheria toxin CTP immunogen is well tolerated, safe, and immunogenic in humans [122]. The results were also promising in clinical trials on pancreatic and colorectal cancer patients [123], although it was not shown at the time whether the immune response would significantly reduce any tumor development. Most recently, phase II clinical trials on colorectal cancer have been completed using the CTP-DT vaccine [124], with only limited success when a strong immune response was associated with improved survival. We have proposed a molecular model that accounts for resistance to radiotherapy and associated poor prognosis for hCG-expressing cancers. We postulated that removing hCG from the circulation should assist in initiating tumor regression in vivo through resumption of apoptosis; however, trials conducted to date have yet to show any significant promise for anti-hCG vaccines as an adjuvant treatment for cancer. If multiple cancers are producing these molecules, and a credible link between their production and tumor growth can be demonstrated, why have attempts to use vaccines failed to deliver any significant responses in patients? As described in this chapter, only about a third of epithelial cancers express hCG. To date, patients have
160
Human Chorionic Gonadotropin (hCG)
not been preselected based on their hCG status when recruited into studies. In the case of the latest trial, vaccine was given to all colorectal cancer patients and was not restricted to those who tested hCG positive. A vaccine that stimulates immune responses against hCG would be expected to benefit only those patients in whom hCG is aberrantly expressed, suggesting that the vaccine might have benefited a higher proportion of the trial group if detection of ectopic hCG expression had been applied as a fundamental inclusion criterion for entry into the trial. Indeed, the Her2 paradigm clearly showed that Herceptin® would never have displayed its smallbut-significant advantage in breast cancer patient survival if it had been given to all breast cancer patients irrespective of their Her2 expression status [125].
Summary In summary, this chapter reviews and reinforces the collective data supporting the notion that the ectopic secretion of hCG by common epithelial tumors is an epiphenomenon. Indeed, expression of hCG by epithelial cancer is approximately 30%, and in bladder cancer, pancreatic cancer, and colorectal cancer, per se, may be as high as 50%. The expression is associated with resistance to radiotherapy, rapid development of metastasis, and thus poor prognosis. hCG influences cell population growth by inhibiting apoptosis in culture—an indirect molecular mechanism consistent with the clinical finding of resistance to radiotherapy. Structural biophysical studies and bioinformatics analysis show that hCG can homodimerize; it shares topological and homological features with the cystine-knot family of growth factor proteins. The topological resemblance of TGF and VEGF to hCG is particularly striking, and hCG has been shown to reverse TGF-induced apoptosis. We hypothesize that any ectopic hCG effect is modulated via the TGF-RII receptor and/or the VEGF KDR receptor, bringing about multiple responses associated with the oncogenic process. Whatever the precise mechanism, hCG has a growth-modulating function in tumorigenesis which can be removed using specific antibodies. This presents a new, adjuvant approach to treatment strategies pertaining to hCG-expressing cancers. One such potential approach is immunodepletive therapy in the form of an anti-hCG vaccine.
References [1] Djewitzi W. Über einen fall von chorionepithelioma der harnblase. Virchows Arch 1904;178:451–64. [2] Aschheim S, Zondek B. Das hormon des hypophysenvorderlappens: testobjekt zum nachweis des hormons. Klin Wochenschr 1927;6:248–52. [3] McFadzean AJS. Feminization associated with carcinoma of the adrenal cortex. Lancet 1946;2:940–3. [4] Iles RK, Chard T. Human chorionic gonadotropin expression by bladder cancers: biological and clinical potential. J Urol 1991;145:453–8. [5] Iles RK, Lee CL, Oliver RTD, Chard T. Composition of intact hormone and free subunits in the hCG-like material found in serum and urine of patients with carcinoma of the bladder. Clin Endocrinol 1990;32:355–64.
Biological Function of the Free -Subunit: Expression and Treatment Target in Cancer
161
[6] Stenman UH, Alfthan H, Hotakainen K. Human chorionic gonadotropin in cancer. Clin Biochem 2004;37:549–61. [7] Iles RK, Chard T. Immunochemical analysis of human chorionic gonadotropin-like material secreted by “normal” and neoplastic urothelial cells. J Mol Endocrinol 1989;2:107–12. [8] Iles RK, Purkis PE, Whitehead PC, Oliver RTD, Leigh I, Chard T. Expression of human chorionic gonadotropin by non-trophoblastic nonendocrine “normal” and malignant epithelial cells. Br J Cancer 1990;61:663–6. [9] Iles RK, Butler SA. Human urothelial carcinomas—a typical disease of the aged: the clinical utility of human chorionic gonadotropin in patient management and future therapy. Exp Gerontol 1998;33:379–91. [10] Dirnhofer S, Koessler P, Ensinger C, Feichtinger H, Madersbacher S, Berger P. Production of trophoblastic hormones by transitional cell carcinoma of the bladder: association to tumor stage and grade. Hum Pathol 1998;29:377–82. [11] Mora J, Gascon N, Tabernero JM, Rodriguez-Espinosa J, Gonzalez-Sastre F. Different hCG assays to measure ectopic hCG secretion in bladder carcinoma patients. Br J Cancer 1996;74:1081–4. [12] El-Ahmady O, Halim AB, Mansour O, Salman T, El-Din AG, Walker RP. Urinary gonadotropin peptide (UGP) in Egyptian patients with benign and advanced malignant urological disease. Br J Cancer 1996;73:1486–90. [13] Pectasides D, Bafaloucos D, Antoniou F, Gogou L, Economides N, Varthalitis J, et al. AFP, -hCG, PSA, SCC and CA 19-9 for monitoring transitional cell carcinoma of the bladder. Am J Clin Oncol 1996;19:271–7. [14] Iles RK, Persad R, Trivedi M, Sharma KB, Dickinson A, Smith P, et al. Urinary concentration of human chorionic gonadotropin and its fragments as a prognostic marker in bladder cancer. Br J Urol 1996;77:61–9. [15] Halim A-B, El-Ahmady O, Hamza S, Aboul-Ela M, Oehr P. Urinary -hCG in benign and malignant urinary tract disease. Disease Markers 1995;12:109–15. [16] Chang E, Lee E, Oh SJ, Kim JS, Kang C. The immunoexpressions and prognostic significance of inhibin alpha and beta human chorionic gonadotropins (HCG) in breast carcinomas. Cancer Res Treat 2005;37:241–6. [17] Bieche I, Lazar V, Nogues C, Poynard T, Giovangrandi Y, Bellet D, et al. Prognostic value of chorionic gonadotropin beta gene transcripts in human breast carcinoma. Clin Cancer Res 1998;4:671–6. [18] Schwarz-Roeger U, Petzoldt B, Waldschmidt R, Walker RP, Bauknecht T, Kiechle M, et al. UGP-A tumor marker of gynecologic and breast malignancies? Specificity and sensitivity in pretherapeutic patients and the influence of hormonal substitution on the expression of UGP. Anticancer Res 1997;17:3041–5. [19] Hoon DS, Sarantou T, Doi F, Chi DD, Kuo C, Conrad AJ, et al. Detection of metastatic breast cancer by beta-hCG polymerase chain reaction. Int J Cancer 1996;69:369–74. [20] Agnantis NJ, Patra F, Khaldi L, Filis S. Immunohistochemical expression of subunit beta HCG in breast cancer. Eur J Gynaecol Oncol 1992;13:461–6. [21] Li D, Wen X, Ghali L, Al-Shalabi FM, Docherty SM, Purkis P, et al. hCG beta expression by cervical squamous carcinoma: in vivo histological association with tumor invasion and apoptosis. Histopathol 2008;53:147–55. [22] Crawford RA, Iles RK, Carter PG, Caldwell CJ, Shepherd JH, Chard T. The prognostic significance of beta human chorionic gonadotropin and its metabolites in women with cervical carcinoma. J Clin Pathol 1998;51:685–8. [23] Grosssmann M, Hoermann R, Gocze PM, Ott M, Berger P, Mann K. Measurement of human chorionic gonadotropin-related immunoreactivity in serum, ascites and tumor cysts of patients with gynaecologic malignancies. Eur J Clin Invest 1995;25:867–73.
162
Human Chorionic Gonadotropin (hCG)
[24] Louhimo J, Carpelan-Holmstrom M, Alfthan H, Stenman UH, Jarvinen HJ, Haglund C. Serum hCG , CA 72–4 and CEA are independent prognostic factors in colorectal cancer. Int J Cancer 2002;101:545–8. [25] Lundin M, Nording S, Carpelan-Holmstrom M, Louhimo J, Alfthan H, Stenman UH, et al. A comparison of serum and tissue hCG as prognostic markers in colorectal cancer. Anticancer Res 2000;20:4949–51. [26] Kido A, Mory M, Adachi Y, Yukaya H, Ishida T, Sugimachi K. Immunohistochemical expression of beta-human chorionic gonadotropin in colorectal carcinoma. Surg Today 1996;26:966–70. [27] Webb A, Scott-Mackie P, Cunningham D, Norman A, Andreyev J, O’Brien M, et al. The prognostic value of CEA, beta HCG, AFP, CA125, CA19-9 and C-erb B-2, beta HCG immunohistochemistry in advanced colorectal cancer. Ann Oncol 1995;6:581–7. [28] Berzal-Cantalejo F, Sabater Marco V, Alonso Hernandez S, Jimenez Pena R, Martorell Cebollada MA. Syncytial giant cell component. Review of 55 renal cell carcinomas. Histol Histopathol 2004;19:113–18. [29] Hotakainen K, Ljunberg B, Paju A, Rasmusen T, Althan H, Stenman UH. The free beta subunit of human chorionic gonadotropin as a prognostic factor in renal cell carcinoma. Br J Cancer 2002;86:185–9. [30] Dirnhofer S, Freund M, Rogatsch H, Krabichler S, Berger P. Selective expression of trophoblastic hormones by lung carcinoma: neurendocrine tumors exclusively produce human chorionic gonadotropin -subunit (hCG). Hum Pathol 2000;31:966–72. [31] Szturmowicz M, Wiatr E, Sakowicz A, Slodkowska K, Filipecki S, RowinskaZakrzewska ER. The role of human chorionic gonadotropin beta subunit elevation in small-cell lung cancer patients. Tumour Biol 1999;20:99–104. [32] Slodkwoska J, Szturmowicz M, Rudzinski P, Giedronowicz D, Sakowitcz A, Androsiuk W, et al. Expression of CEA and trophoblastic cell markers by lung carcinoma in association with histological characteristics and serum marker levels. Eur J Cancer Prev 1998;7:51–60. [33] Yokotani T, Koizumi T, Taniguchi R, Nakagawa T, Isobe T, Yoshimura M, et al. Expression of alpha and beta genes of human chorionic gonadotropin in lung cancer. Int J Cancer 1997;16(71):539–44. [34] Boucher LD, Yoneda K. The expression of trophoblastic cell markers by lung carcinomas. Hum Pathol 1995;26:1201–6. [35] Shah T, Srirajaskanthan R, Bhogal M, Toubanakis C, Meyer T, Noonan A, et al. -fetoprotein and human chorionic gonadotropin- as prognostic markers in neuroendocrine tumor patients. Br J Cancer 2008;99:72–7. [36] Hedstrom J, Grenman R, Ramsey H, Finne P, Lundin J, Haglund C, et al. Concentration of free hCGbeta subunit in serum as a prognostic marker for squamous-cell carcinoma of the oral cavity and oropharynx. Int J Cancer 1999;84:525–8. [37] Bhalang K, Kafrawy AH, Miles DA. Immunohistochemical study of the expression of human chorionic gonadotropin-beta in oral squamous cell carcinomas. Cancer 1999;85:757–62. [38] Vartiainen J, Lassus H, Lehtovirta P, Finne P, Alfthan H, Butzow R, et al. Combination of serum hCG beta and p53 tissue expression defines distinct subgroups of serous ovarian carcinoma. Int J Cancer 2008;122:2125–9. [39] Ind T, Iles RK, Shepherd J, Chard T. Serum concentrations of cancer antigen 125, placental alkaline phosphatase, cancer-associated serum antigen and free beta human chorionic gonadotropin as prognostic markers for epithelial ovarian cancer. Br J Obs Gyn 1997;104:1024–9. [40] Sheaff MT, Martin JE, Badenoch DF, Baithun SI. hCG as a prognostic marker in adenocarcinoma of the prostate. J Clin Pathol 1996;49:329–32.
Biological Function of the Free -Subunit: Expression and Treatment Target in Cancer
163
[41] Otite U, Baithun S, Chinegwundoh F, Nargund VH, Iles RK. Detection of human chorionic gonadotropin-beta in serum or urine of prostate cancer patients is of no clinical significance. Tumor Biol 2006;27:181–6. [42] Louhimo J, Alfthan H, Stenman UH, Haglund C. Serum hCG b and CA 72–4 are stronger prognostic factors than CEA, CA 19-9 and CA 242 in pancreatic cancer. Oncol 2004;66:126–31. [43] Louhimo J, Nordling S, Alfthan H, von Boguslawski K, Stenman UH, Haglund C. Specific staining of human chorionic gonadotropin in benign and malignant gastrointestinal tissues with monoclonal antibodies. Histopathol 2001;38:418–24. [44] Syrigos KN, Fyssas I, Konstandoulakis MM, Harrington KJ, Papadopoulos S, Milingos N. Beta human chorionic gonadotropin concentrations in serum of patients with pancreatic adenocarcinoma. Gut 1998;42:88–91. [45] de Bruijn HW, ten Hoor KA, Krans M, van der Zee AG. Rising serum values of betasubunit human chorionic gonadotropin (hCG) in patients with progressive vulvar carcinomas. Br J Cancer 1997;75:1217–18. [46] Carter PG, Iles RK, Neven P, Ind TEJ, Shepherd JH, Chard T. Measurement of urinary core fragment of human chorionic gonadotropin in women with vulvovaginal malignancy and its prognostic significance. Br J Cancer 1995;71:350–3. [47] Hussa ROThe clinical marker hCG. New York: Praeger Publishers; 1987. [48] lles RK, Czepulkowski BH, Young BD, Chard T. Amplification and re-arrangement of the -human chorionic gonadotropin (-hCG)-human LH gene cluster is not responsible for the ectopic production of -hCG by bladder tumor cells. J Mol Endocrinol 1989;2:113–17. [49] Bellet D, Lazar V, Bieche I, Paradis V, Giovangrandi Y, Paterlini P, et al. Malignant transformation of non-trophoblastic cells in association with the expression of chorionic gonadotropin beta genes transcribed in trophoblastic cells. Cancer Res 1997;57:516–23. [50] Bo M, Boime I. Identification of the transcriptionally active genes of the chorionic gonadotropin gene in vivo. J Biol Chem 1992;267:3179–84. [51] Rull K, Laan M. Expression of -subunit of human chorionic gonadotropin genes during the normal and failed pregnancy. Hum Reprod 2005;20:3360–8. [52] Giovangrandi Y, Parfait B, Asheuer M, Olivi M, Lidereau R, Vidaud M, et al. Analysis of the human CGB/LHB gene cluster in breast tumors by real-time quantitative RT-PCR assays. Cancer Lett 2001;168:93–100. [53] Lazar V, Diez SG, Laurent A, Giovangrandi Y, Radvanyi F, Chopin D, et al. Expression of human chorionic gonadotropin beta subunit genes in superficial and invasive bladder carcinomas. Cancer Res 1995;55:3735–8. [54] Span PN, Manders P, Heuvel JJ, Thomas CM, Bosch RR, Beex LV, et al. Molecular beacon reverse transcription-PCR of human chorionic gonadotropin-beta-3, -5, and -8 mRNAs has prognostic value in breast cancer. Clin Chem 2003;49:1074–80. [55] Braunstein GD, Vaitukaitis JL, Carbone PP, Ross GT. Ectopic production of human chorionic gonadotropin by neoplasms. Ann Intern Med 1973;78:39–45. [56] Dosogne-Geurin M, Stolarczyk A, Borkowski A. Prospective study of the alpha and beta subunits of hCG in the blood of patients with various benign and malignant conditions. Eur J Cancer 1978;14:525–32. [57] Braunstein GD. hCG expression in trophoblastic and nontrophoblastic tumors. In: Fishman WH, editor. Oncodevelopmental markers: biologic, diagnostic and monitoring aspects. New York: Academic Press; 1983:35. [58] Rodenburg CJ, Nieuwenhuyzen-Kruseman AC, DeMaaker HA, Fieuren EJ, Van Oosterom AT. Immunohistochemical localization and chromatographic characterization
164
[59] [60] [61]
[62]
[63]
[64] [65] [66]
[67]
[68]
[69]
[70] [71]
[72]
[73]
[74]
[75] [76]
Human Chorionic Gonadotropin (hCG)
of human chorionic gonadotropin in a bladder carcinoma. Arch Path Lab Med 1985; 109:1046–8. Iles RK, Oliver RTD, Kitau M, Walker C, Chard T. In vitro secretion of human chorionic gonadotropin by bladder tumor cells. Br J Cancer 1987;55:623–6. Iles RK, Jenkins BJ, Oliver RTD, Blandy JP, Chard T. human chorionic gonadotropin in serum and urine. A marker for metastatic urothelial cancer. Br J Urol 1989;64:241–4. Hagen C, Gilby ED, McNeilly AS, Olgaard K, Bondy PK, Rees LH. Comparison of circulating glycoprotein hormones and their subunits in patients with oat cell carcinoma of the lung and uraemic patients on chronic dialysis. Acta Endocrinologica 1976;83:26–35. Kahn CR, Rosen SW, Weintraub BD, Fajans SS, Gorden P. Ectopic production of chorionic gonadotropin and its subunits by islet-cell tumors. A specific marker for malignancy. N Engl J Med 1977;297:565–9. Adejuwon CA, Segal SJ. Apparent chorionic gonadotropin immunoreactivity in human non-placental tissues: reduction by protease inhibitors or heat treatment. Clin Chim Acta 1984;137:245–53. Monteiro JC, Ferguson KM, McKinna JA, Greening WP, Neville AM. Ectopic production of human chorionic gonadotropin-like material by breast cancer. Cancer 1984;53:957–62. Cowley G, Smith JA, Ellison M, Gusterson B. Production of beta-human chorionic gonadotropin by human squamous carcinoma cell lines. Int J Cancer 1985;35:575–9. Heitz PU, von Herbay G, Kloppel G, Komminoth P, Kasper M, Hofler H, et al. The expression of subunits of human chorionic gonadotropin (hCG) by nontrophoblastic, nonendocrine, and endocrine tumors. Am J Clin Pathol 1987;88:467–72. Cole LA, Wang Y, Elliot M, Latif M, Chambers JT, Chambers SK, et al. Urinary human chorionic gonadotropin free -subunit and core fragment: a new marker of gynecological cancers. Cancer Res 1988;48:1356–60. Yamaguchi A, Ishida T, Nishimura G, Kumaki T, Katoh M, Kosaka T, et al. Human chorionic gonadotropin in colorectal cancer and its relationship to prognosis. Br J Cancer 1989;60:382–4. Marcillac I, Troalen F, Bidart J-M, Ghillani P, Ribrag V, Escudier B, et al. Free human chorionic gonadotropin subunit in gonadal and nongonadal neoplasms. Cancer Res 1992;52:3901–7. Marcillac I, Cottu P, Theodore C, Terrier-Lacombe MJ, Bellet D, Droz JP. Free hCG- subunit as tumor marker in urothelial cancer. Lancet 1993;341:1354–5. Daja MM, Aghmesheh M, Ow KT, Rohde PR, Barrow KD, Russell PJ. Beta-human chorionic gonadotropin in semen: a marker for early detection of prostate cancer? Mol Urol 2000;4:421–7. Span PN, Thomas CM, Heuvel JJ, Bosch RR, Schalken JA, vd Locht L, et al. Analysis of expression of chorionic gonadotropin transcripts in prostate cancer by quantitative Taqman and a modified molecular beacon RT-PCR. J Endocrinol 2002;172:489–95. Jiang Y, Zeng F, Xiao C, Liu J. Expression of beta-human chorionic gonadotropin genes in renal cell cancer and benign renal disease tissues. J Huazhong Univ Sci Technolog Med Sci 2003;23:291–3. Moutzouris G, Yannapoulos D, Barbatis C, Zaharof A, Theodorou C. Is human chorionic gonadotropin production by transitional cell carcinoma of the bladder a marker of aggressive disease and resistance to radiotherapy? Br J Urol 1993;72:907–9. Iles RK, Chard T. Molecular insights into structure and function of human chorionic gonadotropin. J Mol Endocrinol 1993;10:217–34. Vaitukaitis JL, Braunstein GD, Ross GT. A radioimmunoassay which specifically measures human chorionic gonadotropin in the presence of human luteinizing hormone. Am J Obstet Gynecol 1972;113:751–8.
Biological Function of the Free -Subunit: Expression and Treatment Target in Cancer
165
[77] Iles RK, Delves PJ, Butler SA. Does hCG or hCG play a role in cancer cell biology? Mol Cell Endocrinol 2010, in press. [78] Gillott DJ, lles RK, Chard T. The effects of -human chorionic gonadotropin on the in vitro growth of bladder cancer cell lines. Br J Cancer 1996;73:323–6. [79] Bagshawe KD, Springer CJ, Searle F, Antoniw P, Sharma SK, Melton RG, et al. A cytotoxic agent can be generated selectively at cancer sites. Br J Cancer 1988;58:700–3. [80] Butler SA, Staite EM, Iles RK. Reduction of bladder cancer cell growth in response to hCG CTP37 vaccinated mouse serum. Oncol Res 2003;14:93–100. [81] Devi GR, Oldenkamp JR, London CA, Iversen PL. Inhibition of human chorionic gonadotropin beta-subunit modulates the mitogenic effect of c-myc in human prostate cancer cells. Prostate 2002;53:200–10. [82] Jankowska A, Gunderson SI, Andrusiewicz M, Burczynska B, Szczerba A, Jarmolowski A, et al. Reduction of human chorionic gonadotropin beta subunit expression by modified U1 snRNA caused apoptosis in cervical cancer cells. Mol Cancer 2008;7:26. [83] Butler SA, Ikram MS, Mathieu S, Iles RK. The increase in bladder carcinoma cell population induced by the free subunit of human chorionic gonadotropin is a result of an anti-apoptosis effect and not cell proliferation. Br J Cancer 2000;82:1553–6. [84] Lapthorn AJ, Harris DC, Littlejohn A, Lustbader JW, Canfield RE, Machin KJ, et al. Crystal structure of human chorionic gonadotropin. Nature 1994;369:455–61. [85] Wu H, Lustbader JW, Liu Y, Canfield RE, Hendrickson WA. Structure of human chorionic gonadotropin at 2.6 Å resolution from MAD analysis of the selenomethionyl protein. Structure 1994;2:545–8. [86] Butler SA, Laidler P, Porter JR, Kicman AT, Chard T, Cowan DA, et al. The beta subunit of human chorionic gonadotropin exists as a homodimer. J Mol Endocrinol 1999;22:185–92. [87] Butler SA, Iles RK. Monomeric human chorionic gonadotropin beta is responsible for autocrine growth effects. Tumor Biol 2004;25:18–23. [88] Muller YA, Li B, Christinger HW, Wells JA, Cunningham BC, De Vos AM. Vascular endothelial growth factor: crystal structure and functional mapping of the kinase domain receptor binding site. Proc Natl Acad Sci USA 1997;94:7192–7. [89] Herr F, Baal N, Reisinger K, Lorenz A, McKinnon T, Preissner KT, et al. HCG in the regulation of placental angiogenesis. Results of an in vitro study. Placenta 2007:S85–93. [90] Reisinger K, Baal N, McKinnon T, Münstedt K, Zygmunt M. The gonadotropins: tissuespecific angiogenic factors? Mol Cell Endocrinol 2007;269:65–80. [91] Zygmunt M, Herr F, Keller-Schoenwetter S, Kunzi-Rapp K, Münstedt K, Rao CV, et al. Characterization of human chorionic gonadotropin as a novel angiogenic factor. J Clin Endocrinol Metab 2002;87:5290–6. [92] Demir R, Kayisli UA, Seval Y, Celik-Ozenci C, Korgun ET, Demir-Weusten AY, et al. Sequential expression of VEGF and its receptors in human placental villi during very early pregnancy: differences between placental vasculogenesis and angiogenesis. Placenta 2004;25:560–72. [93] Arrieta O, Michel Ortega RM, Angeles-Sánchez J, Villarreal-Garza C, Avilés-Salas A, Chanona-Vilchis JG, et al. Serum human chorionic gonadotropin is associated with angiogenesis in germ cell testicular tumors. J Exp Clin Cancer Res 2009;28:120. [94] Coombs LM, Pigott DA, Eydmann ME, Proctor AJ, Knowles MA. Reduced expression of TGF is associated with advanced disease in transitional cell carcinoma. Br J Cancer 1993;67:578–84. [95] Eder IE, Stenzl A, Hobisch A, Cronauer MV, Bartsch G, Klocker H. Transforming growth factors-1 and 2 in serum and urine from patients with bladder carcinoma. J Urol 1996;157:953–7.
166
Human Chorionic Gonadotropin (hCG)
[96] Eder IE, Stenzl A, Hobisch A, Cronauer MV, Bartsch G, Klocker H. Expression of transforming growth factors-1, 2 and 3 in human bladder carcinomas. Br J Cancer 1997;75:1753–60. [97] Fujii D. Transforming growth factor gene maps to human chromosome 19 long arm and to mouse chromosome 7. Somatic Cell Mol Genet 1986;12:281–8. [98] Vitt UA, Hsu SY, Hsueh AJW. Evolution and classification of cystine knot-containing hormones and related extracellular signaling molecules. Mol Endocrinol 2001;15:681–94. [99] Levitsky HI, Mongomery J, Ahmadzadeh M, Staveley-O’Carroll K, Guarnieri F, Longo DL, et al. Immunization with granulocyte-macrophage colony-stimulating factor-transduced, but not B7–1 transduced, lymphoma cells primes idiotype-specific T cells and generates potent systemic antitumor immunity. J Immunol 1996;156:3858–65. [100] Allison JP, Hurwitz AA, Leach DR. Manipulation of costimulatory signals to enhance antitumor T-cell responses. Curr Opin Immunol 1995;7:682–6. [101] Boon T, Cerottini J-C, Van den Eynde B, van der Bruggen P, Van Pel A. Tumour antigens recognized by T lymphocytes. Annu Rev Immunol 1994;12:337–65. [102] Maeurer MJ, Storjus WJ, Kirkwood JM, Lotze MT. New treatment options for patients with melanoma: review of melanoma-derived epitope-based peptide vaccines. Melanoma Res 1996;6:11–24. [103] Strauss HJ. Mutant ras proteins and peptides: bad news for tumors. J Natl Cancer Inst 1995;87:1820–2. [104] Fisk B, Chesak B, Pollack MS, Wharton JT, Ioannides CG. Oligopeptide induction of a cytotoxic T lymphocyte response to HER-2/Neu prot-oncogene in vitro. Cell Immunol 1994;157:415–27. [105] Humphrey PA, Wong AJ, Vogelstein B, Zalutsky MR, Fuller GN, Archer G, et al. Antisynthetic peptide antibody reacting at the fusion junction of deletion-mutant epidermal growth factor receptors in human glioblastoma. Proc Natl Acad Sci USA 1990;87:4207–11. [106] Burchell J, Taylor-Papdimitriou J, Boshell M, Gendler S, Duhig T. A short sequence within the amino acid tandem repeat of a cancer-associated mucin, contains immunodominant epitopes. Int J Cancer 1989;44:691–6. [107] Renard V, Leach DR. Perspectives on the development of a therapeutic HER-2 cancer vaccine. Vaccine 2007;25S:B17–23. [108] Delves PJ, Iles RK, Roitt IM, Lund T. Designing a new generation of anti-hCG vaccines for cancer therapy. Mol Cell Endocrinol 2007;260–262:276–81. [109] Brennan FR, Jones TD, Hamilton WDO. Cowpea mosaic virus as a vaccine carrier of heterologous antigens. Mol Biotechnol 2001;17:15–26. [110] Talwar GP, Sharma NC, Dubey SK, Salahuddin M, Das C, Ramakrishnan S, et al. Isoimmunization against human chorionic gonadotropin with conjugates of processed 3-subunit of the hormone and tetanus toxoid. Proc Nat Acad Sci USA 1976;73:218–22. [111] Griffin PD. Immunisation against hCG. Hum Reprod 1994;9:267–72. [112] Stevens VC. Progress in the development of human chorionic gonadotropin anti-fertility vaccine. Am J Reprod Immunol 1996;35:148–55. [113] Stevens VC, Cinader B, Powell JE, Lee AC, Koh SW. Preparation and formulation of a human chrionic gonadotropin antifertility vaccine: selection of adjuvant and vehicle. Am J Reprod Immunol 1981;1:315–21. [114] Stevens VC, Chou WS, Powell JE, Lee AC, Smoot J. The identification of peptide sequences of human chorionic gonadotropin containing a conformational epitope. Immunol Lett 1986;12:11–18.
Biological Function of the Free -Subunit: Expression and Treatment Target in Cancer
167
[115] Triozzi PL, Stevens VC. Human chorionic gonadotropin as a target for cancer vaccines. Oncol Reports 1999;6:7–17. [116] Talwar GP. Fertility regulating and immunotherapeutic vaccines reaching trials stage. Hum Reprod Update 1997;3(4):301–10. [117] Dirnhofer S, Klieber R, De Leeuw R, Bidart JM, Merz WE, Wick G, et al. Functional and immunological relevance of the COOH-terminal extension of human chorionic gonadotropin beta: implications for the WHO birth control vaccine. FASEB J. 1993;7:1381–5. [118] Berger P, Sturgeon C, Bidart JM, Paus E, Gerth R, Niang M. The ISOBM TD-7 workshop on hCG and related molecules. Towards user-oriented standardization of pregnancy and tumor diagnosis: assignment of epitopes to the three-dimensional structure of diagnostically and commercially relevant monoclonal antibodies directed against human chorionic gonadotropin and derivatives. Tumour Biol 2002;23:1–38. [119] Talwar GP. Vaccines and passive immunological approaches for the control of fertility and hormone-dependent cancers. Immunol Rev 1999;171:173–92. [120] Porakishvili N, Chiesa MD, Chikadze N, Martensen P, Justesen J, Lund T, et al. Elimination of luteinizing hormone cross-reactive epitopes from human chorionic gonadotropin. Vaccine 2002;20:2053–9. [121] Stevens VC. Future perspectives for vaccine development. Scand J Immunol Suppl 1992;11:137–43. [122] Triozzi PL, Martin EW, Gouchnour D, Aldritch W. Phase 1b trial of a synthetic beta human chorionic gonadotropin vaccine in patients with metastatic cancer. Ann N Y Acad Sci 1993;630:358–9. [123] Clinical Triozzi PL. immunological effects of a synthetic beta-hCG vaccine. Int J Oncol 1994;5:1447–53. [124] Moulton HM, Yoshihara PH, Mason DH, Iversen PL, Triozzi PL. Active specific immunotherapy with a beta-human chorionic gonadotropin peptide vaccine in patients with metastatic colorectal cancer: antibody response is associated with improved survival. Clin Cancer Res 2002;8:2044–51. [125] Baselga J. Clinical trials of Herceptin® (trastuzumab). Eur J Cancer 2001;37:S18–24.
15 Use of hCG in Reproductive Dysfunction Francis W. Byrn Obstetrics and Gynecology, University of New Mexico, Albuquerque, NM, USA
Human chorionic gonadotropin (hCG) is a heterodimeric glycoprotein. The primary structures of the - and -subunits were determined in 1972. The hormone is approximately 30% carbohydrate [1]. Its similarity to luteinizing hormone (LH) and its capacity to bind to its receptors has fostered interest in hCG’s clinical utility in human reproductive medicine. Studies were performed that clarified the disappearance rates of hCG following pregnancy delivery [2], after intravenous (IV) and intramuscular (IM) injections to healthy males and females [3], after subcutaneous or IM injections in healthy males [4], as well as calculating metabolic and renal clearance rates with continuous-infusion and single-injection techniques [5]. These studies similarly confirmed that hCG disappears in a rapid phase and a slower phase, taking approximately 6 h and 36 h [5], respectively. Using an injection of 10,000 IU, Midgeley and Jaffe’s early data on hCG disappearance following delivery also estimated a fast, initial halflife of 8.9 h and a longer, trailing half-life of 37.2 h. They discussed the fact that the rate of disappearance at a given time following delivery did not appear to be related to the concentration of hCG, nor to the duration of pregnancy, nor to the route of delivery [2]. With IV rates, Rizkallah demonstrated an initial fast half-life of 3.75.9 h, and a slower 22- to 26.7-h half-life [3]. Studies examining the pharmacokinetics of hCG when administered to humans have been published [3–5]. In males, the peak hCG concentrations are reached 6–8 h after IM injection [4]. When administered intramuscularly to three subjects, the halflife of the tail was longer (27.5, 31.2, and 31.6 h) compared to clearance rates after IV injection, which averaged approximately 26 h. There were no appreciable differences in the disappearance curves between males and females [3]. In this study, the researchers also made other observations, including that the metabolic clearance rate of hCG was more prolonged than that of human LH. The maximal blood concentrations achieved with IV versus IM injections of 10,000 IU of hCG were approximately 5.8 and 2.2 IU/ml, respectively. The peak of hCG was delayed 6 h and somewhat lower compared to IV administration. Using then current second international reference preparation-human menopausal gonadotropins (IRP-hMG) standards, the authors noted that LH in human ovulation reaches 40–200 mIU during a 24-h period. They commented on the fact that the maximum concentration of hCG after an IM Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00015-3 © 2010 Elsevier Inc. All rights reserved.
172
Human Chorionic Gonadotropin (hCG)
injection in one subject was equivalent to 4000 mIU of the second IRP-hMG standard, noting that this was an approximately 20-fold increase over the maximum values of the LH peak (in terms of standard reference) during spontaneous ovulation. The slower disappearance rate of hCG relative to LH was characterized as a disadvantage, in that slower metabolism could be an aggravating factor in multiple ovulation and an independent factor for ovarian hyperstimulation syndrome (OHSS) [3]. A study by Fischer et al. investigated the time interval of ovulation in women subjects undergoing ovulation management with clomiphene after either IM injection of 10,000 or 500 IU provided intravascularly on an alternating protocol for two cycles [6]. Ultrasound assessment and serum samples were obtained every 2 h beginning 30 h after IV administration and 31 h after IM administration. Given the apparent variability of ovulation intervals demonstrated by IM hCG administration, the authors hypothesized that an ovulation-inducing threshold of hCG provided by IV administration would result in less variance in the endpoint of ovulation. Twentyfive women provided data by completing both induction protocols. The mean time to ovulation with IM injection of hCG was 40.4 h, and 38.3 h with IV administration of hCG. The 500 IU dosage utilized in this study had previously been noted to exceed the peak LH concentration by threefold [6]. Ultimately, 48 of 55 (87.3%) studied cycles provided physical or endocrinologic evidence of ovulation. Patterns of serum hCG levels were monitored in four random patients. In comparison to previously mentioned pharmacokinetic studies [1,5], the study by Fischer et al. [6] obtained clinical trial data more practically consistent with actual patient management strategies. Thirty-six hours after IM injection, hCG concentrations had increased to a mean level of 311 IU/l (range 201–359 IU/l) from preadministration hCG concentrations of less than 2 IU/l. Twenty-one hours after IV administration of 500 IU of hCG, serum levels were 45 IU/l, with a range of 36–55 IU/l. The patterns and rates of change of estradiol and progesterone levels from the time of hCG administration until documented ovulation were reported not to be different as a consequence of the routes of administration of the two different doses of hCG. These studies used urinary-derived chorionic gonadotropin to augment or substitute for LH. Early data suggest that hCG has a longer half-life than LH, and that IM administration of 5000 IU (and certainly 10,000 IU) significantly exceeds the maximal concentrations of LH during the peak of the surge during spontaneous ovulation [3]. Although not specifically studied, the authors discussed possible implications for multiple ovulations and/or sustaining supraphysiologic estrogen and progesterone secretion from ovaries stimulated with hMG.
15.1 Historical Overview and Perspective The indications for hCG administration within clinical reproductive medicine most often involve ovulation management. One must keep in mind that this is a very broad category. The most common clinical indication for an hCG injection is to secure
Use of hCG in Reproductive Dysfunction
173
ovulation (as a completing event of natural or induced folliculogenesis). For example, this could be an eventual recommendation for an otherwise anovulatory woman who wishes to conceive. A second hCG use, however, is to prospectively promote the release of one or more oocytes at a time anticipated to coincide with an ancillary procedure such as timed intercourse or artificial insemination. In this circumstance, anovulation might not be the primary problem. Although definitions cannot always be uniformly applied in discussions of medical protocols that involve an injection of hCG, it will be helpful to think of protocols as a continuum in three stages. Ovulation induction (OI) is focused on helping a woman ovulate if, without assistance, she might ovulate infrequently or not at all. Controlled ovarian stimulation (COS) is a more comprehensive regimen of either oral or parenteral medications that promote ovulation. However, these medical regimens might result in the development of more than one ovarian follicle. They are generally more dependent upon the use of hCG and, because of this, necessitate more clinical surveillance to avoid overstimulation of the ovary. The most aggressive protocols, termed controlled ovarian hyperstimulation (COH), are variations of COS, but with a more focused intent of multiple follicle recruitment. Monitoring of COH protocols is quite necessary. These three protocols are the foundation for assisted reproduction protocols such as in vitro fertilization (IVF). The progress of clinical OI over the past 45 years has revolved around protocols that utilize oral agents [7–12] and parenteral agents [11,13,14]. When successful, the former (i.e., clomiphene citrate) do not necessarily depend upon the adjunctive IM injection of hCG; however, reviews of standard treatment plans with clomiphene citrate may include a recommendation for the use of hCG [15] in select instances. Specific clinical indications for hCG use in managing infertility can be gleaned from prior reviews [15–20]. The oldest and most consistently discussed indications are OI (including both clomiphene citrate- and gonadotropin-managed patients) [7,15,19], and as an adjunct for timing artificial insemination procedures, whether using the husband’s or donor sperm samples [21–25]. Less common, secondary indications for use of hCG-component protocols have been in women with unexplained infertility [26], or who have not successfully conceived after surgery or focused medical management of pelvic endometriosis [27]. These secondary categories might involve women who do not necessarily exhibit ovulatory dysfunction, and a partner whose semen samples might be entirely normal, when empirical data suggest that they might otherwise benefit. It should be mentioned that hCG is also used for luteal support in OI protocols, particularly as a component of cycles controlled with hMG [28–30]. Soules et al. noted that it is probably important to maintain threshold levels of plasma LH in the luteal phase of a cycle in order to adequately support luteal function. Physiologically, this is associated with maintaining a concentration of progesterone in the postovulation or postembryo transfer interval of clinical protocols. hCG derived from urinary sources has been variably used for the above indications for more than 30 years [13,31,32]. Publications relating to its earliest employment were not reviewed, but it is evident that by the mid-1970s, administered hCG was being given in doses of 10,000 IU, 5000 IU, or less [14].
174
Human Chorionic Gonadotropin (hCG)
When a woman fails to respond to clomiphene citrate, it is recommended that she consider use of gonadotropin therapy. A general definition of clomiphene resistance implies a failure to ovulate at conventional multiday clomiphene citrate doses (up to 150 or 200 mg/day), or a failure to conceive with an ovulatory dose of clomiphene citrate during the three to six previous cycles [33]. Thus, clomiphene resistance is a common diagnosis that leads to COS. This transition typically requires a timed hCG injection as an integral component of the induction protocol, utilizing menopausal gonadotropins [34–37]. The conventional assumption is that hCG acts directly and independently as the surrogate LH signal; however, studies examining the extent to which an endogenous LH surge is suppressed by gonadotropin induction suggest that it is not entirely suppressed. Given that the LH surge dynamic remains but is otherwise truncated or altered, other studies have examined possible clinical consequences that might result when an ancillary hCG signal is provided in addition to an endogenous LH signal [38]. To summarize, hCG administration is a common adjunct in the contemporary management of human infertility. It can be included in differing protocols involving oral agents; if this first-line therapy is unsuccessful, patients can proceed to gonadotropin therapy. Protocols utilizing exogenous follicle stimulating hormone (FSH) customarily require hCG administration. As hCG injections became more specifically integrated into treatment algorithms, increasing numbers of clinical studies were published. Concomitantly, as steroid and glycoprotein immunoassays became more practically available, and ultrasound technology improved, studies ascertaining the indications for and the utility and practical effectiveness of hCG administration also became more informative. For example, the practical endpoint of determining if ovulation has occurred has been [39] and remains a considerable challenge, whether it is the result of normal hormonal dynamics, or the goal after hCG administration [6,26,40]. Historically, clomiphene OI has been monitored with clinical assessment of hormones [8,19], cervical mucus [41–43], basal body temperature graphing [8,44,45], timed assessment of serum progesterone [15], and eventually transabdominal [39,46,47] and transvaginal ultrasound [28,48]. OI with gonadotropin protocols have been monitored with cervical mucus [49,50], vaginal cytology, cervical mucus changes [49], estrogen levels, and ultrasound [51]. Progressing from imprecise clinical markers like cervical mucus changes [31,49,50] to progressively more accurate methods like timed [39,46] or serial high-resolution transvaginal ultrasound series [52] to determine ovulation has been transformative. As more sophisticated modalities for monitoring OI were developed and incorporated into clinical use, positive and negative predictive accuracy for the event of ovulation improved. As an evolving process, eventually studies were published that could focus on clinical endpoints such as pregnancy rates consequent to variations in stimulation protocols rather than efficacy studies on OI [27,29,53,54]. The following section is a brief review of the use of hCG in the clinical management of reproductive dysfunction. It devotes more attention to the most common indications of OI, COS, and use in artificial insemination protocols.
Use of hCG in Reproductive Dysfunction
175
15.2 C onsiderations of hCG Administration within Clinical Protocols 15.2.1 OI—Oral Agents Ovulatory dysfunction is the first or second most common problem in clinical infertility [55]. A pivotal report was published in 1961 noting effective ovulation with the chemical agent clomiphene citrate, called MRL/41 [13]. This agent, first synthesized in 1956 and approved for use in the United States in 1967 [32], was the basis of more than 2000 publications by 1976. Consensus algorithms using this agent have been published and periodically revised [11,12,15,19,20]. By 1976, the general side effects, risks, and benefits of this agent had been documented and discussed. They included low pregnancy rates, compromise of cervical mucus, follicular luteinization (pseudoovulation), elevated miscarriage rates, and a multiple birth rate of 6–12%. Clomiphene citrate remains the most common oral medication utilized in OI [7,8,12,15,19,36,56]. Collectively, this mixed estrogen agonist/antagonist has been used in various protocols, over a varying number of days [49], and at sequentially higher dosages in nonresponsive patients. More recently, other oral agents, including insulin sensitizing agents [57–59] and aromatase inhibitors [32,60,61], have been employed, either independently or in combination with clomiphene citrate. This chapter will not focus specifically on these agents. By 2003, a Practice Committee Report from the American Society for Reproductive Medicine listed anovulation, luteal phase deficiency, and unexplained infertility as indications for management with clomiphene alone or in combination with other treatment regimens. Because of the increased cost of monitoring, hCG use was cautionary, with the consideration that it be limited to women who require intrauterine insemination (IUI), or in whom a mid-cycle LH surge could not be reliably detected [56]. The most recent publication providing consensus recommendations for OI still lists clomiphene citrate as the initial medication [36].
15.2.2 OI—Parenteral Agents, with hCG The history of nonoral, parenteral agents for ovulation management was initiated in 1959 when hMG was first used for OI in Europe. Four years later, gonadotropins were approved for clinical trials in the United States. hMG was approved for sale in the United States in 1975 [32]. One of the common indications for second-line OI therapy cited in earlier publications was nonresponse or unacceptable side effects of clomiphene citrate [7,20]. hMG were used when patients failed to respond to OI with clomiphene citrate, which resulted in standard protocols covering conventional OI [33,34,37]. Although urinary-derived menotropins had been available since the 1960s, they contained variable activities of FSH, LH, and quantities of potentially allergenic urinary proteins [33]. In the 1980s, more uniformly manufactured products, such as Pergonal®, Humegon®, Urofollitropin®, and Metrodin®, became available. Pergonal
176
Human Chorionic Gonadotropin (hCG)
and Humegon contained fixed activities of FSH and LH; Urofollitropin and Metrodin contained standard dosages of FSH, but with reduced or essentially nil LH activity [62]. They still had a very small amount of urinary proteins. In another 10 years, recombinant FSH (rFSH) products were developed, providing increased purity, absence of urinary proteins, and production consistency. Two preparations of rFSH were developed in the 1990s; Gonal F® (follitropin alpha) became available in 1995, and Puregon® (follitropin beta) became available in 1997. rFSH was first approved for use in the United States the same year. rFSH (1988) was purer, contained no LH or co-purified proteins, had high specific bioactivity, was available independent of urine collection, and boasted absolute source control and batch-to-batch consistency [33,63]. The era of assisted reproduction began with the first IVF births in England (1978) and in the United States (1981) [32]. In retrospect, these conceptions were more exceptional than clinically routine. Subsequently, all contributing factors to success with IVF were actively scrutinized. This included adapting the extant induction protocols that relied on clomiphene or hMG. The need for reliable controlled stimulation protocols for assisted reproductive technology was a strong impetus for the development of the increasingly pure FSH products mentioned earlier. Rather than limited follicle recruitment, the new paradigm was to develop efficient controlled stimulation protocols that would result in consistently high numbers of periovulatory follicles for assisted reproduction [64, 67]. As those same FSH preparations supplanted hMG, they were, out of necessity, incorporated into conventional OI protocols. This resulted in another set of clinical concerns for patients considering more conventional stimulation protocols, because of the increased risks of high-order multiple pregnancies and OHSS [68–74] that were associated with the more aggressive protocols being developed. Some of the information that applies to the timing of hCG injection was developed in the late 1980s as clinical efficacy protocols for controlled stimulation for IVF. These protocols were being clarified and understood [65, 66, 75].
15.2.3 OI Protocols and FSH Preparations Publications appeared that examined and compared the clinical efficacy of the many FSH-containing preparations that were then available (including hMG, urinary FSH (uFSH) products, and rFSH products). In 1984, one publication provided the first study of COS with IUI as an alternative to IVF [76]. The clinical reproductive conditions being addressed with the use of these newer formulations were historically the same: polycystic ovary syndrome (PCOS), clomiphene-resistant OI, adjuncts for artificial insemination, and controlled stimulation or hyperstimulation for unexplained infertility or IVF programs. Entirely new models of OI, however, had to be re-examined in light of the new products. The reader is reminded that the basic formula of controlled stimulation—FSH administration culminating in timed hCG injection—remained the prototypical sequence for clinical studies assessing novel formulations for managing clinical infertility; however, older clinical concerns remained and new concerns were added.
Use of hCG in Reproductive Dysfunction
177
For example, did FSH-only products work as well as products containing both FSH and LH? Did rFSH work differently than urinary-sourced pure FSH? Two representative studies can be mentioned. The availability of pure FSH formulations was expected to benefit the management of patients with PCOS, who often had abnormally high levels of LH. The lack of LH activity had no negative influence on stimulation of PCOS patients [77,78]. Also, a Cochrane Review in 2000 noted that urinary-derived FSH products did not seem to improve pregnancy rates when compared to less-expensive hMG in PCOS patients. The FSH products did, however, reduce OHSS [79]. Another review looked at the efficacy and safety of rFSH versus purified uFSH in patients with PCOS, as well as outcome assessment of either of these preparations compared to more conventional hMG products [33]. Six randomized, controlled trials involving clomiphene-resistant anovulatory women were sufficient for review. The primary outcome of interest was a live birth or ongoing pregnancy per woman. The secondary outcomes included ovulation rate per cycle, ongoing pregnancy rate, live birth, miscarriage rate, incidence of OHSS, incidence of multiple pregnancy, total gonadotropin dose, total duration of stimulation, single-follicle development, and cancellation rate per woman. Chronic low-dose regimens appear to be less likely to cause OHSS when compared to conventional regimens using either uFSH or rFSH. This meta-analysis concluded that “in summary, there is as yet insufficient evidence to conclude that rFSH is more effective than uFSH for OI in women with clomiphene citrate-resistant PCOS.” What must be acknowledged is that hCG was an integral aspect of almost all of the stimulation protocols then being developed that utilized FSH activity. There was no consensus about what dosage of hCG to use, when to administer the medication, or whether more than one injection of hCG was necessary. A sampling of protocol details from the references reviewed for this chapter reinforces this impression. Examples derived from both oral and parenteral protocols include the following synopses. In a relatively early publication on induction of ovulation with menotropins, it was mentioned that up to 25,000 IU of hCG was administered over a 3- to 4-day interval to induce ovulation after deducing sufficient stimulation by serial determinations of total urinary estrogen [80]. One investigator specifically mentioned that 6000–10,000 IU of hCG could be given at mid-cycle to augment the LH surge for luteinized unruptured follicle (LUF) syndrome [19]. Another study examined ovulation response rates in a sample of former clomiphene nonresponders (at 100 mg) when they were stratified to receive either a recommended daily micro dose of 200 IU of hCG given intramuscularly after the largest (ultrasound monitored) follicle reached 12 mm in women who repeated a 100-mg dose of clomiphene. Their responses were compared to ovulation rates in a control group of women, who had also not responded to 100 mg of clomiphene, and who repeated a trial at an increased dosage of 150 mg. The ovulation rates were 57% for the Clomid/hCG group, and 18% for the Clomid-only group [81]. Individualized hMG doses from 75–225 IU/day provided stimulation, and hCG was given at 5000–10,000 IU/day for 2 or 3 days, based on cervical mucus changes [50]. In a clinical trial of rFSH compared to pure uFSH, 10,000 IU of hCG was given when one follicle 18 mm, or two to three follicles 15 mm were seen [82].
178
Human Chorionic Gonadotropin (hCG)
Yarali et al. administered 10,000 IU of hCG when at least one stimulated follicle of at least 17 mm in diameter was noted [83]. Using an induction protocol that began with 150 IU FSH daily but was individually adjusted after ultrasound noting ovarian ultrasound characteristics on cycle day 8, Guzick et al. [25] administered 10,000 IU of hCG when two or more follicles were 18 mm and estradiol levels varied from 500–3000 pg/ml. Similarly, Hedon gave 5000 IU of hCG when a follicle of at least 16 mm was seen [84]. Investigating a rFSH preparation compared to an FSH/LH preparation in a step-up induction regimen, Balasch [85] administered 10,000 IU of hCG when a follicle greater than 17 mm developed. In a study by Fisch et al. [26], patients in two of four randomization arms received either a 100-mg clomiphene citrate cycle on days 5 through 9 and 5000 IU of hCG administered intramuscularly on cycle days 19, 22, 25, and 28; or placebo tablets on days 5 through 9, but hCG injections on the same schedule. One report from 1988 noted that among 52 pregnancies conceived on a specific hMG/hCG protocol, continuing beyond the first trimester was a 27% incidence of multiple gestations. The majority of these multiple pregnancies (9 of 14 or 62%) were twins. The indications for treatment in that study were male factor infertility (sperm density less than 20 million per milliliter, and/or motility less than 40%), abnormalities in cervical mucus or a poor postcoital test, or management of patients receiving donor insemination [86]. When assessed, the data only suggested an intrauterine inseminant density of more than 20 million sperm as correlating with increased multiple pregnancy risk. Parameters that did not significantly correlate included total hMG dose, type or total hCG dose, day of hCG administration, or peak estradiol level. It is interesting to note that, in this study, the timing of hCG administration, usually 10,000 IU, was based upon serial estradiol assessment only. Ultrasound information was not performed per the described protocol. The hCG dose was given 24-h after an estradiol level attained 400 pg/ml, but was less than 1000 pg/ml. Continuing with other examples of hCG inclusion in protocols used in ovulation management for PCOS and unexplained infertility, Tadakoro gave 3000, 6000, or 10,000 IU of hCG as well as providing luteal support with 1500 IU of hCG on days 3, 6, 9, depending upon hormone surveillance [37]. A variation of mid-cycle stimulation was investigated. The authors utilized hMG at 150 IU/day, monitored ovarian response with transvaginal ultrasound until at least one follicle was at least 16 mm in diameter, and randomized patients to receive either 10,000 IU of hCG as a single or divided injection, with the second 5000 IU injection provided 1 week after the initial 5000 IU injection. Data confirmed no significant clinical differences except length of the luteal phase, which was longer in the divided-dose protocol [30]. Luteal support is a variable management step in the OI protocols reviewed [59–61]. Some studies mention specific protocols, such as 5000 IU of hCG when at least one follicle was 17 mm, and serum estradiol consistency (noting on day of determined hCG injection that mean number of follicles 10 mm were 4.7 2.2 and 410 pg/ml 272 pg/ml). The luteal phase was supported with 2500 IU of hCG 4–7 days after the hCG ovulatory injection [70]. In a review of approaches to OI from the early 1980s, the author [20] highlighted use of hCG for gonadotropin protocols, emphasizing an incremental dosing paradigm
Use of hCG in Reproductive Dysfunction
179
for ovulation dose of hCG, derived from the Royal Women’s Hospital. In a first induction cycle, 3000 IU of hCG was given 48 h after last FSH dose. Subsequently, the cycle was monitored by daily estradiol and progesterone values to validate ovulation. If ovulation was determined to have occurred, supplementary doses of hCG (1000 IU on days 9 and 12) following the ovulating dose of hCG were given. If ovulation did not occur in the first cycle, stimulation was repeated, and, if necessary, the initial hCG dose was incrementally increased to 5000, 7500, or 10,000 IU by the fourth cycle. If the 10,000 IU dose was necessary, then supplementary doses were withheld. To summarize, the historical and continuing use of hCG results in its inclusion in a variety of ovulation management protocols. This heterogeneity has occurred because when hCG is being considered as a component of clinical infertility treatment, many interrelated factors have to be taken into account. What is the indication being treated? What is the clinical goal? Is it ovulation of the lowest achievable number of follicles, or maximal stimulation? What medication or combination of medications is being used as the stimulation component? What modalities are available for monitoring and surveillance? What dose will be selected? Will more than one dose be necessary? What are the expected benefits of the management protocol, and how will they be determined? What are the inherent risks associated with the protocol, and how will they be minimized or monitored? What criteria will be used to determine when to administer hCG? Given the confounding variables inherent in clinical studies, the timing of hCG administration has been based upon one or more of the following parameters: consideration of preovulatory serum progesterone, suspicion or early documentation of an incipient LH surge, ovarian follicle size or number, and estradiol levels.
15.3 T iming Administration of hCG—Ultrasound Monitoring, Progesterone Patterns, and Endogenous LH Surge Patterns When ultrasound technology emerged in the early 1980s, it was incorporated into clinical protocols [39,46,87]. A 1981 publication stated that although (transabdominal) ultrasound scan does not supplant estrogen monitoring, it can assist in detection of hyperstimulation, in the more accurate timing of the ovulatory dose of hCG, and in the withholding of hCG. It was recommended that an ultrasound examination be performed on all patients prior to hCG administration [87]. As ultrasound technology advanced, other relevant studies were published. A paper published in 1991 [40] provided data regarding a critical assumption about hCG administration. The purpose was to derive an equation for predicting ovulation, based upon data for correlating ovulation with follicle size on the day of hCG administration in hMG-stimulated cycles. The primary question was: what size follicle (in millimeters) ovulates in response to a timed injection of 10,000 IU of hCG after COH with hMG? In their discussion, the study authors remarked on the thencurrent lack of consensus and the confounding issues related to the understanding
180
Human Chorionic Gonadotropin (hCG)
that follicles reaching 20–25 mm in spontaneous development were likely to ovulate after LH surge induction. None of the previous studies had compared ovulation percentages using induced follicle size as a parameter; it was remarked that oocytes obtained from follicles as small as 16 mm could be fertilized in IVF circumstances. These authors also listed related uncertainties, such as whether a follicle of a certain size would guarantee ovulation, a lack of understanding with regard to the number of ovulated oocytes that was correlated with a chance of pregnancy, and the fact that no optimal E2 range correlating with increased fecundity had been determined (recall that these questions were being raised more than a decade after the use of hCG in many OI protocols; see Dickey in a comprehensive 2009 review) [32]. Patients were stimulated with 150 IU of hMG, initiated daily, and adjusted by protocol per changes in estradiol values and the number of follicle reaching critical sizes. After the hCG injection, ultrasound surveillance was performed to confirm ovulation. No exogenous luteal support was provided. Among 49 patients in 122 cycles, 1344 follicles were studied. Follicles measuring 17 mm or larger on the day of hCG administration ovulated significantly more often than follicles 16 mm or smaller. Follicles measuring 14 mm or smaller rarely ovulated. The average estradiol value on the day of hCG administration was 947 pg/ml. No significant difference in the estrogen value was correlated with cycles resulting in ongoing pregnancy, spontaneous miscarriage, or failed conception. The data from this paper reflect a currentuse paradigm for OI with controlled stimulation.
15.4 Risks of Ovulation Management with hCG Controlled ovulation is the goal, but it must be understood that the preceding induction protocols also carry inherent risks of superovulation, ovarian enlargement, multiple follicles, OHSS, and high-order multiple pregnancies and births [32]. More recent publications are focusing on strategies to minimize these risks. These options include reducing gonadotropin exposure by combining FSH with clomiphene [11], and limiting the dose or the number of days of stimulation via withholding hCG or other maneuvers. One strategy to minimize or avoid OHSS in patients with PCOS was to use lowdose FSH protocols. Formerly, it was unclear if a step-up, step-down, or combination protocol would provide better efficacy and lower OHSS risk. In one well-conducted study, Christin-Maitre [35] examined the benefit of either a step-up or step-down protocol with rFSH in clomiphene citrate-resistant patients with PCOS. The study noted similar cumulative rates of gestation: 38.6% versus 30.8% in the step-up versus the step-down protocols, respectively. The latter protocol, however, was more efficient in promoting monofollicular development and ovulation and had a lower rate of hyperstimulation, but a longer duration of treatment seemed necessary. Details of the protocol included giving 5000 IU of hCG as a single dose when the leading follicle was more than 18 mm; hCG was withheld if four or more 16 mm follicles were present and/or the estradiol value reached 1000 pg/ml or higher.
Use of hCG in Reproductive Dysfunction
181
The protocol also stated specifically that the luteal phase was not supported. The data from this study reported that ovarian follicles exceeding 16 mm were noted in 83% of the inductions with the step-up protocol, compared to only 56% in the step-down protocol. Also, multiple follicles (more than three in number) reaching more than 16 mm were noted in only 4.7% of the step-up cycles, compared to 36% of the stepdown cycles. These favorable comparisons resulted in the administration of hCG in 84.6% of step-up cycles, compared to only 61.8% of step-down cycles. These results suggest that continuing to modify induction protocols can result in risk reduction for multiple follicle recruitment. It was noted that most clinical induction protocols rely on estrogen and ultrasound follicle characteristics to help timed hCG administration. Other studies also examined the confounding variable of endogenous LH secretion. There is debate over whether superovulation protocols inhibit or modify the endogenous LH surge. During the 1980s, when controlled hyperstimulation protocols for IVF programs were studied, the problems of premature luteinization and/or progesterone secretion both compromised success rates. The data derived from clinical assisted reproductive technologies (ART) studies help contribute to an understanding of the larger issue of timing of hCG for non-ART induction protocols; specifically, do endogenous LH levels or progesterone levels influence a patient’s response to an hCG injection? The relationship of hCG administration relative to the LH surge has not been extensively studied. An interesting retrospective data analysis of pregnancy rates following IUI in women who underwent the procedure in 49 unstimulated cycles and in 856 consecutive hMG-stimulated cycles is, however, relevant. Group A had an endogenous LH surge and was not given hCG, Group B was given hCG after an LH surge, and Group C received hCG before an LH surge. Pregnancy rates provided were 16% per cycle; however, the lowest rate was in Group A and the highest was in Group B. The authors concluded that the better pregnancy rate occurs when hCG is given after an LH surge. Their protocol involved hMG at 75 IU of FSH, adjusted as necessary until one to two follicles 16 mm in diameter were detected. With increased hormonal surveillance, LH, and progesterone, 5000 IU of hCG was given under selective conditions per protocol. This related to presence or absence of LH and/or a progesterone shift. Timing of IUI was contingent upon specific administration of hCG. Luteal support was provided: 1000 IU every 3 days for three doses. The analysis of the results suggested that timed administration of hCG following an LH surge could statistically improve cycle pregnancy rate, particularly if hCG was given up to 20 h after detection of the surge and if IUI could be completed within 48 h of the hCG injection [38]. The suggestion that hCG superimposed upon a spontaneously manifested LH surge improved the chances of pregnancy was noted in another publication by Mitwally et al. [60]. In this review of retrospective data, the authors examined the influence of hCG administration in relationship to detected LH surge as a component of pregnancy success in cycles involving ovarian stimulation. First, they concluded that hCG administration was associated with a favorable outcome during stimulation in general. Higher clinical pregnancy rates associated with hCG administration were noted in a variety of stimulation protocols, sperm exposure models (timed intercourse or IUI) of presumed causes of infertility, specifically unexplained infertility,
182
Human Chorionic Gonadotropin (hCG)
or PCOS. In addition, they reported a practical distinction regarding the timing of the hCG injection being pertinent to the stimulating protocol. If clomiphene citrate were used, better pregnancy results were achieved when the LH surge was monitored and hCG was given subsequent to the surge. In FSH-driven protocols, monitoring and using of an LH surge parameter for hCG administration was not as successful as administering hCG based upon more common parameters involving estradiol and ultrasound criteria.
15.5 Efficacy of LUF Syndrome The central question regarding LUF syndrome is whether luteinization of a follicle cyst can occur in the absence of a released oocyte from the mature Graafian follicle [88]. The phenomenon was first considered in 1935 and examined again after 1966 [91,93]. Components of the diagnosis presupposed that the patient showed evidence of failure to ovulate after maturing a follicle, yet exhibited manifestations of corpus luteum function (primarily progesterone secretion) with variable clinical characterization of the luteal phase. The prevalence of this diagnosis remains controversial [91,94,95]. Issues relate to the accuracy with which the condition might be diagnosed; its persistence; and whether it is associated with specific conditions such as endometriosis, unexplained infertility, or OI [89,91]. Relevant points in the discussion about the possible etiologies of LUF also consider deficient LH surge dynamics and concomitant variations in the FSH surge associated with the LH surge [92,95] and its treatment [89,95]. Studies have surveyed the possible occurrence of this syndrome with technology available at the time. Early incidence studies used combinations of assessments of peritoneal fluid concentrations of ovarian steroids, basal body temperature graphing, and serum progesterone levels [88,91]. Hamilton et al. published a study in which they prospectively monitored 600 cycles among 270 infertile patients every other day from cycle day 8 onward, and then with daily ovarian ultrasound when an ovarian follicle exceeded 15 mm in diameter. Ultrasound scanning continued until follicle dynamics met criteria consistent with ovulation, or persisted in size to 30 mm or greater for 3 consecutive days. A progesterone measurement was obtained on each day of ultrasound. The data from this study suggested that approximately 10% of the women and more than 6% of cycles exhibited a follicle growth pattern consistent with failure to rupture by ultrasound, but, with suggestion of luteinization, manifested as shifts in basal temperature patterns and increases in luteal phase progesterone. This pattern was noted in 14 of the 27 women who received medical management during the index cycle producing the data. The medications used for cycles with apparent ovulation failure included clomiphene citrate, alone or in combination with hCG, hMG, or LH-releasing hormone [92]. This well-executed study was the first to characterize the follicular growth pattern of LUF cycles compared to those in normal ovulatory cycles. The growth rates were comparable until the serum-detected LH surge, at which time follicle rupture occurred in the 45 control cycles. In contrast, the LUF pattern was marked by a rapid increase in follicle diameter (reaching 35–36 mm on average) and persisting through
Use of hCG in Reproductive Dysfunction
183
the remainder of the cycle. Additionally, LH levels from LUF cycles surged 2 days later and were ultimately lower than LH values compared to values from normal cycles (despite the fact that the clinical luteal phases were similar in length). The mid-luteal progesterone levels were significantly lower when obtained from cycles exhibiting a LUF pattern. Ovulation documentation has been as direct as laparoscopic assessment of ovarian stigmata [88], with many related studies substituting persistent or enlarging ovarian cyst formation beyond the window of assumed ovulation [88,89]. Based upon data suggesting that concentrations were significantly higher in women who actually ovulated versus those women who did not apparently ovulate, one study proposed substituting estradiol and progesterone concentration values derived from peritoneal fluid obtained within 5 days of suspected ovulation. The data presented suggested overlap in concentrations in peritoneal hormone levels, fluid volume, and protein concentrations from women with and without laparoscopically determined ovulation stigmata, as well as from some women with no evidence of persistent ovarian cysts [88]. In an early study, Check et al. [89] monitored ovarian folliculogenesis via ultrasound in 333 patients. Of these, 89 failed to release an ovum by ultrasound criteria in two consecutive cycles. Relevant criteria included documenting when a mature follicle attained 18–24 mm in diameter, monitoring whether the index follicle decreased in size by 5 mm, free fluid appearing in the cul-de-sac, and cervical mucus regressing and serum progesterone increasing at least 3 ng/ml. Free fluid was variable, and it was not clear if all patients had progesterone assessments. The 89 patients considered to have LUF on two occasions were derived from three groups. Group 1 was not on any medications (n 39). Group 2 was given clomiphene citrate (n 17). Group 3 was treated with hMG (n 33). They were subsequently managed per protocol to induce successful ovulation. They were first treated with 10,000 IU of hCG alone. If unsuccessful, a repeat cycle was managed with 15,000 IU of hCG and 150 IU of hMG. In Group 1, 35 of 39 patients provided ultrasonographic evidence of ovulation. In Group 2, 13 of 16 patients provided ultrasonographic evidence of ovulation. In Group 3, 31 of 33 patients ovulated when specifically given the higher dose of hCG and hMG after failing to ovulate on hCG alone. No control groups were provided, and the authors acknowledged the lack of a placebo arm. In a later publication from the same author [95], the assessment and treatment of LUF were again explored with more comprehensively monitored patients. LUF was characterized by two patterns, both involving prospective monitoring of estradiol, progesterone, and LH starting 17 days before expected menses and continuing until the serum progesterone was greater than 5 ng/ml. Daily ultrasounds were obtained on the same days. LUF was considered when a lead follicle failed to demonstrate a 5-mm reduction in average diameter within 72 h of maturity, and corroborated with LH surge, a drop in estrogen, and an increase in progesterone to above 2.5 ng/ml on two consecutive cycles. Two patterns of LUF were described: a mature-follicle type and a prematureluteinization type. The former type manifested a follicle achieving 18–24 mm average diameter, and estrogen of more than 200 pg/ml while the progesterone remained under 2.5 ng/ml. Subsequently, the follicle size did not diminish and the mid-luteal
184
Human Chorionic Gonadotropin (hCG)
phase progesterone reached a minimum of 8 ng/ml. In the latter type, a follicle attained a diameter of more than 18 mm, but progesterone increased to more than 2.5 ng/ml and the estradiol level never reached 200 pg/ml. As in the former group, the follicle size did not decrease and the mid-luteal phase progesterone value exceeded 8 ng/ml. Patients in both ovulation and non-OI protocols with LUF syndrome were then treated, in sequential cycles if not responding, with triggering doses of 10,000 IU of hCG; 15,000 IU of hCG; 15,000 IU of hCG; and 150 IU of hMG; or a divided-dose treatment of 1000 IU of hCG followed by 9000 IU of hCG 24 h later. Failure to respond resulted in transfer of patients to an active OI arm, sequencing variations of combinations of clomiphene citrate, then serial hMG and the hCG permutations employed in the non-OI arm. Again, there were no controlled cycles. It was noted that there were no conceptions in the 168 cycles in which LUF was noted, as opposed to 24 conceptions in the 198 (12.1%) of cycles that appeared to be corrected with the strategies employed in this study [95]. A related study focused on whether the release efficacy of ova by hCG in hMGtreated cycles was critically related to the late follicular phase progesterone level. In the discussion, the authors pointed out that one of the proposed functions of early progesterone production is to enhance the activity of certain prostaglandins associated with follicular rupture. Because hCG is administered in menotropin OI cycles based upon estrogen levels and follicle size criteria, the authors wondered if critical levels of progesterone had to be evident prior to administration of hCG. Based upon data assessment of progesterone levels in discrete intervals of 0.1–0.5, 0.6–0.9, 1.0–1.5, and greater than 1.6 ng/ml, they concluded that rates of ovulation did not show significant differences following hCG administration of 10,000 IU of hCG (related to the existing pre-injection progesterone level) [95]. The incidence of LUF from different studies varies from 6.7% [92], to 10.7% in spontaneous cycles [96], to more than 25% in some earlier studies [91]. Without being able to control for a variety of factors, the discussions about LUF remained focused on etiologies: whether this condition is repetitive and whether it has higher associations of occurrence with conditions such as unexplained infertility, endometriosis, or pelvic inflammatory disease. In studies that did not use serial ultrasound methods to prospectively monitor follicle dynamics, indirect ascertainment of ovulation (particularly in studies utilizing ovulation management) might provide confounding data on efficacy of response, and, indirectly, pregnancy rates [91,95]. The reasons for including a brief review of this syndrome are, first, that it is germane to patients who might need to be treated with OI protocols; and second, because it might be more important to incorporate postinjection surveillance for documenting the timing and occurrence of ovulation.
15.6 Considerations of hCG Administration for Timing IUI IUI is probably the least standardized infertility treatment. Variables reported to be important for clinical management include different treatment groups and population composition [22], age, clinical etiologies and indications for the procedure [50], previous diagnostic workup, previous treatments, suspicion of cervical mucus
Use of hCG in Reproductive Dysfunction
185
abnormalities [97], cycle management [98], ovarian stimulation, ovarian monitoring, IUI timing, number of inseminations per cycle [99–102], associated treatments in the male and female partner, and number of treatment cycles to be performed [23,102– 106]. The specific methods for sperm collection, storage, and laboratory preparation for IUI [107–110] additionally confound purported endpoints, independent from the stimulation protocol, and the timing determinants for the insemination itself. Given these clinical variables, and the acknowledged heterogeneity of most studies [23,102–111], the focus of an insemination procedure still remains the identification of the day of suspected ovulation [112]. Earlier in this chapter we reviewed the use of methods for urinary LH determination, as well as the elective use of injections of hCG associated with OI and COS. Both of these modalities were seamlessly applied to insemination scheduling. Urinary LH monitoring was rapidly embraced as an adjunct to timing of artificial insemination procedures [53,113]. Elective injection of hCG was used in gonadotropin-controlled stimulation protocols for timing IUI [25,38,54,60,114]. The indications for IUI as first-line therapy include unexplained infertility, male infertility, and ovulatory dysfunctions [115]. Whether an ovulation management protocol is utilized as an adjunct to the timed procedure is variable. These protocols have included both clomiphene citrate [27,52,115,116] and controlled stimulation with gonadotropins [38,54,85,114]. Clomiphene citrate administration has been an initial step in many management strategies for IUI. Clomiphene use results in ovulation during a window of time that cannot be precisely estimated. Elsewhere in this chapter we have commented on the variables of folliculogenesis, preovulatory follicle diameters, and spontaneous LH surge chronology, each of which results in the estimation that the actual event of ovulation can occur 24–56 h after a detected LH surge [53]. It is well understood that detection of the LH surge by urinary detection kits is an independent confounder in identifying the time interval for insemination. Components of this difficulty include variability of the pituitary secretion pattern in a given 24-h interval; rate of urine production (which affects concentrations of LH in urine upon which a semi-quantitative enzyme-linked immuno-spectrometric assay (ELISA) test depends); the inconsistency of detection between early morning, mid-day, or evening assessment among available detection kits; and the interference of augmented LH release when certain patients utilize clomiphene citrate. The use of hCG as a timing agent addresses some of the variance associated with endogenous LH release patterns, but also introduces other considerations. One important consideration is the need for ovarian follicle surveillance to determine the timing of the injection. Although an injection can be empirically timed, one or more ultrasound procedures usually provide a better estimate of when to administer hCG. Follicles have predictable growth trajectories, supporting the possibility of empiric timing. Many carefully done studies, however, have noted ranges in the numbers of follicles starting and completing development (endpoint diameters associated with inducing spontaneous LH surges) that can critically influence when to proceed [38,48,115]. Once the injection is given, the expected time of ovulation has less variance than that associated with spontaneous LH surges; perhaps 36–48 h [115]. In a study of IM versus IV hCG administration after clomiphene OI, Fischer et al. monitored the time interval for ovulation, estrogen, and progesterone values. They noted a mean ovulation time of 40.4 h
186
Human Chorionic Gonadotropin (hCG)
after IM and 38.3 h after IV administration. The range among responses in both groups was less than 36 h and more than 48 h [6]. Anderson et al. noted a mean time to ovulation of 38.3 h, which they derived from a range of 34–46 h among 37 cycles. They also reported that in 66% of cases, the largest follicle was the first to rupture. Three similarly designed studies were published in 1997, 1999 [53], and 2006 [52]. They examined the pregnancy rates in women receiving clomiphene citrate induction who subsequently underwent insemination based upon either urinary LH determination or hCG injection. The protocols were similar, and the two later studies are reviewed here in detail. Deaton et al. published a retrospective review of results for patients with unexplained, anovulatory, or male factor infertility. Among all three clinical categories, 250 were conducted with LH timing and 182 were conducted with hCG timing of a single insemination. There was no significant difference in the clinical pregnancy rates when each clinical category was compared. In 1999, Zreik et al. published the results of a prospective trial in which women with unexplained infertility, ovulatory dysfunction, or male factor infertility, and who were being treated with clomiphene citrate, underwent IUI after being randomized to two different protocols to facilitate insemination timing. The two protocols were urinary LH determination versus hCG injection. Data were derived from 141 cycles performed in 54 couples. Clomiphene was administered on cycle days 3 through 7. Daily LH monitoring began on day 10, and two daily inseminations were performed once a urinary LH surge was detected. An ultrasound was performed once on the day of the first insemination in this group. In the other study group, ultrasound monitoring was conducted. When a lead follicle was 18 mm, 10,000 IU of hCG was administered. Two inseminations were completed beginning the following day. If a pregnancy did not occur, the patient crossed over to the alternate protocol for the next cycle. This continued for up to four cycles. In this study, unexplained infertility included patients who had undergone laparoscopy or had minimal endometriosis. The two protocols used for timing IUI yielded similar pregnancy rates for the three diagnostic groups [53]. Follicle sizes were similar between the LH groups: 22.5 mm compared to 21.1 mm on the day women in the second group were given hCG. The authors stated that 12% of normal ovulatory women might have had a urinary LH peak of less than 20 IU/ml, below the threshold of many urinary LH detection kits. In 2006, Lewis et al. [52] also published a prospective, randomized trial comparing pregnancy rates with two different methods of timing IUI in women who were utilizing clomiphene citrate. One group timed the insemination procedure the day after detecting a positive urinary LH surge. The other group received 10,000 IU of hCG when criteria were met for: endometrial development (at least 8 mm thickness with at least partial layering), acceptable follicle size (20 mm mean diameter), and acceptable follicle number (less than four follicles above 20 mm diameter). The injection was given in the evening and the insemination procedure was completed 33–42 h later. Patients acceptable for study utilized 100 mg oral clomiphene citrate on days 5–9. All subjects underwent vaginal ultrasound each study cycle prior to starting clomiphene. Women randomized to the LH detection group self-monitored urinary LH beginning on day 12.
Use of hCG in Reproductive Dysfunction
187
Women in the hCG group were seen on day 12 for transvaginal ultrasound. Repeat scans were scheduled according to initial follicle dimensions. Patients not meeting criteria for hCG administration in a specific cycle self-monitored urinary LH. Patients remained in their treatment group assignment for three cycles. Subjects in the surge group were dropped if they failed to detect a surge. Subjects who failed to meet hCG injection criteria and failed to detect an LH surge were dropped. This study is noteworthy because it is clinically relevant, and the protocols within the study are similar to common current office practices. The major indication for treatment was unexplained infertility, with a minor proportion of male factor infertility. Of the randomized 150 patients, 129 completed at least one cycle of treatment (58 in the LH surge group; 71 in the hCG group). This resulted in 17 and 23 pregnancies, respectively. A preponderance of pregnancies occurred in the first treatment cycle. Both by intention-to-treat and cumulative pregnancy rate analyses, there was no significant difference in pregnancy rates. Study dropout rates were higher for the LH group: 31% compared to 11% for the hCG group. Failure to detect an LH surge was the most common contributor to noncompletion of a cycle in the home-monitoring group. The authors concluded that ultrasound monitoring with hCG administration did not seem to result in a higher pregnancy rate when compared to home urinary LH monitoring; however, the failure to detect a surge (commented upon in other studies) might be one reason to consider hCG administration [52,56]. Thus, the data from retrospective and prospective controlled studies [52,53] suggest that hCG timing does not appear to increase clinical pregnancy rates as a result of timed insemination in clomiphene-induced ovulation when compared to LH monitoring. Alternatively, Deaton et al. noted that ultrasound-timed injection of hCG after clomiphene induction, followed by IUI, was more efficacious than cycles concluding with clomiphene and hCG-timed intercourse [27]. A more recent meta-analysis [115] of seven studies meeting criteria for clomiphene treatment, IUI, and use of urinary versus hCG-timed IUI procedures was conducted. The data were reviewed separately for different indications: male factor, unexplained infertility, ovulatory dysfunction (the same clinical triad treated in previously mentioned studies). Overall, 1461 patients received hCG and 1162 served as controls. Preparation steps for the semen samples were variable; clomiphene citrate dosages varied from 50 to 150 mg/day among studies. The dose of hCG administered was consistently 10,000 IU among studies. Inseminations were conducted at 18–24 h in the LH monitored groups, and at 36–42 h in the hCG group. Criteria for hCG administration (requiring a lead follicle of 20 mm) were mentioned in four studies. From this meta-analysis, the authors reported that there was a significant difference in pregnancy rates between the hCG-administered group and the LH-recording group that favored LH monitoring [115]. This interpretation was qualified and it pertained only to the groups treated for male factor and unexplained infertility. An opposite, improved trend was described for pregnancy rates resulting from hCG administration rather than LH timing for the clinical indication of ovulatory dysfunction. Given that hCG can be used for induction protocols, one must keep in mind the relative benefits, uncertainties, costs, and risks of its use in contrast to the same considerations for urinary LH monitoring. The advantages of selectively timed
188
Human Chorionic Gonadotropin (hCG)
administration of hCG include a proven biological basis, ease of planning, and clinical predictability [115]. The disadvantages include the cost of the medication and the need for ultrasound monitoring. If IUI is being considered as the first-line approach for couples with unexplained infertility, ovulatory dysfunction, or male factor infertility, the use of LH kits to time IUI could produce important cost savings over the use of ultrasonography [53]. The specific advantages of using urinary LH detection for timing include its costeffectiveness and the fact that it can be used in a private setting [115]. The disadvantages are that the associated false positive and negative rates lead to suboptimally timed procedures, or skipped or cancelled cycles because of nondetection in a truly ovulatory cycle [52]. Because IUI might be an isolated treatment recommendation or an adjunct to both OI and COS protocols [104,114], a derivative set of clinical risks, as well as anticipated benefits, are encountered when multiple follicle development occurs followed by hCG-timed IUI. The primary benefit is a higher pregnancy rate per treatment cycle. Some studies did not confirm this advantage, however. In 1991, Martinez et al. studied 48 patients with male factor or unexplained infertility; they underwent either timed intercourse or timed IUI after hMG stimulation with hCG injection. No significant difference in pregnancy rates between the two options of sperm exposure was detected [114]. Other studies and reviews confirmed an improved chance of pregnancy [37,68]. The derivative risks of controlled stimulation over OI include persistent ovarian cyst formation [40,90], ovarian hyperstimulation [73], and creation of highorder multiple pregnancies [86,32,37]. One study [40] utilized estradiol assays and transvaginal sonography to monitor ovarian follicle size, number, and response to 10,000 IU of hCG administration after controlled stimulation with hMG. The primary purpose of the study was to develop a prediction model for the likelihood of conception. Data derived from the study, however, suggest a high rate of ovarian cyst persistence after controlled stimulation. Fifty-four previously ovulating but nonconceiving patients presented for a consecutive stimulation cycle. It was noted that residual ovarian cysts were found in 28 (51.9%) of those returning patients. Interestingly, data from this study suggested that the likelihood of conception did not correlate with the number of ovulatory follicles, the total number of follicles present at time of hCG administration, or the total hMG dose administered. There was, however, a correlation with serum E2 levels on the day of hCG administration. The continuum of problems associated with hMG/hCG protocols, ranging from persistent cyst formation to iatrogenic OHSS, is an important risk in controlled stimulation protocols, but is not independently addressed in this chapter.
15.7 High-Order Multiple Pregnancies The third relevant risk in controlled stimulation or timed insemination is that of multiple follicle recruitment and high-order multiple pregnancy creation [32]. Use of hCG is now an integral component of various fertility-promoting strategies. A brief
Use of hCG in Reproductive Dysfunction
189
chronologic progression synopsis of the development and incorporation of gonadotropin preparations into clinical protocols has been published [32]. The review provides a cogent discussion about the contribution of non-ART to high-order multiple pregnancies and births that occur as a result of the modern era of OI. For example, it was estimated that approximately 230,000 OI cycles were conducted in the United States in 2003, and that 22% of all twin, 40% of all triplet, and 71% of all higherorder pregnancies in 2004 were the result of controlled stimulation protocols. As a specific example, one publication reported a 27% incidence of multiple gestations (14 of 52) conceived within an hMG/FSH protocol. The majority of conceptions (62%) were twins [86]. Dickey articulated possible factors contributing to the disproportionate increase in multiple pregnancies over the natural background rate. In his opinion, this increase is the result of more clinicians trained to use gonadotropins, more potent gonadotropins, and the adaptation of dosing regimens developed for IVF for more conventional OI–IUI protocols. Publications identify the conundrum of how gonadotropin doses associated with assisted reproduction are being employed in empiric controlled stimulation protocols with IUI, possibly because of higher pregnancy rates [25,54,98,106] relative to clomiphene citrate-based protocols [37,53], adding to the risk of high-order multiple pregnancies. Patient and cycle factors that affect the risk of multiple pregnancies have been identified [32,71,86]. Women under the age of 32 have an increased risk of multiple conceptions, and women over 38 years of age have a significantly reduced risk [32]. Physiological responses to gonadotropins (manifested as elevated periovulatory estradiol levels and increased number of follicles) have been suggested based upon univariate and multivariate analyses in prospectively modeled studies [71,72]. Another set of variables (possibly promoting increased conception numbers) is introduced when IUI is combined with OI [102]. These confounding variables include the indication for treatment (mild male infertility, cervical factor, or unexplained infertility), the stimulation protocol, the criteria for follicle development (number desired, cycles cancelled, converted to IVF), quality of sperm preparation, number of cycles per patient, and outcome focusing on final live birth rate [111]. Other studies relevant to the contribution of the inseminant itself have been published. Only the study by Shelden [86] was positive in concluding that increased sperm density might be associated with increased risk of multiple pregnancies. Others, focusing on fresh versus frozen semen samples, motile count [107,108], and motility longevity [109] did not conclude that these particular factors significantly increased the risk of multiple conception. There is a paucity of controlled trials that comprehensively account for this large set of confounders [111]. Given the above, a consensus goal still remains: to reduce the incidence of multiple pregnancies as a result of the clinical protocol for IUI [32,86]. Strategies to reduce clinical risk of multiple pregnancies have been offered. One review listed three points of focus to help reduce multiple births due to OI. Maximizing the potential effectiveness of first-step treatment algorithms using oral medications (including antiestrogens and aromatase inhibitors) was paramount. When gonadotropin protocols became necessary, it was proposed that the lowest effective doses be used
190
Human Chorionic Gonadotropin (hCG)
[32,35,61,66,68,71,72]. Dickey reiterated that, historically, early use of hMG relied upon daily doses of 75 IU of FSH activity; these met with good success. He particularly emphasized clinical data that reinforced the safety and efficacy of low FSH dose strategies (based upon a review of the pregnancy rate outcomes associated with different protocol doses, estrogen levels, duration of stimulation, number of cycles conducted), and the effect of maternal age (often most pertinent to women less than 32 years of age). A second recommendation was to have contingencies in place to deal with unacceptable multiple follicle development, including cycle cancellation or conversion to IVF. A third recommendation, in sequence, was whether selective pregnancy reduction was available [3].
References [1] O’Connor JF, Birken S, Lustbader JW, Krichevsky A, Chen Y, Canfield RE. Recent advances in the chemistry and immunochemistry of human chorionic gonadotropin: impact on clinical measurements. Endocr Rev 1994;15:650–83. [2] Midgeley AR, Jaffe RB. Regulation of human gonadotropins: II. Disappearance of human chorionic gonadotropin following delivery. J Clin Endocrinol Metab 1968;28:1712–18. [3] Rizkallah T, Gurpide E, Vande Wiele RI. Metabolism of HCG in man. J Clin Endocrinol Metab 1969;29:92–100. [4] Saal W, Glowania H-J, Hengst W, Happ J. Pharmacodynamics and pharmacokinetics after subcutaneous and intramuscular injection of human chorionic gonadotropin. Fertil Steril 1991;56:225–29. [5] Wehmann RE, Nisula BC. Metabolic and renal clearance rates of purified human chorionic gonadotropin. J Clin Invest 1981;68:184–94. [6] Fischer RA, Nakajima ST, Gibson M, Brumsted JR. Ovulation after intravenous and intramuscular human chorionic gonadotropin. Fertil Steril 1993;60:418–22. [7] Wolf LJ. Ovulation induction. Clin Obstet Gynecol 2000;43:902–15. [8] Wu CH. Monitoring of ovulation induction. Fertil Steril 1978;30:617–30. [9] Weseley AC, Melnick H. Tamoxifen in clomiphene-resistant hypothalamic anovulation. Int J Fertil 1987;32:226–8. [10] Tajima C. Endocrine profiles in tamoxifen-induced conception cycles. Fertil Steril 1984; 42:548–53. [11] Kemmann E, Jones JR. Sequential clomiphene citrate-menotropin therapy for induction or enhancement of ovulation. Fertil Steril 1983;39:772–9. [12] Lobo RA, Gysler M, March CM, Goebelsmann U, Mishell DR. Clinical and laboratory predictors of clomiphene response. Fertil Steril 1982;37:168–73. [13] Greenblatt RB, Barfield WE, Jungck ED, Ray AW. Induction of ovulation with MRL/41: preliminary report. J Am Med Assoc 1961;178:101–4. [14] Schwartz M, Jewelewicz R. The use of gonadotropins for induction of ovulation. Fertil Steril 1981;35:3–12. [15] Hammond MG. Monitoring techniques for improved pregnancy rates during clomiphene ovulation induction. Fertil Steril 1984;42:499–509. [16] Keye Jr WR, Chang RJ, Rebar RW, Soules MR. Infertility evaluation and treatment. Philadelphia: WB Saunders Co; 1995. pp. 58, 140–141, 130. [17] Bruno L, Vaclav I, Glezerman M. Diagnosis and treatment of functional infertility. 3rd revised ed. Berlin: Blackwell Wissenschaft; 1993.
Use of hCG in Reproductive Dysfunction
191
[18] Lawrence A, Healy D. Management of women with chronic anovulation. In: Kovacs G, editor. The subfertility handbook. A clinician’s guide, Cambridge UK: Cambridge University Press; 1997. p. 69–82 [Chapter 6]. [19] Wu CH. A rational and practical approach to clomiphene therapy. Clin Obstet Gynecol 1984;27:953–65. [20] Pepperell RJ. A rational approach to ovulation induction. Fertil Steril 1983;40:1–14. [21] Andersen AG, Als-Nielsen B, Hornes PJ, Andersen LF. Time interval from human chorionic gonadotropin (HCG) injection to follicular rupture. Hum Reprod 1995;10:3202–5. [22] Shenfield F, Doyle P, Valentine A, Steele SJ, Tan S-L. Effects of age, gravidity and male infertility status on cumulative conception rates following artificial insemination with cryopreserved donor semen: analysis of 2998 cycles of treatment in one centre over 10 years. Hum Reprod 1993;8:60–4. [23] Bensdorp AJ, Cohlen BJ, Heineman MJ, Vandekerckhove P. Intra-uterine insemination for male subfertility. Cochrane Database Syst Rev 2007;4. CD000360. [24] Cantineau AE, Cohlen BJ, Heineman MJ. Ovarian stimulation protocols (anti-oestrogens, gonadotropins with and without GnRH agonists/antagonists) for intrauterine insemination (IUI) in women with subfertility. Cochrane Database Syst Rev 2007;18:CD005356. [25] Guzick DS, Sullivan MW, Adamson GD, Cedars MI, Falk RJ, Peterson EP, et al. Efficacy of treatment for unexplained infertility. Fertil Steril 1998;70:207–13. [26] Fisch P, Casper RF, Brown SE, Wrixon W, Collins JA, Reid RL, et al. Unexplained infertility: evaluation of treatment with clomiphene citrate and human chorionic gonadotropin. Fertil Steril 1989;51:828–33. [27] Deaton JL, Gibson M, Blackmer KM, Nakamima ST, Badger GJ, Brumsted JR. A randomized, controlled trial of clomiphene citrate and intrauterine insemination in couples with unexplained infertility or surgically corrected endometriosis. Fertil Steril 1990;54:1083–8. [28] Erdem A, Erdem M, Atmaca S, Guler I. Impact of luteal phase support on pregnancy rates in intrauterine insemination cycles: a prospective randomized study. Fertil Steril 2009;91:2508–13. [29] Pritts EA, Atwood AK. Luteal phase support in infertility treatment: a meta-analysis of randomized trials. Hum Reprod 2002;17:2287–99. [30] Grazi RV, Taney FH, Gagliardi CL, Von Hagen S, Weiss G, Schmidt CL. The luteal phase during gonadotropin therapy: effects of two human chorionic gonadotropin regimens. Fertil Steril 1991;55:1088–92. [31] Marshall JR, Jacobson A, Hammond CB. Dose response relationships of ovulation induction with human menopausal gonadotropin. J Clin Endocrinol Metab 1969;29:106–10. [32] Dickey RP. Strategies to reduce multiple pregnancies due to ovulation stimulation. Fertil Steril 2009;91:1–17. [33] van Wely M, Bayram N, van der Veen F. Recombinant FSH in alternative doses or versus urinary gonadotrophins for ovulation induction in subfertilituy associated with polycystic ovary syndrome: a systematic review based on a cochrane review. Hum Reprod 2003;18:1143–9. [34] van Santbrink EJP, Fauser BCJM. Urinary follicle-stimulating hormone for normogonadotropic clomiphene-resistant anovulatory infertility: prospective, randomized comparison between low dose step-up and step-down dose regimens. J Clin Endocrinol Metab 1997;82:3597–602. [35] Christin-Maitre S, Hugues JN. A comparative randomized multicentric study comparing the step-up versus step-down protocol in polycystic ovary syndrome. Hum Reprod 2003;18:1626–31.
192
Human Chorionic Gonadotropin (hCG)
[36] Thessaloniki ESRM/ASRM-sponsored PCOS Consensus Workshop Group. Consensus on infertility treatment related to polycystic ovary syndrome. Hum Reprod 2008;23:462–77. [37] Tadakoro N, Vollenhoven B, Clark S, Baker G, Kovacs G, Burger H, et al. Cumulative pregnancy rates in couples with anovulatory compared with unexplained infertility in an ovulation induction programme. Hum Reprod 1997;12:1939–44. [38] Fuh KW, Wang X, Tai A, Wong I, Norman RJ. Intrauterine insemination: effect of the temporal relationship between the luteinizing hormone surge, human chorionic gonadotropin administration and insemination on pregnancy rates. Hum Reprod 1997;12:2162–6. [39] Seibel MM, McArdle CR, Thompson IE, Berger MJ, Taymor ML. The role of ultrasound in ovulation induction: a critical appraisal. Fertil Steril 1981;36:573–7. [40] Silverberg KM, Olive DL, Burns WN, Johnson JV, Groff TR, Schenken RS. Follicular size at the time of human chorionic gonadotropin administration predicts ovulation outcome in human menopausal gonadotropin-stimulated cycles. Fertil Steril 1991;56:296–300. [41] Bateman BG, Nunley Jr WC, Kulp LA. Exogenous estrogen therapy for treatment of clomiphene-citrate-induced cervical mucus abnormalities: is it effective? Fertil Steril 1990;54:577–9. [42] Moghissi KS. Cervical mucus changes and ovulation prediction and detection. J Reprod Med 1986;31:748–53. [43] Maxson WS, Pittaway DE, Herbert CM, Garner CH, Wentz AC. Antiestrogenic effect of clomiphene citrate: correlation with serum estradiol concentrations. Fertil Steril 1984;42:356–59. [44] de Mouzon J, Testart J, Lefevre B, Pouly J-L, Frydman R. Time relationships between basal body temperature and ovulation or plasma progestins. Fertil Steril 1984;41:254–8. [45] McCarthy JJ, Rockette HE. Prediction of ovulation with basal body temperature. J Reprod Med 1986;31:742–7. [46] Queenan JT, O’Brien GD, Bains LM, Simpson J, Collins WP, Campbell S. Ultrasound scanning of ovaries to detect ovulation in women. Fertil Steril 1980;34:99–105. [47] Batzer FR. Ultrasonic indices of ovulation. J Reprod Med 1986;31:764–9. [48] Baerwald AR, Walker RA, Pierson RA. Growth rates of ovarian follicles during natural menstrual cycles, oral contraception cycles, and ovarian stimulation cycles. Fertil Steril 2009;91:440–9. [49] Marrs RP, March CM, Mishell DR. A comparison of clinical and laboratory methods in monitoring human menopausal gonadotropin therapy. Fertil Steril 1980;34:542–7. [50] Kurachi K, Aono T, Suzuki M, Hirano M, Kobayashi T, Kaibara M. Results of HMG (Humegon)-HCG therapy in 6096 treatment cycles in 2166 Japanese women with anovulatory infertility. Eur J Obstet Gynecol Reprod Biol 1985;19:43–51. [51] Marrs RP, Vargyas JM, March CM. Correlation of ultrasonic and endocrinologic measurements in human menopausal gonadotropin therapy. Am J Obstet Gynecol 1983; 145:417–21. [52] Lewis V, Queenan Jr J, Hoeger K, Stevens J, Guzick DS. Clomiphene citrate monitoring for intrauterine insemination timing: a randomized trial. Fertil Steril 2006;85:401–6. [53] Zreik TG, Garcia-Velasco JA, Habboosh MS, Olive DL, Arici A. Prospective, randomized, crossover study to evaluate the benefit of human chorionic gonadotropin-timed versus urinary luteinizing hormone-timed intrauterine inseminations in clomiphene citrate-stimulated cycles. Fertil Steril 1999;71:1070–4. [54] Karlström P, Bergh T, Lundkvist O. A prospective randomized trial of artificial insemination versus intercourse in cycles stimulated with human menopausal gonadotropin or clomiphene citrate. Fertil Steril 1993;59:554–9. [55] Jaffe SB, Jewelewicz R. The basic infertility investigation. Fertil Steril 1991;56:599–610.
Use of hCG in Reproductive Dysfunction
193
[56] Practice Committee Report. Use of clomiphene citrate in women. Fertil Steril 2003;80: 1302–8. [57] de Leo V, La Marca A, Petraglia F. Insulin lowering agents in the management of polycystic ovary syndrome. Endocr Rev 2003;24:633–67. [58] Baillargeon J-P, Iuorno MJ, Nestler JE. Insulin sensitizers for polycystic ovary syndrome. Clin Obstet Gynecol 2003;46:325–40. [59] Legro RS, Barnhart HX, Schlaff WD, Carr BR, Diamond MP, Carson SA, et al. Clomiphene, metformin, or both for infertility in the polycystic ovary syndrome. N Engl J Med 2007;356:551–66. [60] de Ziegler D. The dawning on the non-cancer uses of aromatase inhibitors in gynaecology. Hum Reprod 2003;18:1598–602. [61] Mitwally MF, Abdel-Razeq S, Casper RF. Human chorionic gonadotropin administration is associated with high pregnancy rates during ovarian stimulation and timed intercourse or intrauterine insemination. Reprod Biol Endocrinol 2004;2:55. [62] Diczfalusy E, Harlin J. Clinical–pharmacological studies on human menopausal gonadotropin. Hum Reprod 1988;3:21–7. [63] Howles CM. Genetic engineering of human FSH (Gonal-F). Hum Reprod Update 1996; 2:172–91. [64] Cohen JJ, Debache C, Pigueau F, Mandelbaum J, Plachot M, Brux de. Sequential use of clomiphene citrate, human menopausal gonadotropin, and human chorionic gonadotropin in human in vitro fertilization. II. Study of luteal phase adequacy following aspiration of the preovulatory follicles. Fertil Steril 1984;42:360–5. [65] Yee B, Vargyas JM. Multiple follicle development utilizing combinations of clomiphene citrate and human menopausal gonadotropins. Clin Obstet Gynecol 1986;29:141–7. [66] Rosenwaks Z, Mausher SJ, Acosta AA. Use of HMG and/or FSH for multiple follicle development. Clin Obstet Gynecol 1986;29:148–57. [67] Stone BA, Vargyas JM, Marrs RP, Quinn PJ, Batzofin JH, Tan T, et al. Levels of steroid and protein hormones in antral fluids of women treated with different combinations of gonadotropins, clomiphene citrate, and a gonadotropin-releasing hormone analog. Fertil Steril 1988;49:249–57. [68] Eijkemans MJC, Imani B, Mulders AGMGJ, Habbema JDF, Fauser BCJM. High singleton live birth rate following classical ovulation induction in normogonadotrophic anovulatory infertility (WHO 2). Hum Reprod 2003;18:2357–62. [69] White DM, Polson DW, Kiddy D, Sagle P, Watson H, Gilling-Smith C, et al. Induction of ovulation with low-dose gonadotropins in polycystic ovary syndrome: an analysis of 109 pregnancies in 225 women. J Clin Endocrinol Metab 1996;81:3821–4. [70] Balasch J, Fabregues F, Penarrubia J, Creus M, Vidal R, Casamitjana R, et al. Follicular development and hormonal levels following highly purified or recombinant folliclestimulating hormone administration in ovulatory women and WHO Group II anovulatory infertile patients. J Assist Reprod Genetics 1998;15:552–9. [71] Tur R, Barri PN, Coroleu B, Buxaderas R, Parera N, Balasch J. Use of a prediction model for high-order multiple implantation after ovarian stimulation with gonadotropins. Fertil Steril 2005;83:116–21. [72] Imani B, Eijkemans MJC, Faessen GH, Bouchard P, Guidice LC, Fauser BCJM. Prediction of the individual follicle-stimulating hormone threshold for gonadotropin induction of ovulation in normogonadotropic anovulatory infertility: an approach to increase safety and efficiency. Fertil Steril 2002;77:83–90. [73] Dourrons NE, Williams DB. Prevention and treatment of ovarian hyperstimulation syndrome. Sem Reprod Endocrinol 1996;14:355–65.
194
Human Chorionic Gonadotropin (hCG)
[74] Golan A, Ron-El R, Herman A, Soffer Y, Weinraub Z, Caspi E. Ovarian hyperstimulation syndrome: an update review. Obstet Gynecol Surv 1989;44:430–40. [75] Bernardus RE, Seegar Jones G, Acosta AA, Garcia JE, Liu HC, Jones DL, et al. The significance of the ratio of follicle-stimulating hormone and luteinizing hormone in induction of multiple follicular growth. Fertil Steril 1985;43:373–8. [76] Sher G, Knutzen VK, Stratton CJ, Montakhab MM, Allenson SG. In vitro sperm capacitation and transcervical intrauterine insemination for the treatment of refractory infertility: phase I. Fertil Steril 1984;41:260–4. [77] Hayden CJ, Balen AH, Rutherford AJ. Recombinant gonadotrophins. Br J Obstet Gynaecol 1999;106:188–96. [78] van Weissenbruch MM, Schoemaker HC, Drexhage HA, Schoemaker J. Pharmacodynamics of human menopausal gonadotrophin (HMG) and follicle-stimulating hormone (FSH): the importance of the FSH concentration in initiating follicular growth in polycystic ovary-like disease. Hum Reprod 1993;8:813–21. [79] Cohlen BJ, Vandekerckhove P, te Velde ER, Habbena JDF. Timed intercourse versus intra-uterine insemination with or without ovarian hyperstimulation for subfertility in men. Cochrane Library. 2002. Cochrane Database Sys review 2000;2:CD000360. [80] Oelsner F, Serr DM, Mashiach S, Blankstein J, Lunenfeld Snyder. The study of induction of ovulation with menotropins: analysis of results of 1897 treatment cycles. Fertil Steril 1978;30:538–44. [81] Branigan EF, Estes A. Use of micro-dose human chorionic gonadotropin (hCG) after clomiphene citrate (CC) to complete folliculogenesis in previously CC-resistant anovulation. Am J Obstet Gynecol 2005;192:1890–6. [82] Coelingh Bennink HJ, Fauser BC, Out HJ. Recombinant follicle-stimulating hormone (FSH; Puregon) is more efficient than urinary FSH (Metrodin) in women with clomiphene citrate-resistant normogonadotropic, chronic anovulation: a prospective, multicenter assessor-blind, randomized clinical trial. European Puregon Collaborative Anovulation Study Group. Fertil Steril 1998;69:19–25. [83] Yarali H, Bukulmez O, Gurgan T. Urinary follicle-stimulating hormone (FSH) versus recombinant FSH in clomiphene citrate-resistant normogonadotropic, chronic anovulation: a prospective, randomized study. Fertil Steril 1999;72:276–81. [84] Hedon B, Hugues JN, Emperaire JC, Chabaud JJ, Barbereau D, Boujenah A, et al. A comparative prospective study of a chronic low dose versus a conventional ovulation stimulation regimen using recombinant human follicle stimulating hormone in anovulatory infertile women. Hum Reprod 1998;13:2688–92. [85] Balasch J, Fabregues F, Creus M, Casamitjina R, Puerto B, Vanrell JA. Recombinant human follicle stimulating hormone for ovulation induction in polycystic ovary syndrome: a prospective, randomized trial of two starting doses in a chronic low-dose step-up protocol. J Assist Reprod Genet 2000;17:561–5. [86] Sheldon R, Kemmann E, Bohrer M, Pasquale S. Multiple gestation is associated with the use of high sperm numbers in intrauterine insemination specimen in women undergoing gonadotropin stimulation. Fertil Steril 1988;49:607–10. [87] O’Herlihy C, Evans JH, Brown JB, Ch de Crespigny LJ, Robinson HP. Use of ultrasound in monitoring ovulation induction with human pituitary gonadotropins. Obstet Gynecol 1982;60:577–82. [88] Koninckx PR, De Moor P, Bosens IA. Diagnosis of the luteinized unruptured follicle syndrome by steroid hormone assays on peritoneal fluid. Br J Obstet Gynaecol 1980;87:929–34.
Use of hCG in Reproductive Dysfunction
195
[89] Check JH, Chase JS, Adelson HG, Dietterich C. New approaches to diagnosis and therapy of the luteinized unruptured follicle syndrome. Int J Fertil 1986;30:29–32. [90] Tummon IS, Henig I, Radwanska E, Binor Z, Rawlins R, Dimowski WP. Persistent ovarian cysts following administration of human menopausal and chorionic gonadotropins: an attenuated form of ovarian hyperstimulation syndrome. Fertil Steril 1988;49:244–8. [91] Katz E. The luteinized unruptured follicle and other ovulatory dysfunctions. Fertil Steril 1988;50:839–48. [92] Hamilton CJCM, Wetzels LCG, Evers JLH, et al. Follicle growth curves and hormonal patterns in patients with luteinized unruptured follicle syndrome. Fertil Steril 1985;43:541–8. [93] Marik J, Hulka J. Luteinized unruptured follicle syndrome: a subtle cause of infertility. Fertil Steril 1978;29:270–4. [94] Kerin JF, Kirby C, Morris D, McEvoy M, Ward B, Cox LW. Incidence of the luteinized unruptured follicle phenomenon in cycling women. Fertil Steril 1983;40:620–6. [95] Check JH, Dietterich C, Nowroozi K, Wu CH. Comparison of various therapies for the luteinized unruptured follicle syndrome. In J Fertil 1992;37:33–7. [96] Killick S, Elstein M. Pharmacologic production of luteinized unruptured follicles by prostaglandin synthetase inhibitors. Fertil Steril 1987;47:773–7. [97] Helmerhorst FM, van Vlier HAAM, Gornas T, Finken MJ, Grimes DA. Intrauterine insemination versus timed intercourse for cervical hostility in subfertile couples. Obstet Gynecol Survey 2006;61:402–14. [98] Hughes EG. “Effective treatment” or “not a natural choice”? Hum Reprod 2003;18: 912–14. [99] Cantineau AEP, Heineman MJ, Cohlen BJ. Single versus double intrauterine insemination in stimulated cycles for subfertile couples: a systematic review based on a Cochrane Review. Hum Reprod 2003;18:941–6. [100] Randall GW, Gantt PA. Double vs. single intrauterine insemination per cycle. Use in gonadotropin cycles and in diagnostic categories of ovulatory dysfunction and male factor infertility. J Reprod Med 2008;53:196–202. [101] Guzick DS. For now, one well-timed intrauterine insemination is the way to go. Fertil Steril 2004;82:30–31. [102] Osuna C, Matorras R, Pijoan JI, Rodriguez-Escudero FJ. One versus two inseminations per cycle in intrauterine insemination with sperm from patients’ husbands: a systematic review of the literature. Fertil Steril 2004;82:17–24. [103] Martinez AR, Bernadus RE, Vermeiden JPW, Shoemaker J. Basic questions on intrauterine insemination: an update. Obstet Gynecol Survey 1993;46:811–28. [104] Ombelet W, Puttemans P, Bosmans E. Intrauterine insemination: a first-step procedure in the algorithm of male subfertility treatment. Hum Reprod 1995;10:90–102. [105] Stewart JA. Stimulated intra-uterine insemination is not a natural choice for the treatment of unexplained infertility. Should the guidelines be changed? Hum Reprod 2003;18:903–14. [106] Botchan A, Hauser R, Gamzu R, Yogev L, Paz G, Yavetz H. Results of 6139 artificial insemination cycles with donor spermatozoa. Hum Reprod 2001;16:2298–304. [107] Horvath PM, Bohrer M, Shelden RM, Kemmann E. The relationship of sperm parameters to cycle fecundity in superovulated women undergoing intrauterine insemination. Fertil Steril 1989;52:288–94. [108] Berg U, Brucker C, Berg FD. Effect of motile sperm count after swim-up on outcome of intrauterine insemination. Fertil Steril 1997;67:747–50. [109] Denil J, Ohal DA, Hurd WW, Menge AC, Hiner MR. Motility longevity of sperm samples processed for intrauterine insemination. Fertil Steril 1992;56:436–8.
196
Human Chorionic Gonadotropin (hCG)
[110] Bordson BL, Ricci E, Dickey RP, Dunaway H, Taylor SN, Curole DN. Comparison of fecundibility with fresh and frozen semen in therapeutic donor insemination. Fertil Steril 1986;46:466–9. [111] Cooke ID. Randomized studies in intrauterine insemination. Fertil Steril 2004;82:27–9. [112] Ragni G, Somigliana E, Vegetti W. Timing of intrauterine insemination: where are we? Fertil Steril 2004;82:25–6. [113] Federman CA, Dumesic DA, Boone WR, Shapiro SS. Relative efficiency of therapeutic donor insemination using a luteinizing hormone monitor. Fertil Steril 1990;54:489–92. [114] Martinez AR, Bernadus RE, Voorhjorst FJ, Vermeiden JPW, Shoemaker J. A controlled study of human chorionic gonadotropin induced ovulation versus urinary luteinizing hormone surge for time of intrauterine insemination. Hum Reprod 1991;6:1247–51. [115] Kosmas IP, Tatsioni A, Fatemi HM, Kolibianakis EM, Tournaye H, Devroey P. Human chorionic gonadotropin administration vs. luteinizing monitoring for intrauterine insemination timing, after administration of clomiphene citrate: a meta-analysis. Fertil Steril 2007;87:607–12. [116] Martinez AR, Bernadus RE, Voorhjorst FJ, Vermeiden JPW, Shoemaker J. Intrauterine insemination does and clomiphene citrate does not improve fecundity in couples with infertility due to male or idiopathic factors: a prospective, randomized, controlled study. Fetil Steril 1990;53:847–53.
16 hCG in Assisted Reproduction Ervin E. Jones Genetics and IVF Institute, Fairfax, Virginia, USA
Both human chorionic gonadotropin (hCG) and luteinizing hormone (LH) are heterodimeric glycoprotein hormones [1]. Both are composed of a common - and -subunits that renders their specificity. Both hormones bind to the same LH/hCG receptor. Classical expression of the LH/hCG receptor has been well established in cells of testis, ovarian theca, granulose, and luteal cells. LH regulates both steroid hormone synthesis and gametogenesis in both males and females. hCG is secreted by the syncytiotrophoblast cells of the early blastocyst and later by the placenta [2]. Many other tissues are also known to produce hCG, including the human pituitary. The circulating half-life of hCG is about 20–37 h, whereas the circulating half-life of LH is only 25 min to 2 h. This difference in half-life and stronger binding affinity account for the approximate 50-fold difference in potency of hCG compared with LH [3]. Although LH is present in all mammalian species, CG is produced only in humans and advanced primates. The biological and immunological activity of hCG is derived from the structural differences in the -subunit. The carboxyl terminal end 24 amino acids unique to hCG render both the biological and immunological specificity of the molecule. This difference allows modern immunoassays to distinguish between the two molecules [4]. More recently, it has become clear that many tissues produce hCG. Recently, it has been shown that menopausal urine contains hCG of pituitary origin. Odell et al. [5] were able to show that hCG is secreted in a pulsatile fashion in adult humans of both sexes. The same authors were able to show that cultured human pituitary cells secrete hCG [6]. These observations are supported by the findings of Muller and Simoni, who showed that mRNA for CG is expressed in sections of marmoset pituitary; sections of marmoset pituitary gland also stained positive for the hormone using immunohistochemical techniques [7]. hCG is widely used in assisted reproduction as an agonist for LH. hCG is easily obtained from urine of pregnant women and placental tissue, and is produced in recombinant Chinese hamster ovary cells. hCG is also less costly when compared to LH. Human menopausal gonadotropin (hMG) is widely used in assisted reproduction and is now known to contain hCG [8].
Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00016-5 © 2010 Elsevier Inc. All rights reserved.
198
Human Chorionic Gonadotropin (hCG)
16.1 T he Ovarian Cycle and hCG Use in Assisted Reproduction In order to describe the use of hCG in assisted reproduction, it is necessary to briefly review the normal ovarian cycle and describe how hCG is used in both natural cycles and in controlled ovarian stimulation cycles for intrauterine insemination or in vitro fertilization (IVF). The role of hCG in the luteal–follicular transition and its potential impact on implantation will be discussed. The use of hCG in the diagnosis and management of abnormal pregnancy will also be reviewed.
16.2 The Follicular Phase: The Role of LH During the natural ovarian cycle, LH/hCG receptors are present in granulosa, theca, and luteal cells of ovarian follicles. Prior to ovulation, LH stimulates the production of estrogen by the production of androgens, which are acted upon by the enzyme aromatase to produce estrogens. LH levels are low during the early follicular phase and begin to rise just prior to ovulation, culminating in the LH surge. Rising LH levels facilitates meiosis in and maturation of oocytes and causes rupture of the follicle. After ovulation, LH induces luteinization of the remaining follicle cells to become a corpus luteum. The corpus luteum is comprised of large luteal cells (derived from the follicular granulosa cells) and small granulosa cells (derived from the theca cells). Follicle stimulating hormone (FSH) receptors are only present on granulosa cells. FSH stimulates follicle growth and selection of the dominant follicle in natural cycles, and increases aromatase activity, which results in increased production of estrogens from androgens. FSH promotes the production of LH receptors on granulosa cells, which, along with cells from the theca compartment, produce the corpus luteum. hCG is often used in ovarian stimulation cycles for IVF. Low levels of LH are secreted during the early follicular phase. Low doses of hCG might be substituted for LH in controlled ovarian stimulation cycles. Low-dose hCG (50–200 mIU/ml) provides a simplified, cost-effective protocol for ovarian stimulation when used in combination with FSH. Several of the combined commercial preparations contain hCG of pituitary origin. Filicori and coworkers studied the impact of low-dose hCG alone and its efficiency in clinical practice to replace LH-containing gonadotropins in controlled ovarian stimulation cycles using a controlled, prospective, and randomized study [8]. They found that the duration of FSH administration was reduced, preovulatory steroid levels were higher, the amount of FSH used was lower, and the number of small (but not large) follicles was reduced. Fertilization rates were higher and the pregnancy outcome did not differ. hCG-stimulated follicle growth and maturation in the presence of FSH administration did not cause premature luteinization and resulted in a more estrogenic intrafollicular environment. They concluded that in low dose in the latter stages of controlled ovarian hyperstimulation, hCG alone reduced recombinant FSH consumption and that the outcomes were comparable to traditional regimens (Figure 16.1).
hCG in Assisted Reproduction
199
GnRH
FSH
FSH
cAMP Induction of FSH. receptors (~ 1500/ cell)
cAMP
LH
Androstenedione Cholesterol Testosterone
Thecal interstitial cell
Induction of aromatase
Androstenedione Testosterone Inhibition of 5α-reductase
aromatase
Estradiol
Estrone
5α-reduced androgens
Granulosa cell
Figure 16.1 A schema depicting the two-cell, two-gonadotropin hypothesis. Modified with permission from Graunwald et al. Manual of assisted reproduction. 2nd ed. Berlin: Springer Verlag; 2000.
Other studies have shown that addition of hCG in cycles of women who seemed to produce high numbers of immature oocytes is a useful maneuver. Huddleston et al. describe a subgroup of patients with a high rate of oocyte immaturity during IVF cycles stimulated with recombinant FSH alone. Addition of LH in a subsequent cycle significantly increased the number of mature oocytes. The study was a matched pair design, where each patient served as their own control. Better quality embryos were also obtained compared with FSH-only cycles [9].
16.3 The Periovulatory Phase and the Mid-Cycle LH Surge In natural cycles, estradiol secreted by the developing follicle sensitizes the anterior pituitary to gonadotropin releasing hormone (GnRH), and the LH surge is triggered. During the preovulatory phase, rising LH levels, which eventually culminate in a surge, cause nuclear maturation of the oocyte, softening and expansion of the oocyte–cumulus complex, and ovulation. The endogenous LH surge mechanism is undesirable in cycles stimulated for IVF and embryo transfer because timing of oocyte collection would be unpredictable. GnRH agonists and antagonists are widely used in assisted reproduction to prevent the endogeneous LH surge. In cycles stimulated for IVF, gonadotropins are administered to increase the number of follicles. In IVF cycles, hCG is administered when certain monitoring criteria are met. Such criteria consist of a certain number of preovulatory follicles of predetermined size. Once such criteria are met, hCG is usually
200
Human Chorionic Gonadotropin (hCG)
administered as a single injection of 5000–10,000 IU to simulate the LH surge. In this instance, hCG injection substitutes for LH and allows for controlled timing of oocyte collection. The hCG injection might be thought of as a “surrogate” LH surge. If hCG is given too late, one or more of the most advanced follicles yield postmature (fragmented) eggs of low viability; on the other hand, if hCG is injected too soon, the eggs might be immature. Ovulation will generally occur 36–39 h after hCG administration. Therefore, retrieval of oocytes is timed for 34–36 h after injection of hCG. In vitro maturation of oocytes is currently a subject of intense investigation. The ability to harvest immature oocytes and mature them in vitro would contribute immensely to the field of reproductive medicine. For example, patients with malignant disease would be able to bank oocytes for future use without the need of undergoing potentially risky stimulation protocols that are commonly used to stimulate multiple follicle development for oocyte retrieval. As described in this chapter, hCG is most often used prior to retrieval of oocytes for IVF. Similarly, hCG is also used to induce maturation of immature oocytes in vivo and in vitro [10]. In vivo, priming with hCG results in more oocytes than in nonprimed and FSH-primed cycles, and in vivo matured oocytes result in better embryo development and higher pregnancy rates. Their findings illustrate the importance of the use of hCG in the induction of oocyte maturation in modern reproductive medicine. In controlled ovarian stimulation cycles for timed intercourse or intrauterine insemination, hCG is also used to trigger ovulation. In such cycles, follicle growth is monitored using serial transvaginal ultrasound scanning along with measurements of serum estradiol levels. Timed administration of hCG helps to assure ovulation and facilitates timing of intercourse or insemination.
16.4 The Follicular–Luteal Transition Following natural ovulation, or oocyte retrieval in IVF cycles, the granulosa and theca cells that remain in the follicle are transformed (under the influence of LH) into luteal cells that make up the corpus luteum. The granulosa cells become the large luteal cells of the corpus luteum, and the theca cells become the small luteal cells. The large cells decline in number in the late luteal phase in concert with a decline in steroidogenesis. The small cells also decrease their ability to respond to LH. Importantly, the large luteal cells that remain seem to increase their ability to respond to hCG. The follicular–luteal transition begins. The corpus luteum secretes primarily progesterone, although it also synthesizes and secretes estradiol. LH is believed to play a key role in the follicular–luteal transition. The follicular– luteal transition is primarily driven by LH. The LH surge induces the formation of the corpus luteum in natural cycles, and low levels of LH are continuously secreted to support the early development of the corpus luteum. hCG binds to LH receptors and facilitates the follicular–luteal transition. In controlled ovarian stimulation cycles for IVF, exogenous hCG is often used to supplement the luteal phase. In conception cycles, hCG produced by the developing blastocyst stimulates prolongation of the life of the corpus luteum.
hCG in Assisted Reproduction
201
16.5 The Luteal–Placental Shift The corpus luteum generally lives for 11–12 days in nonconception cycles; progesterone levels fall, menses follow, and the next menstrual cycle ensues. In conception cycles, the placenta must eventually take over progesterone production from the corpus luteum to prevent menses and to permit the pregnancy to succeed. hCG is secreted exponentially and the long half-life extends the life span of the corpus luteum, maintaining progesterone secretion until the placenta begins to produce significant progesterone at about 7 weeks of gestation [11]. hCG can be detected in the maternal circulation around the time of implantation; it peaks around 9–11 weeks of gestation. hCG then decreases to about 1/10 the peak level. The secretion of hCG is crucial for maintenance of pregnancy until about the seventh week of gestation. Between 7 and 10 weeks, the corpus luteum is replaced by the placenta in terms of progesterone production. This phase of luteal function has been referred to as the “luteal–placental shift.” In conception cycles, the life of the corpus luteum is extended until the luteal–placental shift occurs. In a study of trinucleate embryos obtained from IVF, the eight-cell stage embryo expresses mRNA [12]. hCG can be detected in maternal serum 1–2 days after implantation. Locate et al. studied the differential distribution of mRNA for the and -subunits of CG in the implantation stage blastocyst of the marmoset monkey. CG mRNA was detected mainly in polar and mural trophoblasts 13–15 days postconception [13]. Thus, early in the first trimester, hCG that is manufactured by the syncytiotrophoblast “rescues” the corpus luteum, which is the major source of estrogens and progesterone. This function of the corpus luteum continues well into early pregnancy. By itself, however, the corpus luteum is not capable of generating the very high steroid levels that are characteristic of late pregnancy. The placenta augments the production of progesterone and estrogens such that by 8 weeks of gestation, the placenta has become the major source of these steroids (Figure 16.2). The timing of the luteal–placental shift is crucial, particularly in IVF cycles. At approximately 8 weeks gestation, most centers used either progesterone or hCG to supplement the luteal phase. Lower doses of hCG are usually given (1250–2500 IU) daily after pregnancy has been diagnosed. Some centers combine progesterone and hCG in the luteal phase.
16.6 The Potential Role of hCG in Implantation Precise synchronization between the developing embryo and the maternal endometrium is necessary for successful implantation. hCG is one of the earliest molecules produced by the embryo and is the most specific marker of its presence [14]. A dynamic interchange occurs at the maternal–fetal interface during embryo implantation. Studies in primate models have shown that CG is the major trophoblast signal; it not only rescues the corpus luteum but also modulates the uterine environment in preparation for implantation [11]. This response is characterized by alteration in
202
Human Chorionic Gonadotropin (hCG) (B) 1000
100
800
80 Progesterone (ng/ml)
Estradiol (pg/ml)
(A)
600
400
200
60
40
20
0
0 0
5
10
15
Conception
20
0
5
10
15
20
Nonconception
Figure 16.2 Conception and nonconception from IVF menstrual cycles. Mean ( SEM) serum concentrations of estradiol (A) and progesterone (B) in conception and nonconception cycles from 28 IVF cycles without luteal support. Note luteal rescue as reflected by rising estradiol and progesterone levels under the influence of endogenous hCG in conception cycles. Reproduced with permission from Hutchinson-Williams et al. Fertil Steril, 1990;53:495.
both the morphological and biochemical characteristics in the three major cell types: luminal and glandular epithelium and stromal fibroblasts. Local effects of CG that influenced the immune system might permit survival of the fetal allograft and prevent endometrial cell death. The hCG/LH receptor has received considerable attention for its role as a luteo tropic factor. This role not only relays the inadequate support provided by the reduced rates of LH secretion but also influences the pregnancy in a paracrine fashion, probably by binding with the LH/hCG receptor on the endometrial epithelium. The embryo actively participates in its own implantation, tolerance, and placentation [15] (Figure 16.3). hCG functions as a long-range signal and sustains the corpus luteum in the face of precipitously declining LH levels of maternal origin. hCG is one of the most significant molecules secreted by the developing blastocyst, both prior to and after implantation. In addition to rescuing the corpus luteum, hCG also acts as an immunosuppressive agent and a growth factor that promotes trophoblasts growth and placental development. hCG might also have a role in the adhesion of the trophoblasts to the maternal endometrium; hCG has protease activity, and hCG levels are high in the area where the trophoblasts face and make contact with the endometrium [15]. hCG appears to have a role in uterine angiogenesis and vascular remodeling, both of which are crucial to embryo implantation and normal development of the placenta [16].
hCG in Assisted Reproduction
203
Figure 16.3 A fully expanded 5-day human blastocyst in culture. The trophectoderm and prominent inner cell mass are visible. Courtesy of Andy Dorfmann, Director Embryology Laboratory, Genetics and IVF Institute, Fairfax, VA, USA.
16.7 hCG in the Management of Normal Pregnancy In addition to its therapeutic role in assisted reproduction, hCG is highly useful as a diagnostic tool. Normal serum hCG levels are approximately 100 mIU/ml at the time of missed menses in natural cycles, and reach about 100,000 mIU/ml by 8–10 weeks gestation. The first significant rise in plasma hCG occurs 9–12 days following ovulation. There is a rapid rise in the daily output of hCG by in vitro cultured blastocyst up to about day 11 following fertilization. Gonadotropin production then tends to plateau to day 14 following fertilization. A similar observation was reported using blastocyst recovered from the uterine cavity of the baboon [17]. Catt et al. found that hCG does not appear in the maternal circulation until after implantation of the blastocyst [18]. The serum concentrations of hCG then rise sharply and, with a doubling time of approximately 2.3 days, reach levels of 10,000–100,000 mIU/ml near the end of the first-trimester pregnancy. hCG levels begin to decline after the tenth week of pregnancy. hCG detected in maternal blood is by far the most reliable marker for biochemical diagnosis of early pregnancy. The development of monoclonal antibody assays for hCG provided increased sensitivity and also decreased the risk of cross-reactivity substances such as LH. hCG is ubiquitous even in the nonpregnant state (at low concentrations). Thus, the risks of a false positive test have decreased with the advent of monoclonal antibody assays for hCG. A single estimation for hCG should only be considered indicative of pregnancy if it is in excess of 25 mIU/ml, or pregnancy can be definitively identified if the level of hCG is seen to double in concentration in serial samples obtained at 2- to 3-day intervals. Knowledge of hCG’s chemistry has led to the development of sensitive and specific immunoassays that are used in
204
Human Chorionic Gonadotropin (hCG)
assisted reproduction. hCG diagnostics are used in the management of normal uterine pregnancy, abnormal intrauterine pregnancy (IUP), ectopic pregnancy, and cancers of the reproductive tract. For each laboratory, the detection limit for hCG assays should be determined in order to establish the levels of hCG above which the diagnosis can be confidently made in relation to assay background. hCG measurements should be postponed until the hCG injected prior to ovum pickup has been cleared from the patient’s blood. Clinical assays for beta hCG measure between 1 and 5 mIU/ml; levels less than 5 mIU/ml are considered negative for pregnancy; levels between 5 and 25 mIU/ml might still be considered as spurious for pregnancy and should be repeated depending on the clinical circumstance. All pregnancy tests are forms of immunoassays of some sort; home, slide, and dipstick tests detect 20–50 mIU/ml. Developments of highly specific monoclonal antibodies to the -subunit of hCG have resulted in assays of greater sensitivity and specificity. The -subunit of hCG provides a sensitive test for hCG. Using this method, the diagnosis of normal pregnancy can be made as early as 8–9 days after ovulation.
16.8 Hyperstimulation Syndrome Ovarian hyperstimulation syndrome occurs following luteinization of ovarian follicles. Hyperstimulation syndrome only occurs after the administration of hCG. Clinical signs of hyperstimulation syndrome usually occur 5–10 days after hCG administration or following the LH surge. Hyperstimulation syndrome is rare in women with a spontaneous LH surge. hCG is thought to be causal in initiation of hyperstimulation syndrome. hCG administration provokes profound effects on paracrine parameters of differentiation at implantation—insulin growth factor binding protein-1 and prolactin and implantation. Leukemia inhibitory factor (LIF), macrophage colonystimulating factor (MCSF), and vascular endothelial growth factor(VEGF), cytokines important for neoangiogenesis, were significantly stimulated by hCG, which suggests hCG has a role in controlling endometrial vascularization and placentation [14]. It is believed that ovarian secretion causes increased capillary permeability and increased peritoneal fluid shift. Increased capillary permeability results in massive ascites and hypovolemia. Patients at increased risk for hyperstimulation syndrome should not be treated with luteal phase hCG. Most programs use progesterone supplementation instead of hCG. A major preventive measure is to withhold hCG in the face of impending hyperstimulation syndrome. Interestingly, hyperstimulation syndrome rarely occurs following a spontaneous LH surge induced by rising estrogen levels. Recent advances in the control of ovarian cycles stimulated for IVF have lead to the use of gonadotropin hormone releasing hormone agonists to trigger an LH surge. Patients managed in this manner rarely get hyperstimulation syndrome. This observation suggests that characteristic actions of the hCG molecule cause hyperstimulation syndrome.
hCG in Assisted Reproduction
205
16.9 hCG in the Management of Ectopic Pregnancy Ectopic pregnancy is a major clinical problem, occurring in over 75,000 cases per year in the United States. Ectopic pregnancy can be an unintended consequence of assisted reproduction. Ectopic pregnancy occurs in approximately 2% of IVF cycles. When compared with normal pregnancies, hCG levels increase at different rates in ectopic pregnancies [19]. The rate of rise of hCG in ectopic pregnancy is slower than that of normal pregnancies [19]. Similarly, the rate of fall of hCG is lower for ectopic pregnancies than following completed spontaneous abortions. Thus, the earliest manifestation of ectopic pregnancy might be low, or slowly rising, titers of hCG. Most low or poorly rising hCG titers are due to abnormal intrauterine pregnancies. hCG levels are measured at approximately 9 days after presumed implantation. For example, most pregnancy tests are obtained 7 days after a day-5 transfer, or 10 days after a day-3 transfer. Ectopic pregnancy must be suspected if the first titer is low or when titers fail to double appropriately. The measurement of hCG by radioimmunoassay shows the clinical usefulness of hCG. It is a specific indicator of presence of a viable trophoblastic tissue. The gold standard for monitoring suspected ectopic pregnancy is a combination of serial hCG titers and transvaginal ultrasound. When hCG titers reach 1500–2000 mIU/ ml, a gestational sac should be visible on transvaginal ultrasound. Nyberg et al. looked at 150 women with early pregnancy, 76 intrauterine pregnancies, and 74 ectopic pregnancies [20]. When the hCG titer was 1800 mIU/ml or greater in all women with IUP, a gestational sac was seen in all patients. In contrast, in patients with ectopic pregnancy and hCG titer greater than 1800 mIU/ml, no gestational sac was seen. The authors concluded that when hCG levels exceed 1800 mIU/ml, an intrauterine gestational sac is normally detected and its absence is evidence for ectopic pregnancy. hCG doubles every 2 days during the 2–4 weeks following ovulation. An obstetrician should see a sac if hCG is greater than 1500 mIU/ml. When hCG doubles every 2 days, the situation is reassuring but does not rule out abnormal pregnancy. Failure of hCG titers to double must raise suspicion of ectopic pregnancy. hCG doubles every 2 days under 1200 mIU/ml, every 3 days between 1200 and 6000 mIU/ml, and every 4 days above 6000 mIU/ml. Clinicians must be aware that the discriminatory zone varies with the assay and with the reference preparation. Mixed-effects regression models have also been used to estimate the hCG trajectory in its variability and relationship to pregnancy outcomes [17]. The authors observed a threefold increase in hCG between the day of detection and the next day. They described the average profile of hCG rise and its variability during the 7 days following estimated implantation in a population of naturally conceived pregnancies. The relative rate of rise decreased thereafter, reaching 1.6-fold between days 6 and 7. hCG levels followed a log-quadratic trajectory, and the patterns of rise were unrelated to number of fetuses or risk of spontaneous abortion for up to six babies. hCG titers are also used to monitor treatment of ectopic pregnancy, regardless of the treatment modality (i.e., medical or surgical). Methotrexate is a foliate antagonist
206
Human Chorionic Gonadotropin (hCG)
and is widely used in assisted reproduction for ectopic pregnancy. In all cases, titers must be followed until they become negative. Agostini et al. looked at the change in hCG levels after methotrexate injection and outcome in 129 consecutive patients [21]. A 20% decline in hCG levels between days 1 and 4 of methotrexate treatment had a positive predictive value of 97%. With the evolution of conservative surgery for ectopic pregnancy, the risk of persistent ectopic pregnancy increased [22]. The authors discussed a series of 11 patients, 10 of whom had repeat surgeries for either symptoms or for plateauing hCG levels. While monitoring symptoms, serial measurements of hCG levels constitute optimum treatment for this group of patients. Mock et al. also studied the disappearance rate of hCG in patients treated for ectopic pregnancy by laparoscopic salpingostomy. The disappearance of hCG from the circulation followed biexponential decay with a rapid fall followed by slow disappearance. The disappearance kinetics are characterized by two half-lives: an early and a late. The authors concluded that the late half-life should be used as a parameter to follow ectopic pregnancies after treatment [23]. Clearly, low serum levels of hCG might indicate early pregnancy. Persistently low serum hCG levels might indicate abnormal placentation associated with either ectopic pregnancy or IUP. Patients with suspected ectopic pregnancy often have persistently low serum hCG concentrations (50–500 mIU/ml). Undetectable levels of hCG (less than 5 mIU/ml) rule out the existence of viable trophoblastic tissue and render a medical or surgical procedure unnecessary.
16.10 Conclusions The use of hCG occupies an extraordinarily prominent place in the field of reproductive medicine. The use of hCG can be partitioned into two major categories: diagnostic and therapeutic. Diagnostic uses of hCG include detection, monitoring, and management of normal pregnancy. hCG is also widely used in the diagnosis, management, and monitoring of abnormal pregnancies, including ectopic pregnancies and persistent trophoblastic disease. Therapeutic uses of hCG in reproductive medicine have found greatest utility in controlled ovarian stimulation protocol for IVF (including in vitro maturation of oocytes) and in protocols for ovulation induction for intrauterine insemination. Several isoforms of hCG resulting from posttranslational modification of the native molecule have been recently identified. These will, undoubtedly, find utility in the diagnosis and treatment of both disease states and in ovarian stimulation protocols used in assisted reproduction.
References [1] Lapthorn AJ, Harris DC. Crystal structure of human chorionic gonadotropin. Nature 1994;369:455–61.
hCG in Assisted Reproduction
207
[2] Pierce JG, Parsons TF. Glycoprotein hormones: structure and function. Annu Rev Biochem 1981;50:465–95. [3] Rao CV, Estergreen VL, Carman Jr FR, Moss GE. Receptors for gonadotrophin and prostaglandin F2 alpha in bovine corpora lutea of early, mid and late luteal phase. Acta Endocrinol 1979;91:529–37. [4] Filicori M. Use of luteinizing hormone in the treatment of infertility: time for reassessment? Fertil Steril 2003;79:253–5. [5] Odell WD, Griffin J. Pulsatile secretion of human chorionic gonadotropin in normal adults. N Engl J Med 1987;317:1688–91. [6] Odell WD, Griffin J. Secretion of chorionic gonadotropin by cultured human pituitary cells. J Clin Endocrinol Metab 1990;71:1318–21. [7] Muller TM, Simoni M. Chorionic gonadotropin beta subunit mRNA but not luteinizing hormone beta subunit mRNA is expressed in the pituitary of the common marmoset (Callithrix jacchus). J Mol Endocrinol 2004;32:115–18. [8] Filicori M, Fazleabas AT. Novel concepts of human chorionic gonadotropin: reproductive system interactions and potential in the management of infertility. Fertil Steril 2005;84:275–84. [9] Huddleston HG, Jackson KV. hMG increases the yield of mature oocytes and excellentquality embryos in patients with a previous cycle having a high incidence of oocyte immaturity. Fertil Steril 2009;92:946–9. [10] Son WY, Chung JT, Chian R-C, Herrero B, Demirtas E, Elizur S, et al. A 38 h interval between hCG priming and oocyte retrieval increases in vivo and in vitro oocyte maturation rate in programmed IVM cycles. Hum Reprod 2008;23:2010–16. [11] Srisuparp S, Strakova Z. The role of chorionic gonadotropin (CG) in blastocyst implantation. Arch Med Res 2001;32:627–34. [12] Bonduelle ML, Dodd R. Chorionic gonadotrophin-beta mRNA, a trophoblast marker, is expressed in human 8-cell embryos derived from tripronucleate zygotes. Hum Reprod 1988;3:909–14. [13] Lopata A, Berka J. Differential distribution of mRNA for the alpha- and beta-subunits of chorionic gonadotrophin in the implantation stage blastocyst of the marmoset monkey. Placenta 1995;16:335–46. [14] Licht P, Russu V. On the role of human chorionic gonadotropin (hCG) in the embryoendometrial microenvironment: implications for differentiation and implantation. Semin Reprod Med 2001;19:37–47. [15] Perrier d’Hauterive S, Berndt S. Dialogue between blastocyst hCG and endometrial LH/ hCG receptor: which role in implantation? Gynecol Obstet Invest 2007;64:156–60. [16] Zygmunt M, Herr F, Keller-Schoenwetter S, Kunzi-Rapp K, Munsteadt K, Rao CV, et al. Characterization of human chorionic gonadotropin as a novel angiogenic factor. J Clin Endocrinol Metab. 2002;87:5290–6. [17] Nepomnaschy PA, Weinberg CR. Urinary hCG patterns during the week following implantation. Hum Reprod 2008;23:271–7. [18] Catt KJ, Dufau ML. Appearance of hCG in pregnancy plasma following the initiation of implantation of the blastocyst. J Clin Endocrinol Metab 1975;40:537–40. [19] Silva C, Sammel MD. Human chorionic gonadotropin profile for women with ectopic pregnancy. Obstet Gynecol 2006;107:605–10. [20] Nyberg DA, Filly RA. Ectopic pregnancy. Diagnosis by sonography correlated with quantitative HCG levels. J Ultrasound Med 1987;6:145–50. [21] Agostini A, Blanc K. Prognostic value of human chorionic gonadotropin changes after methotrexate injection for ectopic pregnancy. Fertil Steril 2007;88:504–6.
208
Human Chorionic Gonadotropin (hCG)
[22] Seifer DB, Gutmann JN, Doyle MB, Jones EE, Diamond MP, DeCherney AH. Persistent ectopic pregnancy following laparoscopic linear salpingostomy. Obstet Gynecol 1990; 76:1121–5. [23] Mock P, Chardonnens D, Stamm P, Campana A, Bischof P. The apparent late half-life of human chorionic gonadotropin (hCG) after surgical treatment for ectopic pregnancy. A new approach to diagnose persistent trophoblastic activity. Eur J Obstet Gynecol Reprod Biol 1998;78:99–102.
17 Illicit Use of hCG in Dietary
Programs and Use to Promote Anabolism Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
The most common therapeutic use of hCG is not to promote ovulation nor to promote progesterone production and maintain fertility. It is use hCG illicitly in a dietary programs. There are thousands of websites, books, newspaper advertisements, and TV advertisements selling diets with these ridiculous claims. Some athletes even give themselves regular shots of hCG to promote testosterone and muscle growth. These uses inspire us to ask why? This chapter discusses the pros and cons of illicit hCG applications, and uses and misuses of hCG administration.
17.1 Dietary Programs Advertisements trying to sell prescription hCG and a novel diet come from medical clinics, doctors’ offices, and pharmacies all over the world. These advertisements are in newspapers and even presented by medical correspondents on major TV shows (KSL5 Utah, Fox Morning Show, Mike & Juliet Show, and others). This story all starts with one very small study performed in England in the early 1950s. Dr. Simeons conducted a nonblinded, noncontrolled study in which participants were restricted to 500 kcal/day and given 125 IU supplements of urine-origin hCG [1]. The study had amazing results and led to weight loss. According to Dr. Simeons [1], hCG mobilizes stored fat all over the body and suppresses appetite. If you search the Internet, there are literally thousands of advertisements for the hCG-based diet of Dr. Simeons. In 1959, Sohar [5] argued that it was solely the 500 kcal diet that led to the weight loss and that the hCG supplements did nothing. In the 1960s, Craig et al. [8] and Frank [9] each conducted their own study regarding the Simeons diet. The two separate experiments both agreed that there was no possible relationship between hCG administration and loss of weight or hunger. In the years that followed, physicians profiting from the dietary administration of hCG began performing double-blind studies. These small, inappropriately controlled studies confirmed that the Simeons diet worked [6,7]. Then in the 1970s, several double-blind studies were independently performed by Young et al. [10], Stein et al. [11], Greenway and Bray [12], and Shetty and Kalkoff [13]. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00017-7 © 2010 Elsevier Inc. All rights reserved.
210
Human Chorionic Gonadotropin (hCG)
They all reached the same conclusion as Sohar did in 1959—that hCG had nothing to do with hunger and did not promote weight loss. In the 1980s, Richer et al. [14] reached the same conclusion. In the 1990s, the high-standard controlled studies of Bosch et al. [15] and Lijesen et al. [16] once again confirmed that an hCG diet simply did not and could not work. The amount of medical evidence that began accumulating was overwhelming. Countless studies used clean, double-blind, and carefully controlled data to disprove the claims of Simeons’ hCG-based diet. It simply did not work [17,18]. It should be noted that in 2007, the USA Federal Trade Commission charged Kevin Trudeau with misrepresentation for writing a book praising the hCG diet [19]. Today, the CV Rao Laboratory [2–4] has identified hCG receptors in numerous sites associated with pregnancy (e.g., the uterus, the placenta, the brain). The laboratory has not, however, found a receptor in the digestive tract, in the liver, or in adipose tissues that could explain how hCG mobilizes stored fat and suppresses hunger. Thus, Simeons’s 1954 results are seemingly false and were apparently contrived to support a new enterprise. Nevertheless, Simeons started a dietary fad, and even after 55 years his diet is still sold throughout the world. Dr. Simeons even came out with a support book for followers of his diet. Somehow, Simeons became a public and scientific hero for a nonsense diet. After the accumulation of all this evidence, how and why there are still clinics, doctors’ offices, and pharmacies completely sold on the hCG diet defies all logic. Yes, hCG will promote emesis or nausea and vomiting. Is this the secret to the diet’s claims of exceptional weight loss—extreme nausea and constant vomiting? Today, clinics and pharmacies give patients a choice of injectable hCG, hCG drops that one places under the tongue, or hCG–green tea mix pills [19–23]. They claim that each works as well as the other, but there isn’t any evidence to support that hCG can mobilize fats, suppress hunger, or induce euphoria, as claimed. Surely, the digestive tract destroys the peptide hormone before it is absorbed into the circulation. Injectable hCG preparations range from partially purified human pregnancy urine extracts, called Profasi, Pregnyl, Novarel, Chorex, and Follotein, to super-pure Chinese hamster ovary cell-line recombinant hCG called Ovidrel. All are misused for dietary purposes with claims of great weight loss. It is one thing for a woman to be sold by crazy advert about some miracle diet. It is another to consider the consequence of hCG administration, cessation of menstrual periods, infertility, and hyperemesis gravidarum.
17.2 hCG and Anabolism Promotion Unquestionably, hCG promotes testicular testosterone that acts on muscles to promote growth and anabolism in men. Interestingly, injections of hCG do not promote significant testosterone production in women [24] but rather promote progesterone and androstenedione production [24]. As such, the other illicit use of hCG is in sports and athletics, particularly in major professional sports and world sports such as the Olympic Games. The hormone is also sold on the Internet and in magazine advertisements, supposedly making people strong by aiding muscle growth. Agencies such as the World Anti-Doping Agency (WADA) and the U.S. Anti-Doping Agency (USADA) started to test the urine of Olympic and other international sporting
Illicit Use of hCG in Dietary Programs and Use to Promote Anabolism
211
athletes. In the United States, all major athletic associations carry out random urine tests that detect testosterone, growth hormone, and the hormone hCG, including the National College Athletic Association, National Football League, National Hockey League, National Basketball League, National League Baseball, and the American League Baseball. As shown by Stenman and colleagues [25], hCG can circulate for 7–11 days following injection. The USA hCG Reference Service has shown WADA and USADA that during this period hCG is gradually degraded to free subunit, to nicked hCG, to nicked molecules missing the -subunit C-terminal peptide, and finally to -core fragment. Thus, it is important to measure each of these molecules in urine tests. The USA hCG Reference Service recently examined a professional athlete’s serum and urine at the request of a sports authority doping agency. Mostly urine total hCG or urine fragments were observed, and his level of hCG was very close to the 5 mIU/ ml cutoff. But over a 1-month period, these levels persisted with little change. They were inconsistent with illegally administered hCG levels, which should disappear over time. Interestingly, similar findings were observed upon testing the athlete’s father, and as a result a case of familial hCG was diagnosed [30]. Familial hCG is a rare genetic abnormality that leads to high pituitary hCG production [30]. The USA hCG Reference Service has now observed four total cases of familial hCG. We now understand that it is statistically possible that some athletes might be positive for hCG due to genetic abnormalities. After the USA hCG Reference Service released these findings, a hearing was held, and the professional sports team cleared the athlete’s name. Yes, familial hCG is a rare genetic explanation for positive low-level hCG results. It should not be forgotten, however, that testicular cancer can cause low-level hCG results in men, and gestational trophoblastic disease cause low-level results in women. Cancers of various origins might explain low-level total hCG results in both men and women. Today, most agencies test urine using the Siemens Immulite assay, which is the only automated hCG test that detects all of these degradation products [26,27]. Laidler and colleagues [28] measured hCG in 1400 men and statistically determined that a 5-mIU/ml cutoff was very acceptable in terms of total hCG concentration present in urine. Delbenke et al. [29] examined 5663 men and found hCG reaching 2.28 mIU/ml. They also supported the 5-mIU/ml sensitivity limit. This is now used as cutoff in doping studies by WADA and USADA.
References [1] Simeons ATW. The action of chorionic gonadotropin in the obese. Lancet 1954;2:946–7. [2] Lei ZM, Reshef E, Rao CV. The expression of human chorionic gonadotropin/ luteinizing hormone receptors in human endometrial and myometrial blood vessels. J Clin Endocrinol Metab 1992;75:651–9. [3] Lei ZM, Rao CV, Kornyei J, Licht P, Hiatt ES. Novel expression of human chorionic gonadotropin/luteinizing hormone receptor gene in brain. Endocrinol 1992;132:2262–70. [4] Lei Z, Rao CV. Gonadotropin receptors in human fetoplacental unit: implications for hCG as an intracrine, paracrine and endocrine regulator of human fetoplacental function. Placenta 1992;13:213–24.
212
Human Chorionic Gonadotropin (hCG)
[5] Sohar EA. A forty-day 550 calorie diet in the treatment of obese outpatients. Am J Clin Nutr 1959;5:514–19. [6] Gusman HA. Chorionic gonadotropin in obesity. Further clinical observations. Am J Clin Nutr 1969;22:686–95. [7] Asher WL, Harper HW. Effect of human chorionic gonadotropin on weight loss, hunger and feeling of well being. Am J Clin Nutr 1973;26:211–18. [8] Craig LS, Ray RE, Waxler SH, Madigan HM. Chorionic gonadotropin in the treatment of obese women. J Clin Nutr 1963;12:230–4. [9] Frank BW. The use of chorionic gonadotropin hormone in the treatment of obesity. Am J Clin Nutr 1964;14:133–6. [10] Young RL, Fuchs RJ, Woinjen MJ. Chorionic gonadotropin in weight control. A double blind crossover study. J Am Med Assoc 1976;236:2495–7. [11] Stein MR, Julis RE, Peck CC, Hinshaw W, Sawicki JE, Deller JJ. Ineffectiveness of human chorionic gonadotropin in weight reduction: a double-blind study. Am J Clin Nutr 1976;29:940–8. [12] Greenway FL, Bray GA. Human chorionic gonadotropin (hCG) in the treatment of obesity. West J Med 1977;127:461–3. [13] Shetty KR, Kalkoff RK. Human chorionic gonadotropin (hCG) treatment of obesity. Arch Inter Med 1977;137:151–5. [14] Richer RT, Runnebaum B. Risiko-Nutzen risk-benefit analysis of a hCG-500 hcal reducing diet (cura romana) in females. Geburtshitfe und Frauenhellkunde 1987;47:297–305. [15] Bosch B, Venter I, Stewart RI, Bertram SR. Human chorionic gonadotropin and weight loss. A double-blind placebo-controlled trial. S Afr Med J 1990;77:185–9. [16] Lijesen GK, Theeuwen I, Assendelft WJ, Van Der Wal G. The effect of human chorionic gonadotropin (hCG) in the treatment of obesity by means of the Simeon therapy: a criteria-based meta-analysis. Br J Clin Pharmacol 1995;40:237–43. [17] Barrett S. hCG worthless as a weight-loss aid. Diet Scam Watch 2007 2010. http://www .dietscam.org/reports/hcg.shtml. [18] Peyman T. The hCG myth, http://www.drtarapeyman.com/hCGArticle.pdf [19] FTC: Marketer Kevin Trudeau violated prior court order, charges him with misrepresenting contents of book, http:/www.casewatch.org/ftc/news/2007/trudeau.shtml; 2007. [20] hCG Diet, http://www.hcgdietdirect.com/?gclidCIGSp_60z50CFRESawod9HZFrg; 2010 [21] Your hCG.com, http://www.yourhcg.com/purchase.php; 2010 [22] hCG Medical. Advanced Wellness and Hormone Therapy, http://www.hcgmedical.com/ inthenews/index.html; 2008 [23] OraThin oral hCG alternative, http://www.naturalcuresstore.com/product/ORAL_ HCG?gclidCPG_5-q4z50CFShSagodGjqdrw; 2010 [24] Handelsman DJ. Clinical review: the rationale for banning human chorionic gonadotropin and estrogen blocker in sport. J Clin Endocrinol Metab 2006;91:1646–53. [25] Stenman UH, UnkilaKallio L, Korhonen J, Alfhan H. Immunoprocedures for detecting human chorionic gonadotropin: clinical aspects and doping control. Clin Chem 1997;43:1293–8. [26] Cole LA, Sutton JM, Higgins TN, Cembrowski GS. Between-method variation in hCG test results. Clin Chem 2004;50:874–82. [27] Cole LA, Shahabi S, Butler S, Mitchell H, Newlands ES, Behrman HR, et al. Utility of commonly used commercial hCG immunoassays in the diagnosis and management of trophoblastic diseases. Clin Chem 2001;47:308–15.
Illicit Use of hCG in Dietary Programs and Use to Promote Anabolism
213
[28] Laidler P, Cowan DA, Hider RC, Kickman AT. New decision limits and quality control material for detecting human chorionic gonadotropin misuse in sports. Clin Chem 1994;40:1306–11. [29] Delbenke FT, Van Eenoo P, De Backer F. Detection of human chorionic gonadotropin misuse in sports. Intl J Sports Med 1998;19:287–90. [30] Cole LA, Laidler LL. Inherited hCG. J Reprod Med 2010;55:99–102.
18 Antibodies for Intact hCG, for
Total hCG, for Free Subunits, Glycosylation Variants, and for hCG Fragments Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
In this chapter, we discuss the basics of designing an hCG or hCG-associated assay. The first and most important step is in choosing appropriate antibodies. These same principles are required whether the assay is an in-house manual microtiter plate test, a platform automated test, or a point-of-care/over-the-counter test. For many years, hCG tests have been nicknamed hCG or hCG tests. This dates back to the original radioimmunoassays (RIAs) of the late 1960s and early 1970s. These assays used a polyclonal antibody against hCG dimer. Because of the common hCG and LH - and the -subunits similarity, they detected both hCG and luteinizing hormone (LH). In 1972, Vaitukaitis and colleagues [1] discovered an RIA using an antibody against just the hCG -subunit. This RIA recognized only hCG and not LH. Soon after, all hCG tests adopted this method. This was the hCG or hCG test. Today, there are two distinct classes of immunometric hCG tests: intact hCG assays and total hCG assays. Intact hCG assays detect hCG dimer forms only. Total hCG assay should detect all forms of hCG (-subunit, dimer, free subunits, and fragments). In practice, total hCG tests only detect hCG and its free -subunit. In this chapter, we will introduce the antibodies used for both intact and total hCG tests. We will also introduce the antibodies needed to make a free - and -subunits specific test, hyperglycosylated hCG (hCG-H) and nicked hCG assays, and finally the antibodies needed to detect hCG -core fragment.
18.1 Intact hCG and Total hCG Assay To make an intact hCG assay, a mixture of two antibodies is needed. One antibody acts as a capture antibody and the second as a tracer antibody. Some examples of appropriate antibodies are anti-core 2119; INN-hFSH-72, -98, -100, -179, -132, -158; and anti-core SB6, B108, INN-hCG-2, -22, -24; and ISOBM-265, -274, -284, -273, -276, -285. In Combination 1 (Figure 18.1), antibodies against the cores of - and -subunits Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00018-9 © 2010 Elsevier Inc. All rights reserved.
218
Human Chorionic Gonadotropin (hCG)
Figure 18.1 Configuration and specificity of intact hCG assays. The illustrations of and are representations; they bear no relationship to the actual structures. The listed hCG variants show which variant are recognized (contain antibody binding sites).
are used. Both of these antibodies are generated using - and -subunit standards. Combination 1 detects all forms of hCG dimer: hCG, hCG-H, nicked hCG, or nicked hCG missing the -subunit C-terminal peptide (CTP) (Figure 18.1). In Combination 2 (Figure 18.1), an antibody against hCG dimer is combined with an antibody against the -subunit core, the latter of which is generated using intact hCG dimer as antigen. Some examples of appropriate antibodies include B109, INNhCG-10, -40, -53 and ISOBM-283, -270, -272, -275. As illustrated (Figure 18.1), this combination will detect only hCG and hCG-H. Because of a three-dimensional (3D)
Antibodies for Intact hCG, for Total hCG, for Free Subunits
219
Figure 18.2 Total hCG assays. The illustrations of and are representations; they bear no relationship to the actual structures.
shape change that occurs as a result of nicking on the -subunit, it will generally not detect nicked hCG or nicked hCG missing the CTP. Three antibody combinations are suggested for total hCG assays (Figure 18.2). Each combination uses two antibodies directed to the -subunit. Examples of appropriate antibodies include CCF01, h54, ISOBM-264, -273, -278, -282, -287, -313, and -280. Combination 3 (Figure 18.2) shows an antibody against the core of -subunit (immunogen intact -subunit), and an antibody against a C-terminal fragment of -subunit. Tryptic fragments 123–145, 134–145, and 115–145 have been used as immunogens linked to a carrier protein. The advantage of an antibody against the CTP is that it lacks any cross-reactivity with LH or other glycoprotein hormones. The disadvantage is that the binding site includes as many as four O-linked oligosaccharides. This
220
Human Chorionic Gonadotropin (hCG)
Figure 18.3 Epitopes on the 3D crystal structure of hCG [6]. Epitopes shown are those areas indicated by Berger and colleagues [4].
makes the antibody somewhat carbohydrate-specific, differentially binding regular hCG and hCG-H. The combination of antibodies illustrated in Figure 18.3 recognize hCG, free -subunit, hCG-H, and nicked hCG. It does not, of course, detect molecules missing the CTP. Because of the carbohydrate specificity of Combination 3 (Figure 18.2), some total hCG tests use a synthetic amino acid sequence (residues 130–145) against the CTP [2–4]. As illustrated in Combination 4 (Figure 18.2), this combination should equally detect hCG, free -subunit, hCG-H, and nicked hCG. A new combination using two antibodies directed at the 3D core of -subunit has recently become popular (Figure 18.2, Combination 5). Two commercial total hCG tests use this combination: the Siemens Immulite and the Abbott Architect tests. At the USA hCG Reference Service, we have tried comparing 72 monoclonal antibodies raised against the intact -subunit, checking each of them for pairing in an immunometric assay. We were not successful. As reported by Berger et al. [4], there are multiple core epitopes on the hCG -subunit (Figure 18.3). As such, a combination of antibodies might be feasible (i.e., ISOBM-273, -276, or -285 to one -subunit site with ISOBM-265, -274, or -284 to a second site). We know that some researchers have generated monoclonal antibodies against the innermost part of -core fragment and paired them with a core -subunit antibody [4,5], but finding two independent antibodies against the core is difficult. In Combination 5 (Figure 18.2), both antibodies bind all basic serum and urine variants of hCG -subunit, including hCG, free , hCG, nicked hCG-H, nicked hCG, and -core fragment.
18.2 F ree Subunit, Glycosylation Variant, and Fragment Assay Both monoclonal and polyclonal antibodies have been successfully raised against most hCG variants. Birken and associates [6] generated multiple monoclonal antibodies against a 100% nicked hCG-H preparation. They injected a nicked hCG-H into mice
Antibodies for Intact hCG, for Total hCG, for Free Subunits
221
and were able to generate B152, an hCG-H specific antibody, as well as B151, an antibody specific to nicked hCG [6]. The B152 and B151 antibodies have been paired with core -subunit antibodies to make specific hCG-H only and specific nicked hCG only assays [6,7]. Multiple monoclonal antibodies have also been generated against free -subunit [8–11]. Once again, the intact -subunit was used as immunogen, but this time a clone recognizing the free interface of the -subunit was selected. Antibodies such as FBT11, a free -specific antibody do not detect hCG dimer. As such, pairing these antibodies with an antibody against the core -subunit epitope makes an immunometric assay. Other polyclonal and monoclonal antibodies have been generated against the common free -subunit of the glycoprotein hormones, again using intact -subunit or free -subunit as immunogen [11–13]. These antibodies can be paired with an antibody against core -subunit to make an immunometric assays. Antibodies that specifically detect only -subunit core fragment have also been generated (i.e., B210, INN-hCG-106, and -229). Other antibodies have been generated against -core fragment as immunogen that detect both -core fragment and free -subunit (B201, B204, INN-hCG-64, -68) [14,15]. Each of these antibodies can be paired with an antibody against core -subunit to generate an immunometric assay.
18.3 Generating an Immunoassay Finding the best antibody combination is the most important aspect of designing an hCG or hCG-associated assay. At appropriate antibody dilutions, a microtiter plate immunometric assay can be established. A single, pure antibody can be used as the coating antibody; an antibody linked to an enzyme can be used as the tracer antibody. The tracer antibody can be linked to a color-generating enzyme like horseradish peroxidase or alkaline phosphatase. The coating antibody can be coated on a microtiter plate such as an Immulon plate. An incubation period of 16–24 h in 0.25 M NaHCO3—0.1 M NaCl is necessary for the coating to occur. After coating, the plate should be blocked with 1 mg/ml of bovine serum albumin. The 0.2 ml total volume sample or standard is then added and the plate incubated another 2 h at room temperature. The plate is then washed and 0.2 ml of tracer antibody is added. After an additional 2-h incubation period, the plate is again washed and 0.2 ml of enzyme substrate is added. After generation of a blue or yellow color, absorbance is then read in a microtiter plate-reader. The results are calculated from the standard curve. The same capture and tracer antibody process can be use in an automated assay. They can also be used in a point-of-care or over-the-counter assay, as described in Chapters 22 and 23. It should be noted that every automated assay differs in its procedures, thus standard mechanisms should be followed.
References [1] Vaitukaitis JL, Braunstein GD, Ross GT. A radioimmunoassay which specifically measures human chorionic gonadotropin in the presence of human luteinizing hormone. Am J Obstet Gynecol 1972;113:751–8.
222
Human Chorionic Gonadotropin (hCG)
[2] Caraux J, Chichehian B, Gestin C, Longhi B, Lee AC, Powell JE, et al. Non-crossreactive monoclonal antibodies to human chorionic gonadotropin generated after immunization with a synthetic peptide. J Immunol 1985;134:835–40. [3] Bidart JM, Ozturk M, Bellet DH, Jolivet M, Gras-Masse H, Troalen F, et al. Identification of epitopes associated with hCG and the beta hCG carboxyl terminus by monoclonal antibodies produced against a synthetic peptide. J Immunol 1985;134:457–64. [4] Berger P, Sturgeon C, Bidart J-M, Paus E, Gerth R, Niang M, et al. The ISOBM TD-7 workshop on hCG and related molecules. Tumor Biol 2002;23:1–38. [5] Venkatesh N, Krishnaswamy S, Meuris S, Murthy GS. Epitope analysis and molecular modeling reveal the topography of the C-terminal peptide of the beta-subunit of human chorionic gonadotropin. Eur J Biochem 1999;265:1061–6. [6] Birken S, Krichevsky A, O’Connor J, Schlatterer J, Cole LA, Kardana A, et al. Development and characterization of antibodies to a nicked and hyperglycosylated form of hCG from a choriocarcinoma patient. Endocrine J 1999;10:137–44. [7] Cole LA, Butler SA, Khanlian SA, Giddings A, Muller CY, Seckl MJ, et al. Gestational trophoblastic diseases: 2. Hyperglycosylated hCG as a reliable marker of active neoplasia. Gynecol Oncol 2006;102:150–8. [8] Bidart JM, Troalen F, Salesse R, Bousfield GR, Bohuon CJ, Bellet DH. Topographic antigenic determinants recognized by monoclonal antibodies on human choriogonadotropin beta-subunit. J Biol Chem 1987;262:8551–6. [9] Bidart JM, Troalen F, Lazar V, Berger P, Marcillac I, Lhomme C, et al. Monoclonal antibodies to the free beta-subunit of human chorionic gonadotropin define three distinct antigenic domains and distinguish between intact and nicked molecules. Endocrinol 1992;131:1832–40. [10] Conde IBB, Moreno AIB, Bas CA, Rodríguez OZ, Rodríguez AM, Sanchéz IA, et al. Monoclonal antibody against free -subunit of human chorionic gonadotropin. Hybridoma Hybridomics 2002;21:381–4. [11] Berger P, Klieber R, Panmoung W, Madersbacher S, Wolf H, Wick G. Monoclonal antibodies against the free subunits of human chorionic gonadotropin. J Endocrinol 1990;125:301–9. [12] Madersbacher S, Klieber R, Mann K, Marth C, Tabarelli M, Wick G, et al. Free alphasubunit, free beta-subunit of human chorionic gonadotropin (hCG), and intact hCG in sera of healthy individuals and testicular cancer patients. Clin Chem 1992;38:370–6. [13] Moodley D, Moodley J, Buck R, Haneef R, Payne A. Free alpha-subunits of human chorionic gonadotropin in preeclampsia. Int J Gyn Obstet 1995;49:283–7. [14] Krichevsky A, Birken S, O’Connor J, Bikel K, Schlatterer J, Yi C, et al. Development and characterization of a new, highly specific antibody to the human chorionic gonadotropinbeta fragment. Endocrinol 1991;128:1255–64. [15] Nishimura R, Koizumi T, Das H, Takemori M, Hasegawa K. Enzyme immunoassay of urinary -core fragment of human chorionic gonadotropin as a tumor marker for ovarian cancer. Methods Mol Med 2001;39:135–41.
19 Quantitative hCG Assays Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
There are three basic classifications of quantitative hCG test used today: manual immunometric assays, as used by research and clinical laboratories; automated immunometric assays, as used by clinical laboratories; and competitive radioimmunoassays, as used by research laboratories and specialty clinical facilities. There are also two clear types of quantitative hCG tests used in laboratories. There are intact hCG assays that only measure hCG – dimer, and total hCG tests that measure – dimer and free -subunit, and, to some extent, also measure hCG degradation products: nicked hCG, nicked missing the C-terminal peptide, and -core fragment. The specificity issue is dealt with in Chapter 21. In this chapter, we describe the mechanisms and applications of different quantitative hCG assay.
19.1 Manual Immunometric Assays A simple manual immunometric hCG assay uses a 96-well microtiter plate dish (Figure 19.1), such as an Immulon plate (Thermo-Fisher Scientific, Pittsburgh, PA). Detection can be performed using a microtiter plate detector using an enzyme-labeled antibody and immuno-spectrometric or immuno-flourimetric assessment. Prior to use, plates should be washed using a plate washer or under a faucet. The 96-well plate should be designed so that the wells can be coated with an antibody. A special 0–200 l 8-channel pipette is useful for the coating process (Figure 19.1). Plates should always be coated with a 1- to 20-g/ml antibody concentration. The exact amount depends on the antibody affinity. As described in Chapter 18, the test user will need two antibodies to bind to distant sites on hCG. One antibody will be the coating antibody and the second antibody will be the enzyme-labeled tracer antibody. The tracer antibody should be linked to a tracer enzyme such as peroxidase or alkaline phosphatase. Alternatively, the tracer can be labeled with a radioiodine or other marking agent. Overnight incubation in 0.25 M sodium bicarbonate/0.1 M sodium chloride buffer will coat the microtiter plates with antibody. The binding sites need to be blocked, which is accomplished by incubating the plate for 30 min with 1.0 mg/ml of bovine serum albumin in the same buffer. Plates in blocking buffer can be covered and stored in the refrigerator for up to a week. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00019-0 © 2010 Elsevier Inc. All rights reserved.
224
Human Chorionic Gonadotropin (hCG)
Figure 19.1 Microtiter plate and pipette for manual immunometric assay.
The principle behind the manual immunometric assay is described in Figure 19.2. Samples and standards are added to the microtiter plate. The plate is numbered 1–12 for the columns and labeled A–H for the rows. Generally, columns 1 and 2 are used for the standard curve, and columns 3–12 for the samples. Each sample or standard is added to a well at a volume of 0.1 ml. Samples and standard are diluted using a protein buffer (phosphate-buffered saline or phosphate buffer pH 7.4, containing 0.1 mg/ml ovalbumin or bovine serum albumin). In addition to the 0.1 ml samples or standard, 0.1 ml protein buffer is added to every well, bringing the total volume of all wells to 0.2 ml. If the samples are in serum, then the standard needs to be in serum. This is achieved by adding 0.1 ml normal male serum to the standard curve instead of 0.1 ml protein buffer to each serum sample. Similarly 0.1 ml is added to the standards, which are made in protein buffer. This way, all samples and standard contain 50% serum. The mixture should be incubated in the plates for 2–6 h at room temperature; 2 h for highaffinity coating antibody, 6 h for low-affinity coating antibody. After incubation, plates are carefully washed either using a commercial plate washer or by passing a plate under the faucet and shaking out the water. The plates are then dried by knocking against a clean absorbent surface, such as paper towels. Once the plates are thoroughly dried, 200 l of tracer antibody is added to each of the 96 wells (see Figure 19.2). Tracer antibody is an enzyme-labeled antibody that will bind the antibody-immobilized hCG through a distant binding site. Tracer antibody
Quantitative hCG Assays
225
Figure 19.2 The principle of enzyme-linked spectrometric immunometric hCG assays.
should be prepared in a protein buffer (duplicate or quadruplicate). After incubating 2–3 hours with tracer antibody, plate is again washed and enzyme (tracer) is added substrate is added. After a short incubation plates is read on microtiter plate reader (see Figure 19.2). Ideally, when testing for intact or total hCG, the standard curve should range from 0.1 to 20 ng/ml, or 1 to 200 mIU/ml. Samples should be tested using at least two different concentrations. The pure hCG standard curve should be calibrated against the fourth IS or first reference reagent (RR) international standards (see Chapter 24).
19.2 Automated Immunometric Assays There are numerous brands of automated immunometric tests. They all perform the same procedures as manual assays, but in a robotic manner. In general, the immobilized or coating antibody is bound to a specialized disposable tube or disposable
226
Human Chorionic Gonadotropin (hCG)
bead. Racks of samples are usually placed into the automated immunometric machine. A computer linked to the machine manages the procedures and prints the results. All tests function using the immunometric assay principle (Figure 19.2). Table 19.1 lists the most common automated systems sold in the United States today. The Abbott AxSym and Siemens Dimension RXL tests are enzyme-labeled tracer assays with spectrometric detection. The Siemens Stratus and Tosoh A1A600 test use an enzyme-labeled tracer with fluorimetric detection. The Wallac Delfia uses a europium lanthanide label with fluorimetric detection. The Abbott Architect, Beckman Access, Ortho Vitros, Roche Elecsys, Siemens ACS180, Siemens ADVIA Centaur, and Siemens Immulite assays use a chemilumminescent label and a chememiluminescence detector. Almost every manufacturer of automated immunoassay platforms produces both high and low throughput machines. The Siemens Immulite 1000, for instance, is for low throughput use (approximately 80 tests/hour), whereas the Siemens Immulite 2000
Table 19.1 Properties of Commercial Automated Intact hCG and Total hCG Test. Product
Type of Test
Mechanism
Detection Limit
Abbott Architect
Total hCG
1.2 mIU/ml
Abbott AxSym
Total hCG
Beckman Access 2
Total hCG
Ortho Vitros ECi
Total hCG
Roche Elecsys
Total hCG Intact hCG
Siemens Stratus CS
Total hCG
Siemens ACS180
Total hCG
Siemens ADVIA Centaur Siemens Dimension RXL Siemens Immulite
Total hCG
Tosoh A1A600
Total hCG
Wallac Delfia (Perkins Elmer)
Intact hCG
Two step chemiluminescent assay Microparticle enzyme immunoassay Chemiluminescent immunometric assay Chemiluminescent immunometric assay Electrochemiluminescent immunoassay Electrochemiluminescent immunoassay Fluorimetric enzyme immunoassay Chemiluminescent immunometric assay Chemiluminescent immunometric assay Two step spectrometric immunoassay Chemiluminescent immunometric assay Fluorimetric enzyme immunoassay Europium-labeled fluorimetric immunoassay
Intact hCG Total hCG
Information obtained from company technical support personnel.
2.0 mIU/ml 0.5 mIU/ml 2.39 mIU/ml 0.1 mIU/ml 0.5 mIU/ml
0.5 mIU/ml 2.0 mIU/ml 2.0 mIU/ml 1.0 mIU/ml 1.0 mIU/ml 0.5 mIU/ml 0.5 mIU/ml
Quantitative hCG Assays
227
and 2500 are for higher throughput (approximately 200 tests/hour). We know that the hCG test run on the Immulite 1000 has a longer incubation and can pick up more cross-reacting hCG variants. Those run on the Immulite 2000 and 2500, to accommodate the faster throughput, have shorter antibody incubations and do not detect crossreacting hCG variants as well. As shown in Table 19.1, there are not many intact hCG tests manufactured today. Most have been discontinued because total hCG tests are more comprehensive. They detect free -subunit and degradation products, making them useful for detecting postpartum and cancer-clearing hCG as well.
19.3 Competitive Radioimmunoassays The hCG radioimmunoassay was first described in 1964 [1]. The original hCG radioimmunoassay detected the common -subunit of both hCG and luteinizing hormone. Then, in 1972, Vaitukaitis discovered that generating antibodies against hCG -subunit
Figure 19.3 The principle behind a competitive radioimmunoassay.
228
Human Chorionic Gonadotropin (hCG)
permitted radioimmunoassays to detect hCG alone [2]. The radioimmunoassay was almost totally replaced by the immunometric assay in the 1980s and 1990s. There are still a few centers that continue to use the Vaitukaitis style hCG -subunit radioimmunoassay to this day. We know that Charing Cross Cancer Center, for instance, still uses the radioimmunoassay for monitoring cases of gestational trophoblastic disease, testicular germ cell malignancy, and other cancers. They believe that the hCG -subunit radioimmunoassay detects all forms of hCG -subunit, whereas automated and immunometric assays fail to detect all variants because they use two antibodies. They also claim that the radioimmunoassay is their choice assay for cancer applications [3]. The issue of hCG assay specificity and recognition of hCG variants is discussed in Chapter 21. The principle behind the radioimmunoassay is described in Figure 19.3. A limiting concentration of hCG -subunit polyclonal or monoclonal antibody is placed in a tube. Sample or standard hCG is added to the tube along with a set amount of radioiodine-labeled hCG. The radiodinated hCG and the samples of hCG compete for binding the limiting antibody. The antibody is precipitated by one of a variety of methods and the radioactivity can then be determined. If the samples contain no hCG, then only the radiodinated by the competitive antibody, leading to high radioactivity results. The greater the amount of hCG in the sample or standard, the less radioiodinated hCG will be bound and the lower the radioactivity measurement will be. An inverse log-logit standard curve is observed, with the concentration of hCG increasing with lower radioactivity values.
References [1] Paul WE, Odell WD. Radiation inactivation of the immunological and biological activities of human chorionic gonadotropin. Nature 1964;203:979–80. [2] Vaitukaitis JL, Braunstein GD, Ross GT. A radioimmunoassay which specifically measures human chorionic gonadotropin in the presence of human luteinizing hormone. Am J Obstet Gynecol 1972;113:751–8. [3] Mitchell H, Seckl MJ. Discrepancies between commercially available immunoassays in the detection of tumor-derived hCG. Mol Cell Endocrinol 2007;260:310–13.
20 False Positive hCG Assays Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
False positive hCG results have been a major issue regarding quantitative serum hCG assays. Throughout their history, both immunometric and radioimmunoassays have used animal antibodies to bind hCG [1–3]. Humans produce heterophilic antibodies against human immunoglobulins that can cross-react with animal immunoglobulins. Humans also produce antibodies against the animal antibodies used in vaccinations and other therapies, such as human anti-mouse antibodies (HAMA). Most modern assays use mouse monoclonal antibodies and goat polyclonal antibodies. The presence of HAMA, human anti-animal antibodies (HAAA), and heterophilic antibodies in blood samples can interfere with assays and cause false positive results. Hormones like hydrocortisone, luteinizing hormone (LH), estradiol, and progesterone are always present in serum, thus tests are always positive. They become slightly higher than normal when blood contains interfering HAMA or heterophilic antibodies. In terms of becoming false positive, however, hCG is a unique hormone. When indicating pregnancy, gestational trophoblastic disease, or cancer, hCG is either completely positive or completely negative. When HAMA/HAAA or heterophilic antibodies interfere with an estradiol, progesterone, LH, or hydrocortisone test, slightly elevated results occur, which is a minor issue. In contrast, if hCG is falsely positive due to heterophilic antibodies/HAAA/HAMA, it can create an emergency. False positive results can, for example, be interpreted as positive results indicative of cancer or gestational trophoblastic disease. Heterophilic antibodies and false positive tests can result from a family history of immunoglobulin-A deficiency syndrome or from protein crossing the stomach into the blood, such as in mononucleosis cases [4]. Now we will review the reported false positive hCG test result experiences of the USA hCG Reference Service and other laboratories.
20.1 False Positive hCG Test False positive hCG tests are an old problem. Heterophilic antibodies first became a major false positive concern with the old radioimmunoassay of the 1970s [1,2]. All too often, women were mistakenly identified with pregnancy because of false positive radioimmunoassays. As illustrated in Figure 20.1, radioimmunoassays are considered negative when most of the radioactive hCG is bound by the limiting animal antibody (Figure 20.1A, part 1). When the limiting animal antibody was bound by Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00020-7 © 2010 Elsevier Inc. All rights reserved.
230
Human Chorionic Gonadotropin (hCG)
Figure 20.1 Mechanism of heterophilic antibody interference in radioimmunoassay (A) and immunometric assay (B). The heterophilic antibody is illustrated in gray.
HAMA/HAAA or a heterophilic antibody, the amount of radioactive hCG bound to the antibody was reduced. In so doing, it reduced the radioactivity immobilized by the limiting antibody and falsely indicated a positive test (Figure 20.1A, part 2). As shown, both hCG and HAMA/HAAA/heterophilic antibodies can join the two antibodies together to yield a positive result (Figure 20.1B, part 2). The multi-antibody immunometric test was invented in the early 1980s. It was hoped that these tests would prevent false positive problems in hCG assays [3]. There were no reported troubles until 1999, when false positive cases began to appear throughout the United States [5–10]. The problem created a demand for a laboratory to resolve false positive hCG problems and the USA hCG Reference Service was born. They identified five cases as false positive in 1999, 12 at the end of year 2000, and an alarming 52 cases between 1999 and 2004 (Table 20.1). Of the 52 cases identified, 34 (65%) were given major chemotherapy or surgery for assumed cancer (Table 20.1). Of these 34 cases, 10 young women received a hysterectomy and 3 received a bilateral salpingooophorectomy that left them either sterile or unable to carry a pregnancy [1–6]. The USA hCG Reference Service has established three clear criteria from its experience for identifying false positive hCG (Table 20.1) [5–10]. The first criterion is that there is a negative urine hCG in the presence of positive serum hCG (Table 20.1) [5–10]. Negative urine hCG occurs because HAMA/HAAA and heterophilic antibodies are very large glycoproteins, such as immunoglobulin-A, M, and G. These
Table 20.1 USA hCG Reference Service Diagnosed 83 Cases as False Positive hCG. Immulite Total hCG Serum
96 Well Immulite Total hCG Plate Intact 1:2 hCG Serum
Immulite Total hCG Urine
96 Well Plate Free Serum
96 Well False Positive Plate -Core Assay Used by Serum Referrer
Needless Treatment Given to Patient for Presumed Positive hCG
A. 1999–2004 cases (5 years), false positive quantitative serum hCG assay (n 52) 12 15
8.4 9.2
ND ND
0.15 0.19
0.06 0.07
Abbott AxSym Abbott AxSym
8 11 12 13 16 17 18 20 21 22 24 33 38 50
15 1.0 1.0 1.0 1.0 17 1.0 1.0 1.0 9.4 1.0 1.0 1.0 6.3
8.3 2.1 1.0 1.0 2.3 16 1.0 1.0 1.4 12 1.0 2.3 4.5 4.4
ND 1.0 1.0 1.0 1.0 1.0 1.0 1.0 ND 1.0 1.0 1.0 1.0 1.0
0.19 0.02 0.02 ND 0.11 9.2 0.02 0.02 0.02 0.49 0.02 ND 0.02 0.03
0.06 ND 0.02 0.50 1.1 0.02 0.02 0.02 0.20 0.024 0.02 ND 0.3 ND
Abbott AxSym Abbott Axsym Abbott AxSym Abbott Axsym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym
50 57 60 79 80 81
5.4 23 1.0 1.0 110 1.0
2.1 6.4 1.0 1.7 27 3.4
1.0 1.0 1.0 1.0 1.0 1.0
0.03 0.3 0.02 0.02 0.02 0.02
0.02 0.15 0.02 0.02 80 1.1
Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym
1.0 15
1.0 4.2
1.0 4.8 22
20
Mtx Mtx, ActD, EMA-CO, Hys Mtx Hys, BSO
Mtx, Hys Mtx Mtx, BSO Mtx, ActD Mtx, ActD Hys Mtx Mtx, ActD, EMA-CO, Hys Mtx Mtx, Hys, EMA-CO Mtx, EMA-CO Mtx (continued)
231
6 8
False Positive hCG Assays
Referrers Total hCG Serum
Table 20.1 (Continued)
82 97 120 122 139 151 160 168 190 200 212
15 18 13 1.0 10 9.9 3.6 1.0 1.0 1.0 1.0
275 284 300 300 350 350 351
42 1.0 6.1 1.0 179 1.0 21
36
385 400
13 1.0
3.4
402 1010 40 80 20
1.0 3.3 1.0 1.0 1.0
3.4 15
0.8
0.85 47 6.5
0.2
Immulite Total hCG Urine
96 Well Plate Free Serum
96 Well False Positive Plate -Core Assay Used by Serum Referrer
2.1 16 10 1.0 3.4 5.6 1.0 1.0 1.0 2.4 1.0
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
0.72 0.96 0.24 0.02 0.44 0.02 0.02 0.02 0.02 0.06 0.02
0.12 0.32 0.02 0.02 0.25 2.68 0.04 0.02 ND 1.1 0.02
Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym
1.0 2.1 1.1 1.0 83 2.1 18
1.0 1.0 1.0 1.0 1.0 1.0 1.0
0.02 0.02 0.02 0.02 4.3 0.08 0.9
0.02 0.18 0.13 0.02 1.4 0.55 0.22
Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym Abbott AxSym
15 1.0
1.0 1.0
0.02 0.02
0.17 0.02
Abbott AxSym Abbott AxSym
1.1 1.5 1.0 3.7 4.4
1.0 1.0 1.0 1.0 1.0
0.02 0.02 0.02 1.9 ND
0.02 0.15 ND 0.6 ND
Abbott AxSym Abbott AxSym Beckman Access/2 Ortho Vitros ECI Siemens ACS180
Needless Treatment Given to Patient for Presumed Positive hCG Mtx Mtx Mtx Mtx, ActD Mtx, ActD Mtx, Hys, EMA-CO Mtx Mtx, Hys, EMA-CO, BSO
Mtx Mtx, ActD, EMA-CO, Hys Mtx, ActD, Hys, EMA-CO Mtx Mtx, Hys Mtx Etoposide Mtx
Human Chorionic Gonadotropin (hCG)
Immulite Total hCG Serum
232
Immulite 96 Well Total hCG Plate Intact 1:2 hCG Serum
Referrers Total hCG Serum
1.0 13 1.0 1.0 74
120
2.1 1.6 1.0 1.0 ND
1.0 1.0 1.0 1.0 1.0
0.4 2.16 0.02 ND 1.3
0.22 0.24 0.02 ND 0.43
Siemens Centaur Mtx Siemens Centaur Siemens Centaur Mtx Siemens Dimension Siemens Immuno-1
B. 2005–2009 cases (5 years), false positive quantitative serum hCG assays (n 24) 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.6 1.0 1.0 1.0 1.0
11 11 16 80 14 20 19 11 14.3 11 5.6 7.1
1.0 4.4 1.0 1.0 1.0 1.0 1.0 4.2 4.8 5.7 1.67 5.5
1.0 1.0
1.0 1.9 1.0
1.4 1.0 1.0 1.0 1.0 2.1 1.0 1.7 1.0 1.0 1.0 1.0
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
0.02 0.02 0.02 0.02 ND 0.02 0.02 0.06 0.02 0.02 0.02 0.02
0.02 1.2 ND ND 0.02 1.5 0.31 ND 0.03 ND ND ND
Beckman Access/2 Beckman Access/2 Beckman Access/2 Beckman DXI Ortho Vitros Eci Ortho Vitros Eci Ortho Vitros Eci Roche Elecsys Siemens Centaur Siemens Centaur Siemens Centaur Siemens Centaur
1.0 2.1 1.0 1.6 2.1 1.0 1.0 ND ND ND ND 4.4
1.0 ND 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02
ND ND ND 0.02 ND ND ND ND ND ND ND 0.02
Siemens Centaur Mtx Siemens Centaur Siemens Centaur Siemens Centaur Siemens Dimension Siemens Dimension Mtx Siemens Dimension Siemens Immulite Siemens Immulite Siemens Immulite Siemens Immulite Tosoh Nexia
All results are in IU/l and molar equivalents of hCG (IU/l). Free is free -subunit, -core is -core fragment. ND is not determined.
Mtx, EMA-CO
Mtx Hys
233
84 95 25 50 41 100 9.1 13.2 40 6.1 404 15
False Positive hCG Assays
23 27 37 23 74
234
Human Chorionic Gonadotropin (hCG)
molecules do not cross the glomerular kidney barrier and get into urine. Urine was measured in 71 cases (urine not available in 5 cases). Urine testing identified 71 of 71 (100%) false positive cases and was by far the most sensitive method for detecting false positive hCG. Urine was measured using a sensitive automated immunoassay (Siemens Immulite, sensitivity 1 mIU/ml), and not an insensitive point-of-care test as commonly used in the United States (sensitivity 20 mIU/ml). The second criterion is the observation of different or negative results between the Service’s assay (the Siemens Immulite 1000) and the referring institute’s assay. This identified and confirmed 63 of 76 (83%) cases. The final criterion is the fact that urine -core fragment is made in the kidney and only present in human urine. Positive -core fragment in serum indicated heterophilic antibodies. This was useful in 32 of 56 (57%) measured cases (Table 20.1). We did not find that examining diluted samples was of much use. As shown in Table 20.1, measuring 1:2 diluted samples aided the identification of false positive hCG in just 7 of 23 cases (30%). Laboratories today have a test for heterophilic antibodies in serum. We find that this test is only partially useful in identifying false positive hCG cases. We propose that in order to demonstrate a false positive, at least two of these three methods described here need to show the false positive hCG problem. This was achieved in all USA hCG Reference Service false positive cases (Table 20.1). These methods and the use of heterophilic antibody blocking agents have been described previously [11,12]. Investigations have since shown that the Abbott AxSym test was not appropriately protected against HAMA/HAAA/heterophilic antibodies [13–19]. Abbott added animal serum to its sample diluting buffer to protect against HAMA/heterophilic antibodies, but did not add anything to the antibody preparations. As such, the test was protected using diluted samples, but not when testing undiluted samples with lower hCG concentrations [5–10]. This was the source of their false positive problem. Subsequent to the USA hCG Reference Service’s early publications [5–10], most women receiving needless therapy sued Abbott in individual and class-action lawsuits. This brought the false positive problem to press and to physician attention [16,18,19]. With the suits, the Abbott AxSym moved from being the number one selling assay to a minor selling assay in the United States. With the highly publicized false positive problem laboratories wanted to avoid using the Abbott AxSym hCG test. Abbott eventually fixed their test. Since 2004, the USA hCG Reference Service has not heard about a single false positive result with the Abbott AxSym test or with Abbott’s new platform, the Abbott Architect hCG assay. With all the attention from the press and lawsuits, physicians became weary of false positive cases and started using laboratory tests to check women for HAMA, HAAA, and heterophilic antibodies. As such, the USA hCG Reference Service had only 24 false positive cases referred between 2004 and 2009. It appeared that the crisis was over. The USA hCG Reference Service is aware of false positive results with all major laboratory hCG assays in extreme HAMA/HAAA/heterophilic antibody circumstances. Most quantitative serum hCG tests are protected by animal serum in the antibody preparations, but some HAMA/HAAA/heterophilic antibody cases have high levels of anti-animal antibody, which can still influence otherwise appropriately protected assays and cause false positive results. This was the case with the Siemens
False Positive hCG Assays
235
Centaur, Siemens ACS180, Siemens Dimension, Ortho Vitros, and Beckman Access automated assays. These tests were responsible for 52 out of 76 total false positive quantitative serum hCG assay cases reported during 1999–2004 (Table 20.1). Considering the 76 false positive cases, 39 were given needless chemotherapy or a hysterectomy for an assumed cancer or gestational trophoblastic disease. Since corrections were made to the Abbott AxSym test, we see only the occasional false positive results. We have, however, observed false positive results with the Beckman Access and Beckman Access 2, Beckman DXI, Ortho Vitros ECi, Roche Elecsys, Siemens ACS180, Siemens Centaur, Siemens Dimension RXL, Siemens Immulite 2000, Siemens Immuno-1, and the Tosoh Nexia assays (Table 20.1). Currently, the general occurrence of false positives seems to be at a minimum in laboratory quantitative serum tests, and we hope it will remain this way. Problems are still reported, however, with serum point-of-care tests. These tests are dipstick-like qualitative hCG tests. Because this type of test can only be protected in limited ways, it appears that it might not be appropriate for serum application, and this warrants some caution.
References [1] Vladutiu AO, Sulewski JM, Pudlak KA, Stull CG. Heterophilic antibodies interfering with radioimmunoassay. J Am Med Assoc 1982;248:2489–90. [2] Hussa RO, Rinke ML, Schweitzer PG. Discordant human chorionic gonadotropin results: causes and solutions. Obstet Gynecol 1985;65:211–19. [3] Armstrong EG, Ehrlich PH, Birken S, Canfield R. Use of a highly sensitive and specific immunoradiometric assay for detection of human chorionic gonadotropin in urine of normal, nonpregnant, and pregnant individuals. J Clin Endocrinol Metab 1984;59:867–74. [4] Knight AK, Bingemann T, Cole L, Cunningham-Rundles C. Frequent false positive beta human chorionic gonadotropin in immunoglobulin a deficiency. Clin Exp Immunol 2005;141:333–7. [5] Cole LA, Rinne KM, Shahabi S, Omrani A. False positive hCG levels leading to unnecessary surgery and chemotherapy, and needless occurrences of diabetes and coma. Clin Chem 1999;45:313–14. [6] Butler SA, Cole LA. Falsely elevated hCG leading to unnecessary therapy. Obstet Gynecol 2002;99:515–16. [7] Khanlian SA, Smith HO, Cole LA. Persistent low levels of hCG: a pre-malignant gestational trophoblastic disease. Am J Obstet Gynecol 2003;188:1254–9. [8] Cole LA. Problems with hCG measurements and interpretation of hCG results in gestational trophoblastic diseases. Jpn J Obstet Gynecol 2003;20:11–14. [9] Cole LA. Phantom hCG and phantom choriocarcinoma. Gynecol Oncol 1998;71:325–9. [10] Rotmensch S, Cole LA. False diagnosis and needless therapy of presumed malignant disease in women with false positive human chorionic gonadotropin concentrations. Lancet 2000;355:712–15. [11] Olsen TG, Hubert PR, Nycum LR. Falsely elevated human chorionic gonadotropin leading to unnecessary therapy. Obstet Gynecol 2001;98:843–5. [12] Flam F, Hambraeus-Jonzon K, Hansson L, Kjaeldgaard A. Hydatidiform mole with nonmetastatic pulmonary complication and false low level of hCG. Eur J Obstet Gynecol Reprod Biol 1998;77:235–7.
236
Human Chorionic Gonadotropin (hCG)
[13] Chu J-W, Golman JO. False elevation of serum hCG. Laboratory updates: Detroit Medical Center University Laboratories; 2003; http://www.dmc.org/univlab/jau_janice.htm. [14] Rode L, Daugaard G, Fenger M, Hilsted L, Krag-MØller L, Raaberg L, et al. Serum hCG: still a problematic marker. Acta Obstet Gynecol Scand 2003;82:199–202. [15] Trojan A, Joller-Jemelka H, Stahel RA, Jacky E, Hersberger M. False positive hCG in a patient with germ cell cancer. Oncol 2004;66:336–8. [16] Pesce MA. False elevated hCG results with the AxSym assay. Clin Chem 2003;49:A92. [17] Esfandiari N, Goldberg JM. Heterophile antibody blocking agent to confirm false positive serum human chorionic gonadotropin assay. Obstet Gynecol 2003;101:1144–6. [18] Giannopoulos P, Jones G, Lim E-M. A simple protocol for identification of heterophilic antibody interferences in the Abbott AxSym hCG assay. Aust Assoc Clin Chem 2003. Gold Coast, Queensland. Abstract. [19] Cole LA, Khanlian SA. Easy fix for clinical laboratories for the false positive defect with the Abbott AxSym total -hCG test. Clin Biochem 2004;37:344–9.
21 Specificity of Different hCG Assays Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
This and the following two chapters discuss clinical laboratory serum hCG tests and point-of-care (POC) and over-the-counter (OTC) pregnancy tests. All these tests seem so well designed and might appear to be perfect; however, as we deal with the specificities of these tests, it becomes apparent that they are not perfect. Hyperglycosylated hCG (hCG-H) is the principal hCG form present during the first weeks of pregnancy testing. Two clinical laboratory serum hCG tests do not appear to be suitable for early pregnancy testing because they poorly detected hCG-H. The bulk of clinical laboratory tests, in contrast, do not detect the degradation product hCG missing the -subunit C-terminal peptide (CTP), which makes them unsuitable for monitoring cancer and gestational trophoblastic disease patients. Radical statements need to be me made by manufacturers that assays are not suitable for early pregnancy and not appropriate for monitoring cancer and gestational trophoblastic disease. One then wonders, what use does the assay have? Why do manufacturers produce such useless tests. In this chapter, we will analyze the suitability of different blood and urine hCG tests for pregnancy detection and for monitoring cancer and gestational trophoblastic disease cases. Considering POC and OTC tests, we focus on pregnancy applications. The majority of devices poorly detected hCG-H, which means they are not appropriate early pregnancy tests. In other words, they do not do what they were designed to, and this is the issue dealt with in this chapter. An “X” on a physician order form does not tell the laboratory whether the tests are detecting pregnancy or monitoring cancer or gestational trophoblastic disease, so a laboratory cannot exclude these applications. A physician orders a test regardless of the limitation of specificity. This is not a problem that physicians are familiar with. It is the responsibility of the laboratory director to make sure that the laboratory offers an hCG assay that can appropriately deal with all physicians needs. All the data on clinical laboratory tests described here have been previously published [1–7], as have the data regarding POC [8,9] and OTC testing devices [8,10,11]. All data were determined blindly. Clinical laboratory serum tests were evaluated at 14 independent laboratories in the United States and Canada with blinded samples. Both POC and OTC devices were evaluated blindly (devices and urine samples coded) at the USA hCG Reference Service. No money was received from any company for performing this testing, so it has no commercial or other bias. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00021-9 © 2010 Elsevier Inc. All rights reserved.
238
Human Chorionic Gonadotropin (hCG)
21.1 Clinical Laboratory Tests We investigated 10 major automated assays used in the United States (1% of laboratory market as indicated by College of American Pathologists Excel program). Two examples of manual assay kits were also investigated, along with an example of the hCG radioimmunoassay. As shown in Table 21.1, devices were evaluated with hCG-H, nicked hCG, hCG missing the CTP, and hCG free -subunit standards added to nonpregnant female serum. All tests were calibrated in nanomoles per liter, which was converted to equivalents of milli-international units per milliliter (1 pmol/ ml hCG-H 407 mIU/ml hCG equivalents; 1 pmol/ml free -subunit 244 mIU/ml hCG equivalents). As shown (Table 21.1), 3 out of 13 assays poorly detected hCG-H, and two assays overdetected hCG-H. hCG-H accounted for 90 11% of the total hCG (hCG plus hCG-H plus free -subunit) produced in the third complete week of pregnancy (the week following implantation) and 54 27% of the total hCG in serum in the fourth complete week of pregnancy (the week following missing the menstrual period) [4]. Table 21.1 Detection of hCG-H Serum Standard (465 IU/l hCG Equivalency), Nicked hCG (3930 IU/l hCG Equivalency), hCG Missing CTP (hCG Minus CTP; 260 IU/l hCG Molar Equivalency), and hCG Free -Subunit (Free ) Serum Standard (3840 IU/l hCG Molar Equivalency in Serum). hCG-H Nicked hCG mIU/ml (%) mIU/ml (%)
hCG Minus CTP hCG Free mIU/ml (%) mIU/ml (%)
400 (86%) 630 (135%) 370 (80%) 770 (165%) 385 (83%) 474 (102%) 500 (108%) 232 (50%) 467 (100%) 449 (97%)
4071 (104%) 4042 (103%) 2780 (71%) 2736 (70%) 3691 (94%) 3097 (79%) 3048 (78%) 3658 (93%) 4240 (108%) 3827 (97%)
<2 (<1%) <2 (<1%) <2 (<1%) <2 (<1%) 190 (73%) <2 (<1%) <2 (<1%) <2 (<1%) 273 (105%) <2 (<1%)
3278 (85%) 5841 (152%) 4445 (116%) 3220 (84%) 3744 (98%) 5235 (136%) 5535 (144%) 3300 (86%) 3843 (100%) 3187 (83%)
17 (3.6%) 249 (64%)
908 (23%) 4007 (102%)
147 (54%) <2 (<1%)
123 (3.2%) 1625 (42%)
3280 (77%)
182 (70%)
2767 (73%)
A. Popular Tests Abbott AxSym Beckman Access 2 Ortho Vitros ECi Roche Elecsys Siemens Stratus CS Siemens ACS180 Siemens ADVIA Centaur Siemens Dimension RXL Siemens Immulite Tosoh A1A600
B. Manual Kits Serono MAIAclone Wako hCG-CTP
C. Specialized Assay Radioimmunoassay hCG 406 (87%)
Assays yielding results more than 25% different from actual concentrations are shown in parentheses in bold, italic font. Samples were evaluated blindly by 14 independent commercial laboratories in the United States and Canada.
Specificity of Different hCG Assays
239
This is the period during which most women test for pregnancy. We wonder whether tests like the Siemens Dimension and Serono MAIAclone (both of which only detect hCG-H with 50% and 3.6% efficiency, respectively) can really be called pregnancy tests because of their ineffectiveness in detecting early pregnancy. How many thousands of women were analyzed with these tests and mistakenly diagnosed as being not pregnant? Examining nicked hCG (Table 21.1), the Serono MAIAclone assay detects this molecule with 23% efficiency. Nicked hCG is most evident during clearance of hCG after parturition of pregnancy. It is particularly evident during clearance following evacuation of hydatidiform mole. Care is needed using this assay in cancer and gestational trophoblastic disease cases. The Siemens Immulite assay is one of the few assays that appropriately detect hCG missing the CTP (Table 21.1). Surprisingly, poor detection is noted with the manual hCG radioimmunoassay. hCG missing the CTP is occasionally detected in pregnancy, particularly during clearing of hCG following parturition. It is most commonly detected in monitoring cancer cases. Seemingly, only the Siemens Immulite assay is appropriate for managing these cases. By definition, a total hCG test is a test that detects hCG and its free -subunit. Commonly, nontrophoblastic malignancies and particularly bladder cancers and ovarian dysgerminomas produce only free -subunit [12]. As shown in Table 21.1, several tests either poorly detect or exaggerate actual free -subunit concentrations; however, the Serono MAIAclone and the Wako hCG-CTP detect free -subunit with 3.2% and 42% efficiency, respectively. Free -subunit measurement might also be critical in detecting recurrence of choriocarcinoma [12]. Figures 21.1–21.5 illustrate the importance in detecting hCG-H, nicked hCG, hCG missing the CTP, and free -subunit.
Figure 21.1 Serum samples (33 total) collected in the 2 weeks following implantation of pregnancy. Total hCG was measured using the Siemens Immulite 1000 assay and hCG-H measured using the B152 microtiter plate assay. Solid circles are total hCG (regular plus hCG-H) and clear circles are hCG only (total hCG less hCG-H).
240
Human Chorionic Gonadotropin (hCG)
Figure 21.2 Choriocarcinoma case monitored with total hCG (Siemens Immulite 1000) (solid black circles) and Serono MAIAclone hCG assay (white circle with black line). The Serono assay poorly detects hyperglycosylated hCG.
Figure 21.1 shows 22 serum samples from 3 to 5 weeks of pregnancy (the time when hCG-H predominates). Black circles are total hCG (hCG-H plus hCG plus free -subunit). Clear circles are hCG only. The results yielded by the Serono MAIAclone test (3.6% detection of hCG-H) would be similar to those indicated by the clear circles (hCG only and not total hCG). The results yielded by the Siemens Dimension test (50% detection of hCG-H) would fall approximately halfway between the concentrations indicated by the white and black circles (see Figure 21.1). The samples detected by the Siemens Immulite (100% detection of hCG-H) are the black circles. As indicated, an assay with poor hCG-H sensitivity would poorly detect the pregnancies illustrated in Figure 21.1. Considering the worst case of only 3.6% detection of hCG-H, if the assay had a sensitivity of 1 mIU/ml, it would detect only 15 of 33 (45%) of total early pregnancies. Choriocarcinoma produces primarily or only hCG-H [12]. It is classed and treated according to the World Health Organization’s protocol for measuring the level of hCG in serum. Figure 21.2 compares the hCG-H sensitivity of the Siemens Immulite (100% detection) with the Serono MAIAclone (3.6% detection) in monitoring a choriocarcinoma case. As shown, unduly low values are recorded with the Serono MAIAclone, approximately 1/50th of the actual values. The patient with the low hCG values, as determined by the Serono MAIAclone test, would incorrectly receive methotrexate chemotherapy. Similarly, the patient with the higher values, as determined by the Siemens Immulite test, would receive a more potent 5-agent combination therapy. Ovarian dysgerminoma cases, or a nontrophoblastic neoplasm, only produce the free -subunit of hCG [12]. Figure 21.3 illustrates the levels of free -subunit
Specificity of Different hCG Assays
241
Figure 21.3 Patient with dysgerminoma (germ cell ovarian malignancy) monitored with Siemens Immulite 1000 intact hCG assay (black circles) and with B109 intact hCG assay (white circles with black line).
detected in a patient undergoing chemotherapy for treatment of ovarian dysgerminoma. As shown, the Siemens Immulite assay successfully detected 100% of free -subunit concentrations, but negative results (0% detection of free -subunit) were recorded with the B109 microtiter plate intact hCG assay. Thus, it is assumed that assays with poor free -subunit detection would give results between those of the B109 and Immulite assays. Figure 21.4 illustrates a patient with partial hydatidiform mole. Samples were taken during the weeks following dilation and curettage. As illustrated and commonly observed, when hCG drops below 100 mIU/ml in hydatidiform mole cases, it is nicked and gives very different results when measured with an assay that does not detect nicked hCG. Questionable results would be expected with the Roche Elecsys (70% detection of nicked hCG) and once again with the Serono MAIAclone (23% detection of nicked hCG) assays. Figure 21.5 illustrates the importance of detecting hCG that is missing the CTP. Shown are the data collected by the USA hCG Reference Service using Siemens Immulite assays (100% detection). A referring laboratory collected data using Siemens ACS180 assays (1% detection). As observed, three recurrences of disease are completely missed by the Siemens ACS180. Similar results can be expected using other assays that do not detect the CTP (Abbott AxSym, Beckman Access, Ortho Vitros ECi, Roche Elecsys, Siemens ADVIA Centaur, Siemens Dimension RXL, and Tosoh A1A600 assays). The assays that do detect the CTP are the radioimmunoassay (70% detection), Siemens Stratus CS (73% detection), and the Siemens Immulite (105% detection) assay (see Table 21.1). In summary, the Serono MAIAclone and Siemens Dimension should be avoided for early pregnancy testing. All other assays are appropriate for detecting early
242
Human Chorionic Gonadotropin (hCG)
Figure 21.4 Patient with partial hydatidiform mole monitored in the weeks after evacuation of mole with the Siemens Immulite 1000 total hCG test (clear circles with black line) and the B109 microtiter plate assay (black circles with dotted line). Whereas the Immulite assay detects nicked hCG, the B109 assay does not detect nicked hCG.
Figure 21.5 Management of a choriocarcinoma case using the Siemens Immulite total hCG assay (black circles with dotted line) and the Siemens ACS180 assay (clear circles with black line). The ACS180 assay does not detect hCG -subunit C-terminus.
pregnancy. For gestational trophoblastic disease and cancer cases, only the Siemens Immulite is recommended because of its appropriate recognition of degradation products like hCG missing the CTP. The radioimmunoassay can be considered a second choice for this application.
Specificity of Different hCG Assays
243
21.2 Point-of-Care Tests POC tests are rapid (3–5 min) urine pregnancy tests used in physician’s offices, pregnancy clinics, and clinical laboratories. They have very limited uses in monitoring cancer and gestational trophoblastic disease cases. As such, we can only examine these tests strictly as pregnancy tests, and disregard the sensitivities needed for gestational trophoblastic disease and cancer applications. As shown in Table 21.2, these tests vary in sensitivity from 11 to 88 mIU/ml. The U.S. Food and Drug Administration (FDA) demands that urine hCG tests do not exceed a sensitivity of 20 mIU/ml. The FDA’s primary concerns are biochemical pregnancies and positive pregnancy tests in women in perimenopause and menopause (who normally produce pituitary hCG). Stenman et al. [13] suggest a similar sensitivity limitation for the assays that are used to detect biochemical pregnancies and diagnose perimenopausal and menopausal women. USA hCG Reference Service recently performed a study of background urine hCG [14] using various POC tests. The College of American Pathologists suggests that the 97.5th percentile of total background urine hCG concentrations is an appropriate cutoff for diagnosing biochemical pregnancies and perimenopause and menopause cases. We examined 11,991 urine samples from women during their menstrual cycles, from women with biochemical pregnancies (incidence 1 in 4 pregnancies) and from perimenopausal and menopausal women. We found that the overall urine Table 21.2 Sensitivity of OTC and POC Pregnancy Tests. Device
Sensitivity for hCG
Sensitivity for hCG-H
Sensitivity for Free
11 IU/l 11 IU/l 11 IU/l 11 IU/l 5.5 IU/l 5.5 IU/l
44 IU/l 44 IU/l 44 IU/l 44 IU/l 5.5 IU/l 5.5 IU/l
Not Detected Not Detected Not Detected Not Detected 5.5 IU/l 5.5 IU/l
11 IU/l 22 IU/l 22 IU/l 22 IU/l 5.5 IU/l 44 IU/l
11 IU/l 44 IU/l 44 IU/l 44 IU/l 22 IU/l 88 IU/l
88 IU/l Not Detected Not Detected Not Detected Not Detected 44 IU/l
A. OTC Pregnancy Tests ClearBlue Easy ClearBlue Easy Digital e.p.t e.p.t Certainty Digital First Response Early Result First Response Gold Digital
B. POC Pregnancy Tests Beckman-Coulter Icon 25 Inverness Acceava hCG Basic II Mainline Maxie hCG Mainline Confirms hCG Quidel QuickVue One-Step Siemens Clinitest, Digital Analyzer
Four devices were tested at each of five concentrations of pure first Reference Reagent hCG (88, 44, 22, 11, 5.5 mIU/ml), five matching concentrations of hCG-H (hCG equivalency in IU/l), and five matching concentrations of hCG free -subunit (hCG molar equivalency of IU/l). All standards were prepared in normal male urine. Lowest concentrations leading to 4 of 4 consecutive positive results are reported.
244
Human Chorionic Gonadotropin (hCG)
hCG cutoff (or 97.5th percentile) was 1.2 mIU/ml. We concluded that the concerns expressed by Stenman et al. [13] and by the FDA seemed to be needless and excessively cautious. We evaluated six random POC devices. Two tests, The Beckman-Coulter Icon 25 and the Quidel QuickVue One-Step (sensitivities 11 and 22 mIU/ml, respectively), equally detected hCG and hCG-H levels. These tests were determined to be appropriate for detecting early pregnancy. The other four tests (Inverness Acceava hCG, Mainline Maxie hCG, Mainline Confirms hCG, and Siemens Clinitest hCG) failed to appropriately detect hCG-H. These tests should be avoided in early pregnancy testing and should be limited to testing after 6 weeks gestation.
21.3 Over-the-Counter Tests OTC, or home, tests are also rapid (3–5 min) urine early pregnancy detection tests. They do not have any specified use in cancer or gestational trophoblastic disease cases. As such, we can examine OTC tests strictly as pregnancy tests and disregard the sensitivities needed for gestational trophoblastic disease and cancer applications. As shown in Table 21.2, OTC tests are generally more sensitive than POC tests. They vary in sensitivity from 5.5 to 11 mIU/ml. Stenman et al. [13] suggests that these tests should have a lower sensitivity of 25 mIU/ml when being used to detect biochemical pregnancies and diagnose perimenopause or menopause. We also conducted a large study of background urine hCG [14], using various OTC tests. Considering biochemical pregnancies and perimenopause and menopause cases, the 97.5th percentile was just 1.2 mIU/ml. In this regard, are the concerns expressed by Stenman et al. [13] justified? We evaluated the six biggest-selling OTC tests sold in the United States. Two of these tests (First Response Early Results and First Response Gold Digital) equally detected hCG and hCG-H. These tests are very appropriate for detecting early pregnancies; however, we question the reliability of the other four tests prior to 6 weeks gestation (2 weeks after missing the menstrual period). The ClearBlue Easy and First Response pregnancy tests claim accurate pregnancy detection up to 4 days prior to missing the menstrual period. Because First Response equally detects both hCG and hCG-H, this test might be appropriate for detecting early pregnancies. ClearBlue Easy, on the other hand, poorly detects hCG-H and thus is inappropriate for detecting early pregnancies.
References [1] Cole LA, Seifer DB, Kardana A, Braunstein GD. Selecting human chorionic gonadotropin immunoassays: consideration of cross-reacting molecules in first trimester pregnancy serum and urine. Am J Obstet Gynecol 1993;168:1580–6. [2] Cole LA, Shahabi S, Butler S, Mitchell H, Newlands ES, Behrman HR, et al. Utility of commonly used commercial hCG immunoassays in the diagnosis and management of trophoblastic diseases. Clin Chem 2001;47:308–15.
Specificity of Different hCG Assays
245
[3] Cole LA, Sutton JM, Higgins TN, Cembrowski GS. Between-method variation in hCG test results. Clin Chem 2004;50:874–82. [4] Cole LA. Human Chorionic Gonadotropin and Associated Molecules. Expert Rev Molec Diag 2009; 9:51–73. [5] Sturgeon CM, Berger P, Bidar J-M, Birken S, Burns C, Norman RJ, et al. Differences in recognition of the 1st WHO international references reagent for hCG-related isoforms by diagnostic immunoassays for human chorionic gonadotropin. Clin Chem 2009;55:1484–91. [6] Rinker AD, Tietz NW. Beta-hCG vs. intact hCG assays in the detection of trophoblastic disease. Clin Chem 1989;35:1799–800. [7] Higgins TN, Hanna AN, Hofer TL, Cenbrowski GS. Measurement of inaccuracy and imprecision of HCG methods using dilutions of the WHO 4th IS-HCG standard and a pregnant patient’s serum. Clin Biochem 2004;37:152–4. [8] Cole LA. The utility of over-the-counter (home) pregnancy test: comprehensive study. US Pharm 2010; in press. [9] Stenman U-H, Tiitinen A, Alfthan H, Valmu L. The classification, functions and clinical use of different isoforms of HCG. Hum Reprod Update 2006;12:1–16. [10] Cole LA, Sutton-Riley JM, Khanlian SA, Borkovskaya M, Rayburn BR, Rayburn WF. Sensitivity of over-the-counter pregnancy tests: comparison of utility and marketing messages. J Am Pharm Assoc 2005;45:608–15. [11] Cole LA, Khanlian SA, Sutton JM, Davies S, Rayburn WF. Accuracy of home pregnancy tests at the time of missed menses. Am J Obstet Gynecol 2004;190:100–5. [12] Muller C, Cole LA. The quagmire of hCG and hCG testing in gynecologic oncology. Gynecol Oncol 2009;112:663–72. [13] Stenman U-H, Alfthan H. Optimal sensitivity for pregnancy tests. The most sensitive pregnancy tests are not necessarily the most accurate. IVD Technol 2003; http://www .devicelink.com/ivdt/archive/03/04/010.html. [14] Cole LA, Ladner DG. Background hCG in non-pregnant individuals: need for more sensitive point-of-care and over-the-counter pregnancy tests. Clin Biochem 2009;42:168–75.
22 Point-of-Care Pregnancy Tests Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
Point-of-care (POC) pregnancy tests are available at physician’s offices and prenatal care clinics around the world. They are also used as urine hCG tests in clinical laboratories. Most of today’s POC tests work on the dual antibody immunometric principle. Most can be used with urine or serum samples. They are qualitative assays with a sensitivity ranging from 20 to 100 mIU/ml hCG. Figure 22.1 shows the workings of a modern POC hCG test. Multiple manufacturers produced POC hCG tests (Figure 22.2). These tests range from urine-drop devices to dipstick and
Figure 22.1 Principle operation of modern point-of-care tests. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00022-0 © 2010 Elsevier Inc. All rights reserved.
248
Human Chorionic Gonadotropin (hCG)
Figure 22.2 Examples of point-of-care tests. Mainline confirm test (A), Beckman-Coulter Icon 25 (B), Quidel QuickVue (C), Inverness Acceava (D), and Siemens Clinitest (E). Every test shown is positive, with a clear control line “C” and test line “T” shown in windows. Table 22.1 Sensitivity and Specificity of Point-of-Care Pregnancy Tests. Device
hCG Standard IU/l
hCG-H Standard IU/l
Free -subunit Standard IU/l
Quidel QuickVue One-Step Beckman-Coulter Icon 25 Siemens Clinitest, Digital Analyzer Inverness Acceava hCG Basic II Mainline Maxine hCG Mainline Confirms hCG
24 12 50 18 24 24
24 12 100 50 50 50
88 88 500 500 500 500
Six devices were tested at each of six concentrations of hCG (first Reference Reagent): 6, 12, 24, 50, 100, and 500 IU/l. Lowest concentrations leading to six of six positive results are reported. hCG-H is hyperglycosylated hCG.
Digital devices (Figure 22.2). The principal manufacturers of multiple POC products include Quidel Corp. (San Diego, CA), Beckman-Coulter Corp. (Fullerton, CA), and Inverness Medical Innovations Inc. (Waltham, MA). Other manufacturers include Mainline Inc. (Ann Arbor, MI), Siemens Healthcare Diagnostic Ltd. (New York, NY), Genzyme (Cambridge, MA), Abbott Rapid Diagnostics Ltd. (Abbott Park, IL), SCI International (Frederick, MD), and multiple others.
22.1 Point-of-Care Tests We have carefully evaluated the specificities, sensitivities, and abilities to detect pregnancy of commonly used POC products [1]. We examined six commonly used POC tests. Each of these six tests is a representation of the multiple types of test currently used at physician’s offices, prenatal care clinics, and clinical laboratories. As shown in Table 22.1, the POC devices vary widely in sensitivity for regular hCG, ranging from 12 to 50 IU/l. The Quidel QuickVue and the Beckman-Coulter Icon 25 tests equally detect regular and hyperglycosylated hCG (hCG-H); however, the Mainline
Point-of-Care Pregnancy Tests
249
Maxie and Mainline Confirms, Inverness Accreava hCG Basic II, and the Siemens Clinitest tests poorly detect hCG-H relative to hCG (Table 22.1). As stated, a test that poorly detects hCG-H is not appropriate for testing the first 6 weeks of pregnancy (see Chapters 21 and 23). None of the POC devices tested detected free -subunit. We investigated the pregnancy detection sensitivity of various POC devices [1]. The Beckman-Coulter Icon 25 test had equal detection of regular hCG and hCG-H and faired the best (Table 22.1). This test had the best sensitivity, detecting 77% of the 30 day-of-missing-menses pregnancy urines (Table 22.2). The Quidel QuickVue, Table 22.2 Evaluation of POC Tests with 30 Urine Samples from Women on the Day of Missing their Menstrual Period. Total hCG in IU/l
Quidel QuickVue One-Step
BeckmanCoulter Icon 25
Siemens Clinitest
Inverness Acceava Basic II
Mainline Maxie hCG
Mainline Confirms hCG
41 142 56 179 138 24 103 61 31 224 34 176 18 68 135 5.8 89 33 8.9 260 85 103 32 150 122 173 31 147 37 196 Mean
1 1 1 1 1 0 1 1 0 0 1 0 0 1 1 0 0 1 0 1 1 1 0 1 1 1 0 1 1 1 67%
1 1 1 1 1 0 0 1 0 0 1 1 1 1 1 0 0 1 0 1 1 1 1 1 1 1 1 1 1 1 77%
1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 0 0 0 0 1 0 1 0 1 1 0 0 1 0 1 43%
0 1 1 1 1 0 0 1 1 0 1 1 0 1 1 0 0 1 0 1 1 1 1 1 1 1 0 1 1 0 67%
0 0 1 1 1 0 1 0 0 0 0 0 0 1 1 0 0 1 1 1 0 1 0 1 1 1 0 1 0 1 50%
0 0 1 1 1 0 0 0 0 0 0 0 0 1 1 0 0 0 0 1 0 1 0 1 1 1 0 1 0 1 40%
Calculated from length of three previous menstrual cycles. Devices were run exactly as specified by manufacturer (Siemens device measured in Digital device analyzer), 0 indicates negative results and 1 indicates positive results.
250
Human Chorionic Gonadotropin (hCG)
which also equally detects regular hCG and hCG-H, faired reasonably and detected 67% of pregnancies on the day of missing menses (Table 22.2). The other four POC devices analyzed (Siemens Clinitest, Digital; Inverness Acceava; Mainline Maxie; and Mainline Confirms) poorly detected hCG-H (Table 22.1) and detected just 43%, 67%, 50%, and 40% of pregnancies on the day of missing menses, respectively. All POC tests claim 99% accuracy and claim use as early as 3 days prior to the day of missing menses. The results of this study suggest that the 99% claims are erroneous and confusing. The hook effect is a common problem with POC tests. The hook effect occurs when an anti- core antibody in the POC device is combined with an anti--subunit antibody. Excess amounts of free -subunit in urine and serum samples can completely occupy the -subunit antibody binding site, blocking any region for hCG to bind. This leads to a negative pregnancy result. The hook effect can occur at any point during pregnancy. Similarly, if tests use a -subunit antibody to site 1 and a -subunit antibody to site 2 a hook effect can occur. -core fragment can predominate in urine and fully occupy one of the antibody sites, which can cause a false negative test. This problem primarily occurs during the second and third trimesters of pregnancy [2].
References [1] Cole LA. Inappropriate pregnancy tests. Int J Gynecol Obstet 2010; in press. [2] Gronowski AM, Cervinski M, Stenman UH, Woodworth A, Ashby A, Scott MG. False negative results in point-of-care qualitative human chorionic gonadotropin (hCG) devices due to excess hCG -core fragment. Clin Chem 2009;55:1885–1886.
23 Over-the-Counter Pregnancy Tests Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
In 1976, the FDA approved the first home, or over-the-counter (OTC), pregnancy test. The product, called e.p.t. (abbreviation of “early pregnancy test”), claimed to detect pregnancy 4 days after missing the menstrual period [1]. Since then, OTC devices have undergone many changes, including the adaption of the immunometric assay format [2], which all OTC tests now use. Many of today’s OTC tests are digital. Instead of relying on the presence or absence of a faint test line to determine pregnancy, digital results are clearly stated: “Yes” or “No;” “Pregnant” or “Not pregnant.” As examined in Chapter 21, not all OTC tests detect the same things. Some tests, like the Church and Dwight (Princeton, NJ) devices First Response and Answer, equally detect hCG, hyperglycosylated hCG (hCG-H), and free -subunit. Other tests like the analogous Johnson & Johnson (Skillman, NJ) e.p.t and the Inverness (Waltham, MA) ClearBlue Easy devices detect hCG, but hCG-H is poorly detected and free -subunit is not detected at all. To conduct studies, the USA hCG Reference Service purchased and tested OTC devices from 12 pharmacies in Albuquerque, NM (three Walgreens stores, three CVS pharmacies, two Wal-Mart pharmacies, two grocerybased pharmacies, and two independent pharmacies). These tests include First Response and First Response Gold Digital, ClearBlue Easy Manual and ClearBlue Easy Digital, e.p.t Manual and e.p.t Certainty Digital, and other generic-brand devices. In this chapter, we examine and explore the advantages and disadvantages of these commercial products.
23.1 How OTC Tests Work The principle behind immunometric OTC tests is illustrated in Figure 23.1. Urine is rapidly absorbed by a porous nitrocellulose stick. The urine concentrates as it enters a plastic-covered section of the device at an interface of high-/low-porosity nitrocellulose. The concentrated urine is then mixed with an antibody against hCG site 1 and labeled with a blue, red, or gold dye. The mixture of dyed antibody and concentrated urine slowly rises inside the plastic-covered section of the device. When it reaches the windowed area of the device, the mixture is exposed to an immobilized antibody Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00023-2 © 2010 Elsevier Inc. All rights reserved.
252
Human Chorionic Gonadotropin (hCG)
Figure 23.1 Principle of modern OTC tests. Flowing urine is rapidly concentrated as it reaches interface of high-/low-porosity nitrocellulose under plastic of device. Urine is then mixed with a dyed/colored antibody against hCG site 1, and incubates as they slowly rise in the stick. The first immobilized band in the window of the stick is the antibody against hCG site 2. If hCG is present, a colored line forms (test line). This creates the dyed antibody site 1: hCG: immobilized antibody site 2. The dyed antibody continues to rise and is immobilized in the second dye band (control line) by an immobilized anti-animal immunoglobulin.
against hCG site 2. If hCG is present in the mixture, it would bind the freely mobile site-1 dyed hCG antibody. The dyed complex then binds the stationary hCG antibody against hCG site 2. This creates a dyed antibody-1: hCG: antibody 2 complex, which then forms a dyed colored test line at the site of the immobilized antibody and indicates pregnancy. Any residual antibody would be bound by a control line containing stationary anti-animal immunoglobulin, forming an additional dyed/colored line, indicating that the device worked appropriately. As such, a single control line indicates a negative test, and an additional result line indicates a positive test. Figure 23.2 shows examples of positive and negative results on various OTC devices.
23.2 Accuracy of OTC Tests For the first study [3], the USA hCG Reference Service prepared a concentrate resembling urine at 6 weeks of pregnancy. This was made by mixing on a molar basis 50% regular hCG (recombinant CHO-cell hCG), 30% hCG-H (batch C5), 10% free -subunit, and 10% -core fragment into pools of healthy, nonpregnant female urine [3]. The urine pools were prepared at 100, 50, 25, 12.5, and 6.3 mIU/ml total hCG concentration and were heated to 37 °C in an incubator before each test. Each device
Over-the-Counter Pregnancy Tests
253
Figure 23.2 Illustration of results windows. Manual ClearBlue Easy device on left is negative; other two devices are positive (A); Manual First Response device on the right is positive; other two devices are negative (B); and of the three digital devices, e.p.t Certainty, First Response, and ClearBlue, only the middle device is positive (C).
was tested exactly according to manufacturers’ instruction sheets. Each device was dipped into urine for 5–20 s and incubated 2–5 min before reading. Readings were taken in duodecuplicate. The lowest concentration of total hCG that yielded 12 of 12 positive results was considered the sensitivity of the device (Table 23.1). Two laboratory technicians and two nurses read the blinded device against the coded urines and all results were decoded by Dr. Laurence Cole.
Table 23.1 Evaluation of OTC tests. Day Relative to Missed Period
Achieved Pregnancy
hCG First Median Response mIU/ml Manual
e.p.t. First Response Manual Gold Digital
e.p.t. Certainty Digital
ClearBlue Easy Manual
ClearBlue Easy Digital
Accu-Clear
CVS One-Step Manual
Study 1. Twelve urines prepared in the laboratory, 50% hCG, 30% hCG-H, 20% free -subunit, 5 total hCG concentrations 100 mIU/ml 50 mIU/ml 25 mIU/ml 12.5 mIU/ml 6.3 mIU/ml Sensitivity
Lab test Lab test Lab test Lab test Lab test
100 50 25 12.5 6.3
12 of 12 11 of 11 12 of 12 12 of 12 12 of 12 6.3 mIU/ml
12 of 12 11 of 12 8 of 12 0 of 12 0 of 12 50 mIU/ml
12 of 12 12 of 12 12 of 12 8 of 12 0 of 12 25 mIU/ml
10 of 12 4 of 12 0 of 12 0 of 12 0 of 12 100 mIU/ml
12 of 12 2 of 12 0 of 12 0 of 12 0 of 12 100 mIU/ml
Study 2. Forty sets of urine samples collected from individuals with term clinical pregnancy, mean day of missing menses (day 0) 12.8 days 0 3
100% 100%
41 154
100% 100%
98% 95%
55% 80%
65% 80%
58% 75%
60% 75%
Study 3. Eighty sets of urine samples collected from individuals with term clinical pregnancy, mean day of missing menses (day 0) 13.0 days 6 5 4 3 2 1 0 1 2 3 4
29% 40% 76% 88% 92% 99% 99% 100% 100% 100% 100%
Three individual studies [3,8,9].
2.1 2.9 5.2 12 21 40 70 143 227 302 534
25% 33% 58% 74% 76% 93% 96% 100% 100% 100% 100%
25% 25% 42% 68% 81% 91% 96% 96% 99% 100% 100%
0% 8.3% 6.1% 14% 29% 42% 53% 64% 77% 80% 100%
0% 16.7% 6.1% 18% 31% 55% 68% 71% 79% 86% 100%
0% 16.7% 9.1% 27% 29% 57% 67% 74% 81% 87% 100%
0% 16.7% 3.0% 12% 28% 51% 51% 69% 77% 84% 100%
Over-the-Counter Pregnancy Tests
255
As shown in Table 23.1, First Response Manual was found to have a sensitivity of 6.3 mIU/ml, whereas e.p.t Manual was 50 mIU/ml and ClearBlue Easy Manual was 25 mIU/ml. The remaining two devices tested, CVS One-Step Manual and AccuClear, had sensitivities of 100 mIU/ml and 100 mIU/ml, respectively. The USA hCG Reference Service concluded that First Response was by far the most sensitive OTC device. It is 4 times more sensitive than its closest competitor, ClearBlue Easy. The high sensitivity of First Response is likely the result of it equally recognizing hCG, hCG-H, and free -subunit, whereas e.p.t. and ClearBlue Easy detect intact hCG only, and not free or hCG-H (Chapter 21). Stenman and colleagues [10], claim that sensitive OTC tests might not be the best at detecting pregnancy and that sensitivity should be restricted to 25 mIU/ml. This is because of low hCG tests occurring in women in perimenopause and menopause, and erroneous positive tests occurring in biochemical pregnancies [10]. Recently, we performed a large study of background urine hCG. Our study tested 11,921 menstrual cycle urine samples and a proportional number of perimenopausal, menopausal, and biochemical pregnancies [11]. Considering hCG levels in biochemical pregnancies, perimenopause, and menopause cases, the 97.5th percentile was just 1.2 mIU/ml. In this regard, the concerns expressed by Stenman and colleagues might be overly cautious [10]. The only appropriate way to properly evaluate OTC tests is to obtain urine samples from women newly achieving pregnancy and from those samples, determine the proportion of gestations detected. Accurately collecting large numbers of such samples was a mammoth task. We accumulated 120 clinically pregnant urine samples from 220 women eager to achieve pregnancy. Once all the samples were obtained, luteinizing hormone (LH) was measured in the participants’ urine in order to pinpoint the exact day of ovulation. Daily urine samples were collected until extremely low levels of hCG were detected, which indicated the exact day of pregnancy implantation [4]. Daily urine samples were collected for an additional 4 weeks to provide pregnancy sample for this and future projects. In a second study of OTC tests sponsored by the USA hCG Reference Service, we evaluated the urine of 40 women on the day of their missed period [8,9]. The actual first day of their missed period was calculated from the three to five previous menstrual cycles by determining days from LH peak to commencement of menstrual bleeding. This average was then used to calculate the individual timing of the day of missing a period. In the 40 women participating in this study, the average time from LH peak to commencement of menstrual bleeding was 12.8 days. Urines were also evaluated 3 days after missing a period. Once again all urines were coded at testing. As shown in Study 2 (Table 23.1), Manual First Response detected 100% of the 40 women on the estimated day of missing menses (day 0) and 100% 3 days later; First Response Gold Digital detected 98% of urines on day 0 and 95% 3 days later; Manual e.p.t detected 55% of urine on day 0 and 80% 3 days later; e.p.t Certainty Digital test detected 65% on day 0 and 80% 3 days later; Manual ClearBlue Easy detected 58% on day 0 and 75% 3 days later; and ClearBlue Easy Digital detected 60% on day 0 and 75% 3 days later. The first conclusion was that digital tests have similar if not better sensitivity than Manual tests. The second conclusion was that
256
Human Chorionic Gonadotropin (hCG)
once again, the First Response test was the most sensitive test, detecting 100% of pregnancies on the day of missing menses. The other brands detected just 55–65% of pregnancies; however, all of these devices claim to detect 99% of pregnancies on the day of missing menses. This does not appear to be true for the e.p.t and ClearBlue Easy brands of tests. We have spoken with customer service at Inverness Medical Innovations, the manufacturer of ClearBlue Easy. We were informed that the calculation of day 0 was LH peak 16 days for this test, and that this measurement was made in accordance with Inverness criterion. No publication, however, supports this LH peak 16 time frame. All published studies show day 0 to be around 13 days from LH peak to menses [4–6], which is very close to the 12.8-day mean reported here. Ironically, Inverness Medical Innovation’s own publication, which measured 895 menstrual cycles, calculated day 0 to be 13.2 2.0 days [7]. Could this be the source of their erroneous 99% claims for ClearBlue Easy tests? It soon became obvious that another study would be necessary in order to confirm our results. Church and Dwight sponsored this study with the understanding that all samples would be coded so that test readers did not know if they were testing a sample from 6 days before or 4 days after the missed menstrual period. Church and Dwight had no role in decoding or in publishing the results. They agreed to these criteria and were willing to pay the ~$40,000 for the 5280 OTC tests needed to perform this extremely large and comprehensive study. This was the most thorough study ever completed. It examined urine samples between 6 days before the menstrual cycle until 4 days after in 80 additional women who achieved pregnancy. All urine samples were provided by the USA hCG Reference Service. Four samples at a time were thawed from storage at –20°C and brought to 37°C immediately before testing. The study proceeded over 6 long days with four test readers, two laboratory technicians, and two nurses (Study 3, Table 23.1). Study 3 (Table 23.1) reports the results of the study sponsored by Church and Dwight. On day 0, 99% of women had achieved pregnancy, and the Manual and Digital First Response tests detected 96% of these pregnancies; Manual and Digital e.p.t tests detected 53% and 68% of pregnancies, respectively; and Manual and Digital ClearBlue Easy devices detected 67% and 51% of pregnancies, respectively. At 6 days before missing the menstrual period (a time when just 29% of women had achieved implantation of pregnancy), Manual and Digital First Response tests detected just 25% of pregnancies, whereas at 1 day after day 0 (when 100% of women had implanted pregnancy), they detected 100% of pregnancies (Table 23.1). Manual and digital results for the ClearBlue Easy and e.p.t tests consistently ranged from 0% detection at 6 days prior to day 0, to 64–74% detection at day 1 (the time of 100% implantation), to 100% detection at 4 days after day 0. The results in this 80-person study were extremely similar to those in the 40-person study, and once again showed near absolute results on day 0 with First Response devices, but just 53–67% with the e.p.t and ClearBlue Easy devices. This study is being prepared for publication. According to the three studies conducted by the USA hCG Reference Service (Table 23.1), the claims made by ClearBlue Easy, such as being able to use the
Over-the-Counter Pregnancy Tests
257
product up to 4 days prior to missing menses, are seemingly without foundation. At 4 days prior to missing menses, ClearBlue Easy tests detect 9.1% (manual) and 3.0% (digital) of pregnancies. The claims made by First Response pregnancy tests 4 days prior to missing menses, which detect 58% (manual) and 42% (digital) of pregnancies, might be closer to reality (Table 23.1). Once again we caution about the exaggerated 99% claims made by Inverness Medical Innovations and Johnson & Johnson Healthcare and Pharmaceuticals for e.p.t. and ClearBlue Easy devices. Care is needed when purchasing home, or OTC, pregnancy tests.
References [1] Anonymous. Home pregnancy test simple to use, reasonably accurate, especially if result is positive. Fam Plann Persp 1979;11:190–1. [2] Joshi UM, Roy R, Sheth AR, Shah HP. A simple and sensitive color test for the detection of human chorionic gonadotropin. Obstet Gynecol 1981;57:252–4. [3] Cole LA, Sutton-Riley JM, Khanlian SA, Borkovskaya M, Rayburn BR, Rayburn WF. Sensitivity of over-the-counter pregnancy tests: comparison of utility and marketing messages. J Am Pharm Assoc 2005;45:608–15. [4] Wilcox AJ, Baird DD, Weinberg CR. Time of implantation of the conceptus and loss of pregnancy. N Engl J Med 1999;340:1796–9. [5] Wilcox AJ, Baird DD, Dunson D, McChesney R, Weinberg CR. Natural limits of pregnancy testing in relation to the expected menstrual period. J Am Med Assoc 2001;286:1759–61. [6] Cole LA, Ladner DG, Byrn FW. The normal variabilities of the menstrual cycle and the menstrual cycle leading to pregnancy. Fertil Steril 2009;91:522–7. [7] Johnson SR, Miro F, Barrett S, Ellis JE. Levels of urinary chorionic gonadotropin (hCG) following conception and variability of menstrual cycle length in a cohort of women attempting to conceive. Curr Med Res Opin 2009;25:741–8. [8] Cole LA. Human chorionic gonadotropin and associated molecules. Expert Rev Molec Diag 2009; 9:51–73. [9] Cole LA. The utility of over-the-counter (home) pregnancy test: comprehensive study. US Pharm 2010; in press. [10] Stenman U-H, Alfthan H. Optimal sensitivity for pregnancy tests. The most sensitive pregnancy tests are not necessarily the most accurate. IVD Technol 2003. http://www .devicelink.com/ivdt/archive/03/04/010.html [11] Cole LA, Ladner DG. Background hCG in non-pregnant individuals: need for more sensitive point-of-care and over-the-counter pregnancy tests. Clin Biochem 2009;42:168–75.
24 hCG Standards Ulf-Håkan Stenman Helsinki University Central Hospital, Finland
Standardization of hormones became important when pharmaceutical preparations of human hormone preparations became available in the 1930s. Most of these were crude tissue or urine extracts, which were standardized with biological assays. The first international standard (first IS) for hCG was introduced in 1939 and was assigned its value in arbitrary international units (IU), one tablet of the standard containing 10 mg of dry substance; its potency was thus 100 IU/mg [1]. The second IS was calibrated against the first IS, and the third and fourth against the second [2,3]. Thus, the hCG standardization can be traced back to the first IS. The introduction of immunoassays for hormones into clinical use in the 1960s revealed problems caused by discrepancies between bioactivity and immunoreactivity. Consisting of partially purified urinary hCG preparations, the first and second standards contained degradation products and subunits of hCG, which lacked bioactivity but were immunoreactive. Thus, they were not suited for standardization of immunoassays, which, to a variable degree, recognized hCG subunits and other gonadotropins. The third IS was a highly purified preparation of hCG, but because the second IS contained subunits, the value for the third IS had to be assigned on the basis of bioactivity [3,4]; however, the third IS also contained significant amounts of degraded forms of hCG (nicked hCG, hCGn), which were recognized by many hCG assays. With the development of better purification techniques, it became possible to prepare more pure preparations of hCG, its subunits, and partially degraded forms. The values of these were assigned in substance concentrations—the first time for a protein hormone. In 2006, these preparations were approved by the WHO as International Reference Reagents (IRR) [5,6]. The properties of the standards are summarized in Table 24.1.
Table 24.1 IS and Reference Preparations for hCG and Related Substances. Standard
Source
Potency (IU/mg)
Established
Reference
First IS Second IS IRP Third and fourth IS IRR
Placental extract Pregnancy urine Pregnancy urine Pregnancy urine Pregnancy urine
10 782 12,000 9300 13,000
1939 1964 1976 1980 2005
[1] [2] [4] [3] [6]
Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00024-4 © 2010 Elsevier Inc. All rights reserved.
260
Human Chorionic Gonadotropin (hCG)
24.1 First IS During the third Conference on the Standardization of Hormones, held under the auspices of the Permanent Commission on Biological Standardization in Geneva in August of 1938, it was decided to arrange for the establishment of an IS for the gonadotropic substance of human urine of pregnancy. The first IS for hCG was established in 1939, under the auspices of the League of Nations. It consisted of a mixture of urinary hCG preparations donated by six companies. The dry preparations were diluted with lactose to a mass of 800 g and further dried over P2O5 in vacuo. From this material, 10 mg tablets were prepared and batches of 25 tablets were packed in tubes that were each enclosed in 2854 ampoules. The definition of the unit was “the specific gonadotropic activity of 0.1 mm (100 gamma) of the standard preparation shall be the IU for recording the activities of all gonadotropic preparations of human urine of pregnancy, but only of such.” Thus, the potency of this preparation was 10 International Units/mg (10 IU/mg) [1]. It should be noted that most of the dry substance in this preparation consisted of lactose, and the activity per mg protein is not known. This standard was successfully used to calibrate pharmaceutical preparations of hCG for the next few decades.
24.2 Second IS Preparation of a new standard was initiated in 1960, when supplies of the first IS were running low. The Expert Committee on Biological Standardization of WHO authorized the National Institute for Medical Research in London to arrange establishment of a new standard. Because highly purified preparations of hCG were considered unstable, it was decided that material of medium potency should be used. The second IS was prepared from a single batch of 10 g of urinary hCG with a potency 2180 IU/ml donated by N.V. Organon (Oss, The Netherlands). This preparation was dissolved in 4 l of 0.5% lactose solution and 1-ml aliquots distributed into 3850 ampoules, which were freeze dried. Each ampoule contained 6.78 mg of dry substance consisting of lactose and the urinary hCG preparation. This standard was calibrated in an international collaborative study performed in nine laboratories using six different bioassays and one immunoassay. Based on this study, the standard was assigned a value of 5300 IU/ampoule and the IU of hCG was defined as the activity contained in 0.001,279 mg of the Second IS for hCG [2]. This corresponds to a potency of 782 IU/mg, which means that most of the mass of the standard consisted of lactose.
24.3 Third IS When the second IS was used for standardization of immunoassays, the presence of free subunits that, to variable degrees, were recognized by different immunoassays
hCG Standards
261
Table 24.2 Comparison of WHO Units (IU), Mass Units (g), and Substance Concentrations (pmol). Abbreviation
MW
IU
g
pmol
hCG hCG hCG hCGcf
37,500 23,500 14,000 12,000
1 1 1
0.11 1 1 1
2.9 42.5 71.4 83.3
Modified from [28]. The WHO IU for hCG is based on bioactivity, whereas the corresponding values for hCG and hCG are based on mass.
caused problems. In 1972, highly purified preparations of hCG and its subunits encoded CR119 were prepared by Canfield and Ross and were available from the Center for Population Research and the Reproduction Branch of the National Institute of Child Health and Human Development in the United States. CR119 consisted of three highly pure preparations: CR119 hCG, CR119 hCG, and CR119 hCG isolated from urine. CR119 hCG had a nominal activity of 13,450 IU/mg. A total of 550 mg of hCG and 280 mg of each subunit were made available to the WHO and used to prepare new standards [3,4]. These preparations were initially called reference preparations and after thorough international evaluation organized by the WHO laboratory in the United Kingdom, National Institute of Biological Standards and Control (NIBSC), and in 1980, CR119 was adopted as a WHO standard with the code [3]. It was assigned a value of 650 IU/ampoule based on its bioactivity and the ratio of bioactivity to mass (i.e., potency), was 9286 IU/mg. The same batch of hCG was used to prepare the fourth IS, thus the third IS 75/537 and fourth IS 75/589 are identical. The concentrations of hCG in IU of hCG can be converted to pmol by multiplying by 2.93 [7]. Standards for the free subunits of hCG were prepared by dissociating intact hCG and purifying the subunits. Because these lacked bioactivity, the standards for hCG (75/551) and hCG (75/569) were assigned values based on mass with 1 g corresponding to 1 IU [3]. Thus, their units are not comparable to those of hCG. Whereas 1 IU of hCG corresponds to 2.9 pmol, one IU of hCG is 42.5 pmol (Table 24.2). Many commercial hCG assays measure hCG and hCG together, and the results are based on the units for hCG and expressed in IU/l [8]. It should be recognized that this unit does not correspond to the WHO IU for hCG. Groups that measure subunits by specific assays usually express the concentrations of hCG and hCG in substance concentrations as pmol/l [8]. In practice, the concentrations of hCG subunits are not expressed using the official WHO units.
24.4 IRR for hCG and Related Substances Methods to improve immunoassays standardization were discussed at two Bergmeyer conferences organized by the International Federation of Clinical Chemistry (IFCC)
262
Human Chorionic Gonadotropin (hCG)
Table 24.3 IFCC Nomenclature and WHO Designations for the First IRR for hCG and Clinically Important hCG-Related Molecules. Molecule
Symbol
Description
Ampoule Code
Human chorionic gonadotropin, intact Human chorionic gonadotropin, nicked
hCG
hCG devoid of nicked forms and free subunits Partially degraded hCG, missing peptide bonds in the hCG-40–50 region Purified hCG, dissociated from hCG Purified hCG, dissociated from hCG Partially degraded hCG, missing peptide bonds in the hCG-40–50 region Residues hCG-6–40, joined by disulfide bonds to hCG-55–92
99/688
Human chorionic gonadotropin, -subunit Human chorionic gonadotropin, -subunit Human chorionic gonadotropin, nicked -subunit Human chorionic gonadotropin, -core fragment
hCGn
HCG hCG hCGn
hCGcf
99/642
99/720 99/650 99/692
99/708
in 1990 and 1992. Participants were invited to suggest projects, and Ulf-Håkan Stenman’s proposal to develop standardization of hCG immunoassays was approved. The project plan was elaborated by an expert group and published in 1993. The main goals of the project were as follows: (1) establishing a uniform nomenclature and abbreviations for the various molecular forms of hCG; (2) preparing new calibrators for the various molecular forms of hCG and establishing methods for determining the substance concentrations (mol/l) of these; (3) improving quality assessment materials and procedures; and (4) defining methods for characterization of hCG measurement procedures [8]. In 1995, IFCC appointed a working group (WG) for standardization of hCG determinations. The group consisted of Ulf-Håkan Stenman, Helsinki (chairman); Steven Birken, New York; Jean-Michel Bidart, Paris; Bruce Nisula, Bethesda, MD; Peter Berger, Innsbruck; and Rob Norman, Adelaide. Nisula soon retired from the group and was replaced by Catharine Sturgeon, who acted as secretary until 2001. In 2001, she replaced Stenman as chairman when he was appointed member of the Scientific Division of the IFCC with responsibility to supervise the WG. The first goal of the WG, establishment of a nomenclature, was accomplished in 1993 (Table 24.3) [8]. The need to prepare new standards arose when the third IS was found to contain partially degraded hCG (i.e., hCGn) and some other impurities, causing problems when the preparations were used for immunoassay standardization. The decision to assign values to the new standards in substance concentrations was based on the reasoning that immunoassays reflect molar concentrations of protein rather than bioactivity [8]. Furthermore, the use of substance concentrations (i.e., mol/l) is the most appropriate way of comparing concentrations of two (or several) variants of the same molecule with different molecular weights (MW) and to express
hCG Standards
263
the sum of their concentrations. Molar concentrations can reliably be determined by amino acid analysis [5]. Financing for the process of creating new hCG standards became possible through support from the IFCC and 13 diagnostics companies: Abbott Diagnostics Division; Bayer PLC (Diagnostics Division); Beckman Coulter Ltd; Bioclone (Australia); BioMe’rieux Ltd; Chiron Diagnostics; CIS International; Dade Behring Ltd; Diagnostic Products Corporation; Ortho-Clinical Diagnostics; PerkinElmer Life Sciences; and Randox Laboratories. Together they provided US$155,000 in exchange for early access to the hCG preparations [5]. Two lots of pharmaceutical grade hCG with a biological activity of 4200 IU/mg of dry weight from Diosynth (Division of Akzo/Nobel) were used as starting material for preparation of the standards. Steven Birken at Columbian University, NY, undertook purification of the four forms of hCG that were considered clinically important [i.e., hCG, hCG, hCG, and core fragment of hCG (hCGcf)]. In addition, standards were also prepared for two partially degraded or nicked forms (i.e., hCGn and hCGn) (Table 24.3). The nicked variants were prepared because they were needed for characterization of antibody and assay specificity. Certain antibodies that do not recognize hCGn will miss part of hCG that may form a substantial part of hCG in urine and also in serum [9–11]. Nicked hCG (hCGn) is also useful for characterization of the epitope specificity of hCG antibodies [12]. A major part of the hCG immunoreactivity in urine of pregnant women consists of hCGcf, and it is of potential clinical value as a urine marker for cancer [13]. Therefore, it is important for characterization of pregnancy tests and other immunoassays for detection of hCG immunoreactivity in urine. The purification process and value assignment of the six new preparations have been described. Removal of the nicked forms of hCG was achieved by use of hydrophobic interaction chromatography and final purification of subunits by reverse phase chromatography. With the exception of hCGn, which contained some ribonuclease, the final preparations displayed a spectrum of isoelectric points reflecting the carbohydrate heterogeneity of the native forms. Contamination with other proteins and other forms of hCG was negligible [5]. The standards were assigned values in substance concentrations by amino acid analysis and the specific absorbance at 276 nm was determined. The activity of the hCG preparation determined by the seminal weight method in two laboratories was 10,300 and 15,400 IU/mg, while the Leydig cell testosterone production assay gave a value of 14,100 IU/l [5]. The mean value of all assays was 13,300 IU/mg, thus its potency was about 43% higher than that of the third and fourth IS, but there is a strong confidence interval. Lyophilized preparations of the purified material were delivered to the National Institute of Standards and Control (Potters Bar, Herts, UK) where they were dissolved in a buffer with a carrier protein, lyophilized, and subjected to stability and recovery studies in a collaborative study. All preparations were found to be stable and the recovery after reconstitution was normal. The six preparations were approved by WHO as the First International Reference Preparations (IRP) for six hCG-related molecules (Table 24.3). The IRRs are primarily intended for characterization of hCG immunoassays [6].
264
Human Chorionic Gonadotropin (hCG)
The first IRRs have been used to characterize the reactivity of commonly used sandwich-type assays and two in-house radioimmunoassay (RIA) methods. In 14 assays, mean recovery of the fourth IS was 107% (range 93–126%) while that for IRR 99/680 hCG was 139% (range 109–164%). The higher values and the larger variation for the IRR reveal problems caused by the presence of hCGn in the fourth IS for assay calibration; hCGn is differently recognized by different assays. Recognition of hCG by assays designed to measure this form together with hCG was more variable and all but one sandwich assay overestimated it, one by 150%. Four sandwich assays recognized hCGcf, but they all underestimated it. RIAs, on the other hand, overestimated it [14]. The results of a similar study comprising eight assays gave very similar results [15]. These studies demonstrate the value of the new IRRs. The results of these studies reveal problems that will be encountered when the presently used fourth standard has to be replaced. It is obvious that value assignment on the basis of the bioactivity of a new standard will result in a substantial change, about 40% in assay calibration. The original aim of the hCG standardization project was to prepare standards calibrated in substance concentrations. The metrologically correct way to accomplish this would be to use the IRRs to calibrate immunoassays hCG, hCG, hCG, and hCGcf and to express the concentrations in mol/l [8]. When the IRRs were approved, WHO recommended that hCG immunoassays should still be calibrated against the fourth IS; however, because the fourth IS has to be replaced soon, standardization inevitably will change when a new standard is introduced. Based on preliminary information, the new standard will be based on IRR 99/680 (Chris Burns, personal communication). Because of the heterogeneity of Chinese hamster ovary cells (CHO) moieties, the MW displays a spectrum of values. The average MW of hCG determined by MALDI-TOF mass spectrometry is 37,500; that of hCG is 14,000; that of hCG is 23,500; and that of hCGcf is 10,000 [5]. The calculated mass of the peptide moiety of hCG is 10,206 and that of the glycated subunit containing two N-linked biantennary sialylated CHO chains is 14,165. The MW calculated on the basis of the amino acid composition of hCG is 15,532 and that containing two N-linked biantennary CHO chains and four type 1 o-core carbohydrates is 24,316. Thus the calculated MW of a “typical” hCG molecule would be 38,931 [11]. The difference between the calculated and measured average MW indicates that the CHO chains are shorter than those used to calculate the theoretical MW. The MW of hCG produced by trophoblastic cancer is higher than that of pregnancy hCG [16], which is explained by the larger carbohydrate chains of hyperglycosylated hCG (hCG-H) [17,18].
24.5 Future Prospects hCG produced by choriocarcinoma often contains larger than normal carbohydrate moieties. This form of hCG, called hCG-H, can be determined by the use of a monoclonal antibody (B152) raised against hCG-H isolated from urine of a patient with choriocarcinoma [17,19]. Most hCG secreted during early pregnancy consists of
hCG Standards
265
hCG-H [20]. Several studies suggest that specific determination of hCG-H is useful for diagnosis of certain pregnancy complications [21,22] and for differentiation between benign and malignant trophoblastic disease [23]. Some hCG assays and pregnancy tests have been shown to underestimate hCG-H [24,25]. Final evaluation of this would require a standard for hCG-H. When the goals of the hCG WG were defined, the potential utility of hCG-H was not recognized and a standard was not prepared; however, identification of a suitable source hCG-H is problematic; although it occurs in serum and urine during early pregnancy, it is not yet known whether hCG-H from these sources is comparable to that occurring in patients with trophoblastic and other types of cancer. Studies of urine from pregnant women and patients with trophoblastic and testicular tumors have revealed large differences in the glycan structure of hCG [11]. Therefore, hCG-H is not a well-defined entity, and the form of hCG-H to be used as a standard remains to be determined.
24.6 Recombinant hCG Recombinant hCG (rhCG) has been produced in CHO cells, and a pharmaceutical preparation is available from Serono. rhCG is also a potential new standard and thus its biological and immunological properties are of interest. When rhCG is used for ovulation induction, a dose of 250 g of rhCG is equivalent to 5000–10,000 IU of urinary hCG [26,27]. The hCG content of this preparation is expressed on the basis of the mass of the peptide moiety disregarding the carbohydrate content [27]. Thus 250 g of rhCG correspond to approximately 360 g of urinary hCG. It can thus be calculated to have a specific activity of 13,900 IU/mg [28], which is similar to the potency of about 12,000–15,000 IU/mg of the WHO hCG RR 98/688 [5]. These calculations are based on the typical carbohydrate structure of urinary hCG, but the carbohydrates of rhCG are slightly different; the N-linked carbohydrates are the same, but the O-linked carbohydrate chains on the CTP are slightly different [27]. rhCG is also available as a laboratory product, and it is a potential calibrator for immunoassay purposes; however, considerable differences in the immunoreactivity pituitary and recombinant growth hormone have been observed [29]. Therefore, introduction of a recombinant standard for hCG would require careful evaluation of its reactivity in various immunoassays.
References [1] The Department of Biological Standards NIfMR. Memorandum on the international standard for the gonadotropic substance of urine of pregnancy (chorionic gonadotropin). Bull World Health Org L o N 1939;8:884–6. [2] Bangham DR, Grab B. The second international standard for chorionic gonadotropin. Bull World Health Org 1964;31:111–25.
266
Human Chorionic Gonadotropin (hCG)
[3] Storring PL, Gaines-Das RE, Bangham DR. International reference preparation of human chorionic gonadotropin for immunoassay: potency estimates in various bioassay and protein binding assay systems; and International reference preparations of the alpha and beta subunits of human chorionic gonadotropin for immunoassay. J Endocrinol 1980; 84:295–310. [4] Canfield RE, Ross GT. A new reference preparation of human chorionic gonadotropin and its subunits. Bull World Health Org 1976;54:463–72. [5] Birken S, Berger P, Bidart JM, Weber M, Bristow A, Norman R, et al. Preparation and characterization of new WHO reference reagents for human chorionic gonadotropin and metabolites. Clin Chem 2003;49:144–54. [6] Bristow A, Berger P, Bidart JM, Birken S, Norman R, Stenman UH, et al. Establishment, value assignment, and characterization of new WHO reference reagents for six molecular forms of human chorionic gonadotropin. Clin Chem 2005;51:177–82. [7] Alfthan H, Haglund C, Dabek J, Stenman U-H. Concentrations of human chorionic gonadotropin, its b-subunit and the core fragment of the -subunit in serum and urine of men and nonpregnant women. Clin Chem 1992;38:1981–7. [8] Stenman U-H, Bidart JM, Birken S, Mann K, Nisula B, O’Connor J. Standardization of protein immunoprocedures. Choriogonadotropin (CG). Scand J Clin Lab Invest Suppl 1993;216:42–78. [9] Cole LA, Kardana A, Ying FC, Birken S. The biological and clinical significance of nicks in human chorionic gonadotropin and its free beta-subunit. Yale J Biol Med 1991;64:627–37. [10] Hoermann R, Berger P, Spoettl G, Gillesberger F, Kardana A, Cole LA, et al. Immunological recognition and clinical significance of nicked human chorionic gonadotropin in testicular cancer. Clin Chem 1994;40:2306–12. [11] Valmu L, Alfthan H, Hotakainen K, Birken S, Stenman UH. Site-specific glycan analysis of human chorionic gonadotropin beta-subunit from malignancies and pregnancy by liquid chromatography–electrospray mass spectrometry. Glycobiology 2006;16:1207–18. [12] Berger P, Sturgeon C, Bidart JM, Paus E, Gerth R, Niang M, et al. The ISOBM TD-7 Workshop on hCG and related molecules. Towards user-oriented standardization of pregnancy and tumor diagnosis: assignment of epitopes to the three-dimensional structure of diagnostically and commercially relevant monoclonal antibodies directed against human chorionic gonadotropin and derivatives. Tumour Biol 2002;23:1–38. [13] Wehmann RE, Nisula BC. Characterization of a discrete degradation product of the human chorionic gonadotropin beta-subunit in humans. J Clin Endocrinol Metab 1980; 51:101–5. [14] Sturgeon CM, Berger P, Bidart JM, Birken S, Burns C, Norman RJ, et al. Differences in recognition of the first WHO International Reference Reagents for hCG-related isoforms by diagnostic immunoassays for human chorionic gonadotropin. Clin Chem 2009; 55:1484–91. [15] Whittington J, Fantz CR, Gronowski AM, McCudden C, Mullins R, Sokoll L, et al. The analytical specificity of human chorionic gonadotropin assays determined using WHO International Reference Reagents. Clin Chim Acta 2010;411:81–5. [16] Mann K, Karl HJ. Molecular heterogeneity of human chorionic gonadotropin and its subunits in testicular cancer. Cancer 1983;52:654–60. [17] Elliott MM, Kardana A, Lustbader JW, Cole LA. Carbohydrate and peptide structure of the alpha- and beta-subunits of human chorionic gonadotropin from normal and aberrant pregnancy and choriocarcinoma. Endocrine 1997;7:15–32.
hCG Standards
267
[18] Birken S. Specific measurement of O-linked core 2 sugar-containing isoforms of hyperglycosylated human chorionic gonadotropin by antibody b152. Tumour Biol 2005; 26:131–41. [19] Birken S, Krichevsky A, O’Connor J, Schlatterer J, Cole L, Kardana A, et al. Development and characterization of antibodies to a nicked and hyperglycosylated form of hCG from a choriocarcinoma patient: generation of antibodies that differentiate between pregnancy hCG and choriocarcinoma hCG. Endocrine 1999;10:137–44. [20] Kovalevskaya G, Birken S, Kakuma T, O’Connor JF. Early pregnancy human chorionic gonadotropin (hCG) isoforms measured by an immunometric assay for choriocarcinomalike hCG. J Endocrinol 1999;161:99–106. [21] Cole LA, Shahabi S, Oz UA, Bahado-Singh RO, Mahoney MJ. Hyperglycosylated human chorionic gonadotropin (invasive trophoblast antigen) immunoassay: a new basis for gestational Down syndrome screening. Clin Chem 1999;45:2109–19. [22] Kovalevskaya G, Birken S, Kakuma T, Ozaki N, Sauer M, Lindheim S, et al. Differential expression of human chorionic gonadotropin (hCG) glycosylation isoforms in failing and continuing pregnancies: preliminary characterization of the hyperglycosylated hCG epitope. J Endocrinol 2002;172:497–506. [23] Cole LA, Butler S. Detection of hCG in trophoblastic disease. The USA hCG Reference Service experience. J Reprod Med 2002;47:433–44. [24] Butler SA, Khanlian SA, Cole LA. Detection of early pregnancy forms of human chorionic gonadotropin by home pregnancy test devices. Clin Chem 2001;47:2131–6. [25] Cole LA, Kohorn EI. The need for an hCG assay that appropriately detects trophoblastic disease and other hCG-producing cancers. J Reprod Med 2006;51:793–811. [26] Chang P, Kenley S, Burns T, Denton G, Currie K, DeVane G, et al. Recombinant human chorionic gonadotropin (rhCG) in assisted reproductive technology: results of a clinical trial comparing two doses of rhCG (Ovidrel) to urinary hCG (Profasi) for induction of final follicular maturation in in vitro fertilization-embryo transfer. Fertil Steril 2001; 76:67–74. [27] Gervais A, Hammel YA, Pelloux S, Lepage P, Baer G, Carte N, et al. Glycosylation of human recombinant gonadotropins: characterization and batch-to-batch consistency. Glycobiology 2003;13:179–89. [28] Stenman UH, Tiitinen A, Alfthan H, Valmu L. The classification, functions and clinical use of different isoforms of HCG. Hum Reprod Update 2006;12:769–84. [29] Stenman UH. Detection of illicit use of growth hormone. Clin Chem 2009;55:387–8.
25 Background hCG Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
Background hCG is one of the poorest studied scopes of hCG research, but it is still important. Only by knowing what background levels of female hCG are can we say fairly what concentration of serum and urine hCG constitutes a pregnancy or the successful treatment of gestational trophoblastic disease. We now understand that hCG is primarily produced by trophoblast cells during both pregnancy and gestational trophoblastic cases. We understand that background hCG is produced by the pituitary gland as part of the menstrual cycle. We can also consider menopausal pituitary hCG as background hCG [1–3]. We also know that the free -subunit of hCG can be produced by any type of cancer cell in the body [4–6] and can also be considered as background hCG. Elevated levels of pituitary hCG, and its free subunits and fragments, may be part of a rare genetic disorder in humans (7, see Chapter 31), which might also constitute background hCG. It was thought that hCG was made by bacteria, fungi, plants, and almost every cell in the body [8–15], but today we have learned a great deal about hCG and no longer consider such sites as hCG sources [16]. In short, the pituitary gland seems responsible for all human background hCG, whether from menstrual cycles, menopause, or genetic disorders [1,3,7]. Why does the gonadotrope cell of the pituitary gland produce hCG? Under the control of pulsed gonadotropin releasing hormone (GnRH) produced by the hypothalamus, the gonadotrope cells of the pituitary gland primarily express luteinizing hormone (LH). These cells promote pulses of LH. They also produce pulses of hCG. As shown [17], normal women produce 0.68 0.09 pulses/h of hCG in the follicular phase, and 0.42 0.16 pulses/h in the luteal phase of the menstrual cycle. These rates are very similar to the pulses/h of LH during these same periods, 0.62 0.08 and 0.38 0.08, respectively [17]. Why? Like trophoblast cells, the gonadotrope cells of the pituitary gland produce an excess amount of the common hCG or LH -subunit so that -subunit is limiting and controls the synthesis of LH and hCG dimer [18]. There are seven back-to-back hCG -subunit genes in the human genome. One LH -subunit gene is in the midst of all the other hCG -subunit genes. GnRH promotes pituitary LH throughout the length of the menstrual cycle and seems to be unable to avoid promoting some of the excess surrounding hCG -subunit genes. As such, the pulses of hCG supplement the pulses of LH [17]. When women enter menopause or perimenopause, they lose progesterone and estrogen feedback control of GnRH pulses. As observed in menopausal or perimenopausal women, GnRH resorts to continuous, uncontrolled, nonstop rapid pulses, which lead to the unduly high levels of both LH and follicle Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00025-6 © 2010 Elsevier Inc. All rights reserved.
272
Human Chorionic Gonadotropin (hCG)
stimulating hormone (FSH) as well as very measurable levels of hCG [19]. According to the USA hCG Reference Service experiences, pituitary hCG levels detected during menopause or perimenopause can range from 1 to 39 mIU/ml of hCG in serum [16] (see Chapter 32). According to Snyder et al. [1], hCG ranges from 2 to 13.1 mIU/ ml in serum in menopausal women. According to Odell and Griffin [2], hCG is produced throughout the length of the menstrual cycle. The USA hCG Reference Service recently showed that detectable levels of hCG (1 mIU/ml) were present at the time of the mid-cycle LH peak in all menstrual cycles [20]. This is background hCG, and in this chapter we examine background hCG in detail.
25.1 Serum hCG As established by Odell and Griffin [2], low levels of hCG are produced by men throughout life and by women throughout their menstrual cycle and in menopause. An average of 8.9 pg/ml of hCG (equivalent of 0.098 mIU/ml) is detectable in men’s serum throughout life. These levels have been calculated using a very sensitive hCG assay (sensitivity 3 pg/ml or 0.033 mIU/ml). In women’s serum, hCG ranging from 3.0 to 160 pg/ml (equivalent of 0.033–1.7 mIU/ml) is detected. These levels mirror the rises and falls in LH levels during the menstrual cycle. As such, the highest level of hCG is detected at the mid-cycle LH peak that precedes ovulation; the lowest hCG levels are detected in the follicular phase and the lowest levels in the luteal phase of pregnancy. The USA hCG Reference Service recently tested the menstrual cycle urine of 9123 women [3,20]. LH peak hCG was detectable in all menstrual cycles at the LH peak, mean 1.52 0.91 mIU/ml. Much lower levels were detected in the follicular and luteal phases of the menstrual cycle. The serum studies of Odell and Griffin [2] suggest a background hCG of 1.7 mIU/ml [3,20]. The mean 1.52 0.91 mIU/ml as determined by the USA hCG Reference Service is very similar to the Odell and Griffin value. A more recent study conducted by Snyder and colleagues [1] suggests a range of total hCG of 2–4.6 mIU/ml; however, it should be noted that these values were calculated using menstrual serum samples from only 240 women of 18–40 year old. In the field of clinical chemistry, and with all clinical laboratory tests, the generally accepted upper reference limit for background limits is the 97.5th centile [21,22]. According to Snyder et al. [1], the 97.5th centile for hCG during menstrual years is 2.5 mIU/ml. The Snyder study calculated this value based on the average of two assays (Roche Elecsys total hCG and the Siemens Advia Centaur total hCG). Automated hCG tests generally have a sensitivity of 1 or 2 mIU/ml. The generally accepted cutoff for detecting pregnancy is 5 mIU/ml, but according to Snyder et al. [1], the cutoff for these tests could be lowered to 2.5 mIU/ml. The USA hCG Reference Service used the Siemens Immulite assay to test 428 male and 369 menstruating female serum samples. We found 2.7 mIU/ml to be the 97.5th centile, which is very similar to the Snyder and colleagues value of 2.5 mIU/ml finding. Thus it can be concluded that a cutoff of 2.5–2.7 mIU/ml is the most appropriate cutoff for use with pregnancy detection. There are very few other published studies on background serum hCG. We question what sources of hCG need to be included in a serum background hCG mixture. There is normal menstrual pituitary hCG and biochemical pregnancy
Background hCG
273
Figure 25.1 Serum samples from multiple menstrual cycles courtesy of Snyder et al. [1], tested for total hCG.
hCG. A biochemical pregnancy is one that fails to properly implant and leads to a transient rise in hCG, which is called background hCG. The pregnancy fails a few days later and women usually have menses on time or up to 3 days late [23]. Women have no knowledge that they are very briefly pregnant. Biochemical pregnancies are normal [23–26] and most reports indicate that biochemical pregnancies account for 25% of all pregnancies. In 2006, there were 90.9 million U.S. women ages 20–40 [27]. There were 4,256,996 births that year in the United States, or 1 birth per 16.6 women. If biochemical pregnancies account for one in four pregnancies as published [23–26], then there would be one biochemical pregnancy per 66.4 women (16.6 4). The study by Snyder and colleagues [1] involves 240 serum samples from women ages 18–40. If this study included a representative proportion of biochemical pregnancies, it should account for 240 ÷ 66.4, or approximately 3–4, serum samples. The median hCG in a biochemical pregnancy is 2.9 mIU/ml [23]. The Snyder study involves seven positive samples close to this range (2.0–4.2 mIU/ml). As such, it is probably representative of proportionate biochemical pregnancies (Figure 25.1).
25.2 Urine hCG There are even less data published on background urine hCG. Here we present the USA hCG Reference Service’s extensive data on 9306 urine samples from 416 women 20–40 years age. Determining the cutoff level of urine hCG is important for over-the counter (OTC) and point-of-care (POC) pregnancy tests, but what concentration represents pregnancy? The U.S. Food and Drug Administration (FDA) is currently somewhat wary about urine as a media for measuring hCG. They approve the use of automated serum hCG tests for urine measurements, but only for quantitative use at a sensitivity of 30 mIU/ml. They also approve POC urine tests, but again,
274
Human Chorionic Gonadotropin (hCG)
only for quantitative use at a sensitivity of 20 mIU/ml. There is, however, no sensitive urine hCG test in use today. With a grant from Church and Dwight Inc., the USA hCG Reference Service collected daily urine samples from 399 women during as many as five menstrual cycles. The USA hCG Reference Service provided the funds needed to measure hCG and LH. The mean cycle was 28.6 3.8 days, with the LH peak occurring at 14.6 3.1 days [3,20]. The USA hCG Reference Service measured total urine hCG using the Siemens Immulite automated hCG test. Prior to testing, this assay was validated for use with urine samples (Table 25.1). LH was measured on the Siemens Immulite without validation.
Table 25.1 Comparison of Serum and Urine standards and Calibrators on the Siemens Immulite 1000 Analyzer.
A. Prepared in serum Siemens, 95th centile 8.7–13.7 IU/l internal control Siemens, 95th centile 35–49 IU/l internal control Siemens, 95th centile 342–510 IU/l internal control Standard curve, first RR hCG standard, 0 pmol/l (0.0 IU/l) Standard curve, first RR hCG standard, 10.8 pmol/l (4.4 IU/l) Standard curve, first RR hCG standard, 54.5 pmol/l (22.2 IU/l) Standard curve, first RR hCG standard, 272 pmol/l (111 IU/l) Precision, 600 IU/l, 10 times Precision, 40 IU/l, 10 times Linearity, 1000 IU/l in serial twofold dilutions to 7.75 IU/l Sensitivity, a single normal male serum tested 100 times Interassay variance
10.1 0.75 IU/l 37.8 1.8 IU/l 390 14.7 IU/l 1, 1, 1.4, 1.0 IU/l 4.3 0.14 IU/l b 22.3 2.8 IU/l b 115 4.5 IU/l b Coefficient of variation 4.4% Coefficient of variation 4.2% Coefficient of variation 6.1% 98 results 1 IU/l, 1.1, 1.4 IU/l 7.3%
B. Prepared in urine Siemens, 95th centile 8.7–13.7 IU/l internal control Siemens, 95th centile 35–49 IU/l internal control Siemens, 95th centile 342–510 IU/l internal control Standard curve, first RR hCG standard, 0 pmol/l (0.0 IU/l) Standard curve, first RR hCG standard, 10.8 pmol/l (4.4 IU/l) Standard curve, first RR hCG standard, 54.5 pmol/l (22.2 IU/l) Standard curve, first RR hCG standard, 272 pmol/l (111 IU/l) Precision, 600 IU/l, 10 times Precision, 40 IU/l, 10 times Linearity, 1000 IU/l in serial twofold dilution to 7.75 IU/l Sensitivity, a single normal male urine tested 100 times Interassay variance
9.5 4.2 IU/la 37.5 3.1 IU/la 379.5 11.5 IU/la 1, 1, 1, 1 IU/l 4.4 0.73 IU/la,b 22.2 0.68 IU/la,b 114.5 14 IU/la,b Coefficient of variation 4.8% Coefficient of variation 6.9% Coefficient of variation 2.6% 99 results 1 IU/l, 1.4 IU/l 5.4%
Siemens hCG calibration controls were extracted in commercial pooled control serum and urine in quadruplicate (Sigma). These are supplied with ranges for the 95th centile of results. Standards were made using the first RR hCG standard, both in serum and urine. a No statistical difference observed between serum and urine results (t test, P 0.2). b The equivalent concentration in hCG international units (IU/l) was 0.407 the molar concentration of standard (pmol/l).
Background hCG
275
The Siemens Immulite total hCG test worked equally as well with serum samples as it did with urine samples, with similar sensitivity and similar accuracy. As shown in Figure 25.2, only 31 of 3662 total urine samples (0.80%) were found positive during the follicular phase of the menstrual cycle; 3631 samples were 1 mIU/ ml, or below the sensitivity of the assay. The maximum hCG in any urine sample from the follicular phase was 4.0 mIU/ml. The 95th, 97.5th, and 99th centiles were 1, 1, and 1 mIU/ml, respectively. A much greater number of urine hCG elevations were detected around the LH peak or ovulatory phase of the menstrual cycle. As stated earlier, the average LH peak in urine was 1.52 0.91 mIU/ml of hCG. In total, 1 mIU/ml of hCG was detected in 206 of 2289 urine samples (9.0%) (Figure 25.3).
Figure 25.2 Urine samples, 3662 total, from 399 women providing daily urine during the follicular phase of the menstrual cycle, days 1–10, tested for hCG.
Figure 25.3 Urine samples, 2289 total, from 399 women providing daily urine during the ovulatory phase of the menstrual cycle, days 11–17, tested for hCG.
276
Human Chorionic Gonadotropin (hCG)
The highest hCG recorded during the LH peak was 9.3 mIU/ml. The 95th, 97.5th, and 99th centiles were 1.2, 1.6, and 2.3 mIU/ml, respectively. Only 22 of 3172 (0.69%) urine samples were positive for hCG in the luteal phase of the cycle (Figure 25.4). The highest hCG recorded during the luteal phase was 2.7 mIU/ml. We considered the proportionate percentage of biochemical pregnancies by testing 183 urine samples provided by 17 women (Figure 25.5). As shown, hCG values ranged from 1.0 to 27 mIU/ml. Totaling all of these data together, 9306 urines were examined. A total of 442 urines (menstrual cycle plus biochemical pregnancies) were positive (1 mIU/ml) or 4.7%, with the highest value 27 mIU/ml coming from a biochemical pregnancy. The 95th, 97.5th, and 99th centiles were 1, 1.2, and 1.7 mIU/ml, respectively.
Figure 25.4 Urine samples, 3172 total, from 399 women providing daily urine during the luteal phase of the menstrual cycle, days 18–28, tested for hCG.
Figure 25.5 Urine samples, 183 total, from 17 women in the process of a biochemical pregnancy, tested for hCG.
Background hCG
277
Why is urine background hCG (97.5th centile, 1.2 mIU/ml) less than serum background hCG (97.5th centile, 2.5 mIU/ml)? In Chapter 5, we examined the concentrations of 345 serum and 903 urine samples collected throughout the duration of a pregnancy. Comparing the serum to urine values, the median urine result was 63% of the median serum result. Because 1.2 mIU/ml is approximately 63% of 2.5 mIU/ml, the difference is suggested to be a simple case of urine being more dilute than serum. These data indicate that 1.2 mIU/ml is an appropriate cutoff for all test systems that detect pregnancy in urine samples. These data should surmount the FDA’s concerns and systems like the Siemens Immulite should be permitted for quantitative urine hCG measurements. We believe the FDA should be less wary about more sensitive POC and OTC pregnancy tests.
25.3 The Cancer Predicament Total hCG measurements are not solely used as pregnancy tests; they are also used to test for gestational trophoblastic disease and as a broad spectrum tumor marker [4–6]. Total hCG measurements could potentially be used as a screening test for all forms of human cancer [6]. Should background hCG include cancer cases? Like pregnancies, cancer needs to be detected by an hCG test as well. If a person has positive hCG test results, pregnancy will be the first diagnosis. If the hCG does not rise sharply like in a normal, healthy pregnancy, ectopic pregnancy will be considered. After excluding ectopic pregnancy via ultrasound and methotrexate treatment, other possibilities will be investigated, including gestational trophoblastic disease and cancer. Ultimately, MRI and CT/PET will identify the malignancy. Nontrophoblastic malignancies produce up to 200 mIU/ml equivalents of free -subunit or hCG [4–6]. As such, cancer hCG is seemingly not a background measurement like menstrual cycle and biochemical pregnancies. It is a foreground measurement; however, cancer hCG cannot be ignored as a background hCG because it raises red flags and alerts healthcare providers to potentially life-threatening diseases.
25.4 The Menopausal hCG Quandary A major quandary is whether menopausal hCG should be considered as background hCG. Yes, if someone is 60 year old and has a pregnancy test, it may be significantly positive, but it makes no sense for the woman to be pregnant. Is this a false positive hCG test result? In cases like these, a higher concentration of pituitary hCG is a background hCG measurement. According to the serum studies of Snyder et al. [1], menopausal hCG significantly raises the 97.5th centile levels of hCG to 7.7 mIU/ml. We understand that the pituitary gland produces low levels of hCG, but this production increases significantly during perimenopause and menopause [1–3,19]. If a 45-year-old woman is tested for pregnancy prior to surgery and has positive hCG results, she would think that she has become pregnant. If a perimenopausal
278
Human Chorionic Gonadotropin (hCG)
woman had a positive test result, it might indicate pituitary hCG, which is a natural source of hCG in all perimenopausal and menopausal women. As recently published by Gronowski et al. [28], an FSH test (FSH 30 mIU/ml) can distinguish when a woman is in perimenopause or menopause. For all women over the age of 40, an FSH test performed at the same time as a total hCG test can determine whether the woman is pregnant (2.5 mIU/ml) or is in perimenopause at the time of testing. When the FSH test is administered alongside a total hCG test, we do not consider the naturally occurring hCG present in peri- and postmenopause to be a form of background hCG. Today, all medical centers perform presurgery pregnancy tests in order to prevent harm to the fetus if a woman is unknowingly pregnant. Problems arise when 60-yearold woman tests positive for hCG in a presurgery pregnancy test. The surgery is cancelled regardless of its importance. The physician then investigates the possibility of pregnancy and ectopic pregnancy and, after excluding both, considers the possibility of cancer and gestational trophoblastic disease. Only after excluding all possibilities is natural menopausal pituitary hCG considered. The originally scheduled surgery can then be reconsidered, but this process can sometimes take months to complete, which in some cases is simply unacceptable. Therefore, we strongly suggest that medical centers around the world heed the recommendations of Gronowski et al. [28] and run parallel FSH tests when performing hCG tests on women. This one simple test may help to avoid the inconveniences that thousands of women have suffered due to erroneous hCG test results and cancellation of surgery.
References [1] Snyder JA, Haymond S, Parvin CA, Gonowski AM, Grenache DG. Diagnostic considerations in the measurement of human chorionic gonadotropin in aging woman. Clin Chem 2005;51:1830–5. [2] Odell WD, Griffin J. Pulsatile secretion of human chorionic gonadotropin in normal adults. N Engl J Med 1987;317:1688–91. [3] Cole LA, Ladner DG. Background hCG in non-pregnant individuals: need for more sensitive point-of-care and over-the-counter pregnancy tests. Clin Biochem 2009;42:168–75. [4] Muller C, Cole LA. The quagmire of hCG and hCG testing in gynecologic oncology. Gynecol Oncol 2009;112:663–72. [5] Cole LA, Wong Y, Latif M, Chambers JT, Chambers SK, Schwartz PE. Urinary human chorionic gonadotropin free -subunit and -fragment: new markers gynecologic cancers. Cancer Res 1988;48:1356–60. [6] Regelson W. Have we found the “definitive cancer biomarker”? The diagnostic and therapeutic implications of human chorionic gonadotropin-beta statement as a key to malignancy. Cancer 1995;76:1299–301. [7] Cole LA. The utility of over-the-counter (home) pregnancy test: comprehensive study. US Pharm; 2010; in press. [8] Cohen H, Strampp A. Bacterial synthesis of substances similar to human chorionic gonadotropin. Proc Soc Exper Biol Med 1976;152:408–10. [9] Dominigue G, Johson E. Isolation of subcellular fractions containing immunogenic endobacterial common antigen. Z Immunitatsforsh Exper Klinical Immunol 1874;148:23–8.
Background hCG
279
[10] Backus BT, Affronti LF. Tumor-associated bacteria capable of producing human choriogonadotropin-like substance. Infect Immun 1981;32:1211–15. [11] Maruo T, Segal SJ, Koide SS. Studies on the apparent human chorionic gonadotropinlike factor in crab Ovalipes ocellatus. Endocrinol 1979;104:932–9. [12] Yoshimoto Y, Wolfsen AR, Odell WD. Human chorionic gonadotropin-like substance in nonendocrine tissues of normal subjects. Science 1977;197:1977–81. [13] Braunstein GD, Kamdar V, Rasort J, Swaminathan N, Wade ME. Widespread distribution of chorionic gonadotropin-like substance in normal human tissues. Endocrinol 1978;102(Suppl.):96. [14] Chen H-C, Hodgen GC, Matsuura S, Lin LJ, Gross E, Reichert LE, et al. Evidence for a gonadotropin from nonpregnant subjects that has physical immunological and biological similarities to human chorionic gonadotropin. Proc Natl Acad Sci USA 1976; 73:2885–9. [15] Acevedo HF, Slifkin M, Pouchet GR, Pardo M. Immunochemical localization of a choriogonadotropin-like protein in bacteria isolated from cancer patients. Cancer 1978;41:1217–29. [16] Cole LA. Human chorionic gonadotropin and associated molecules. Expert Rev Mol Diagn 2009;9:51–73. [17] Odell WD, Griffin J. Pulsatile secretion of chorionic gonadotropin during the normal menstrual cycle. J Clin Endocrinol Metab 1989;69:528–32. [18] Ruddon RW, Hanson CA, Addison NJ. Synthesis and processing of human chorionic gonadotropin subunits in cultured choriocarcinoma cells. Proc Natl Acad Sci USA 1979;76:5143–7. [19] Cole LA, Khanlian SA, Sasaki Y, Muller CY. Normal production of hCG in menopause: a medical management dilemma. N Engl J Med 2007;356:1184–6. [20] Cole LA, Gutierrez JM. Production of hCG during the menstrual cycle. J Reprod Med 2009;54:245–250. [21] Solberg HE. Statistical treatment of collected reference values and determination of references limits. In: Grasbeck R, Alstrom T, editors. Reference values in laboratory medicine: the current state of the art. Chichester, England: Wiley; 1981. p. 193–205. [22] Rustad P, Petersen HP. Effect of analytical quality on establishing common reference intervals and their use. Scand J Clin Lab Invest 2004;64:399–406. [23] Wilcox AJ, Baird DD, Weinberg CR. Time of implantation of the conceptus and loss of pregnancy. N Engl J Med 1999;340:1796–9. [24] Wilcox AJ, Weinberg CR, Baird DD. Timing of sexual intercourse in relation to ovulation— effects on the probability of conception, survival of the pregnancy, and sex of the baby. N Engl J Med 1995;333:1517–21. [25] Levy T, Goldman JA, Dicker D, Ashkenazi J, Feldberg D. Very early pregnancy wastage in vitro fertilization and embryo transfer (IVF-ET). J Assist Reprod Genet 1991; 8:250–3. [26] Mansour FW, Buckett WM. Pregnancy loss following treatment with assisted reproductive technologies. Expert Rev Obstet Gynecol 2008;3:767–74. [27] US Census Bureau Data, 2008. [28] Gronowski AM, Fantz CR, Parvin CA, Sokoll LJ, Wiley CL, Wener MH, et al. Use of serum FSH to identify perimenopausal women with pituitary hCG. Clin Chem 2008;54:652–6.
26 Pregnancy Testing Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
hCG is, by definition, “the pregnancy test” molecule and its prime use is indisputably in pregnancy diagnosis. Every year, 209 million women around the world achieve pregnancy; of those, 134 million pregnancies lead to live births (2009 WHO statistics). Most of these pregnancies were first diagnosed using hCG tests. Table 26.1 shows hCG levels in serum and urine during pregnancy [1]. As shown, during the first week of pregnancy, or 3 weeks to 3 weeks and 6 days following the last menstrual bleeding, all of the total hCG that is produced is hyperglycosylated hCG (hCG-H) and free -subunit, and not regular hCG. As such, hCG-H and free -subunit are as much pregnancy markers as regular hCG because they are critical components of total hCG. Menstrual cycles greatly fluctuate in length (Table 26.2). Women do not commonly have regular 28-day menstrual cycle; they range from 20 to 34 days and vary in length each month [2]. Timing of LH peaks and ovulation varies greatly as well (Table 26.2). Between menstrual cycles, LH peaks and ovulation occur at an average 14.7 days and begin between day 9 and day 20 of a menstrual cycle [2]. Women normally have a variable menstrual cycle (i.e., 28 days, then 25 days, then 31 days, then 28 days, then 26 days, then 30 days, then 26 days, etc.), but they average 27.7 days in length [2]. The average luteal phase is 13.2 days, but varies from 9 to 20 days between menstrual cycles. In a menstrual cycle during which a woman achieves pregnancy, great variation also occurs. On average, the timing of pregnancy implantation occurs at 24.6 days of the menstrual cycle (Table 26.3), but can vary between day 16 and day 30 of the menstrual cycle, or over a period of 14 days. If a person has an early LH peak on day 9, immediate fertilization of the ovum, and rapid implantation of the hatched blastocyst, implantation might occur at day 16; however, if a person has a late LH peak on day 20, slow fertilization, and slow hatching of a mature blastocyst, implantation might occur as late as day 30 of the menstrual cycle, or around the expected day of missing menstrual bleeding (Table 26.3) [2]. Women’s hCG results vary dramatically between pregnancies. In a study of 80 women who achieved pregnancy, total hCG results varied from 4.3 to 4020 mIU/ml (932-fold) on the expected day of missing a menstrual period. Any results that fall between 4.3 and 4020 mIU/ml must be considered as normal hCG levels. hCG is the only measurable human molecule that can vary to such extremes and still have results considered to be normal. The widely varying results simply means that one woman with an hCG level of 4.3 mIU/ml literally implanted very late, or on the day of missing her menstrual period. A second woman with an hCG level of 4020 mIU/ml might Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00026-8 © 2010 Elsevier Inc. All rights reserved.
282
Human Chorionic Gonadotropin (hCG)
Table 26.1 Detection of hCG Variants in 345 Pregnancy Serum Samples, and 779 Urine Samples. Gestation Age (Weeks Since Last Menses)
n
Median Total hCG (mIU/ml)
hCG-H (% mole/ mole) mean SD
Free hCG -Subunit (% mole/mole) mean SD
12 33 25 16 22 33 7 8 21 57 62 49
12 72 521 1879 16,049 47,641 64,150 102,750 95,650 32,275 21,250 21,025
90 11% 54 27% 42 27% 29 38% 16 13% 7.0 5.4% 5.1 4.4% 4.3 3.1% 2.3 1.5% 1.3 0.61% 0.65 0.60% 0.36 0.16%
12 22% 7.3 5.9% 1.7 0.75% 1.4 0.63% 1.0 0.24% 0.99 0.51% 0.92 0.35% 0.68 0.47% 0.67 0.33% 0.62 0.26% 0.55 0.42% 0.47 0.19%
80 62 62 62 62 50 47 23 1288 124 39 50
13 (1.1) 139 (1.9) 902 (1.7) 2,730 (1.5) 7,500 (0.46) 16,650 (0.34) 50,458 (0.78) 81,625 (0.79) 19,130 (0.20) 6,367 (0.20) 4,337 (0.20) 5,224 (0.25)
83 92% 72 51% 38 46% 22 25% 15 13% 13 10% 8.2 8.5% 5.0 6.0% 4.2 3.8% 3.0 3.8% 1.4 1.3% 1.6 2.3%
18 8.7% 13 5.9% 7.0 4.1% 5.7 3.2% 4.1 2.9% – 3.6 3.2% 2.7 2.3% 3.8 3.0% 3.2 0.59% 3.8 2.2% 6.3 5.2%
A. Serum samples 3 weeks–3 weeks, 6 days 4 weeks–4 weeks, 6 days 5 weeks–5 weeks, 6 days 6 weeks–6 weeks, 6 days 7 weeks–7 weeks, 6 days 8 weeks–8 weeks, 6 days 9 weeks–9 weeks, 6 days 10 weeks–10 weeks, 6 days 11 weeks–13 weeks, 6 days 14 weeks–17 weeks, 6 days 18 weeks–26 weeks, 6 days 27 weeks–40 weeks, 6 days
B. Urine samples 3 weeks–3 weeks, 6 days 4 weeks–4 weeks, 6 days 5 weeks–5 weeks, 6 days 6 weeks–6 weeks, 6 days 7 weeks–7 weeks, 6 days 8 weeks–8 weeks, 6 days 9 weeks–9 weeks, 6 days 10 weeks–10 weeks, 6 days 11 weeks–13 weeks, 6 days 14 weeks–17 weeks, 6 days 18 weeks–26 weeks, 6 days 27 weeks–40 weeks, 6 days
Proportion hCG-H and free -subunit measured as a proportion of total hCG. SD is standard deviation. Values in parentheses are the relationship between serum and urine values.
have implanted on day 16 of her menstrual period, or as many as 2 weeks prior to the time she missed her menstrual period (determined from the average length of three to five previous cycles). During that time, hCG levels could have doubled almost every day, rising as high as 4020 mIU/ml. Both these cases must be considered to be variations of normal and acceptable hCG levels. The problem is that serum and urine hCG levels remain widely variable throughout the length of pregnancy by as much as 932-fold, regardless of whether they are 1 month (serum median 46 mIU/ml, range 4.3–4020 mIU/ml), 3 months (serum median
Pregnancy Testing
283
Table 26.2 Variation in Menstrual Cycles Measured from 408 Healthy Menstrual Cycles from 167 Women.
Mean SD Range
Length of Menstrual Cycle
Within Within Time of Individual Individual LH Peak Variance (Length of Variance Follicular Phase)
Length of Luteal Phase (CycleFollicular Phase)
Within Individual Variance
27.7 2.4 days 20–34 days
2.8 1.6 daysa 0.8–6.7 days
13.2 2.0 days 9–20 days
2.6 3.1 daysc 0.2–15 days
14.7 2.4 days 9–20 days
3.9 3.7 daysb 0.5–18 days
Timing of implantation is calculated from the start of the menstrual period (from time of first menstrual bleeding). Within individual variance is the mean statistical variance among individuals with up to five monitored, sequential menstrual cycles. Mean values are standard deviation (SD). a The individual variance for timing of the menstrual cycle is 5.2 days. b The individual variance for timing of the presumed ovulatory window (follicular phase) is 6.6 days. c The individual variance for length of luteal phase is 4.0 days.
Table 26.3 Menstrual Cycles Leading to Pregnancy Measured from 111 Cycles from 111 Women Who Became Pregnant.
Mean SD Range
Day of Missing Expected Onset of Menses
Time of LH Peak
Timing of Implantation
Implantation Relative to Timing of Ovulatory Window
15.2 2.4 days 10–20 days
24.6 3.1 daysa 16–30 days
9.4 2.0 days 28.7 2.5 days 4–14 days 24–34 days
Implantation Relative to Day of Missing Expected Menses –4.3 2.2 days –10 to 0 days
Timing of implantation is calculated as day of first total hCG detection (1 mIU/ml), per Wilcox et al. [3–5]. Day of missing expected onset of menses is calculated from average length of menstrual cycle for previous 3–5 cycles. Values are standard deviation (SD). a The individual variance in the timing of implantation was 10.1 days.
95,650 mIU/ml, range 2070–265,000 mIU/ml), 6 months (serum median 21,250 mIU/ml, range 310–168,000 mIU/ml), or 9 months pregnant (serum median 21,025 mIU/ml, range 160–115,000 mIU/ml). These widely variable hCG levels are due to varying times of initial implantation [2]. Generally, the same, single pregnancy will yield the highest levels of hCG at 1 month, 3 months, 6 months, and 9 months. Similarly, the same, single pregnancy will yield the lowest levels of hCG at 1 month, 3 months, 6 months, and 9 months. These wide variations must all be considered normal hCG levels. The extreme variations in hCG levels during pregnancy are clearly due to the time of pregnancy implantation—the real starting point of pregnancy. If pregnancy were
284
Human Chorionic Gonadotropin (hCG)
dated from the date of implantation, there would be only minimal variations in hCG levels from person to person. This might allow the process of estimating the day of delivery to be much more accurate; far more accurate than our current estimation method, which is based on the first day of the last menstrual period. If a woman’s 3-month hCG level are as high as 26,5000 mIU/ml, gestational trophoblastic disease would be considered and an ultrasound urgently ordered. Yet this case is just part of the extreme range of what is normal, just a normal pregnancy would be detected. Similarly, if a woman has hCG levels as low as 4.3 mIU/ml on the day of missing her menstrual period, or 2070 mIU/ml at 3 months, the treating obstetrician will predict spontaneous abortion, but it will not occur. In these instances, we are talking about the extremities of normal hCG levels at different weeks of gestation. Questions of when can one first definitively detect pregnancy arise. One has to consider multiple issues, including wide variation in hCG results between pregnancies, timing of implantation, the sensitivity and specificity of a pregnancy test, and biochemical pregnancies or early pregnancy losses. All of these issues must be considered when evaluating the best time to test for pregnancy. As shown [2], approximately 5% of pregnancies do not implant and produce hCG until the predicted day of missing the menstrual period. As such, the next day (day after missing menstrual period) has to be considered the earliest time that one can accurately detect pregnancy. Today, over-the-counter home pregnancy tests publicize use as early as 4 days prior to missing the menstrual period. This is a time at which approximately only 50% of pregnancies have implanted. There is a good possibility that a home test at this time will falsely not detect a future implanting pregnancy. There are many problems and complications that can arise in the reverse situation. A woman who does not want to get pregnant might have a negative home test 4 days prior to missing her menstrual period. These negative test results and heavy advertising from the pregnancy testing company have allowed her to convince herself that she is definitely not pregnant. This woman might then choose to drink and party. Three weeks later, she goes to her obstetrician because she has not had a menstrual period. The obstetrician tests her for pregnancy and concludes that she is in fact pregnant; she simply had not implanted at the time of testing with the home pregnancy test. The drinking and partying might have caused fetal alcohol syndrome. This is an example of the unintentional complications that can result as a consequence of not being properly informed about or prepared for pregnancy. When evaluating the timing of accurate pregnancy testing, biochemical pregnancies (also called early pregnancy losses) have to be considered. Biochemical pregnancies account for 25% of all pregnancies [4,6]. Classically, in these types of pregnancies, the embryo implants inappropriately and only lasts 5–10 days. The inappropriate implantation then fails and leads to a slightly late menstrual period. Most women do not even know that they have had a biochemical pregnancy. The problem is that biochemical pregnancies lead to a transient 5- to 20-day rise in hCG, which might extend into the time of the anticipated menstrual period. One cannot determine pregnancy until approximately 5 weeks past the start of the last period (5 weeks gestation), or until approximately 1 week after missing the menstrual period (in order to exclude biochemical pregnancies and diagnose a definitive, term-type, clinical pregnancy). All
Pregnancy Testing
285
pregnancy tests (home, clinical, and laboratory) are performed at 3 and 4 weeks of gestation all the time all over the world. These pregnancy tests, therefore, must be repeated at 5 weeks gestation to be certain whether a pregnancy is an actual clinical pregnancy versus a late implanting, biochemical pregnancy. Specificity is also a problem with pregnancy testing. As shown in Table 26.1, during the first 2 weeks of pregnancy, hCG-H and free -subunit are predominantly produced. As discussed in Chapter 21, many hCG tests poorly detect or do not detect hCG-H and free -subunit. The ClearBlue Easy and e.p.t. home pregnancy tests, for example, do not detect free -subunit and poorly detect hCG-H. Yet the manufacturers of ClearBlue Easy advertise use as early as 4 days prior to missing the menstrual period. Clearly, these home tests should not be used for detecting pregnancy prior to 5 weeks of gestation, or 1 week after missing the menstrual period. The Siemens Dimension automated laboratory serum hCG test poorly detects free -subunit and hCG-H (see Chapter 21). Similarly, the Wako hCG-CTP and Serono MAIAclone manual hCG tests very poorly detect free -subunit and hCG-H. Once again, we say that laboratories using these three test brands should tell physician subscribers that their hCG test cannot accurately detect pregnancy prior to 5 weeks of gestation. Numerous point-of-care or clinical pregnancy tests do not detect free -subunit and very poorly detect hCG-H, including the Inverness Acceava, the Quidel QuickVue, and Mainline Maxie hCG tests. These brands of tests should not be used until 5 weeks of gestation. We examine serum and urine hCG throughout the length of gestation in Table 26.1. As shown, hCG rises sharply from the time of implantation and peaks at 10 weeks gestation (the point since the start of the last menstrual period). Over the following 5 weeks, serum hCG levels fall to approximately 1/5 of peak levels; urine hCG levels fall to about 1/10 of peak levels. Serum and urine hCG levels then remain relatively constant until term. At the time of implantation, urine hCG levels are approximately equal to serum hCG. Up to approximately 6 weeks gestation, urine hCG is equivalent to or exceeds serum hCG. Urine hCG then drops to approximately 0.8 serum hCG until 12 weeks of pregnancy. Urine hCG then drops down to 0.2 serum hCG throughout the remainder of pregnancy. As established, as hCG advances throughout the months of pregnancy, it become less charged or sialylated [7,8]. As hCG becomes less charged or sialylated, an increasingly higher proportion of molecules are removed from the circulation by the liver’s galactose receptor. This might explain the lower concentrations of hCG observed in urine in the later trimesters of pregnancy (see Chapter 21). As shown in Table 26.1, the proportion of free -subunit is notably higher in urine than in serum samples. This might be due to the much more rapid clearance of free -subunit than hCG dimer from the circulation [9,10]. Free -subunit is generated from the dissociation of regular hCG dimer and hCG-H dimer in the circulation and rapidly cleared by the kidneys [9,10]. Urine also contains significant amounts of -core fragment; concentrations that exceed hCG levels in pregnancy urine between 7 weeks of gestation until term (not shown in Table 26.1). -core fragment is the terminal degradation product of free -subunit in serum. This is the secondary component of free -subunit in urine.
286
Human Chorionic Gonadotropin (hCG)
References [1] Cole LA. Human chorionic gonadotropin and associated molecules. Expert Rev Mol Diagn 2009;9:51–73. [2] Cole LA, Ladner DG, Byrn FW. The normal variabilities of the menstrual cycle. Fertil Steril 2009;91:522–7. [3] Wilcox AJ, Baird DD, Dunson D, McChesney R, Weinberg CR. Natural limits of pregnancy testing in relation to the expected menstrual period. J Am Med Assoc 2001; 286:1759–61. [4] Wilcox AJ, Baird DD, Weinberg CR. Time of implantation of the conceptus and loss of pregnancy. N Engl J Med 1999;340:1796–9. [5] Wilcox AJ, Weinberg CR, Baird DD. Timing of sexual intercourse in relation to ovulation. Effects of the probability of conception, survival of the pregnancy, and sex of the baby. N Engl J Med 1995;333:1517–21. [6] Dickey RP, Olar TT, Taylor SN, Curole DN, Harrigill K. Relationship of biochemical pregnancy to pre-ovulatory endometrial thickness and pattern in patients undergoing ovulation induction. Hum Reprod 1993;2:327–30. [7] Kovalevskaya G, Birken S, Kakuma T, O’Connor JF. Early pregnancy human chorionic gonadotropin (hCG) isoforms measured by an immunometric assay for choriocarcinomalike hCG. J Endocrinol 1999;161:99–106. [8] Ashitaka Y, Tokura Y, Tane M, Mochizuki M, Tojo S. Studies on the biochemical properties of highly purified hCG. Endocrinol 1970;87:233–44. [9] Wehmann RE, Nisula BC. Metabolic and renal clearance rates of purified human chorionic gonadotropin. J Clin Invest 1981;68:184–94. [10] Nisula BC, Wehmann RE. Distribution, metabolism, and excretion of human chorionic gonadotropin and its subunits in man. In: Segal SJ, editor. Chorionic Gonadotropin. New York: Plenum Press; 1981. p. 199–215.
27 Predicting Spontaneously
Aborted (SAB) Pregnancies Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
There are two major types of pregnancy failures in humans: spontaneous abortions (SABs) and biochemical pregnancies. SABs (miscarriages) occur within the first or second trimesters of pregnancy. After the second trimester, they are considered stillborn pregnancies. Worldwide, first- and second-trimester miscarriages account for 15–20% of all pregnancies [1–3]. In the United States, the miscarriage rate accounts for 16% of all pregnancies [1–2]. Biochemical pregnancies are pregnancies that fail to implant appropriately, rapidly fail within 7 days of implantation, and lead to a late menstrual bleeding (1–3 days late). Biochemical pregnancies account for 25% of all pregnancies in the United States [4,5]. Together, biochemical pregnancies and SABs account for approximately 41–45% of human pregnancies. Woman who are eager to have a baby are also eager to avoid biochemical pregnancies and SABs. For example, a woman is now 6 weeks pregnant, has missed her menstrual period, and has passed the time period when biochemical pregnancies occur, but what is the chance of an SAB terminating her pregnancy versus her making it to term? I know that my own daughter was in this situation recently after having multiple SABs, scared to tell anyone that she is pregnant until she has gotten through the first trimester. Fertility specialists have devised numerous test protocols in order to predict pregnancy failures. Some say that measurement of spot serum progesterone at 4–8 weeks pregnancy is the best indicator [6–8], but others claim that spot hCG tests are a better indicator [9–11]. Multiple fertility specialists now measure the hCG doubling rate over a 2-day period as the best indicator [12–15]. In truth, none of these tests are really a good predictor of term pregnancy versus SAB outcome. All too often, the USA hCG Reference Service hears from women who were told they had a deficient hCG doubling rate or low progesterone and that an SAB was impending. All the USA hCG Reference Service can say is that it is all statistics and that the tests are not very accurate. All too often we hear that a woman with unduly low hCG doubling rate and low progesterone had a normal term pregnancy. The advice, therefore, is to hope and pray, which is not much comfort for a woman dealing with low doubling rate. In early pregnancy, hCG biologically promotes progesterone production. Thus, progesterone is just a secondary measurement of hCG. We ask ourselves, does such a hormone really have anything to do with pregnancy failure? In this chapter, we also look at hyperglycosylated hCG (hCG-H) as another pregnancy predictor. This Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00027-X © 2010 Elsevier Inc. All rights reserved.
288
Human Chorionic Gonadotropin (hCG)
molecule biologically controls implantation of pregnancy and could be an excellent indicator of failing pregnancies or more correctly a marker of pregnancy success.
27.1 Pregnancy Outcome Markers In 2005, the USA hCG Reference Service performed a large blind study, comparing spot hCG measurements, spot progesterone measurements, and the hCG doubling rate. Serum samples from 4 to 8 weeks gestation were collected from women who were attending the University of New Mexico infertility clinic. Progesterone and hCG were blindly measured using the automated Siemens Immulite assay. In total, 67 pregnancies were followed (Table 27.1), 36 pregnancies ended in a SAB and 31 went to term normal delivery. An average of 12 ng/ml optimum cutoff was taken from the 10–15 ng/ml cutoff of the spot progesterone serum assay used by other centers [6–8]. The 12 ng/ml cutoff would detect 25% of term pregnancies and predict 67% of SAB cases (Table 27.1). The predictive value positive (adjusted for the 16% incidence of SABs) was just 33%. The area under the receiver operating characteristic (ROC) curve (ROC test) indicated 77% accuracy independent of the cutoff. As shown in Table 27.1, spot hCG was also Table 27.1 A Comparison Between hCG and hCG-H Doubling Times, and Single Measurements of hCG and hCG-H, in Predicting Pregnancy Outcome at 4–8 Weeks Gestation. Single Progesterone Number of term outcome 67 cases Cutoff used to predict SAB 12 ng/ml False positive rate, term outcome samples Number of SAB outcome cases SAB outcome cases detection rate Adjusted prediction valuea Area under ROC curve (accuracy independent of cutoff) a
Single hCG (mIU/ml)
hCG Doubling Single hCG-H Rates (mIU/ml) (ng/ml)
67
30
67
22
9.0 ng/ml
25%
15%
2.2 days double 36%
36
36
18
36
67%
38%
28%
89%
33%
34%
28%
89%
77%
78%b
60%b
92%b
2.9%
Predictive value for detecting a failure, considering 16% of pregnancies end in SAB. Percent prediction of SAB, considering these statistics, is divided by total prediction (SABs plus term pregnancy false positive). b A statistically significant improvement shown in ROC area under the curve, between single serum hCG-H, single serum hCG, and hCG doubling test. P 0.00005; hCG-H versus. hCG, hCG-H versus hCG doubling tests, and hCG and hCG doubling tests.
Predicting Spontaneously Aborted (SAB) Pregnancies
289
evaluated and used at an optimum cutoff of 22 mIU/ml. This falsely detected 15% of term pregnancies and only predicted 38% of SAB cases. The predictive value positive of hCG was 34%, similar to the predictive value positive of spot progesterone. According to the ROC test, the accuracy rate was 78% and similar to the rate of progesterone. In terms of prediction, we inferred that spot hCG is similar to progesterone. We carefully considered the hCG doubling rate as an indicator for SAB. We examined doubling over 2.2 days, as determined by others [9–11]. As shown in Table 27.1, 36% of term pregnancy cases were positive in this test and 73% of SAB cases were predicted. The predictive value positive was just 28% and the area under the ROC was just 60%. These results showed that the hCG doubling test was no better than spot hCG or progesterone measurement. As indicated earlier, spot hCG tests had the best predictive value (38%), but those results were only marginally better than guess work. Progesterone and hCG results are interrelated. All the samples that gave positive results with hCG, for example, gave positive results with the progesterone. Thus, we looked at combining test values of spot progesterone and spot hCG, and unsurprisingly the combination did not improve prediction statistics at all. None of the tests was predictive enough for a physician to be able to confidently tell a woman that she will have an SAB or to suggest a pregnancy termination, yet numerous women have been told to terminate pregnancies due to low hCG and progesterone concentrations [1,2]. It is well established that hCG values vary greatly in every human pregnancy. A normal term pregnancy can have hCG values ranging from 23 to 4653 mIU/ml (202) at 4 weeks gestation and 114–45,800 mIU/ml (402) at 5 weeks gestation [16]. A study was published in 2006 [17]. Physicians need to realize that extremely low hCG values and secondary low progesterone concentrations can both be normal. New tests are urgently needed to accurately predict failing pregnancies.
27.2 Early Pregnancy hCG We have known since 1987 that the hCG produced in choriocarcinoma and other gestational trophoblastic invasive disease conditions was hCG-H [18]. A relationship between extent of invasion in gestational trophoblastic disease cases and the presence of hCG-H was confirmed in hundreds of cases examined using a specific hCG-H immunoassay [19–21]. John O’Connor and Galina Kovalevskaya showed and confirmed the predominant presence of choriocarcinoma-like hCG (hCG-H) in pregnancy at the time of implantation [23,24]. This led to the assumption that the same molecule that controlled trophoblast invasion and malignancy in gestational trophoblastic disease cells (hCG-H) also promoted trophoblast invasion during implantation of pregnancy. This concept was later proven [22]. As first shown by O’Connor, hCG-H is the predominant form of hCG produced in early pregnancy [23]. Table 27.2 shows the most recent serum and urine data for term pregnancies as determined by the USA hCG Reference Service [25]. These data were accumulated over 13 years of examining hCG and hCG-H via serum and urine testing, as well as first- and second-trimester Down syndrome screening
290
Human Chorionic Gonadotropin (hCG)
Table 27.2 Detection of Total hCG and hCG-H in Early Pregnancy; from 90 Pregnancy Serum and 266 Pregnancy Urine Samples. Gestation Age (Weeks Since Last Menses)
n
Median Total hCG (mIU/ml)
hCG-H (%) mean SD
12 33 25 16
12 72 521 1879
90 11% 54 27% 42 27% 29 38%
80 62 62 62
13 139 902 2730
83 92% 72 51% 38 46% 22 25%
A. Serum samples 3 weeks–3 weeks, 6 days 4 weeks–4 weeks, 6 days 5 weeks–5 weeks, 6 days 6 weeks–6 weeks, 6 days
B. Urine samples 3 weeks–3 weeks, 6 days 4 weeks–4 weeks, 6 days 5 weeks–5 weeks, 6 days 6 weeks–6 weeks, 6 days
Proportion hCG-H is measured as a proportion of total hCG. SD is standard deviation.
studies. As shown, hCG-H accounts for 90 11% of total serum hCG, and 83 92% of total urine hCG in week 3 of gestation when hCG is first produced. It accounts for 54 27% of total hCG in serum and 72 51% of total hCG in urine during week 4 of gestation (the week following a missed menstrual period). A sample’s proportion of hCG-H greatly declines thereafter. It is suggested that hCG-H is the only form of hCG produced in very early pregnancy (for the purpose of promoting implantation). Once implantation is successful, hCG-H levels begin to decline.
27.3 Testing for hCG-H Following in the footsteps of O’Connor and Kovalevskaya, we considered whether hCG-H could be a marker for biochemical pregnancy and SABs. Using IVF and normal pregnancies, they had shown that hCG-H predicted failure with a predictive value of 80% [23,24]. If hCG-H is the true promoter of implantation, the root event in preventing failures, then failure of implantation would surely be marked by this agent. As shown in Table 27.1, an hCG-H cutoff 9.0 ng/ml predicted future SABs at 4–8 weeks of gestation (73% detection rate at 2.9% false positive rate). It was calculated that the positive predictive value was 89%, and 92% accuracy as indicated by the area under the ROC curve [17]. In terms of predicting pregnancy failures, spot hCG-H is a great improvement over spot hCG, spot progesterone, and hCG doubling tests. Figure 27.1 illustrates the use of hCG-H to predict future SAB at 9 and 13 ng/ml cutoffs. The USA hCG Reference Service recently examined urine samples from 42 term pregnancies and 20 failures; 10 of these failures were biochemical pregnancies, 10 were SABs. The samples were examined on the day of pregnancy implantation, when low levels of hCG and hCG-H are first detectable [26]. An absolute difference
Predicting Spontaneously Aborted (SAB) Pregnancies
291
Figure 27.1 Scattergram showing single serum test hCG-H results in 67 term outcome (black circles) and 36 failing (SAB) pregnancies (clear circles) at 4–8 week gestation. The cutoffs illustrated are at 9 and 13 ng/ml.
Figure 27.2 Term pregnancy urine samples (n 42) and SAB pregnancy urine samples (n 20) on day of pregnancy implantation (first appearance of hCG).
292
Human Chorionic Gonadotropin (hCG)
was found (Figure 27.2). On the day of implantation, 100% of term pregnancies exceeded 50% hCG-H (proportion of total hCG). Most cases had 100% hCG-H; a smaller number of cases were in the 55–100% hCG-H range. Of the 20 failed pregnancies, 13 (65%) were below 50% hCG-H because they failed to produce the amount of hCG-H needed to implant appropriately. This is consistent with the findings of Norwitz et al. [27] and Semprini and Simoni [28]; both groups showed 67% of failed pregnancies are due to failed implantation. The remainder, and assumingly the 7 of 20 failure cases that exceeded 50% hCG-H, have genetic or immune causes of failure. These data have been published [29]. On the day of pregnancy implantation, it is hCG-H that can make an absolute differentiation between pregnancies that will go to term and pregnancies that will fail (Figure 27.2). During 5–8 weeks gestation, hCG-H can also make a 92% accurate prediction of pregnancy outcome. hCG-H is presently the best single test for examining pregnancy outcome—so why is it not being used by all pregnancy and infertility specialists? The first application discovered for hCG-H was an improved means of detecting Down syndrome pregnancies [30]. At that time, all rights to hCG-H and its only monoclonal antibody (B152) were exclusively leased to Quest Diagnostics. In our opinion, the discovery of its use in applications such as SAB screening suggests a major application. Quest Diagnostics, however, has not adopted this application. We have been told that Quest Diagnostics leased the antibody to Focus Diagnostics and that the test will be available for other laboratories to use. Once available, we hope that spot hCG-H assays will become a well-established SAB screening method.
References [1] Puscheck EE, Pradhan A. First-Trimester pregnancy loss, http://emedicine.medscape. com/article/266317-overview; 2010 [2] Liliane C, Parazzini F, Di Cintio E, Zanconato G, Benzi G, Bortolus R, et al. Odds of miscarriage with specific risk factors: breakdown of pregnancy loss odds by groups, http://miscarriage.about.com/od/riskfactors/a/miscrates.htm; 2009 [3] Liu DTY, Jeavons B, Preston C, Pearson D. A prospective study of spontaneous miscarriage in ultrasonically normal pregnancies and relevance to chorionic villus sampling. Prenat Diagn 2005;7:2237. [4] Levy T, Goldman JA, Dicker D, Ashkenazi J, Feldberg D. Very early pregnancy wastage in in vitro fertilization and embryo transfer (IVF-ET). J Assist Reprod Genet 1991;8:250–3. [5] Wilcox AJ, Weinberg CR, O’Connor JF, Baird DD, Schlatterer JP, Canfield RE, et al. Incidence of early loss of pregnancy. N Engl J Med 1988;319:189–94. [6] Lin YS, Liu CH. Prediction of early pregnancy outcomes. Int J Gynecol Obstet 1995; 51:133–8. [7] Riss PA, Radivojevic K, Bieglmayer C. Serum progesterone and human chorionic gonadotropin in very early pregnancy: implications for clinical management. Eur J Obstet Gynecol Reprod Biol 1989;32:71–7. [8] Lower AM, Yovich JL. The value of serum levels of oestradiol, progesterone and -human chorionic gonadotropin in the prediction of early pregnancy loss. Hum Reprod 1992;7:711–17.
Predicting Spontaneously Aborted (SAB) Pregnancies
293
[9] Muleba N, Casey B, Wells CE, Mcintire D, Leveno K. Human chorionic gonadotropin levels following spontaneous abortion in the first trimester. Am J Obstet Gynecol 2005; 193:S116–17. [10] Frishman GN, Hackett RJ, McGrath M, Metheny WP. Evaluation of the prognosis of in vitro fertilization pregnancies with initially low serum human chorionic gonadotropin levels. Am J Obstet Gynecol 2001;185:2392–5. [11] Homan G, Brown S, Moran J, Homan S, Kerin J. Human chorionic gonadotropin as a predictor of outcome in assisted reproductive technology pregnancies. Fertil Steril 2000;73:270–4. [12] Daneshmand S, Shapiro B, Garner F, Aguirre M, Thomas S, Udoh A. Very early hCG dynamics predict pregnancy loss at up to 10 weeks gestation. Fertil Steril 2007;88:S159. [13] Hamori M, Stuckensen JA, Rumpf D, Kniewald T, Kniewald A, Marquez M. Early pregnancy wastage following late implantation of embryos after in-vitro fertilization and embryo transfer. Hum Reprod 1989;4:714–17. [14] Zegers-Hochschild F, Altieri E, Fabres C, Fernández E, Mackenna A, Orihuela P. Predictive value of human chorionic gonadotropin in the outcome of early pregnancy after in-vitro fertilization and spontaneous conception. Hum Reprod 1994;9:1550–5. [15] Bjercke S, Tanbo T, Dale PO, Mørkrid L, Åbyholm T. Human chorionic gonadotropin concentrations in early pregnancy after in-vitro fertilization. Hum Reprod 1999; 14:1642–6. [16] Cole LA, Khanlian SA, Sutton JM, Davies S, Stephens N. Hyperglycosylated hCG (invasive trophoblast antigen, ITA) a key antigen for early pregnancy detection. Clin Biochem 2003;36:647–5. [17] Sutton-Riley JM, Khanlian SA, Byrn FW, Cole LA. A single serum test for measuring early pregnancy outcome with high predictive value. Clin Biochem 2006;39:682–7. [18] Cole LA. The O-linked oligosaccharides are strikingly different on pregnancy and choriocarcinoma hCG. J Clin Endocrinol Metab 1987;65:811–13. [19] Cole LA, Kohorn EI, Kim GS. Detecting and monitoring trophoblast disease: new perspectives in measuring human chorionic gonadotropin levels. J Reprod Med 1994; 39:193–200. [20] Cole LA. Immunoassay of hCG, its free subunits and metabolites. Clin Chem 1997;43:2233–43. [21] Cole LA, Dai D, Leslie KK, Butler SA, Kohorn EI. Gestational trophoblastic diseases: 1. Pathophysiology of hyperglycosylated hCG-regulated neoplasia. Gynecol Oncol 2006; 102:144–9. [22] Cole LA, Khanlian SA, Riley JM, Butler SA. Hyperglycosylated hCG (hCG-H) in gestational implantation, and in choriocarcinoma and testicular germ cell malignancy tumorigenesis. J Reprod Med 2006;51:919–29. [23] O’Connor J, Ellish N, Kakuma T, Shlatterer J, Kovalevskaya G. Differential urinary gonadotropin profiles in early pregnancy and early pregnancy loss. Prenat Diagn 1999; 18:1232–40. [24] Kovalevskaya G, Birken S, Kakuma T, Ozaki N, Sauer M, Lindheim S, et al. Differential expression of human chorionic gonadotropin (hCG) glycosylation isoforms in failing and continuing pregnancies: preliminary characterization of the hyperglycosylated hCG epitope. J Endocrinol 2002;172:497–506. [25] Cole LA. Human Chorionic Gonadotropin and Associated Molecules. Expert Rev Molec Diag 2009;9:51–73. [26] Wilcox AJ, Baird DD, Weinberg CR. Time of implantation of the conceptus and loss of pregnancy. N Engl J Med 1999;340:1796–9.
294
Human Chorionic Gonadotropin (hCG)
[27] Norwitz ER, Schust DJ, Fisher SJ. Implantation and the survival of early pregnancy. N Engl J Med 2001;345:1400–8. [28] Semprini AE, Simoni G. Not so efficient reproduction. Lancet 2000;356:257–8. [29] Sasaki Y, Ladner DG, Cole LA. Hyperglycosylated human chorionic gonadotropin and the source of pregnancy failure. Fertil Steril 2008;89:1781–6. [30] Cole LA, Shahabi S, Oz UA, Bahado-Singh RO, Mahoney MJ. Hyperglycosylated hCG (invasive trophoblast antigen) immunoassay: a new basis for gestational Down syndrome screening. Clin Chem 1999;45:2109–19.
28 hCG, Hyperglycosylated hCG,
and Free ß-Subunit in Predicting Down Syndrome Pregnancies and Preeclampsia Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
Back in 1995, with new discovery of hyperglycosylated hCG (hCG-H), I presented the story of the discovery to the Obstetrics and Gynecology faculty at Yale University. I was approached by a member of the Maternal Fetal Medicine division at Yale University to test a library of second-trimester Down syndrome pregnancy samples and controls. Screening pregnancies for Down syndrome was the first application found for hCG-H [1]. Why should hCG-H be a marker of Down syndrome pregnancy? hCG and hCG-H take on different profiles in Down syndrome pregnancies. The limited fusion of trophoblast cells in Down syndrome causes an unduly high proportion of cytotrophoblast cells to remain unfused. This leads to an unduly high production of hCG-H in Down syndrome pregnancies [2–4]. In this chapter, we look at the procedures involved in screening for Down syndrome pregnancies.
28.1 Down Syndrome Screening Trisomy 21, also referred to as Down syndrome, is the most common genetic abnormality that complicates pregnancies. Trisomy 21 and other genetic monosomies, trisomies, and triploidies can be detected through a cytogenetic analysis procedure. Amniocentesis is performed at 14–26 weeks of gestation and can lead to late pregnancy termination between 16 and 30 weeks of gestation (for those choosing to abort). Amniocentesis is complicated by an approximate 1 in 300 failure rate in which the fetus is aborted by the procedure. Chorionic villous sampling can determine cytogenetics at 9–11 weeks gestation. The prenatal risk for analysis by chorionic villous sampling is approximately 2 times higher; approximately 1 in 150 failures, depending on the expertise of the center performing the procedure. The advantage of chorionic villous sampling is that abortion can be performed at 12–16 weeks of gestation. As age advances, risk for Down syndrome pregnancies increases significantly, occurring in 1 in 105 pregnancies in women age 40 or older [5]. Women below age Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00028-1 © 2010 Elsevier Inc. All rights reserved.
296
Human Chorionic Gonadotropin (hCG)
30 have a 1 in 1000 Down syndrome risk [5]. Because more women achieve pregnancy prior to age 30, more Down syndrome fetus cases occur in women younger than age 30. Thus, it was statistically more appropriate not to take the higher risks of amniocentesis (1 in 300 failures) or chorionic villous sampling (1 in 150 failures) to determine their cytogenetic outcomes. New tests and supplemental procedures were needed for the younger women warranting the risks of cytogenetic screening. In 1984, Chard and colleagues [6] showed that -fetoprotein was a rough indicator of Down syndrome pregnancy. In 1987, Bogart and colleagues [7] showed that Down syndrome pregnancies were associated with 2 times higher hCG levels in the second trimester of pregnancy. Wald and colleagues [8] then showed that unconjugated estriol was also a useful marker. This led to the development of the “triple test,” which used a combination of hCG, -fetoprotein, and unconjugated estriol as markers for predicting those at high risk for Down syndrome pregnancies. This triple test normally had a 60% detection rate of Down syndrome pregnancy and a 5% false positive rate. Even though there was a low false positive rate and minimal risks for women under 30 years old, a woman having a negative result from a positive triple test still had an approximate 1 in 100 chance of having a Down syndrome fetus. The chance was enough to justify amniocentesis, which has a 1 in 300 pregnancy failure rate. The addition of inhibin A as a fourth marker in the quadruple test improved the sensitivity for identifying Down syndrome pregnancies (68% detection rate with a 5% chance of false positive results) [9]. Major improvements also arose from using higher sensitivity hCG-H in place of regular hCG (80% detection rate with a 5% chance of false positive results) [10]. All these biochemical tests were, however, still limited by the 5% false positive results. There was still a 1 in 50 to 1 in 100 chance of a positive test resulting in a Down syndrome pregnancy, and each test was limited to the second trimester of pregnancy. In the late 1990s, serum markers were identified for first-trimester Down syndrome screening; hCG free ß-subunit was preferred over regular hCG [11]. These serum markers required high-risk chorionic villous sampling, which was required in order to determine cytogenetics in those indicated as being at risk for positive Down syndrome tests. A woman positive in first-trimester biochemical testing had a 1 in 100 chance of having a Down syndrome pregnancy, and chorionic villous sampling had a 1 in 150 chance of destroying the pregnancy. For women with genetic predispositions to Down syndrome, it was a gamble. In 1992, Nicolaides and colleagues [12] found that fetal nuchal translucency (measured using ultrasound) was a powerful indicator of Down syndrome pregnancy. The combination of ultrasound nuchal translucency with laboratory measurement of free ß-subunit and pregnancy-associated plasma protein-A became the standard combination of tests to assess risk for a Down syndrome fetus. In 2003 [13], free ß-subunit was replaced with hCG-H and the sensitivity of the Nicolaides test combination improved. Today, the nuchal translucency combinations with pregnancyassociated plasma protein-A (PAPP-A) and free ß-subunit have a 94% detection rate with a 5% chance of false positive results. Combinations using hCG-H in place of free ß-subunit have a 95% accurate detection rate with a 5% chance of false positive results. Regardless of whether the test is performed in the first or second trimester,
hCG, Hyperglycosylated hCG, and Free ß-Subunit
297
the hCG variant (hCG, hCG-H, or free ß-subunit) is a major part of the test. Both first- and second-trimester combinations are used throughout the world to predict Down syndrome. In terms of predicting a Down syndrome pregnancy, the hCG-H, PAPP-A, and nuchal translucency combination are the best methods available today [13–15]. Although this combination of markers can yield a positive result for Down syndrome pregnancy, it still only has a limited chance of indicating a Down syndrome fetus. Furthermore, these markers require chorionic villous sampling to give a definitive prediction, and chorionic villous sampling is a high-risk procedure. New tests that examine fetal chromosomes in separate maternal blood are on the horizon [16]. Such complex methodologies will eventually replace biochemical testing, ultrasound methods, and amniocentesis, or chorionic villous sampling will be improved and the associated risks removed.
28.2 Preeclampsia Screening Hypertensive disorders, pregnancy induced hypertension (PIH), preeclampsia, eclampsia and hemolysis, elevated liver enzymes, and low platelet count (HELP syndrome) are the most deadly complications of pregnancy. Each can lead to hemorrhage and death. They are most common in nulliparous, or first-time, pregnancies. All told, they occur in approximately 7% of all pregnancies. As discussed in Chapter 36, hypertensive disorders might be a complication deriving from adequate hCG-H production, appropriate invasion, and establishment of efficient hemochorial placentation during pregnancy [17–19]. As such, impending hypertensive disorders can be predicted by measuring hCG-H in serum or urines during the first and second trimesters of pregnancy [20]. Ray Bahado-Singh and colleagues tested 27 samples from women genetically destined to develop hypertensive disorders and 510 controls [20]. As illustrated in Figure 28.1, using multiples of the 500 300 200
n = 510 n=3
hCG-H (ng/ml)
100 50 30 20
50th cen
tile
10 5 3 2
n = 24
1 14
15
16
17
18
19
Weeks of gestation
Figure 28.1 hCG-H measurements as a marker of forthcoming hypertensive disorder.
298
Human Chorionic Gonadotropin (hCG)
median (MoM) algorithms, hCG-H is an outstanding indicator of hypertensive disease. A progressive increase in risk is noted with decreasing MoM. As such, hCG-H presents itself as an ideal marker for predicting pregnancy hypertensive disorders.
References [1] Cole LA, Cermik D, Bahado-Singh R. Oligosaccharide variants of hCG-related molecules: potential screening markers for Down syndrome. Prenat Diagn 1997;17:1188–90. [2] Kato Y, Braunstein GD. Discordant secretion of placental protein hormones in differentiating trophoblasts in vitro. J Clin Endocrinol Metab 1989;68:814–20. [3] Knöfler M. Regulation of hCG during normal gestation and in pregnancies affected by Down’s syndrome? Mol Hum Reprod 1999;5:895–7. [4] Frendo JL, Vidaud M, Guibourdenche J, Luton D, Muller F, Bellet D, et al. Defect of villous cytotrophoblast differentiation into syncytiotrophoblast in Down’s syndrome. J Clin Edcrinol Metab 2000;85:3700–7. [5] Newberger DS. Down syndrome: prenatal risk assessment and diagnosis. Am Fam Physician 2000;62:837–8. [6] Chard T, Lowings C, Kitau MJ. Alphafetoprotein and chorionic gonadotropin levels in relation to Down’s syndrome. Lancet 1984;29:750–1. [7] Bogart MH, Pandian MR, Jones OW. Abnormal maternal serum chorionic gonadotropin levels in pregnancies with fetal chromosome abnormalities. Prenat Diagn 1987; 7:623–30. [8] Wald NJ, Cuckle HS, Densem JW, Nanchahal K, Canick JA, Haddow JE, et al. Maternal serum unconjugated oestriol as an antenatal screening test for Down’s syndrome. Br J Obstet Gynaecol. 1988;95:334–41. [9] Wald NJ, Densem JW, Muttukvishna S, Knight PG. Prenatal screening for Down syndrome using inhibin A as a serum marker. Prenat Diagn 1996;16:143–53. [10] Cole LA, Shahabi S, Oz UA, Bahado-Singh RO, Mahoney MJ. Hyperglycosylated hCG (invasive trophoblast antigen) immunoassay: a new basis for gestational Down syndrome screening. Clin Chem 1999;45:2109–19. [11] Macri JN, Kasturi RV, Krantz DA, Cook EJ, Moore ND, Young JA, et al. Maternal serum Down syndrome screening: free beta protein is a more effective marker than human chorionic gonadotropin. Am J Obstet Gynecol 1990;163:1248–53. [12] Nicolaides KH, Azar G, Byrne D, Mansur C, Marks K. Fetal nuchal translucency: ultrasound screening for chromosomal defects in the first trimester of pregnancy. Br Med J 1992;304:867–89. [13] Palomaki GE, Knight GJ, Roberson MM, Cunningham GC, Lee JE, Strom CM, et al. Invasive trophoblast antigen (hyperglycosylated human chorionic gonadotropin) in second-trimester maternal urine as a marker for Down syndrome: preliminary results of an observational study on fresh samples. Clin Chem 2003;50:182–9. [14] Palomaki GE, Knight GJ, Neveux LM, Pandian R, Haddow JE. Maternal serum invasive trophoblast antigen and first-trimester Down syndrome screening. Clin Chem 2005; 51:1499–504. [15] Quest Diagnostics, Maternal Serum Screen, First Trimester. Quest Diagnostics Incorporated, http://www.questdiagnostics.com/hcp/topics/genetictesting/mss_t1.html; 2000–10. [16] Poon L, Leung T, Lau T, Lo Y. Prenatal detection of fetal Down’s syndrome from maternal plasma. Lancet 2000;356:1819–20.
hCG, Hyperglycosylated hCG, and Free ß-Subunit
299
[17] Robillard PY, Hulset TC, Deka CG. Preeclampsia and human reproduction. An essay of a long term reflection. J Reprod Immunol 2003;59:93–100. [18] Burton GJ. Placental oxidative stress: from miscarriage to preeclampsia. J Soc Gynecol Invest 2004;11:342–52. [19] Salas SP. What causes pre-eclampsia? Clin Obstet Gynecol 1999;13:141–57. [20] Bahado-Singh RO, Oz AU, Kingston JM, Shahabi S, Hsu CD, Cole LA. The role of hyperglycosylated hCG in trophoblast invasion and the prediction of subsequent preeclampsia. Prenat Diagn 2002;22:478–83.
29 hCG in Monitoring Gestational Trophoblastic Diseases Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
In terms of managing gestational trophoblastic diseases, there is no better example of a perfect tumor marker than total hCG (hCG plus all hCG variants). In hydatidiform mole cases, invasive mole cases, gestational trophoblastic neoplasm (choriocarcinoma without histology), and choriocarcinoma, total hCG measures the exact extent of disease, estimates the number of cells, gages the success of therapy, and provides the first indication of disease recurrence (Figures 29.1–29.3). It is an extremely sensitive test, showing the presence of disease when tumors can be palpated, scanned, or visualized. There is no known case of a hydatidiform mole or choriocarcinoma not producing total hCG, suggesting 100% sensitivity and 100% specificity as a tumor marker. Total hCG tests are generally sensitive down to 1 mIU/ml, yet tumors cannot normally be visualized by MRI until hCG exceeds approximately 2000 mIU/ml [1]. You cannot manage gestational trophoblastic disease cases without measuring total hCG. In monitoring cases of complete and partial hydatidiform mole, the concentration of hCG in serum and urine is an exact indicator of the quantity of remaining tissue. With a new mole, hCG levels might start as high as 100,000–1,000,000 mIU/ml
Figure 29.1 USA hCG Reference Service referral. The figure shows use of hCG in monitoring evacuation of complete hydatidiform mole. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00029-3 © 2010 Elsevier Inc. All rights reserved.
302
Human Chorionic Gonadotropin (hCG)
Figure 29.2 USA hCG Reference Service referral. The figure shows use of hCG in monitoring an invasive complete hydatidiform mole. Methotrexate and Actinomycin D chemotherapy are given.
Figure 29.3 USA hCG Reference Service referral. The figure shows use of hCG in monitoring choriocarcinoma, EMA-CO (etoposide, methotrexate and actinomycin D cycling weekly with cyclophosphamide and onvocin (vincristine)), and EMA-EP (etoposide and cisplatin) are given.
and decline by orders of magnitudes as hCG clears following dilation and curettage (D&C) (Figure 29.1). A plateau in hCG decline or a new rise in hCG from an undetectable hCG level is an absolute indicator of persistent or invasive mole and confirms the need for chemotherapy. In choriocarcinoma, invasive mole, and gestational trophoblastic neoplasm cases, hCG is once again a direct indicator of the presence and extent of disease. Monitoring hCG follows the disappearance of malignancy accurately (Figures 29.2 and 29.3). It is normal to diagnose recurrence of gestational trophoblastic disease solely based on the appearance of measurable hCG.
hCG in Monitoring Gestational Trophoblastic Diseases
303
I remember being at the Liaoning Tumor Institute in Manchuria, China, 20 years ago. At that time, they had as many as 30 choriocarcinoma cases in this hospital, possibly more cases than any hospital in the world. I particularly remember watching their total hCG test, an hCG radioimmunoassay, being run each day on each choriocarcinoma case. Each person that ran the hCG tests only had the responsibility of running one test on each case, as stated by the employment contract. In my opinion, this is communism in practice because the employee did nothing else. The laboratory technician ran only undiluted hCG, not hCG diluted 1/10, 1/100, and 1/1000 (which cover hCG levels that stretch as high as 1,000,000 mIU/ml). As a result, what were actually very high hCG levels were recorded as just 1000 mIU/ml. As hCG levels diminished slowly over time from 1,000,000 mIU/ml down to 1000 mIU/ml, the test results would be reported as the same, 1000 mIU/ml every day for the first month of treatment. Only after the first month would hCG levels drop below 1000 mIU/ml. The Institute likely missed many recurrences of disease (see Figure 29.2). hCG has to be measured over the full range (1–1,000,000 mIU/ml) by testing serum at varying dilutions; of pure (no dilution), 1/10, 1/100, and 1/1000 in serum. When monitoring hCG levels in cases of gestational trophoblastic diseases, as a general rule, three consecutive rising hCG levels (Figure 29.2) warrant the use of chemotherapy. Approximately one-third of complete hydatidiform mole cases need chemotherapy because of the presence of residual tissue after D&C evacuation. Commonly, an “elbow,” or plateau is observed during hCG decline (as indicated in Figure 29.2). Once again, three stagnant hCG results in the declination of hCG levels warrants the need for chemotherapy. Measuring total hCG and not intact hCG (hCG dimer only) is essential for the proper management of gestational trophoblastic disease cases. As a general rule, invasive mole, gestational trophoblastic neoplasm, and choriocarcinoma cases primarily produce hyperglycosylated hCG (hCG-H). In these cases, whether invasion through the uterus (as occurs in pregnancy) or invasion throughout the body (as occurs in these malignancies), hCG-H is the driving force of invasion (see Chapter 13). It is essential that the total hCG test used measures hCG-H (Table 29.1). Very often during clearance, hCG becomes dissociated as free subunits and degrades. As such measurement of free -subunit, nicked hCG, and nicked hCG missing the -subunit C-terminal peptide (CTP) are also important. As shown in Table 29.1, hCG assays vary greatly in their abilities to detect hCG-H, free -subunit, nicked hCG, nicked hCG missing the CTP in serum, and -core fragment in urine. Some commercial assays do not detect these hCG variants. Because of this variation in detecting gestational trophoblast disease and cancer variants, the literature of all commercial assays claims that they should only be used for normal pregnancy applications. As shown in Table 29.1, only the Siemens Immulite, the Siemens Stratus, and the hCG radioimmunoassay detect all pertinent variants. These are the only tests that should be used for monitoring patients with gestational trophoblastic disease. When managing cases of gestational trophoblastic disease, physicians should be aware of what assay their hospital or clinical chemistry center uses. If it is not the Siemens Immulite, Siemens Stratus, or an hCG radioimmunoassay, they need to
304
Human Chorionic Gonadotropin (hCG)
Table 29.1 Blind Evaluation of Commercial hCG Test with hCG-H Standard, 465 mIU/ml; Nicked hCG Standard, 3926 mIU/ml; Free -Subunit Standard, 3840 mIU/ml Molar Equivalents of hCG; hCG Missing -Subunit C-Terminal Peptide (hCG Minus CTP), 260 mIU/ml Molar Equivalents of hCG; and -Core Fragment Standard, 1092 mIU/ml Molar Equivalents of hCG. Popular Tests
hCG-H mIU/ml, %
Nicked hCG mIU/ml, %
hCG Free mIU/ml, %
hCG Minus CTP mIU/ml, %
-Core Fragment mIU/ml, %
Abbott AxSym Beckman Access Ortho Vitros ECi Roche Elecsys Siemens Stratus Siemens ACS180 Siemens Centaur Siemens Dimension Siemens Immulite Tosoh A1A600 Serono MAIAclone Wako hCG-CTP hCG RIA
400, (86%)
4071, (104%)
3278, (85%)
2, (1%)
2, (1%)
630, (135%)
4042, (103%)
5841, (152%)
2, (1%)
2, (1%)
370, (80%)
2780, (71%)
4445, (116%)
2, (1%)
2, (1%)
770, (165%) 385, (83%)
2736, (70%) 3691, (94%)
3220, (84%) 3744, (98%)
2, (1%) 190, (73%)
2, (1%) 2, (1%)
474, (102%)
3097, (79%)
5235, (136%)
2, (1%)
2, (1%)
500, (108%)
3048, (78%)
5535, (144%)
2, (1%)
2, (1%)
232, (50%)
3658, (93%)
3300, (86%)
2, (1%)
2, (1%)
467, (100%)
4240, (108%)
3843, (100%)
273, (105%)
273, (25%)
449, (97%)
3827, (97%)
3187, (83%)
2, (1%)
2, (1%)
17, (3.6%)
908, (23%)
123, (3.2%)
147, (54%)
87, (8%)
249, (64%) 406, (87%)
4007, (102%) 3280, (77%)
1625, (42%) 2767, (73%)
2, (1%) 182, (70%)
2, (2%) 170, (16%)
Percentages are proportions of standard concentration [2,3]. RIA is radioimmunoassay.
request that the patient’s serum and urine samples be managed by an outside laboratory that specifically uses these appropriate tests. Figures 29.4 and 29.5 are representative examples of gestational trophoblastic disease cases monitored using an inappropriate assay. The USA hCG Reference Service was contacted because these patients’ hCG results made no sense. In the case shown in Figure 29.4, the Serono MAIAclone assay did not appropriately detect hCG-H, which resulted in misleading results. In the case shown in Figure 29.5, the Siemens ACS180 assay (like multiple other assays) did not detect hCG missing the CTP. As such, it missed multiple recurrences of disease. hCG is a perfect marker for gestational trophoblastic diseases because it has 100% sensitivity and 100% specificity—but only if care is taken to select an appropriate assay that adequately recognizes all forms of the molecule.
hCG in Monitoring Gestational Trophoblastic Diseases
305
Figure 29.4 A sample choriocarcinoma case monitored by the USA hCG Reference Service. The referring center used the Serono MAIAclone assay, which does not detect hCG-H (Table 29.1), whereas the Reference Service used the Siemens Immulite, which equally detects hCG and hCG-H (Table 29.1).
Figure 29.5 A sample choriocarcinoma case monitored by the USA hCG Reference Service. The referring center used the Siemens ACS180 assay, which does not detect hCG missing the -subunit C-terminal peptide (hCG-CTP)( Table 29.1), whereas the Reference Service used the Siemens Immulite, which equally detects hCG and hCG-CTP (Table 29.1).
306
Human Chorionic Gonadotropin (hCG)
References [1] Seckl MJ. Investigation and treatment of patients with persistent gestational trophoblastic disease and gestational trophoblastic tumors/neoplasia in the United Kingdom. In: Hancock BW, Seckl MJ, Berkowitz RS, Cole LA, editors. Gestational trophoblastic disease, vol 3; 2010. p. 335–65. http://www.isstd.org/isstd/chapter12.html. [2] Cole LA, Shahabi S, Butler S, Mitchell H, Newlands ES, Behrman HR, et al. Utility of commonly used commercial hCG immunoassays in the diagnosis and management of trophoblastic diseases. Clin Chem 2001;47:308–15. [3] Cole LA, Sutton JM, Higgins Higgins TN, Cembrowski GS. Between-method variation in hCG test results. Clin Chem 2004;50:874–82.
30 Use of Hyperglycosylated hCG as a Unique Marker of Gestational Trophoblastic Neoplasms Laurence A. Cole and Carolyn Y. Muller USA hCG Reference Service, Albuquerque, NM, USA
In many ways, gestational trophoblastic disease is governed and regulated by the presence and predominance of hyperglycosylated hCG (hCG-H). As described in the evolution chapter at the end of this book (Chapter 36), it is the presence of unregulated hCG-H in humans that drives gestational trophoblastic neoplasms and causes people to get these human-specific diseases. As discussed in other chapters of this book (e.g., Chapter 13), hCG-H drives pregnancy implantation through regulatory mechanisms involving blockage of apoptosis, metalloproteinases, and collagenases. It is these same processes that drive invasion in gestational trophoblastic neoplasms including choriocarcinoma. After a woman has a normal childbirth, a few deeply implanted root cytotrophoblast cells remaining in the uterus lose their ability to fuse into syncytiotrophoblast cells. These cytotrophoblast cells are then transformed into permanently invasive choriocarcinoma cells which can grow very rapidly. Within as little as a 6-week period, that same woman can end up in the emergency room with trouble breathing (lung metastases) and seizures (brain metastases) due to choriocarcinoma or gestational trophoblastic neoplasm (choriocarcinoma without histology). This disease is the most malignant disease known to humans. It appears to be totally driven by hCG-H produced by cytotrophoblast cells and the pregnancy implantation-like invasion that it drives [1]. If the woman is immediately treated with chemotherapy, this fast-growing disease can respond well and regress completely, with a 90% 5-year survival rate [1]. In this chapter, we review the potential uses of hCG-H tests in gestational tropho blastic disease management. We examine an inactive form of disease called quiescent gestational trophoblastic disease. This is a relatively common inactive form of gestational trophoblastic disease and is seen all over the world. It is possibly the most common explanation for persistent low levels of hCG outside of pregnancy. We also examine a widespread variant of gestational trophoblastic neoplasm and chorio carcinoma called minimally aggressive gestational trophoblastic neoplasm. New treatment protocols are proposed for this chemorefractory disorder. We then look at hCG-H, the promoter of trophoblast disease invasion, as a new alternative marker to grade gestational trophoblastic diseases as opposed to the present World Health Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00030-X © 2010 Elsevier Inc. All rights reserved.
308
Human Chorionic Gonadotropin (hCG)
Organization (WHO) grading system. We also note how hCG-H determines which cases of gestational trophoblastic disease will respond to chemotherapy.
30.1 Quiescent Gestational Trophoblastic Disease The USA hCG Reference Service has now consulted on 134 cases of quiescent gestational trophoblastic disease. In the Reference Service’s experience, it is probably the most common cause for persistent low hCG levels outside of pregnancy in a woman of menstrual age. Each year, the USA hCG Reference Service is referred approximately 90 cases of persistently low hCG levels with nonrising hCG levels outside of pregnancy. Approximately 35 are pituitary hCG in menopausal women, 5 are cancer, 5 are false positive hCG (very rare today), and 45 are quiescent gestational trophoblastic disease. As shown in Figure 30.1, 62 of these 134 cases of quiescent gestational trophoblastic disease cases followed the evacuation of complete hydatidiform mole. After evacuation, hCG levels never diminished to zero, and subsequent chemotherapy did not help. In the 32 cases that followed spontaneous abortion of pregnancy, the aborted pregnancy was probably a hydatidiform mole. Eleven cases followed chemotherapy treatment for gestational trophoblastic neoplasm. The nine cases that occurred follo wing methotrexate termination of an ectopic pregnancy were likely ectopic hydatidiform moles. Eight cases followed evacuation of a partial mole. Seven cases were choriocarcinoma cases treated with chemotherapy. Five cases occurred after therapeutic evacuation of gestation, which were also probably a hydatidiform mole. Quiescent gestational trophoblastic disease can be diagnosed after at least 3 consecutive months of persistent low hCG levels with minimal oscillations in the levels (no more than twofold variation in a month, levels ranging from 2 average to 0.5 average) [1–4]. Two cases that we have consulted received a transabdominal hysterectomy. Examining the histology, a very small amount of highly differentiated syncytiotrophoblast tissue was observed [3]. Quiescent gestational trophoblastic disease seemingly represents highly differentiated syncytiotrophoblast cells with no significant cytotrophoblast cells. Because there are no significant cytotrophoblast cells, there is no detectable hCG-H, or no invasion promoter [1–5]. To date, we have observed cases with hCG ranging from 1.1 to 207 mIU/ml. Higher levels have not yet been observed. The median hCG level in these cases was 16.05 mIU/ml (Figure 30.1). In every case, quiescent gestational trophoblastic disease was marked by the absence of hCG-H. If your center does not have access to the hCG-H test to diagnose quiescent disease, it is still easy to identify. Three months of constant low hCG levels (1.1– 207 mIU/ml) with no more than twofold oscillation in hCG results can be observed in all cases of quiescent disease. In addition, patients have to have had a recent gestation within the past year, whether the origin of the disease was a failed pregnancy or gestational trophoblastic disease. Quiescent gestational trophoblastic disease has no hCG-H and no rapidly growing cells. We have never observed a case of quiescent gestational trophoblastic disease
Use of Hyperglycosylated hCG as a Unique Marker of Gestational Trophoblastic Neoplasms
309
Figure 30.1 Quiescent gestational trophoblastic disease in 134 individuals following history of hydatidiform mole, spontaneous abortion, ectopic pregnancy, and gestational trophoblastic neoplasm.
that responds to any kind of chemotherapy. In the USA hCG Reference Service’s experience, quiescent gestational trophoblastic disease disappears within 6 months in most cases. We have observed rare exceptions where prolonged low hCG levels lasted as long as 9 years. In cases with a history of hydatidiform mole or spontaneously aborted pregnancy, 10% of cases led to invasive trophoblastic disease, with hCG levels doubling, then doubling again over a period of time. In cases with a history of gestational trophoblastic neoplasm or choriocarcinoma, the incidence of recurrence of disease is notable higher, 25%. The USA hCG Reference Service first identified quiescent gestational trophoblastic disease in patients with no hCG-H or no invasion promoter. Their findings have now been confirmed in England and by Yale University [6,7]. If quiescent gestational trophoblastic disease is identified, hCG levels should be followed closely over the ensuing months until hCG disappears. The patient must take an oral contraceptive pill to ensure avoidance of pregnancy. The hCG-H produced by the pregnancy might activate the latent gestational trophoblastic neoplasm. In the USA hCG Reference Service’s experience, 131 of 134 cases had no detectable hCG-H. In three cases, the proportion of hCG-H was less than 10% of total hCG. In quiescent gestational trophoblastic disease cases, it is notable that hCG is 1.1– 207 mIU/ml instead of zero. This is a result of the death of most cytotrophoblast cells; the remnants of a tiny amount of syncytiotrophoblast cells produce only hCG. In
310
Human Chorionic Gonadotropin (hCG)
many cases, the tiny amount of cells might reside outside of the uterus so that a hysterectomy is not helpful and, in our experience, does not resolve this inactive disease.
30.2 M inimally Aggressive Gestational Trophoblastic Neoplasm It has been established that hCG-H is an autocrine molecule produced by cytotrophoblast cells. hCG-H is the promoter of invasion, growth, and malignancy in gestational trophoblastic diseases [4,8–15]. Invasion and malignancy occurs in the presence of hCG-H [4,8–15]. At implantation of pregnancy, for example, hCG-H is the molecular stimulus that drives invasion of the placenta [3,14,16]. In choriocarcinoma and gestational trophoblastic neoplasms, hCG-H production occurs and leads to mass invasion or mass malignancy. As described earlier, the absence of hCG-H (1% hCG-H) in choriocarcinoma or gestational trophoblastic neoplasm is indicative of quiescent gestational trophoblastic disease or inactive disease. In contrast, the presence of high proportions of hCG-H of total hCG (40% hCG-H) seemingly demonstrates a highly aggressive gestational trophoblastic neoplasm [3,6,7,10,15]. Here we consider borderline cases; malignancies with low proportions of hCG-H—a condition between highly invasive disease and quiescent disease, which we call minimally aggressive gestational trophoblastic disease. Patients seldom die from highly aggressive gestational trophoblastic disease. Most fatalities occur as a direct complication of the prolonged use of aggressive chemotherapy in resistant or multiagent chemorefractory cases [17–19]. An example of an extreme chemorefractory case referred to the USA hCG Reference Service is shown in Figure 30.2. One chemotherapy regimen followed another due to lack of hCG resolution. In total, 14 different regimens were given, each failing to completely diminish the hCG. The USA hCG Reference Service has consulted on three unfortunate cases of chemorefractory choriocarcinoma. Each of the women in these cases ultimately died from complications of the treatment. The first and second cases were unsuccessfully treated with 9 and 11 sequential chemotherapy regimens, respectively. In both of these cases, total hCG never rose above 1000 mIU/ml throughout the multiple regimens of chemotherapy. Both women expired from pulmonary fibrosis complications of bleomycin used in varying combination regimens. In the blood samples sent to the USA hCG Reference Service throughout the courses of the diseases, it was noted that the percent hCG-H was borderline positive (3.5% and 6%). This indicated very borderline aggressive disease, lying between quiescent or inactive (1%) and highly invasive (30%) disease. In the third case, the patient was unsuccessfully treated with a different sequence of seven chemotherapy regimens and one surgical procedure (Figure 30.3). During this time, the patient’s total hCG never rose above 1000 mIU/ml (Figure 30.3). The patient turned to God, and in her extreme faith refused further therapy. Over 6 months, her hCG rose to 930,000 mIU/ml. She then went to the Emergency Room
Use of Hyperglycosylated hCG as a Unique Marker of Gestational Trophoblastic Neoplasms
311
Figure 30.2 Example of chemorefractory minimally invasive gestational trophoblastic neoplasm, case 447 (Table 30.1). Patient received 14 chemotherapy regimens over 4 years. Abbreviations are those used in Table 30.1.
with seizures and trouble breathing. Brain, liver, and lung metastases of choriocarcinoma were found. Her general oncologists chose to use Xeloda to treat her very widespread disease, a rather obscure choice for treating highly advanced choriocarcinoma. It failed to rapidly abate the aggressively growing malignancy and the patient died. Table 30.1 shows the USA hCG Reference Service’s experience with 35 cases of aggressive disease and 31 cases of what we now call slow-growing or minimally aggressive disease (hCG-H 40%). The two categories are like two different diseases; the differences are outstanding. In the aggressive disease group, the average hCG-H proportion was 75 11%. Out of 35 cases of aggressive disease, 34 were successfully treated with a single regimen of chemotherapy. In the minimally aggressive group, the average hCG-H was 14 11%, and none were successfully treated by a single regimen of chemotherapy. Seemingly, the minimally aggressive group has a very slowgrowing and very slow-spreading disease, with hCG levels doubling every 2–6 weeks. It is clearly too slow-growing to respond to chemotherapy, which is likely the reason why all these cases are chemorefractory. Clearly, the proportion of hCG-H should be higher in minimally aggressive patients before chemotherapy will work. Is it not,
312
Human Chorionic Gonadotropin (hCG)
Figure 30.3 Minimally invasive gestational trophoblastic neoplasm case refusing further chemotherapy; agreed to permit monthly monitoring of hCG and hCG-H at USA hCG Reference Service. This is case 198 in Table 30.1. Abbreviations are those used in Table 30.1.
however, obvious to the oncologists and gynecologic oncologists treating these cases; chemotherapy will not work if there is very slow-growing disease. After enduring year after year of ineffective chemotherapy for suspected minimally aggressive disease, three women chose to halt further therapy but agreed to provide monthly blood samples during their rejection of therapy. The results of these three patients are shown in Figures 30.3–30.5. In the first case (Figure 30.3), the proportion of hCG-H started at 19%. As the disease slowly spread, total hCG climbed to around 1000 mIU/ml, and the proportion of hCG-H climbed from 19% to 26% to 33%. The disease spread within months, and when the hCG was close to 10,000 mIU/ml, the hCG-H was 48%, which is the level corresponding to aggressive disease when chemotherapy response would be appropriate. At this point in time, this patient continued to refuse chemotherapy because she was asymptomatic and off chemotherapy; however, her hCG rose to around 100,000 mIU/ml (64% hCG-H) and almost reached 1 million mIU/ml; the proportion of hCG-H now was 100% of total hCG. At this late point in her advanced disease, this patient was unsuccessfully treated. The second case (Figure 30.4) also came from a patient with a history of ineffective chemotherapy response who refused further therapy. Her hCG-H rose from
Table 30.1 USA hCG Reference Service Early (2000 IU/l) Gestational Trophoblastic Disease Cases. Code
Age
Total hCG (IU/l)
hCG-H (%)
Ineffective Chemotherapy
History
Notes
A. Aggressive disease, hCG-H 40%, 35 cases 92 140 186 198 211 214 258 259 271 298 302 313 314 347 359 371 395 408 409 417 420 424
53 37 22 20 23 29 35 21 42 31 34 35 35 36 26 39 31 34 31 30 34 29
863 4.7 31.9 52.5 1993 1504 330 670 542 305 98.6 341 600 521 38.1 1929 639 901 60 143 36 454
100 100 42 61 94 100 83 100 69 43 100 49 81 100 61 100 100 60 100 62 58 47
Choriocarcinoma Choriocarcinoma Partial mole Choriocarcinoma Complete mole Choriocarcinoma GTN Complete mole Choriocarcinoma GTN Choriocarcinoma GTN Choriocarcinoma GTN Choriocarcinoma Complete mole GTN Complete mole Complete mole Choriocarcinoma Complete mole Partial mole (Continued)
Table 30.1 (Continued) Code
Age
Total hCG (IU/l)
hCG-H (%)
441 492 493 494 495 521 522 542 544 545 552 557 Average
28 29 43 27 26 39 31 33 N/A 37 20 34
1926 214 1208 220 415 1113 18.1 869 11.1 9.03 821 48 335
60 39 60 95 100 85 41 46 100 100 74 100 75 11
Ineffective Chemotherapy
Mtx, ActD
History
Notes
GTN Choriocarcinoma Partial mole GTN Complete mole Complete mole Complete mole Complete mole Complete mole Choriocarcinoma Choriocarcinoma GTN
34 of 35 cases effective
B. Minimally aggressive disease, hCG-H 40%, 31 cases 198
20
53
19
200 204 267 271
21 24 27 35
113 23 42 600
26 15 35 5.0
301
40
42.2
5.7
Mtx, ActD, EMA-CO, EMA-EP, Taxol-Ifosamide, BEP, ICE, TAH, BSO, Xeloda ActD, EMA-CO, EMA-EP EMA-CO, Taxol, BEP, Xeloda Mtx, ActD, EMA-CO, EMA-EP EMA-CO, EMA-EP, Taxol-Ifosamide, Cisplatin, BEP, Xeloda, ICE Mtx, ActD, EMA-CO, EMA-EP, Taxol-Ifosamide, BEP, Xeloda
Complete mole Complete mole GTN GTN Choriocarcinoma Complete mole
See Figure 30.3
370 380 395 402 423 431 436 447
26 46 31 36 30 38 24 38
1220 238 639 331 28.4 663 725 18
1.2 9.2 20 2.5 11 4.1 0.38 18
459
24
20
13
464 475 481 483 490 491 496 516 551 553 554 568 572 574 580 585 Average
34 24 31 27 50 34 27 46 37 26 33 29 38 19 34 23
320 147 18.2 445 1596 2362 60 71 740 270 26 208 824 97 150 31 223
27 16 22 38 27 3.9 2.5 1.4 12 6.4 16 0.17 5.1 4.7 1 8.1 13 11
Mtx, TAH, ActD TAH, ActD, Etoposide Mtx, ActD, EMA-CO, EMA-EP Mtx Mtx Mtx, ActD, EMA-CO, EMA-EP, Taxol, BEP, ICE Mtx, ActD Mtx, ActD, EMA-CO, Taxol, Carboplatin, Cytoxan, BEP, Cisplatin, Genctabine, Doxil-Bleomycin, Avastin, Xeloda ActD, EMA-CO, EMA-EP, Taxol-Ifosamide, ICE, Xeloda, TAH-BSO Mtx, ActD TAH EMA-CO EMA-EP Taxol Mtx Mtx-Folinic acid Taxol-Ifosamide, Xeloda Cytoxan Mtx Mtx, ActD, EMA-CO, EMA-EP, Taxol-Ifosamide, ICE, BEP Mtx EMA-CO, EMA-EP Taxol-Ifosamide, BEP, Xeloda Mtx Mtx, ActD Mtx, Mtx Mtx, TVH, ActD EMA-EP, BEP, Carboplatin Mtx, ActD, EMA-CO, TAH Mtx, Mtx 0 of 31 cases effective
Complete mole Complete mole Choriocarcinoma GTN Partial mole Choriocarcinoma GTN Choriocarcinoma
See Figure 30.2
Choriocarcinoma
See Figure 30.4
Complete mole Choriocarcinoma GTN SAB Choriocarcinoma Choriocarcinoma Complete mole Choriocarcinoma Complete mole GTN Complete mole Complete mole Complete mole GTN Complete mole Complete mole
See Figure 30.9
See Figure 30.8 See Figure 30.7 See Figure 30.5 See Figure 30.6
GTN is gestational trophoblastic neoplasm. Mtx is methotrexate; ActD is actinomycin D; TAH is transabdominal hysterectomy; EMA-CO and EMA-EP are weekly etoposide, methotrexate, and actinomycin alternating with cyclophosphamide and onvocin (vincristine) or etoposide and cisplatin; TVH is transvaginal hysterectomy; BSO is bilateral salphingo-oophorectomy; SAB is spontaneous abortion history.
316
Human Chorionic Gonadotropin (hCG)
Figure 30.4 Minimally invasive gestational trophoblastic neoplasm case refusing further chemotherapy. This is case 459 in Table 30.1. Abbreviations are those used in Table 30.1.
13% to 19% to 24% at 3000 mIU/ml, and then rose to 41% to 64% as hCG levels rose higher. Her total hCG rose to over 100,000 mIU/ml. At that time, she had a seizure from brain metastases and dyspnea from lung metastases and agreed to consider more treatment. She was treated very successfully with EMA-CO chemotherapy (etoposide, methotrexate, and actinomycin D cycled weekly with cyclophosphamide and onvocin [vincristine]). The third case (Figure 30.5) was a similar patient who declined further treatment after several courses of failed chemotherapy. Her hCG-H rose from 2.5% to 28% to 48% at 12,000 mIU/ml. She was having symptoms of lung metastases when she saw her gynecologic oncologist. She responded well to a repeat course of EMA-EP chemotherapy (etoposide, methotrexate, and actinomycin D cycled weekly with etoposide and cisplatin), a regimen that previously failed when hCG-H was in the range consistent with minimally aggressive disease. Although these data reflect just three cases, the temporal data recovered were extremely helpful in supporting the notion that hCG-H is not only a critical biomarker of trophoblastic disease status but also a predictor of chemotherapy responsiveness. The experiences with these patients suggest that allowing some level of disease progression will ultimately be more successful when the treatment is given in
Use of Hyperglycosylated hCG as a Unique Marker of Gestational Trophoblastic Neoplasms
317
Figure 30.5 Minimally invasive gestational trophoblastic neoplasm case. This is case 496 in Table 30.1. Abbreviations are those used in Table 30.1.
an active tumor growth phase. As a practitioner, this is often counterintuitive to standard clinical practice and difficult to do as a result. These three cases suggest that minimally aggressive trophoblastic disease (H-hCG 40%) will not be eradicated effectively with chemotherapy; however, previously resistant regimens (used during a minimally aggressive disease phase) can be effective at eradicating the active aggressive disease. As disease advances, hCG-H rises and leads to aggressive disease, which responds appropriately to chemotherapy. In these three cases, the patients’ refusal of treatment allowed disease progression to advance significantly, with total hCG measuring 10,000, 100,000, and 1,000,000 mIU/ml before starting therapy. It appeared that when disease advanced to around 3000 mIU/ml, hCG-H was close to 40% (our definition of aggressive disease). As shown by Charing Cross Hospital in London, on average, once hCG levels reach 2000 mIU/ml, metastases might be identified radiographically [20]. If we planned a clinical research protocol to follow individuals to 3000 mIU/ml, liver and lung metastases might be evident and might present danger. An important clinical trial, such as this, needs to be done in order to validate these findings. In such a trial, minimally aggressive cases would be allowed to advance to 3000 mIU/ml and the
318
Human Chorionic Gonadotropin (hCG)
Figure 30.6 Case testing let rise to 3000 mIU/ml protocol. This is case 551 in Table 30.1. Abbreviations are those used in Table 30.1.
proven EMA-CO or EMA-EP chemotherapy regimens would follow. Patients were not resistant to these previously failed regimens; they had a form of disease that produced insufficient hCG-H (seemingly the invasion signal) that did not respond to chemotherapy appropriately. Our protocol of waiting until total hCG advances to approximately 3000 mIU/ml before starting chemotherapy was successful in 20 of 23 cases of slow-growing, minimally invasive disease (minimally aggressive disease cases, hCG-H 40% of total hCG, or total hCG taking 1 week to double). A multiagent chemotherapy regimen (either EMA-CO or EMA-EP), totally ablated all disease in 20 of 20 women. The success rate can be expanded to 22 of 22 cases if the two cases initially refusing chemotherapy are also included. Examples of successful cases are shown in detail in Figures 30.6–30.9. Why 20 of 23? The protocol failed in three cases. Three cases developed lung nodules (as shown by MRI) when total hCG was in the 800–2000 mIU/ml range. It is unknown whether the nodules truly were choriocarcinoma nodules or scar tissue, but the patients were advised to restart chemotherapy by their oncologists. Two of the three failed cases did not respond appropriately to EMA-CO or EMA-EP after departing the trial. These cases had returned to the chemorefractory scenario and are still receiving therapy today. One case responded fully to EMA-CO. We regret the
Use of Hyperglycosylated hCG as a Unique Marker of Gestational Trophoblastic Neoplasms
319
Figure 30.7 Case testing let rise to 3000 mIU/ml protocol. This is case 490 in Table 30.1. Abbreviations are those used in Table 30.1.
inability to biochemically monitor these patients over the course of their continued chemorefractory disease. It is concluded that patients with a low percent of hCG-H (40% of total hCG) are what we call chemoresistant, chemorefractory, or minimally aggressive disease cases. These types of cases need to be placed on a protocol of waiting until hCG is 3000 mIU/ml before treating with chemotherapy. According to Table 30.1, it appears these cases are primarily recurrent mole, choriocarcinoma, and gestational trophoblastic neoplasm cases. Moving against standard practice and allowing disease progression to occur before recommending treatment is a paradigm shift in terms of invasive cancer cases; however, the findings presented here show that chemotherapy appeared to be ineffective and led to toxicity rather than tumor reduction (as measured by a decline in total hCG) when given to patients with relatively low levels of hCG associated with a low percentage of hCG-H. Because of this persistent finding, a new management strategy is advocated for such situations. In patients with trophoblastic neoplasia (with low levels of hCG-H and usually low levels of total hCG), no active therapy should be undertaken until the total hCG becomes elevated to 3000 mIU/ml and the percentage of hCG-H
320
Human Chorionic Gonadotropin (hCG)
Figure 30.8 Case testing let rise to 3000 mIU/ml protocol. This is case 483 in Table 30.1. Abbreviations are those used in Table 30.1.
rises to 40% of total hCG. This report describes 20 cases successfully managed in this way. Clearly, the precise limits for initiating therapy in such situations require research. If a center cannot measure hCG-H, how can one identify minimally aggressive gestational trophoblastic disease? It can be easily determined by multiple parameters. First, by slowly rising hCG levels, with hCG doubling roughly every 2–6 weeks; second, by finding of a patient with gestational trophoblastic disease that is chemorefractory. Is allowing cases to advance to near 3000 mIU/ml total hCG a high-risk approach? This is the greatest fear of the practicing clinician. In choriocarcinoma, gestational trophoblastic neoplasm, and invasive mole cases, a tumor is rarely seen by CT or MRI prior to the tumor attaining significant size correlating to an hCG 2000 mIU/ ml [21]. As such, allowing the tumor to increase production of hCG to 3000 mIU/ ml is strictly allowing malignancy to advance to an imageable state. Will this have untoward consequences for the patient? Further experience with this strategy will help resolve this issue; however, sequential treatment of disease in the refractory
Use of Hyperglycosylated hCG as a Unique Marker of Gestational Trophoblastic Neoplasms
321
Figure 30.9 Case testing let rise to 3000 mIU/ml protocol. This is case 464 in Table 30.1. Abbreviations are those used in Table 30.1.
period can have fatal consequences, as is demonstrated by the three cases described in this article. We believe that treating women with invasive mole or choriocarcinoma and a low total hCG concentration can have untoward consequences. Yet waiting until the total hCG is around 3000 mIU/ml (or the percent hCG-H crosses the 40% threshold) allows a much more effective response to treatment. We know of three women who, after a prolonged ineffective treatment of choriocarcinoma, refused additional chemotherapy until they became symptomatic with seizures and respiratory distress. At the time of hospitalization for these cases, the total hCG reached 22,000, 130,000, and 932,000 mIU/ml, respectively. Two of the women were then successfully treated by reinitiation of multiagent chemotherapy. We certainly do not advocate waiting until clinical symptoms occur, but these cases do illustrate the evolution of chemorefractory, minimally invasive disease to chemoresponsive highly invasive disease. Cancer is a dynamic disease. Trophoblastic disease is perhaps the best human cancer model. It allows biochemical assessment of the changing invasiveness of disease. We infer that hCG-H is an exceptional tumor marker because it is the actual promoter
322
Human Chorionic Gonadotropin (hCG)
signal of tumor invasion [4,8–15]. The percent of hCG-H in total hCG marks invasiveness (or absence thereof) of a tumor, thus defining quiescent or noninvasive tumors, minimally invasive tumors, and highly invasive tumors as described here. It seems to mark a gradient of disease invasiveness between quiescent and highly invasive disease; this is bidirectionally dynamic, responding to pressures of treatment. It might be a much more accurate method of tumor classification (or risk of tumor) than FIGO 2000 risk scoring parameters (hCG concentration, interval since index pregnancy event, age, tumor size, and site of metastases). As shown in this chapter, hCG-H is this exceptional marker. It identifies minimally invasive or slow-growing malignancies, and warrants waiting for a tumor to advance (measured by a biochemical threshold) before chemotherapy will most likely be effective. If hCG-H is not available to identify minimally invasive gestational trophoblastic neoplasm, then cases can be judged by slow growth of malignancy (hCG doubling time 1 week). It is likely that the phenomenon of quiescent and minimally aggressive disease is a function of most, if not all, solid tumors. Patients with recurrent disease will frequently experience a partial clinical response after reinitiation of chemotherapy, then plateau into a “stable” disease state after treatment until the disease progresses by radiologic criteria (target lesion progression or development of new lesions) or biochemical parameters (doubling of respective nonspecific tumor markers). Continued chemotherapy given during this “stable” period is thought to be a success in maintenance of the disease which is otherwise incurable and likely to rapidly progress; however, patients who choose not to have treatment, or those who experience a “chemo holiday,” will frequently demonstrate similar periods of “stable disease” without receiving any treatment. This “stable” time frame likely represents an inherent quiescent to minimally invasive cancer biology that is part of the dynamic life of solid tumors, perhaps as a response to chemotherapeutic interventions. Therefore, chemotherapy given during this inherent biological quiescent time should be considered refractory treatment in parallel to what is seen in the cases presented here. Proving this paradigm in other solid tumors is difficult because there are no known sensitive biomarkers of invasiveness. The proportion of hCG-H of total hCG is a sensitive biomarker of invasiveness in trophoblastic disease and should serve as a model to help us better understand the dynamic changes in solid tumor biology. Although the data presented here are limited to three cases initially refusing chemotherapy and 20 cases successfully treated after reaching 3000 mIU/ml, the data are very much supported by the 35 aggressive disease cases presented, 34 of 35 successfully treated and 31 minimally aggressive cases presenting with untreatable chemorefractory disease. We believe that these findings are profoundly important in the treatment of gestational trophoblast diseases. Given the minor risks associated with this delayed therapy protocol in compliant patients (tumor is not visible until hCG 2000 mIU/ml) [20] and the potentially fatal outcomes of prolonged exposure to multiagent chemotherapy, we encourage the oncology community to adopt this new treatment paradigm. Further experience will continue to test this recommendation and might suggest consideration of a centralized registry to further explore the unique dynamics of this disease.
Use of Hyperglycosylated hCG as a Unique Marker of Gestational Trophoblastic Neoplasms
323
30.3 Gestational Trophoblastic Neoplasm Classification As described earlier, hCG-H, the physical promoter of gestational trophoblastic neoplasms [4,8,9], is an outstanding marker of disease. It allows cases to be divided aggressive or chemotherapy responsive, minimally aggressive or chemotherapy nonresponsive, and quiescent categories. Nothing like this can be done with the WHO characterization system. hCG-H proportions are a lot more meaningful than the current WHO categorization system believes; they say nothing about how tumors can be appropriately treated with chemotherapy. Through Focus Diagnostics, hCG-H is now being sold by Quest Diagnostic. All centers treating gestational trophoblastic diseases should adopt this assay.
References [1] Muller CY, Cole LA. The quagmire of hCG and hCG testing in gynecologic oncology. Gynecol Oncol 2009;112:663–72. [2] Cole LA. Use of hCG tests for evaluating trophoblastic diseases: choosing an appropriate hCG assay, false detection of hCG, unexplained elevated hCG, and quiescent trophoblastic disease. In: Hancock BW, Newland ES, Berkowitz RS, Cole LA, editors. Gestational trophoblastic disease. 2nd ed. London: Chapman and Hall; 2002. [3] Cole LA, Butler SA, Khanlian SA, Giddings A, Muller CY, Seckl MJ, et al. Gestational trophoblastic diseases: 2. Hyperglycosylated hCG as a reliable marker of active neo plasia. Gynecol Oncol 2006;102:150–8. [4] Cole LA, Butler SA. Hyperglycosylated hCG and its free -subunit, tumor markers and tumor promoters: a review. J Reprod Med 2008;53:499–510. [5] Cole LA, Khanlian SA, Giddings A, Butler SA, Muller CY, Hammond C, et al. Gestational trophoblastic diseases: 4. Presentation with persistent low positive human chorionic gonadotropin. Gynecol Oncol 2006;102:116:3–9. [6] Cole LA, Muller Y. hCG in the Management of Quiescent and Chemorefractory Gestational Trophoblastic Diseases. Gynecol Oncol 2010, in press. [7] Hancock BW, Tidy JA. Clinical management of persistent low-level hCG elevation. Trophobl Dis Update 2003;4:5–6. [8] Kohorn EI. Persistent low-level “real” human chorionic gonadotropin: a clinical cha llenge and a therapeutic dilemma. Gynecol Oncol 2002;85:315–20. [9] Cole LA, Dai D, Leslie KK, Butler SA, Kohorn EI. Gestational trophoblastic diseases: 1. Pathophysiology of hyperglycosylated hCG-regulated neoplasia. Gynecol Oncol 2006;102:144–9. [10] Handshuh K, Guibourdenche J, Tsatsaris V, Guesnon M, Laurendeau I, Evain-Brion D, et al. Human chorionic gonadotropin produced by the invasive trophoblast but not the villous trophoblast promotes cell invasion and is down-regulated by peroxisome proliferator-activated receptor-a. Endocrinol 2007;148:5011–19. [11] Khanlian SA, Smith HO, Cole LA. Persistent low levels of hCG: A pre-malignant gestational trophoblastic disease. Am J Obstet Gynecol 2003;188:1254–9. [12] Kelly LS, Birken S, Puett D. Determination of hyperglycosylated human chorionic gonadotropin produced by malignant gestational trophoblastic neoplasias and male germ
324
[13]
[14]
[15]
[16] [17]
[18]
[19]
[20]
[21]
Human Chorionic Gonadotropin (hCG)
cell tumors using a lectin-based immunoassay and surface plasmon resonance. Mol Cell Endorcinol 2007;260:33–9. Kovalevskaya G, Genbacev O, Fisher SJ, Caceres E, O’Connor JF. Trophoblast origin of free isoforms: cytotrophoblasts are the primary source of choriocarcinoma-like free. Mol Cell Endocrinol 2002;194:147–55. Hamade AL, Nakabayashi K, Sato A, Kiyoshi K, Takamatsu Y, Laoag-Fernandez JB, et al. Transfection of antisense chorionic gonadotropin gene into choriocarcinoma cells suppresses the cell proliferation and induces apoptosis. J Clin Endocrinol Metab 2005;90:4873–9. Cole LA, Khanlian SA, Riley JM, Butler SA. Hyperglycosylated hCG in gestational implantation and in choriocarcinoma and testicular germ cell malignancy tumorigenesis. J Reprod Med 2006;51:919–25. Sasaki Y, Ladner DG, Cole LA. Hyperglycosylated hCG the source of pregnancy failures. Fertil Steril 2008;89:1871–86. Cole LA, Khanlian SA, Kohorn EI. Evolution of the human brain, chorionic gonadotropin and hemochorial implantation of the placenta: insights into origins of pregnancy failures, preeclampsia and choriocarcinoma. J Reprod Med 2008;53:449–57. Goto S, Ino K, Mitsui T, Kikkawa F, Suzuki T, Nomura S, et al. Survival rates of patients with choriocarcinoma treated with chemotherapy without hysterectomy: effects of anticancer agents on subsequent births. Gynecol Oncol 2004;93:529–35. Turan T, Karacay O, Tulunay G, Boran N, Koc S, Bozok S, et al. Results with EMA/CO (etoposide, methotrexate, actinomycin D, cyclophosphamide, vincristine) chemotherapy in gestational trophoblastic neoplasia. Int J Gynecol Oncol 2006;16:1432–8. Berkowitz RS, Goldstein DR. Presentation and management of persistent gestational trophoblastic disease and gestational trophoblastic tumors in the USA. In: Hancock BW, Newland ES, Berkowitz RS, Cole LA, editors. Gestational trophoblastic disease. 2nd ed. London: Chapman and Hall; 2007. Newlands E. Presentation and management of persistent gestational trophoblastic disease (GTD) and gestational trophoblastic tumor (GTT) in the United Kingdom. In: Hancock BW, Newland ES, Berkowitz RS, Cole LA, editors. Gestational trophoblastic disease. 2nd ed. London: Chapman and Hall; 2002.
31 Pituitary hCG and Familial hCG Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
Numerous older publications from the 1960s, 1970s, and 1980s suggested bacteria, crabs, and other unlikely sources to explain detection of hCG outside of pregnancy [1–8]. We now have a better understanding of the various sources of hCG when a person is not pregnant. This is explained under background hCG in Chapter 25, as quiescent gestational trophoblastic disease in Chapter 30, as false positive hCG assays in Chapter 20, and here as pituitary and familial hCG. We now know that men and women can have positive hCG tests due to circulating heterophilic or interfering antibodies. These cause false positive hCG results and might explain some of the strange hCG tests [9–11]. We also now know that men and women with cancer normally produce the total hCG assay immunoreactive free -subunit of hCG [12–14]. Here we examine the accumulation of evidence that shows the pituitary gland normally producing hCG during the menstrual cycle and in menopause [15–18]. We also examine the data showing that some people might have an inherited or familial hCGproducing disorder [19]. It has now been 29 years since hCG production was first demonstrated to come from the pituitary gland [20]. Since then, over 40 publications have confirmed this observation and described how extremely low low levels of hCG (,2 mIU/ml) accompany luteinizing hormone (LH) during the menstrual cycle [15–18,20–29]. In amenorrhea or menopause, the ovary fails to provide estrogen to feed back to inhibit gonadotropin releasing hormone produced by the hypothalamus. This leads to gonadotropin releasing hormone constantly stimulating the pituitary, and to raised production of LH, follicle stimulating hormone, and to raised production of human chorionic gonadotropin (0 to 33.6 mIU/ml). As such, hCG is evident in all women with amenorrhea or menopause (Ages 48), and in women following bilateral salpingooophorectomy (BSO), in women during oligomenorrhea in perimenopause [15–18, 21,24–26,28]. In medical practice, a positive hCG test prior to menopause suggests pregnancy or gestational trophoblastic disease [23,24,27,28]. A positive hCG test in perimenopausal and menopausal women, however, suggest pituitary hCG. Today, testing for pregnancy is routinely performed before surgical procedures, prior to imaging procedures, prior to administration of certain medications that might harm a fetus, and with admission to emergency departments. When hCG is detected, all planned procedures, whether critical or noncritical, are halted pending investigation of pregnancy and the source of hCG, seemingly regardless of patient age or menopausal status. After excluding regular or ectopic pregnancies, most cases, regardless of age or menopausal status are referred to gynecologic oncologists. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00031-1 © 2010 Elsevier Inc. All rights reserved.
326
Human Chorionic Gonadotropin (hCG)
Numerous older obstetrics and gynecology textbooks state that the only source of hCG, if pregnancy and ectopic pregnancy have been eliminated, is gestational trophoblastic disease. Many physicians still adhere to these guidelines, which can lead to needless hysterectomies or unnecessary chemotherapy in menopausal patients naturally producing pituitary hCG. Studies show that very rarely it is a malignancy or gestational trophoblastic disease first diagnosed as a result of a positive hCG test [30–34]. In most cases it is pituitary hCG. Sixty million women are now in perimenopause or menopause in the United States. Considering the frequency of hCG testing prior to surgery, at the emergency room, prior to administration of certain drugs, and prior to X-rays, many of these 60 million must have been through the stress and alarm of a questionable pregnancy. Here we carefully consider all aspects of pituitary hCG, a key form of hCG present in menopausal women. We examine its structure, its biological function, and its occurrence.
31.1 Structure of Pituitary hCG Pituitary hCG has an identical amino acid structure to placental hCG [25]. It is unique, however, in having a variable portion of sulfated oligosaccharides [25]. The sulfate groups are attached to N-acetylgalactosamine residues, which replace galactose residues in N- and O-linked oligosaccharides (Figure 31.1). As published, the -subunit of pituitary hCG has 2.7 mole/mole oligosaccharides terminating in sulfate groups, and 4.6 mole/mole oligosaccharides terminating in sialic acid groups (37% sulfation) [25]. The -subunit of pituitary hCG has 0.8 mole/mole oligosaccharides terminating in sulfate and 1.7 mole/mole oligosaccharides terminating in sialic acid groups (32% sulfation).
31.2 Biological Function of Pituitary hCG The USA hCG Reference Service recently examined over 8300 urine samples from women with normal menstrual periods [18]. As found, low concentrations (1 mIU/ ml) of hCG were detectable at the time of the mid-cycle LH peak in 232 of 277 (84%) menstrual cycles [18]. The mean hCG level was 1.54 0.90 mIU/ml and the range was 1–9.2 mIU/ml. Considering the mean and the distribution of the range of hCG values, it is logical to assume that the 45 of 277 menstrual cycles in which hCG was not detected produced lower hCG concentrations (1 mIU/ml)—too low to be detected by the Siemens Immulite assay used. It was inferred that low concentrations of hCG supplement LH during all human menstrual cycles and are part of normal physiology. As shown early by Odell and Griffin [35,36], using an ultrasensitive sandwich assay to hCG, pituitary hCG is produced at very low levels (mean 0.01 mIU/ml) in men, with a wide range of 0.03–1.7 mIU/ml [35,36]. Pituitary hCG was detected in women in pulses of the luteal and follicular phases of the menstrual cycle which paralleled LH levels [35,36]. Injections of gonadotropin releasing
Pituitary hCG and Familial hCG
327
Figure 31.1 Sulfated oligosaccharide on pituitary hCG. (A) N-linked oligosaccharides, (B) O-linked oligosaccharides. The figure shows variants in N-linked oligosaccharides with GalNAc-sulfate and Gal-SA nonreducing terminals, and O-linked oligosaccharides with GalNAc-sulfate and Gal-SA nonreducing terminals. Man is mannose, GlcNAc is N-acetylglucosamine, GalNAc is N-acetylgalactosamine, Fuc is fucose, Gal is Galactose, and SA is sialic acid. A dotted line indicates variable presence.
hormone (GnRH) were shown to directly promote circulating pituitary hCG levels in men and women, just as it similarly promotes LH levels [35,36]. It is inferred that pituitary hCG supplements pituitary LH in men and women [18,25,36]. As yet, it is unknown whether there is a specific function for pituitary hCG during the menstrual cycle. Pituitary hCG could have functions separate from those of LH. But even if pituitary hCG has no specific function, there is a natural explanation for its production. There is a single LH -subunit gene buried among the seven or eight back-to-back hCG -subunit genes in human chromosomes [37]. hCG and LH share a single common -subunit. It is possible that a small amount of hCG is promoted by GnRH alongside specific LH -subunit stimulation in pituitary gonadotrope cells during normal menstrual cycle physiology in women, and in men. LH has multiple functions during the LH peak and ovulation period. We know that both LH and hCG act on the LH/hCG receptor to promote progesterone production by corpus luteum cells [26,29,38]. We assume that LH and hCG act on the same receptor on follicles. As established, with the appearance of an LH/hCG receptor on granulosa cells of the Graafian follicle or primary follicle, LH first promotes follicular growth [39,40], then stimulates diploid cell meiosis [41,42]. LH causes the follicle to form a stigma or protrusion [43] and then promotes collagenase production
328
Human Chorionic Gonadotropin (hCG)
to degrade and penetrate the stigma [43,44]. The penetration of the stigma causes bursting of the follicle (ovulation) to occur. LH then acts to differentiate the burst or ovulated follicle into a corpus luteum [45,46]. It is not clear whether hCG just incidentally assists LH in each of these five steps or has specific functions of its own in one or more of these steps. The mass of hCG in an individual human pituitary gland secretes 0.5–1.1 g/ gland; approximately 25- to 50-fold less than the mass of LH [29]. Publications show that pituitary hCG has approximately half the biological activity in promoting progesterone production of placental hCG [25]. In conclusion, pituitary hCG seems to parallel LH and supplements LH activity in both men and women.
31.3 Occurrence of Pituitary hCG As shown by the USA hCG Reference Service, all women produce 1.54 0.90 mIU/ ml (means standard deviation) of pituitary hCG (detected in urine) during the LH peak of the normal menstrual cycle [18]. Odell and Griffin indicate that the miniscule level of 0.01 mIU/ml (average) pituitary hCG is detectable in the serum of all men [35,36]. Similar miniscule levels of pituitary hCG are detected in serum during the luteal and follicular phase of the menstrual cycle [35,36]. Extremely low levels of hCG can be detected in healthy men and in women (during the menstrual cycle). These tiny concentrations are close to or below the detection limits of most laboratory hCG/pregnancy assays. Rarely do these tiny levels cause alarm regarding possible pregnancy or halt scheduled surgeries or necessary procedures. During menopause and perimenopause, the ovary partially or completely fails, thus reducing or halting production of progestagens and estrogens. Similarly, if a woman receives a BSO, she will halt production of progestagens and estrogens and be placed in a state of menopause. Progestagens and estrogens control GnRH pulses by the hypothalamus through feedback. In the absence or reduced presence of progestagens and estrogen, GnRH pulses go wild and become as frequent as possible. This leads to highly raised serum LH and follicle stimulating hormone (FSH) concentrations. Pituitary hCG is raised significantly as well. In menopausal women, the average pituitary hCG level in serum is 9.66 6.52 mIU/ml (mean standard deviation); the range is 1.3–33.6 mIU/ml. As defined, perimenopausal women are still fertile but have raised FSH levels (30 mIU/ml) [17] and oligomenorrhea (extremely variable menstrual periods) [15]. Menopausal woman have amenorrhea or no menstrual cycle. BSO and primary amenorrhea induces the onset of menopause at an early age. The USA hCG Reference Service’s experience is shown in Table 31.1. As shown, the mean serum concentration of pituitary hCG was 5.5 3.2 mIU in perimenopausal women (range 1.4–12 mIU/ml), 11.0 6.3 mIU/ml in menopausal women (range 2.1–32.9 mIU/ml), and 9.8 9.5 mIU/ml in women with BSO or primary amenorrhea (range 1.8–33.6 mIU/ml). A significant difference was observed between perimenopause and menopause cases (P 0.000064), and between perimenopause and BSO plus primary amenorrhea cases (P 0.029).
Pituitary hCG and Familial hCG
329
Table 31.1 The USA hCG Reference Service’s Experience with 103 Women Demonstrated to Have Pituitary hCG. Case
Age
hCG (mIU/ml)
hCG-H (mIU/ml)
History
38 39 39 39 40 40 40 40 40 40 42 42 43 43 44 44 44 44 46 46 46 46 47 Mean
2.0 2.0 5.8 2.8 6.0 7.6 10.1 12.0 4.4 11.6 8.3 7.2 8.4 3.8 3.2 4.2 9.2 2.2 3.0 3.3 3.7 5.0 1.4 5.5 3.2
0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 Range
1.4–12
42 42 43 43 44 45 47 48 48 48 48 49 49 49
20 12.9 16.1 12.7 6.4 7.2 10.6 12.2 4.1 7.7 6.8 2.1 19.2 7.7
0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05
A. Perimenopause 455 339 194 199 563 150 561 36 216 427 422 442 377 421 317 305 410 257 173 3 313 506 534 23 cases
B. Menopause 331 538 397 489 499 17 432 187 350 450 328 300 406 533
(Continued)
330
Human Chorionic Gonadotropin (hCG)
Table 31.1 (Continued) Case
Age
hCG (mIU/ml)
hCG-H (mIU/ml)
540 22 412 425 306 376 169 191 201 79 206 411 413 434 564 21 32 151 193 143 344 469 164 183 192 20 90 329 478 49 336 517 467 476 127 320 565 445 465 269 357 498
49 50 50 50 50 51 51 51 51 51 51 51 51 51 51 52 52 52 52 52 52 52 53 53 53 53 53 53 53 54 54 54 54 54 55 55 55 55 55 55 55 55
9.9 16.2 7.5 12.5 7.7 9.8 7.4 13.7 4.3 13.7 12.3 8.1 4.1 14.6 7.3 4.7 11.0 9.0 10.2 10.6 14.5 7.5 27.2 5.9 10.6 6.0 10.4 7.0 9.3 6.1 6.7 39.0 6.0 12.1 5.3 8.0 8.2 9.2 4.6 9.6 14.2 11.5
0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05
History
(Continued)
Pituitary hCG and Familial hCG
331
Table 31.1 (Continued) Case
Age
hCG (mIU/ml)
hCG-H (mIU/ml)
History
115 188 34 162 429 341 514 517 543 548 182 18 428 69 cases
56 57 57 59 59 60 61 62 66 68 69 70 54 Mean
7.3 7.9 14.9 15.6 16.1 32.9 28.5 15.2 17.1 2.5 4.5 31.5 16 11.4 7.1
0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 Range
2.1–39.0
0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 Range
BSO BSO BSO BSO BSO BSO Men Men Oop Oop Oop 1.8–33.6
C. BSO and primary amenorrhea 444 349 519 454 460 322 468 415 319 319 471 11 cases
34 35 35 36 37 40 29 42 28 28 34 Mean
3.2 11.5 2.3 3.9 6.9 1.8 6.1 19.4 12.7 33.6 6.4 9.8 9.5
BSO is bilateral salpingo-oophorectomy, Oop is bilateral oophorectomy, Men is amenorrhea. All woman had FSH 30 mIU/ml [17].
The USA hCG Reference Service consults on numerous pituitary hCG cases each year (Table 31.1). These cases seemingly mostly derive from routine hCG testing performed on women prior to surgery, X-ray procedures, and emergency room visits. It appears to the USA hCG Reference Service that these services wrongly conclude from a positive hCG test that these menopausal women are pregnant. Women are routinely tested for pregnancy by ultrasound, tested further for ectopic pregnancy by ultrasound and other methods, given methotrexate to eliminate ectopic pregnancy, and finally referred to an oncologist or gynecologic oncologist for cancer investigation. All too often these women receive an unnecessary hysterectomy or chemotherapy when hCG levels do not disappear [15,16].
332
Human Chorionic Gonadotropin (hCG)
Over 40 papers have described the normality of pituitary hCG in menopausal women, but it appears that physicians either have not read these publications or they cannot believe that a positive hCG test can mean anything but pregnancy. Time and again, after excluding cancer and choriocarcinoma, we conclude that a patient has pituitary hCG (Table 31.1). These are the kind of cases that get referred to the USA hCG Reference Service. We know of four cases in which menopausal women were waiting years for a renal transplant. Finally, a matching kidney came, and when they entered hospital for the transplant they were given the hCG test. It was positive. Even though it is not physically possible for them to be pregnant, their transplants were cancelled and the kidneys given to other patients all because of the positive hCG test. The USA hCG Reference Service has no specific assay for sulfated pituitary hCG because there is no known specific test for sulfated pituitary hCG; we simply exclude other possibilities like gestational trophoblastic disease and cancer. Hyperglycosylated hCG (hCG-H) does not appear to be produced by the pituitary (Table 31.1). There are, however, two other tests that are invaluable. The first test is the FSH test. As shown by Gronowski et al. [17], a raised serum FSH of 30 mIU/ml illustrates perimenopause and menopause, or a lack of sufficient steroid feedback to the hypothalamus. An FSH of 30 mIU/ml means that there is likely to be elevated pituitary hCG present. The second test involves treating patients with a high-estrogen oral contraceptive pill for 3 weeks, which suppresses pituitary hCG and proves whether the hCG is in fact from a pituitary origin [15,16,24,28]. These two methods make it easy to prove pituitary hCG is the source of circulating hCG. We emphasize again, however, that production of pituitary hCG in menopause is normal and natural, and does not mean pregnancy or cancer. Hospitals need to start using the FSH test alongside the hCG tests when evaluating women prior to surgery. If the FSH is 30 mIU/ml, they need to consider that a positive hCG result is due to pituitary hCG. All pregnancies will still be identified. Doing this will avoid numerous needless investigations into cancer and other explanations for elevated hCG.
31.4 Familial hCG Syndrome The USA hCG Reference Service has seen four cases that defied every criterion. These cases did not have cancer, gestational trophoblastic neoplasm, pituitary hCG, quiescent gestational trophoblastic disease, minimally invasive gestational trophoblastic disease, or false positive hCG. Their serum FSH levels were normal. It was only after excluding every normal possibility and not getting any meaningful results that the USA hCG Reference Service investigated the patients’ family and discovered a familial hCG syndrome, a seemingly rare genetic disorder. Medical investigation revealed no obvious clinical complaints. Both the male and female cases were fertile; women were having normal menstrual cycles. The subject in the first case had 41 mIU/ml hCG detected in serum and 3.5 mIU/ml in urine (Table 31.2). The patient was excluded from surgery because of an apparent positive pregnancy test. The patient realized that she could not possibly be pregnant and joined the false positive legal action that was ongoing at that time. They referred her to
Pituitary hCG and Familial hCG
333
the USA hCG Reference Service. The USA hCG Reference Service tested her and found that her hCG levels were real and not false positive. It was then revealed to us that her sister had the same problem. One week later, we tested her a second time and observed no obvious changes. We then tested serum and urine from her sister and her mother, both of which were similarly positive for hCG (Table 31.2). The concept of familial hCG syndrome, or inherited hCG production syndrome, was born. In this case, the subjects were asymptomatic, with no obvious disease or obscure condition. Because both sisters were young with positive hCG, a genetic disorder without symptoms was concluded. The subject in the second case was a male football player and sports star. He was disqualified from national sports as a result of a positive hCG spot test, which indicated anabolic abuse, because hCG promotes testosterone production. This athlete’s positive hCG level remained static over multiple months (Table 31.2). Interestingly, hCG was only detected in urine, which was the only fluid tested by the athletic testing service. The urine contained 6.7 mIU/ml; 6.9 mIU/ml when retested 2 months later (Table 31.2). The athlete profoundly stated to us and to the press that he never took hCG. The USA hCG Reference Service’s view of the case was that he had background hCG as opposed to rising and falling hCG as one would see in someone administering hCG to themselves. We tested his father and he was also positive for hCG in urine only. This was the second documented case of familial hCG syndrome. We defended his appeal to the national sport authority, and he was reinstated and rejoined the sports national leagues. Table 31.2 Concentration of Total hCG (mIU/ml) and of Molar hCG Equivalents (mIU/ml) of hCG-H, Free -Subunit (Free ), and -Core Fragment (-Core) in Familial hCG Serum and Urine Samples. Case
Age
Total hCG Serum (mIU/ml)
hCG-H Serum Molar (mIU/ml)
Free Serum Molar (IU/l)
Total hCG Urine Molar (mIU/ml)
Free Urine Molar (mIU/ml)
-Core Urine Molar (mIU/ml)
Case 1 Case 1 repeat Sister Mother Case 2 Case 2 repeat Father Case 3 Case 3 repeat Father Brother Case 4 Case 4 repeat Father
32 32 34 54 28 28 52 32
41 44 34.2 35.1 1 1 1 2.0 1.8 1.3 1.1 2.8 2.7 3.3
0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5
34 28 17 16 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 1.4
3.5 2.5 2.0 2.0 6.7 6.9 10 27 25 1.59 4.0 54 27 27
0.3 0.3 0.3 0.3 0.3 0.3 0.3 6.5 4.2 0.6 0.6 3.0 1.5 2.3
0.8 0.8 0.8 0.8 0.8 0.8 0.8 7.6 6.6 5.2 4.0 3.6 3.6 3.8
50 28 22 22 45
334
Human Chorionic Gonadotropin (hCG)
The USA hCG Reference Service had two additional cases that fit into the emerging story of familial hCG syndrome (Table 31.2). In each case, low hCG levels persisted and similar results were observed among family members. The hCG variant produced in familial hCG syndrome was a bit peculiar. Case 1 was marked by serum-free -subunit (Table 31.2), whereas cases 2, 3, and 4 were marked by urine total hCG, with little or no hCG detected in serum. It appears that these three cases were producing degraded hCG, which was being rapidly cleared from the circulation (Table 31.2). It was assumed that this inherited hCG was being produced by the pituitary gland; however, this is an assumption and can in no way be confirmed with current methods. We concluded that we found a new and rare genetic disorder characterized only by expression of low levels of hCG-related molecules. This is clearly another explanation of positive hCG in individuals not pregnant, not having gestational trophoblastic disease, and not having cancer. The only way to confirm this oddity is by showing consistency in hCG concentration and by performing hCG tests on parents and kin. Ulf Stenman, Ph.D. in Finland, asked if he could test the serum and urine from the familial hCG cases. Intriguingly, he derived very different results. He was observing a high proportion of free -subunit in every case and minimal levels of total hCG. We stared at the conflicting results with amazement. Then, after questioning the specificity of hCG and free -subunit assay used, the conflict was resolved. The USA hCG Reference Service uses the Siemens Immulite as total hCG assays. This assay recognizes hCG molecules missing the -subunit C-terminal peptide (hCG-CTP) and free -subunit missing the -subunit C-terminal peptide (free CTP). Stenman’s total hCG test did not detect these molecules. Our free -subunit assay (the FBT11 monoclonal antibody assay) required the C-terminal peptide to be present, whereas Stenman’s free -subunit assay detected molecules missing the C-terminal peptide. These familial cases were all producing free -CTP, which was detected as total hCG by our assay, and not as free -subunit in our free -subunit test. The free -CTP was detected as free -subunit by Stenman’s test. This was a new discovery: The principal type of hCG produced in familial hCG was free -CTP. Free -CTP is like the hCG variant produced by cancer cells (see Chapter 32) rather than the intact sulfated hCG produced by the pituitary. Cancer cells produce a free -subunit; macrophages and leukocyte elastase enzyme cleave the C-terminal peptide. We might wonder if the familial free -CTP is coming from a familial tumor. There is still much to learn in this area of study.
References [1] Acevedo HF, Slifkin M, Pouchet GR, Pardo M. Immunochemical localization of a choriogonadotropin-like protein in bacteria isolated from cancer patients. Cancer 1978; 41:1217–29. [2] Cohen H, Strampp A. Bacterial synthesis of substances similar to human chorionic gonadotropin. Proc Soc Exper Biol Med 1976;152:408–10.
Pituitary hCG and Familial hCG
335
[3] Dominigue G, Johson E. Isolation of subcellular fractions containing immunogenic endobacterial common antigen. Z Immunitatsforsh Exper Klinical Immunol 1974;148:23–8. [4] Backus BT, Affronti LF. Tumor-associated bacteria capable of producing human choriogonadotropin-like substance. Infect Immun 1981;32:1211–15. [5] Maruo T, Segal SJ, Koide SS. Studies on the apparent human chorionic gonadotropinlike factor in crab Ovalipes ocellatus. Endocrinol 1979;104:932–9. [6] Yoshimot Y, Wolfsen AR, Odell WD. Human chorionic gonadotropin-like substance in nonendocrine tissues of normal subjects. Science 1977;197:1977–81. [7] Braunstein GD, Kamdar V, Rasort J, Swaminathan N, Wade ME. Widespread distribution of chorionic gonadotropin-like substance in normal human tissues. Endocrinol (Suppl.) 1978;102:96. [8] Chen H-C, Hodgen GC, Matsuura S, Lin LJ, Gross E, Reichert LE, et al. Evidence for a gonadotropin from nonpregnant subjects that has physical immunological and biological similarities to human chorionic gonadotropin. Proc Natl Acad Sci USA 1976;73:2885–9. [9] Knight AK, Bingemann T, Cole L, Cunningham-Rundles C. Frequent false positive beta human chorionic gonadotropin in immunoglobulin A deficiency. Clin Exp Immunol 2005;141:333–7. [10] Rotmensch S, Cole LA. False diagnosis and needless therapy of presumed malignant disease in women with false-positive human chorionic gonadotropin concentrations. Lancet 2000;355:712–15. [11] Hussa RO, Rinke ML, Schweitzer PG. Discordant human chorionic gonadotropin results: causes and solutions. Obstet Gynecol 1985;65:211–19. [12] Regelson W. Have we found the “definitive cancer biomarker”? The diagnostic and therapeutic implications of human chorionic gonadotropin-beta statement as a key to malignancy. Cancer 1995;76:1299–301. [13] Butler SA, Iles RK. Ectopic human chorionic gonadotropin secretion by epithelial tumors and human chorionic gonadotropin -induced apoptosis in Karposi’s sarcoma. Is there a connection? Clin Cancer Res 2003;9:4666–73. [14] Iles RK. Human chorionic gonadotropin and its fragments as markers of prognosis in bladder cancer. Tumor Marker Update 1995;7:161–6. [15] Cole LA, Khanlian SA, Muller CY. Detection of hCG peri- or post-menopause an unnecessary source of alarm. Am J Obstet Gynecol 2008;198:275–9. [16] Cole LA, Sasaki Y, Muller CY. Normal production of hCG in menopause: a medical management dilemma. N Engl J Med 2007;356:1184–6. [17] Gronowski AM, Fantz CR, Parvin CA, Sokoll LJ, Wiley CL, Wener MH, et al. Use of serum FSH to identify perimenopausal women with pituitary hCG. Clin Chem 2008;54:652–6. [18] Cole LA, Ladner DG. Background hCG in non-pregnant individuals: need for more sensitive point-of-care and over-the-counter pregnancy tests. Clin Biochem 2009; 42:168–75. [19] Cole LA, Laidler LL. Inherited hCG. J Reprod Med 2010;55:99–102. [20] Matsuura S, Ohashi M, Chen HC, Shownkeen RC, Hartree AS, Reichert Jr LE, et al. Physicochemical and immunological characterization of an hCG-like substance from human pituitary glands. Nature 1980;286:740–1. [21] Huang SC, Chen HC, Chen RJ, Hsieh CY, Wei PY, Ouyang PC. The secretion of human chorionic gonadotropin-like substance in women employing contraceptive measures. J Clin Endocrinol Metab 1984;58:646–53. [22] Wahlstrom T, Stenman UH, Lundqvist C, Tanner P, Schroder J, Seppala M. The use of monoclonal antibodies against human chorionic gonadotropin for immunoperoxidase
336
[23] [24] [25]
[26]
[27]
[28]
[29]
[30] [31]
[32]
[33]
[34]
[35] [36] [37] [38]
[39] [40]
Human Chorionic Gonadotropin (hCG)
staining of normal placenta, pituitary gland, and pituitary adenomas. J Histochem Cytochem 1981;29:864–5. Seki K, Matsui H, Sekiya S. Advances in the clinical laboratory detection of gestational trophoblastic disease. Clin Chim Acta 2004;349:1–13. Cole LA, Khanlian SA. Inappropriate management of women with persistent low hCG results. J Reprod Med 2004;49:423–32. Birken S, Maydelman Y, Gawinowicz MA, Pound A, Liu Y, Hartree AS. Isolation and characterization of human pituitary chorionic gonadotropin. Endocrinol 1996;137:1402–11. Hoermann R, Spoettl G, Moncayo R, Mann K. Evidence for the presence of human chorionic gonadotropin (hCG) and free beta-subunit of hCG in the human pituitary. J Clin Endocrinol Metab 1990;71:179–86. Snyder JA, Haymond S, Parvin CA, Gronowski AM, Grenache DG. Diagnostic considerations in the measurement of human chorionic gonadotropin in aging women. Clin Chem 2005;51:1830–5. Cole LA, Khanlian SA, Giddings A, Butler SA, Muller CY, Hammond C, et al. Gestational trophoblastic diseases: 4. Presentation with persistent low positive human chorionic gonadotropin. Gynecol Oncol 2006;102:165–72. Hartree AS, Shownkeen RC, Stevens VC, Matsuura S, Ohashi M, Chen H-C. Studies of human chorionic gonadotropin-like substance of human pituitary glands and its significance. J Clin Endocrinol Metab 1983;96:115–26. Louhimo J, Kokkola A, Alfthan H, Stenman UH, Haglund C. Preoperative hCG and CA 72–4 are prognostic factors in gastric cancer. Int J Cancer 2004;111:929–33. Carpelan-Holmstrom M, Louhimo J, Stenman UH, Alfthan H, Jarvinen H, Haglund C. Estimating the probability of cancer with several tumor markers in patients with colorectal disease. Oncol 2004;66:296–302. Louhimo J, Alfthan H, Stenman UH, Haglund C. Serum hCG beta and CA 72–4 are stronger prognostic factors than CEA, CA 19–9 and CA 242 in pancreatic cancer. Oncol 2004;66:126–31. Hotakainen K, Haglund C, Paju A, Nordling S, Alfthan H, Rintala E, et al. Chorionic gonadotropin beta-subunit and core fragment in bladder cancer: mRNA and protein expression in urine, serum and tissue. Eur Urol 2002;41:677–85. Hotakainen K, Ljungberg B, Paju A, Rasmuson T, Alfthan H, Stenman UH. The free beta-subunit of human chorionic gonadotropin as a prognostic factor in renal cell carcinoma. Br J Cancer 2002;86:185–9. Odell WD, Griffin J. Pulsatile secretion of human chorionic gonadotropin in normal adults. N Engl J Med 1987;317:1688–91. Odell WD, Griffin J. Pulsatile secretion of chorionic gonadotropin during the normal menstrual cycle. J Clin Endocrinol Metab 1989;69:528–32. Policastro PF, Daniels-McQueen S, Carle G, Boime I. A map of the hCG beta-LH beta gene cluster. J Biol Chem 1986;13:5907–16. Lei ZM, Reshef E, Rao CV. The expression of human chorionic gonadotropin/ luteinizing hormone receptors in human endometrial and myometrial blood vessels. J Clin Endocrinol Metab 1992;75:651–9. Gougeon A. Dynamics of follicular growth in the human: a model from preliminary results. Human Reprod 1986;1:81–7. Bogovich K, Richards JS, Reichert Jr LE. Obligatory role of luteinizing hormone (LH) in the initiation of preovulatory follicular growth in the pregnant rat: specific effects of
Pituitary hCG and Familial hCG
[41] [42]
[43] [44]
[45] [46]
337
human chorionic gonadotropin and follicle-stimulating of hormone on LH receptors and steroidogenesis in theca, granulosa, and luteal cells. Endocrinol 1981;109:860–7. Motola S, Popliker M, Tsafriri A. Are steroids obligatory mediators of LH/hCG triggered resumption of meiosis in mammals? Biol Reprod 2007;77:167–8. Hegele-Hartung C, Grütznera M, Lessla M, Grøndahl C, Ottesen JL, Brännström M. Activation of meiotic maturation in rat oocytes after treatment with follicular fluid meiosis-activating sterol in vitro and ex vivo. Biol Reprod 2001;64:418–24. Robkera RL, Russella DL, Yoshiokab S, Sharmaa SC, Lydona JP, O’Malleya BW, et al. Ovulation: a multi-gene, multi-step process. Steroids 2000;65:559–70. Butler TA, Zhu C, Mueller RA, Fuller GC, Lemaire WJ, Woessner Jr JF. Inhibition of ovulation in the perfused rat ovary by the synthetic collagenase inhibitor SC 44463. Biol Reprod 1991;44:1183–8. Acosta JT, Miyamoto A. Vascular control of ovarian function: ovulation, corpus luteum formation and regression. Anim Reprod Sci 2004;82-83:127–40. Nalbandov AV, Bahr JM. Ovulation, corpus luteum formation, and steroidogenesis. Basic Life Sci 1974;4:399–407.
32 hCG, Free -Subunit, and
-Core Fragment as Markers of Malignancies Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
In many ways detecting the free -subunit in the blood of a person with cancer is like finding an indicator of poor prognosis. This means the cancer is producing the free -subunit that promotes invasion and metastaes in the malignancy, thus causing it to grow and become more aggressive [1–5]. Fortunately, there are hCG vaccines currently undergoing clinical trials worldwide. These vaccines will generate antibodies in humans that will block the free -subunit and improve the outcome and longevity of a patient’s life (see Chapter 14) [6–8]. Cancer cells primarily produce hyperglycosylated hCG (hCG-H) free -subunit, an overly glycosylated, uncombined free -subunit. Free -subunit is quickly removed from the circulation. As such, low levels of free -subunit can best be detected as their degradation product, -core fragment, in urine. As claimed [9,10], every cancer actually secretes free -subunit, which is detectable on the cancer’s surface. This is part of the process of carcinogenesis. As published [11,12], approximately 50% of all cancers actually produce free -subunit in serum or its -core fragment in urine. In this chapter, we describe free -subunit and -core fragment and analyze their practical use as tumor markers.
32.1 hCG and Tumor Markers Early studies describe the production of free -subunit by cervical cancer and breast, bladder, ovarian, cervical, brain, colorectal, uterine, and lung malignancy cell lines [1,13–20]. As reported by Acevedo et al. [9] and by Regelson [24], almost every human malignancy generates hCG free -subunit, but this claim is rather contentious. At best, the diagnostic potential for even the most recognizable free -subunit producing tumor is around 30% (Table 32.1). Recent studies clearly show that free -subunit is not a high-sensitivity diagnostic marker in these malignancies, but it does seem to have value as being an excellent marker for poor prognosis [22]. When free -subunit is degraded, it generates -subunit core fragment, which is a more sensitive tumor marker in urine samples (Table 32.2) [2,23]. Serum hCG free Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00032-3 © 2010 Elsevier Inc. All rights reserved.
340
Human Chorionic Gonadotropin (hCG)
Table 32.1 Use of Serum Free -Subunit as a Tumor Marker When Detecting Malignancies. Malignancy
Patients
Sensitivity (>3 fmol/ml)
Ovarian cancer Cervical cancer Endometrial cancer Vulvar cancer Bladder cancer Lung cancer Colorectal cancer Total
150 60 55 50 170 243 436 1164
38%, average of 3 reports 37%, average of 2 reports 33%, average of 2 reports 38%, a single report 35%, average of 3 reports 18%, average of 2 reports 17%, average of 2 reports 30%, average of 16 reports
Averages are determined by combining total positive cases from multiple reports [2,11,15,21,22,26,27].
Table 32.2 Use of Urine -Subunit Core Fragment as a Tumor Marker for Detecting Malignancies. Malignancy
Patients
Sensitivity (>3 fmol/ml)
Ovarian cancer Cervical cancer Endometrial cancer Pancreatic cancer Bladder cancer Lung cancer Total
207 410 157 29 102 122 1027
66% 48% 47% 55% 48% 24% Mean 48%
Data from multiple reports [11,13,19,28,29].
-subunit and urine -subunit core fragments have both proven useful in the detection and management of a wide range of malignancies [2,15,18]. As shown in Tables 32.1 and 32.2, 48% of 1027 malignancy cases tested positive for urine -subunit core fragment; 30% of 1164 malignancy cases tested positive for serum free . The expression of the free - or -subunit core fragment in a patient with malignancy correlates with poor grade and stage of the tumor [24,25]. This indicates that expression of free -subunit is associated with a negative outcome in human malignancies. Tables 32.1 and 32.2 show that gynecologic and urinary tract malignancies appear to be the biggest producers of free -subunit or -core fragment. Free -subunit is detected in serum of 38% of ovarian malignancies; -subunit core fragment is detected in the urine of 68% of these same malignancies. Similarly, free -subunit is detected in the serum of 38% of bladder malignancies, and -subunit core fragment is detected in the urine of 48% of these same malignancies. In contrast, free -subunit is detected in the serum of just 18% of lung cancer cases, and -subunit core fragment in the urine of just 24% of these malignancies. Serum free -subunit and urine -core fragment are nonspecific tumor markers. They detect a proportion of all malignancies. The urine -core fragment test might be
Table 32.3 hCG Variants in 44 Patients with Nontrophoblastic Malignancies Referred to in the USA hCG Reference Service. Age
Total hCG (ng/ml)
hCG-H (ng/ml)
41 55 27 41 43 50 33 50 28 48 26 25 17 34 19 25 51 39 14 15 14 26 19 43 40 41
0.75 0.92 0.60 1.1 0.65 0.62 1.7 3.9 0.13 3.3 18 1.5 0.1 14 19 64 29 11 0.39 0.23 0.45 2.3 4.4 31 25 45
0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.26 0.2 0.05 0.05 0.44 0.05 0.05 0.05 0.05 0.05 11.3 0.37 0.05
Molar (%)
1.9 1.1
4.0
36
Free -Subunit (ng/ml)
Molar (%) -Core Fragment Molar (ng/ml) (%)
4.6 0.29 0.37 0.61 0.38 0.58 0.83 2.3 0.08 1.5 8.06 0.66 0.03 9.6 1.4 14.5 1.4 6 0.23 0.03 0.17 0.96 1.9 18.3 1.2 12.3
1014 52 102 92 97 155 81 97 102 75 74 73 113 12 37 8 90 97 22 62 69 71 98 8 45
0.49 0.52 1.17 0.93 0.13 0.053 0.1 0.02 0.064 30 9.67 0.02 0.02 3 0.83 34 0.18 22.6 0.1 0.1 0.4 0.3 1.4 21 1 490
Final Diagnosis
240 207 716 310 73 31 22 181 3336 197 Endometrial 79 16 195 2 754 94 160 326 48 117 249 15 3996
Dysgerminoma
Dysgerminoma Dysgerminoma
Dysgerminoma
(continued)
Table 32.3 (Continued) Age
Total hCG (ng/ml)
hCG-H (ng/ml)
37 32 21 49 40 34 27 20 26 52 27 50 54 17 45 49 21 46
1.1 30 3.9 0.42 21 17 7.0 22 22 25 19 11 55 10 3.9 0.88 12 1.0
0.05 0.05 0.05 0.05 0.08 0.3 0.05 0.05 0.05 0.13 0.13 0.05 0.05 0.13 2.7 0.05 1.54 0.05 Median
Molar (%)
0.4 1.8
0.5 0.7
1.3 69 13 1.8
Free -Subunit (ng/ml)
Molar (%) -Core Fragment Molar (%) (ng/ml)
0.73 1.8 2.2 0.14 9.3 7.4 3.3 10.2 11 11 7.6 28 23.8 4 2.73 0.02 8.5 0.59 Median
110 10 93 55 73 72 78 77 83 73 66 421 72 66 116 117 98 79
0.9 0.22 0.8 0.02 520 23 54 67 223 120 256 121 167 2 44 0.31 2.07 0.02 Median
300 3 75 17 9088 497 2831 1118 3720 1762 4945 4037 1114 73 4141 129 63 202
Final Diagnosis
Dysgerminoma Bladder Ovarian
Bladder cancer Dysgerminoma Renal Colon cancer Dysgerminoma
hCG, Free -Subunit, and -Core Fragment as Markers of Malignancies
343
useful as a general nonspecific screening test for all malignancies, as it detects 48% of all malignancies. A positive test, however, because the test is nonspecific, would need to be followed by a complete body malignancy scan in order to detect the site of the malignancy (MRI scan of head and pelvis and CT scan of chest). The USA hCG Reference Service has consulted on 44 cases that were later proven to have a nontrophoblastic neoplasm (Table 32.3). In 40 of these cases, cancer was not identified until the USA hCG Reference Service consultation. MRI and CT were then performed. From feedback to the Service, we know that most consulted cases were ovarian germ cell tumors or dysgerminomas (Table 32.3). Germ cell tumors might be one of the biggest producers of free -subunit. They can lead to cases with total hCG levels of 64 ng/ml (704 mIU/ml) in serum and -core fragment levels of 520 ng/ml (molar equivalent of 20,800 mIU/ml of hCG) in urine samples (Table 32.3). Among the 44 cases, most of what was detected was free -subunit and -core fragment (molar equivalent of total hCG). This clearly identified these molecules as the principal forms being produced by malignant cells. hCG-H had no value at all as a nontrophoblastic tumor marker.
References [1] Butler SA, Ikram MS, Mathieu S, Iles RK. The increase in bladder carcinoma cell population induced by the free beta subunit of hCG is a result of an anti-apoptosis effect and not cell proliferation. Br J Cancer 2000;82:1553–6. [2] Iles RK. Ectopic hCG expression by epithelial cancer: malignant behavior metastasis and inhibition of tumor cell apoptosis. Mol Cell Endocrinol 2007;260:264–70. [3] Carter WB, Sekharem M, Coppola D. Human chorionic gonadotropin induces apoptosis in breast cancer. Breast Cancer Res Treat 2006;100:S243–4. [4] Iles RK. hCG beta expression by cervical squamous carcinoma—in vivo histological association with tumour invasion and apoptosis. Histopathology 2008;53:147–55. [5] Crawford RA, Iles RK, Carter PG, Caldwell CJ, Shepherd JH, Chard T. The prognostic significance of beta human chorionic gonadotropin and its metabolites in women with cervical carcinoma. J Clin Pathol 1998;51:685–8. [6] Delves PJ, Iles RK, Roitt IM, Lund T. Designing a new generation of anti-hCG vaccines for cancer therapy. Mol Cell Endocrinol 2007;260:276–81. [7] Iversen PL, Mourich DV, Moulton HM. Monoclonal antibodies to two epitopes of -human chorionic gonadotropin for the treatment of cancer. Curr Opin Mol Ther 2003;5:156–60. [8] Moulton HM, Yoshihara PH, Mason DH, Iversen PL, Triozzi PL. Active specific immunotherapy with -human chorionic gonadotropin peptide vaccine in patients with metastatic colorectal cancer: antibody response is associated with improved survival. Clin Cancer Res 2002;8:2044–51. [9] Acevedo HF, Tong JY, Hartsock RJ. Human chorionic gonadotropin-beta subunit gene statement in cultured human fetal and cancer cells of different types and origins. Cancer 1995;76:1467–75. [10] Acevedo HF, Hartstock RJ. Metastatic phenotype correlates with high expression of membrane-associated complete -human chorionic gonadotropin in vivo. Cancer 1996;78:2388–99. [11] Muller C, Cole LA. The quagmire of hCG and hCG testing in gynecologic oncology. Gynecol Oncol 2009;112:663–72.
344
Human Chorionic Gonadotropin (hCG)
[12] Cole LA. New discoveries on the biology and detection of human chorionic gonadotropin. Reprod Biol Endocrinol 2009;7:1–37. [13] Cole LA. -core fragment (-core UGP or UGF). Tumor Marker Update 1994;6:69–75. [14] Nishimura R, Baba S, Hasegawa K, Kinugasa M, Okamura M, Kimura A, et al. Characterization of immunoreactive hCG beta-subunit in cultured fluids of the cell lines derived from gynecologic malignant tumors. Nippon Sanka Fujinka Gakkai Zasshi 1990;42:1471–6. [15] Marcillac I, Cottu P, Theodore C, Terrier-Lacombe MJ, Bellet D, Droz JP. Free hCG beta subunit as tumour marker in urothelial cancer. Lancet 1993;341:1354–5. [16] Bepler G, Jaques G, Oie HK, Gazdar AF. Human chorionic gonadotropin and related glycoprotein hormones in lung cancer cell lines. Cancer Lett 1991;58:145–50. [17] Offner FA, Ott G, Povey S, Knuechel R, Preisler V, Fuezesi L, et al. Characterization of the new bladder cancer cell line HOK-1: expression of transitional squamous and glandular differentiation. Virchow Arch 1991;428:99–106. [18] Ozturk M, Bellet D, Isselbacher KJ, Wands J. Ectopic beta-human chorionic gonadotropin production by a human hepatoma cell line (FOCUS): isolation and immunochemical characterization. Endocrinol 1987;120:559–66. [19] Rosen SW, Calvert I, Weintraub BD, Tseng JS, Rabson AS. Stimulation of N6O29-dibutyryl cyclic adenosine 39:59-monophosphate of ectopic production of the free beta subunit of chorionic gonadotropin by a human brain tumor cell line. Cancer Res 1980;40:4325–8. [20] Udagawa Y, Nozawa S, Chin K, Sakayori M, Mikami M, Ohta K, et al. Biological properties of two newly established cell lines (SKG- 3a,3b) from a human uterine cervical epidermoid carcinoma. Nippon Sanka Fujinka Gakkai Zasshi 1984;36:237–46. [21] Cole LA, Hartle RJ, Laferla JJ, Ruddon RW. Detection of the free subunit of human chorionic gonadotropin in cultures of normal and malignant trophoblast cells, pregnancy sera, and sera of patients with choriocarcinoma. Endocrinol 1983;113:1176–8. [22] Iles RK. Human chorionic gonadotropin and its fragments as markers of prognosis in bladder cancer. Tumor Marker Update 1995;7:161–6. [23] Cole LA, Schwartz PE, Wong Y. Urinary gonadotropin fragments (UGF) in cancers of the female reproductive system: I sensitivity and specificity comparison with other markers. Gynecol Oncol 1988;31:82–90. [24] Regelson W. Have we found the “definitive cancer biomarker”? The diagnostic and therapeutic implications of human chorionic gonadotropin-beta statement as a key to malignancy. Cancer 1995;76:1299–301. [25] Butler SA, Iles RK. Ectopic human chorionic gonadotropin secretion by epithelial tumors and human chorionic gonadotropin -induced apoptosis in karposi’s sarcoma: is there a connection? Clin Cancer Res 2003;9:4666–73. [26] Szturmowicz M, Slodkowska J, Zych J, Rudzinski P, Sakowicz A, RowinskaZakrzewska E. Frequency and clinical significance of -subunit human chorionic gonadotropin expression in non-small cell lung cancer patients. Tumor Biol 1999;20:99–104. [27] Cook AM, Huddart RA, Jay G, Norman A, Dearnaley DP, Horwich A. The utility of tumor markers in assessing the response to chemotherapy in advanced bladder cancer. Br J Cancer 2000;82:1952–7. [28] Kinugasa M, Nishimura R, Koizumi T, Morisue K, Higashida T, Natazuka T, et al. Combination assay of urinary beta-core fragment of human chorionic gonadotropin with serum tumor markers in gynecologic cancers. Jpn J Cancer Res 1995;86:783–9. [29] Cole LA, Tanaka A, Kim GS, Park SY, Koh MW, Schwartz PE, et al. Beta-core fragment (beta-core/UGF/UGP), a tumor marker: a 7-year report. Gynecol Oncol 1996;60:264–70.
33 hCG and Hyperglycosylated hCG
Purification and Analysis from Serum, Urine, and Culture Fluids Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
Over the past 20 years, the USA hCG Reference Service has used a variety of methods for purifying hCG and hyperglycosylated hCG (hCG-H) from serum, urine, and culture fluids. Most of these methods are published [1–3] and were used to obtain pure hCG and hCG-H for carbohydrate structure and peptide cleavage studies and for generating antibodies. Our basic purification protocol was derived from the purification methods of Birken et al. [4], which are used for hCG amino acid analysis and hCG standards. Over the past 20 years, we have purified approximately 40–45 hCG preparations from serum, urine, and culture fluids.
33.1 Urine hCG and hCG-H The USA hCG Reference Service has purified more hCG and hCG-H preparations from large volumes of urine than any other source [1,2]. Generally, 4–10 mg of hCG is required as starting material, which can be obtained from about 3–10 L of urine from pregnancy peak, choriocarcinoma, or hydatidiform mole cases. It is essential to have at least 4 mg of hCG because lower concentrations of pure hCG are very sticky and often lost in glassware. The first stage of purification is removal of lipids, salts, sugars, and small peptides. This is performed by precipitating the urine with acetone (one volume urine to two volumes acetone). The acetone is first cooled to 4°C before adding it to the urine. Once added to the urine, it is stirred continuously overnight in a refrigerator. In the morning, the mixture is carefully decanted down to a 1 L volume, transferred to 250 or 500 ml centrifuge bottles, and spun at 2000 rpm for 30 min. After aspirating all the liquid, the powder is dried overnight under a stream of desiccant-dried air. This is a simple method in which very little hCG or hCG-H is lost in the final powder. The powder is slowly dissolved in 100 ml of distilled water and centrifuged to remove any precipitate (hCG is extremely soluble in water). The precipitate is discarded. Nine volumes of ethanol are now added (9 volumes ethanol to 1 volume hCG Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00033-5 © 2010 Elsevier Inc. All rights reserved.
348
Human Chorionic Gonadotropin (hCG)
solution). Once again, the mixture is continuously stirred overnight, allowed to settle for 30 min. This liquid is then aspirated and then centrifuged in the 250 or 500 ml centrifuge tubes. Small volumes of pure ethanol are used to extract the precipitate from the multiple centrifuge bottles. All the precipitates are then transferred into one bottle. The single bottle of precipitate is recentrifuged and allowed to dry overnight under a stream of desiccant-dried air. The second purification stage uses a crude gel filtration to separate hCG and hCG-H from most other proteins. The single precipitate in the centrifuge vessel is taken up in 20 ml of distilled water and transferred to a smaller centrifugation vessel (25 or 50 ml). Any precipitate is discarded. A large column (2.6 cm 100 cm) of Sephacryl S200 High Resolution gel is prepared. This should be equilibrated with ammonium bicarbonate 0.05 M (no pH adjustment needed); ammonium bicarbonate must be used because it is vaporized upon lyophilization. The 20 ml preparation is slowly passed through the 500 ml volume column and pumped at approximately 20 ml/h. Fractions of 100 5 ml are collected and assayed for total hCG. All peak hCG tubes are collected, pooled, and lyophilized. The lyophilizate is taken up in a total of 10 ml of distilled water and centrifuged to remove any precipitate. It is then placed on a 200-ml DEAE-Sepharose column. The column is equilibrated with ammonium bicarbonate 0.05 M. After pumping the sample onto the column, elute the column with 250 ml ammonium bicarbonate 0.05 M; then elute with the same volume of buffer containing sequentially 0.05, 0.10, 0.15, 0.20, 0.25, and 1.0 M sodium chloride. Volumes are either collected in tubes or in beakers (changing beakers with each change in buffer concentration). hCG normally elutes with 0.10 M sodium chloride and hCG-H with 0.15 M sodium chloride. The 250 ml concentrates are rapidly dialyzed and lyophilized. A fine gel filtration step is then necessary. hCG and hCG-H should now be purified separately. The lypohilizate is taken up in a 5 ml volume of distilled water and applied to a 100 cm 1.3 cm gel filtration column. This should be Sephacryl S200 HR or S300 HR. This final gel filtration should be pumped slowly over 2 days and equilibrated with ammonium bicarbonate 0.05 M. The final pure hCG and hCG-H peaks should be collected and rapidly lyophilized, then tested for purity as described later. In our experience, this procedure yields approximately 60–70% of the starting material.
33.2 Culture Fluid hCG and hCG-H We have found that the methods used to purify urine (described earlier) also work to generate pure hCG and hCG-H from culture fluids. It is noted that the culture fluids must be serum free in order for purification to be possible (otherwise, see serum hCG purification, described later). Culture fluids have to be such a volume as to contain 2–10 mg of hCG or hCG-H in the starting material. The culture fluids might have to be as great as 100 l in volume. Volumes of hCG or hCG-H in culture fluids can be reduced by using a Millipore Minitan or other tangential flow dialysis protein concentration machine, using cutoff molecular weight 10,000.
hCG and Hyperglycosylated hCG Purification and Analysis from Serum, Urine, and Culture Fluids
349
33.3 Serum hCG and hCG-H Affinity purification can be considered for serum samples. The hCG Reference Service has had some success in purifying hCG from serum samples using this method. In our studies, we had 200 ml of serum from an individual with a Down syndrome pregnancy and a total hCG of 165,000 mIU/ml (about 15 g/ml total hCG, or 3.0 mg hCG and hCG-H total). In another study, we had 50 ml of serum from a hydatidiform mole case with a total hCG of 550,000 mIU/ml (about 50 g/ml total hCG, or 2.5 mg total). In both studies, we prepared 40 ml of affinity gel with a monoclonal antibody against the folded core of the ß-subunit of hCG. The antibody was completely pure. Complete purity is essential for this purification process and thus must be checked. We bound the antibody to the gel at an approximate concentration of 1 mg/ml using cyanogen bromide–activated Sepharose. Two affinity chromatography procedures were required to purify hCG and hCG-H from serum. In the first procedure, we placed the serum in a beaker with gel and mixed it overnight very slowly (do not mix quickly or it will destroy the gel matrix). We then poured the gel into a small chromatography column and allowed it to drain. We washed the column with 100 ml ammonium bicarbonate 0.05 M. We eluted the gel very slowly (10 ml/h) with 20 ml guanidine hydrochloride 4 M. All solutions should be at refrigerator temperature in order to avoid dissociation of hCG subunits. We then dialyzed the eluate overnight in ammonium bicarbonate 0.05 M. The dialyzate was then lyophilized. We found that the hCG in the dialyzate was only about 20% pure (mass of hCG versus mass of protein). We then repeated the affinity chromatograph procedure a second time, once again eluting with guanidine hydrochloride 4 M. This method gave approximate yield of 40% hCG or hCG-H.
33.4 Analysis All samples are analyzed for purity by reducing sodium dodecyl sulfate polyacrylamide gel electrophoresis with coomassie blue staining [2]. Only two bands corresponding to the - and ß-subunits of hCG (elution at molecular sizes 18,000 and 32,000) will be observed. Alternatively, the observation of three bands could be consistent with nicked hCG ß-subunit and an -subunit, with two ß-subunit bands corresponding to ß1–ß47 and ß48–ß145 (elution at molecular sizes 18,000 and 28,000).
References [1] Kardana A, Elliott ME, Gawinowicz MA, Birken S, Cole LA. The heterogeneity of hCG: I. Characterization of peptide variations in 13 individual preparations of hCG. Endocrinol 1991;129:1541–50. [2] Elliott MM, Kardana A, Lustbader JW, Cole LA. Carbohydrate and peptide structure of the - and -subunits of human chorionic gonadotropin from normal and aberrant pregnancy and choriocarcinoma. Endocrine 1997;7:15–32.
350
Human Chorionic Gonadotropin (hCG)
[3] Cole LA. Distribution of O-linked sugar units on hCG and its free alpha-subunit. Mol Cell Endocrinol 1987;50:45–57. [4] Birken S, Canfield RE. Isolation and amino acid sequence of COOH-terminal fragments from the -subunit of human chorionic gonadotropin. J Biol Chem 1977;252: 5386–92.
34 Dissociation, Desialylation, and Cleavage of hCG Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
This chapter describes the procedures used to generate standards, desialylated molecules, and cleaved molecules. Generally, these procedures are useful when studying the biology of hCG and hyperglycosylated hCG (hCG-H). These procedures are not published and represent methods used by the USA hCG Reference Service.
34.1 Dissociation The following procedure dissociates hCG and hCG-H into - and -subunits. First, a solution with 6 M guanidine hydrochloride in sodium citrate buffer (pH 4.0) is prepared. Next, 3 ml of this solution is added to 1–10 mg of solid hCG or hCG-H. This mixture is then incubated for 1 h at 37 °C, then diluted with distilled water twofold because 3 M guanidine hydrochloride is less viscous. The diluted mixture is then placed on a 1.3 cm 100 cm column of Sephacryl S200 equilibrated with 0.05 M ammonium bicarbonate. After the hCG-3M guanidine hydrochloride sample has entered the column, another 20 ml of 3 M guanidine hydrochloride in 0.05 M ammonium bicarbonate solution is added in order to prevent the rapid recombination of hCG subunits. Finally, the column is washed with 0.05 M ammonium bicarbonate and 100 2 ml fractions are collected. Peaks corresponding to the dissociated - and -subunits of hCG are collected.
34.2 Desialylation hCG, hCG-H, or their free - and -subunits are desialylated by treatment with neuraminidase. We particularly recommend Clostridium perfringens neuramindase, most notably the lyophilized powder sold by Sigma Chemical Co., called Neuraminidase Type-X. A 10 buffer is prepared using 0.5 M sodium acetate (pH 5.0). Five units of enzyme are added to 0.25 ml of the 10 buffer. hCG or hCG-H is prepared by adding a 0.1–10 mg mass in 0.8 ml distilled water, then adding 0.1 ml neuraminidase (in 0.5 M buffer). This mixture is incubated for 24 h at 37°C. After incubation, another 0.1 ml of neuraminidase is added and the mixture is incubated for an additional 24 h at 37°C. The hCG or hCG-H is now desialylated. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00034-7 © 2010 Elsevier Inc. All rights reserved.
352
Human Chorionic Gonadotropin (hCG)
34.3 Cleavage of -Subunit C-Terminal Peptide Thermolysin can cleave the C-terminal peptide of hCG or hCG-H without dissociation of subunits, specifically at residues 114/115 of the 145 amino acid -subunit. Once cleaved, 0.2–2 mg of hCG or hCG-H can be extracted using 0.1 ml distilled water. Thermolysin (6 mg) is prepared in 0.9 ml of 0.1 M sodium chloride, 0.05 M Tris (pH 7.5) plus 0.01 M calcium chloride. The hCG and thermolysin are then mixed and incubated for 2 h at 25°C. After incubation, 0.025 ml glacial acetic acid is added to stop the reaction from proceeding. This mixture is immediately placed on a 1.3 cm 100 cm Sephacryl S200 HR gel filtration column equilibrated with 0.05 M ammonium acetate in order to separate the core hCG and C-terminal peptide.
34.4 Reduction and S-Carboxymethylation of hCG Reduction and S-carboxymethylation separates the subunits of hCG and hCG-H; cleaves all disulfide linkages in the subunits; and, through S-carboxymethylation of the cysteine residues, fixes the molecule in an unfolded state. This is useful for protease studies, for cleaving oligosaccharides, and for other experiments requiring unfolded subunits. First, hCG and hCG-H are dissociated and separated using 3 M guanidine hydrochloride. Then 1–10 mg of hCG or hCG-H are taken up in 1 ml of a 0.05 M Tris–HCl buffer (pH 7.5) solution containing 3 M guanidine hydrochloride, 0.5 mM disodium EDTA, and 65 mM dithiothreitol (to break disulfide bridges). This mixture is incubated overnight at 37 °C. Iodoacetic acid (0.35 M) is prepared in 0.05 M Tris–HCl buffer (pH 7.5). A 1 ml volume of iodoacetic acid is added to the now dissociated mix. This mixture is shaken for 3 h at 37°C. The 2 ml mixture is then dialyzed overnight against 0.05 M ammonium bicarbonate and then lyophilized.
34.5 Tryptic Digest of Reduced and S-Carboxymethylated hCG Subunit Making small tryptic peptides of reduced and S-carboxymethylated hCG is useful for investigating hCG structure by mass spectrometry and other sequencing methods. Reduced and S-carboxymethylated subunits are digested with sequence grade trypsin (10% w/w) in 10 mM ammonium bicarbonate at 37°C for 16 h [1].
References [1] Valmu L, Alfthan H, Hotakainen K, Birken S, Stenman UH. Site-specific glycan analysis of human chorionic gonadotropin -subunit from malignancies and pregnancy by liquid chromatography–electrospray mass spectrometry. Glycobiol 2006;16:1207–18.
35 hCG and Free -Subunit Producing Cell Lines Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
Using an hCG and free -subunit producing cell line is the easiest way to study hCG and free -subunit biology. These types of cell lines can be useful when studying hCG and free -subunit synthesis, structure, invasion, and receptor binding. In general, all choriocarcinoma, testicular germ cell, and pregnancy extravillous trophoblast cell lines produce hyperglycosylated hCG (hCG-H), just like in the in vivo situation [1,2], whereas the majority of cancer cell lines primarily produce free -subunit, just like the in vivo picture [4–8]. As published, hCG-H promotes growth and invasion by choriocarcinoma and testicular germ cell lines and inhibits cell apoptosis [1–3], whereas free -subunit seemingly inhibits apoptosis and promotes growth and invasion in cancer cell lines [4–8]. Most of the cell lines listed can be maintained in Dulbecco’s modified Eagle’s medium with 10% fetal calf serum. Some need to be maintained in either McCoys 5A medium with 10% fetal calf serum, or RPMI-1640 cell culture medium containing 10% fetal calf serum. Almost every cell line can be obtained from American Type Culture Collection (ATCC) in Manassas, Virginia. A wide variety of cell lines (57 total) have been shown to produce hCG or the free -subunit from cancer sources, choriocarcinoma sources, and pregnancy trophoblast sources (Table 35.1). Table 35.1 hCG and Free -Subunit Producing Cell Lines, 57 Total. Cell Line
Tissue Source
Bladder Cancer SCABER T24 J82 RT112 RT4 5637 SW800
Squamous cell Transitional epithelial cell Transitional epithelial cell Epithelial cell Transitional papillary cell Transitional epithelial cell Transitional epithelial cell (Continued)
Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00035-9 © 2010 Elsevier Inc. All rights reserved.
354
Human Chorionic Gonadotropin (hCG)
Table 35.1 (Continued) Cell Line
Tissue Source
Bone Cancer G292 KHO2/NP
Osteosarcoma Osteosarcoma
Brain Cancer CBT
Glioblastoma
Breast Cancer MDA-MB436 MCF7 BT549
Adenocarcinoma cell Ductal carcinoma cell Papillary carcinoma
Cervical Cancer DOT CASKI HELA
Epidemoid cell Epidemoid cell Epithelial cell
Choriocarcinoma JAR JEG-3 BEWO HOCC
Cytotrophoblast Cytotrophoblast Cytotrophoblast Cytotrophoblast
Colon Cancer T84 WIDR LOVO
Colorectal carcinoma Colorectal adenocarcinoma Colorectal adenocarcinoma
Endometrial Cancer HEC1B AN-3-CA KLE ISHIKAWA
Adenocarcinoma Poorly differentiated endometrial Adenocarcinoma Ademocarcinoma
Karposi’s Sarcoma KS-SLK
Karposi’s sarcoma
Leukemia HL60 CCRF-CEM MOLT4 U937
Promyleucytic leukemia Lymphoblastic leukemia Myeloid leukemia Myeloid leukemia
Lymphoma U937 JIYOYE
Histiocytic lymphoma Burkitt lymphoma (Continued)
hCG and Free -Subunit Producing Cell Lines
355
Table 35.1 (Continued ) Cell Line
Tissue Source
Lung Cancer CHAGO A549 SKLU-1 RF-8150 NCI-H69 NCI-H526
Epithelial, nonsmall cell Adenocarcinoma, nonsmall cell Adenocarcinoma, nonsmall cell Squamous, nonsmall cell Small cell Small cell
Melanocarcinoma HT144 SK-MEL-1
Fibrobastic Lymphatic
Nasopharynx Cancer Detroit 562 RPMI-2650
Carcinoma of pharynx Squamous cell
Oral Cavity TR146
Buccal carcinoma
Ovary NCC-IT OVCAR3
Germinoma Epithelial
Prostate Cancer PC3 DU145 LN-CAP-FGC
Adenocarcinoma Epithelial Carcinoma
Pregnancy Trophoblast HTR8/SVNeo ACH-3P
Extravillous trophoblast First trimester trophoblast/choriocarcinoma fusion
Retinobastoma Y79
Eye retinoblastoma
Stomach IGSK
Adenocarcinoma
Testicular Cancer NTERA N2102 NEC8 JKT-1
Germ cell embryonal cancer Germ cell embryonal cancer Germ cell embryonal cancer Seminoma
356
Human Chorionic Gonadotropin (hCG)
A publication by Acevedo et al. found hCG and free -subunit in the membranes of cell cultures, using flow cytometric methods combined with immunoassays. A total of 74 cell lines were shown to be producing hCG or its free -subunit [9].
References [1] Cole LA, Schwartz PE, Wong Y. Urinary gonadotropin fragments (UGF) in cancers of the female reproductive system: I. Sensitivity and specificity, comparison with other markers. Gynecol Oncol 1988;31:82–90. [2] Cole LA, Khanlian SA, Riley JM, Butler SA. Hyperglycosylated hCG (hCG-H) in gestational implantation and in choriocarcinoma and testicular germ cell malignancy tumorigenesis. J Reprod Med 2006;51:919–29. [3] Hamade AL, Nakabayashi K, Sato A, Kiyoshi K, Takamatsu Y, Laoag-Fernandez JB, et al. Transfection of antisense chorionic gonadotropin gene into choriocarcinoma cells suppresses the cell proliferation and induces apoptosis. J Clin Endocrinol Metab 2005;90:4873–9. [4] Iles RK. Ectopic hCG expression by epithelial cancer: malignant behavior metastasis and inhibition of tumor cell apoptosis. Mol Cell Endocrinol 2007;260:264–70. [5] Li D, Wen X, Ghali L, Al-Shalabi FM, Docherty SM, Purkis P, et al. hCG expression by cervical squamous carcinoma in vivo histological association with tumor invasion and apoptosis. Histopathology 2008;53:147–155. [6] Butler SA, Ikram MS, Mathieu S, Iles RK. The increase in bladder carcinoma cell population induced by the free beta subunit of hCG is a result of an anti-apoptosis effect and not cell proliferation. Br J Cancer 2000;82:1553–6. [7] Butler SA, Iles RK. Ectopic human chorionic gonadotropin secretion by epithelial tumors and human chorionic gonadotropin -induced apoptosis in Karposi’s sarcoma is there a connection? Clin Cancer Res 2003;9:4666–73. [8] Jankowska A, Gunderson SI, Andrusiewicz M, Burczynska B, Szczerba A, Jarmolowski A, et al. Reduction of human chorionic gonadotropin beta subunit expression by modified U1 snRNA caused apoptosis in cervical cancer cells. Mol Cancer 2008;7:26–34. [9] Acevedo HF, Krichevsky A, Campbell-Acevedo EA, Galyon JC, Buffo MJ, Hartsock RJ. Expression of membrane-associated human chorionic gonadotropin, its subunits, and fragment by culture human cancer cells. Cancer 1992;69:1829–42.
36 Evolution of hCG, Evolution of
Humans, and Evolution of Human Pregnancy Disorders and Cancer Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
The evolution of chorionic gonadotropin (CG) is a biological marvel in which two separate molecules evolved from a single deletion mutation in early primates. Dozens of incredible changes in amino acid sequences occurred during the development of CG and hyperglycosylated CG (CG-H) in early primates, advanced primates, and humans. These changes took CG and CG-H from nonacidic molecules with minimal activity in early primates to very acidic molecules with a long circulating half-life and maximal activity in humans. In so doing, they made human CG and CG-H ultrapotent. Although CG promotes differentiation of cytotrophoblast cells to syncytiotrophoblast cells, CG-H promotes the formation of cytotrophoblast columns and their extensions into primitive villi in early primates (Chapter 3). Together, CG and CG-H drive the formation of trophoblast villi, the root structures of hemochorial placentation. CG drives angiogenesis in uterine vasculature so that invading villi are met with maternal blood. Both CG and CG-H drive the efficient nutrient system that we call hemochorial placentation. It is seemingly the evolution of ultraefficient CG and CG-H that drove hemochorial placentation to its maximum potential. In turn, an increasingly efficient transfer of fetal nutrients eventually led to an increased amount of nutrients needed to form the human brain. Although the generation of ultraefficient hemochorial placentation allowed the development of the human brain, it came with its deficits. For humans, the demanding requirements of CG and CG-H led to an extraordinarily high incidence of pregnancy failures. As an added complication, the extreme hemochorial placentation efficiency in humans might have led to the evolution of pregnancy disorders such as preeclampsia and hypertensive pregnancy. The generation of highly efficient CG-H gave humans a resourceful invasion driver. The transformation of the cells that generate this invasion driver leads to choriocarcinoma, one of the worst malignancies known to humans. All human cancers can express the free -subunit of hCG. The free -subunit of hCG becomes hyperglycosylated and promotes the growth and invasion of human malignancies. In this chapter, we deal with CG evolution, the evolution of humans and human CG, and the complications humans now face because of the tremendous evolution of these molecules. Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00036-0 © 2010 Elsevier Inc. All rights reserved.
360
Human Chorionic Gonadotropin (hCG)
36.1 CG Evolution The story of CG evolution was first described by Fiddes and Goodman in 1979 [1,2]. It was detailed, refined, and extended by Maston and Ravulo [3], by Hallast and colleagues [4], and by Policastro and colleagues [5]. Prior to the molecules produced by primates, no CG or CG-H form was ever made by another species. Even the most primitive prosimian primates (e.g., Lemurs) do not produce any form of CG; however, both New World and Old World early simian primates produce a form of CG [1–4]. CG forms came about with the evolution of the early simian primates as a deletion mutation in the gene coding for the -subunit of luteinizing hormone (LH) (Figure 36.1). As illustrated, this deletion mutation in the codon corresponding to LH -subunit amino acid residue-114 caused a frame shift in the LH -subunit gene. The termination codon TAA (immediately after codon for amino acid 121) was lost. Read through occurred until an additional TAA or termination codon was reached. This next termination codon was immediately after the codon for amino acid 145 (Figure 36.1) [1]. This led to the syntheses of a 145 amino acid hCG -subunit on CG compared to the 121 amino acid -subunit originally on LH. This 145 amino acid C-terminal extension is critical
Figure 36.1 Deletion mutation in the -subunit of LH (DNA codon corresponding to residue 114), with the evolution of early simian primates causing the evolvement of CG [1].
Evolution of hCG, Evolution of Humans, and Evolution of Human Pregnancy Disorders and Cancer
361
to the biological activity of CG. With the evolution of advanced simian primates and humans, the hCG -subunit gene became duplicated, such that today, humans have as many as seven or eight back-to-back copies of the hCG-subunit gene [5]. In early simian primates, the CG -subunit gained two O-linked oligosaccharides on the new C-terminal peptide sequence in addition to the single N-linked oligosaccharide inherent to its LH -subunit origin. While the LH -subunit was produced in pituitary gonadotrope tissue, CG was produced by placental or trophoblast tissues. In pituitary tissue, oligosaccharides were sulfated, but in placental tissue they were terminated with very acidic sialic acid sugar residues (Figure 36.2). In advanced
Figure 36.2 Glycosylation units attached to CG in early simian and advanced simian primates and humans. Illustrated negative charges on N- and O-oligosaccharides are due to sialic acid, principal source of acidic pI.
362
Human Chorionic Gonadotropin (hCG)
simian primates, mutations occurred in the C-terminal extension (additional Ser residues). CG -subunit gained a third acidic O-linked oligosaccharide (Figure 36.2). With the evolution of humans, further mutations occurred in the C-terminal extension (additional Ser residues), and the CG -subunit gained a fourth acidic O-linked oligosaccharide. With further human evolution mutations (additional Asn residue), the CG -subunit also gained a second N-linked oligosaccharide (Figure 36.2) [3]. Starting with early simian primates, the entire evolutionary story is centered on the generation of human CG—an ultrapotent form of LH produced by the placenta. Root placenta cells are cytotrophoblast cells that fuse to form multinucleated syncytiotrophoblast cells (Chapter 3). Due to the presence of specific N-acetylglucosaminyltransferases, cytotrophoblast cells glycosylate molecules differently than syncytiotrophoblast cells [11]. As such, the cytotrophoblast cells of simian primates make a CG-H [12]. CG-H has large, double-sized oligosaccharide structures that are produced in cytotrophoblast cells (see Chapter 6). Because CG-H binds a different receptor [15,16], it seems to have functions completely different from those of regular CG. CG-H promotes pregnancylike invasion and growth and malignancy like choriocarcinoma. Regular CG functions like LH, promoting corpus luteal progesterone production for 3 weeks in very early pregnancy. Regular CG also has critical functions in differentiating trophoblast cells to form villi, in promoting uterine spiral artery growth to provide fetal nutrition, and in uterine development throughout the course of pregnancy [13,14]. The two molecules, CG and CG-H, seemingly work together in pregnancy to promote invasion at implantation, uterine growth, villi formation, and provide the nutrition needed from maternal spiral arteries. Humans evolved a CG and CG-H with long circulating half-lives and high biological potency, which developed into the ultraefficient nutritional transfer system we have today. As illustrated in Figure 36.2, the oligosaccharides (3, 4, and 6 oligosaccharides on the -subunit) added to the evolving CG molecule generated increasingly acidic molecules. This ranged from pI 8.0 on the originating LH molecule, to pI 6.3 on early simian CG, to pI 4.9 on advanced simian CG, and finally to pI 3.5 on human CG [3,6–8] (Figure 36.2, Table 36.1). The result was that LH (pI 8.0) had a circulating half-life of 25 min in humans, yet CG (pI 3.5) had a circulating half-life of 36.7 h [8–10]. Extending the circulating half-life extends the biological activity of a molecule [9,10], so human CG has 88 times greater potency than LH. This is why women are given hCG and not LH to promote ovulation in IVF clinics.
36.2 CG, CG-H, and the Evolution of Humans Figure 36.3 shows pictures of a prosimian primate (lemur), an early simian primate (cebus monkey), an advanced simian primate (orangutan), and a human. Very clearly, the larger cranium or brain in the orangutan and the exceptionally larger cranium or brain in the human sets them apart. We start the issues of CG, CG-H, and the evolution of humans by considering the craniums or brains of primates. In most mammals, brain weights average 0.1% of total body weight (Table 36.1). For example, a whale, horse, elephant, cow, or sheep brain (0.08%, 0.11%, 0.14%, 0.09%, and 0.15%, respectively)
Evolution of hCG, Evolution of Humans, and Evolution of Human Pregnancy Disorders and Cancer
363
Table 36.1 Parallelisms between Placental Implantation and Invasion Characteristics, Presence and Structure of CG and CG-H, and Relative Brain Masses [3,6–10,17–20]. Species
Placentation Characteristics
Depth of Implantation in Uterus
CG Structure Produced pI (Isoelectric Point)
Brain Mass (% of Body Mass)
Hemochorial
Through decidua 4 O-linked, pI 3.5 to 30% myometrium Through decidua 3 O-linked, pI 4.9 to 10% myometrium
2.4%
Through decidua 2 O-linked, pI 6.3 only
0.17%
No invasion
No CG or CG-H produced
0.07%
No CG or CG-H produced No CG or CG-H produced No CG or CG-H produced No CG or CG-H produced No CG or CG-H produced
0.08%
A. Primates Human
Advanced simian Hemochorial primate (e.g., baboon or orangutan) Hemochorial Early simian primate (e.g., cebus monkey or platyrrhine) Prosimian primate Epitheliochorial (e.g., lemur)
0.74%
B. Nonprimate mammals Whale
Epitheliochorial
No invasion
Horse
Epitheliochorial
No invasion
Elephant
Epitheliochorial
No invasion
Cow
Synepitheliochorial No invasion
Sheep
Synepitheliochorial No invasion
0.11% 0.14% 0.09% 0.15%
Figure 36.3 From the left: ringtail lemur (prosimian primate), cebus monkey (early simian primate), orangutan (advanced simian primate), and human. Clearly the biggest difference is larger cranial cavity in orangutan and much larger cranial cavity in human.
364
Human Chorionic Gonadotropin (hCG)
averages around 0.1% of body weight average. Compared with other primates and other mammals, humans have a uniquely large brain to body mass ratio [17–23] (Table 36.1). In the earliest primates, the prosimians, the brain is tiny and just 0.07% of body weight. The evolution of larger brains began with an expansion of the brain to 0.17% of body weight in early simian primates, then to 0.74% of body weight with advanced primates, and finally to an extreme 2.4% of body weight with humans. The greater brain size of advanced primates and the still greater brain size of humans correlate with the large energy demand or nutritional demands of the developing fetuses [17–23] and interestingly, the evolution of CG and CG-H parallels the evolution of the brain. CG and CG-H first appeared with early simian primates and was followed by the lowering of CG pI and extension of CG circulating half-life. This led to the generation of an increasingly efficient fetal nutrient transfer system. In humans, .60% of the fetal energy and nutrient intake during pregnancy is used to support the developing brain’s nutritional needs; other primates and mammalian species use approximately 20% [17–23]. It is the development of an ultraefficient CG and CG-H that gave humans the required nutrient system that developed the large brain. Numerous studies support the concept that advanced primates and, to an even greater extent, humans have evolved unique and increasingly efficient placental– fetal nutrition transfer mechanisms to support the demands of the increasingly larger embryonic brains (Table 36.1) [17–22,25,26]. How did this occur? How did humans evolve the extremely efficiency fetal nutrition system needed to support the development of a large embryonic brain? As it happens, the human brain is approximately 11% of the newborn body mass [24]. As discussed earlier, CG and CG-H evolved with simian primates. As they evolved, they gained additional oligosaccharide units. The -subunit from LH evolved into a -subunit with one acidic oligosaccharide, to CG -subunit with three acidic oligosaccharides in an early simian, then to CG -subunit with four acidic oligosaccharides in a baboon, and finally to CG -subunit with six acidic oligosaccharides in humans. The acidity of the molecules increased from pI 8.0 to pI 6.3, to pI 4.9, and finally to pI 3.5, respectively [3,6–8]. The effective biological potency of CG increased 88-fold from LH to simian CG to human CG [3,6–8]. For prosimian primates and earlier mammals that existed prior to simian primates, placental transfer of nutrients was not very efficient. As shown in Figure 36.4, the epitheliochorial placentation process in these animals involves a placenta attaching to an uninvaded uterine wall. Nutrients would have to travel across layers of maternal connective tissue, endometrial epithelial tissue, chorionic epithelial tissue, and fetal connective tissue in order to reach the fetal circulation. As detailed in Chapter 3 of this book, CG and CG-H promoted formation of an alternative placentation system called hemochorial placentation that has a minimal one syncytiotrophoblast cell barrier between maternal and fetal circulations. Hemochorial placentation also evolved parallel to CG in simian primates [1,3,19,27,28]. CG-H promoted the invasion of trophoblast cells deeper and deeper into the uterus (Chapter 3). Regular CG promoted the fusion of trophoblast cells to syncytiotrophoblast cells [29]. This fusion is critical to the formation of placental villi and the generation of floating villi, both of which are responsible for an efficient
Evolution of hCG, Evolution of Humans, and Evolution of Human Pregnancy Disorders and Cancer Endometrial artery
Decidua
En
365
Endometrial epithelial cells
do
m
et
Endometrial artery
ria
lg
la
nd ve
ecti
con
lls
ue tiss
ic
n rio
lia
he
it ep
o
Ch Fetal venis
Fetal venis
e lc
e
su
tis
ec
nn
Co
e tiv
Umbilical cord
Figure 36.4 Epitheliochorial implantation in lemurs. Nutrients have to be transferred across maternal connective tissue, endometrial epithelial cells, chorionic epithelial cells, and fetal connective tissue cells to reach fetal circulation.
transfer system. Once the floating villi are flooded by maternal blood, the transfer system uptakes the maternal nutrients (Chapter 3). Regular CG also promoted development of the maternal uterine blood supply, which supports hemochorial placentation and growth of the uterus [12,13]. CG and CG-H initially generated a more efficient nutrient transfer system in simian primates. With the evolution of the potency of CG and CG-H, an effective system developed in early simian primates. This system then evolved into an efficient system in advanced simian primates, and finally into an ultraefficient system in humans. This puts into place the fetal nutrition system needed to support the development of the human embryonic brain. As illustrated in Figure 36.5 and Table 36.1, villous trophoblast cells are only implanted in the outer decidua in early simian species. This permits only limited maternal bleeding and inefficient nutrient transfer. In advanced simian species, villous trophoblast cells implant deeper, traveling through the decidua and approximately 10% of the thickness of the myometrium. In humans, placental growth and deep implantation of villous trophoblast cells settle as deep as one-third the thickness of the myometrium. This permits full maternal blood flow onto the floating villi and the most efficient nutrient transfer process. These extreme nutrition demands allow the human brain to develop, but this is just a small part of human brain evolution. How the brain generated human logic, affection, and reasoning are different questions. Recently, hCG and an hCG receptor were discovered in the fetal brain [32]. Could CG also have played a part in human brain development? Everything appears to correlate: brain size, evolution of CG and CG-H, increasing biological activity of CG and CG-H, deeper and deeper implantation, and greater and
Figure 36.5 Diagram of ultraefficient hemochorial placentation in humans (A), effective hemochorial placentation in advanced simian primates (e.g., baboon, orangutan) (B), and inefficient hemochorial placentation in early simian primates (e.g., cebus monkey or platyrrhine) (C).
Evolution of hCG, Evolution of Humans, and Evolution of Human Pregnancy Disorders and Cancer
367
greater efficiency in hemochorial placentation (Table 36.1). The evolution of CG, CG-H, and hemochorial placentation are linked together and form the roots of evolution of humans. The development of CG clearly offers an explanation of how humans evolved [30,31].
36.3 C G Evolution and the Development of Pregnancy Disorders Human evolution took the biological activities of CG and CG-H, and placentation and fetal nutrition methods to extremes. It was only by moving to such extremes that the very large human brain was able to develop. Now we must examine the consequences of these extreme evolutionary strides. For instance, humans have an extremely high preponderance of failing pregnancies. Pregnant women get preeclampsia in pregnancy, which is a condition unique to humans. In addition, only humans get invasive gestational trophoblastic diseases. Of the greatest concern are human cancers. They express a hyperglycosylated -subunit, which, like hCG-H, might act on an autocrine receptor to block apoptosis and further promote malignant cell growth and invasion or metastases (see Chapter 14). All of these problems appear to have evolved in humans with the generation of ultrapotent CG and CG-H. As shown in Figure 36.6, a total hCG concentration with >50% CG-H is needed on the day of pregnancy implantation in order for a human pregnancy to extend to
Figure 36.6 CG-H as a proportion of total hCG in 42 term pregnancies (clear circles), 20 first-trimester spontaneous abortion pregnancies (black circles) [31].
368
Human Chorionic Gonadotropin (hCG)
term. Of 42 term pregnancies, all 42 had >50% CG-H; the majority of these term pregnancies had 95–100% CG-H (Figure 36.6) [33]. If CG-H was not adequately produced, biochemical pregnancy or spontaneous abortion consistently occurred [33]. It was determined that 65% of pregnancy failures produced 50% hCG-H. This corresponds with the 67% of failing pregnancies shown to not implant appropriately [34,35]. The remaining pregnancies apparently implanted properly but later miscarried because of genetic or immune reasons [34,35]. As shown in Table 36.2, 40–45% of human pregnancies fail. This extremely high failure incidence among humans has to be an attribute of their evolutionary demands on sufficient CG-H in the implantation process. For a species with less extreme invasive hemochorial placentation or CG-H demands, the failure rates are considerably lower. For example, 12% of pregnancies among baboons or advanced simian primates fail, 10% of pregnancies fail in cebus monkeys or early simian primates, and 6% fail among primitive prosimian monkeys or lemurs [26,30,36]. When the human failure rate is compared to the failure rates of other primate species, it is inferred that human pregnancies fail because of inadequate supply of hCG-H, which is essential for successful implantation and nutritional support. Some humans have difficulties meeting these CG-H implantation demands. Experiments should be conducted in which women are administered human CG-H on the day of pregnancy implantation to ensure appropriate pregnancy implantation and avoid biochemical pregnancies and miscarriages. Hypertensive pregnancy disorders like preeclampsia are unique to humans (Table 36.2) [37–42]. Preeclampsia arises from inadequate hemochorial implantation and leads to anoxia and nutritional limitations [37–41]. Maternal blood pressure rises to accommodate these inadequacies, thus causing hypertensive disorders. As demonstrated, forthcoming preeclampsia is marked by unusually low blood concentrations of CG-H at the end of the first-trimester pregnancy (39, two other studies submitted). In fact, CG-H is an outstanding marker for predicting women likely to generate preeclampsia (39, two other studies submitted). It appears that preeclampsia might be a complication of CG-H and invasion efficiency. If women were given injections of CG-H at the time of pregnancy implantation and at the end of the first-trimester of pregnancy, we wonder whether failing pregnancy and preeclampsia could be prevented.
36.4 hCG and Cancer In humans only, cytotrophoblast cells transform at parturition or following evacuation of a hydatidiform mole and choriocarcinoma or a gestational trophoblastic neoplasm occurs as a result. These are some of the most malignant and rapidly growing forms of cancer. Choriocarcinoma is similar to pregnancy implantation in that it produces 50–100% CG-H. Just as CG-H drives the invasion at implantation of pregnancy, it also drives invasion in choriocarcinoma, first into invading the full width of the uterus and then into invading the body, the liver, the lungs, and the brain [15,16]. The evolution of CG-H permits a deep implantation of pregnancy and efficient placentation. This same molecule can drive invasion of a malignancy or choriocarcinoma in
Evolution of hCG, Evolution of Humans, and Evolution of Human Pregnancy Disorders and Cancer
369
Table 36.2 Proportion hCG-H and Possible Origin of Uniquely Human Obstetric Disorders. Species
Depth of Invasion in Uterus
hCG-H
Requirement for hCG-H
Pregnancy Disorders
Human
Through decidua to 30% myometrium
Essential
Advanced simian primate (e.g., baboon or orangutan)
Through decidua to 10% myometrium
Essential
Early simian primate (e.g., cebus monkey or platyrrhine)
Through decidua only
Necessary
Not produced
No
Preeclampsia 5–10%, SAB biochemical 40–45%, hydatidiform mole 1/900, choriocarcinoma 1/24,096 Preeclampsia 0%, SAB biochemical 12%, hydatidiform mole 0, choriocarcinoma 0 Preeclampsia 0%, SAB biochemical 10%, hydatidiform mole 0, choriocarcinoma 0 Preeclampsia 0%, SAB biochemical 6%, hydatidiform mole 0, choriocarcinoma 0
Prosimian primate No invasion (e.g., lemur)
humans, which is an unfortunate consequence. This is unique to humans; choriocarcinoma or gestational trophoblastic neoplasms do not occur in any other primates with less potent forms of CG-H. As discussed in Chapter 14, the free -subunit of CG is hyperglycosylated, like CG-H. Free -subunit has functions that are very similar to those of CG-H. It is an autocrine, acting on the same cells that produce it, and it promotes apoptosis and cell invasion or malignancy [41–45]. Human cancer cells retrodifferentiate and produce free -subunit [46–52]. This advances cancer cell growth and malignancy [41–46]. Fortunately, we are now developing hCG/free -subunit vaccines; these are significantly aiding the treatment of human cancers [53,54]. All told, evolution has not made humans a perfect species. CG, CG-H, and hemochorial placentation had to be taken to extremes in order for humans to evolve into what we are today. As a result of these extremes, we have been left with major complications, pregnancy failures, preeclampsia, choriocarcinoma, and more aggressive cancers.
370
Human Chorionic Gonadotropin (hCG)
36.5 Personal Thoughts The findings and theories discussed in this chapter are somewhat extraordinary [30,31]. The human molecules CG and CG-H appear to have had critical roles in human evolution. It was seemingly just a single point mutation that created CG and CG-H. In turn, the evolution of these molecules permitted the development of intelligent human life. What’s more bizarre is that these same molecules also have critical roles in human birth, controlling implantation of pregnancy and miscarriage of pregnancy; and in human death, modulating cancer cell growth and cancer invasion. One might wonder if these could be God’s extraordinary molecules. How, as biblically conceived, God facilitated human evolution or human creation, and how God could influence human life (as in child birth) and human death (as in cancer)? As taught by the Bible, God holds every person’s story in the book of life. This is a science article, and further discussion of this issue here is not appropriate. The relationship between hCG, hCG-H, and human creation, birth, and death might be point of much discussions in both religious and scientific circles for many years to come.
References [1] Fiddes JC, Godman HM. The cDNA of the -subunit of human chorionic gonadotropin suggests evolution of a gene by readthrough into the 3 untranslated region. Nature 1979; 281:351–6. [2] Talmadge K, Vamvakopoulos NC, Fiddes JC. Evolution of the genes for the -subunits of human chorionic gonadotropin and luteinizing hormone. Nature 1984;307:37–40. [3] Maston GA, Ruvolo M. Chorionic gonadotropin has a recent origin within primates and an evolutionary history of selection. Mol Biol Evol 2002;19:320–35. [4] Hallast P, Rull K, Laan M. The evolution and genomic landscapes of CGB1 and CGB2 genes. Mol Cell Endocrinol 2007;260:2–11. [5] Policastro P, Ovitt CE, Hoshina M, Fukuoka H, Boothby MR, Boime I. The -subunit of human chorionic gonadotropin is encoded by multiple genes. J Biol Chem 1983; 258:11492–9. [6] Bambra CS. Purification and properties of baboon chorionic gonadotropin. J Reprod Fertil 1987;19:421–30. [7] Crawford CRJ, Tegear GW, Niall HD. The nucleotide sequence of baboon chorionic gonadotropin -subunit genes have diverged from the human. Gene 1986;46:161–9. [8] Baldwin DM, Highsmith RF, Ramey JW, Krummen LA. An in vitro study of LH release, synthesis and heterogeneity in pituitaries from proestrous and short-term ovariectomized rats. Biol Reprod 1986;34:304–15. [9] Gam L-H, Latiff A. SDS-PAGE electrophoretic property of human chorionic gonadotropin (hCG) and its -subunit. Int J Biol Sci 2005;1:103–9. [10] Nisula BC, Wehmann RE. Distribution, metabolism, and excretion of human chorionic gonadotropin and its subunits in man. In: Segal SJ, editor. Chorionic gonadotropin. New York: Plenum Press; 1980. p. 199–212. [11] Endo T, Nishimura R, Mochizuki M, Kochibe N, Kobata A. Altered glycosylation is induced in both - and -subunits of human chorionic gonadotropin produced by choriocarcinoma. J Biochem 1988;103:1035–8.
Evolution of hCG, Evolution of Humans, and Evolution of Human Pregnancy Disorders and Cancer
371
[12] Kovalevskaya G, Genbacev O, Fisher SJ, Cacere E, O’Connor JF. Trophoblast origin of hCG isoforms: cytotrophoblasts are the primary source of choriocarcinoma-like hCG. Mol Cell Endocrinol 2002;194:147–55. [13] Zygmunt M, Herr F, Munstedt K, Lang U, OD Liang. Angiogenesis and vasculogenesis in pregnancy. Eur J Obstet Gynecol Reprod Biol 2003;110:S10–8. [14] Zygmunt M, Herr F, Keller-Schoenwetter S, Kunzi-Rapp K, Munstedt K, Rao CV, et al. Characterization of human chorionic gonadotropin as a novel angiogenic factor. J Clin Endocrinol Metab 2002;87:5290–6. [15] Cole LA. Hyperglycosylated hCG. Placenta 2007;28:977–86. [16] Cole LA, Khanlian SA. Hyperglycosylated hCG: a variant with separate functions to regular hCG. Mol Cell Endocrinol 2007;260:228–36. [17] Gibbons A. Solving the brain’s energy crisis. Science 1998;280:1345–7. [18] Robillard PY, Hulset TC, Deka CG. Preeclampsia and human reproduction. An essay of a long term reflection. J Reprod Immunol 2003;59:93–100. [19] Martin RD. Scaling of the mammalian barrier: the maternal energy hypothesis. News Physiol Sci 1996;4:149–54. [20] Cunnane SC, Herbige LS, Crawford MA. The importance of energy and nutrient supply in human brain evolution. Nutr Health 1993;9:19–35. [21] Martin RD. Human reproduction: a comparative background for medical hypotheses. J Reprod Immunol 2003;59:111–35. [22] Martin RD. Relative brain size and basal metabolic rate in terrestrial vertebrates. Nature 1981;293:57–60. [23] Pijnenborg R. The placental bed. Hypertens Pregn 1996;15:7–23. [24] Jordaan HVF. Newborn brain: body weight ratios. Am J Phys Anthropol 2005;44:2279–84. [25] Kliman HJ. Uteroplacental blood flow. The story of decidualization, menstruation, and trophoblast invasion. Am J Pathol 2000;157:1759–68. [26] Jauniaux E, Poston L, Burton GJ. Placental-related diseases of pregnancy: involvement of oxidative stress and implications in human evolution. Hum Reprod Update 2006;12:747–55. [27] Burton GJ. Early placental development. Placenta 2006;27:A2. [28] Luckett WP. Comparative development and evolution of the placenta in primates. Contrib Primatol 1974;3:142–234. [29] Shi QJ, Lei ZM, Rao CV, Lin J. Novel role of human chorionic gonadotropin in differentiation of human cytotrophoblasts. Endocrinol 1993;132:1387–95. [30] Cole LA, Khanlian SA, Kohorn EI. Evolution of the human brain, chorionic gonadotropin and hemochorial implantation of the placenta. J Reprod Med 2008;53:549–57. [31] Cole LA. hCG and hyperglycosylated hCG in the establishment and evolution of hemochorial placentation. J Reprod Immunol 2009;82:112–18. [32] Lei ZM, Rao CV, Kornyei J, Licht P, Hiatt ES. Novel expression of human chorionic gonadotropin/luteinizing hormone receptor gene in brain. Endocrinol 1993;132:2262–70. [33] Sasaki Y, Ladner DG, Khanlian SA, Cole LA. Hyperglycosylated hCG and the source of pregnancy failures. Fertil Steril 2008;89:1871–86. [34] Norwitz ER, Schust DJ, Fisher SJ. Implantation and the survival of early pregnancy. N Engl J Med 2001;345:1400–8. [35] Semprini AE, Simoni G. Not so efficient reproduction. Lancet 2000;356:257–8. [36] Jauniaux E, Gulbis B, Burton GJ. The human first-trimester gestational sac limits rather than facilitates oxygen transfer to the fetus—a review. Trophobl Res 2003;24:S86–93. [37] Burton GJ. Placental oxidative stress: from miscarriage to preeclampsia. J Soc Gynecol Invest 2004;11:342–52.
372
Human Chorionic Gonadotropin (hCG)
[38] Robillard PY, Chaline J, Chaouat G, Hulsey TC. Preeclampsia/eclampsia and the evolution of the human brain. Curr Anthropol 2003;44:130–5. [39] Bahado-Singh RO, Oz AU, Kingston JM, Shahabi S, Hsu CD, Cole LA. The role of hyperglycosylated hCG in trophoblast invasion and the prediction of subsequent preeclampsia. Prenat Diagn 2002;22:478–83. [40] Solomon CG, Seely EW. Preeclampsia—searching for the cause. N Engl J Med 2004;350:641–2. [41] Salas SP. What causes pre-eclampsia? Clin Obstet Gynecol 1999;13:141–57. [42] Butler SA, Ikram MS, Mathieu S, Iles RK. The increase in bladder carcinoma cell population induced by the free beta subunit of hCG is a result of an anti-apoptosis effect and not cell proliferation. Br J Cancer 2000;82:1553–6. [43] Iles RK. Ectopic hCG expression by epithelial cancer: malignant behavior metastasis and inhibition of tumor cell apoptosis. Mol Cell Endocrinol 2007;260:264–70. [44] Carter WB, Sekharem M, Coppola D. Human chorionic gonadotropin induces apoptosis in breast cancer. Breast Cancer Res Treat 2006;100:S243–4. [45] Iles RK. hCG expression by cervical squamous carcinoma—in vivo histological association with tumour invasion and apoptosis. Histopathologyology 2008;53:147–55. [46] Crawford RA, Iles RK, Carter PG, Caldwell CJ, Shepherd JH, Chard T. The prognostic significance of beta human chorionic gonadotropin and its metabolites in women with cervical carcinoma. J Clin Pathol 1998;51:685–8. [47] Acevedo HF, Hartstock RJ. Metastatic phenotype correlates with high expression of membrane-associated complete -human chorionic gonadotropin in vivo. Cancer 1996;78:2388–99. [48] Cole LA. New discoveries on the biology and detection of human chorionic gonadotropin. Reprod Biol Endocrinol 2009;7:1–37. [49] Cole LA. -core fragment (-core UGP or UGF). Tumor Marker Update 1994;6:69–75. [50] Nishimura R, Baba S, Hasegawa K, Kinugasa M, Okamura M, Kimura A, et al. Characterization of immunoreactive hCG beta-subunit in cultured fluids of the cell lines derived from gynecologic malignant tumors. Nippon Sanka Fujinka Gakkai Zasshi 1990;42:1471–6. [51] Marcillac I, Cottu P, Theodore C, Terrier-Lacombe MJ, Bellet D, Droz JP. Free hCG beta subunit as tumour marker in urothelial cancer. Lancet 1993;341:1354–5. [52] Delves PJ, Iles RK, Roitt IM, Lund T. Designing a new generation of anti-hCG vaccines for cancer therapy. Mol Cell Endocrinol 2007;260:276–81. [53] Iversen PL, Mourich DV, Moulton HM. Monoclonal antibodies to two epitopes of -human chorionic gonadotropin for the treatment of cancer. Curr Opin Mol Ther 2003;5:156–60. [54] Moulton HM, Yoshihara PH, Mason DH, Iversen PL, Triozzi PL. Active specific immunotherapy with -human chorionic gonadotropin peptide vaccine in patients with metastatic colorectal cancer: antibody response is associated with improved survival. Clin Cancer Res 2002;8:2044–51.
37 Summary: hCG a Remarkable Molecule
Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
37.1 hCG Extremities hCG -subunit is quite a remarkable and versatile polypeptide. It is the source of three or four independent molecules: hCG the hormone has multiple functions in controlling placental and fetal growth during pregnancy (Chapters 3, 5, and 11); hyperglycosylated hCG (hCG-H) is an autocrine that controls placental growth and invasion (Chapters 5 and 13); hCG free -subunit, an autocrine which modulates cancer cell growth and malignancy (Chapter 14); and pituitary hCG, a sulfated hormone with questionable functions (Chapters 5 and 31). hCG -subunit is the only polypeptide known to be the root of multiple independent molecules like hCG, hyperglycosylated hCG, pituitary hCG and free -subunit. In some ways, hCG is a one-of-a-kind molecule. It is the most glycosylated glycoprotein known. It is also the only known glycoprotein that can be so extremely glycosylated, with sugar accounting for 25% (in hCG) to 41% (in hCG-H) of the molecular weight (Chapters 5 and 6). hCG is coded for by at least seven separate genes, and we question why—why such multiplicity (Chapter 4)?
37.2 Structure of hCG The primary structure of hCG-related molecules (hCG, hCG-H, free -subunit, free -subunit, nicked free -subunit, and -core fragment) are well-known. Both the amino acid sequences and carbohydrate side chain sequences are now established (Chapters 5 and 6). The three dimensional structures of hCG and hCG-H, however, are still somewhat of a quandary (Chapter 10), and the effects of carbohydrate side chains on the structures still somewhat of a mystery.
37.3 Biological Functions In physiological utility, in roles in evolution, and in roles in disease, the biological functions of the hCG polypeptide group of molecules appear to be somewhat extreme. hCG has a wide variety of roles in pregnancy, including promoting progesterone production, Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00037-2 © 2010 Elsevier Inc. All rights reserved.
374
Human Chorionic Gonadotropin (hCG)
promoting uterine artery and vein angiogenesis, differentiation and fusion of trophoblast cells, abatement of uterine muscle contraction during pregnancy, suppression of immune rejection of invading foreign placental cells, and preparation of the endometrium for receipt of placental cells (Chapter 5). hCG has also been shown to have major roles in fetal brain and other fetal organ growth and development (Chapter 11). hCG-H has been shown to have separate functions, which are promoting implantation of pregnancy and growth of cytotrophoblast cells (Chapter 13). Pituitary hCG seemingly acts in the growth process of follicles and ovulation of uterine ova (Chapter 31). Free -subunit promotes the growth, invasion, and malignancy of cancer cells (Chapter 14). All told, the biological functions of this hCG polypeptide group are very diverse. hCG and hCG-H have clear roles in the development of efficient hemochorial placentation and the associated development of a large brained species, humans (Chapters 3 and 36). Humans were created with or evolved to have an extremely high incidence of pregnancy miscarriages and biochemical pregnancies when compared to other primate species. hCG-H is the apparent source of these complications. Humans uniquely get potentially fatal hypertensive disorders in pregnancy, such as preeclampsia. This can also be blamed on hCG and hCG-H action (Chapter 36). The evolution of hCG-H has led humans to develop invasive trophoblast diseases (Chapters 29, 30, and 36), and the evolution of free -subunit has led to humans to develop advanced malignancies (Chapter 14). The hCG polypeptide group of molecules appears to have wide-reaching biological and clinical functions.
37.4 Detection of hCG Polypeptide Group of Molecules Measurement of the hCG polypeptide group of molecules has innumerable diagnostic functions. hCG measurement marks pregnancy, pituitary hCG production in menopausal cases, hydatidiform mole cases, and quiescent gestational trophoblastic disease (Chapters 29, 30, and 31). hCG-H measurements are used to detect early pregnancies (Chapter 26). The hCG-H assay is a test used to identify failing or miscarrying pregnancies (Chapter 27), predicting Down syndrome and preeclampsia pregnancies (Chapter 28), and most of all as a test for identifying invasive gestational trophoblastic diseases or choriocarcinoma (Chapter 29). Measurements of serum and urine free -subunit and its degradation product (-core fragment) are invaluable as general markers that identify germ cell and other malignancies (Chapter 32). All told, the measurements of the hCG polypeptide group of molecules have a wide spectrum of clinical applications.
37.5 Administration of hCG Administration of hCG has wide use in infertility and IVF clinics (Chapters 15 and 16). Applications range from promoting follicle development to ovulation and poly-ovulation during the menstrual cycle (Chapters 15 and 16). Other rather illicit
Summary: hCG a Remarkable Molecule
375
applications of hCG administration include anabolism in males, as hCG promotes testosterone production in men (Chapter 17). hCG may also be considered as a dietary agent and can be administered to individuals during certain weight-loss regimens (Chapter 17); however, there is no evidence to support its use as a weight-loss agent.
37.6 Detection of hCG Outside Pregnancy There are numerous explanations for hCG polypeptide production outside of pregnancy, ranging from false positive hCG assays (Chapter 20) to quiescent gestational trophoblast disease, a benign or inactive form of gestational trophoblastic neoplasm (Chapter 30). Persistent low levels of hCG can derive from cancer or gestational trophoblastic disease neoplasm (Chapters 29, 30, and 32). Interestingly, recent studies cite familial hCG, common to an entire family, as a new and rare cause of persistent low levels of hCG (Chapter 31).
37.7 Synthesis and Degradation of hCG The synthesis of hCG, hCG-H, and free -subunit is somewhat complicated. In response to -subunit and -subunit mRNA, the peptides are produced in the endoplasmic reticulum. During synthesis, three early critical disulfide bridges are added to the -subunit (Chapters 5 and 10), and root Dolichol Man-9 Glc-3 N-linked oligosaccharides are added to - and -subunits. The Man-8 Glc-3 N-linked oligosaccharides are degraded to a Man-5 root oligosaccharide (Chapter 8). N- and O-linked oligosaccharide processing, disulfide bridge formation, and combination are completed in the Golgi apparatus (Chapters 5 and 8). Excess -subunit and/or excess -subunit are alternatively processed in the Golgi apparatus, where they receive larger or more complex N-linked oligosaccharides (Chapter 7). The larger oligosaccharides prevent the free subunits from combining with hCG or hCG to form dimer (Chapter 7). Free -subunit can form – dimers (Chapters 7 and 14). The degradation and clearance processes of hCG and hCG-H begin upon secretion. Macrophages and monocyte human leukocyte elastase cleave or nick the -subunit and -subunit (Chapter 9). With time, nicking leads to cleavage to the -subunit C-terminal peptide. hCG and hCG-H continually undergo dissociation to generate free subunits in the circulation. Free -subunit is more rapidly nicked than intact hCG or hCG-H (Chapter 9). Finally, the liver and kidneys clear hCG, hCG-H, and their degradation and dissociation products from the circulation. In the kidneys, nicked degradation products are further cleaved by exo- and endoproteinases to form -core fragments, which is the terminal degradation product (Chapter 9).
38 hCG and the Future Laurence A. Cole USA hCG Reference Service, Albuquerque, NM, USA
At the time Robert Hussa wrote his comprehensive book on hCG in 1987, nobody could have predicted that the hCG field would have changed as much as it has over the past 33 years. As highlighted by Robert Hussa in Chapter 1 in this book, the number of citations regarding hCG has grown exponentially year by year. In 1987, hCG was one molecule; a hormone produced by the placenta with a single function in promoting progesterone production by corpus luteal cells. We now believe hCG exists as three or four separate molecules, each with independent hormonal, autocrine, and paracrine functions. Can we predict what will be known about hCG in a similar 33-year period by 2043? We might discover new variants with further functions. Of particular interest would be discovering an independent function for free -subunit. In general, scientific knowledge expands twofold every 10 years. Therefore, it is inevitable that a great deal of new information on hCG will emerge over the next three decades, and it would be foolish to attempt to predict what these new discoveries might be. It is tantalizing, however, to speculate based on what we now know and to attempt to answer the numerous questions raised in this text. In the final chapter of this book, we take the liberty of looking into this future. Recently, new receptor sites for hCG have been found in fetal organs and the placenta and on the uterus. Is there significance to these findings? If so, will they reveal new biological functions of this hormone? Is it involved in the growth and development of fetal organs and the fetal brain? Does hCG alone control placental growth and function? This information might be critical to our understanding of the hormone hCG. The future might see hCG measurement in assessment of pregnancy progress, and administration of hCG to promote feto-placental development. Today, we know that hyperglycosylated hCG (hCG-H) somehow promotes placental invasion. Based on the results of multiple research studies, we speculate that hCG-H antagonizes a TGF receptor. This has to be confirmed. How is hCG-H action controlled and modulated? We also must determine how and whether it promotes metalloproteinases and collagenases and other invasive enzymes. How does it work with associated cytokines? Are there methods for blocking hCG-H actions and preventing invasion in gestational trophoblastic diseases? In the future, we might see hCG-H or its derivatives being used to prevent biochemical pregnancies and miscarriages, and hCG-H administration during the course of pregnancy to prevent onset of hypertensive disorders.
Human Chorionic Gonadotropin (hCG). DOI: 10.1016/B978-0-12-384907-6.00038-4 © 2010 Elsevier Inc. All rights reserved.
378
Human Chorionic Gonadotropin (hCG)
Free -subunit has a role in human malignancies. The role and the biology of free -subunit need to be fully understood. How can we control free -subunit expression to control human malignancies? Can we extend lives by inhibiting gene expression? There has been some success using hCG vaccines in treating advanced human malignancies. We hope the day will come in which the treatment of numerous advanced human malignancies can be aided by hCG vaccines. New total hCG tests that equally detect all hCG variants are needed. When Hussa’s book on hCG was published in 1987, the only hCG test at that time was the radioimmunoassay. Today, the machine-automated immunometric assay is the current choice of tests. Over the next 33 years, we hope to see the development of new tests other than immunometric assays. New tests might equally recognize hCG, hCG-H, and other hCG-related molecules. New standards are urgently needed for hCG-H, for pituitary hCG, and for hCG and free -subunit missing the -subunit C-terminal peptide. These new standards will aid in the development of new assays. The pituitary produces a sulfated form of hCG during the menstrual cycle and menopause. Could this have independent functions? This area of research has not yet been explored but might teach us new information regarding the roles of hCG during the menstrual cycle and menopause. We hope it will be instigated in the years to come. Currently, 3D structure of hCG is limited to X-ray crystallography. For example, only deglycosylated hCG has been examined because of the limitations of N- and O-linked oligosaccharides in crystal formation. The oligosaccharides on hCG, and particularly on hCG-H, might dramatically affect our understanding of the molecule’s structure. With time, we hope new procedures like electron microscopy will evolve and allow us to examine the complete 3D structures of whole hCG, whole hCG-H, and whole hyperglycosylated free -subunit. It is only with this information that we will truly understand each of the molecule’s functions. We also hope new antibodies and assays for hCG-H will be developed. These tests would be useful in pregnancy management and management of gestational trophoblastic diseases, and in accurately predicting biochemical pregnancies, pregnancy failures, and hypertensive disorders during pregnancy. We still do not fully understand how hCG gene expression is controlled. It is only with this understanding that the meaning of hCG expression can be fully appreciated. Over the past 50 years, the discoveries made in the hCG field have opened doors to other hCG-related discoveries. We hope this will continue for many years to come.