Microbial Conversions Of Raw Glycerol

MICROBIAL CONVERSIONS OF RAW GLYCEROL

No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form or by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in rendering legal, medical or any other professional services.

MICROBIAL CONVERSIONS OF RAW GLYCEROL

GEORGE AGGELIS EDITOR

Nova Biomedical Books New York

Copyright © 2009 by Nova Science Publishers, Inc.

All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted in any form or by any means: electronic, electrostatic, magnetic, tape, mechanical photocopying, recording or otherwise without the written permission of the Publisher. For permission to use material from this book please contact us: Telephone 631-231-7269; Fax 631-231-8175 Web Site: http://www.novapublishers.com NOTICE TO THE READER The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained in this book. The Publisher shall not be liable for any special, consequential, or exemplary damages resulting, in whole or in part, from the readers’ use of, or reliance upon, this material. Any parts of this book based on government reports are so indicated and copyright is claimed for those parts to the extent applicable to compilations of such works. Independent verification should be sought for any data, advice or recommendations contained in this book. In addition, no responsibility is assumed by the publisher for any injury and/or damage to persons or property arising from any methods, products, instructions, ideas or otherwise contained in this publication. This publication is designed to provide accurate and authoritative information with regard to the subject matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in rendering legal or any other professional services. If legal or any other expert assistance is required, the services of a competent person should be sought. FROM A DECLARATION OF PARTICIPANTS JOINTLY ADOPTED BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A COMMITTEE OF PUBLISHERS. Library of Congress Cataloging-in-Publication Data

ISBN 978-1-61728-015-3 (E-Book) Available upon request

Published by Nova Science Publishers, Inc.Ô New York

Contents Preface

vii

Chapter I

Glycerol Waste from Biodiesel Manufacturing Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

1

Chapter II

Pathways to Aerobic Glycerol Catabolism and their Regulation Stylianos Fakas, Anna Makri, Stamatina Bellou and George Aggelis

9

Chapter III

Citric Acid Production from Raw Glycerol by Yarrowia Lipolytica Wratislavia 1.31 Anita Rywińska and Waldemar Rymowicz

19

Biodiesel By-Products Used as Substrates for Oxalic Acid Production by Aspergillus Niger Izabela Musiał and Waldemar Rymowicz

31

Chapter IV

Chapter V

Production of Omega-3 Polyunsaturated Fatty Acids from Biodiesel-derived Crude Glycerol by Microalgal and Fungal Fermentation Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

Chapter VI

Biotechnological Production of Bioplastics from Raw Glycerol G. Mothes and O. Otto

Chapter VII

Single Cell Oil and Gamma-linolenic Acid Production by Thamnidium elegans Grown on Raw Glycerol Stylianos Fakas, Stamatina Bellou, Anna Makri and George Aggelis

Chapter VIII

Chapter IX

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids by Fungi F. Mantzouridou Characterization of Microbial Biomass Production from Glycerin Waste by Various Yeast Strains Piotr Juszczyk and Waldemar Rymowicz

41 65

85

101

vi Chapter X

Chapter XI

Index

Contents Microbial Conversion of Glycerol into 1,3-propanediol: Glycerol Assimilation, Biochemical Events Related with 1,3-Propanediol Biosynthesis and Biochemical Engineering of the Process Seraphim Papanikolaou Capital and Manufacturing Cost Estimation of a Bioprocess Converting Raw Glycerol to 1,3-Propanediol Anastasia A. Apostolakou, Ioannis K. Kookos and Apostolis A. Koutinas

137

169

183

Preface Glycerol is a simple carbon source than can be easy assimilated by numerous eukaryotic and prokaryotic microorganisms. However, this compound had been neglected as substrate for microbial fermentations for many years due mainly to its high cost. Nevertheless, recent developments in the fuel market, led to the production in large scale of biodiesel derived from vegetable oil, reversed this situation. Currently glycerol is available in large quantities in the biodiesel production units. This industry produces enormous quantities of glycerol as a by-product, during transesterification process of the oil with methanol or ethanol. Therefore, glycerol turns into an attractive feedstock. Biotechnology provides a broad range of methods for the valorisation of raw glycerol. Important commodities and high-added value products of industrial interest, such as organic acids, bioplastics, polyunsaturated fatty acids, carotenoids etc could be produced from raw glycerol. Glycerol can also be recycled after its conversion into microbial triacylglycerols and then to biodiesel. In this book, the biochemical pathways of glycerol metabolism in prokaryotic and eukaryotic cells cultivated under various conditions are discussed. The various methods, proposed so far in the international literature for the valorisation of glycerol by biotechnological means, are shown and discussed by using scientific and technological criteria. Evaluation of the economic viability of the 1,3-propanediol production processes is also presented. All chapters (research articles and reviews) were reviewed by experts. Although the examples presented are, doubtless, the more important, there are also a lot of other applications based on raw glycerol. Chapter I - Crude glycerol represents the major byproduct of the biodiesel industry. Currently, biodiesel production in the United States is experiencing a rapid expansion; as a result, the market is being flooded with excess crude glycerol. Due to various impurities contained in the crude glycerol, it is not cost-effective to purify this waste stream for use in the food, pharmaceutical, or cosmetics industries. Various alternative methods are being developed for utilizing this crude glycerol. If this waste stream can be utilized economically, the biodiesel production process will become more profitable and more prevalent. Chapter II - The catabolic glycerol pathways have long been elucidated, and the regulatory properties of the enzymes involved in the major pathways have been studied in some detail. The advent of molecular biology allowed for the identification and

viii

George Aggelis

characterization of the genes coding for the enzymes of the major catabolic pathways. Characterization of the glycerol genes, still in its infancy, produced rather confusing results and thus many of these findings are subject to revision. Chapter III - The effects of different kinds of raw glycerol on yeast growth and citric acid biosynthesis in fed-batch cultures were studied. The following types of raw glycerol were used: purified and un-purified raw glycerol from rape seed methyl ester production and unpurified raw glycerol from ethyl ester production. All of them were a very good carbon and energy source for citric acid production by the acetate mutant strain of Y. lipolytica Wratislavia 1.31. Salt and other impurities in raw glycerol slightly influenced the production of citric acid. The highest citric acid concentration (146 g/L) was obtained with purified glycerol from methyl ester production during a fed-batch culture lasting 148 h. In this process, the yield of citric acid (0.73 g/g) was also the highest. Chapter IV - The aim of the studies was to evaluate the dynamics and yield of oxalic acid production from biodiesel by-products, such as pure glycerol, fatty acids and glycerin waste (a mixture of glycerol and fatty acids) by Aspergillus niger XP in submerged cultivations. The comparative studies included: product yields, volumetric productivity and concentration of citric acid (an unwanted by-product). The maximum concentration of oxalic acid (55.7 gdm-3) was obtained in the medium containing 50 gdm-3 of fatty acids, which also resulted in the highest oxalate yield and volumetric productivity (1.25 gg-1 and 0.29 gdm-3h-1, respectively). In contrast, the concentration of citric acid was very low (<1 gdm-3). When the medium contained 65 gdm-3 of glycerin waste, A. niger XP produced only 42.0 gdm-3 of oxalic acid (without undesired by-products) at oxalate yield of 0.8 gg-1. The lowest oxalic acid concentrations were obtained in the medium containing pure glycerol. Chapter V - Crude glycerol represents the major byproduct of the biodiesel industry. As biodiesel production skyrockets, the market is being flooded with excess crude glycerol. Producing the omega-3 polyunsaturated fatty acids docosahexaenoic acid (DHA, C22:6, ω-3) and eicosapentaenoic acid (EPA, C20:5, ω-3) provides an opportunity to utilize this undervalued material. With many therapeutic benefits, DHA and EPA have been used as supplements in various human foods or animal feeds. Fish oil as the main source of DHA/EPA has several limitations such as undesirable taste and odor, heavy metal contamination, and potential shortage due to overfishing; thus, it is necessary to seek alternative sources to produce these two fatty acids. Our laboratory has been developing a microbial fermentation process to produce DHA and EPA from crude glycerol. The microalga Schizochytrium limacinum was used as DHA producer; the fungus Pythium irregulare as EPA producer. It was found that the major impurities contained in crude glycerol, methanol and soap, were inhibitory to algal/fungal culture, but they can be easily removed from the crude glycerol medium. The culture conditions, including medium composition and culture temperature, for S. limacinum and P. irregulare were optimized in flask cultures, with a high DHA yield of 4.91 g/L from S. limacinum and an EPA yield of 182 mg/L from P. irregulare. Overall, the works presented in this chapter show that biodieselderived crude glycerol can serve as a good carbon source for microbial production of omega3 fatty acids. Future research should focus on (1) elucidating the mechanisms of the inhibitory effects of soap on algal/fungal growth, (2) controlling fungal morphology to create

Preface

ix

the desired pellet form, (3) developing optimal fermenter cultures of the two species, and (4) implementing the process on a large scale. Chapter VI - A wide variety of bacteria are able to synthesize polyhydroxyalkanoates (PHA) as an intracellular storage compound. The best-known and most common representative one is poly (3-hydroxybutyrate) (PHB). The properties of these isolated biodegradable polymers are very similar to polypropylene. As a rule their synthesis takes place, if the multiplication is limited by nutrients other than the carbon source. Depending on the metabolism of the involved bacteria, quite different carbon substrates can be used; the most common ones are sugars. In contrast to petrochemically produced plastics PHA can be produced from renewable carbon resources, which become more and more important in respect to the conservation of finite fossil resources like mineral oil and coal and their largely neutrality with regard to the emission of CO2. The authors could show that crude glycerol, a by-product of the biodiesel production from rape, is a promising feedstock for the production of PHB. By co-feeding of appropriate precursors we also could produce copolymers of poly (3-hydroxybutyric acid-co-3hydroxyvaleric acid) with co-monomer contents of 12-25 mol% resulting in improved polymer properties. For the isolation of the PHA we used an enzymatic method. The polymer composition was not changed during this procedure. Applying an additional mechanical disruption of the bacteria by French press, the use of the expensive enzyme lysozyme could be minimized or omitted. The molecular weight of the isolated polymer (700.000 – 900.000 g/mol) is sufficiently high, to be processed by common methods of polymer industry. Chapter VII - The oleaginous Zygomycete Thamnidium elegans was grown on raw glycerol producing high amounts of single cell oil (SCO) rich in gamma-linolenic acid (GLA). Produced biomass contained more than 40% w/w lipid, which contained 7.3% (w/w) GLA, giving a GLA yield of 664.3 mg/l GLA. This yield indicates that raw glycerol is an efficient substrate for SCO production. Lipid analysis showed that at the beginning of growth the mycelium was rich in phospholipids and glycolipids plus sphingolipids, while neutral lipids accumulated as growth proceeded. Fatty acid analysis of the major lipid fractions revealed that phospholipids were rich in linoleic acid and GLA, while the other two fractions had similar fatty acid composition. Chapter VIII - Importance and fermentation technologies for the production of foodgrade carotenoids by fungi are presented with focus on raw glycerol employment. Information about fungal growth, substrate assimilation and carotenoid production parameters such as yield, selectivity, productivity and process economics are detailed. Toxicological aspects of the raw glycerol are also discussed. Chapter IX - The conversion of glycerin waste from ethyl ester biodiesel production into a protein source for animal feed was examined. In the study, six yeast strains were used for biomass production: Yarrowia lipolytica ATCC 8661, Y. lipolytica ATCC 8661 UV’1, Y. lipolytica A-101, Y. lipolytica Z, Candida robusta ATCC 60 559, and C. utilis ATCC 60 558. The glycerin waste, a main by-product of biodiesel industry, contained (w/w) 45% of raw glycerol, 44% of fatty acids, 3–4% of ethyl esters, and large quantities of potassium soaps. Since the pH of the substrate was 9.6, glycerin waste was added batchwise during yeast growth, in order to prevent the pH value from exceeding 4.0. All experiments were conducted in a stirred tank reactor, on media containing 30 g/L of glycerin waste as a substrate.

x

George Aggelis

The results of the experiments have shown that the strain Y. lipolytica 8661 UV’1 is the most suitable for biomass production from this substrate. During batch cultivation, the strain simultaneously utilized glycerol and fatty acids. The biomass yield and biomass production rate obtained with this strain were the highest at 0.89 g/g and 2.7 g/Lh, respectively. Protein concentration in the biomass varied from 26.5 to 36.5% (w/w), depending on the yeast strain used. The content of essential amino acids was in compliance with the FAO/WHO standards for fodder yeast, valine and isoleucine occurring in the yeast biomass in higher quantities. The nutritional value of the yeast obtained in a submerged culture on glycerin waste ranged between 64 and 70% according to Oser's Essential Amino Acid Index (EAAI). Analysis of technological parameters (biomass yields, volumetric biomass production rate and protein content has also revealed that the strain Y. lipolytica ATCC 8661 UV’1 is the most efficient biomass producer from glycerin waste. The results are very promising, as these findings may lead to a low-cost process of fodder yeast biosynthesis from a waste generated during biodiesel production. Chapter X - The ongoing energy crisis has resulted in increasing demands for renewable fuels in the market, and this has as an inevitable effect bio-diesel production. This situation will soon lead to the accumulation of tremendous quantities of crude – impure glycerol in every country utilizing bio-diesel. Therefore, glycerol valorization should have much to offer in the cost reduction of the overall bio-diesel production process. The most important studies that are related with the conversion of (crude) glycerol into higher added-value chemical compounds are referred to its conversion into 1,3-propanediol, a substance of noticeable importance for the chemical and the textile industry. This conversion is carried out with the aid of various prokaryotic microorganisms (principally strains belonging to the family Enterobacteriaceae, to the lactic acid group and to the genus Clostridium sp.) principally under anaerobic conditions. This chapter presents a brief survey of studies that have been carried out by various research teams (including our team – Department of Food Science and Technology of the Agricultural University of Athens and Department of Biology, Division of Genetics, Cell and Development Biology of the University of Patras) in relation with the assimilation of glycerol by bacterial strains and its conversion into 1,3-propanediol. To this end, physiological approaches related with the anaerobic assimilation of glycerol, biochemical aspects related with the biosynthesis of 1,3-propanediol and biotechnological aspects concerning the feasibility of the process in various fermentation configurations will be considered and discussed. Chapter XI - The aim of this chapter is to present the calculations of the capital cost and the total manufacturing cost of a bioprocess that is used to convert raw glycerol to 1,3propanediol. To this end a representative process flow diagram is developed based on well known heuristics. Data from the literature are used to solve the material and energy balances. The process equipment is then designed and the fixed capital cost estimated. Finally, the total manufacturing cost is estimated. The results of the economic analysis are particularly helpful in identifying research direction that will improve process economics and can contribute in transforming the results of fundamental research into successful industrial projects.

In: Microbial Conversions of Raw Glycerol Editor: George Aggelis

ISBN 978-1-60692-392-4 © 2009 Nova Science Publishers, Inc.

Chapter I

Glycerol Waste from Biodiesel Manufacturing Zhiyou Wen1,2, Denver J. Pyle1 and Sneha K. Athalye1 1

Department of Biological Systems Engineering, Institute for Critical Technology and Applied Science, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA 2.

Abstract Crude glycerol represents the major byproduct of the biodiesel industry. Currently, biodiesel production in the United States is experiencing a rapid expansion; as a result, the market is being flooded with excess crude glycerol. Due to various impurities contained in the crude glycerol, it is not cost-effective to purify this waste stream for use in the food, pharmaceutical, or cosmetics industries. Various alternative methods are being developed for utilizing this crude glycerol. If this waste stream can be utilized economically, the biodiesel production process will become more profitable and more prevalent.

1. Introduction As oil prices reach historical highs and public support for alternative fuels broadens, biodiesel as a renewable energy source has attracted increasing attention. In general, for every 100 pounds of biodiesel produced, approximately 10 pounds of crude glycerol are created. Because this glycerol is expensive to purify for use in the food, pharmaceutical, or cosmetics industries, biodiesel producers must seek alternative methods for its disposal. Various methods for disposal and utilization of this crude glycerol have been attempted, including combustion, composting, anaerobic digestion, animal feeds, and thermochemical/biological conversions to value-added products. The objective of this chapter

2

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

is to provide a general background in terms of biodiesel production and waste glycerol utilization.

2. Current Status of Biodiesel Production Biodiesel production has increased tremendously in last few years. Taking the United States as an example, the National Biodiesel Board has projected the annual U.S. biodiesel production in 2007 as 450 million gallons, a sharp increase from less than 100 million gallons prior to 2005 (NBB, 2008). On December 19, 2007, the Energy Independence and Security Act was signed into law in the U.S. Under this legislation, annual biodiesel production is supposed to increase to 1 billion gallons in 2012 (RFA 2008). Although this increment is relatively small compared to that of cellulosic ethanol, the absolute amount of biodiesel production will be increased significantly. As biodiesel production skyrockets, the market is being flooded with crude glycerol. Crude glycerol prices have dropped from 25 cents/lb in 2004 to 2.5-5 cents/lb in 2006 (Johnson and Taconi, 2007; Yazdani and Gonzalez, 2007) because the current U.S. demand for glycerol is not large enough for all of this crude glycerol. It is clear that new uses for this byproduct are needed.

3. How Biodiesel is Made Biodiesel is made through a catalyzed transesterification between oils or fats (triglycerides) and an alcohol (usually methanol) (Figure 1). Common feedstocks are pure vegetable oil (e.g., soybean, canola, sunflower), rendered animal fats, or waste vegetable oils. The theoretical ratio of methanol to triglyceride is 3:1; which corresponds to having one methanol molecule for each of the three hydrocarbon chains present in the triglyceride molecule, and is equivalent to approximately 12% methanol by volume. In practice, this ratio needs to be higher in order to drive the reaction towards a maximum biodiesel yield; 25% methanol by volume is recommended.

Figure 1. Transesterification reaction to produce biodiesel.

The catalyst can be alkalis, acids, or enzymes (e.g., lipase). The majority of biodiesel produced today is made using an alkali (such as NaOH or KOH) catalyzed reaction because

Glycerol Waste from Biodiesel Manufacturing

3

this reaction (1) requires only low temperature and pressure, (2) has a high conversion yield (98%) with minimal side reactions and a short reaction time, (3) is a direct conversion to biodiesel with no intermediate compounds, and (4) does not require specific construction materials. The glycerol backbone of the triglyceride remains as a waste product after the reaction is completed.

4. Characterizations of Glycerol Waste Crude glycerol generated from biodiesel production is impure and of little economic value. In general, glycerol makes up 60% to 85% (w/w) of the crude stream (GonzalezPajuelo et. al., 2005; Mu et. al., 2006, Pyle et. al., 2008). The wide range of the purity values can be attributed to different glycerol purification methods or different feedstocks used by biodiesel producers. For example, Thompson and He (2006) have characterized the glycerol produced from various biodiesel feedstocks. The authors found that mustard seed generated a lower level (62%) of glycerol, while soy oil had 67.8 % glycerol, and waste vegetable oil had the highest level (76.6 %) of glycerol. Methanol and free fatty acids (soaps) are the two major impurities contained in crude glycerol (Thompson and He, 2006). The existence of methanol is due to the fact that biodiesel producers use excess methanol to drive the chemical transesterification and do not recover all the methanol. The soaps, which are soluble in the glycerol layer, originate from a reaction between the free fatty acids present in the initial feedstock and the catalyst (base).i.e.,

In addition to methanol and soaps, crude glycerol also contains a variety of elements such as calcium, magnesium, phosphorous, or sulfur. Thompson and He (2006) reported that the elements present in the glycerol of different feedstock sources (such as canola, rapeseed, and soybean) were similar. Calcium was in the range of 3-15 ppm, magnesium was 1-2 ppm, phosphorous was 8-13 ppm, and sulfur was 22-26 ppm. However, when crambe (a perennial oilseed plant) was used as feedstock, crude glycerol contained the same elements, but at vastly different concentrations (Thompson and He, 2006). Schröder and Südekum (1999) also reported the elemental composition of crude glycerol from rapeseed oil feedstock. Phosphorous was found to be between 1.05 % and 2.36 % (w/w) of the crude glycerol. Potassium was between 2.20 % and 2.33%, while sodium was between 0.09% and 0.11%. Cadmium, mercury, and arsenic were all below detectable limits. The crude glycerol derived from alkali-catalyzed transesterification usually has a dark brown color with a high pH (11-12). When used in microbial fermentations, crude glycerol is

4

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

dissolved in the medium solution and the pH is usually adjusted to a neutral range. Under this condition, soaps will be converted into free fatty acids, as shown in the following equation

Figure 2. Appearance of different glycerol streams. A: pure glycerol; B: raw crude glycerol with pH at 12; C: crude glycerol after pH adjusted to 7.0; D: crude glycerol after pH adjusted to 7.0 and phase separation by centrifugation.

After pH adjustment, the free fatty acids in the crude glycerol stream resulted in a cloudy solution. After centrifugation, this cloudy solution will be separated into two clear phases, with the top layer being the free fatty acid phase, and bottom layer the glycerol phase. Figure 2 shows the appearance of pure glycerol, raw crude glycerol (with pH 12), crude glycerol after pH adjustment (pH 7), and crude glycerol after pH adjustment and phase separation by centrifugation.

5. Utilization of Glycerol Waste As a waste material, crude glycerol can be utilized through a variety of methods such as combustion (Johnson and Taconi, 2007), composting, or anaerobic digestion (Holm-Nielsen et. al., 2008). Crude glycerol has also been used as a feed additive for various animals such as pigs (Lammers et. al., 2008b), broiler chickens (Cerrate et. al., 2006), and laying hens

Glycerol Waste from Biodiesel Manufacturing

5

(Lammers et. al., 2008a). It was found that the metabolizable to digestible energy ratio of glycerol is similar to that of corn or soybean oil when fed to pigs (Lammers et. al., 2008b). Birds fed 2.5 % to 5% glycerol-diets had higher breast yield than the control group (Cerrate et. al., 2006). Crude glycerol has also been used to feed dairy cows in order to prevent ketosis, but the result was not positive (DeFrain et. al., 2004). Converting crude glycerol into valued-added products through thermo-chemical methods or biological methods is an alternative for utilizing this waste stream. It has been reported that glycerol can be thermochemically converted into propylene glycol (Alhanash et. al., 2008; Dasari et. al., 2005), acetol (Chiu et. al., 2006), or a variety of other products (Johnson and Taconi 2007). Cortright et al. (2002) have developed an aqueous phase reforming process that transforms glycerol into hydrogen. Virent Energy Systems is currently trying to commercialize this technology and claim that sodium hydroxide, methanol, and high pH levels within crude glycerol help the process (Nilles 2005). For biological conversions of crude glycerol, the glycerol serves as a feedstock in various fermentation processes. For example, Lee et al. (2001) have used glycerol in the fermentation of Anaerobiospirillum succiniciproducens for the production of succinic acid. The fermentation of E. coli on glycerol leads to the production of a mixture of ethanol, succinate, acetate, lactate, and hydrogen (Dharmadi et. al., 2006). Glycerol can also be converted to citric acid by the yeast Yarrowia lipolytica. It has been reported that this organism produces the same amount of citric acid when grown on glucose or on raw glycerol (Papanikolaou et. al., 2002). Rymowicz et al. (2006) found that acetate mutant strains of Y. lipolytica can produce high levels of citric acid while producing very little isocitrate. Furthermore, it has been shown that Clostridium butyricum can utilize biodiesel-derived glycerol to produce 1,3-propanediol (an important chemical building block with many industrial uses) in both batch and continuous cultures. During the fermentation process, the organism also produces byproducts of acetic and butyric acid (Papanikolaou et. al., 2004).

6. Conclusion Because glycerol is the major byproduct of the biodiesel manufacturing process, the disposal of crude glycerol has been a major issue faced by biodiesel producers. Research has shown that biodiesel-derived crude glycerol can be utilized to form a multitude of products. There are several thermochemical and biological methods that have been investigated as possible value-added outlets for this currently under-utilized and under-valued byproduct.

References Alhanash, A., Kozhevnikova, E.F. and Kozhevnikov, I.V. (2008). Hydrogenolysis of glycerol to propanediol over Ru: Polyoxometalate bifunctional catalyst. Catalysis Letters, 120, 307-311.

6

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

Cerrate, S., Yan, F., Wang, Z., Coto, C., Sacakli, P. and Waldroupand, P.W. (2006). Evaluation of glycerine from biodiesel production as a feed ingredient for broilers. International Journal of Poultry Science, 5, 1001-1007. Chiu, C.W., Dasari, M.A., Sutterlin, W.R. and Suppes, G.J. (2006). Removal of residual catalyst from simulated biodiesel's crude glycerol for glycerol hydrogenolysis to propylene glycol. Industrial and Engineering Chemistry Research, 45, 791-795. Cortright, R.D., Davda, R.R. and Dumesic, J.A. (2002). Hydrogen from catalytic reforming of biomass-derived hydrocarbons in liquid water. Nature, 418, 964-967. Dasari, M.A., Kiatsimkul, P.P., Sutterlin, W.R. and Suppes, G.J. (2005). Low-pressure hydrogenolysis of glycerol to propylene glycol. Applied Catalysis a-General 281, 225231. DeFrain, J.M., Hippen, A.R., Kalscheur, K.F. and Jardon, P.W. (2004). Feeding glycerol to transition dairy cows: Effects on blood metabolites and lactation performance. Journal of Dairy Science, 87, 4195-4206. Dharmadi, Y., Murarka, A. and Gonzalez, R. (2006). Anaerobic fermentation of glycerol by Escherichia coli: A new platform for metabolic engineering. Biotechnology and Bioengineering, 94, 821-829. Gonzalez-Pajuelo, M., Meynial-Salles, I., Mendes, F., Andrade, J.C., Vasconcelos, I. and Soucaille, P. (2005). Metabolic engineering of Clostridium acetobutylicum for the industrial production of 1,3-propanediol from glycerol. Metabolic Engineering, 7, 329336. Holm-Nielsen, J.B., Lomborg, J., Oleskowicz-Popiel, P. and Esbensen, K.H. (2008). On-line near infrared monitoring of glycerol-boosted anaerobic digestion processes: Evaluation of process analytical technologies. Biotechnology and Bioengineering, 99, 302-313. Johnson, D.T. and Taconi, K.A. (2007). The glycerin glut: Options for the value-added conversion of crude glycerol resulting from biodiesel production. Environmental Progress, 26, 338-348. Lammers, P.J., Kerr, B.J., Honeyman, M.S., Stalder, K., Dozier, W.A., Weber, T.E., Kidd, M.T. and Bregendahl, K. (2008a). Nitrogen-corrected apparent metabolizable energy value of crude glycerol for laying hens. Poultry Science, 87, 104-107. Lammers, P.J., Kerr, B.J., Weber, T.E., Dozier, W.A., Kidd, M.T., Bregendahl, K. and Honeyman, M.S. (2008b). Digestible and metabolizable energy of crude glycerol for growing pigs. Journal of Animal Science, 86, 602-608. Lee, P.C., Lee, W.G., Lee, S.Y. and Chang, H.N. (2001). Succinic acid production with reduced by-product formation in the fermentation of Anaerobiospirillum succiniciproducens using glycerol as a carbon source. Biotechnology and Bioengineering, 72, 41-48. Mu, Y., Teng, H., Zhang, D.J., Wang, W. and Xiu, Z.L. (2006). Microbial production of 1,3propanediol by Klebsiella pneumoniae using crude glycerol from biodiesel preparations. Biotechnology Letters, 28, 1755-1759. NBB. (2008). National Biodiesel Board. Available from: http://www.biodiesel.org/pdf_files/ fuelfactsheets/Production_Graph_Slide.pdf. (Accessed on Feburary 11, 2009) Nilles, D. (2005). A glycerin factor. Biodiesel Magazine August/September. Grand Forks, ND: BBI International.

Glycerol Waste from Biodiesel Manufacturing

7

Papanikolaou, S., Muniglia, L., Chevalot, I., Aggelis, G. and Marc I. (2002) Yarrowai lipolytica as a potential producer of citric acid from raw glycerol. Journal of Applied Microbiology, 92, 737-744. Papanikolaou, S., Fick M. and Aggelis., G. (2004). The effect of raw glycerol concentration on the production of 1,3-propanediol by Clostridium butyricum. Journal of Chemical Technology and Biotechnology, 79, 1189-1196. Pyle, D. Garcia, R. and Wen, Z. (2008). Producing docosahexaenoic acid (DHA)-rich algae from biodiesel-derived crude glycerol: effects of impurities on DHA production and algal biomass composition. Journal of Agricultural and Food Chemcistry, 56, 3933-3939. RFA. (2008). Renewable Fuels Association. Available from http://www.ethanolrfa.org/ resource/standard/ (Accessed on Feburary 11, 2009) Rymowicz, W., Rywińska, A., Żarowska B. and Juszczyk P. (2006). Citric acid production from raw glycerol by acetate mutants of Yarrowia lipolytica. Chemical Papers, 60, 301304. Schröder, A. and Südekum, K.H. (1999). Glycerol as a by-product of biodiesel production in diets for ruminants. In: Proceedings of the 10th International Rapeseed Conference. Canberra, Australia: The Regional Institute, Ltd. Thompson, J.C. and He, B.B. (2006). Characterization of crude glycerol from biodiesel production from multiple feedstocks. Applied Engineering in Agriculture, 22, 261-265. Yazdani, S.S. and Gonzalez, R. (2007). Anaerobic fermentation of glycerol: a path to economic viability for the biofuels industry. Current Opinion in Biotechnology, 18, 213219.

In: Microbial Conversions of Raw Glycerol Editor: George Aggelis

ISBN 978-1-60692-392-4 © 2009 Nova Science Publishers, Inc.

Chapter II

Pathways to Aerobic Glycerol Catabolism and their Regulation Stylianos Fakas, Anna Makri, Stamatina Bellou and George Aggelis1 Division of Genetics, Cell and Development Biology, Department of Biology, University of Patras, Patras 26504, Greece

Abstract The catabolic glycerol pathways have long been elucidated, and the regulatory properties of the enzymes involved in the major pathways have been studied in some detail. The advent of molecular biology allowed for the identification and characterization of the genes coding for the enzymes of the major catabolic pathways. Characterization of the glycerol genes, still in its infancy, produced rather confusing results and thus many of these findings are subject to revision.

Introduction Glycerol had been largely neglected as substrate for microbial fermentations, the main reason being glycerol’s high cost. Recent developments, however, reversed this situation, and nowadays glycerol is becoming one of the preferred fermentation substrates. This reversal was due to the thriving biodiesel industry, which produces large amounts of glycerol as a byproduct. Thus, glycerol’s cost became almost zero, making glycerol an attractive feedstock. Obviously, use of glycerol as fermentation feedstock requires profound knowledge of the pathways involved in glycerol assimilation. The first step of glycerol assimilation is its transport from the fermentation medium to the cell’s interior. Glycerol transport has some unique features that differentiate it from the transport mechanisms of usual microbial substrates (e.g. sugars). Glycerol can cross 1

E-mail: [email protected].

10

Stylianos Fakas, Anna Makri, Stamatina Bellou et al.

biological membranes by passive diffusion, while sugars can not (Romano, 1986). Simple diffusion though can not concentrate glycerol inside the cells, while facilitated diffusion, under certain circumstances, can. Facilitated diffusion of glycerol is the only diffusion mechanism that has been found to effect substrate uptake in many microorganisms, mostly bacteria (Eze and McElhaney, 1981; Castro and Loureiro-Dias, 1991; Truniger and Boos, 1993). In fungi, however, an active transport system operates, which is usually coupled to proton uptake (i.e. proton symport). The facilitated diffusion and active transport systems capture glycerol inside microbial cells, where glycerol enters some catabolic pathway. Microorganisms use several pathways to transform glycerol into intermediates of the glycolytic pathway. The most common pathway is the phosphorylation pathway, found in most eukaryotes (Gancedo et al., 1968; Courtright, 1975b) and many bacteria (Heath and Gaudy, 1978; Deutscher et al., 1993; Pasteris and Strasser de Saad, 2005). Next in significance is the oxidative pathway, mainly distributed among bacteria (Rush et al., 1957; Lin, 1976). Finally, both fungi and bacteria use some other minor pathways, which are essentially variations of the two major pathways (Uwajima et al., 1984). It should be stated though that more than one pathway may operate in a microorganism. All pathways employed, however, are regulated to ensure efficient glycerol assimilation. The regulation of glycerol assimilation involves both its transport and its catabolism, though the latter mechanism is the most common one. The major regulatory mechanism, employed by both fungi and bacteria, is the induction by glycerol substrate of the enzymes of the catabolic pathway (Gancedo et al., 1968; Lin, 1976; Hondmann et al., 1991). Bacteria, however, that employ phosphoenolopyruvate-phosphotransferase system (PEP-PTS) for sugar assimilation use a more complicated mechanism to regulate sugar and glycerol assimilation (Deutscher et al., 1993). In addition, Aspergillus nidulans, which transports glycerol by facilitated diffusion, exceptionally regulates glycerol assimilation by regulating its transport (Castro and Loureiro-Dias, 1991). Detailed regulatory studies, however, only appeared after the identification of the genes involved in glycerol assimilation. Identification of glycerol assimilation genes in yeast has lagged behind that of bacteria (Pavlik et al., 1993; Ronnow and Kielland-Brandt, 1993). This fact probably accounts for the limited regulatory studies of glycerol assimilation in yeast. In glycerol transport, there exists much controversy concerning the genes involved, and much of the earlier findings are under revision (Oliveira et al., 2003; Ferreira et al., 2005). In glycerol catabolism, the situation seems clearer, as several factors that affect gene expression have been identified (Grauslund et al., 1999; Grauslund and Ronnow, 2000). There are, however, many details of glycerol assimilation yet to be elucidated. In this chapter, an overall description of aerobic glycerol catabolism in bacteria and fungi will be presented. Rather than narrative, the chapter attempts to summarize our current knowledge on the major aspects of glycerol metabolism: transport and catabolism and their regulation. Glycerol catabolism in anaerobes is dealt in chapter 10.

Pathways to Aerobic Glycerol Catabolism and their Regulation

11

Glycerol Transport Though a polar molecule, glycerol can cross the cell membrane through passive diffusion. Actually, in some microorganisms like the yeast Saccharomyces cerevisiae, passive diffusion was thought as the only mechanism involved in glycerol transport (Sutherland et al., 1997). Passive diffusion does not necessitate energy expenditure, and it can only operate when there is a concentration gradient across the cellular membrane (Romano, 1986). Generally, passive diffusion predominates when concentration gradient is very high, and thus is of little physiological significance during usual growth conditions. Passive diffusion may, however, function as a complementary mechanism to other more competent transport systems, such as facilitated diffusion. Facilitated diffusion, as passive diffusion, takes place only down a concentration gradient and requires no energy (Romano, 1986). Facilitated diffusion is mediated by specific carriers, which are channel proteins, named facilitator proteins, located in cell membranes. The specificity of glycerol facilitators is not absolute, as they allow for the passage of other polyols and molecules such as urea and glycine; charged molecules, however, like glycerol-3phosphate can not pass through glycerol facilitators (Moat et al., 2002). The glycerol facilitators are regulated by membrane lipid composition, as changes in the lipid composition affect the rate of glycerol transport (Truniger and Boos, 1993; Sutherland et al., 1997). In Escherichia coli, the rates of glycerol transport by passive and facilitated diffusion increase with increasing membrane fluidity (Eze and McElhaney, 1981). The effect of cellular membrane composition on glycerol transport is not without precedent, as membrane proteins are known to be affected by both the physical state and lipid composition of the membrane (Truniger and Boos, 1993). Membrane fluidity, which mainly depends on its lipid composition, affects the conformational stability and hence the proper function of many membrane proteins, including those proteins that mediate facilitated diffusion (Truniger and Boos, 1993). Generally, transport of substrates by facilitated diffusion is rare, because most microorganisms live in environments where substrates’ concentration is low. Glycerol, however, is one of the few exceptions among usual growth substrates, as many microorganisms solely use facilitated diffusion to transport glycerol. Bacteria are the best known example of microorganisms that usually transport glycerol using facilitator proteins. Thus bacteria need an efficient mechanism to maintain the glycerol concentration gradient, even at low extracellular glycerol concentrations. This mechanism is based on the coupling of glycerol transport to its phosphorylation by glycerol kinase, so as the glycerol that enters the cells is rapidly transformed to 3-P-glycerol. This then allows for the unabated glycerol transport (Romano, 1986). Fusarium oxysporum is one the very few moulds known to transport glycerol solely by facilitated diffusion (Castro and Loureiro-Dias, 1991). The latter authors, after trying several potential competitive inhibitors of the glycerol facilitator, concluded that the facilitator is highly specific, as only 1,2-propanediol affected glycerol’s uptake (Castro and Loureiro-Dias, 1991). In most fungi, however, glycerol uptake takes place by active transport. Active systems transport glycerol against a concentration gradient, thus they require energy to operate (Romano, 1986). In S. cerevisiae, active transport results in intracellular

12

Stylianos Fakas, Anna Makri, Stamatina Bellou et al.

glycerol concentrations that are 10 times higher than the extracellular ones (Lages and Lucas, 1997). Active transport is mediated by permeases, which are highly specific membrane proteins. In S. cerevisiae and Pichia sorbitophila, a glycerol/H+ symport operates, where one proton and one mole of glycerol are co-transported inwards (Lages and Lucas, 1995; Lages and Lucas, 1997). This mechanism involves the establishment of a proton gradient across the cytoplasmic membrane by proton extrusion from cells (Romano, 1986). The energy required for proton extrusion is provided by ATP hydrolysis. Once the proton gradient is established, glycerol can be transported inwards along with a proton, which flows down its electrochemical gradient (Romano, 1986).

Glycerol Catabolism There are two major pathways involved in aerobic glycerol catabolism: the phosphorylation pathway, which is the most common, and the oxidation pathway (Figure 1). The phosphorylation pathway is generally found in eukaryotes and involves glycerol phosphorylation by glycerol kinase to yield 3-P-glycerol, which is then oxidized to 3-Pdihydroxyacetone by a NAD-linked dehydrogenase (Courtright, 1975a; Sprague and Cronan, 1977; Ermakova and Morgunov, 1988). In the oxidative pathway, glycerol is first oxidized to yield dihydroxyacetone, which is then phosphorylated to produce again 3-Pdihydroxyacetone (Rush et al., 1957; Tom et al., 1978). Generally, microorganisms use either one pathway to assimilate glycerol, but in Neurospora crassa both pathways operate (Tom et al., 1978). Some microorganisms, however, use an alternative pathway, in which glycerol is oxidized by a NADP-linked dehydrogenase to glyceraldehyde (Figure 1) (Vasiliadis et al., 1987). Glyceraldehyde may then be converted to 3-P-glyceraldehyde or glycerate, and then 3P-glycerate (Viswanath-Reddy et al., 1977). Yet another pathway for glycerol catabolism operates in strains of Aspergillus, Neurospora, and Penicillium, and involves glycerol oxidase that oxidizes glycerol to D-glyceraldehyde and hydrogen peroxide using molecular oxygen (Uwajima et al., 1984; Lin et al., 1996). The glycerol oxidase pathway has the unique feature that glycerol oxidation is not coupled to the reduction of NAD or FAD (Uwajima et al., 1984). Whatever the pathway used, glycerol is finally transformed to intermediates of the glycolytic pathway. Then, the second branch of glycolytic pathway is employed to transform these intermediates to pyruvic acid and thence to acetyl-CoA. Assuming that glycerol is catabolized by one of the major pathways, the overall stoichiometry of glycerol catabolism is:

Thus one mole of each ATP and pyruvate are produced per mole of glycerol consumed, while two moles of each ATP and pyruvate are produced per mole of glucose consumed. The produced pyruvate is further metabolized to acetyl-CoA, which is the biosynthetic precursor of various microbial products (e.g. citric acid, lipids).

Pathways to Aerobic Glycerol Catabolism and their Regulation

13

Figure 1. Major pathways of aerobic glycerol metabolism.

Regulation of Glycerol Assimilation Most studies in the regulation of glycerol’s assimilation concern bacteria that use the phosphorylation pathway. Early studies showed that glycerol kinase is cytosolic, while glycerol-3-phosphate dehydrogenase is found in cell membrane (Kistler and Lin, 1972). In most bacteria that employ the phosphorylation pathway the regulation of glycerol kinase and glycerol-3-phosphate dehydrogenase is by glycerol induction (Lin, 1976). In Pseudomonas

14

Stylianos Fakas, Anna Makri, Stamatina Bellou et al.

aureginosa, for example, both enzyme activities are induced some 20 times during growth on glycerol (McCowen et al., 1981). In bacteria that use the PEP-PTS to transport sugars, the induction of glycerol kinase is mediated by phosphorylation, which increases ninefold the enzyme activity (Deutscher et al., 1993). The presence of PEP-PTS substrates in the culture medium, however, causes glycerol kinase to donate its phosphate group to the proteins of the PEP-PTS system, which are then phosphorylated (Deutscher et al., 1993). The phosphorylation activates the PEP-PTS proteins and allows for the sugar transport to commence. This then reduces the activity of glycerol kinase, which in turn reduces the uptake and catabolism of glycerol (Deutscher et al., 1993). Studies on the regulation of glycerol catabolism in fungi are rather scarce, and most of them concern yeasts. As in bacteria, the regulation of the phosphorylation pathway enzymes has attracted most interest. In fungi, glycerol kinase is a cytosolic enzyme, while glycerol-3phosphate dehydrogenase is located in the mitochondrial membrane (Gancedo et al., 1968; Courtright, 1975b). Both enzymes are usually induced during growth on glycerol. In C. utilis, glycerol kinase activity increases sevenfold (Gancedo et al., 1968), while in N. crassa the activities of both glycerol kinase and glycerol-3-phosphate dehydrogenase increase 10 to 15fold (Courtright, 1975b). On the contrary, in Saccharomyces cerevisiae, the regulation is principally via catabolite repression by glucose (dextrose), as in glycerol-grown cells there is only a twofold increase in the activities glycerol kinase and glycerol-3-phosphate dehydrogenase (Sprague and Cronan, 1977). In Aspergillus nidulans, which contains the enzymes of both the phosphorylation and the NADP-depended pathway, a more flexible regulatory mechanism has been evolved. The induction of the enzymes implicated in glycerol catabolism is pathway-specific; that is only the enzymes of the phosphorylation pathway are induced during growth on glycerol (Hondmann et al., 1991). In addition, the latter enzymes are subject to catabolite repression by glucose (Hondmann et al., 1991). In Fusarium oxysporum, which transports glycerol solely by facilitated diffusion, the regulation of glycerol catabolism is quite distinct (Castro and Loureiro-Dias, 1991). This fungus employs the phosphorylation pathway to assimilate glycerol, but both glycerol kinase and glycerol-3-phosphate dehydrogenase are constitutive. On the contrary, the facilitator proteins are regulated by glucose repression and inactivation. This then means that the regulation of glycerol catabolism is mediated solely by the regulation of glycerol transport (Castro and Loureiro-Dias, 1991).

Genes Implicated in Glycerol Assimilation and their Regulation A good deal of work has been done to clarify the regulation of passive and facilitated diffusion of glycerol in Saccharomyces cerevisiae. Glucose grown cells of S. cerevisiae transport glycerol solely by diffusion mediated by the Fps1 product, a channel protein which, until recently, was thought as a low affinity facilitated diffusion transporter (Lages and Lucas, 1997). Interestingly, deletion of Fps1 affects membrane lipid composition, the deletion mutant having 25% less phospholipids and 62% more glycolipids than the wild type, which results in lower membrane permeability (Sutherland et al., 1997). The role of Fps1in

Pathways to Aerobic Glycerol Catabolism and their Regulation

15

glycerol transport remains dubious, as a recent report indicates that the Fps1channel is involved in passive rather than facilitated diffusion (Oliveira et al., 2003). The latter authors maintain that the earlier assumptions were based on the saturation kinetics of glycerol transport, which was in fact an artifact caused by the action of glycerol kinase that entraps glycerol in the cell (Oliveira et al., 2003). It seems, however, that the task of passive glycerol transport is not accomplished by Fps1 alone, as there is at least a Fps1 homologous gene having a somehow different regulation (Oliveira et al., 2003). Whatever the genes involved in glycerol diffusion, this mechanism acts only complementary to active transport. As with glycerol diffusion, there is been much debate about the genes involved in glycerol active transport. Early studies by Holst et al. (2000) in Saccharomyces cerevisiae suggested that two genes (named GUP1 and GUP2 after glycerol uptake) that encode membrane proteins are involved in glycerol active transport. Deletion of GUP1 slows growth on glycerol, while deletion of GUP2 does not affect the growth rate (Holst et al., 2000). Extensive expression studies of GUP1 and GUP2 showed that both genes are constitutively expressed (Oliveira and Lucas, 2004). This fact, however, contradicts earlier findings which suggested that the proteins involved in active uptake are induced by glycerol and repressed by glucose (Lages and Lucas, 1997), thus the implication of GUP1 and GUP2 in glycerol active transport has been disputed. Recently, it was suggested that the STL1 gene product, which belongs to the sugar permease family, is the glycerol/H+ symporter in S. cerevisiae (Ferreira et al., 2005). These authors based their suggestions on the following facts: STL1 deletion mutants can not actively transport glycerol, nor grow efficiently on glycerol, while STL1 expression correlates directly with glycerol uptake activity (Ferreira et al., 2005). Identification of STL1 protein as an active transporter should not come as a surprise, as yet another member of the permease family is the active transporter of maltose (Nelissen et al., 1997). Whether the STL1 glycerol/H+ symporter is of widespread occurrence in fungi remains to be seen. In most fungi, the first committed step of glycerol assimilation is its phosphorylation by glycerol kinase, which is the product of GUT1 gene (Pavlik et al., 1993). This gene was identified by Sprague and Cronan (1977) who isolated GUT1 mutants that were defective in glycerol assimilation. Conclusive evidence, however, were presented by Pavlik et al. (1993) who showed that disruption of the open reading frame of GUT1 results in mutants that do not produce glycerol kinase and hence can not grow on glycerol. The regulation of GUT1 expression depends on the carbon source and is mediated by activation and repression systems. More specific, GUT1 expression is repressed during growth on glucose and derepressed during growth on non-fermentable carbon sources, such as glycerol and ethanol (Grauslund et al., 1999). Derepression is mediated by the cis-acting transcriptional activator Adr1 and by two trans-acting transcriptional activators Ino2 and Ino4 (Grauslund et al., 1999). These activators bind to specific DNA sites, called upstream activator sequences (UAS). In addition, the product of the negative regulator OPI1, which negatively controls the expression of INO2 and INO4, represses the GUT1 promoter in cells grown on glucose (Grauslund et al., 1999). The second step of the phosphorylation pathway involves the mitochondrial glycerol-3phosphate dehydrogenase, which is the product of GUT2 gene (Ronnow and Kielland-Brandt, 1993). As the GUT1 gene, GUT2 was originally identified by Sprague and Cronan (1977).

16

Stylianos Fakas, Anna Makri, Stamatina Bellou et al.

Later, these results were confirmed by disruption studies, which resulted in glycerol nonutilizing phenotype (Ronnow and Kielland-Brandt, 1993). The regulatory properties of GUT2 are very much the same those of GUT1, that is, GUT2 is repressed during growth on glucose and derepressed during growth on non-fermentable carbon sources, such as glycerol, ethanol, and lactate (Grauslund and Ronnow, 2000). Derepression, however, is mediated through the protein kinase Snf1p and the heteromeric protein complex, Hap2/3/4/5, while repression is again regulated by the OPI1 negative regulator (Grauslund and Ronnow, 2000).

Conclusion Despite the large number of studies concerning glycerol’s catabolism, there are still many details that remain as yet unknown. These details concern mostly the genes involved in glycerol assimilation and their regulation. Most studies carried out thus far have been about the genes coding for the enzymes of the phosphorylation pathway, while nothing is known about the regulation of the genes involved in the oxidative catabolic pathway. Future studies in the molecular biology of glycerol assimilation in fungi other than Saccharomyces will be awaited with much interest.

References Castro, I.M., Loureiro-Dias, M.C., (1991). Glycerol utilization in Fusarium oxysporum var. lini: regulation of transport and metabolism. Journal of General Microbiology, 137, 1497-1502. Courtright, J.B., (1975a). Differential rates of synthesis of glycerokinase and glycerophosphate dehydrogenase in Neurospora crassa during induction. Archives of Biochemistry and Biophysics, 167, 34-44. Courtright, J.B., (1975b). Intracellular localization and properties of glycerokinase and glycerophosphate dehydrogenase in Neurospora crassa. Archives of Biochemistry and Biophysics, 167, 21-33. Deutscher, J., Bauer, B., Sauerwald, H., (1993). Regulation of glycerol metabolism in Enterococcus faecalis by phosphoenolpyruvate-dependent phosphorylation of glycerol kinase catalyzed by enzyme I and HPr of the phosphotransferase system. Journal of Bacteriology, 175, 3730-3733. Ermakova, I.T., Morgunov, I.G., (1988). Pathways of glycerol metabolism in Yarrowia (Candida) lipolytica. Microbiology (Moscow), 57, 533-546. Eze, M., McElhaney, R.N., (1981). The effect of alterations in the fluidity and phase state of the membrane lipids on the passive permeation and facilitated diffusion of glycerol in Escherichia coli. Journal of General Microbiology, 124, 299-307. Ferreira, C., van Voorst, F., Martins, A., Neves, L., Oliveira, R., Kielland-Brandt, M.C., Lucas, C., Brandt, A., (2005). A member of the sugar transporter family, Stl1p is the glycerol/H+ symporter in Saccharomyces cerevisiae. Molecular Biology of the Cell, 16, 2068-2076.

Pathways to Aerobic Glycerol Catabolism and their Regulation

17

Gancedo, C., Gancedo, J.M., Sols, A., (1968). Glycerol metabolism in yeasts. Pathways of utilization and production. European Journal Of Biochemistry / FEBS, 5, 165-172. Grauslund, M., Lopes, J.M., Ronnow, B., (1999). Expression of GUT1, which encodes glycerol kinase in Saccharomyces cerevisiae, is controlled by the positive regulators Adr1p, Ino2p and Ino4p and the negative regulator Opi1p in a carbon source-dependent fashion. Nucl. Acids Res., 27, 4391-4398. Grauslund, M., Ronnow, B., (2000). Carbon source-dependent transcriptional regulation of the mitochondrial glycerol-3-phosphate dehydrogenase gene, GUT 2, from Saccharomyces cerevisiae. Canadian Journal of Microbiology, 46, 1096-1100. Heath, H.E., Gaudy, E.T., (1978). Relationship between catabolism of glycerol and metabolism of hexosephosphate derivatives by Pseudomonas aeruginosa. Journal of Bacteriology, 136, 638-646. Holst, B., Lunde, C., Lages, F., Oliveira, R., Lucas, C., Kielland-Brandt, M.C., (2000). GUP1 and its close homologue GUP2, encoding multimembrane-spanning proteins involved in active glycerol uptake in Saccharomyces cerevisiae. Molecular Microbiology, 37, 108124. Hondmann, D.H., Busink, R., Witteveen, C.F., Visser, J., (1991). Glycerol catabolism in Aspergillus nidulans. Journal of General Microbiology, 137, 629-36. Kistler, W.S., Lin, E.C.C., (1972). Purification and Properties of the Flavine-Stimulated Anaerobic L-{alpha}-Glycerophosphate Dehydrogenase of Escherichia coli. Journal of Bacteriology, 112, 539-547. Lages, F., Lucas, C., (1995). Characterization of a glycerol/H+ symport in the halotolerant yeast Pichia sorbitophila. Yeast, 11, 111-119. Lages, F., Lucas, C., (1997). Contribution to the physiological characterization of glycerol active uptake in Saccharomyces cerevisiae. Biochimica et Biophysica Acta (BBA) Bioenergetics, 1322, 8-18. Lin, E.C.C., (1976). Glycerol dissimilation and its regulation in bacteria. Annual Review of Microbiology, 30, 535-578. Lin, S.-F., Chiou, C.-M., Tsai, Y.-C., (1996). Purification and characterization of a glycerol oxidase from Penicillium sp. TS-622. Enzyme and Microbial Technology, 18, 383-387. McCowen, S.M., Phibbs, P.V., Feary, T.W., (1981). Glycerol catabolism in wild-type and mutant strains of Pseudomonas aeruginosa. Current Microbiology, 5, 191-196. Moat, A.G., Foster, J.W., Spector, M.P., 2002. Microbial physiology: Energy production and metabolite transport. (4th ed). New York, ST: John Wiley and Sons. Nelissen, B., Wachter, R., Goffeau, A., (1997). Classification of all putative permeases and other membrane plurispanners of the major facilitator superfamily encoded by the complete genome of Saccharomyces cerevisiae. FEMS Microbiology Reviews, 21, 113134. Oliveira, R., Lages, F., Silva-Graca, M., Lucas, C., (2003). Fps1p channel is the mediator of the major part of glycerol passive diffusion in Saccharomyces cerevisiae: artefacts and re-definitions. Biochimica et Biophysica Acta (BBA) - Biomembranes, 1613, 57-71. Oliveira, R., Lucas, C., (2004). Expression studies of GUP1 and GUP2, genes involved in glycerol active transport in Saccharomyces cerevisiae, using semi-quantitative RT-PCR. Current Genetics, 46, 140-146.

18

Stylianos Fakas, Anna Makri, Stamatina Bellou et al.

Pasteris, S.E., Strasser de Saad, A.M., (2005). Aerobic glycerol catabolism by Pediococcus pentosaceus isolated from wine. Food Microbiology, 22, 399-407. Pavlik, P., Simon, M., Schuster, T., Ruis, H., (1993). The glycerol kinase (GUT1) gene of Saccharomyces cerevisiae: cloning and characterization. Current Genetics, 24, 21-25. Romano, A.H., (1986). Microbial sugar transport systems and their importance in biotechnology. Trends in Biotechnology, 207-213. Ronnow, B., Kielland-Brandt, M.C., (1993). GUT2, a gene for mitochondrial glycerol 3phosphate dehydrogenase of Saccharomyces cerevisiae. Yeast, 9, 1121-1130. Rush, D., Karibian, D., Karnovsky, M.L., Magasanik, B., (1957). Pathways of glycerol dissimilation in two strains of Aerobacter aerogenes: enzymatic and tracer studies. Journal of Biological Chemistry, 226, 891-899. Sprague, G.F., Cronan, J.E., (1977). Isolation and characterization of Saccharomyces cerevisiae mutants defective in glycerol catabolism. Journal of Bacteriology, 129, 13351342. Sutherland, F.C., Lages, F., Lucas, C., Luyten, K., Albertyn, J., Hohmann, S., Prior, B.A., Kilian, S.G., (1997). Characteristics of Fps1-dependent and -independent glycerol transport in Saccharomyces cerevisiae. Journal of Bacteriology, 179, 7790-7795. Tom, G.D., Viswanath-Reddy, M., Howe, H.B., (1978). Effect of carbon source on enzymes involved in glycerol metabolism in Neurospora crassa. Archives of Microbiology, 117, 259-263. Truniger, V., Boos, W., (1993). Glycerol uptake in Escherichia coli is sensitive to membrane lipid composition. Research in Microbiology, 144, 565-574. Uwajima, T., Shimizu, Y., Terada, O., (1984). Glycerol oxidase, a novel copper hemoprotein from Aspergillus japonicus. Molecular and catalytic properties of the enzyme and its application to the analysis of serum triglycerides. Journal of Biological Chemistry, 259, 2748-2753. Vasiliadis, G.E., Sloan, J., Marshall, J.H., May, J.W., (1987). Glycerol and dihydroxyacetone metabolizing enzymes in fission yeasts of the genus Schizosaccharomyces. Archives of Microbiology, 147, 263-267. Viswanath-Reddy, M., Bennett, S.N., Howe, H.B., (1977). Characterization of glycerol nonutilizing and protoperithecial mutants of Neurospora. Molecular and General Genetics, 153, 29-38.

In: Microbial Conversions of Raw Glycerol Editor: George Aggelis

ISBN 978-1-60692-392-4 © 2009 Nova Science Publishers, Inc.

Chapter III

Citric Acid Production from Raw Glycerol by Yarrowia Lipolytica Wratislavia 1.31 Anita Rywińska2 and Waldemar Rymowicz Department of Biotechnology and Food Microbiology, Faculty of Food Science, University of Environmental and Life Sciences, Wroclaw, PL–50-375 Wroclaw, Poland

Abstract The effects of different kinds of raw glycerol on yeast growth and citric acid biosynthesis in fed-batch cultures were studied. The following types of raw glycerol were used: purified and un-purified raw glycerol from rape seed methyl ester production and un-purified raw glycerol from ethyl ester production. All of them were a very good carbon and energy source for citric acid production by the acetate mutant strain of Y. lipolytica Wratislavia 1.31. Salt and other impurities in raw glycerol slightly influenced the production of citric acid. The highest citric acid concentration (146 g/L) was obtained with purified glycerol from methyl ester production during a fed-batch culture lasting 148 h. In this process, the yield of citric acid (0.73 g/g) was also the highest.

Keywords: Yarrowia lipolytica, citric acid, raw glycerol, fed-batch culture

Introduction Citric acid is the most important organic acid, the global production of which has now reached 1.4 million tonnes and annual growth of 3.5–4.0% in demand/consumption of citric acid is anticipated. As a result of adverse market conditions, only big producers have 1

e-mail: [email protected]

20

Anita Rywińska and Waldemar Rymowicz

survived. Citric acid is widely used to impart a pleasant, tart flavor to foods and beverages. It also finds applications as a functional detergent additive, in pharmaceuticals, cosmetics and toiletries. Recent prices of citric acid have been around $1 to $1.3 per kilo (Soccol1 et al. 2006). Large-scale production of citric acid is carried out mainly by the filamentous fungus Aspergillus niger and a few other fungi in submerged fermentations on beet or cane molasses, sucrose or glucose syrup. During the past four decades many yeast belonging to the species Yarrowia (Candida-Saccharomycopsis-Endomycopsis) lipolytica, C. quilliermondii, C. oleophila, C. intermedia, have been used for the production of citric acid using renewable resources or waste materials as substrates. Y. lipolytica, when grown under nutrient-limited conditions, is able to produce citric acid from a variety of carbon sources, including sugars (glucose, glucose syrups), alcohols (industrial ethanol, methanol, glycerol), acetate, nhydrocarbons, plant oils (crude rapeseed oil, canola oil) or mixtures of industrial free fatty acids of animal origin (Anastassiadis and Rhem 2005, Anastassiadis and Rhem 2006, Crolla and Kennedy (2004), Fickers et al. 2005, Finogenova et al. 2005, Papanikolaou et al. 2006, Rywińska et al. 2006, Wojtatowicz et al. 1991, Venter et al. 2004). Yeasts can generally tolerate higher substrate concentrations than fungi, with comparable conversion rates and a greater tolerance to metal ions, thus allowing the use of less refined substrates such as byproducts and residues from the agro-industry (Meers and Milsom 1987, Venter et al. 2004). The use of agro-industrial wastes in submerged fermentation is economically important and minimizes environmental problems. A cost reduction in citric acid production can be achieved by using less expensive substrates. Glycerol water from either rape seed methyl or ethyl esters production, for example, would be a particularly interesting raw material, lowcost carbon substrate, for the production of citrates by yeasts, if they could be utilized in fermentation processes without any pretreatment. It is estimated that around 1 million tones/year of raw glycerol will have been generated, as a by-product of biofuel production, by the end of the decade (Pachauri and He 2006). However, only a few reports can be found in literature regarding citric acid production from glycerol (Imandi et al. 2007, Levinson et al. 2007, Papanikolaou et al. 2002, Papanikolaou et al. 2003, Rymowicz et al. 2006, Rymowicz et al. 2008). The objective of the study in this chapter was to investigate the suitability of different types of raw industrial glycerol as a substrate for citric acid production by the yeast strain of Yarrowia lipolytica Wratislavia 1.31 under fed-batch culture conditions.

Materials and Methods Microorganism The yeast strain of Yarrowia lipolytica Wratislavia 1.31 was originally obtained from the yeast culture collection of the Department of Biotechnology and Food Microbiology, Wroclaw University of Environmental and Life Sciences in Poland. Originally, the Wratislavia 1.31 strain was obtained from the wild strain A-101 by exposure to UV radiation and it is acetate negative (ace-) mutant (i.e. it can not grow on acetate as the sole carbon and

Citric Acid Production from Raw Glycerol by Yarrowia Lipolytica Wratislavia… 21 energy source). In comparison to the parent strain, the Wratislavia 1.31 strain is characterized by markedly enhanced purity of citrate fermentation in glucose medium (production of isocitric acid, unwanted by-product in this process, is very low - citric acid to isocitric acid ratio as 95 to 5). The yeast strain was maintained on YM slants with paraffin at 4 °C.

Media and Culture Conditions The growth medium for inoculum preparation contained per litre of tap water: 50 g glycerol (MERCK), 3g yeast extract (DIFCO), 3g malt extract (MERCK) and 5g bactopepton (MERCK). The culture was grown in 0.3 L flasks containing 0.05 L of growth medium on an Elpan (Poland) rotational shaker at 30 °C for 3 days. An inoculum of 0.1 L was introduced into bioreactor. Citric acid was produced in the production medium with following composition (g/L in tap water): glycerol, 80 – 100; NH4Cl, 3; MgSO4 x 7H2O, 1; KH2PO4, 0.2; yeast extract, 1. The culture broth medium was fed with two portions of glycerol solution (about 50 – 200 mL, it depended on density and concentration of glycerol – mind the types of glycerol below) after 24 h and 48 h of cultivation (Figures 1-4), to obtain final glycerol concentration of 200 g/L and total working volume of 1.3 L. The following types of glycerol: pure glycerol (purity 98%-POCh)–P98%Gly; purified (ME35%Gly) from methyl ester production (SG BODDINS GmBH, Germany) containing 350 g/L glycerol and 0.65 g/L NaCl; un-purified (ME85%Gly) raw glycerol from methyl ester production (SG BODDINS GmBH, Germany) containing 850 g/L glycerol and 65 g/L NaCl and un-purified raw glycerol from ethyl ester production (EE56%Gly) (Faculty of Chemistry, University of Wroclaw, Poland), containing 560 g/L glycerol and 50 g/L KCl were used as a carbon and energy sources in the media for citric acid production. Cultivations were carried out in 3.5 L of a Bioflo III stirred tank reactor (New Brunswick Scientific Co., USA) with a working volume of 1.3 L at 30 °C. The aeration rate was fixed at 0.2 L/min. The stirrer speed was adjusted to 600 rpm and the pH was maintained automatically at 5.5 by the addition of NaOH solution (40% w/v).

Analytical Methods The biomass was determined by the dry weight method. 10 mL samples were centrifuged at 5000 rpm, washed with distilled water, filtered on Millipore filters (0.45 μm) and dried to constant weight at 105 ºC. Isocitric acid was determined, using an enzymatic method according to Goldberg and Ellis (1983). The concentrations of citric, malic and fumaric acids, glycerol, erythritol and mannitol were determined by HPLC on an Aminex HPX87H Organic Acid column coupled to a UV (λ=210 nm) using RI detector. The column was eluted with 20 mM of H2SO4 at room temperature at a flow rate of 0.6 mL/min.

Anita Rywińska and Waldemar Rymowicz

22

Results and Discussion

150 135 120 105 90 75 60 45 30 15 0

50 40 30 20

X, (g/L)

CA, ICA, Gly, (g/L)

The results of our earlier investigations show that acetate negative mutants of Y. lipolytica are the best producers of citric acid from glycerol in batch and fed-batch experiments (Rymowicz et al. 2006, Rymowicz et al. 2008). In this chapter, three kinds of raw glycerol, varying in purity and the presence of salt, were used to enhance citric acid production by Y. lipolytica Wratislavia 1.31 strain, under fed-batch conditions. Pure glycerol was used as a control. As shown in Figures 1-4, the curves for cell growth were similar to those observed in conventional batch fermentation. However, significant differences were observed in cell biomass production between cultures, depending on the type of glycerol, although the initial nitrogen concentrations (3 g/L NH4Cl) were the same in each process. After every single feeding with glycerol a great decrease of biomass as well as all metabolites concentrations was observed, which was the result of increasing of the entire culture volume. The highest biomass (19 g/L) was obtained in the process with P98%Gly. When raw glycerol were used as a carbon source, lower concentrations of biomass, ranging from 14.5 g/L (ME35%Gly) to 16 g/L (ME85%Gly) were achieved. These data suggest that salts and other impurities inhibit the yeast growth.

10 0 0

20

40

60

80

100

120

10 8 6 4 2 0

5 4 3 2 1 0 0

20

40

60

80

100

MAL, FUM, (g/L)

MAN, ER, (g/L)

Figure 1A. Biomass, X (●), glycerol, Gly (▲), citric acid, CA (□) and isocitric acid, ICA (■) changes during fed-batch production by Y. lipolytica Wratislavia 1.31 strain growing in P98%Gly medium.

120

Time (h)

Figure 1B. Variation of erythritol, ER (○), mannitol, MAN (+), fumaric acid, FUM (■), and malic acid, MAL (♦) concentration with time. Culture conditions: Gly 200 g/L, 3 g/L NH4Cl at 30 ºC and pH = 5.5. Data are the average of duplicate cultures of Y. lipolytica Wratislavia 1.31.

150 135 120 105 90 75 60 45 30 15 0

50 40 30 20

X, (g/L)

CA, ICA, Gly, (g/L)

Citric Acid Production from Raw Glycerol by Yarrowia Lipolytica Wratislavia… 23

10 0 0

20

40

60

80

100

120

140

160

20

5

16

4

12

3

8

2

4

1

0

MAL, FUM, (g/L)

MAN, ER, (g/L)

Figure 2A. Biomass, X (●); glycerol, Gly (▲); citric acid, CA (□); and isocitric acid, ICA (■) during fed-batch production of citric acid by Y. lipolytica Wratislavia 1.31 in ME35%Gly medium.

0 0

20

40

60

80 Time (h)

100

120

140

160

Figure 2B. Variation of erythritol, ER (○); mannitol, MAN (+); fumaric acid, FUM (■); and malic acid, MAL (♦) concentration with time. Culture conditions: Gly 200 g/L, 3 g/L NH4Cl at 30 ºC and pH = 5.5. Data are the average of duplicate cultures of Y. lipolytica Wratislavia 1.31.

In the control culture (P98%Gly), the final concentration of citric acid was 128 g/L after 109 h (Figure 1A). In the fermentation processes with the use of raw glycerol, the best result was observed with the ME35%Gly culture. As a result, 146 g/L of citric acid was achieved after 148 h, corresponding to a 0.73 g/g yield (Figure 2A and Table 1). The same mutant strain gave a yield of citric acid of around 0.62 g/g when the yeast grew in batch mode with glycerol (Rymowicz et al. 2006). In the processes with EE56%Gly (Figure 3A) and ME85%Gly (Figure 4A), the final concentrations of citric acid were similar (137 g/L and 130 g/L, respectively). The results show that the effects of raw glycerol on fed-batch cultures were scarce and did not interfere with citric acid production. According to Imandi et al. (2007), when 54.4 g/L of the by-product containing glycerol (obtained during biodiesel production) was used, the resultant citric acid quantity was 77.39 g/L, but the percentages of glycerol in this byproduct were not reported. The values obtained in the present study were markedly higher than those reported by Papanikolaou et al., (2002), who used Y. lipolytica LGAM S (7) 1 strain, grown on raw glycerol. However, similar very high final citric acid concentrations, ranging from 134 g/L to 166.5 g/L, were reported by Anastassiadis and Rhem

Anita Rywińska and Waldemar Rymowicz

24

150 135 120 105 90 75 60 45 30 15 0

50 40 30 20

X, (g/L)

CA, ICA, Gly, (g/L)

(2006) in batch and repeated batch cultures of Y. lipolytica ATCC 20177 on glucose media. According to Kamzolova et al. (2005), Y. lipolytica 187/1 strains produced around 135 g/L citric acid when vegetable fat was used as substrate in batch cultures. Simultaneously with citric acid accumulation, the presence of by-products, such as isocitric, malic, fumaric acids and polyols were observed. Moreover, the amount of isocitric acid produced by Wratislavia 1.31 strain from pure and raw glycerol was very low, only about 3.5 g/L (Table 1). According to Anastassiadis et al. (2002) and Kamzolova et al. (2005) C. lipolytica DSM 3286 and Y. lipolytica 187/1 strains produced similar amount of isocitric acid. In contrast, the wild strain of Y. lipolytica A-101 produced more than 10 g/L of this acid (Wojtatowicz et al. 1993). Levinson et al. (2007) reported that among twenty seven of Y. lipolytica strains, selected for citric acid production from glycerol, only one produced less than 10% of isocitric acid.

10

0

20

40

60

80

100

120

0 140

15

10

12

8

9

6

6

4

3

2

0

0 0

20

40

60

80

100

120

MAL, FUM, (g/L)

MAN, ER, (g/L)

Figure 3A. Biomass, X (●), glycerol, Gly (▲), citric acid, CA (□) and isocitric acid, ICA (■) changes during fed-batch production of citric acid Y. lipolytica Wratislavia 1.31 in EE56%Gly.

140

Time (h) Figure 3B. Variation of erythritol, ER (○), mannitol, MAN (+), fumaric acid, FUM (■), and malic acid, MAL (♦) concentration with time. Culture conditions: Gly 200 g/L, 3 g/L NH4Cl at 30 ºC and pH = 5.5. Data are the average of duplicate cultures of Y. lipolytica Wratislavia 1.31.

Citric Acid Production from Raw Glycerol by Yarrowia Lipolytica Wratislavia… 25 50 40 30 20

X, (g/L)

CA, ICA, Gly, (g/L)

150 135 120 105 90 75 60 45 30 15 0

10

0

20

40

60

80

100

120

0 140

15

10

12

8

9

6

6

4

3

2

0

0 0

20

40

60

80

100

120

MAL, FUM, (g/L)

MAN, ER, (g/L)

Figure 4A. Biomass, X (●), glycerol, Gly (▲), citric acid, CA (□) and isocitric acid, ICA (■) changes during fed-batch production of citric acid by Y. lipolytica Wratislavia 1.31 in ME85%Gly medium.

140

Time (h) Figure 4B. Variation of erythritol, ER (○), mannitol, MAN (+), fumaric acid, FUM (■), and malic acid, MAL (♦) concentration with time. Culture conditions: Gly 200 g/L, 3 g/L NH4Cl at 30 ºC and pH = 5.5. Data are the average of duplicate cultures of Y. lipolytica Wratislavia 1.31.

Table 1. Results of the citric acid production from various kinds of glycerol by Y. lipolytica Wratislavia 1.31 strain Parameter Cultivation time, t/h Biomass (g/L) Citric acid (g/L) Isocitric acid (g/L) Total yield of citric acid produced, Y/(g/g)

Glycerol P98%Gly 109 19 ± 0.424 128 ± 2.262 3.0 ± 0.070

ME35%Gly 148 14.5 ± 0.989 146 ± 5.656 3.5 ± 0.127

EE56%Gly 136 14,8 ± 0.282 137 ± 2.828 2.3 ± 0.070

ME85%Gly 117 16 ± 0.254 130 ± 3.818 3.4 ± 0.127

0.64 ± 0.011

0.73 ± 0.028

0.685 ± 0.014

0.65 ± 0.019

It is interesting to note that polyols, such as erythritol and mannitol were produced in the culture broth (Figures 1-4B). The highest concentrations of erythritol (15.1 g/L) and mannitol

26

Anita Rywińska and Waldemar Rymowicz

(8.7 g/L) were determined after 120 h, in the process with ME35%Gly (Figure 2). In the cultures containing unpurified glycerol, EE56%Gly and EM85%Gly, the accumulation of erythritol reached about 14 g/L, but mannitol was synthesized to a small extent (about 2.5 g/L) (Figure 3B and 4B). Erythritol is present in fruit (such as melons, grapes and pears), mushrooms and some fermented foods (Park et al. 1998). Mannitol is the most abundant polyol in natural environment and it is present in bacteria, yeast, fungi, algae and several plants, e.g. celery, onions, grasses and olives (Wisselink et al. 2002). Some microorganisms, including bacteria, yeast and fungi (Aoki et al. 1993) can selectively produce mannitol and erythritol from glucose and fructose. Production of these sugar alcohols from glycerol by Y. lipolytica yeast is not very common. As has been observed, sugar alcohols including mannitol and erythritol, protect plants, fungi, yeast and bacteria from osmotic stress (Kets et al. 1996). It is quite likely that in this study, polyols were produced as a consequence of the strain exposure to high concentrations of glycerol and citric acid. In this study, the concentration of sugar alcohols gradually increased with citric acid production but after utilization of glycerol, the Wratislavia 1.31 strain was able to convert polyols into citric acid (Figures 1-4). This in turn, prolonged the effective phase of citric acid production. The amount of these by-products was below 0.5 g/L at the end of the process. For example, the shortest was the process with the use of pure glycerol (after 92 h), while polyols were utilized after 109 h (Figure 1). In addition to sugar alcohol, at the early stage of the production phase, the yeast produced intermediates of a tricarboxylic cycle, such as malic and fumaric acid (Figures 14B). The concentrations of fumaric acid were very low, but high amount of malic acid (about 6 g/L) was observed in the cultures with EE56%Gly and ME85%Gly (Figures 3 and 4B). These metabolites, like polyols, were utilized at the end of the cultures. Figure 5 shows that Y. lipolytica Wratislavia 1.31 strain produced citric acid with specific (q) and volumetric (Q) rates, different at the three stages. The values of these kinetic parameters were the highest during the first production phase, since citric acid was partially produced by growing yeast cells (Rywińska et al. 2006, Bubbico et al. 1996), but next, they gradually decreased in the second and third phases. Similar changes in the dynamics of citric acid production from glucose were reported earlier (Wojtatowicz et al. 1991, Enzminger and Asejno 1986, Behrens et al. 1982). However, the type of glycerol did not exert a significant impact on q and Q in successive stages. At the end of the process, the specific citric acid production rate ranged from 0.036 g/gh (EE56%Gly) to 0.048 g/gh (ME35%Gly), while the volumetric productivity was from 0.54 g/Lh (EE56%Gly) to 0.76 g/Lh (P98%Gly). The values of specific and volumetric rates of citric acid synthesis were comparable with those reported when glucose media were used (Rywińska et al. 2006, Wojtatowicz et al. 1993, Rane and Sims 1994, Rane and Sims 1996, Klasson et al. 1989). However, when alternative carbon sources for citric acid production were used, the specific production rate and volumetric productivity were higher than those reported above and obtained in our study. According to Kamzolova et al. (2005) the specific rate of citric acid synthesis reached 0.127 g/gh. The volumetric productivity with Y. lipolytica 187/1 grown on rapeseed oil was 1.25 g/Lh. When Y. lipolytica N1 strain was grown on ethanol, the values were 0.120 g/gh and 1.15 g/gh, respectively (Kamzolova et al. 2003).

Citric Acid Production from Raw Glycerol by Yarrowia Lipolytica Wratislavia… 27 0,25

2,5 Q (g/Lh)

0,15

1,5

0,1

1

q (g/gh)

0,2

2

0,05

0,5

0

0 0

20

40

60

80

100

120

140

160

A. 2,5

0,25 0,2 0,15

1

0,1

0,5

q (g/gh)

Q (g/Lh)

2 1,5

0,05

0 0

20

40

60

80

100

120

140

0 160

B. 0,25

2

0,2

1,5

0,15

1

0,1

0,5

0,05

0

q (g/gh)

Q (g/Lh)

2,5

0 0

20

40

60

80

100

120

140

160

2,5 2 1,5 1 0,5 0

0,25 0,2 0,15 0,1 0,05 0 0

20

40

60

80

100

120

140

q (g/gh)

Q (g/Lh)

C.

160

Time (h)

D. Figure 5. The specific CA production rate, q (∆) and the volumetric CA productivity, Q (▲) in fed batch citric acid production by Y. lipolytica Wratislavia 1.31 strain growing on glycerol. Culture conditions: Gly 200 g/L, 3 g/L NH4Cl at 30 ºC and pH = 5.5. A –P98%Gly, B- ME35%Gly, CEE56%Gly, D- ME85%Gly. Data are the average of duplicate cultures of Y. lipolytica Wratislavia 1.31.

Studies on citric acid production from glycerol are scarce and mainly report the experiments carried out in shake flasks (Imandi et al. 2007, Papanikolaou et al. 2002, Papanikolaou et al. 2003, Levinson et al. 2007). Papanikolaou et al., (2002) reported that citric acid fermentation by Y. lipolytica LGAM S(7)1 growing on raw glycerol yielded 0.42 g citric acid per g glycerol consumed. The same microorganism (LGAM S(7)1) showed a yield

28

Anita Rywińska and Waldemar Rymowicz

of citric acid produced per glycerol consumed of around 0.5 g/g when cultivated on mixtures containing raw glycerol and industrial derivative of tallow (Papanikolaou et al. 2003). As reported by Levinson et al., (2007), the strain of Y. lipolytica NRRL YB-423 produced citric acid from pure glycerol with a yield of 0.54 g/g. In general, the values of the kinetic parameters, as well as those of the efficiency of citric acid production by the strain under investigation, were found within the range reported in literature for other strains in batch cultivations on glucose media (Enzminger and Asejno 1986, Rane and Sims 1996, Klasson et al. 1989, Antonucci et al. 2001). A markedly higher value of citric acid yield, equal to 1.55 g/g, has been reported with the strain of Y. lipolytica 187/1 grown on rapeseed oil in fed batch culture (Kamzolova et al. 2005). Wojtatowicz et al., (1993) used the strain of Y. lipolytica A101 grown on n-hexadecane and obtained a yield of 1.44 g/g of total acids (citric and isocitric). Kamzolova et al., (2003) reported a citric acid yield of 0.87 g/g with the strain of Y. lipolytica N1 in the culture with ethanol as a carbon source.

Conclusion The results obtained in the present study and in earlier investigations (Rymowicz et al. 2006) suggest that raw glycerol, a by-product from biodiesel production, could be a promising substrate for citric acid production by ace- mutants of Y. lipolytica yeast. The data obtained with the strain Wratislavia 1.31 are the best reported in international literature. It is apparent that fed-batch mode is suitable to increase the final citric acid concentration and product yield. In conclusion, this process seems to be very effective for large-scale production of citric acid.

Acknowledgement This work was financed by the Ministry of Sciences and Higher Education of Poland under Project No. 2P06T 044 30.

References Anastassiadis, S. and Rhem, H. (2005). Continuous citric acid secretion by a high specific pH dependent active transport system in yeast Candida oleophila ATCC 20177. Electronic Journal of Biotechnology, 8, 147-161. Anastassiadis, S. and Rhem, H. (2006). Citric acid production from glucose by yeast Candida oleophila ATCC 20177 under batch, continuous and repeated batch cultivation. Electronic Journal of Biotechnology, 9, 26-39. Anastassiadis, S. and Aivasidis, A. and Wandrey, C. (2002). Citric acid production by Candida strains under intracellular nitrogen limitation. Applied Microbiology and Biotechnology; 60, 81-87. Antonucci, S. and Bravi, M. and Bubbico, R. and Di Michele, A. and Verdone, N. (2001). Selectivity in citric acid production by Yarrowia lipolytica. Enzyme and Microbial Technology, 28, 189-195.

Citric Acid Production from Raw Glycerol by Yarrowia Lipolytica Wratislavia… 29 Aoki, MAY and Pastore, G. M. and Park, Y. K. (1993). Microbial transformation of sucrose and glucose to erythritol. Biotechnology Letters, 15, 383-388. Behrens, U. and Schulze, E. and Weissbrodt, E. and Schelfel, B. (1982). Kinetics of byproduct formation during production of citric acid by Saccharomycopsis lipolytica. Acta Biotechnology, 2, 171-177. Bubbico, R. and Presti, S. and Bravi, M. and Moresi, M. and Spinosi, M. (1996). Repeated batch citrate production by Yarrowia lipolytica using yeast recycling by cross-flow microfiltration. Agro Food Industry Hi-Tech, 3, 35-38. Crolla, A., Kennedy, K. J. (2004). Fed-batch production of citric acid by Candida lipolytica grown on n-parrafins. Journal Biotechnology, 110, 73-84. Enzminger, J. D. and Asejno, J. A. (1986). Use of cell recycle in the aerobic fermentative production of citric acid by yeast. Biotechnology Letters, 8, 7-12. Fickers, P., Benneti P. H., Wache, Y., Marty, A., Mauersberger, S., Smit, M. S., et al. (2005). Hydrophobic substrate utilisation by the yeast Yarrowia lipolytica, and its potential applications. FEMS Yeast Research, 5, 527-43. Finogenova, T. V. and Morgunov, I. G. and Kamzolova, S. V. and Chernyavskaya, O. G. (2005). Organic acid production by the yeast Yarrowia lipolytica: a review of prospects. Applied Biochemistry and Microbiology, 41, 418-425. Goldberg, D. M. and Ellis, G. (1983). Isocitrate dehydrogenase. In: Bergmeyer HU (ed) Methods of Enzymatic Analysis. Weinheim, 3, 183-190. Imandi, S. B. and Bandaru, V. V. R. and Somalanka, S. R. and Garapati, H. R. (2007). Optimization of medium constituents for the production of citric acid from byproduct glycerol using Doehlert experimental design. Enzyme and Microbial Technology, 40, 1367-1372. Kamzolova, S. V. and Morgunov, I. G. and Aurich, A. and Perevoznikova, O. A. and Shishkanova, N. V. and Finogenova, T. V. and Stottmeister, U. (2005). Lipase secretion and citric acid production in Yarrowia lipolytica yeast grown on animal and vegetable fat. Food Technology and Biotechnology, 43, 113-122. Kamzolova, S. V. and Shishkanova, N. V. and Morgunov, I. G. and Finogenova, T. V. (2003). Oxygen requirements for growth and citric acid production of Yarrowia lipolytica. FEMS Yeast Research, 3, 217-222. Kets, E. P. W. and Galinski, E. A. and De Wit M. and De Bont, J. A. M. and Heipieper, H. J. (1996). Mannitol, a novel bacterial compatible solute in Pseudomonas putida. Journal of Bacteriology, 178, 6665-6670. Klasson, T. K. and Clausen, E. C. and Gaddy, J. L. (1989). Continuous fermentation for the production of citric acid from glucose. Applied Biochemistry and Biotechnology, 21, 491-509. Levinson, W. E. and Kurtzman, C. P. and Kuo, T. M. (2007). Characterization of Yarrowia lipolytica and related species for citric acid production from glycerol. Enzyme and Microbial Technology, 41, 292-295. Meers, J. L., Milsom P. E. (1987). Organic acids and amino acids. In: Bu’Lock J, Kristiansen B., editors. Basic biotechnology. Orlando, FL: Academic Press, 359-83.

30

Anita Rywińska and Waldemar Rymowicz

Pachauri, N. and He, B. (2006). Value – added utilization of crude glycerol from biodiesel production: A survey of current research activities. American Society of Agricultural and Biological Engineers ASABE, Paper No 066223, 2-15. Papanikolaou, S. and Galiotou-Panayotou, M. and Chevalot, I. and Komaitis, M. and Marc, I. and Aggelis, G. (2006). Influence of glucose and saturated free-fatty acid mixtures on citric acid and lipid production by Yarrowia lipolytica. Current Microbiology, 52, 134142. Papanikolaou, S. and Muniglia, L. and Chevalot, I. and Aggelis, G. and Marc, I. (2002). Yarrowia lipolytica as a potential producer of citric acid from raw glycerol. Journal of Applied Microbiology, 92, 737-744. Papanikolaou, S. and Muniglia, L. and Chevalot, I. and Aggelis, G. and Marc, I. (2003). Accumulation of a cocoa-butter-like lipid by Yarrowia lipoytica cultivated on agroindustrial residues. Current Microbiology, 46, 124-130. Park, J. B. and Seo, B. C. and Kim, J. R. and Park, Y. K. (1998). Production of erythritol in fed-batch cultures of Trichosporon sp. Journal of Fermentation and Bioengineering, 86, 577-580. Rane, K. D. and Sims, K. A. (1994). Oxygen uptake and citric acid production by Candida lipolytica Y 1095. Biotechnology Bioengineering, 43, 131-137. Rane, K. D. and Sims, K. A. (1996). Citric acid production by Yarrowia lipolytica: effect of nitrogen and biomass concentration on yield and productivity. Biotechnology Letters, 18, 1139-1144. Rymowicz, W. and Rywińska, A. and Gładkowski, W. (2008). Simultaneous Production of Citric Acid and Erythritol from Crude Glycerol by Yarrowia lipolytica Wratislavia K1. Chemical Papers 62 (3) 239-246. Rymowicz, W. and Rywińska, A. and Żarowska, B. and Juszczyk, P. (2006). Citric acid production from raw glycerol by acetate mutants of Yarrowia lipolytica. Chemical Papers, 60, 391-394. Rywińska, A. and Wojtatowicz, M. and Rymowicz, W. (2006). Citric acid biosynthesis by Yarrowia lipolytica A-101-1.31 under deficiency of various medium macrocomponents. Electronic Journal of Polish Agricultural Universities Biotechnology, 9 (1). Soccol1, C. R. and Vandenberghe, L. P. S. and Rodrigues, C. and Pandey, A. (2006). Citric acid production. Food Technology and Biotechnology, 44, 141-149. Venter, T. and Kock, J. L. F. and Botes, P. J. and Smit, M. S. and Hugo A. and Joseph, M. (2004). Acetate enhances citric acid production by Yarrowia lipolytica when grown on sunflower oil. Systematic and Applied Microbiology, 27, 135-138. Wisselink, H. W. and Weusthuis, R. A. and Eggink, G. and Hugenholtz, J. and Grobben G. J. (2002). Mannitol production by lactic acid bacteria: a review. International Dairy Journal, 12, 151-161. Wojtatowicz, M. and Rymowicz, W. and Kautola, H. (1991). Comparison of different strains of the yeast Yarrowia lipolytica for citric acid production from glucose hydrol. Applied Biochemistry and Biotechnology, 31, 165-174. Wojtatowicz, M. and Marchin, G. L. and Erickson, L. E. (1993). Attempts to improve strain A-101 of Yarrowia lipolytica for citric acid production from n-paraffins. Process Biochemistry, 28, 453-460.

In: Microbial Conversions of Raw Glycerol Editor: George Aggelis

ISBN 978-1-60692-392-4 © 2009 Nova Science Publishers, Inc.

Chapter IV

Biodiesel By-Products Used as Substrates for Oxalic Acid Production by Aspergillus Niger Izabela Musiał3 and Waldemar Rymowicz Department of Biotechnology and Food Microbiology, Faculty of Food Science, University of Environmental and Life Sciences, Wroclaw, PL–50-375 Wroclaw, Poland

Abstract The aim of the studies was to evaluate the dynamics and yield of oxalic acid production from biodiesel by-products, such as pure glycerol, fatty acids and glycerin waste (a mixture of glycerol and fatty acids) by Aspergillus niger XP in submerged cultivations. The comparative studies included: product yields, volumetric productivity and concentration of citric acid (an unwanted by-product). The maximum concentration of oxalic acid (55.7 gdm-3) was obtained in the medium containing 50 gdm-3 of fatty acids, which also resulted in the highest oxalate yield and volumetric productivity (1.25 gg-1 and 0.29 gdm-3h-1, respectively). In contrast, the concentration of citric acid was very low (<1 gdm-3). When the medium contained 65 gdm-3 of glycerin waste, A. niger XP produced only 42.0 gdm-3 of oxalic acid (without undesired by-products) at oxalate yield of 0.8 gg-1. The lowest oxalic acid concentrations were obtained in the medium containing pure glycerol.

Introduction Organic acids, especially oxalic acid, are commonly used for leaching of heavy metals from minerals, ore materials and solid waste (Strasser et al. 1994; Ambikadevi and Lalithambika 2000; Mulligan et al. 2004; Santhiya and Ting 2005; Wu and Ting 2006). Bio-

1

e-mail: [email protected].

32

Izabela Musiał and Waldemar Rymowicz

hydrometallurgy groups are interested in the production of high amounts of oxalic acid at low pH, on a large industrial scale. Up to the date, this acid has not been produced on an industrial scale using biotechnological methods. As compared to conventional techniques, microbial bioleaching is generally considered a green technology, which requires lower costs and energy consumption. Oxalic acid can be produced by several microorganisms, including A. niger, which was found to be very efficient (Guru and Bilges 2001) in producing not only oxalic acid, but also citric and laconic acids, depending on the medium composition and biosynthesis conditions, especially the pH. Oxalic acid production by A. niger can be stimulated by maintaining high pH, within the range of 6 - 7, when the culture medium is enriched with appropriate amounts of nitrogen and phosphorus (Bahaman et al. 1998; Camisole et al. 1998; Lebanon et al. 1999). Kubiczek et al. (1988) proved that pyruvate from glycolysis was transformed to oxalacetate, which was hydrolysed to oxalate and acetate by cytosolic oxaloacetate hydrolase (EC 3.7.1.1). When the pH is low, the main product of A. niger biosynthesis is citric acid, with the exception of oxalic acid production from lipid substrates (Leangon et al. 1999; Santoro et al. 1999). Our recent investigation showed that some strains of A. niger produced a great amount of oxalic acid from lipids at low pH (between 4 and 5) (Rymowicz and Lenart 2003). Lower costs of microbial oxalic acid production are generated by modification of the microorganism, introduction of new technologies and use of less expensive substrates. The amount of oxalic acid produced by oxaloacetate hydrolase gene (oahA) overexpressing the strain of A. niger EOAH-1 reached 28.6 gdm-3 from 30 gdm-3 of glucose, which was twice that of parental strain of A. niger WU-2223L (Hattori et al. 2007). Mandal and Banerjee (2005) show that mycelia of A. niger NCIM 548 immobilized in polyurethane foam produced from 3 to 3.7 times more oxalic acid from glucose than free cells, under identical shake-flask conditions. The data in literature show that the use of renewable products from agriculture and food industry for the oxalic acid production, such as beet molasses, green corn syrup, sweet potato or whey have been studied extensively (Guru and Bilges 2001; Leangon et al. 1999; Strasser et al. 1994; Santoro et al. 1999). According to Strasser et al. (1994), sucrose and lactose were suitable carbon sources for oxalic acid production, although A. niger in the sucrose medium additionally produced high amounts of gluconic and oxalic acids, whereas in lactose permeate medium only oxalic acid was produced and the yield of its production was high. The oxalate yield from sucrose was about 0.2 gg-1 while from lactose it ranged from 0.4 to 0.5 gg-1 (Strasser et al. 1994; Cameselle et al. 1998; Santoro et al. 1999). Raw lipid materials can be considered as efficient substrates for the biosynthesis of oxalic acid by A. niger. After 7 days, its production reached 68 gdm-3 in the medium containing 50 gdm-3 of crude rapeseed oil and post-refining fatty acids. High yield of the product, up to 1.4 gg-1, was reported by Rymowicz and Lenart (2003). There are many reports on oxalic acid fermentation in synthetic media containing lactose, sucrose or lipids, but the production of oxalic acid by A. niger from glycerol has not been studied yet. A great amount of glycerol is generated during biodiesel production. Production of biodiesel has grown dramatically during the last few years. Biodiesel is recommended for use as a substitute for petroleum-based diesel mainly because it is a renewable, domestic resource

Biodiesel By-Products Used as Substrates for Oxalic Acid Production…

33

with environmentally friendly emission profile and it is biodegradable (Bozbas 2008). The most common way to produce biodiesel is transesterification, i.e. catalyzed chemical reaction involving vegetable oil and alcohol to yield biodiesel – methyl or ethyl fatty acid esters and glycerin waste (Al-Zuhair 2007). Glycerin waste contains a mixture of glycerin, methanol or ethanol, water, inorganic salts and free fatty acids. The production of biofuel from triglycerides gives about 10 % (w/v) of glycerin waste as a by-product. Glycerin waste with its impurities is of low value (Pachauri and He 2006). The aim of the present investigation was to study the dynamics and yields of oxalic acid biosynthesis from biodiesel by-products, such as glycerin waste, pure glycerol and fatty acids by A. niger XP in submerged cultivations.

Materials and Methods The strain used in this study was Aspergillus niger XP obtained from the Wroclaw University of Economics (Poland). It was maintained on potato dextrose agar (Difco) at 4°C and subcultured every month. The following types of substrates: pure glycerol (purity 98% -POCh, Poland); un-purified glycerin waste containing 24% of glycerol and 68% of fatty acids from methyl ester production (obtained from the Faculty of Chemistry, University of Wroclaw, Poland) and fatty acids from methyl ester production (obtained from the Faculty of Chemistry, University of Wroclaw, Poland) were used as carbon and energy sources in the media for oxalic acid production. The growth medium for inoculum preparation contained: 50g of pure glycerol, 20g of fatty acids, 2.0g of KH2PO4; 1g of NH4NO3; 0.3g of MgSO4 x 7H2O; 0.1g of FeSO4 x 7H2O; 0.2g of ZnSO4 x 7H2O; 15g of methanol per dm3 tap water. The production medium contained: 80-105 g of pure glycerol or 37-50 g of fatty acids (since fatty acids are insoluble in water, total amount of this substrate was increased to 50 gdm-3 in two stages: at the beginning of the process 30 gdm-3 and after 48 h 20 gdm-3) or 47-65 g of glycerin waste (when the total amount of the substrate was increased to 65 gdm-3, its addition was also divided into two stages: the initial portion of 48.7 gdm-3 was followed by 16.3 gdm-3 after 48h); 2.5 g of KH2PO4; 1 g of NH4NO3; 0.3 g of MgSO4 x 7H2O; 0.1 g of FeSO4 x 7H2O; 0.2 g of ZnSO4 x 7H2O and 0.75 g of SPAN 20 per dm3 tap water. The strain of A. niger XP was maintained on potato dextrose agar. The spores from a slant agar were suspended in a 0.1 % Tween 80 solution. The cultures were inoculated with a spore solution to a final concentration of 106 spores in 1 cm3. The inoculation culture was grown in shake flasks at 30 o C, 160 rpm for 72 h. Oxalic acid production was carried out in a 5.5 dm3 Bioflo III stirred tank reactor (New Brunswick Scientific Co., USA) with a working volume of 2.5 dm3 at 30° C. The aeration rate was fixed at 1.5 dm3min-1. The stirrer speed was adjusted to 500 rpm. The pH was maintained automatically at 4.5 by adding KOH solution (40% w/v). The dry weight of the biomass was determined by harvesting the mycelium by filtration through a pre-weighed membrane filter (cellulose nitrate filter, 1.2 µm pore size, Millipore) and drying at 80 °C to constant weight. Unconsumed lipids were extracted from the culture

34

Izabela Musiał and Waldemar Rymowicz

medium twice, using petroleum ether as a solvent. The organic phase was dried at 50 °C to constant weight. The concentrations of oxalic and citric acids and glycerol were determined using HPLC on an Aminex HPX87H Column Organic Acids coupled to a UV detector at 210 nm and using an RI detector. The column was eluted with 20 mM H2SO4 at room temperature and a flow rate of 0.6 mL/ min. Retention time of oxalic acid, citric acid, and glycerol was 6.4, 7.9 and 12.9 min., respectively.

Results and Discussion The strain of A. niger XP was selected by Rymowicz and Lenart (2004), as it was considered the most efficient producer of oxalic acid from lipid substrates, e.g. rapeseed oil or fatty acids. The ability of this strain to produce oxalic acid with the use of biodiesel production by-products was studied on submerged cultures containing pure glycerol and fatty acids or glycerin waste. Figure 1 shows oxalic acid production with the use of pure glycerol. Initial substrate concentrations were 80.0 and 105.0 gdm-3. After 8 days of biosynthesis, the medium contained more than 40 and more than 70% of the initial amount of the substrate, respectively. A. niger XP in the medium containing glycerol produced low amounts of oxalic acid (from 18.5 to 24.0 gdm-3) and its growth was also at a low level. The biomass concentration was found within the range of 6.5 to 11.0 gdm-3. On the other hand, higher accumulation of oxalic acid and biomass was observed in the cultures containing fatty acids and glycerin waste, used as carbon sources (Figures 2 and 3). When the media contained 37.0 gdm-3 and 50 gdm-3 of fatty acids, the production of oxalic acid was twice as high (41.8 and 55.7 gdm-3, respectively). On day 5, the presence of citric acid (approx. 1 gdm-3) was observed in each process. The biomass concentration in the two cultures containing fatty acids was comparable and amounted to 14.5 and 15.5 gdm-3, respectively. Two other cultures were performed on the medium containing glycerin waste at initial concentration of 47 and 65 gdm-3. The processes in the media containing glycerin waste exhibited both high biomass concentration and oxalic acid production (23.1-25.3 gdm-3 and 34.0-42.0 gdm-3, respectively) (Figure 3). The increase in biomass observed with the strain of A. niger XP was likely due to the presence of organic nitrogen and phosphorus in the raw material. Besides, the glycerin fraction, in addition to glycerin and fatty acids, also contained methyl and ethyl alcohol residues. Methanol and ethanol stimulate many fermentation processes (Röhr et al. 2000; Haq et al. 2003; Kumar et al. 2003). Methanol affects metabolic processes of lipids of A. niger, enhances permeability of cellular membranes, facilitates secretion of metabolites from cells and increases the production of oxalic acid. In addition, methanol affects the morphology of mycelium, stimulating the formation of mycelial pellets. Rymowicz and Lenart (2004) found the highest concentration of oxalic acid on lipid substrate when they used 2% methanol in the culture medium. Higher concentrations inhibited the fermentation processes. The presence of citric acid (<1 gdm-3) was found on day 3 in the two processes in the media containing glycerin waste, but later on it was utilized and not found on further days (Figure 3). It is characteristic of oxalic acid biosynthesis by A. niger that in the media containing sucrose or

Biodiesel By-Products Used as Substrates for Oxalic Acid Production…

35

glucose, such by-products as gluconic and citric acid are produced in large quantities. Cameselle et al. (1998) reported that A. niger 1120 in the medium containing sucrose produced 27.2-33.8 gdm-3 of oxalic acid and also high amounts of gluconic and citric acids. When whey was used in the medium containing 120.0 gdm-3 of lactose, 25.0 gdm-3 of pure oxalic acid was obtained. Bohlmann et al. (1998) used A. niger 1120 on whey containing 60% of lactose and after 13 days they obtained a final oxalate concentration of 41.4 gdm-3. Rymowicz and Lenart (2003) studied A. niger XP in continuous cultures with crude rapeseed oil and post-refining fatty acids and found that it produced from 66.1 to 68.1 gdm-3 of oxalic and from 1.4 to 4.8 gdm-3 of citric acid. GLY

120

20

80

10

40

0 0

X, OA [gdm -3]

B

50

100

150

0 200

30

120

20

80

10

40

0 0

50

100

150

GLY [gdm-3]

X, OA [gdm-3]

OA

30

GLY [gdm -3]

X

A

0 200

time [h] -3

Figure 1. Kinetics of biomass (X, gdm ) and oxalic acid (OA, gdm-3) production and glycerol (GLY, gdm-3) consumption of A. niger XP during growth in batch culture. Culture conditions: (A) S0=80 gdm-3 pure glycerol and (B) S0=105 gdm-3 pure glycerol; growth on Bioflo III stirred tank reactor (New Brunswick Scientific Co., USA), working volume of 2.5 dm3, temperature 30°C, aeration rate 1.5 dm3 min-1, stirrer speed 500 rpm, pH 4.5.

Table 1 shows that both the substrate per se and its concentration affected the yield of oxalic acid biosynthesis by A. niger XP. The highest yields (from 1.11 to 1.25 gg-1), were obtained in the media containing fatty acids, lower parameters (<1 gg-1) were in media containing glycerin waste, whilst lowest were observed in the media with glycerin. Rymowicz and Lenart (2003) performed biosynthesis of oxalic acid by the strain of A. niger XP using rapeseed oil and post-refining fatty acids and obtained the yield of 1.66 gg-1. When oxalic acid is biosynthesized in the media containing carbohydrates, its production yields are lower. The highest oxalate yields obtained with carbohydrate substrates have been observed

Izabela Musiał and Waldemar Rymowicz

36

with lactose. Bohlmann et al. (1998) and Santoro et al. (1999) used lactose in their studies and obtained the yields of oxalic acid production of 0.54 and 0.4 gg-1, respectively. A

X

FA

OA

CA

-3

X, FA, OA, CA [gdm ]

50 40 30 20 10 0 0

50

100

150

200

Figure 2. Kinetics of biomass (X, gdm-3), oxalic acid (OA, gdm-3) and citric acid (CA, gdm-3) production and fatty acids (FA, gdm-3) consumption of A. niger XP during growth in batch culture. Culture conditions: (A) S0=40 gdm-3 fatty acids and (B) S0=30 gdm-3 fatty acids and 20 gdm-3 of fatty acids added after 48h of cultivation process; growth on Bioflo III stirred tank reactor (New Brunswick Scientific Co., USA), working volume of 2.5 dm3, temperature 30°C, aeration rate 1.5 dm3 min-1, stirrer speed 500 rpm, pH 4.5. Figure 4 shows time profiles of the volumetric oxalic acid production rates in all cultures under investigation, in the media containing various substrates. The highest parameters (0.56 and 0.63 gdm-3h-1) were obtained on day 4 when glycerin waste of 47.0 and 65.0 gdm-3 was used, respectively. On day 5, this parameter decreased to 0.14 gdm-3h-1. The lowest values (0.43-0.49 gdm-3h-1) were noted on days 4 and 5 in the media containing fatty acids. Rymowicz and Lenart (2003) used crude rapeseed oil and post-refining fatty acids for oxalic acid production and obtained high volumetric productivities, ranging from 12 to 16 gdm-3d-1,

Biodiesel By-Products Used as Substrates for Oxalic Acid Production…

37

which were maintained for 2 to 4 days. The volumetric oxalic acid production rate (when the maximum concentration of oxalic acid had been achieved) in the media containing pure glycerin, fatty acids and glycerin waste was found within the range of 0.10 gdm-3h-1 to 0.29 gdm-3h-1 (Table 1). A

X

FA

GLY

OA

CA

-3

X, FA, GLY, OA, CA [gdm ]

50 40 30 20 10 0 0

50

100

150

16.3 gdm-3 glycerin waste

Figure 3. Kinetics of biomass (X, gdm-3), oxalic acid (OA, gdm-3) and citric acid (CA, gdm-3) production and consumption of fatty acids (FA, gdm-3) and glycerol (GLY, gdm-3) from glycerin waste of A. niger XP during growth in batch culture. Culture conditions: (A) S0=47.0 gdm-3 glycerin waste containing 13.0 gdm-3 glycerol and 34.0 gdm-3 fatty acids; (B) S0=48.7 gdm-3 of glycerin waste containing 13.0 gdm-3 glycerol and 35.7 gdm-3 fatty acids and 16.3 gdm-3 of glycerin waste added after 48 h of cultivation process, containing 2.0 g dm-3 glycerol and 14.3 gdm-3 fatty acids; growth on Bioflo III stirred tank reactor (New Brunswick Scientific Co., USA), working volume of 2.5 dm3, temperature 30°C, aeration rate 1.5 dm3 min-1, stirrer speed 500 rpm, pH 4.5.

The lowest volumetric oxalic acid production rate was obtained in the medium containing 105.0 gdm-3 of glycerin, while the highest was found in the process performed with the use of 50.0 gdm-3 of fatty acids. In batch cultures, the volumetric productivities in

Izabela Musiał and Waldemar Rymowicz

38

the media containing carbohydrates were higher and ranged from 3.4 to 6,8 gdm-3d-1 (Strasser et al. 1994; Bohlmann et al. 1998; Cameselle et al. 1998; Santoro et al. 1999). Table 1. Quantitative data of A. niger XP originated from kinetics in media with various kinds of substrates and various initial substrate (S0) concentrations Substrate

Time of fermentation (h) 185 191 191 192 167 166

Pure glycerol Fatty acids Glycerin waste

S0

Sr

X max

OA

CA

QOA

YOA

YC

(gdm-3) 80.0 105.0 37.0 30.0+20a 47.0b 48.7+16.3c

(gdm-3) 45.0 72.0 0.0 5.3 0.0 12.8

(gdm-3) 6.5 11.0 14.5 15.5 23.1 25.3

(gdm-3) 24.0 18.5 41.8 55.7 34.0 42.0

(gdm-3) 0.0 0.0 0.7 0.9 0.0 0.0

(gdm-3h-1) 0.13 0.10 0.22 0.29 0.20 0.25

(gg-1) 0.69 0.56 1.13 1.25 0.72 0.80

(gg-1) 0.30 0.18 1.13 1.11 0.72 0.65

productivity [gdm -3 h -1 ]

Representation of maximum concentration biomass (X max, gdm-3), remaining substrate quantity (Sr, gdm-3), maximum concentration of oxalic acid (OA, gdm-3), maximum concentration of citric acid (CA, gdm-3), oxalic acid volumetric productivity (QOA, gdm-3h-1), total conversion yield of oxalic acid produced per substrate consumed (YOA, gg-1) and total conversion yield of oxalic acid produced per substrate added (YC, gg-1). Sr, CA, QOA, YOA and YC values and fermentation time are presented when the maximum concentration of oxalic acid had been achieved. Culture conditions: growth on Bioflo III stirred tank reactor (New Brunswick Scientific Co., USA), working volume of 2.5 dm3, temperature 30°C, aeration rate 1.5 dm3 min-1, stirrer speed 500 rpm, pH 4.5. a 20 gdm-3 of fatty acids was added after 48h of cultivation process b 47.0 gdm-3 of glycerin waste containing 13.0 gdm-3 glycerol and 34.0 gdm-3 fatty acids c 48.7 gdm-3 of glycerin waste containing 13.0 gdm-3 glycerol and 35.7 gdm-3 fatty acids, 16.3 -3 gdm of glycerin waste containing 2.0 gdm-3 glycerol and 14.3 gdm-3 fatty acids was added after 48h of cultivation process.

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0

20

40

60

80

100

120

140

160

180

200

time [h]

Figure 4. Time profiles of volumetric oxalic acid productivity (over each period of 24 h) during batch culture of A. niger XP in medium containing 8% (○) and 10% (●) pure glycerol, 4% (■) and 5% (□) fatty acids, 5% (+) and 7% (×) glycerin waste.

Biodiesel By-Products Used as Substrates for Oxalic Acid Production…

39

Conclusions Since large amounts of glycerin waste and its components are obtained in biodiesel production, it is justifiable to direct attention to possible ways of management of these waste products. There is no data in literature on glycerin waste obtained from biodiesel production, used as a substrate for oxalic acid production. The results of our studies show that both the fatty acids and the glycerin waste can be used for this purpose. Fatty acids proved to be the best carbon and energy sources for this biosynthesis. Slightly lower was the dynamics and efficiency of oxalic acid production from glycerin waste. However, it seems quite likely that glycerin waste can be an alternative for oxalic acid production, because it is easily available and cheap, and for this reason, it is justifiable to carry out research making headway to innovative and profitable biotechnologies with the use of this waste product. However, it should be noted that by-products of this type of biosynthesis, e.g. gluconic and citric acids can be a serious problem. When oxalic acid was produced with the use of glycerin waste, the broth contained only the desired product, i.e. oxalic acid. The concentration of glycerin waste is limited by insolubility in water of some glycerin components. Fed batch cultures or repeated cultures can markedly improve the technological parameters of oxalic acid production with the use of this substrate. The results obtained in our study are very promising. The use of glycerin waste as substrate for oxalic acid production will not only lower the production costs, but is undoubtedly a valuable biotechnological method for management of this waste product.

References [1] [2] [3]

[4] [5]

[6] [7]

Al-Zuhair, S. (2007).Production of biodiesel possibilities and challenges. Biofuels Bioproducts and Biorefining, 1(1), 57-66. Ambikadevi, V.R. and Lalithambika, M. (2000). Effect of organic acids on ferric iron removal from iron – stained kaolinite. Applied Clay Science, 16, 133-145. Bohlmann, J. T. and Cameselle, C. and Núñez, M. J. and Lema, J. M. (1998). Oxalic acid production by Aspergillus niger, part II: Optimisation of fermentation with milk whey as carbon source. Bioprocess Engineering, 19, 337 – 342. Bozbas, K. (2008). Biodiesel as an alternative motor fuel: Production and policies in the European Union. Renewable and Sustainable Energy Reviews, 12, 542-552. Cameselle, C. and Bohlmann, J. T. and Núńez, M. J. and Lema, J. M. (1998). Oxalic acid production by Aspergillus niger, part I: Influence of sucrose and milk whey as carbon source. Bioprocess Engineering, 19, 247 – 252. Guru, M. and Bilges, A. Y. (2001). Production of oxalic acid from sugar beet molasses by formed nitrogen oxides. Bioresource Technology, 77, 81-86. Haq, IU. and Ali, S. and Qadeer, M. A. and Iqbal, J. (2003). Stimulatory effect of alcohols (methanol and ethanol) on citric acid productivity by a 2-deoxy D-glucose resistant culture of Aspergillus niger GCB-47. Bioresource Technology, 86, 227–233.

40 [8]

[9]

[10]

[11]

[12] [13]

[14]

[15]

[16] [17]

[18]

[19]

[20]

[21]

[22]

Izabela Musiał and Waldemar Rymowicz Hattori, T. and Takahashi, S. and Kino, K. and Kirimura, K. and (2007). Production of oxalic acid by overexpresion of oxalacetate hydrolase gene (oahA) in Aspergillus niger WU-2223L. Journal of Biotechnology, 131(2), 175. Kubicek, C. and Schreferl – Kunar, G. and Wohrer, W. and Rohr, M. (1988). Evidence for a cytoplasmic pathway of oxalate biosynthesis in Aspergillus niger. Applied and Environmental Microbiology, 3, 633 – 637. Kumar, D. and Jain, VK. and Shanker, G. and Srivastava A. (2003). Utilization of fruits waste for citric acid production by solid state fermentation. Process Biochemistry, 38, 1725-1729. Leangon, S. and Maddox, I. S. and Brooks, J. D. and (1999). Influence of the glycolytic rate on production of citric acid and oxalate acid by Aspergillus niger in solid state fermentation. World Journal of Microbiology and Biotechnology, 15, 493 – 495. Mandal, S.K. and Banerjee P. C. (2005). Submerged production of oxalic acid from glucose by immobilized Aspergillus niger. Process Biochemistry, 40, 1605-1610. Mulligan, C.N. and Kamali, M. and Gibbs, B.F. (2004). Bioleaching of heavy metals from a low - grade mining ore using Aspergillus niger. Journal of Hazardous Materials, 110, 77-84. Pachauri, N, and He, B. (2006). Value – added utilization of crude glycerol from biodiesel production: A survey of current research activities. American Society of Agricultural and Biological Engineers ASABE, Paper No 066223, 2-15. Röhr, M. and Fesle, P. A. and Sinkha, J. and Panda, T. (2000). Comparative studies on citric acid production by Aspergillus niger and Candida lipolytica using molasses and glucose. Bioprocess Engineering, 22, 353–361. Rymowicz, W. and Lenart D. (2003). Oxalic acid production from lipids by a mutant of Aspergillus niger at different pH. Biotechnology Letters, 25, 955 – 958. Rymowicz, W. and Lenart, D. (2004). Comparison of different strains of Aspergillus niger for oxalic acid production from lipid substrates. Electronic Journal of Polish Agricultular Universities, Series Biotechnology, Volume 7, Issue 2. Rymowicz, W. and Lenart, D. (2004). Enhanced production of oxalic acid in Aspergillus niger by the addition of methanol. Electronic Journal of Polish Agricultural Universities, Series Biotechnology, Volume 7, Issue 2. Santhiya, D. and Ting, Y.-P. (2005). Bioleaching of spent refinery processing catalyst using Aspergillus niger with high-yield oxalic acid. Journal of Biotechnology, 116, 171-184. Santoro, R. and Cameselle, S. and Rodriguez – Couto, S., and Sanromán, Á. (1999). Influence of milk whey, nitrogen and phosphorus concentration on oxalic acid production by Aspergillus niger. Bioprocess Engineering, 20, 1 – 5. Strasser, H. and Burgstaller, W. and Schinner, F. (1994). High – yield production of oxalic acid for metal leaching processes by Aspergillus niger. FEMS Microbiology Letters, 119, 365-370. Wu, H-Y. and Ting, Y.-P. (2006). Metal extraction from municipal solid waste (MSW) incinerator fly ash – chemical leaching and fungal bioleaching. Enzyme and Microbial Technology, 38, 839-847.

In: Microbial Conversions of Raw Glycerol Editor: George Aggelis

ISBN 978-1-60692-392-4 © 2009 Nova Science Publishers, Inc.

Chapter V

Production of Omega-3 Polyunsaturated Fatty Acids from Biodiesel-derived Crude Glycerol by Microalgal and Fungal Fermentation Zhiyou Wen1,2, Denver J. Pyle1 and Sneha K. Athalye1 1

Department of Biological Systems Engineering, Institute for Critical Technology and Applied Science, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA 2

Abstract Crude glycerol represents the major byproduct of the biodiesel industry. As biodiesel production skyrockets, the market is being flooded with excess crude glycerol. Producing the omega-3 polyunsaturated fatty acids docosahexaenoic acid (DHA, C22:6, ω-3) and eicosapentaenoic acid (EPA, C20:5, ω-3) provides an opportunity to utilize this undervalued material. With many therapeutic benefits, DHA and EPA have been used as supplements in various human foods or animal feeds. Fish oil as the main source of DHA/EPA has several limitations such as undesirable taste and odor, heavy metal contamination, and potential shortage due to overfishing; thus, it is necessary to seek alternative sources to produce these two fatty acids. Our laboratory has been developing a microbial fermentation process to produce DHA and EPA from crude glycerol. The microalga Schizochytrium limacinum was used as DHA producer; the fungus Pythium irregulare as EPA producer. It was found that the major impurities contained in crude glycerol, methanol and soap, were inhibitory to algal/fungal culture, but they can be easily removed from the crude glycerol medium. The culture conditions, including medium composition and culture temperature, for S. limacinum and P. irregulare were optimized in flask cultures, with a high DHA yield of 4.91 g/L from S. limacinum and an EPA yield of 182 mg/L from P. irregulare. Overall, the works presented in this chapter show that biodiesel-derived crude glycerol can serve as a good carbon source for

42

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye microbial production of omega-3 fatty acids. Future research should focus on (1) elucidating the mechanisms of the inhibitory effects of soap on algal/fungal growth, (2) controlling fungal morphology to create the desired pellet form, (3) developing optimal fermenter cultures of the two species, and (4) implementing the process on a large scale.

1. Introduction The beneficial effects of omega-3 polyunsaturated fatty acids (ω-3 PUFAs) on human health have been well understood. In aquacultural industry, ω-3 PUFAs is an essential nutrient for farm-raise fish. Fish oil as a conventional source of ω-3 PUFAs is facing several challenges such as odor/taste problems, heavy metal contaminations, and limited supply. Microorganisms, including microalgae and lower fungi are the primary producers of ω-3 PUFAs. Producing ω-3 PUFAs by microalgae/fungi fermentation on biodiesel-derived crude glycerol provide a new way of producing these high value products, and also an alternative for utilizing the waste materials generated by biodiesel industry.

1.1. Structure and Significance of ω-3 PUFAs Omega-3 polyunsaturated fatty acids are fatty acids containing more than two double bonds with the last double bond located at the 3rd carbon atom from the methyl end. Eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA) are two common ω-3 PUFAs. The chemical structures of these two fatty acids are shown in Figure 1. In living cells, EPA and DHA are normally esterified to form complex lipid molecules.

Figure 1. Chemical structures of EPA and DHA.

Omega-3 fatty acids have proven beneficial effects in the prevention of cardiovascular diseases, cancers, Alzheimer’s, and schizophrenia (Emsley et al. 2003; Peet 2004; Simopoulos 2002; Su et al. 2003). The American Dietetic Association and Dietitians of Canada have recommended that 20 to 35% of daily energy should come from dietary fat, with an emphasis on consuming ω-3 PUFAs (Kris-Etherton and Innis 2007). Similarly, the American Heart Association has recommended that healthy consumers eat two fatty fish

Production of Omega-3 Polyunsaturated Fatty Acids…

43

meals per week and encourages patients with coronary heart disease to consume 1 g/day of EPA and DHA (Kris-Etherton et al. 2002). EPA is a precursor of a group of eicosanoids that are hormone-like substances such as prostaglandins (PG), thromboxanes (TX) and leukotrienes (LT); therefore, it plays an important role in regulating developmental and regulatory physiology. There is evidence that EPA is a potential anticachexia and anti-inflammatory agent (Babcock et al. 2000). It also possesses therapeutic activity against various cardiovascular diseases such as atherosclerosis (Bonaa et al. 1992) and thrombosis (Hostmark et al. 1988). DHA is a component of the photoreceptor cells of infants’ retina, and is also involved in the development of infants’ brain tissues such as synaptic vesicles, myelin, and mitochondria (Nettleton 1995); therefore, it is crucial in the development of the brain and vision in infants. For example, it has been shown that infants fed with DHA-supplemented formula exhibited improved visual acuity over infants fed with standard formula (Makrides et al. 1995). 1.2. Biosynthesis of ω-3 PUFAs Omega-3 fatty acids are synthesized through two steps (Figure 2). First is the de novo synthesis from acetate to oleic acid (18:1 ω-9), which is further desaturated by a Δ12 desaturase to form linoleic acid (18:2 ω-6) and a Δ15 desaturase to form α-linolenic acid (18:3 ω-3). The second is the formation of the ω-9, ω-6 or ω-3 fatty acid families from oleic acid, linoleic acid and α-linolenic acid through a number of stepwise desaturation and elongation steps. Nearly all biological systems, including microorganisms, insects, and higher plants and animals, are capable of de novo synthesis from acetate to short chain fatty acids, with oleic acid as the major product. Some higher plants are capable of synthesizing linoleic acid and αlinolenic acid. However, most higher plants and animals lack the requisite enzymes for longchain fatty acid synthesis and, thus, rarely contain fatty acids above C18 (Gill and Valivety 1997). Some algae and fungi species possess the desaturases and elongases required for the synthesis of various unsaturated fatty acids above C18. They are the primary producers of these fatty acids in nature (Apt and Behrens 1999; Blomquist et al. 1991; Chiou et al. 2001; Ratledge 1993). 1.3. Sources of ω-3 PUFAs The major source of ω-3 PUFAs available on the market is fish oil (Alonso and Maroto 2000). However, there have been major concerns about consumer acceptance of fish oil as a supplement or ingredient because of particular taste and odor (Kris-Etherton et al. 2002) or potential heavy metal contamination (AHA 2008; USEPA 2004). The sustainability of fish oil is another concern. The aquaculture industry extensively uses fish oil as the major form of ω-3 PUFAs in dietary formulations for farm-raised marine fish. At present, approximately 50% of fish oil on the market is used by the aquaculture industry (Tidwell and Allan 2001). The Food and Agriculture Organization of the UN predicts that global fish oil demand in

44

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

2015 will be 145% of historical global production capacity and will only continue to grow, while supply remains relatively stable (New and Wijkström 2002).

Figure 2. Biosynthetic pathways for formation of long chain fatty acids.

Because of the concerns associated with the consumption of fish oil as a source of ω-3 PUFAs, there has been extensive research into developing alternative sources of these important fatty acids. Fish cannot synthesize ω-3 PUFAs de novo. Some microorganisms such as algae or fungi are the primary producers of ω-3 PUFAs. Therefore, these microorganisms have been studied as potential sources of fatty acids (see Section 1.4). The fatty acids from these microbial sources can be extracted and used as ingredients in omega-3

Production of Omega-3 Polyunsaturated Fatty Acids…

45

fortified foods (Simopoulos 1999) or the organisms can be used directly as feed additives for animals such as laying hens (Chin et al. 2006) or farm-raised fish (Harel et al. 2002).

1.4. Microbial Production of Omega-3 Fatty Acids Production of omega-3 fatty acids through fermentation of microorganisms has been extensively investigated. Among DHA-producing microorganisms, the microalgae Crypthecodinium cohnii and Schizochytrium sp. are commonly used because they are capable of producing high levels of DHA under heterotrophic conditions. C. cohnii has been studied as a producer of DHA by several research groups (Ward and Singh 2005) and is used by Martek Biosciences to commercially produce DHA (Martek 2008). Schizochytrium sp. have also been identified and studied as a DHA producer. Biomass from these algae is currently sold as a DHA supplement for fish feed by Aquafauna Bio-Marine Inc. (Hawthorne, CA; http://www.aquafauna.com/Profiles-AlgaMac-3000.htm) and Advanced BioNutrition Corp. (Columbia, MD; http://www.abncorp.com/html/abn_dha.pdf). For EPA production, algae such as Nitzschia, Nannochloropsis, Navicula, Phaeodactylum, and Porphyridium have been investigated. The majority of these EPA-producing species are autotrophic and must be grown in photobioreactors (Ward and Singh 2005). The costs associated with these growing conditions are not favorable for industrial use. Some heterotrophic algal species such as Nitzschia laevis can produce EPA under heterotrophic conditions (Wen and Chen 2003). Fungi are another group of microorganisms capable of producing high levels of omega-3 fatty acids. For example, Mortierella alpina has been studied as a producer of EPA and an important omega-6 fatty acid, ARA (20:4, ω-6). It has been reported that high ARA yields, ranging from 4.5 g/L to 11.1 g/L, have been achieved by this organism (Ward and Singh 2005). Fungal species like Pythium, Thraustochytrium, and Entomophthora contain omega-3 fatty acids anywhere from 1% to 6.4% of dry biomass (Gill and Valivety 1997). It should be noted that recently the taxonomic definitions of Pythium and Thraustochytrium have been changed to the Kingdom Chromista (algae) (Alexopoulos et al. 1996). In this chapter, however, we still use the term fungus to define Pythium in order to keep our terminology consistent with other reports that treated this organism as a fungus.

2. Producing DHA from Microalgae Grown on Crude Glycerol Among the microalgae identified as producers of omega-3 fatty acids, Schizochytrium limacinum is a prolific producer of DHA. The alga was first screened from a mangrove area in the Yap Islands of Micronesia in 1994 (Honda et al. 1998). Further studies showed that this strain was a prolific producer of docosahexaenoic acid under heterotrophic culture conditions (Nakahara et al. 1996; Yaguchi et al. 1997). Both glucose and glycerol have been identified as good carbon sources for heterotrophic culture of this species (Yokochi et al. 1998). This leads us to investigate the possibility of using biodiesel-derived crude glycerol for the culture of this alga.

46

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

2.1. Experimental Protocols 2.1.1. Algal Strain, Medium and Culture Conditions Schizochytrium limacinum SR21 (ATCC MYA-1381) was maintained in ATCC 790 By+ medium containing 5 g/L glucose, 1.0 g/L yeast extract and 1.0 g/L peptone in artificial seawater. The composition of artificial seawater included NaCl (18 g/L), MgSO4 (2.44 g/L), KCl (0.6 g/L), NaNO3 (1.0 g/L), CaCl2. 2H2O (0.3g/L), KH2PO4 (0.05 g/L), Tris buffer (1.0 g/L), NH4Cl (0.027 g/L), vitamin B12 (15.0 x 10-8 g/L), chelated iron solution (3 mL/L), and PI metal solution (10 mL/L) (Starr and Zeikus 1993). The PI solution includes trace amounts of boron, cobalt, manganese, zinc, and molybdenum (Starr and Zeikus 1993). The medium was adjusted to pH 8, and then autoclaved at 121oC for 15 min. The cells were grown in 250mL Erlenmeyer flasks each containing 50 mL medium and incubated at 20oC in an orbital shaker set to 170 rpm. Subcultured cells were used as inoculum for future studies. The inoculum size was 10% of the total liquid volume in each flask. In later cell growth and DHA production studies, glucose was replaced as the carbon source by crude glycerol obtained from biodiesel production. All other components were the same as those used in the subculture. 2.1.2. Crude Glycerol Characterization and Pretreatment Crude glycerol was obtained from Virginia Biodiesel Refinery (West Point, VA). The refinery used alkali-catalyzed transesterification to produce biodiesel from soybean oil. The crude glycerol had a high pH level (11-12) with dark brown color. Because the producer used excess methanol to drive the transesterification towards a maximum biodiesel yield, the crude glycerol contained methanol as a major residue which accounted for ~12.8% (w/w) of the crude glycerol stream. Soap was also found in the crude glycerol stream due to side-reactions. The soaps can be split into free fatty acids and salt by adding a strong acid, i.e.,

(1) Considering the difficulty in determining the exact amount of soap dissolved in the crude glycerol solution, we used free fatty acid precipitated from the crude glycerol (Equation 1) as an estimation of the soap residue. It was found that soap accounted for 25.2% (w/w) of the crude glycerol stream. When the crude glycerol was mixed with artificial seawater (pH 7.5), some soap also precipitated from the solution. Depending on the experimental conditions, the soap either remained in or was removed from the medium. To prepare soap-free medium, the following procedures were used: (i) the glycerol was mixed with distilled water at a ratio of 1:4 (v/v) to reduce the viscosity of the fluid, (ii) the pH of the fluid was adjusted to 3 with hydrochloric acid to convert soap into free fatty acids that precipitated from the liquid, (iii) precipitated free fatty acids were separated from the crude glycerol solution after centrifugation at 5000

Production of Omega-3 Polyunsaturated Fatty Acids…

47

rpm, and (iv) other nutrients (mineral salts, nitrogen source, etc.) were added to the glycerol solution at the desired levels and pH was adjusted to 7.0. 2.1.3. Analysis Cell dry weight was determined by transferring a 5 mL cell suspension into a preweighed centrifuge tube that was then centrifuged at 3444 g for 5 min. The cell pellet was washed twice with distilled water, and then dried at 80oC to constant weight. Glucose concentration was determined by the 3,5-dinitrosalicylic method (Miller 1959). Glycerol concentration was determined by a Shimadzu Prominence HPLC System (Shimadzu Scientific Instruments, Inc. Columbia, MD) with a pulsed refractive index detector. An Aminex HPX-87H (Bio-Rad, Sunnyvale, CA) column was used with 0.1% (v/v) H2SO4 solution as mobile phase. The flow rate was controlled at 0.6 mL/min, and the column temperature was 65oC. Algal cells were harvested and freeze-dried for fatty acid analysis. Fatty acids methyl esters (FAME) were prepared by direct-methylation with 5% methanolic HCl (Christie 2003; Schreiner 2006; Ulberth and Henninger 1992) and determined by a Shimadzu 2010 gas chromatograph (Shimadzu Scientific Instruments, Columbia, MD) equipped with a flameionization detector and a SGE SolGel-WaxTM capillary column (30m×0.25mm×0.25um). The injector was kept at 250oC, with an injection volume of 1μl by split injection mode (10:1 ratio). The column temperature profile was as follows: 80oC for 0.5 min; raised to 175oC at 30oC/min; raised to 260 oC at 5oC/min; maintained for 6 min; raised to 280oC at 30oC/min; maintained for 1 min. The detector temperature was kept at 300oC. The fatty acids of the algae sample were identified by comparing the retention times with those of fatty acid standards (Nu-Chek Prep Inc., MN), and quantified by comparing their peak areas with that of the internal standard (C17:0) (Chi et al. 2007).

2.2. Feasibility of Producing DHA from Crude Glycerol S. limacinum was grown in medium containing crude glycerol; its growth performance and DHA production were compared with those of algae grown in glucose and pure glycerol cultures. As shown in Figure 3A, the cells grew well in crude glycerol when compared to glucose and pure glycerol. The cell dry weight reached the highest level in day 5-6. However, there were still relatively high levels of residual glucose or glycerol when the cells ceased growth (Figure 3B). It is believed that nitrogen was the limiting factor, as its level was lower than that reported by Yokochi et al. (1998). Cell growth kinetics and DHA production by S. limacinum on different substrates were also compared. As shown in Table 1, the specific growth rate, maximum cell dry weight, and biomass productivity of the crude glycerol culture were higher than those of the pure glycerol culture. However, the growth yield on glycerol (Yx/s) was lower than that on glucose, suggesting the algal cells could utilize glucose more efficiently for their growth. The fatty acid profile and DHA production of S. limacinum are shown in Table 2. The three carbon sources resulted in a similar fatty acid composition. C16:0 and C22:6 (DHA) were the major fatty acids, accounting for about 90% of total fatty acid (TFA). Small amounts

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

48

of C14:0, C18:0 and C22:5 (DPA) were also detected in the algal biomass. The total fatty acid content was about half of the cell dry weight. This fatty acid profile was similar to those reported previously for this algal species (Nakahara et al. 1996; Yokochi et al. 1998). In terms of DHA production, DHA cellular content of the three cultures was similar. However, due to the low biomass in pure glycerol culture, the DHA yield and productivity from the pure glycerol algal culture were lower than the other two cases (Table 2).

Cell dry weight (g/L)

25 Glucose Pure glycerol Crude glycerol

20 15 10 5 0

(A) 0

1

2

3

4

5

6

7

8

Culture time (day) Glucose or glycerol (g/L)

120 Glucose Pure glycerol Crude glycerol

100 80 60 40 20 0

(B) 0

1

2

3

4

5

6

7

8

Culture time (day) Chi et al., 2007, with permission from Elsevier. Figure 3. Time course of cell grwoth (A) and substrate consumption (B) of S. limacinum with glucose, pure glycerol, and crude glyerol as substrate.

2.3. Effects of Glycerol Impurities on Algal DHA Production 2.3.1. Effects of Methanol Methanol is one major impurity contained in crude glycerol streams. It has been reported that alcohols such as ethanol are beneficial for the growth of some microalgae producing DHA (de Swaaf et al. 2003). However, the effect of methanol on algal growth is unknown. This leads us to explore the effects of methanol on algal DHA production. Our initial experimental protocol called for the addition of methanol to the glycerol-containing medium,

Production of Omega-3 Polyunsaturated Fatty Acids…

49

Table 1. Cell growth, substrate consumption, and DHA production of S. limacinum using glucose, pure glycerol, and crude glycerol (at 90 g/L) as a substrate

Parameters a Specific growth rate, μ (day-1) Maximum cell dry weight, Xmax (g/L) Biomass productivity (g/L-day) Growth yield, Y x/s (g/g)

Substrate Crude glycerol 0.685±0.042 18.04±1.02 3.06 ± 0.17 0.284±0.016

Pure glycerol 0.562±0.009 14.43±0.14 2.41 ± 0.05 0.280±0.003

Glucose 0.575±0.012 18.47±0.25 3.08 ± 0.06 0.356±0.005

Chi et al., 2007, with permission from Elsevier. a Data are expressed as mean ± SD of three replicates.

Table 2. Fatty acid composition, total fatty acid (TFA) content, and DHA production parameters of S. limacinum with glucose, pure glycerol, and crude glycerol as substratea

Fatty acid 14:0 16:0 18:0 22:5 (n-6) 22:6 (n-3) TFA content DHA content DHA yield DHA productivity

Unit %TFA %TFA %TFA %TFA %TFA mg/g DW mg/g DW g/L g/L-day

Substrate Crude glycerol 5.16±0.04 54.70±0.41 1.04±0.03 5.48±0.10 33.62±0.34 505.69±45.14 170.45±11.27 3.07±0.19

Pure glycerol 4.29±0.05 52.45±3.38 0.47±0.06 4.97±0.04 37.82±3.43 447.93±44.14 171.37±0.96 2.47±0.03

Glucose 4.74±0.19 55.54±1.79 0.98±0.02 5.55±0.20 33.20±1.35 499.71±31.37 165.74±7.82 3.05±0.14

0.51±0.04

0.42±0.01

0.51±0.03

Chi et al., 2007, with permission from Elsevier. a. Data are expressed as mean ± SD of three replicates.

which was then autoclaved (at 121oC for 15 min). This protocol resulted in significant evaporation of methanol because of the autoclave process. To avoid the loss of methanol, we changed the protocol by spiking varying levels of methanol (sterilized by passing through a 0.2 μm filter) into medium which had been previously autoclaved. We used pure glycerol in the medium in order to avoid any possible interferences of other impurities contained in the crude glycerol while evaluating the effects of methanol. As shown in Figure 4, both the cell dry weight and DHA yield decreased as methanol concentration increased from 0 to 20 g/L. The results clearly indicate the negative effects of methanol on growth and DHA production of S. limacinum. Fortunately, under the typical crude glycerol levels used for algal culture (e.g., 60-90 g/L), autoclaving the media can evaporate methanol contained in the original glycerol stream. This allows the algae to thrive on the methanol-free crude glycerol.

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

50

Cell dry weight (g/L)

15 (A)

12 9 6 3 0 0

10

15

20

Methanol (g/L) 2

DHA yield (g/L)

1.6

(B)

1.2 0.8 0.4 0 0

10

15

20

Methanol (g/L)

Figure 4. Effects of methanol concentration on cell growth (A) and DHA production (B) of S. limacinum (grown in 70 g/L of pure glycerol). Data are means of three replicates and error bars show the standard deviations.

2.3.2. Effects of Soap The effects of soap on algal DHA production were investigated at different crude glycerol levels. The soap in the crude glycerol was either kept in the medium (soapcontaining); or removed from the medium (soap-free). As shown in Figure 5, in the lower range of crude glycerol concentrations (10 and 20 g/L), the cell dry weight on soap-free and soap-containing medium was similar. When crude glycerol concentration increased to 40 g/L and 60 g/L, the maximum cell dry weight and biomass productivity for soap-containing culture were lower than those for soap-free cultures. When glycerol concentration exceeded 60 g/L, the growth in soap-containing medium was totally inhibited.

Production of Omega-3 Polyunsaturated Fatty Acids…

51

10

Cell dry weight (g/L)

8

Soap-free Soap-containing

6 4 2 0 10

20

40

60

Crude glyceorl (g/L)

Figure 5. Comparison of cell growth of S. limacinum growing in crude glycerol medium with and without-soap. Data are means of three replicates and error bars show the standard deviations.

The soap had a significant effect on the fatty acid profile of the algal biomass. As shown in Table 3, the algal biomass from soap-free culture had a negligible proportion (<1% of TFA) of C18 fatty acids. When soap was included in the medium, the proportions of C18 fatty acids accounted for 25-35% of TFA. Considering the fact that the soap itself contains a significant amount of C18s (Table 3), the difference in fatty acid profiles between the soapfree and soap-containing algal biomass indicate that S. limacinum was capable of absorbing (or binding) those soap-C18s into the cells. However, the proportion of long chain fatty acids (C22:5 and C22:6) in the soap-containing algae was lower than those in the soap-free algae (Table 3), suggesting that those absorbed C18s were not further elongated by the algal cells. Table 3. Fatty acid composition (% of total fatty acid, TFA) of the soap contained in crude glycerol and algal biomass growing in soap-free and soap containing medium a

a

Fatty acid

Soap

b

Algal biomass in soap-free medium

c

14:0 16:0 18:0 18:1 18:2 18:3 (n-3) 22:5 (n-6) 22:6 (n-3)

11.61 ± 1.05 4.01 ± 0.89 22.97 ± 0.39 54.35 ± 1.47 7.05 ± 0.11

3.32 ± 0.24 44.11 ± 0.53 0.93 ± 0.05

-

7.35 ± 0.27 44.30 ± 0.53

3.14 ± 0.32 42.35 ± 4.28 1.91 ± 0.50 7.43 ± 1.07 17.68 ± 3.21 2.72 ± 0.61 3.55 ± 0.36 21.23 ± 2.03

-

Algal biomass in soap-containing medium

Data are expressed as mean ± SD of three replicates. b,c 20 g/L crude glycerol was used; the biomass from soap-free and soap-containing medium has a similar total fatty acid content of 280-290 mg/g dry biomass.

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

52

This phenomenon was also reported by Hauvermale et al. (2006) in which radiolabeled C16:0, C18:1, or C18:3 fatty acids were fed to Schizochytrium sp. but none of the labels appeared in longer-chain PUFA (Hauvermale et al. 2006). Figure 6 shows the algal DHA production in soap-containing and soap-free medium. Similar to the trends in algal growth (Figure 5), the DHA yield from the soap-containing algae was lower than the yield obtained from the soap-free algae (Figure 6).

DHA yield (g/L)

1.60

1.20

Soap-free Soap-containing

0.80

0.40

0.00 10

20

40

60

Crude glycerol (g/L)

Figure 6. Comparison of DHA production of S. limacinum growing in crude glycerol medium with and without-soap. Data are means of three replicates and error bars show the standard deviations.

The results clearly indicated negative effects of soap on growth and DHA production of S. limacinum. It has been reported that surfactants, including soap, inhibited the growth of various algal species (Ukeles 1965; Yamane et al. 1984), which might be caused by the complex interaction between the cell wall/membrane and the surfactants (Ukeles 1965). However, the mechanisms for the negative effects of soap on S. limacinum are still not well understood, further study of the interaction between soaps and the cell membrane of this species are needed.

2.4. Optimization of Algal Culture Conditions for Producing DHA from Crude Glycerol The above results clearly show that S. limacinum is capable of producing DHA from biodiesel-derived crude glycerol. The culture conditions for this alga were further optimized to enhance DHA production. Both methanol and soap were not included in the medium as they have inhibitory effects on algal DHA production. Methanol can be removed during the autoclave process, while soaps were removed according to the procedures described in Section 2.1.2.

Production of Omega-3 Polyunsaturated Fatty Acids…

53

The effects of crude glycerol concentration on cell growth and DHA production were studied. The crude glycerol stream (after soap removal, but before methanol removal) was added to the medium at 75 g/L to 150 g/L level. The “real” glycerol concentrations in the medium were about 15% lower than the corresponding crude glycerol levels due to loss of methanol. As shown in Table 4, crude glycerol at 75 g/L to 100 g/L resulted in high cell dry weight and biomass productivity. Algal growth was inhibited when the crude glycerol concentration exceeded 100 g/L. The DHA yield and productivity had a trend similar to that of cell dry weight. Table 4. Effects of crude glycerol concentration on cell growth and DHA production by S. limacinum a Crude glycerol (g/L) 75 100 125 150

“Real” glycerol (g/L)

Cell dry weight (g/L)

64 85 106 128

18.7± 1.2 18.6 ± 0.9 7.7 ± 0.5 6.3 ± 0.6

Biomass productivity (g/L-day) 3.11±0.20 3.10±0.15 1.28±0.08 1.05±0.10

DHA yield (g/L)

DHA productivity (g/L-day)

3.05±0.28 2.71±0.35 1.43±0.22 0.96±0.17

0.50±0.05 0.44±0.06 0.23±0.04 0.16±0.03

Chi et al., 2007, with permission from Elsevier. a Cell dry weight and DHA yield are expressed as mean ± SD of three replicates.

The optimal level of crude glycerol obtained here was lower than that reported by (Yokochi et al. 1998), in which 90 g/L to 120 g/L glycerol resulted in highest DHA production. The reason might be that 100% seawater salt were used in this work, while 50% seawater salt was used by Yokochi et al. (1998). The low salt level provided a lower osmotic pressure and thus, allowed more glycerol to be added. We further used a statically-based experimental design for optimizing other parameters influencing the algal DHA production. The detailed experimental procedures have been described previously (Chi et al., 2007). In summary, a Plackett-Burman design was first used to screen the factors that significantly influence the algal culture, the factors to be screened included the concentrations of NaCl, KH2PO4, corn steep liquor, ammonium acetate, MgSO4.7H2O, KCl, NaNO3, CaCl2.2H2O, Tris Buffer, NH4Cl, Vitamin B12, trace metal solution, chelated iron solution, medium pH, and culture temperature. It was found that temperature, trace metal solution, ammonium acetate, and NH4Cl had significant effects (P < 0.1) on DHA yield. Then, a central composite design was used to optimize these four parameters, their optimal values were determined as 30 mL/L of PI solution, 0.04 g/L of NH4Cl, 1.0 g/L of ammonium acetate, and 19.2oC culture temperature. The optimized culture conditions resulted in 22.1 g/L of algal biomass and a DHA yield of 4.91 g/L. Lastly, the algal culture was performed in the optimal conditions as predicted by the experimental design; the deviation between the experimental data and the predicted values was less than 5% (Chi et al., 2007). In summary, the above results indicate that producing DHA-containing algae from less expensive crude glycerol from the biodiesel industry is a feasible and durable option for

54

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

utilizing this glycerol. However, from a nutritional point of view, balanced levels of both DHA and EPA in human food or animal feed are desired. In the following section, the potential of producing EPA from crude glycerol was reported.

3. Producing EPA from Fungi Grown on Crude Glycerol During the development of a process to produce EPA from crude glycerol, we first tested the alga Phaeodactylum tricornutum as glycerol was reported to be a good carbon source for this species in mixotrophic growth conditions (Ceron Garcia et al. 2006; Garci et al. 2000; Sevilla et al. 2004). The alga was grown in medium containing glucose, pure glycerol, or biodiesel-derived crude glycerol, respectively. It was found that the algal growth performance in all three substrates was poor; the cell growth in crude glycerol-containing medium was the worst, which might be due to inhibitory effects of unknown impurities, such as various ions, contained in the crude glycerol (data not shown). Therefore, P. tricornutum was not further investigated in this chapter. Some fungal species such as Mortierella and Pythium are capable of producing high levels of EPA (Bajpai et al. 1992; Obrien et al. 1993; Shimizu et al. 1988a; Shimizu et al. 1988b; Stinson et al. 1991). Compared to algal cultures, fungal cells are more resistant to negative growing environments and require simpler medium nutrients. In addition, fungal culture avoids light limitation, which often happens in algal photoautrophic or mixotrophic cultures. It has been reported that Pythium irregulare can grow and produce EPA on various substrates such as crude soybean oil, sucrose waste stream, and soybean waste stream (Cheng et al. 1999). The wide adaptability of this species leads us to investigate the potential of using crude glycerol for EPA production by P. irregulare.

3.1. Experimental Protocols 3.1.1. Fungal Strain and Culture Conditions Pythium irregulare (ATCC 10951) was grown on an agar slant (with 20 g/L glucose and 10 g/L yeast extract) at 25oC for 7 days. The spores on the slant were suspended with sterilized water and maintained at 4oC for later use. To investigate EPA production from crude glycerol, the spore solution was inoculated into medium containing different concentrations of crude glycerol and yeast extract. The medium was adjusted to pH 6.0 before being autoclaved at 121oC for 15 min. The cells were grown in 250-mL Erlenmeyer flasks, each containing 50 mL of medium, incubated at 25oC in an orbital shaker set to 170 rpm. 3.1.2. Crude Glycerol Characterization and Pretreatment The crude glycerol used for fungal culture was the same as that used for algal DHA production. Its characterization has been reported in Section 2.1.2. The procedures for

Production of Omega-3 Polyunsaturated Fatty Acids…

55

preparation of medium containing crude glycerol (soap-free) were the same as those described in Section 2.1.2, except that no micronutrients were included in the medium. 3.1.3. Analysis The fungal mycelium in each flask was filtered through filter paper, rinsed twice with distilled water, and freeze-dried to constant weight. The freeze-dried fungal biomass was used for measuring the cell dry weight and then used for fatty acid analysis. The FAME preparation and subsequent GC analysis were the same as for algal biomass as described in Section 2.1.3. The determination of glycerol concentration was also the same as that described in Section 2.1.3.

3.2. Feasibility of Producing EPA from Crude Glycerol Growth performance and EPA production of P. irregulare were evaluated by growing the fungus in media containing glucose, pure glycerol, or crude glycerol as a carbon source. As shown in Figure 7A, the glucose-containing medium resulted in the highest cell growth; crude glycerol and pure glycerol also supported cell growth, although the biomass was lower than in glucose culture. In terms of substrate consumption, the initial concentrations of the three carbon sources were around 20 g/L, the “real” glycerol concentration in crude glycerol culture was lower than 20 g/L due to the impurities contained in the crude glycerol fluid. As shown in Figure 7B, the fungi fed with crude glycerol consumed all the substrate, while the other two cultures left residual substrate when the cells reached the stationary phase. This result indicates glycerol is the limiting substrate; therefore, increasing the crude glycerol level may further enhance cell growth. The fatty acid profiles of the fungal biomass derived from different carbon sources were also compared. As shown in Table 5, the major fatty acids were C16:0 and C18s. EPA accounted for 6-10% of total fatty acids in the biomass. The EPA level in terms of content, yield, and productivity from crude glycerol and pure glycerol cultures were similar, with both being lower than those from glucose medium. In the following sections, we optimized the culture conditions for the crude glycerol-culture to further increased the EPA production level from crude glycerol.

3.3. Optimization of Fungal Culture Conditions for Producing EPA from Crude Glycerol The growth conditions, including medium composition and temperature, for the fungal culture were further optimized to enhance the EPA production. Our preliminary experiments have shown that no fungal growth was observed in soap-containing medium, thus, soap was removed from the crude glycerol medium. Also, the methanol residue within the crude glycerol was completely evaporated during autoclaving. Unlike the algae S. limacinum that has a complex nutrients requirement, P. irregulare requires a relatively simple medium composition, with glycerol and yeast extract being the only two components. The cell growth and EPA production in medium containing different

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

56

combinations of these two components were investigated. As shown in Table 6, at low yeast extract levels (e.g., 5 and 10 g/L), the cell dry weight increased with increasing glycerol concentration. When yeast extract was at high levels (15 and 20 g/L), the cell dry weight remained stable, independent of the glycerol concentration. In terms of EPA production, the EPA content was significantly influenced by the yeast extract concentration; low levels (e.g., 5 g/L) resulted in a higher EPA content. This result agreed with algal cultures such as Botryococcus braunii, Dunaliella bardawil, and Dunaliella salina in which a higher percentage of EPA was obtained under low nitrogen levels (Benamotz et al. 1985). Combining cell dry weight and EPA content, the highest EPA yield (ca. 80-90 mg/L) and EPA productivity (11-12 mg/L-day) were obtained in the range of 30-40 g/L crude glycerol and 5-10 g/L yeast extract (Table 6); therefore, we selected 30 g/L crude glycerol and 10 g/L yeast extract in the following temperature study.

6 (A)

Cell dry weight (g/L)

5 4 3 2

Glucose Pure glycerol Crude Glycerol

1 0 0

1

2

3 4 5 6 7 Culture time (day)

8

9

Glucose or glycerol (g/L)

25 Glucose Pure glycerol Crude glycerol

20 15

(B) 10 5 0 0

1

2

3 4 5 6 Culture time (day)

7

8

9

Figure 7. Time course of cell growth (A) and substrate consumption (B) of P. irregulare with glucose, pure glycerol, and crude glycerol as substrate.

Production of Omega-3 Polyunsaturated Fatty Acids…

57

Table 5. Fatty acid composition, total fatty acid (TFA) content, and EPA production parameters of P. irregulare with glucose, pure glycerol, and crude glycerol as substratesa

Fatty acid 14:0 16:0 16:1 18:0 18:1 18:2 20:4 (ARA) 20:5 (EPA) TFA content EPA content EPA yield

EPA productivity a.

Unit %TFA %TFA %TFA %TFA %TFA %TFA %TFA %TFA mg/g DW mg/g DW mg/L mg/L-day

Substrate Crude glycerol 7.44 ± 0.056 25.07 ± 0.51 15.12 ± 0.73 1.41 ± 0.04 20.69 ± 0.33 16.81 ± 0.35 6.21 ± 0.04 7.26 ± 0.5 260.63 ± 23.3 18.92 ± 0.11 47.86 ± 2.02 7.99 ± 0.34

Pure glycerol 5.56 ± 1.95 25.51 ± 0.81 10.05 ± 2.21 18.66 ± 14.27 23.92 ± 12.1 6.43 ± 8.39 3.35 ± 2.01 6.53 ± 1.01 212.62 ± 21.17 14.08 ± 0.84 41.81 ± 1.56 6.97 ± 0.26

Glucose 8.23 ± 0.36 26.1 ± 0.79 7.10 ± 0.64 2.68 ± 0.10 17.72 ± 0.95 18.73 ± 0.75 8.7 ± 0.55 9.90 ± 0.42 198.19 ± 7.81 19.71 ± 1.12 90.09 ± 5.65 15.01 ± 0.92

Data are expressed as mean ± SD of three replicates. DW: cell dry weight.

Table 6. Effect of crude glycerol and yeast extract concentrations on growth and EPA production of P. irregulare a Crude glycerol (g/L) 10 20 30 40 20 30 40 50 20 30 40 50 20 30 40 a

Yeast extract (g/L) 5 5 5 5 10 10 10 10 15 15 15 15 20 20 20

Biomass (g/L)

EPA content (mg/g DW)

EPA yield (mg/L)

EPA productivity (mg/L-day)

2.31 ± 0.01 2.80 ± 0.28 4.62 ± 0.22 5.78 ± 0.43 4.22 ± 0.45 6.27 ± 0.51 6.48 ± 0.46 5.34 ± 0.28 4.96 ± 0.05 4.46 ± 0.24 5.04 ± 0.88 4.83 ± 0.30 6.04 ± 0.14 6.63 ± 0.21 6.65 ± 0.79

18.86 ± 0.89 18.66 ± 1.21 18.16 ± 0.76 16.39 ± 1.19 13.91 ± 1.41 12.67 ± 1.47 13.50 ± 1.24 13.17 ± 1.89 12.50 ± 0.74 11.02 ± 0.66 12.13 ± 0.22 10.77 ± 0.69 10.20 ± 1.09 9.54 ± 1.02 9.02 ± 0.43

43.35 ± 2.14 52.98 ± 3.89 83.84 ± 6.11 88.32 ± 6.57 58.01 ± 6.78 86.20 ± 8.85 87.54 ± 8.08 70.44 ± 9.52 62.06 ± 3.07 49.19 ± 1.50 61.15 ± 8.83 47.18 ± 4.02 69.85 ± 5.77 62.38 ± 6.56 60.02 ± 6.05

7.22 ± 0.35 7.56 ± 0.95 10.49 ± 0.77 11.04 ± 0.82 10.75 ± 1.13 12.31 ± 1.26 12.50 ± 1.15 10.06 ± 1.36 8.87 ± 0.44 7.02 ± 0.21 8.74 ± 1.21 6.74 ± 0.57 11.41 ± 0.82 8.91 ± 0.93 8.57 ± 0.98

The “real” glycerol content was 85% of crude glycerol concentration. DW: cell dry weight.

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

58

Temperature is another important factor influencing cell growth and lipid composition of many EPA-producing microorganisms. In general, higher temperatures lead to rapid cell growth while lower temperatures result in the accumulation of omega-3 fatty acids. Low temperature stress usually leads to a higher level of intracellular O2, which favors the synthesis of long chain omega-3 fatty acids (Feng and Johns 1991; Singh and Ward 1997). A temperature shift strategy has been used to enhance the overall EPA production of P. irregulare grown on glucose (Stinson et al. 1991). Here, we used this “temperature shift” strategy to investigate the effect of temperature on the growth and EPA production of P. irregulare when crude glycerol was used as a carbon source. The fungal cells were grown at 25oC for the first 5 days and then switched to 20oC or 15oC or kept at 25°C. As shown in Figure 8, when the temperature was switched to 20oC, the fungal cells continued to grow for the next two days and reached the highest level of 7.33 g/L. This fungal biomass was even higher than that obtained at 25oC. The cell dry weight began to drop when the temperature was switched to 15oC, indicating that cell growth was inhibited at this temperature level. As for the EPA production (Table 7), 20oC also resulted in a much higher EPA content than the other temperature levels. The EPA content at 15oC was just slightly higher than the EPA content at 25oC. The EPA yield and productivity at different temperature levels were similar to the fungal growth and EPA content, i.e., 20oC resulted in the highest EPA yield and production. Overall, the above results clearly show that using the temperature switching strategy, 20oC was the ideal temperature at the later stage of fungal culture when using crude glycerol as a carbon source, achieving a highest EPA yield and productivity of 182 mg/L and 26 mg/L-day. 10

Cell dry weight (g/L)

8 6 4 2 0 0

1

2

3

4

5

6

7

8

9

Culture time (day)

Figure 8. Cell grwoth of P. irregulare at different temperature levels. (The fungi were grown at 25oC for 5 days and then switched to 15oC, 20oC, or 25 oC). Symbols for different temperature levels: -U-: 25oC; -¡- 15oC; -„- 20oC.

Production of Omega-3 Polyunsaturated Fatty Acids…

59

Table 7. Effects of temperature strategy on EPA production of P. irregulare (DW: cell dry weight)

EPA conte nt (mg/ g DW) EPA yield (mg/ L) EPA prod uctiv ity (mg/ Lday)

Temperature 25oC 14.2 4± 0.01

89.3 6± 7.25 14.8 9± 1.21

20oC 24.93 ± 2.17

15oC 15.6 3± 1.81

182.5 0 ±10.0 6 26.07 ± 1.44

84.6 5± 8.27 14.1 1± 1.38

The above results demonstrate that crude glycerol can be used as a carbon source for EPA production by the fungi P. irregulare. The production level is comparable to that from glucose culture. The EPA-fungal biomass, together with the DHA-algal biomass, provides an alternative use for biodiesel-derived crude glycerol by producing a balanced omega-3 source for fortified food or animal feed.

4. Prospects for Producing DHA/EPA Using Crude Glycerol The work presented in this chapter provides a “proof-of-concept” that crude glycerol from the biodiesel industry can be used for producing high value omega-3 polyunsaturated fatty acids by microalgal and fungal cultures. The optimized flask cultures of the microalga Schizochytrium limacinum resulted in 4.91 g/L of DHA yield and 0.82 g/L-day of DHA productivity; while the fungus Pythium irregulare produced 182 mg/L of EPA with a productivity of 26 mg/L-day. In the future, research efforts need to concentrate on the commercialization of omega-3 fatty acid production from crude glycerol. First, the mechanisms of the inhibitory effects of soap on algal/fungal growth need to be studied. Indeed, soap contained a significant amount of C16 and C18 length fatty acids (Table 3) that could be ideal precursors for synthesis of long chain fatty acids, including EPA and DHA. Although soap had a negative effect on algal/fungal growth and DHA/EPA production,

60

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

elucidation of the mechanism of this negative effect will help to develop a strategy to utilize the C16 and C18 fatty acids contained in the soap. Second, the morphological properties of the fungus P. irregulare need to be studied. During the fungal culture described in this chapter, the fungal cells tended to form undesirable cotton-like (filamentous) mycelia which will eventually hinder large-scale fermenter culture due to heterogeneous growth conditions and oxygen/nutrient transfer limitations. Therefore, transforming the fungal mycelia into small and uniform pellets is desirable and needs to be studied. Third, an optimized culture protocol in fermenter operations should be developed, which will contribute to developing pilot- and large-scale algal/fungal fermentation processes. In fermenter culture, parameters such as dissolved oxygen and shear stress/stir speed of impellers can be controlled to lead to greater EPA/DHA production.

5. Conclusion As glycerol is the major byproduct in the biodiesel manufacturing process, the disposal of crude glycerol has been a major issue faced by biodiesel producers. The work presented in this chapter demonstrates that crude glycerol can be converted to high value omega-3 polyunsaturated fatty acids through algal and fungal fermentation. It provides a promising alternative use for crude glycerol and a cost effective way to produce omega-3 fatty acids. In the future, the successful implementation of algal/fungal omega-3 production processes depends on (i) fully utilizing the soap residue to boost the algal/fungal culture (ii) morphological control of the fungal biomass and (iii) development of optimal fermenter cultures. As a continuation of this study, the development of omega-3 fortified food or animal feeds from the glycerol-derived EPA-fungal or DHA-algal biomass need to be explored.

References AHA. (2008). Fish, levels of mercury, and omega-3 fatty acids. Dallas TX: American Heart Association. Available from: http://www.americanheart.org/presenter.jhtml?identifier=3013797 (Accessed on Feburary, 12, 2009) Alexopoulos, C.J., Mims, C.W., and Blackwell, M. (1996). Introductory Mycology (4th Edition), New York, New York: John Wiley and Sons, Inc. Alonso, D.L. and Maroto, F.G. (2000). Plants as 'chemical factories' for the production of polyunsaturated fatty acids. Biotechnology Advances, 18, 481-497. Apt, K.E. and Behrens, P.W. (1999). Commercial developments in microalgal biotechnology. Journal of Phycology, 35, 215-226. Babcock, T., Helton, W.S, and Espat, N.J. (2000). Eicosapentaenoic acid (EPA): An antiinflammatory omega-3 fat with potential clinical applications. Nutrition, 16, 1116-1118.

Production of Omega-3 Polyunsaturated Fatty Acids…

61

Bajpai, P.K., Bajpai, P. and Ward, O.P. (1992). Optimization of culture conditions for production of eicosapentaenoic acid by Mortierella elongata NRRL-5513. Journal of Industrial Microbiology, 9, 11-17. Benamotz, A., Tornabene, T.G. and Thomas, W.H. (1985). Chemical profile of selected species of microalgae with emphasis on lipids. Journal of Phycology, 21, 72-81. Blomquist, G.J., Borgeson, C.E. and Vundla, M. (1991). Polyunsaturated fatty-acids and eicosanoids in insects. Insect Biochemistry, 21, 99-106. Bonaa, K.H., Bjerve, K.S. and Nordoy, A. (1992). Habitual fish consumption, plasma phospholipid fatty-acids, and serum-lipids - the tromso study. American Journal of Clinical Nutrition, 55,1126-1134. Ceron Garcia, M.C., Camacho, F.G., Miron, A.S., Sevilla, J.M.F., Chisti, Y. and Grima, E.M. (2006). Mixotrophic production of marine microalga Phaeodactylum tricornutum on various carbon sources. Journal of Microbiology and Biotechnology, 16, 689-694. Cheng, M.H., Walker, T.H., Hulbert, G.J. and Raman, D.R. (1999). Fungal production of eicosapentaenoic and arachidonic acids from industrial waste streams and crude soybean oil. Bioresource Technology, 67, 101-110. Chi, Z., Pyle, D., Wen, Z., Frear, C. and Chen, S. (2007). A laboratory study of producing docosahexaenoic acid from biodiesel-waste glycerol by microalgal fermentation. Process Biochemistry, 42, 1537-1545. Chin, H.J., Shen, T.F., Su, H.P. and Ding, S.T. (2006). Schizochytrium limacinum SR-21 as a source of docosahexaenoic acid: optimal growth and use as a dietary supplement for laying hens. Australian Journal of Agricultural Research, 57, 13-20. Chiou, S.Y., Su, W.W. and Su, Y.C. (2001). Optimizing production of polyunsaturated fatty acids in Marchantia polymorpha cell suspension culture. Journal of Biotechnology, 85, 247-257. Christie, W.W. (2003). Lipid Analysis: Isolation, Separation, Identification and Structural Analysis of Lipids. Bridgwater, U.K.: The Oily Press. de Swaaf, M.E., Pronk, J.T. and Sijtsma, L. (2003). Fed-batch cultivation of the docosahexaenoic-acid-producing marine alga Crypthecodinium cohnii on ethanol. Applied Microbiology and Biotechnology, 61, 40-43. Emsley, R., Oosthuizen, P., van Rensburg, S.J. (2003). Clinical potential of omega-3 fatty acids in the treatment of schizophrenia. CNS Drugs, 17, 1081-1091. Feng, C. and Johns, M.R. (1991). Effect of C/N ratio and aeration on the fatty-acid composition of heterotrophic Chlorella sorokiniana. Journal of Applied Phycology, 3, 203-209. Garci, M.C.C., Sevilla, J.M.F., Fernandez, F.G.A., Grima, E.M. and Camacho, F.G. (2000). Mixotrophic growth of Phaeodactylum tricornutum on glycerol: growth rate and fatty acid profile. Journal of Applied Phycology, 12, 239-248. Gill, I. and Valivety, R. (1997). Polyunsaturated fatty acids 1. Occurrence, biological activities and applications. Trends in Biotechnology, 15, 401-409. Harel, M., Koven, W., Lein, I., Bar, Y., Behrens, P., Stubblefield, J., Zohar, Y. and Place, A.R. (2002). Advanced DHA, EPA and ArA enrichment materials for marine aquaculture using single cell heterotrophs. Aquaculture, 213, 347-362. Hauvermale, A., Kuner, J., Rosenzweig, B., Guerra, D., Diltz, S. and Metz, J.G. (2006). Fatty acid production in Schizochytrium sp.: Involvement of a polyunsaturated fatty acid synthase and a type I fatty acid synthase. Lipids, 41, 739-747.

62

Zhiyou Wen, Denver J. Pyle and Sneha K. Athalye

Honda, D., Yokochi, T., Nakahara, T., Erata, M. and Higashihara, T. (1998). Schizochytrium limacinum sp. nov., a new thraustochytrid from a mangrove area in the west Pacific Ocean. Mycological Research, 102, 439-448. Hostmark, A.T., Bjerkedal, T., Kierulf, P., Flaten, H. and Ulshagen, K. (1988). Fish Oil and Plasma-Fibrinogen. British Medical Journal, 297, 180-181. Kris-Etherton, P.M., Harris, W.S. and Appel, L.J. (2002). Fish consumption, fish oil, omega-3 fatty acids, and cardiovascular disease. Circulation, 106, 2747-2757. Kris-Etherton, P.M. and Innis, S. (2007). Position of the American dietetic association and dietitians of Canada: Dietary fatty acids. Journal of the American Dietetic Association, 107, 1599-1611. Makrides, M., Neumann, M., Simmer, K., Pater, J. and Gibson, R. (1995). Are long-chain polyunsaturated fatty-acids essential nutrients in infancy? Lancet, 345, 1463-1468. Martek. (2008). History of Martek Columbia, MD: Martek Biosciences Corporation. Available from: http://aboutmartek.martek.com/history/.(Accessed on Feburary 12, 2009). Miller, G. (1959). Use of dinitrosalicylic acid reagent for determination of reducing sugar. Analytical Chemistry, 31, 426-429. Nakahara, T., Yokochi, T., Higashihara, T., Tanaka, S., Yaguchi, T. and Honda, D. (1996). Production of docosahexaenoic and docosapentaenoic acids by Schizochytrium sp isolated from Yap islands. Journal of the American Oil Chemists Society, 73, 1421-1426. Nettleton, J.A. (1995). Omega-3 Fatty Acids and Health. New York: Chapman and Hall. New, M.B. and Wijkström, U.N. (2002). Use of fishmeal and fish oil in aquafeeds: further thoughts on the fishmeal trap. Rome, Italy. Report nr FAO Fisheries Circular No. 975. Obrien, D.J., Kurantz, M.J. and Kwoczak, R. (1993). Production of eicosapentaenoic acid by the filamentous fungus Pythium irregulare. Applied Microbiology and Biotechnology, 40, 211214. Peet, M. (2004). Nutrition and schizophrenia: beyond omega-3 fatty acids. Prostaglandins Leukotrienes and Essential Fatty Acids, 70, 417-422. Ratledge, C. (1993). Single-cell oils - have they a biotechnological future. Trends in Biotechnology, 11, 278-284. Schreiner, M. (2006). Optimization of solvent extraction and direct transmethylation methods for the analysis of egg yolk lipids. International Journal of Food Properties, 9, 573-581. Sevilla, J.M.F., Garcia, M.C.C., Miron, A.S., Belarbi, E., Camacho, F.G. and Grima, E.M. (2004). Pilot-plant-scale outdoor mixotrophic cultures of phaeodactylum tricornutum using glycerol in vertical bubble column and airlift photobioreactors: Studies in fed-batch mode. Biotechnology Progress, 20, 728-736. Shimizu, S., Kawashima, H., Shinmen, Y., Akimoto, K. and Yamada, H. (1988a). Production of eicosapentaenoic acid by mortierella fungi. Journal of the American Oil Chemists Society, 65, 1455-1459. Shimizu, S., Shinmen, Y., Kawashima, H., Akimoto, K. and Yamada, H. (1988b). Fungal mycelia as a novel source of eicosapentaenoic acid - activation of enzyme(s) involved in eicosapentaenoic acid Production at low-temperature. Biochemical and Biophysical Research Communications, 150, 335-341. Simopoulos, A.P. (1999). Essential fatty acids in health and chronic disease. American Journal of Clinical Nutrition, 70, 560S-569S.

Production of Omega-3 Polyunsaturated Fatty Acids…

63

Simopoulos, A.P. (2002). The importance of the ratio of omega-6/omega-3 essential fatty acids. Biomedicine and Pharmacotherapy, 56, 365-379. Singh, A. and Ward, O.P. (1997). Microbial production of docosahexaenoic acid (DHA, C22:6). Advances in Applied Microbiology, 45, 271-312. Starr, R.C. and Zeikus, J.A. (1993). UTEX - the culture collection of algae at the University of Texas at Austin 1993 List of Cultures. Journal of Phycology, 29, 1-106. Stinson, E.E., Kwoczak, R. and Kurantz, M.J. (1991). Effect of cultural conditions on production of eicosapentaenoic acid by Pythium irregulare. Journal of Industrial Microbiology, 8, 171178. Su, K.P., Huang, S.Y., Chiu, C.C. and Shen, W.W. (2003). Omega-3 fatty acids in major depressive disorder - A preliminary double-blind, placebo-controlled trial. European Neuropsychopharmacology, 13, 267-271. Tidwell, J.H. and Allan, G.L. (2001). Fish as food: aquaculture's contribution - Ecological and economic impacts and contributions of fish farming and capture fisheries. EMBO Reports, 2, 958-963. Ukeles, R. (1965). Inhibition of unicellular algae by synthetic surface-active agents. Journal of Phycology, 1, 102-110. Ulberth, F. and Henninger, M. (1992). One-step extraction methylation method for determining the fatty-acid composition of processed foods. Journal of the American Oil Chemists Society, 69, 174-177. USEPA. What you need to know about mercury and shellfish. (March 2004). U.S. Environmental Protection Agency. Available from: http://www.cfsan.fda.gov/~dms/admehg3.html. (Accessed on Feburary 12, 2009) Ward, O.P. and Singh, A. (2005). Omega-3/6 fatty acids: Alternative sources of production. Process Biochemistry, 40, 3627-3652. Wen, Z.Y. and Chen, F. (2003). Heterotrophic production of eicosapentaenoic acid by microalgae. Biotechnology Advances, 21, 273-294. Yaguchi, T., Tanaka, S., Yokochi, T., Nakahara, T. and Higashihara, T. (1997). Production of high yields of docosahexaenoic acid by Schizochytrium sp. strain SR21. Journal of the American Oil Chemists Society, 74, 1431-1434. Yamane, A.N., Okada, M. and Sudo, R. (1984). The growth-inhibition of planktonic algae due to surfactants used in washing agents. Water Research, 18, 1101-1105. Yokochi, T., Honda, D., Higashihara, T. and Nakahara, T. (1998). Optimization of docosahexaenoic acid production by Schizochytrium limacinum SR21. Applied Microbiology and Biotechnology, 49, 72-76.

In: Microbial Conversions of Raw Glycerol Editor: George Aggelis

ISBN 978-1-60692-392-4 © 2009 Nova Science Publishers, Inc.

Chapter VI

Biotechnological Production of Bioplastics from Raw Glycerol

1

G. Mothes1∗ and O. Otto2 Saxon Institute for Applied Biotechnology (SIAB), Leipzig, Germany 2 BASF Schwarzheide GmbH, Schwarzheide, Germany

Abstract A wide variety of bacteria are able to synthesize polyhydroxyalkanoates (PHA) as an intracellular storage compound. The best-known and most common representative one is poly (3-hydroxybutyrate) (PHB). The properties of these isolated biodegradable polymers are very similar to polypropylene. As a rule their synthesis takes place, if the multiplication is limited by nutrients other than the carbon source. Depending on the metabolism of the involved bacteria, quite different carbon substrates can be used; the most common ones are sugars. In contrast to petrochemically produced plastics PHA can be produced from renewable carbon resources, which become more and more important in respect to the conservation of finite fossil resources like mineral oil and coal and their largely neutrality with regard to the emission of CO2. We could show that crude glycerol, a by-product of the biodiesel production from rape, is a promising feedstock for the production of PHB. By co-feeding of appropriate precursors we also could produce copolymers of poly (3-hydroxybutyric acid-co-3hydroxyvaleric acid) with co-monomer contents of 12-25 mol% resulting in improved polymer properties. For the isolation of the PHA we used an enzymatic method. The polymer composition was not changed during this procedure. Applying an additional mechanical disruption of the bacteria by French press, the use of the expensive enzyme lysozyme could be minimized or omitted. The molecular weight of the isolated polymer (700.000 – 900.000 g/mol) is sufficiently high, to be processed by common methods of polymer industry.

*Corresponding author: Permoserstr.15 D-04318 Leipzig, Germany Phone: +49 341 236 2073, Fax. +49 341 236 2073. E-mail:[email protected].

66

G. Mothes and O. Otto

Introduction Conventional plastics are produced from byproducts of the fossil fuel processing industry. These materials are not degradable in the nature. In contrast bioplastics are of biological origin. In the rule they are biodegradable, compostable or recyclable and can be produced from renewable carbon sources. Thus the production and the use of bioplastics are generally regarded as more sustainable when compared with petrochemical produced plastics. The raw materials of which are limitless when new bioconversion technologies like biorefinery emerge. While, in the past, plastics from alternate sources were significantly more expensive as well as less efficient than those derived using petroleum technology, higher oil prices have rendered these alternatives cheaper in comparison. Moreover, mounting environmental concerns and legislative incentives, particularly in the European Union (EU), are stimulating keen interest in the adoption of biodegradable plastics. This is in turn spurring research to improve bioplastics technology, and developments in plant breeding and processing are expected to further narrow the cost differential between bioplastics and synthetic ones. The European Bioplastics e.V. estimated that the potential of bioplastics could reach about 10% of the total plastics market (Busch, 2007). According to Helmut Kaiser Consultancy, (ACON AG, Zürich, Switzerland) Europe will become one of the most important markets for bioplastics, due to its limited amount of crude oil reserves. In recent years, bioplastics have been used in the food and packaging industry, medical, toys and textile industries. The biobased materials currently cover approximately 10 to 15% of the plastics market there and growth of the bioplastics market to 25 or even 30% by 2020 is predicted. Currently, bioplastics production companies own relatively small dedicated plants and are still in the early stages of development. In the future, they could become integrated in true 'biorefineries' that produce a wide range of products from biomass - from fuels and green platform chemicals, to fiber products and biopolymers. Technically speaking, bioplastics have overcome most of their initial problems and are now just as durable, workable and flexible as normal plastics. In this chapter we will exclusively look at Polyhydroxyalkanoates (PHA), intracellular polyesters, which are produced by bacteria.

Occurrence of PHAs The best-known and most spread PHA is poly (3-hydroxybutyrate) (PHB), first discovered and isolated from Bacillus megaterium by Lemoigne (1926). For many years, the physiological role of this polyester in the bacterial metabolism and the importance of the isolated polymer as bioplastics were largely overlooked (Lenz and Marchessault, 2005). Figure 1 shows a transmission electron micrograph of Methylobacterium rhodesianum with enclosed PHB granules, which appear as electron lucent bodies. At present, a wide variety of bacteria of different genera and metabolism are known, which can synthesize PHA as a carbon and energy resource (Haywood et al., 1989; Brandl et al., 1990). The maximum PHA content varies in the different bacteria strains and can reach up to 80% of the cell mass. As a

Biotechnological Production of Bioplastics from Raw Glycerol

67

rule, the synthesis takes place if a suitable carbon substrate is available and the cellular growth is limited by another nutrient, for instance ammonia or phosphate (Haywood et al, 1989; Anderson and Dawes, 1990). PHAs are stereo regular polyesters of [R]-3hydroxyalkanoic acids. The R (-) configuration is essential for their biodegradability and biocompatibility.

Figure 1. Transmission electron micrograph of Methylobacterium rhodesianum accumulating large amounts of PHB. PHB granules appear as electron lucent bodies.

The typical molecular weights range from about 200.000 – 1.000.000 Da, depending from the used bacteria and carbon substrates (Doi, 1990). Generally, PHAs are deposited as osmotic neutral, light refracting, insoluble cytoplasmatic inclusions (granula), which are surrounded by lipids, and proteins like enzymes (PHA polymerase, PHA depolymerase) and structure and regulatory proteins (phasins). The number of the granules seems to be strain specific (Jurasek and Marchessault, 2002, Pötter and Steinbüchel, 2005).

Diversity of PHAs Figure 2 demonstrates the general formula for PHAs. According to the composition of the monomers PHAs were divided into two classes: short chain lengths (scl) PHA and medium chain lengths (mcl) PHA. The former group with a monomer chain length of 4 –6 is most spread in bacteria. The most synthesized polymers contain both 3-hydroxybutyric acid (3HB) and 3-hydroxyvaleric acid (3HV) as monomer constituents. By feeding of carbon precursor substrates that are structurally related to various target monomer constituents over 100 different PHAs could be isolated (Steinbüchel and Valentin, 1995). In contrast, PHA of the mcl group typically contain a mixture of monomers of 6-14 carbon atoms, which are usually linked via 3-hydroxy ester linkages, but can occasionally also exhibit 2-, 4-, 5-, or 6hydroxy ester linkages. They are solely synthesized in Pseudomonades via sequences involved in the fatty acid metabolism. Their physical properties differ markedly from sclPHAs in being elastomers (de Koning et al., 1996; Kessler et al., 2001; Westhuis et al., 2002). In the rule, bacteria synthesize either scl-PHAs containing primarily 3HB units or mclPHAs containing mainly monomers from 3-hydroxyoctanoate and 3-hydroxydecanoate. This chapter will be restricted predominantly to the scl-PHAs.

68

G. Mothes and O. Otto

Figure 2. General formula for PHAs.

Applications and Properties of PHAs PHAs are of special commercial interest, due to their thermoplastic properties and biodegradability. They can be processed to films, fibers, and bottles, foamed or compounded by common methods of the polymer processing industry. Films made from PHA possess excellent gas and liquid barrier properties, which make them very interesting for special packaging applications like in the pharmaceutical field. They are also be promising for use in food packaging, since they act as good barrier to light incidence in the UV range. In contrast to materials made from poly(lactic acid), PHAs possess a considerably higher hydrolytic stability. Films of PHA can be used to make laminates with other polymers or coatings can be produced with PHA as latex. Other applications are the replacement of petrochemical polymers in toner and developer compositions. A very interesting field is the use as exchanger for affinity chromatography to isolate proteins or antibodies from complex solutions or the development of immunosticks. A wide variety of applications provide nanocomposites made from PHAs. Moreover, the monomers of PHA can be obtained by chemical or enzymatic degradation and used for the synthesis of other chiral compounds (Holmes, 1985; Cox, 1994; Schrecklies, 2001; Van der Walle et al., 2001; Maidi et al., 2003; Bucci et al., 2007; Nissen et al., 2007; Tokiwa and Ugwu, 2007). A cradle-to grave environmental life cycle assessment of a few PHB based composites has been performed and was compared to commodity petrochemical polymers (Pietrini et al., 2007). The results obtained show that it is possible to reach lower environmental impacts than by use of conventional petrochemical polymers. Furthermore, PHAs are very well suitable for biomedical applications. Materials from PHB can be steam sterilized without loosing properties and are resistant to alcohol. If sterilized they may be left in the body and are slowly absorbed. The degradation rate is much more slow than that of poly(lactic acid) or poly(glycolic acid). The degradation products are 3-hydroxybutyric acids, which are natural constituents of human blood and are less acidic and less inflammatory than many currently used synthetic absorbable polymers. Typical medical applications are cardiovascular products like pericardial patches, cardiovascular stents, vascular grafts, sutures and dressings, controlled drug delivery systems, barrier membranes for guided tissue and bone regeneration, scaffolds for tissue engineering (for instance heart valves). Leaching techniques can produce porous structures to guide and support cell adhesion growth and differentiation, as well as permit the passage of nutrients and waste products. Than the engineered cells and tissues can be implanted together with the supporting scaffold.

Biotechnological Production of Bioplastics from Raw Glycerol

69

Partly on account of the piezoelectric properties of PHB, interest has arisen in the use of this polymer for nerve repair, bone implants and surgical fixations like bone pins and plates (Williams and Martin, 2005). The mechanical and thermal properties (melting- and glass-transition temperature, Young’s modulus, tensile strength) of PHB, the most common synthesized representative of PHAs, are close those of isotactic polypropylene (Tab1). PHB and polypropylene display similar degrees of crystallinity; PHB possesses worse solvent resistance but much better resistance to UV weathering. Nevertheless, a significant drawback of PHB is its low elasticity; the extension to break (5%) is markedly lower than that of polypropylene (400%). Above all, the thermal instability of PHB at temperature above 190 °C results in a very small processing window. In addition, melt-processed PHB undergoes a slow aging process under ambient condition and becomes brittle. These properties can be altered and improved by different methods like a simple annealing treatment (De Koning and Lemstra, 1993), blending with other polymers (Sharma and Ray, 1995; Scandola, 1995; Koyama and Doi, 1997), or the addition of plasticizers. The physical properties of PHB can be distinctively improved by incorporating other hydroxyalkanoate units to form PHA copolymers. As above-mentioned, the PHA composition produced by a bacterium depends on the substrate specifity of the enzymes in the PHA biosynthesis pathway. When copolymers with 3HB and 3HV or 4HB monomer units are produced, the properties of these materials also alter as a consequence of the decreased crystallinity and melting temperature, which resulted in a decreased stiffness and increased toughness, the polymer gets more flexible. Thereby the effect of the incorporation of 4HB is much more pronounced (Table 1). Table 1. Thermal and physical properties of PHA1 Sample

Poly(3HB) Poly(3HB-co-9 mol% 3HV) Poly(3HB-co-20mol% 3HV) Poly(3HB-co- 3 mol% 4HB) Poly(3HB-co-10 mol% 4HB) Polypropylene 1

Glass-transition Young’s Melting modulus temperature temperature (°C) 179 162 145 166 159 176

(°C) 4 6 -1

(GPa) 3.5 1.9 1.2 2.9

-10 1.7

Tensile strength

Elongation to break

(MPa) 43 37 31 28 24 34

(%) 5 54 45 242 400

Doi 1990; Saito and Doi, 1994; Yamane et al., 1996.

In contrast to the multitude of potential PHB producers, only a few bacteria are used for the synthesis of PHB and PHB/HV in a larger scale, e.g. Ralstonia eutropha, Alcaligenes latus, Methylobacterium organophilum, Azotobacter vinelandii.

70

G. Mothes and O. Otto

Degradation of PHAs The isolated PHA can be degraded by a number of aerobic and anaerobic bacteria and fungi, which excrete extracellular PHA depolymerases to hydrolyze the water-insoluble PHAs into water soluble monomers and oligomers to use them as carbon source (Jendrossek et al., 1996; 2002). The end products of degradation in aerobic environment are carbon dioxide and water whereas methane is produced under anaerobic conditions. The rate of biodegradation depends on many factors like temperature, moisture level, pH, the composition, thickness and surface area of the PHA and the microbial population in a given environment. As a rule, the degradation rate is best in anaerobic sewage, followed by soil and seawater, respectively (Doi et al., 1992, 1996; Mergaert et al., 1993). Extracellular PHA, without the surrounding granule surface layer of proteins and phospholipids, tends to develop more and more crystalline phase due to the adopted ordered helical conformation of the polyester chains (Cornibert and Marchessault, 1975; de Koning and Lemstra, 1992). In contrast to extracellular PHA depolymerases, which also accept crystalline PHAs, intracellular depolymerases only hydrolyze PHAs in amorphous state (highly mobile polymer chains in disordered conformation) as found in vivo in intact granules (Saito and Kobayashi, 2002). Thus they are involved in the active mobilization of the carbon and energy storage compound for the cell metabolism of the synthesizing bacterium itself and do not contribute to the degradation of plastic materials made from PHA. In addition to the degradation of such materials due to microbial enzyme activities a simple hydrolysis can take place in aqueous environment even though it proceeds very slowly. These properties are important for biomedical applications of PHA. Typically, PHB is completely absorbed in vivo in 24-30 months (Malm et al., 1992; Hazari et al., 1999, Hasirci, 2000). The monomeric degradation products are part of the human metabolism and the cause of the biocompatibility of these polymers.

Pathway and Regulation of scl PHA Synthesis The synthesis of scl PHA starts from acetyl-CoA, a central metabolic intermediate and involves the activity of three enzymes (figure 3). The β-ketothiolase condenses two molecules of acetyl-CoA to acetoacetyl-CoA, which is reduced using NADPH to [R]hydroxybutyryl-CoA by the acetoacetyl-CoA reductase and finally polymerized to PHB by the PHA synthase. The first two enzymes differ from the enzymes involved in the fatty acid pathway in their substrate- and coenzyme specifity and stereo-selectivity and are found in the cytoplasm while the active PHB synthase is located at the surface of the granule. This pathway is found in most PHB synthesizing bacteria (Tomita et al., 1983; Senior and Dawes, 1973; Steinbüchel and Schlegel, 1991; Manchak and Page, 1994; Mothes and Babel, 1994; 1997) excepting a few strains, where a NADH-dependent acetoacetyl-CoA reductase produces [S]-hydroxybutyryl-CoA, which is further metabolized to the [R]-isomer by the action of [R]- and [S]-specific enoyl-CoA hydratases (Moskowitz and Merrick, 1969; Mothes and Babel; 1995). The PHB synthesis is mainly regulated at the metabolic level: the availability of NADPH (as cofactor for the acetoacetyl-CoA reductase), a high ratio of

Biotechnological Production of Bioplastics from Raw Glycerol

71

NADH/NAD (to inhibit the competing citrate synthase) and a high ratio of acetyl-CoA/CoA or rather a low level of CoA are preconditions for PHB synthesis (Senior and Dawes, 1973; Mansfield et al., 1995; Mothes et al., 1997; 1998).

Figure 3. Pathway of PHB synthesis.

The genes coding for the enzymes of biosynthesis of PHA have been cloned from Ralstonia eutropha and were shown to be active in Escherichia coli. Currently up to 60 PHA synthase structural genes as well as many genes encoding proteins and enzymes related to PHB synthesis have been cloned and characterized from different bacteria. Thus, applying the methodology of metabolic engineering new strains could be constructed for effective production of various PHAs (Lee et al, 1994; Rehm and Steinbüchel, 2002; Rehm, 2003).

Fermentative Production of PHAs from Crude Glycerol Depending on the metabolism of the involved bacteria quite different carbon substrates like sugars, alcohols, organic acids, methane or hydrogen gas can be used, the most common ones are sugars. Actually the use of renewable carbon resources, based on agriculture become more and more important in respect to the conservation of finite fossil resources like mineral oil and coal and their largely neutrality with regard to the emission of CO2. Furthermore, from an economical point of view, the costs of the carbon substrate contribute significantly to

72

G. Mothes and O. Otto

the overall production cost of PHA (Choi and Lee, 1997). These two aspects are a good reason for using crude glycerol as carbon source. We could show that crude glycerol, a by-product of the biodiesel production from rape, is a promising feedstock for the production of PHB (Mothes et al., 2007). Crude glycerol samples of various German manufacturers differ mainly in their contamination with up to 6% (w/w) of salts (NaCl or K2SO4), 0-0.5% methanol, residual fatty acids and the pH (4.1-6.8), due to the applied process. All these feedstocks contain glycerol as a main component (80– 90%, w/w) with very low amounts of fatty acid soaps and residual fatty acid methyl esters. That’s why we focused on scl-PHA accumulating bacteria in contrast to Ashby et al. (2004), who applied Pseudomonades by reason of using a crude glycerol feedstock containing 40 % of glycerol and 34 % of fatty acid soaps and fatty acid methyl esters. The employed bacteria by us were Paracoccus denitrificans, Cupriavidus necator H16 and Cupriavidus necator JMP 134, which accumulate PHB from pure glycerol to a content of 70% of cell dry mass (3 g/l) in batch experiments applying nitrogen limitation. Fed batch fermentation was performed to reach a higher cell concentration of 50-g/l dry masses without additional oxygen supply. During growth phase and production phase the pH was maintained by adding NH4OH and KOH, respectively. During the growth phase the NH4+ concentration in the medium was maintained at 0.5 g/l; NH4OH was replaced with KOH when the cell concentration reached 23 g/l and the PHB synthesis started upon nitrogen depletion, when the cell concentration reached 25 g/l. The glycerol was fed at concentration of 10 g/l according the experimentally determined consumption rates. When using crude glycerol containing 5.5 % NaCl, the PHB content of the bacteria was reduced to 48 % of the dry mass at high cell density fermentation with a bacterial dry mass of 50g/l, probably due to the accumulation of the salts in the culture medium (figure 4A). Furthermore the PHB yield coefficient was reduced (figure 4B), obviously due to the use of a part of the glycerol for osmoregulation. The effect of glycerol contaminated with K2SO4 on the PHA content and the yield was less pronounced (Mothes et al., 2007). In these experiments the biomass yield in the growth phase was 0.5 g/g glycerol, irrespective of the mode of crude glycerol. Following we studied the ability of these bacteria to synthesize different copolymers from glycerol and additional relevant carbon substrates, which were fed only during production phase after limitation of the growth by ammonia (Table 2). From propionic acid, valeric acid, propanol, n-pentanol, or levulinic acid co-monomers of 3-hydroxyyvaleric-acid were incorporated into the polymer. All these substrates were consumed in common with glycerol. When applying propionic acid or propanol a significant lower co-monomer content was found, due to the partial decarboxylation of these substrates during its pathway. As a result they were metabolized as well as to acetyl-CoA and propionyl-CoA yielding 3hydroxybutyric acid monomers and 3-hydroxyvaleric acid monomers, respectively. Valeric acid, n-pentanol and levulinic acid yielded polymers with much higher 3-HV content (Doi et al., 1986; 1987). In contrast to Valentin et al. (1992), who demonstrated the synthesis of a terpolyester of 3HB, 3HV and 5-6 mol% 4HV in C. necator (formerly Alcaligenes eutrophus), when cultivated solely with 4-hydroxyvaleric acid or γ-valerolactone, we could only detect 3HB and 3HV constituents in the polymer synthesized by co-feeding of glycerol and levulinic acid by 13C NMR (not shown). The incorporation of 4HV units into the polymer should have

Biotechnological Production of Bioplastics from Raw Glycerol

73

much more powerful effect on the mechanical properties and the thermal stability than 3HV units (Gorenflo et al., 2001).

Figure 4. A: PHB synthesis from crude glycerol of different manufacturers at high cell density (25 g/l residual biomass) with C. necator JMP 134 at 35 °C and pH 7,0. ● glycerol pA (1), ∆ crude glycerol with 5.5 % NaCl (2), ■ crude glycerol with 0.8 % K2SO4 and 0.03 % Methanol (3), □ crude glycerol with 1.6 % K2SO4 and 0.5 % Methanol (4), ▲ mixture (1:2) of crude glycerol (2) and (3), [reproduced from Mothes et al., 2007]. B: Yield of PHB synthesis from crude glycerol at high cell density (25 g/l residual biomass) with C. necator JMP 134. ● glycerol pA (1), ∆ crude glycerol (2), ■ crude glycerol (3), [reproduced from Mothes et al., 2007].

No copolymers from 3HB and 4HB could be detected in C. necator or P. denitrificans, when feeding glycerol and typical precursors like 4-hydroxybutyric acid, 1,4-butanediol or γbutyrolactone in nitrogen-free medium (Table 2). The second substrates were not cometabolized with glycerol. In contrast, when co-feeding acetic acid and γ-butyrolactone (in the ratio 2:1 (w/w)) copolymers with 3.6 mol% 4HB were found. Kunioka et al. (1989) also demonstrated the synthesis of copolymers of 3HB and 25-36 mol% 4HB from 4hydroxybutyric acid, 1,4-butanediol or γ-butyrolactone in C. necator grown in nutrient-rich media, when transferred into nitrogen-free media in batch experiments. This indicated that

G. Mothes and O. Otto

74

the precursors are partially degraded to acetyl-CoA and/or 3-hydroxybutyryl-CoA. The reached polymer concentration were low (8-19%) but could be increased to 50-63%, when co-feeding butyric acid with 2-37mol% 4HB depending on the used kind and amount of precursor substrate. Table 2. Copolymer synthesis from glycerol and a second carbon substrate during batch experiments in flasks using C. necator JMP 134. Glycerol grown cells were transferred into N-free medium to a concentration equivalent to 0.2-0.3 g/l of dry mass and cultivated with 1 g/l glycerol and 0-0.5 g/l of the second substrate at 30°C, pH 7.0 and an agitation rate of 120 rpm 2. Carbon substrate Propionic acid Valeric acid Propanol n-Pentanol Levulinic acid γ- Valerolactone γ- Butyrolactone

PHA (%) 24 h 48h 50 70 51 70 52 68 50 62 55 78 55 71 61 76

Type of comonomer 3-HV 3-HV 3-HV 3-HV 3-HV -

Comonomer (mol%) 24h 7 38 8 37 27 0 0

Figure 5 shows a typical profile of the PHA and HV content and the residual cell concentration during the fed-batch fermentation of C. necator using crude glycerol (contaminated with 0.8% K2SO4) and levulinic acid (p.a) as carbon substrates.

Figure 5. Time profile of the content of PHA (●), HV (▲) and residual cell concentration (□) during the fed-batch fermentation of C. necator JMP 134 at 35 °C and pH 7.0 using crude glycerol as carbon substrate. NH4OH was replaced with KOH when cell concentration reached 13 g/l. Nitrogen depletion occurred at 26.5 h. After that, a mixture of glycerol and 10% levulinic acid (w/w) was fed.

Biotechnological Production of Bioplastics from Raw Glycerol

75

During the growth phase solely glycerol was employed. When the cell multiplication was limited by ammonium, levulinic acid was fed together with glycerol as a substrate mixture with a resulting total concentration of 10 g/l. Preliminary experiments showed that both substrates were consumed simultaneous as well in the growth phase as in production phase. At the beginning of PHA synthesis the uptake rate of levulinic acid was 1.14-fold higher than the uptake rate of glycerol. The feeding profile was calculated according the experimentally determined glycerol consumption rate. Thus a target copolymer content could be reached despite of the higher consumption rate of levulinic acid in comparison to glycerol. Within 48 h of nitrogen limitation the PHA content reached 60% of the cell dry mass. The HV content reached its maximum value after 5 hours, which remained largely stable. The residual cell mass did not increase during production phase. The total cell mass was 34 g/l. The yield of PHB on glycerol in this fermentation was 0.3 g/g of glycerol; the yield of HV was 0.4 g/g of levulinic acid. This process could be scaled up in laboratory scale (1.5- 4 L; Biostate MD fermenter (Braun Biotech)) to a cell concentration of 65 g/l (dry mass) without additional oxygen supply. To avoid additional salt accumulation due to the pH correction fluids, ammonia was used as correction agent during the growth phase, which was changed against 4N KOH to trigger conditions for PHA synthesis. Further fermentations with carbon mixtures with a different portion of levulinic acid, ranging from 10% (w/w) to 20% (w/w), were carried out. In the rule, the HV content in the polymer was in good correlation to the proportion of this second carbon substrate in the fed (Figure 6A). Levulinic acid is a suitable substrate for the production of 3HB/3HV copolymers because it can be produced cost effectively from cellulose containing forest- and agricultural waste residues using the process of biorefinery (Bozell et al., 2000; Kamm 2007). Similar experiments were done using glycerol plus pentanol, which also resulted in a correlation of the reached copolymer content to the portion of added pentanol (Figure 6B). The yield of HV on pentanol was 0.5 g/g of pentanol. Because of the bad miscibility of these two substrates, pentanol had to be fed by a separate pump in accordance to the glycerol consumption. In both cases deviations from the target composition of the copolymer in respect to the carbon substrate proportion in the fed could appear. Possible reasons are the use of parts of glycerol for the synthesis of by-products or for osmoregulation as a result of accumulating salts in the course of production phase. The accumulation of antifoam solution in the medium (when above 5 g/l) also reduces the yield of the both carbon substrates to a different degree. But for the production of copolymers from 3HB and 3HV deviations of 5 mol% of the copolymer content don‘t have a pronounced effect on the polymer properties. In contrast, when focusing on copolymers of for instance 3HB and 4HB or mcl PHAs such deviations have a significant higher effect and in this case for the production of polymers with target composition and resulting target properties the chemostatic cultivation has to be applied (Mothes et al., 2005; Hartmann et al., 2006). The molecular weight of the isolated PHA produced with P. denitrificans or C. necator from crude glycerol varied between 620,000 and 750,000 g/mol. There were no correlations of the molecular weight to the purity of the used glycerol or the portion of HV in the polymer. However, these values were somewhat lower compared to the polymer produced

76

G. Mothes and O. Otto

from acetic acid or fructose (970,000 and 1,160,000, respectively), but sufficiently high for processing by common techniques used in the polymer industry (Mothes at al., 2007).

Figure 6. Fed batch fermentation of C. necator JMP 134 using glycerol plus levulinic acid (A) or pentanol (B) as carbon substrate during production phase. Correlation between the portion of the second substrate in the feed and the molar copolymer content in the polymer.

These results are in accordance with Taidi et al. (1994) and Madden et al. (1999), who demonstrated, that glycerol acts as a chain transfer agent in the chain termination step of the polymerization and that this effect is more pronounced at a high concentration of glycerol in the medium (more than 10 g/L). Furthermore, the molecular mass of PHA is dependent on the producing bacteria (Taidi et al., 1994). Thus, Koller et al. (2005) isolated PHA with a significant lower molecular weight (about 250,000 g/mol) from a glycerol grown halophilic archaeon bacterium (Haloferax mediterranei). Such low molecular weight PHA should not be well applicable for processing by extruder but may have some special fields of application,

Biotechnological Production of Bioplastics from Raw Glycerol

77

for example, as softeners (Koller et al., 2005). Furthermore, the kind of the polymer extraction has a significant influence on its resulting molecular weight. Methods applying a high basic pH, hydroxyapatite or hydrogen peroxide are known to result in polymers with significant lower molecular weight (Ramsey et al., 1990). The melting point of the isolated copolymers was reduced to 153°C-132°C according the increasing HV content (14 mol%-24 mol%) whereas the glass-transition temperature varied between 8°C and –3°C. The Young’s modulus was between 0.8 and 1.45 GPa. The tensile strength (15-21 MPa) and the elongation to break (6.6 – 11%) were somewhat lower than measured by Doi (1990) and Yamane et al. (1996), possibly due to small-incorporated impurities of the polymer resulting from our applied enzymatical recovery method. Nevertheless the copolymers showed improved physical properties in respect to PHB and getting more flexible with better processing properties.

Recovery of the PHAs For the recovery of the PHAs from the bacterial cells several methods have been developed which principally divide into solvent based or water based ones, which should be brief presented. The solvent-based extraction is limited by the poor solubility of the polymer. PHAs are insoluble in the most convenient solutes; suitable solvents are halogenated hydrocarbons like dichloroethane, trichloroethane or methylene chloride. The concentrated biomass is first freeze-dried or spray-dried, the lipids are removed by extraction with methanol or ethanol or acetone in the heat followed by dissolving of the PHA in a selected solvent. The insoluble cellular components are separated by filtration and the PHA is further purified by precipitation in cold ethanol or methanol. With this method PHB with a high purity and yield can be obtained. A big drawback is the very bad solubility of the PHAs, at polymer concentrations more than 5% by weight the resulting solutions tend to become very viscous which results in problems of separating the non-PHA cell materials. Newer inventions describes the extraction with non halogenated solvents like lower chain ketones, acyclic and cyclic esters, cyclic or acyclic alcohols with at least four carbon atoms and cyclic or acyclic alcohols with fife or more carbon atoms (Kurdikar et al., 1998; Narasimhan et al., 2005; Van Walsem et al., 2006; Phb Ind., 2006). Cell lysis by using hypochlorite represents a cheap water-based recovery method (Ramsey et al., 1990). However, hypochlorite reduces the molecular weight of PHA, which could be unfavorable for some applications. This effect of molecular weight reduction could be somewhat reduced by using dispersions of sodium hypochlorite and chloroform for extraction which was due to a shielding effect of the chloroform (Hahn, 1994). A quite other approach uses enzyme cocktails like different proteases, phospholipases and lysozyme for digesting and separating the cell constituents to yield the PHA granules. In the rule, the bacterial cell wall of gram-negative bacteria, characterized by an inner peptidoglucane layer, has first to be damaged by detergents and complexing agents to be accessible for the enzymes. In addition to the complexing of divalent cations like Mg2+ and Ca2+ a substantial part of the lipopolysaccharides from the outer membrane were extracted by

78

G. Mothes and O. Otto

this treatment. Some workers inserted upstream a heat treatment, to denature the DNA, which avoids viscosity problems. Furthermore the denaturation of proteins, like PHA depolymerase diminished the reduction of molecular weight during the course of the recovery process and make the proteins more susceptible to the subsequent protease digestion (deKoning and Witholt, 1997; Schumann and Müller, 2001; Kapritchkoff et al., 2006; Yasotha et al., 2006). We applied the enzymatic approach for isolating the PHA from C. necator and P. denitrificans, which are both gram-negative bacteria. The used enzymes were in a technical grade (ASA Spezialenzyme GmbH, Wolfenbüttel, Germany) and included an alkaline protease (from Bacillus) and lysozyme (from chicken egg). The bacterial cells were pretreated by heat (80-120°C for 15 minutes), washed with water and concentrated by centrifugation or filtration equivalent to 350 g/l bacterial dry mass. After that a complexing agent was added to make the cell wall accessible to lysozyme, which degrades the inner peptidoglucane layer of the cell wall. In laboratory scale we used EDTA for this. By reason that the use of EDTA in industrial scale meets a problem in respect to the discharging of the wastewater into a sewage plant we tested alternative complexing agents. Good results were achieved when using Trilon®M, a tri-sodium salt of methylglycine diacetic acid, produced by BASF (Ludwigshafen, Germany), which is characterized by good biodegradability that the aqueous supernatants arising in the course of PHA recovery can be supplied directly to waste water treatment plant without additional pretreatment. Trilon®M was added at a concentration of 4 mg/g residual cell mass (without PHA). After washing with water lysozyme FL (8.000.000 U/ml; 125 µl/l) was added for 1 hour at 35°C and pH 7.5; followed by addition of the protease A-01 (1.100 U/ml; 0,05 ml/l; with a bacterial cell concentration equivalent to 350 g/l dry mass) at pH 8.5 for additional 1 h at 40°C. The resulting purity of the polymer was between 92-96%. The applied reaction temperature, pH, residence time and enzyme concentration can be further adapted to the intended process parameter. The overall recovery yield was about 98%. In a quite other approach we introduced a mechanical step to disintegrate the cell wall by high pressure using a French press cell press applying a pressure between 10.000 to 18.000 PSI (690-1.241 bar). The best results were at two passages at 14.000-18.000 PSI (965-1.241 bar). The second passage was introduced to compensate for insufficient disintegration. In technical scale a continuous process can realize this pressure step. After washing with water the enzymes could be applied without pretreatment with complexing agents. The concentration of lysozyme could be further reduced. The resulting purity of the PHA was 9193% (figure 7). When omitting lysozyme a somewhat lower purity (88%) was achieved. However, this could be possibly due to inhomogeneities of the used bacterial cell suspension as a consequence of previous washing and concentration steps by discontinuous centrifugation and resuspending yielding blockages of the nozzle of French press accompanied by pressure drop and insufficient cell disintegration. In the technical scale for the washing steps continuous centrifuges or filtration techniques can be used, which ensure a more consistent suspension of the cells during the whole recovery process. We estimate that when applying this technique the lysozyme could be omitted without a drop of the purity of the PHA.

Biotechnological Production of Bioplastics from Raw Glycerol

79

Figure 7. Protocol of PHB recovery from C. necator JMP 134. A: step 2 with complexing agent and lysozyme, B: Step 2 without complexing agent, C: step 2 omitted.

The recovered PHA could be further purified either by washing with methanol, or addition of SDS (0,15 g/g residual dry mass (non PHA)), or sodium hypochlorite (1%, w/w; 10 min at pH 8.5). All this steps were similarly effective and yielded a polymer of 98-100% purity as determined by gas chromatography. It could be shown, that when applying these enzymatic procedure for the recovery of copolymers from 3HB and 3HV no significant change of the molar monomer composition took place (figure 8).

Figure 8. HV content of the PHA in the course of the different purification steps.

80

G. Mothes and O. Otto

Conclusion Glycerol is a promising feedstock for the biotechnological synthesis of PHB and copolymers. In the future materials made from polyhydroxyalkanoates could become a more and more environmental alternative to usual plastics since the continuously increasing price of oil improves the competitiveness of biopolymers and will push the development and the use of renewable resources.

Acknowledgments This work was supported by Agency of Renewable Resources (Germany), Proj. No. 220005-05.

References Anderson, AJ; Dawes, EA; 1990. Occurrence, metabolism, metabolic role, and industrial uses of bacterial polyhydroxyalkanoates. Microbiol. Rev. 54: 450-472. Ashby, RD; Solaiman, DKY; Foglia, TA; 2004. Bacterial poly(hydroxyalkanoate) polymer production from biodiesel coproduct stream. J. Polym. Environ. 12: 105-112. Bozell, JJ; Moens, L; Elliott, DC; Wang, Y; Neuenschwander, GG; Fritzpatrick, SW; Bilski, RJ; Jarnefeld, JL; 2000. Production of levulinic acid and use as a platform chemical for derived products. Resour. Conserv. Recycl. 28: 227-239. Brandl, H; Gross, RA; Lenz, RW; Fuller, RC; 1990. Plastics from Bacteria and for bacteria: poly(β-hydroxyalkanoates) as natural, biocompatible, and biodegradable polyesters, in: Advances in Biochem. Engin./Biotechnol. (Fischer, A; ed) Springer-Verlag, Berlin, Heidelberg , vol. 41: 77-93. Bucci, DZ; Tavares, LBB; Sell, I; 2007. Biodegradation and physical evaluation of PHB packaging. Polymer. Testing 26: 908-915. Busch, R; 2007. Polymere der Zukunft. GIT, 3: 220-221. Choi, J; Lee, SY; 1997. Process analysis and economic evaluation for poly(3hydroxybutyrate) production by fermentation. Bioprocess Eng. 17: 335-342. Cornibert, J; Marchessault, RH; 1975. Conformational isomorphism. A general 2(1) helical conformation for poly(β-hydroxyalkanoates). Macromol. 8: 296-305. Cox, MK; 1994. Properties and applications of polyhydroxyalkanoates. Studies in Polymer Science, 12(Biodegradable Plastics and Polymers), 120-35. de Koning, GJM; Lemstra, PJ; 1992. The amorphous state of bacterial poly[(Rhydroxyalkanoate] in vivo. Polymer 33: 3304-3306. de Koning, GJM; Lemstra, PJ; 1993. Crystallization phenomena in bacterial poly([R]-3hydroxybutyrate): 2. Embrittlement and rejuvenation. Polymer 34: 4089-4094. de Koning, GJM; Witholt, B; 1997. A process for the recovery of poly(hydroxyalkanoates) from Pseudomonads. Part 1: Solubilization. Bioprocess. Ing. 17: 7-13.

Biotechnological Production of Bioplastics from Raw Glycerol

81

Doi, Y; Kunioka, M; Nakamura, Y; Soga, K; 1986. Biosynthesis of polyesters by Alcaligenes eutrophus: incorporation of 13C-labelled acetate and propionate. J. Chem. Soc. Chem. Commun. 1059: 1696-1697. Doi, Y; Tamaki, A; Kunioka, M; Soga, K; 1987. Biosynthesis of an unusual copolyester (10 mol% 3-hydroxybutyrate and 90 mol% 3-hydroxyvalerate units) in Alcaligenes eutrophus from pentanoic acid. J. Chem. Soc. Chem. Commun. 762: 1635-1636. Doi, Y; 1990. Microbial polyesters. VCH Pub., Inc, NY., pp 11-31. Doi, Y; Kanesawa, Y; Tanahashi, N; Kumagai, Y; 1992. Biodegradation of microbial polyesters in the marine environment. Polym. Degrad. Stab. 36: 173-177. Doi, Y; Kasuya, K; Abe, H; Koyama, N; Ishiwatari, S; Tagaki, K; Yoshida, Y; 1996. Evaluation of biodegradabilities of biosynthetic and chemosynthetic polyesters in river water. Polym. Degrad. Stab 51: 281-286. de Koning, GJM; Kellerhals, M; van Meurs, C; Witholt, B; 1996. Poly(hydroxyalkanoates) from fluorescent Pseudomonades in retrospect and prospect. J. Environ. Polym. Degrad. 4: 243-252. Gorenflow, V; Schmack, G; Vogel, R; Steinbüchel, A; 2001. Development of a process for the biotechnological large-scale production of 4-hydroxyvalerate-containing polyesters and characterization of their physical and mechanical properties. Biomacromol. 2: 45-57. Hahn, SK; Chang, YK; Kim, BS; Chang, HN; 1994. Optimization of microbial poly(3hydroxybutyrate) recovery using dispersions of sodium hypochlorite solution and chloroform. Biotechnology and Bioengineering 44: 256-61. Hartmann, R; Hany, R; Pletscher, E; Ritter, A; Witholt, B; Zinn, M; 2006. Tailor-made olefinic medium-chain-length poly[(R)-3-hydroxyalkanoates] by Pseudomonas putida GPo1: batch versus chemostat production. Biotechnol. Bioeng. 93: 737-746. Hasirci, V; 2000. Biodegradable biomedical polymers. In: Biomaterials and bioengineering Handbook (Wise, DL, ed.), Marcel Dekker, New York, pp 141-155. Haywood, GW; Anderson, AJ; Dawes, EA; 1989. A survey of the accumulation of novel polyhydroxyalkanoates by bacteria. Biotechnol. Lett. 11: 471-476. Hazari, A; Johannson-Ruden, G; Jumeno-Bostrom, K; Ljundberg, C; Terenghi, G; Green, C; Wiberg, M; 1999. A new resorbable wrap-around implant as an alternative nerve repair technique. J. Hand Surg. 24B: 291-295. Helmut Kaiser Consultancy, ACON AG, Zürich, Switzerland, 2007. Bioplastics Market Worldwide 2007-2025. Applications, methods, competition, materials, technologies, development, recycling, renewable energy, production, consumption. hkc22.com market study. Holmes, PA; 1985. Applications of PHB - a microbially produced biodegradable thermoplastic. Phys. Technol. 16: 32-6. Jendrossek, D; Schirmer, A; Schlegel, HG; 1996. Biodegradation of polyhydroxyalkanoic acids. Appl. Microbiol. Biotechnol. 46: 451-463. Jendrossek, D; 2002. Extracellular polyhydroxyalkanoate depolymerases: the key enzymes of PHA degradation. in: Biopolymers 3b (Polyesters II), (Doi, Y; Steinbüchel, A, eds), WILEY-VCH, Weinheim, pp 41-83. Jurasek, L; Marchessault, RH; 2002. The role of phasins in the morphogenesis of poly(3hydroxybutyrate) granules. Biomacromol. 3: 256-261.

82

G. Mothes and O. Otto

Kamm, B; Kamm, M; 2007. International biorefinery systems. Pure Appl. Chem. 79: 19831997. Kapritchkoff, FM; Viotti, AP; Alli, RCP; Zuccolo, M; Pradella, JGC; Maiorano, AE; Miranda, EA; Bonomi, A; 2006. Enzymatic recovery and purification of polyhydroxybutyrate produced by Ralstonia eutropha. J. Biotechnol. 122: 453-462. Kessler, B; Weusthuis, R; Witholt, B; Eggink, G; 2001. Production of microbial polyesters: fermentation and downstream processes. Adv. Biochem. Engineer./Biotechnol, SpringerVerlag, Berlin, Heidelberg, 71(Biopolyesters): pp 159-182. Koller, M; Bona, R; Braunegg, G; Hermann, C; Horvat, P; Kroutil, M; Martinz, J; Neto, J; Pereira, L; Varila, P; 2005. Production of polyhydroxyalkanoates from agricultural waste and surplus materials. Biomacromol. 6: 561-565. Koyama, N; Doi, Y; 1997. Miscibility of binary blends of poly([R]-3-hydroxybutyric acid) and poly ([S]-lactic acid). Polymer 38: 1589-1593. Kunioka M; Kawaguchi, Y; Doi, Y; 1989. Production of biodegradable copolyesters of 3hydroxybutyrate and 4- hydroxybutyrate by Alcaligenes eutrophus. Appl. Microbiol. Biotechnol. 30: 569-573. Kurdikar, DL; Strauser, FE; Solodar, AJ; Paster, MD; Asrar, J; 1998. Methods for polyhydroxyalkanoate extraction and recovery using non-halogenated solvents. PCT Int. Appl. WO 9846782 A1 19981022 . Lee, SY; Lee, KM; Chang, HN; Steinbüchel, A., 1994. Comparison of Escherichia coli strains for synthesis and accumulation of poly-(3-hydroxybutyric acid), and morphological changes. Biotehnol. Bioeng. 44: 1337-1347. Lemoigne, M; 1926. Produits de deshydration et depolymerisation de l’acide β-oxybutyrique. Bull. Soc. Chim. Biol. (Paris), 8: 770-782. Lenz, RW; Marchessault, RH; 2005. Bacterial polyesters: Biosynthesis, biodegradable plastics and biotechnology. Biomacromol. 6: 1-8. Madden, LA; Anderson, AJ; Shah, DT; Asrar, J; 1999. Chain termination in polyhydroxyalkanoate synthesis: involvement of exogenous hydroxy-compounds as chain transfer agents, Int. J. Biol. Macromol. 25: 43-53. Maiti, P; Batt, CA; Giannelis, EP; 2003. Renewable plastics: Synthesis and properties of PHB nanocomposites. PMSE Preprints 88: 58-59. Malm, T; Bowald, S; Bylock, A; Busch, C; 1992. Prevention of postoperative pericardial adhesions by closure of the pericardium with absorbable polymer patches. An experimental study. J. Thorac. Cardiovasc. Surg. 104: 600-607. Manchak, J; Page, WJ; 1994. Control of polyhydroxyalkanoate synthesis in Azotobacter vinelandii strain UWD. Microbiol. 140: 953-963. Mansfield, DA; Anderson, AJ; Naylor, LA; 1995. Regulation of PHB metabolism in Alcaligenes eutrophus. Can. J. Microbiol. 41: 44-49. Megaert, J; Webb, A; Anderson, C; Wouters, A; Swings, J; 1993. Microbial degradation of poly(3-hydroxybutyrate) and poly(3-hydroxybutyrate-co-3-hydroxyvalerate) in soils. Appl. Environ. Microbiol. 59: 3233-3238. Moskowitz, GJ; Merrick, JM; 1969. Metabolism of poly-β-hydroxybutyrate. II Enzymatic synthesis of D(-)-β-hydroxybutyryl-CoA by an enoyl hydrase from Rhodospirillum rubrum. Biochemistry 8: 2748-2755.

Biotechnological Production of Bioplastics from Raw Glycerol

83

Mothes, G; Babel, W; 1994. Methylobacterium rhodesianum MB126 possesses two acetoacetyl-CoA reductases. Arch. Microbiol. 161: 277-280. Mothes, G; Babel, W; 1995. Methylobacterium rhodesianum MB 126 possesses two stereospecific crotonyl-CoA hydratases. J. Can. Microbiol. 41: 68-72. Mothes, G; Skinfill-Rivera, I; Babel, W; 1997. Competition between ß-ketothiolase and citrate synthase during poly(ß-hydroxybutyrate) synthesis in Methylobacterium rhodesianum MB 126. Arch. Microbiol. 166: 405-410. Mothes, G; Ackermann, J-U; Babel, W; 1998. Regulation of the PHB synthesis in Methylobacterium rhodesianum MB 126 growing on methanol or fructose. Arch. Microbiol. 169: 360-363. Mothes, G; Ackermann, J-U; 2005. Synthesis of Poly(3-hydroxybutyrate-co-4-hydrobutyrate) with target mole fraction of 4 hydroxybutyric acid units by two-stage continuous cultivation of Delftia acidovorans P4a. Eng. Life Sci. 5: 58 – 62. Mothes, G; Schnorpfeil, C; Ackermann, J-U; 2007. Production of PHB from crude glycerol. Eng. Life Sci. 7: 1-6. Narasimhan, K; Cearley, AC; Gibson, MS; Welling, SJ; 2005. Process for the solvent-based extraction of polyhydroxyalkanoates from biomass. U.S. Pat. Appl. Publ. (2005), US 2005287654, A1 20051229. Nissen, N; Mothes, G; Müller, J; Petermann, C; Scherzer, D. Applications of Biopolymers in Electronic Products, 6. Int. Symposium Renewable Resources (naro.tech), Erfurt, 6.-7.9. 2007. Phb Industrial, S.A., Brazil, 2006. Process for extracting and recovering polyhydroxyalkanoates from cellular biomass. PCT Int. Appl. (2006), WO 2006092033, A2 20060908. Pietrini, M; Roes, L; Patel, MK; Chiellini, E; 2007. Comparative life cycle studies on poly(3hydroxybutyrate)- based composites as potential replacement for conventional petrochemical plastics. Biomacromol. 8: 2210-2218. Pötter, M; Steinbüchel, A; 2005. Poly(3-hydroxybutyrate) granule-associated proteins: impacts on poly(3-hydroxybutyrate) synthesis and degradation. Biomaromol. 6: 552-560. Ramsey, JA; Berger, E; Ramsey, BA; Chavarie, C; 1990. Recovery of poly(3-hydroxybutyric acid) granules by surfactant-hypochlorite treatment. Biotechnol. Tech. 4: 221-226. Rehm, BHA; 2003. Polyester synthases: natural catalysts for plastics. Biochem. J. 376: 15-33. Rehm, BHA; Steinbüchel, A; 2002. PHA synthases: the key enzymes of PHA synthesis. in: Biopolymers 3a (Polyesters I) (Doi, Y; Steinbüchel, A; eds), Wiley-VCH, Weinheim, Germany, pp 173-215. Scandola, M; 1995. Polymer blends based on bacterial poly(3-hydroxybutyrate). Can. J. Microbiol. 41: 310-315. Schrecklies, E; 2001. PHB - ein vielseitiges Biopolymer für die Immunologie. Immunologie Aktuell, 2: 42-44. Schumann, D; Müller, RA; 2001. Verfahren zur Gewinnung von Polyhydroxyalkanoaten (PHA) oder deren Copolymeren. DE 10013514 A1. Senior, PS; Dawes, EA; 1973. The regulation of poly-β-hydroxybutyrate metabolism in Azotobacter beijerinckii. Biochem. J. 134: 225-238.

84

G. Mothes and O. Otto

Saito, Y; Doi, Y; 1994. Microbial synthesis and properties of poly(3-hydroxybutyrate-co-4hydroxybutyrate) in Comamonas acidovorans. Int. J. Biol. Macromol. 16: 99-104. Saito, T; Kobayashi, T; 2002. Intracellular degradation of PHAs. in: Biopolymers 3b (Polyesters II), (Doi, Y., Steinbüchel, A; eds), WILEY-VCH, Weinheim, pp 23-39. Sharma, R; Ray, AR; 1995. Polyhydroxybutyrate, its copolymers and blends. J. Macromol. Sci., Rev. Macromol. Chem. Phys., C35: 327-359.

Steinbüchel, A; Schlegel, HG; 1991. Physiology and molecular genetics of poly(βhydroxyalkanoic acid) synthesis in Alcaligenes eutrophus. Mol. Microbiol. 5: 535-542. Steinbüchel, A; Valentin, HE; 1995. Diversity of bacterial polyhydroxyalkanoic acids. FEMS Microbiol. Lett. 128: 219-228. Taidi, B; Anderson, AJ; Dawes, EA; Byrom, D; 1994. Effect of carbon source and concentration on the molecular mass of poly(3-hydroxybutyrate) produced by Methylobacterium extorquens and Alcaligenes eutrophus. Appl. Microbiol. Biotechnol. 40: 786–790. Tokiwa, Y., Ugwu, C.U., 2007. Biotechnological production of (R)-3-hydroxybutyric acid monomer. J. Biotechnol.132: 264-272. Tomita, K; Saito, T; Fukui, T; 1983. Bacterial metabolism of poly-β-hydroxybutyrate. In: Biochemistry of metabolic processes (Lennon, DLF; Stratman, FW; Zahlten, RN; eds), Elsevier Sci. Pub., Inc., pp 353-366. Valentin, HE; Schönebaum, A; Steinbüchel, A; 1992. Identification of 4-hydroxyvaleric acid as a constituent of biosynthetic polyhydroxyalkanoic acids from bacteria. Appl. Microbiol. Biotechnol. 36: 507-514. Van der Walle, GAM; De Koning, GJM; Weusthuis, RA; Eggink, G; 2001. Properties, modifications and applications of biopolyesters. Advances in Biochemical Engineering/Biotechnology (2001), 71(Biopolyesters), 263-291. VanWalsem, J; Anderson, E; Licata, J; 2006. Methods for solvent extraction of polyhydroxyalkanoates. US. Pat. Appl. Publ. 0060058501, A1 20060316. Weusthuis, RA; Kessler, B; Dielissen, MPM; Witholt, B; Eggink, G; 2002. Fermentative production of medium-chain length poly(3-hydroxyalkanoates); in: Biopolymers 3a (Polyesters I) (Doi, Y; Steinbüchel, A; eds), Wiley-VCH, Weinheim, Germany, pp 291316. Williams, SF; Martin, D; 2005. Applications of PHAs in medicine and pharmacy; in: Biopolymers for medical applications (Steinbüchel, A; Marchessault, RH; eds), WileyVCH, Weinheim, Germany, vol.1, pp 89-125. Yamane, T; Chen, X-F; Ueda, S; 1996. Growth associated production of poly(3hydroxyvalerate) from n-pentanol by methylotrophic bacterium, Paracoccus denitrificans. Appl. Environ. Microbiol. 62: 380-384. Yasotha, K; Aroua, MK; Ramachandran, KB; Tan, IKP; 2006. Recovery of medium-chainlength polyhydroxyalkanoates (PHAs) through enzymatic digestion treatments and ultrafiltration. Biochem. Eng. J. 30: 260-268.

In: Microbial Conversions of Raw Glycerol Editor: George Aggelis

ISBN 978-1-60692-392-4 © 2009 Nova Science Publishers, Inc.

Chapter VII

Single Cell Oil and Gamma-linolenic Acid Production by Thamnidium elegans Grown on Raw Glycerol Stylianos Fakas, Stamatina Bellou, Anna Makri and George Aggelis4 Division of Genetics, Cell and Development Biology, Department of Biology, University of Patras, Patras 26504, Greece

Abstract The oleaginous Zygomycete Thamnidium elegans was grown on raw glycerol producing high amounts of single cell oil (SCO) rich in gamma-linolenic acid (GLA). Produced biomass contained more than 40% w/w lipid, which contained 7.3% (w/w) GLA, giving a GLA yield of 664.3 mg/l GLA. This yield indicates that raw glycerol is an efficient substrate for SCO production. Lipid analysis showed that at the beginning of growth the mycelium was rich in phospholipids and glycolipids plus sphingolipids, while neutral lipids accumulated as growth proceeded. Fatty acid analysis of the major lipid fractions revealed that phospholipids were rich in linoleic acid and GLA, while the other two fractions had similar fatty acid composition.

Notation X Sm Xf=X+Sm L Glol N

4

Functional part of biomass (g/l) Storage material (g/l) Lipid free biomass (g/l) Lipid (g/l) Glycerol (g/l) Total extracellular nitrogen (g/l)

E-mail address: [email protected].

Stylianos Fakas, Stamatina Bellou, Anna Makri et al.

86 ΚN ΚGlol qLmax

Saturation constant for nitrogen (g/l) Saturation constant for glycerol (g/l) Maximum specific lipid production rate (g/gh)

qSmmax

Maximum specific storage material production rate (g/gh)

μmax YX/N YX/Glol YL/Glol YSm/Glol

Maximum specific growth rate (g/gh) Biomass yield on nitrogen consumed (g/g) Biomass yield on glycerol consumed (g/g) Lipid yield on glycerol consumed (g/g) Storage material yield on glycerol consumed (g/g)

Introduction Oleaginous microorganisms have long been used for the production of lipids designated as single cell oils (SCOs) to denote their microbial origin. The term oleaginous was used to describe all these microorganisms that have the ability to accumulate more than 20% w/w lipid in their biomass (Ratledge, 1991). This accumulation threshold, however, is somehow arbitrary, while a more accurate definition would be that a microorganism may be termed oleaginous if it produces a SCO that can be economically extracted and sold (Ratledge, 1991). Indeed, several yeast and mould species have long been considered as potential SCO producers (Woodbine, 1959), while oleaginous algae arrived sometime later on the scene (Bajpai and Bajpai, 1993). Oleaginous microorganisms comprise a small party having distinct biochemical features than the rest microbial species. In oleaginous species, the enzyme isocitrate dehydrogenase (ICDH) of the citric acid cycle is absolutely dependent on AMP for its activity (Ratledge and Wynn, 2002). Under nitrogen limitation, AMP is cleaved by AMP deaminase to produce NH4+ needed for cell functioning:

where IMP is inosine monophosphate. The drop in AMP concentration inhibits ICDH, which catalyzes the following reaction in the mitochondria:

Thus, isocitrate, and thence citrate accumulate in the mitochondria. Then, citrate exits the mitochondria to be cleaved in the cytosol by ATP:citrate lyase, an enzyme that is found only in oleaginous species:

Single Cell Oil and Gamma-linolenic Acid Production by Thamnidium elegans… 87 The acetyl-CoA produced by ATP:citrate lyase is channeled to lipid biosynthesis. This sequence of reactions ensures the efficient channeling of carbon to SCO synthesis. SCOs are the major sources of polyunsaturated fatty acids (PUFAs) having great pharmaceutical and nutraceutical interest (Certik and Shimizu, 1999). Microbial PUFA production has been established because of the lack of alternative sources of these fatty acids. Indeed, some PUFA, like arachidonic acid, are produced nowadays in large scale using oleaginous moulds, while industrial processes for the production of other PUFAs are under development (Ratledge, 2005). Gamma linolenic acid (GLA) is among PUFAs that have potential economical interest, because there are no abundant sources of this fatty acid in nature. Actually, GLA was the first PUFA to be biotechnologically produced in large scale to be used as a food additive, but the advent of plant oils having higher GLA content than the microbial source forced the GLA production process out of business (Ratledge, 2005). Interest in GLA production, however, has reemerged lately, after its selective anticancer properties were established (Kenny et al., 2000). However, the GLA to be used in cancer treatment should be of the highest purity which then makes SCOs the best GLA sources, since plant oils contain high amounts of other PUFAs along with GLA, making its purification very difficult (Ratledge and Wynn, 2002). Thus, several researchers sought to develop processes for GLA production by waste media because these media have zero or even negative cost (Certik et al., 1999; Papanikolaou et al., 2007; Fakas et al., 2008a; Fakas et al., 2008b). Several wastes have been used as substrates for SCO production originating mainly from agro-industries. The wastes used more often comprise molasses produced during sugar manufacture (Almazan et al., 1981; Misra et al., 1984; Johnson et al., 1995), and whey, a byproduct of cheese creameries (Bednarski et al., 1986; Ykema et al., 1988). On the other hand, raw glycerol accumulating during biodiesel manufacture has been rarely used as a substrate for the production of microbial lipids, despite its low cost and wide availability. In the only paper reporting SCO production on raw glycerol, the oleaginous mould Mortierella isabellina was employed to transform raw glycerol to GLA-rich SCO (Papanikolaou et al., 2008). The results of this study were quite encouraging, contradicting earlier reports stating that glycerol is a poor substrate for SCO production (Sajbidor et al., 1988; Chen and Chang, 1996). This then means that the search for efficient strains of oleaginous moulds that could transform raw glycerol into PUFA-rich microbial lipids is more than justified. Besides being a PUFA source, lipids play a major part in microbial physiology having an array of functional roles. Microbial lipids may be categorized by their polarity in three major groups: neutral lipids (NL), glycolipids and sphingolipids (G+S), and phospholipids (P). Each group has distinct physiological roles, but in some cases these roles may overlap. P are the building blocks of cell membranes regulating their function and stability (Carman and Henry, 1999), and are also the major site of PUFA biosynthesis (Certik and Shimizu, 1999). Glycolipids have not been extensively studied, so their function remains somehow obscure, but they have been implicated in membrane building and energy storage (Certik and Shimizu, 1999; Fakas et al., 2006). On the other hand, neutral lipids (NL) are considered as storage lipids, functioning as depositories of energy and biosynthetic precursors (Wagner and Daum, 2005; Fakas et al., 2007).

88

Stylianos Fakas, Stamatina Bellou, Anna Makri et al.

In this chapter, we report the production of GLA-rich SCO by the oleaginous Zygomycete Thamnidium elegans grown on raw glycerol and the development of a numerical model to describe this bioprocess. At the same time, we studied the composition of T. elegans lipids in an ongoing effort to draw the general picture of lipid composition during growth of oleaginous moulds and its physiological implications.

Materials and Methods Microorganisms and Culture Conditions Strains of Cunninghamella echinulata ATHUM 4411, Mortierella isabellina ATHUM 2935, M. ramanniana ATHUM 2922, Mucor sp. LGAM 365, and Thamnidium elegans CCF1465 were maintained on potato dextrose agar (PDA) at 6 ± 1 oC. Growth medium contained (g/l): Raw glycerol, 100; (NH4)2SO4, 2.0; yeast extract, 4.0; KH2PO4, 7; Na2HPO4, 2; MgSO4·7H2O, 1.5; CaCl2·2H2O, 0.1; FeCl3·6H2O, 0.08; ZnSO4·7H2O, 0.001; CuSO4·5H2O, 0.0001; Co(NO3)3·H2O, 0.0001, and MnSO4·5H2O, 0.0001. Raw glycerol was obtained from the Hellenic Industry of Glycerine and Fatty Acids, Athens, Greece. Its purity was 80% (w/w), with impurities composed mainly of potassium and sodium salts (4%, w/w), methanol (1%, w/w), heavy metals and lignin (1%, w/w), non-glycerol organic materials (0.5%, w/w) and water (13.5%, w/w). After sterilization (121 0C/20 min) flasks were inoculated with 1 ml of spore suspension containing 2 × 104 spores, which were produced by growing the strains on PDA for 8 days at 28 0C. All cultivation experiments were performed in 250 ml Erlenmeyer flasks, containing 50 ml of the above medium, incubated in a rotary shaker at 180 r.p.m. and 28 0C. Screening procedure was done by cultivating the moulds in the above medium for 120h and then collecting and determining their biomass and its lipid content.

Analytical Methods Total biomass was harvested by filtration under vacuum through Whatman No1 paper, washed with cold distilled water, dried at 80 oC for 12h, and then gravimetrically determined. Lipid-free biomass was calculated after subtraction of cellular lipids from total biomass. Glycerol was determined by HPLC (Dionex Ultimate 3000 ) in a Shodex SH1011 column coupled to a differential refractometer (RI), under the following conditions: sample volume 20 μl, mobile phase 0.001 N H2SO4, flow rate 1.0 ml/min and column temperature 500 C. Inorganic ammonium ion concentration (ΝΗ4+) was measured with a selective electrode (51927-00, Hach, Colorado, USA). Dissolved oxygen concentration and specific oxygen consumption rate were measured as described in (Papanikolaou et al., 2004b), with a selective electrode (Hana HI9146-04). Dissolved oxygen was > 70 % v/v of the saturation value during all growth phases.

Single Cell Oil and Gamma-linolenic Acid Production by Thamnidium elegans… 89 Fungal Lipid Content Determination The dried mycelia were ground into a fine powder and extracted three times with 100 ml of chloroform/methanol (2:1, v/v) for 48 hours at room temperature (Folch et al., 1957). The lipid extract was washed with 0.88 % w/v KCl (3 × 20 ml) and dried over anhydrous Na2SO4; the solvent was removed by evaporation.

Fractionation of Fungal Lipids A known weight of extracted lipid (approx. 100 mg) was dissolved in chloroform (1 ml) and fractionated by using a column (25 × 100 mm) of silicic acid (1 g), activated by heating overnight at 110 0C (Latge and De Bievre, 1980). Successive applications of 1,1,1trichloroethane (100 ml), acetone (100 ml), and methanol (50 ml) produced fractions containing neutral lipids (NL), glycolipids plus sphingolipids (G+S), and phospholipids (P), respectively. The weight of each fraction was determined after evaporation of the respective solvent. Lipid fractions were stored at –30 0C under a nitrogen atmosphere. All three lipid fractions were analyzed by thin layer chromatography (TLC) using various solvent systems (Fakas et al., 2006).

Gas Chromatography (GC) Analysis Trans-methylation of lipid extracts was performed according to the AFNOR method (AFNOR, 1984). GC analysis was carried out in an Agilent 7890Α device equipped with a HP-88 JandW Scientific column (60 m × 0.25 mm) and a FID detector; helium was the carrier gas (2 ml/min). The analysis was run at 200 0C with the injector and detector at 250 0 C. Fatty acid methyl esters were identified by reference to authentic standards.

Results Screening of Oleaginous Zygomycetes on Raw Glycerol Table 1 shows the biomass production and lipid accumulation in various Zygomycetes grown on raw glycerol. All moulds grew well on raw glycerol and produced significant amounts of biomass, while the lipid content in the produced biomass varied among the moulds. Mortierella isabellina and Thamnidium elegans accumulated high amounts of lipid in their biomass, while the other moulds could not surpass the 20% (w/w) lipid accumulation threshold. T. elegans, however, presented the best growth and lipid yield, and thus it was selected for further study.

Stylianos Fakas, Stamatina Bellou, Anna Makri et al.

90

Table 1. Screening of oleaginous Zygomycetes on raw glycerol. Culture conditions: initial pH 6 and temperature 28o C. Two lots of independent cultures were conducted by using different inocula, while the biomass was collected 120h after inoculation Microorganisms

Biomass (g/l)

Lipid content (%, w/w)

Cunninghamella echinulata

8.2

19.5

Mortierella isabellina

6.2

53.2

Mortierella ramanniana

7.0

22.0

Mucor sp

7.9

19.5

Thamnidium elegans

20.9

38.3

Model Presentation In order to study growth and lipogenesis in T. elegans on raw glycerol a mathematical model was developed (Table 2). Since lipid-free biomass formation is routinely observed for a considerable time after nitrogen exhaustion in cultures of oleaginous moulds, we assumed that this increase in lipid-free biomass is due to the synthesis of storage material (Sm) other than lipid. Thus, total lipid-free biomass was considered as comprising from a functional part (X) and storage material. This functional part should be synthesized when nitrogen is available, while after nitrogen exhaustion the carbon source should be channeled to the synthesis of storage material and lipid.

Model Fitted on the Experimental Data and Evaluation of Parameters The model was fitted on experimental data from cultures of T. elegans grown on raw glycerol (Figure 1). Runge–Kutta (4th order) and Mid-Point integration methods were used and the parameter values were optimized using the least squares method. The Marquardt iterative search algorithm was used in order to determine the parameter values that minimized the residual sum of squares. The model fitted the data very well, as indicated by the very high R2 value, while most of the predicted parameter values were close to the experimental ones (Table 3). Growth of T. elegans on Raw Glycerol T. elegans grew very well on raw glycerol and produced fair amounts of biomass having high lipid content (Figure 1). Glycerol was gradually consumed during growth and exhausted

Single Cell Oil and Gamma-linolenic Acid Production by Thamnidium elegans… 91 at the end of growth. Nitrogen was exhausted 48h after inoculation, but model predicted that nitrogen exhaustion occurred a little sooner, at 40h. Lipid accumulation, however, lagged far behind nitrogen exhaustion, starting 120h after inoculation. Table 2 Model equations describing growth, substrate consumption, and reserve lipid accumulation in Thamnidium elegans growing on raw glycerol Parameter Lipid free-biomass

Equation

Specific growth rate

μ = μ max ⋅

Functional biomass formation Lipid formation rate

dX =μ⋅X dt

X f = X + Sm N N + KN

dL = qL ⋅ X dt

No − N Glol ⋅ No K Glol + Glol

Specific rate of lipid formation Storage material formation rate Specific rate of storage material formation Raw glycerol consumption

q L = q L max ⋅

−

dGlol 1 1 1 =μ⋅X ⋅ + qL ⋅ X ⋅ + qP ⋅ X ⋅ dt Y X / Glol YL / Glol YS / Glol

Nitrogen consumption

−

dN 1 =μ⋅X ⋅ dt YX / N

dS m = q Sm ⋅ X dt

q S m = qSm max ⋅

No − N Glol ⋅ No K Glol + Glol

20

120

sampling

100

Glol (g/l)

60

10

40

sampling

Xf, L, N (g/l)

15

sampling

80

5

20 0

0 0

50

100

Glol (g/l)

150 Time (h) Xf (g/l) L (g/l)

200

250

300

N*10 (g/l)

Figure 1. Model fitted on the experimental data from cultures of Thamnidium elegans growing on raw glycerol. Two lots of independent cultures were conducted by using different inocula.

92

Stylianos Fakas, Stamatina Bellou, Anna Makri et al.

Table 3. Optimized values of growth parameters of Thamnidium elegans cultivated on raw glycerol. Culture conditions: initial pH 6 and temperature 28o C. Two lots of independent cultures were conducted by using different inocula Parameter

Value

ΚN (g/l)

0.15

ΚGlol (g/l)

20

qLmax (g/gh)

0.028

qSmmax (g/gh)

0.037

μmax (1/h)

0.089

YX/N (g/g)

3.93

YX/Glol (g/g)

0.08

YL/Glol (g/g)

0.43

YSm/Glol (g/g)

0.17

R2

0.98

After the completion of lipid accumulation, the produced biomass contained 42% (w/w) lipid. It should be noted, however, that lipid free biomass was produced throughout cultivation time. Actually, after 168h of cultivation, the rate of lipid free biomass synthesis was higher than the rate of lipid synthesis, resulting in the decrease of the lipid content in the produced biomass. Thus, the maximum lipid content was obtained at 168h being 47.9% (w/w). The produced amount of lipid, however, was only 6.8 g/l, while at the end of cultivation 9.1 g/l of lipid were produced.

Lipid Composition during Growth of T. elegans Figure 2 shows the lipid composition during the various growth phases of T. elegans (sampling time is shown in Figure 1). At the beginning of growth, the produced lipids were rich in polar lipids (glycolipids plus sphingolipids (G+S) and phospholipids (P)). Neutral lipids (NL), however, were the major constituent of T. elegans lipids. During growth, NL accumulated in the mycelia, which resulted at the production of lipids containing more than 80% (w/w) NL. Lipid analysis by TLC showed that NL comprised mainly from triacylglycerol, while diacylglycerol and monoacylglycerol were minor components. G+S fraction comprised mostly of monogalactosyl- and digalactosyl-diacylglycerol. P fraction

Single Cell Oil and Gamma-linolenic Acid Production by Thamnidium elegans… 93 contained phosphatidylcholine, phosphatidyl-ethanolamine, and lower quantities of phosphatidylinositol and phosphate-dylserine.

Fatty Acid Composition of the Produced Lipids Thamnidium lipids contained mostly oleic acid (C18:1Δ9), followed by palmitic (C16:0) and linoleic (C18:2 Δ9,12) acids, while stearic (C16:0) and γ-linolenic (GLA, C18:3 Δ6,9,12) acids were found in lower amounts. The GLA content of the oil was higher during active growth (9.9% w/w), while in the lipogenic phase GLA content started to decrease, reaching 7.3% (w/w) in the produced oil. The final GLA yield was 664.3 mg/l. Fatty acid analysis of the lipid fractions in late exponential phase showed that the P fraction was particularly enriched in PUFAs, while NL and G+S contained lower and almost equal amounts of PUFAs (Figure 3). NL contained very high amounts of C18:1 (almost 50% w/w), while G+S had the highest C16:0 content among lipid fractions. 100 NL

G+S

P

Percentage in total lipids (w/w)

80

60

40

20

0 Mid Exponential

Late Exponential

Stationary

Growth phase

Figure 2. Lipid composition during growth of Thamnidium elegans on raw glycerol. NL: Neutral lipids; G+S: Glycolipids plus sphingolipids; P: Phospholipids. Samples were taken for lipid analysis at time points indicated by arrows in Figure 1.

Stylianos Fakas, Stamatina Bellou, Anna Makri et al.

94

Percentage in NL, G+S, P (w/w)

60 50 40 30 20 10 0 C16:0

C18:0

C18:1

C18:2

GLA

Fatty acid

Figure 3. Fatty acid composition of lipid fractions (NL, G+S, P, see Fig 2) in late exponential phase.

Discussion Despite its significance, raw glycerol has been largely neglected as a substrate for SCO production by oleaginous microorganisms. Actually, as far as we are aware, there is only one study reporting SCO production by an oleaginous mould growing on raw glycerol (Papanikolaou et al., 2008). These authors cultivated Mortierella isabellina in media containing high amounts of raw glycerol, and reported that after a prolonged fermentation the mould produced 8.5g/l of biomass containing more than 50% w/w lipid. This lipid content corresponded to the production of 4.4 g/l SCO, which is much less than the corresponding amount of SCO produced by T. elegans. Turning to oleaginous yeasts, Yarrowia lipolytica grown on raw glycerol in a continuous system, it produced biomass that contained 43% w/w lipid, but the produced amount of SCO was only 3.5 g/l (Papanikolaou and Aggelis, 2002). However, the conversion efficiency of raw glycerol to SCO was almost the same for T. elegans, M. isabellina, and Y. lipolytica (0.08-0.09 g/g) indicating that these values may represent the conversion threshold of raw glycerol to SCO. Thus, the increased SCO production by T. elegans may be attributed to its ability to assimilate larger amounts of raw glycerol from the fermentation medium. Fermentation media formulated by wastes have occasionally been used for SCO production with some success. A comparison of the SCO yields obtained in these bioprocesses with the yields on raw glycerol will be made to put SCO production from raw glycerol into perspective. In most papers reporting SCO production on waste media, molasses are the substrate of choice because they contain high sugar amounts (almost 50% w/w) that may be transformed to lipid (Johnson et al., 1995). The red yeast Rhodotorula glutinis grown

Single Cell Oil and Gamma-linolenic Acid Production by Thamnidium elegans… 95 on molasses it produced biomass containing 28% (w/w) lipid (Almazan et al., 1981), while in a subsequent study, Misra et al. (1984) using the same yeast reported much higher yields the lipid content being 50% (w/w). In the latter study, however, the produced amount of SCO was only 2.5 g/l (Misra et al., 1984). This SCO amount was lower than the one reported by Johnson et al. (1995) who cultivated another strain of R. glutinis in an optimized molasses medium and produced 4 g/l of SCO. Whey and molasses were used as co-substrates for SCO production by the yeast Candida curvata, which produced 8.5 g/l of SCO. Even higher SCO amounts, however, were obtained by growing Apiotrichum curvatum on whey permeate; under optimized culture conditions, this yeast produced 11.5 g/l of SCO (Ykema et al., 1988). Whey was also used as a nitrogen source in SCO production by Cunninghamella echinulata grown on glucose-rich media, but SCO yields were as low as 2.4 g/l (Fakas et al., 2008a).On the contrary, much higher SCO yields were obtained when tomato waste was substituted for whey. In a detailed study, tomato waste was evaluated as a co-substrate for SCO production on glucose, and very high SCO yields were obtained reaching 8.7 g/l (Fakas et al., 2008b). Substitution of glycerol for glucose in the tomato waste medium, however, resulted in lower SCO yields (4.5 g/l). Comparing the SCO yield of T. elegans on raw glycerol with the yields obtained on various waste substrates, it can be concluded that raw glycerol is a very efficient substrate for SCO production. Indeed, the amount of SCO produced on raw glycerol is second only to that produced on whey. Given that the latter bioprocess was optimized to achieve such high yields, the potential of raw glycerol for lipid production seems even greater. Fatty acid composition of T. elegans lipids was typical of oleaginous Zygomycetes. The decrease in GLA content of lipids during lipid accumulation is routinely observed in oleaginous Zygomycetes (Kennedy et al., 1993; Kavadia et al., 2001). An explanation has been put forward for this phenomenon stating that Zygomycetes synthesise only as much GLA as needed for proper membrane functioning (Kavadia et al., 2001). This GLA is synthesised mainly during the exponential phase (Fakas et al., 2006), when the fast growing mycelia need GLA to support their expanding cellular membranes. Therefore, GLA synthesis correlates strongly with microbial growth, while lipid accumulation takes place under nongrowth conditions. The produced amount of GLA by T. elegans is considered satisfactory, as it is about the average yields usually reported. C. echinulata grown on a glucose medium it produced 720 mg/l GLA (Gema et al., 2002), while M. isabellina produced 801 mg/l GLA on high sugar content media (Papanikolaou et al., 2004a). However, optimization of GLA production by C. echinulata on starch resulted in yields as high as 1349 mg/l (Chen and Liu, 1997). It should be stated though that most studies report GLA production from costly semi-synthetic media, while studies on GLA production from low value waste media are scarce. Recently, Papanikolaou et al. (2008) reported GLA production by M. isabellina grown on raw glycerol, but the GLA yield obtained (120 mg/l) was much lower than the one achieved in the present study. However, cultivation of C. echinulata on a medium containing tomato waste supplemented with glycerol yielded some 530 mg/l GLA, while substitution of glucose for glycerol resulted in the production of more than 1g/l GLA (Fakas et al., 2008b). Nonetheless, GLA yields on raw glycerol indicate that it could be used for production this high-valued fatty acid.

96

Stylianos Fakas, Stamatina Bellou, Anna Makri et al.

Most of our knowledge in microbial lipid physiology comes from studies in Saccharomyces cerevisiae (Carman and Henry, 1999; Wagner and Daum, 2005), while lipids of oleaginous moulds have been rarely investigated. Detailed lipid analysis in T. elegans showed that NL accumulate as the mycelium ages, while triacylglycerol seemed to be the major NL class. These results are in agreement with the generally accepted physiological role of NL in microorganisms that of energy storage (Wagner and Daum, 2005). This role was further supported by the work of Fakas et al. (2007) who showed that NL are preferentially consumed under carbon starvation conditions. On the contrary, P are considered as structural lipids being major components of cell membranes (Carman and Henry, 1999), and this fact explains why they are synthesised during active growth. Work with C. echinulata showed that P are synthesised during the exponential phase, while phosphatidylcholine and phosphatidylethanolamine were the major P classes (Fakas et al., 2006). Also in C. echinulata, G+S content of lipids was higher at the mid exponential phase, decreasing at the late exponential phase, while a further slight decrease was observed at the stationary phase (Fakas et al., 2006). Overall, the pattern of lipid composition in T. elegans resembled closely that of C. echinulata, which is probably the only oleaginous mould whose lipids were analysed in detail. The fatty acid profile of the various lipid fractions was somehow distinct. The increased C18:2 and GLA concentrations in P fraction are in good agreement with the data reported on C. echinulata (Fakas et al., 2006). In the latter mould, however, P and G+S contained higher amounts of unsaturated fatty acids, the GLA content ranging from 18%-25% w/w in each fraction during growth (Fakas et al., 2006). This would then mean that GLA contributes more to the structure of cell membranes in C. echinulata than in T. elegans. However, the extent of the effect that GLA content of cellular membranes has on their function is not known, besides that increasing the PUFA content, membrane fluidity increases (Certik and Shimizu, 1999). Nevertheless, the latter statement concerning membrane fluidity is considered as an oversimplification of membrane dynamics.

Conclusion Raw glycerol seems like a promising substrate for SCO and GLA production by Thamnidium elegans. Taking into account the zero cost of raw glycerol, a bioprocess for SCO production using raw glycerol as feedstock may be visualized. This bioprocess could provide an alternative to raw glycerol management, which is currently one of the major drawbacks of biodiesel manufacture. Study of lipids during growth of T. elegans showed that their composition follows the same trend as in C. echinulata. This resemblance may indicate a general pattern in oleaginous Zygomycetes, but further research is needed to prove this trend. There is no doubt that these proofs would then have much to contribute in our understanding of the physiology of oleaginous moulds.

Single Cell Oil and Gamma-linolenic Acid Production by Thamnidium elegans… 97

Acknowledgments Financial support was provided by the project ‘‘Kinetics of growth of oleaginous microorganisms and dynamics of biosynthesis of polyunsaturated fatty acids” funded by the University of Patras (Project K. Karatheodori) and the project “Biodiesel production from agro-industrial by-products’’ funded by the Greek Fuel Company DRACOIL SA.

References AFNOR, 1984. Recueil des normes francaises des corps gras, grains oleagineux et produits derives Association Francaise pour Normalisation. 3rd ed, Paris, pp. 95. Almazan, O., Klibansky, M., Otero, M.A., (1981). Microbial fat synthesis by Rhodotorula glutinis from blackstrap molasses in continuous culture. Biotechnology Letters, 3, 663666. Bajpai, P., Bajpai, P.K., (1993). Eicosapentaenoic acid (EPA) production from microorganisms: a review. Journal of Biotechnology, 30, 161-183. Bednarski, W., Leman, J., Tomasik, J., (1986). Utilization of beet molasses and whey for fat biosynthesis by a yeast. Agricultural Wastes, 18, 19-26. Carman, G.M., Henry, S.A., (1999). Phospholipid biosynthesis in the yeast Saccharomyces cerevisiae and interrelationship with other metabolic processes. Progress in Lipid Research, 38, 361-399. Certik, M., Megova, J., Horenitzky, R., (1999). Effect of nitrogen sources on the activities of lipogenic enzymes in oleaginous fungus Cunninghamella echinulata. Journal of General and Applied Microbiology, 45, 289-293. Certik, M., Shimizu, S., (1999). Biosynthesis and regulation of microbial polyunsaturated fatty acid production. Journal of Bioscience and Bioengineering, 87, 1-14. Chen, H., Chang, C., (1996). Production of γ-linolenic acid by the fungus Cunninghamella echinulata CCRC 31840. Biotechnology Progress, 12, 338-341. Chen, H., Liu, T., (1997). Inoculum effects on the production of γ-linolenic acid by the shake culture of Cunninghamella echinulata CCRC 31840. Enzyme and Microbial Technology, 21, 137-141. Fakas, S., Certik, M., Papanikolaou, S., Aggelis, G., Komaitis, M., Galiotou-Panayotou, M., (2008a). [gamma]-Linolenic acid production by Cunninghamella echinulata growing on complex organic nitrogen sources. Bioresource Technology, 99, 5986–5990. Fakas, S., Galiotou-Panayotou, M., Papanikolaou, S., Komaitis, M., Aggelis, G., (2007). Compositional shifts in lipid fractions during lipid turnover in Cunninghamella echinulata. Enzyme and Microbial Technology, 40, 1321–1327. Fakas, S., Papanikolaou, S., Galiotou-Panayotou, M., Komaitis, M., Aggelis, G., (2006). Lipids of Cunninghamella echinulata with emphasis to γ-linolenic acid distribution among lipid classes. Applied Microbiology and Biotechnology, 73, 676-683. Fakas, S., Papanikolaou, S., Galiotou-Panayotou, M., Komaitis, M., Aggelis, G., (2008b). Organic nitrogen of tomato waste hydrolysate enhances glucose uptake and lipid

98

Stylianos Fakas, Stamatina Bellou, Anna Makri et al.

accumulation in Cunninghamella echinulata. Journal of Applied Microbiology, 105, 1062 - 1070. Folch, J., Lees, M., Sloane-Stanley, G., (1957). A simple method for the isolation and purification of total lipides from animal tissues. Journal of Biological Chemistry, 199, 833-841. Gema, H., Kavadia, A., Dimou, D., Tsagou, V., Komaitis, M., Aggelis, G., (2002). Production of γ-linolenic acid by Cunninghamella echinulata cultivated on glucose and orange peel. Applied Microbiology and Biotechnology, 58, 303-307. Johnson, V.W., Singh, M., Saini, V.S., Adhikari, D.K., Sista, V., Yadav, N.K., (1995). Utilization of molasses for the production of fat by an oleaginous yeast, Rhodotorula glutinis IIP-30. Journal of Industrial Microbiology and Biotechnology, 14, 1-4. Kavadia, A., Komaitis, M., Chevalot, I., Blanchard, F., Marc, I., Aggelis, G., (2001). Lipid and γ-linolenic acid accumulation in strains of Zygomycetes growing on glucose. Journal of the American Oil Chemists Society, 78, 341-346. Kennedy, M., Reader, S., Davies, J., (1993). Fatty acid production characteristics of fungi with particular emphasis on gamma linolenic acid production. Biotechnology and Bioengineering, 42, 625-634. Kenny, F., Pinder, S., Ellis, I., Gee, J., Nicholson, R., Bryce, R., Robertson, J., (2000). Gamma linolenic acid with tamoxifen as primary therapy in breast cancer. International Journal of Cancer, 85, 643-648. Latge, J., De Bievre, C., (1980). Lipid composition of Entomophthora obscura Hall and Dunn. Journal of General Microbiology, 121, 151-158. Misra, S., Ghosh, A., Dutta, J., (1984). Production and composition of microbial fat from Rhodotorula glutinis. Journal of the Science of Food and Agriculture, 35, 59-65. Papanikolaou, S., Aggelis, G., (2002). Lipid production by Yarrowia lipolytica growing on industrial glycerol in a single-stage continuous culture. Bioresource Technology, 82, 4349. Papanikolaou, S., Fakas, S., Fick, M., Chevalot, I., Galiotou-Panayotou, M., Komaitis, M., Marc, I., Aggelis, G., (2008). Biotechnological valorisation of raw glycerol discharged after bio-diesel (fatty acid methyl-esters) manufacturing process: production of 1,3propanediol, citric acid and single cell oil. Biomass and Bioenergy, 32, 60-71. Papanikolaou, S., Galiotou-Panayotou, M., Fakas, S., Komaitis, M., Aggelis, G., (2007). Lipid production by oleaginous Mucorales cultivated on renewable carbon sources. European Journal of Lipid Science and Technology, 109, 1060-1070. Papanikolaou, S., Komaitis, M., Aggelis, G., (2004a). Single cell oil (SCO) production by Mortierella isabellina grown on high-sugar content media. Bioresource Technology, 95, 287-291. Papanikolaou, S., Sarantou, S., Komaitis, M., Aggelis, G., (2004b). Repression of reserve lipid turnover in Cunninghamella echinulata and Mortierella isabellina cultivated in multiple-limited media. Journal of Applied Microbiology, 97, 867-875. Ratledge, C., (1991). Microorganisms for lipids. Acta Biotechnologica, 11, 429-438. Ratledge, C., 2005. Single Cell Oils for the 21st Century. in: Z. Cohen, C. Ratledge (Eds.), Single Cell Oils. ed. AOCS Press, Champaign, Illinois, pp. 1-20.

Single Cell Oil and Gamma-linolenic Acid Production by Thamnidium elegans… 99 Ratledge, C., Wynn, J., (2002). The biochemistry and molecular biology of lipid accumulation in oleaginous microorganisms. Advances in Applied Microbiology, 51, 151. Sajbidor, J., Certik, M., Dobronova, S., (1988). Influence of different carbon sources on growth, lipid content and fatty acid composition in four strains belonging to mucorales. Biotechnology Letters, 10, 347-350. Wagner, A., Daum, G., (2005). Formation and mobilization of neutral lipids in the yeast Saccharomyces cerevisiae. Biochemical Society Transactions, 33, 1174-1177. Woodbine, M., (1959). Microbial fat: Microorganisms as potential producers. Progress in Industrial Microbiology, 1, 179-245. Ykema, A., Verbree, E.C., Kater, M.M., Smit, H., (1988). Optimization of lipid production in the oleaginous yeast Apiotrichum curvatum in wheypermeate. Applied Microbiology and Biotechnology, 29, 211-218.

In: Microbial Conversions of Raw Glycerol Editor: George Aggelis

ISBN 978-1-60692-392-4 © 2009 Nova Science Publishers, Inc.

Chapter VIII

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids by Fungi F. Mantzouridou5 Laboratory of Food Chemistry and Technology, School of Chemistry, Aristotle University of Thessaloniki, 541 24 Thessaloniki, Greece

Abstract Importance and fermentation technologies for the production of food-grade carotenoids by fungi are presented with focus on raw glycerol employment. Information about fungal growth, substrate assimilation and carotenoid production parameters such as yield, selectivity, productivity and process economics are detailed. Toxicological aspects of the raw glycerol are also discussed.

Keywords: raw glycerol; food-grade carotenoids; fungi; Blakeslea trispora

Abbreviations Used Acetyl-coA, acetyl-coenzyme A; ADP, adenosine diphosphate; ATP, adenosine triphosphate; NAD, nicotinamide adenine dinucleotide; NADP, nicotinamide adenine dinucleotide phosphate; RP-HPLC, reversed-phase high-pressure liquid chromatography

5

Tel. ++30 2310 997791, Fax. ++30 2310997779, E-mail: [email protected].

102

F. Mantzouridou

Introduction Carotenoids, the most common, naturally occurring terpenoid pigments, are of great interest in many scientific fields because of their wide distribution, diverse functions, and interesting properties. Several carotenoids such as β-carotene, astaxanthin, lutein, canthaxanthin, and lycopene are important industrially as nutrient supplements, food colorants, and feed additives. Scientific interest in dietary carotenoids has increased in recent years due to claims for beneficial effects on human health, such as a reduced risk of cancer and enhancement of immune system function ((Hinds et al., 1997), attributed to their antioxidative potential. Carotenoids used in industry are mostly manufactured using chemical synthesis or are natural extracts or concentrates. Penetration of microbial carotenoids into the food sector has accelerated during the last years in relation to consumer preferences for natural additives. The main carotenoids obtained so far through fermentation that are already used in food/feed or have great potential for future commercialization are presented in Table 1. Among microorganisms, fungi appear most promising industrial sources of carotenoids since fungal cells can be cultivated heterotrophically in large-scale fermentors to accumulate high levels of carotenoids. However, the economic feasibility of the bioprocess depends strongly on the magnitude of costs involved for the production and commercialization of the final product and its intended utilization. Process innovations concerning microbial upgrading of residues and wastes for carotenoid production will improve the economy of the bioprocess. Raw glycerol is the primary by-product of oils/fats processing for the production of soaps, waxes, fatty acids and detergents. Currently, another source for raw glycerol is the production of bio-fuels. Despite that purified glycerol is used in many applications including foods and beverages, cosmetics and pharmaceuticals, refining of raw glycerol is costly. Moreover, the increase in glycerol supplies has created a significant glut in the glycerol global market with a concomitant reduction of its price. Currently, glycerol-rich streams from oleochemical and biodiesel plants are treated as waste materials (Johnson and Taconi, 2007; Yazdani and Gonzalez, 2007). Thus, upgrading processes are particularly valid for the exploitation of these by-products. The introduction of new efficient technologies for the biological conversion of raw glycerol into value-added products seems promising (biorefining approach). The gamut of products deriving from the biological exploitation of raw glycerol ranges from biodegradable plastics and platform chemicals to fine chemicals. Still, research on (raw) glycerol employment in the production of fungal carotenoids is limited. The chapter provides the scientific foundation about fungal growth and carotenoid biosynthesis and describes the status of research efforts to convert (raw) glycerol into commercially valued fungal carotenoids. Trying to facilitate critical appreciation of published data, a discussion topic about the prospect to produce food-grade carotenoids with desired properties together with economical profit using raw glycerol is also included. Moreover, due to current legislative demands on food safety precautions, the last part of the chapter is devoted to the toxicological aspects of raw glycerol employed that may direct more efficiently design of the bioprocess in future trials in both laboratory and large scale ensuring that the desired product is a safe choice for use in food.

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids... 103 Table 1. Microbial carotenoids already in use or with potential of commercialization in food/feed sector

Carotenoid

β-Carotene

Lycopene

Lutein

Zeaxanthin

Canthaxanthin

Astaxanthin

Structure

Major commercial uses

Microbial source

Maximum yield reported

Reference

Blakeslea trispora (fungus)

4000-5000 mg/L of culture medium

Feofilova et al. (2006)

Dunaliella sp. (algae)

400 mg/m2 of cultivation area

Dufosse´ et al. (2005)

food colorant, dietary supplement

Blakeslea trispora (fungus)

1300-1500 mg/L of culture medium

Feofilova et al. (2006)

dietary supplement, poultry feed

Chlorella protothecoides (algae)

225 mg/L of culture medium

Shi et al. (2002)

Flavobacterium multivorum (bacterium)

11 mg/L of culture medium

Bhosale et al. (2004)

Dunaliella salina

6 mg/g of dry biomass

Jin et al. (2003)

Brevibacterium sp. (bacterium)

9.3 mg/L of culture medium

Nelis & De Leenheer (1989)

Haematococcus pluvialis (algae)

30 mg/g of dry biomass

Lorenz & Cysewski (2000)

Phaffia rhodozyma (red yeast)

4 mg/g of dry biomass

Jacobson et al. (2000)

food colorant, dietary supplement

dietary supplement, poultry feed, aquaculture

poultry feed, aquaculture

aquaculture

104

F. Mantzouridou

Fungal Carotenoids Carotenoids are well known fungal secondary metabolites. Several fungal species within Zygomycetes, Ascomycetes, Basidiomycetes and Deuteromycetes have been reported as potential sources of industrially important carotenoids such as β-carotene, lycopene and astaxathnin (Goodwin, 1952; Nelis and Leenheer, 1991; Duffossé, 2006). Lower filamentous fungi of the order Mucorales (class Zygomycetes) are carotene hyper-producers. Among them, Blakeslea trispora is an active producer of β-carotene (up to 4–5 g/L of culture medium) and lycopene (0.7-1.3 g/L of culture medium) on an industrial scale (e.g. Ciegler et al., 1963a; López-Nieto et al., 2004; Feofilova et al., 2006). Phycomyces blakesleeanus on the other hand was found to be an excellent system for the elucidation of the effects of several nutritional and environmental factors as well as of genetic changes on carotenogenesis (e.g. Garton et al., 1951; Goodwin and Willmer, 1952; Murillo and CerdáOlmedo, 1976; Murillo et al., 1978). Strain improvement in this fungus has led to an increase in the β-carotene content (up to 10 mg/g of biomass dry weight), although this concentration is notably lower than that reported for B. trispora (Murillo et al., 1978). Research studies also have focused on regulation of carotenogenesis in dimorphic (i.e. yeast-mycelium dimorphism) species of Mucor (e.g. Mucor circinelloides, Mucor rouxii), which could be useful for industrial production of carotenoids (e.g. Mosqueda-Cano and Gutierrez-Corona, 1995; Velayos et al., 2000; 2004). Neurospora crassa (class Ascomycetes) is another carotenoid-producing fungus that has been used to study the biosynthesis of carotenoids and regulation of the carotenoid pathway (e.g. Zalokar, 1954; Kritsky et al., 1982; Iigusa et al., 2005). Using metabolic engineering, the expression in the fungus Fusarium sporotrichioides (class Deuteromycetes) of carotenoid biosynthetic genes from Erwinia uredovora has provided a convenient system to produce transformed strains of the fungus that produce lycopene (Leathers et al., 2004). However, the yield of the target carotenoid was low (0.5 mg/g of dry biomass). Among the imperfect fungi, Phaffia rhodozyma (currently known as Xanthophyllomyces dendrorhous, class Deuteromycetes) has become an important tool for large-scale production of astaxanthin, with respect to use for animal feed (Jacobson, 2000). Moreover, β-carotene production has been made successful in Rhodotorula spp. (class Basidiomycetes) through mutation (Wang et al., 2007). Torularhodin and for a smaller part β-carotene were the two carotenoids identified in the red yeast Sporobolmyces ruberrimus (Razavi et al., 2006). Advanced knowledge of fungal genetics and molecular biology of carotenoid biosynthesis allowed also carotenoid production in non-carotenogenic microbial systems, e.g. Saccharomyces cerevisiae and Candida utilis (Misawa and Shimada, 1998; Verwaal et al., 2007). However, as recently described by Duffossé and collaborators (2005), “There is a long way from the Petri dish to the market place.”. The practical obstacles for the commercial exploitation of fungal carotenoids are production cost, extensive toxicological testing and regulatory issues, and acceptance by consumers. To date, only few fungal carotenoids have been authorized for food use by the EU. These are β-carotene (Commission directive (EC) No 50/2001) and lycopene (Commission regulation (EC) No 721/2006) derived from B. trispora. In Europe, astaxanthin from P. rhodozyma is allowed for use as an additive in

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids... 105 salmon and trout diet (Commission regulation (EC) No 1288/2004). The continuous interest in B. trispora as an industrial source of carotenoids is related to its legal status worldwide along with fungal ability to produce efficient amount of carotenoids that encourages further studies for the optimization of the bioprocess. Therefore, our attention will be mainly drawn to B. trispora.

Food-grade Carotenoids by Blakeslea Trispora B. trispora exists in two sexual forms [(+) and (-) mating type] (Feofilova et al., 1997). Both strains were not found to have pathogenic and toxigenic effect (SCF, 2000; JEFCA, 2001). Hyper-production of β-carotene by B. trispora occurs when mycelia of the two sexes are co-cultivated (mating) (Anderson et al., 1958; Ciegler et al., 1965). During mating the synthesis of trisporic acids (Figure 1) is favoured, which in turn stimulates carotenogenesis (Lampila et al., 1985a). COOH

O

R

Trisporic acid B: R = -C-CH3 Trisporic acid C: R = -CHOCH3 Figure 1. Chemical structure of trisporic acids.

B. trispora contains β-carotene and its precursor forms that are phytoene, phytofluene, lycopene, and γ-carotene (Hsu et al., 1972). β-Carotene is synthesized as the main carotenoid under the conventional cultivation conditions (Lampila et al., 1985a; Bhosale, 2004). To promote the accumulation of lycopene, inhibition of lycopene cyclase activity by chemical or genetic means is required. The use of special chemical compounds such as substituted amines and nitrogenous heterocyclic bases for the regulation of lycopene formation in B. trispora cells has been revealed in our recent review (Mantzouridou and Tsimidou, 2008). The process steps for β-carotene production by B. trispora are shown in Figure 2. Separate seed cultures for the original (+) and (-) strains are prepared and used to inoculate the production medium. The bioprocess is performed in aerobic submerged batch fermentation under culture conditions which can yield a fungal biomass with high β-carotene content. In the recovery process, biomass is harvested and β-carotene is extracted with the aid of food grade solvents (e.g. ethyl acetate), highly concentrated and crystallized. The final product is either crystalline β-carotene (purity > 96.0 %) or it is formulated as a 30 % suspension in vegetable oil. Analytical evidence (i.e. HPLC analysis, stability tests and microbiological tests) has confirmed that the final crystalline product conforms to all the specifications set out in Directive 95/45/EC for colouring agent in foodstuffs (SCF, 2000).

106

F. Mantzouridou

The production process of lycopene from B. trispora is nearly identical to that used to manufacture β-carotene. The only difference is that imidazole is added to the culture medium in order to inhibit lycopene cyclase (Commission regulation (EC) No 721/2006).

Figure 2. Flow diagram of production of carotenoids by B. trispora.

Following the fermentation phase, the extraction and purification processes are the same used during recovery of β-carotene, which have been approved by the SCF (SCF, 2000). Purity specifications for the final lycopene product ensure that its use as a novel food ingredient in foodstuffs is safe (Commission regulation (EC) No 721/2006). Commercial

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids... 107 lycopene preparations are formulated either as α-tocopherol containing suspensions in edible oils or as water-dispersible powders. Currently, the companies that produce carotenoids from B. trispora for commercial purposes, mainly as colour in foods and beverages, are DSM (formerly Gist-Brocades), Vitan Ltd in Russia and Ukraine, and Vitatene in León (Spain). Additional applications with economical importance are to be found in nutraceutical industries (i.e. pharmaceutical) or as a low-fiber feed supplement (Purcell and Walter, 1971; Flegal et al., 1971; Salikova et al., 1979). Biosynthesis of Carotenoids A schematic representation of the reactions producing carotenoids is shown in Figure 3. Carotenoids in B. trispora derive from mevalonate (MVA) with acetyl-CoA as the precursor (Ciegler, 1965). As shown in Figure 3, after a sequence of reactions (R 1-3), MVA is converted to isopentenyl pyrophosphate (IPP), which is the crucial C5 terpene precursor. This latter compound in the presence of IPP isomerase forms dimethylallyl pyrophosphate (DMAPP) (R 4); the reaction is freely reversible with the equilibrium mixture of IPP and DMADP being approximately 1:9 (Ladygin, 2000). Successive addition of three IPP molecules to DMAPP catalysed by prenyl transferase leads to the formation of the ubiquitous halfway metabolite precursor of C40 carotenoids that is geranylgeranyl pyrophosphate (GGPP) (R 5-7). A tail to tail condensation of two GGPP molecules by phytoene synthase yields the primal carotenoid in the family, phytoene ((R 8). Phytoene lacks the most appealing characteristic of carotenoids, color. The successive desaturation of the phytoene skeleton leads to lycopene synthesis through the intermediates phytofluene, ζ-carotene and neurosporene (R 9-12) (Mehta and Cerdá-Olmedo, 1995). The desaturation process is catalysed by phytoene desaturase. Finally, lycopene cyclase, using lycopene as substrate, forms the rings located at both ends of the β-carotene molecule with γ-carotene as the intermediate in the cyclization (R 13, 14).

Carotenoid Production in Various Medium Formulations Literature data concerning the utilization of various carbon and nitrogen sources and different culture conditions are briefly presented below. These examples are used to point out the range of media used, but are not necessarily the best media in current use for carotenoid production by B. trispora. Carbon Sources

Sugars and other Pure Carbon Sources Literature presents numerous scientific efforts in understanding the effect of individual nutrients on carotenogenesis by B. trispora. Among the various carbohydrates tested, glucose has been reported to be an excellent carbon source for fungal growth. However, fast growth due to high level of rapidly metabolized glucose is associated with low carotenoid yields (up to 100 mg/L of culture medium) (Ciegler et al., 1959b; Dandekar and Modi, 1980;

108

F. Mantzouridou

Mantzouridou et al., 2002; Choudhari and Singhal, 2008). Mechanistically, this has been attributed to catabolite repression (Dandekar and Modi, 1980). Substitution of glucose by cellobiose increased carotenoid yield by about 10 times (Dholakia and Modi, 1982). Results indicated that the activity of β-glucosidase, which cleaves cellobiose to glucose, is low. As a result, the enzymatic hydrolysis of cellobiose and, thus, the rate of glucose realize into the medium is slow. This was considered to help derepression of the catabolite repression in fungal cells and accumulation of higher carotene level. Lower yields were obtained in media containing other readily asssimilable carbohydrates such as fructose and maltose, although biomass formation was satisfactory (Choudhari and Singhal, 2008). Sucrose and lactose were found to be poor substrates with respect to both cell growth as well as carotenoid production (Mantzouridou et al., 2002; Choudhari and Singhal, 2008). This was ascribed to the inability of the fungus to produce invertase and βgalactosidase that break down sucrose and lactose, respectively, into their constituents. In addition to sugars, B. trispora can utilize a number of “non-conventional” carbon sources such as saccharide biopolymers (e.g. starch), lipids, fatty acids, and hydrocarbons. Experimentation trials using dextrins, potato infusion, tapioca dextrin, and soluble starch as sole carbon source showed that these carbon sources did not affect carotenoid production, although biomass production was satisfactory (Choudhari and Singhal, 2008). In B. trispora, triggering of carotenoid synthesis can take place in culture media enriched with vegetable oils or individual fatty acids (Ciegler et al., 1959a; Pazola et al., 1967; 1968; Bekhtereva and Dedyukhina, 1970; Mantzouridou et al., 2002). Evidence so far indicates that, fungal cells grown in media containing vegetable oils (e.g. soybean oil, cottonseed oil) as supplementary carbon source to glucose (or hydrolyzed starch) stimulated growth to a certain extent but stimulated carotenogenesis to a much greater degree (Ciegler et al., 1959a; Mantzouridou et al., 2006). Further improvement in carotenoid yield can be obtained in media containing an equal ratio of lipid to hydrocarbon (kerosene) (Ciegler et al., 1962). However, most studies have focused merely on carotenogenesis in terms of determination of the total carotenoid content. In a recent investigation (Mantzouridou and Tsimidou, 2007), RP-HPLC procedures were employed to follow changes in the carotenoid and triacylglycerol (TAG) composition of the intracellular lipids produced with respect to TAG composition of crude olive pomace (COPO) and soybean oil (CSO). From results it was revealed the interdependence of carotenoid pattern with the type and amount of extracellular oils. In particular, β-carotene production was found to be in expense to lycopene and γ-carotene formation in cells grown on COPO and CSO at two levels of addition (10.0 and 30.0 g/L of culture medium). Still, in culture media enriched with low oil level, CSO was more effective in stimulating β-carotene synthesis than COPO (75 versus 62% of total carotenoids and 130 versus 36 mg/g of biomass dry weight, respectively). Moreover, a shift to γ-carotene synthesis was observed at increased oil level.

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids... 109 Biosynthesis of mevalonate

Mevalonate 1 2 3 4 POPO

POPO

H

Isopentenyl pyrophosphate

POPO

Dimethylallyl pyrophosphate (DAPP) 5 6 2IPP 7

OPOP Geranylgeranyl pyrophosphate (GGPP) 8

GGPP

Phytoene 9 10 11 12

Lycopene 13

γ-carotene 14

β-carotene

Figure 3. Biosynthetic route of carotenoids in B. trispora cells.

Isopentenyl pyrophosphate (IPP)

F. Mantzouridou

110

Waste-based Media A variety of economical substrate constituents available from food processing operations have been tested for their effectiveness on carotenoid production by B. trispora. Traditionally, production media used for carotenoid production by B. trispora contained starch from different types of grains (e.g. hydrolyzed or unhydrolyzed ground whole soy, soybean oil meal, soybean whey, oats, wheat, barley, corn, rice and rye) (Ciegler et al., 1959b). The acid-hydrolyzed whole ground soy, the acid-hydrolyzed hexane extracted soybean oil meal and the unhydrolyzed corn were found to stimulate carotene production (350-400 mg/L of culture medium). However, their effects are not conclusively elucidated due to their complex nutrient composition. Moreover, in a research paper of the same group (Ciegler et al., 1963b) β-ionone, a stimulatory compound of carotenogenesis in B. trispora, was replaced successfully with low-cost agricultural byproducts like citrus oils, cirus pulp, or citrus molasses. Carotene yield was found to be greater in those treatments than in treatment supplied with β-ionone (up to 129 mg versus 98 mg/100 mL of culture medium, respectively). The stimulatory compound was thought to be d-limonene. Molasses is a by-product of the sugar industry readily available at relatively low cost. It contains water, approximately 50% (w/w) sugars (sucrose, glucose, fructose, raffinose), nitrogen compounds, organic acids, amino acids, heavy metals, etc. Whey is a by-product of cheese manufacture and a potential environmental pollutant. It consists of water, lactose (4.5– 5.5%, w/v), proteins (a-lactalbumins, b-lactoglobulins, bovine serum albumin, immunoglobulins), vitamins and mineral salts (Roukas and Lazarides, 1991). Very little published information is available on the production of β-carotene from molasses and cheese whey by B. trispora (Lampila et al., 1985b; Goksungur et al., 2002; Roukas et al., 2003; Choudhari and Singhal, 2008). In these studies whey and molasses were found to be poor substrates with respect to both cell growth as well as β-carotene production. The main disadvantage of molasses and whey is the sucrose and lactose content, respectively.

Nitrogen Sources It has been shown that carotenoid production by B. trispora is influenced by the type and the concentration of the nitrogen source in the culture medium. Carotenoid production has been found to be negatively affected by nitrogenous compounds derived from animal sources (e.g. beef extract, skimmed milk) and inorganic nitrogen sources (e.g. ammonium salts) (Ciegler et al., 1959b; Mantzouridou et al., 2002). On the other hand, the use of complex nitrogen sources such as corn-steep liquor, soybean meal, cotton-seed meal, distiller’s solubles and yeast extract has been common practice for carotenoid production by B. trispora.

Carbon-to-Nitrogen Ratio Carotenoid production requires a relatively high carbon-to-nitrogen ratio (C/N). Indeed, Mantzouridou and collaborators (2006) pointed out the importance of low nitrogen content in

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids... 111 the presence of excess organic carbon substrates in hyper-production of carotenoids by B. trispora. The results showed that providing sufficient carbon and nitrogen within the early growth phase maximal fungal growth is ensured, whereas carotenoid synthesis is slow; a sharp increasing trend of carotenoid synthesis is evidenced within the second half of the fermentation cycle (4-8 days) (late growth phase), when fungal metabolism lowers, and nitrogen content is exhausted. During this phase, there was a suggestion of a relationship between carotenoid synthesis and the amount of carbon available. The explanation was that when growth is complete there is an increase of carbon flux towards carotenoid synthesis. Obviously, in chemostat the culture medium can be designed so that non-limiting carbon conditions to be ensured. Other pre-determined factors can increase carotenogenesis in B. trispora such as α- and β-ionones and some natural terpenoids that do not affect growth, but directly induce key enzyme activities (Lampila et al., 1985a), and various antioxidants that act against the oxidation of carotenoids during fermentation (Ciegler et al., 1961; Mantzouridou et al., 2002).

Performance of Pure Glycerol as a Carbon Source for Fungal Growth and Carotenoid Production In general, several alternative routes are involved with glycerol metabolism in fungi (Deacon, 1997): a) it can be utilized as a sole carbon source under aerobic conditions, b) it is the vital structural component of the major classes of biological lipids, and c) it is an important intermediate in TAG catabolism. When fungi are grown on glycerol as a sole carbon source, after uptake, glycerol can be fed into the glucolytic pathway to yield pyruvate. Subsequently, oxidative decarboxylation of pyruvate generates acetyl-CoA, the starting product for many metabolic routes. Different pathways involved in the initial metabolic steps of glycerol dissimilation are known. The main ones are the phosphorylative and the oxidative pathway. In the first, ATP is used to phosphorylate glycerol to glycerol 3-phosphate (3-P-glycerol) by glycerol kinase; 3-Pglycerol is then oxidised to dihydroxyacetone phosphate (3-P-dihydroxyacetone), an intermediate of glycolysis, by glycerol 3-phosphate dehydrogenase. In the second pathway, glycerol is converted to dihydroxyacetone by NAD+ or NADP+-linked glycerol dehydrogenase, and then phosphorylated to 3-P-dihydroxyacetone by dihydroxyacetone kinase. In general, the phosphorylative pathway is found in eukaryotes. However, in N. crassa both the phosphorylative and oxidative pathways may be operational (Courtright, 1975a,c; North, 1973; 1974), whereas in Schizosaccharomyces pombe and Candida valida H122 only the oxidative pathway may be operational (May and Sloan, 1981; Babel and Hofmann, 1982; May et al., 1982). Several other minor pathways for glycerol utilization have been proposed such as that involving NADP+-linked GDH and Dglyceraldehyde kinase or glycerate kinase (Tom et al., 1978). Considering that growth of fungi necessitates a large amount of sugars for the synthesis of complex structural polysaccharides, in the case of a fungus growing on glycerol as a sole carbon source the process of gluconeogenesis allows generation of glucose from glycerol.

112

F. Mantzouridou

Using Pure Glycerol for Growth and Carotenoid Production by Blakeslea Trispora In B. trispora system, the enzymatic activities involved with glycerol metabolism have not been characterized and literature on fungal cultivation on glycerol is rather limited. In this section, to describe available data, experimental approaches are categorized as follows: Glycerol as a Sole Carbon Source The use of glycerol as a sole carbon substrate by B. trispora is reported for the first time in the production of β-carotene in the paper of Dandekar and Modi (1980). It was observed that if well-formed mycelia of opposite mating types of B. trispora were transferred to media containing glycerol, mated cultures produced much higher yield of trisporic acids and βcarotene than if cultures mated in media containing glucose. Experimentation within the range 0.1-0.4 M level of addition of glucose/glycerol revealed the concentration dependent phenomenon of trisporic acid synthesis by glucose, whereas in glycerol supplemented cultures the amount of trisporic acids was not affected by substrate concentration. Also, the difference in β-carotene level between mated cultures grown on glucose and glycerol increased with an increase in the amount of glucose and glycerol. In particular, mated culture grown in medium supplemented with 0.3 M glucose contained about 61% less β-carotene compared with the one in medium supplemented with 0.3 M glycerol. Mechanistically, the inhibitory effect of glucose on carotenogenesis during mating was attributed to catabolite inactivation of mevalonate kinase, a key early enzyme in isoprenoid synthesis. It has been established that trisporic acids play a central role in regulating carotenogenesis in B. trispora system by derepression of this enzyme synthesis (Desai and Modi, 1977). Dandekar and Modi (1980) also reported that the action of trisporic acid synthesis is also under glucose catabolite repression. Thus, the reason for higher β-carotene production from glycerol can be the overcoming of catabolite repression effects by growing the cells on a poorly utilizable carbon source like glycerol. Although the substitution of glucose by glycerol resulted in 1.6 times higher yield of β-carotene (1.0 mg/g of dry biomass), the latter cannot be considered competitive to others reported in literature for practical applications. Twenty years later, within the frame of studies for media optimization for the production of β-carotene by B. trispora, different researchers tested the glycerol as a sole carbon substrate at levels between 30-70 g/L of culture medium (Mantzouridou et al., 2002; Choudhari and Singhal, 2008). In contrast to the results obtained by Dandekar and Modi (1980), it was not possible to increase β-carotene yield by glycerol. Failure of B. trispora to produce β-carotene on glycerol was primarily due to poor growth of the fungus on this carbon source. The different results might be due to the fact that in the work of Dandekar and Modi (1980) well-formed mycelia precultured on glucose were used to inoculate the main culture medium containing glycerol, whereas in the latest studies experiments were conducted without preculture treatment. It is noted that N. crassa exhibits a form of conditional acetate auxotrophy during growth on minimal glycerol medium due to the insufficient amounts of acetyl-CoA resulting from a decreased synthesis of mitochondrial pyruvate dehydrogenase complex (PDC) (Courtright, 1975a). In addition, Papanikolaou and collaborators (2008) have considered for the case of the oleaginous Mucorales Mortierella isabellina that the reason for

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids... 113 the lower production of biomass on glycerol compared with growth on glucose can be poor regulation of the enzymes GK and 3-P-GDH, or decreased activity of gluconeogenic enzymes. One or more of the above aspects might be also attributed to poor growth of B. trispora on glycerol rather than the absence of the enzymes necessary for glycerol dissimilation. In view of these, further research is needed to elucidate both the glycerol metabolic pathway and the relevant regulatory mechanism in B. trispora cells. Glycerol as co-Substrate of Glucose To avoid poor growth of B. trispora on media containing glycerol as a sole carbon source, it has been recently proposed the simultaneous utilization of glycerol and glucose for β-carotene production by this fungus (Mantzouridou et al., 2008). The rationale behind this medium design was that glucose is appropriate for optimal growth of the fungus in the first stage (early growth phase) of the bioprocess, and in the second stage (late growth phase) glycerol is appropriate for optimal production of carotenoids. This approach was found to be successful in the case of N. crassa; when cells were grown in media containing glycerol plus acetate no reduction in PDC activity was found coinciding with sufficient amount of mycelia growth (Courtricht, 1975a). In the work of Mantzouridou and collaborators (2008), the kinetic behavior of B. trispora (cell growth, substrate assimilation, product formation) was studied in media containing pure glycerol at different levels (10-180 g/L of culture medium) and constant glucose and total nitrogen content (50 and 1.2 g/L of culture medium, respectively). The obtained results are summarised in Table 2. Table 2. Values of the fermentation parameters measured at the time of maximum βcarotene production by B. trispora grown in control medium in the presence/absence of different initial levels of pure glycerol

Medium formulation

Control medium1

Addition level of glycerol (g/L of culture medium) -

Total Biomass Dry Weight (g/L of culture medium)

Maximum volumetric productivity

β-Carotene

Cellular lipids

(% of total carotenoids)2

(mg/Ld)

16

(mg/g of biomass dry weight) 1.5

35

2.0

(g/g of biomass dry weight) 0.095

10

18

2

64

3.0

0.11

30 60 90 180

25 33 41 60

5 15 7 3

66 82 82 85

10.0 41.0 24.0 15.0

0.12 0.26 0.28 0.30

+ Glycerol

Mantzouridou et al. 2008. 1 Control medium contains glucose at 50 g/L of culture medium as a sole carbon source; 2total carotenoids produced: lycopene, γ-carotene, β-carotene.

114

F. Mantzouridou

The experimental results showed satisfactory growth in all cases. Growth and lipogenesis dependence on the level of glycerol addition was clear. Maximum value of total biomass (lipid-free material plus cellular lipids) ranged from 20.0 g/L to 60.0 g/L on 10.0 and 180.0 g glycerol/L of culture medium, respectively. The highest quantity of total lipid presented the value of 0.30 g/g of dry biomass (or 18 g/L of culture medium), which was satisfactory, given that glycerol is not one of the most favourable substrates for lipid production by various oleaginous Zygomyces (e.g. Sajbidor et al., 1988; Chen and Chang, 1996). Additionally, the above value is comparable with that occurred from glucose supplemented with substrates favouring production of microbial lipid by this fungus (e.g. vegetable oils) (up to 0.60 g/g of dry biomass or 15 g/L of culture medium) (Mantzouridou et al., 2006). At relatively low glycerol concentration (10 and 30 g/L of culture medium) carbon limitation was observed during the late growth phase (i.e. 60-288 h). However, in culture containing the highest level of glycerol employed (i.e. 180.0 g glycerol/L of culture medium) noticeable substrate inhibition was exerted following by a significantly long adaptation period for growth among with large substrate amounts remained unconsumed in the medium after long incubation periods. In these batch experiments, the strain-dependent specificity on glucose and glycerol was verified by the sequential order of carbon substrate uptake; glucose was primarily assimilated and when around 40.0% of initial glucose content was consumed glycerol metabolism occurred. Theoretically, for useful application on carotenoid production, co-substrates of glucose should influence cellular metabolism so that carbon flow to be channelled to carotenoid biosynthetic route. The findings of Mantzouridou and collaborators (2008) indicated that glycerol in the presence of glucose critically controls carotenoid biosynthesis in B. trispora cells. Maximum β-carotene production and productivity achieved in batch experiments of this study (15.0 mg/g and 41 mg/Ld, respectively, at initial glycerol concentration of 60.0 g/L) is indeed noteworthy in particular when compared with yields reported from batch cultures growing on glucose as a sole carbon source or on other inexpensive hydrophilic carbon substrates (3-8 mg/g of dry biomass) (e.g. Ciegler et al., 1959b; 1963b; Dholakia and Modi, 1982; Lampila et al., 1985b; Roukas et al., 2003). Under the above fermentation conditions the selectivity of the bioprocess was also found to be improved (β-carotene > 80% of total carotenoids).

Using Pure Glycerol for Growth and Carotenoid Production by other Fungi Glycerol was found ineffective as carbon substrate for carotenogenesis in P. blakesleeanus (Garton et al., 1951), merely because the fungus could not grow on media containing glycerol as a sole carbon source, and for photocarotenogenesis in M. rouxii (Mosqueda-Cano and Gutiérrez-Corona, 1995). In both cultures, glucose caused the highest carotene content (Garton et al., 1951; Mosqueda-Cano and Gutiérrez-Corona, 1995). On the other hand, in the case of Rhodotorula strains glycerol has been shown to be the most effective carbon source for β-carotene production. Results from experimental trials using glycerol as a sole carbon and energy source for growing R. lactosa showed that the highest biomass yield achieved (0.53 g/g substrate) was obtained at 21.5 g glycerol/L of culture

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids... 115 medium after 20 h (Martelli et al., 1992). At this point glycerol was exhausted. After 48 h, βcarotene production was 1.8 mg/g of biomass dry weight (or 22 mg/L of culture medium). The work of Martelli and collaborators (1992) also indicated that more β-carotene (2.66 mg/g of biomass dry weight or 28 mg/L of culture medium) could be produced if cells from the 20 h culture were suspended in distilled water. The above findings imply that no external substrate is required for β-carotene biosynthesis. At higher initial glycerol level (i.e. 28 g/L of culture medium) growth inhibition along with changes in cell morphology were observed (Martelli et al., 1992). Authors attributed the inhibitory effect of glycerol to plasmolysis of the cell membrane. Glycerol was also found a potential substrate for astaxanthin production by P. rhodozyma in rich medium containing yeast extract and peptone (Kusdiyantini et al., 1998). The results obtained from the work of Kusdiyantini and collaborators (1998) revealed that maximum specific growth rate (μmax value around 0.24 h-1) was independent of the initial level of glycerol employed in the culture medium. Kinetic data on astaxanthin production using different initial glycerol levels showed a maximum specific astaxanthin production rate (vAst value of 0.09 mg/g h) occurring for optimal μ value of 0.075 h-1 (Kusdiyantini et al., 1998). However, with these observations alone authors were not able to conclude whether the reason of maximum vAst is a critical μ value or is a critical cell concentration corresponding to the appearance of certain physiological conditions that is primarily limitation of glycerol. Importantly, Kusdiyantini and collaborators (1998) also observed that astaxanthin percentage in total pigment content and its yield from glycerol were not affected by the initial substrate level (78 % and 0.97 mg/g, respectively). For optimal glycerol level of 37.8 g/L of culture medium, astaxanthin content reached 1.8 mg/g of biomass dry weight (or 33.7 mg/L of culture medium) after a fermentation period of 168 h (Kusdiyantini et al., 1998). To improve yield of the bioprocess authors suggested a strategy of astaxanthin production in fed batch culture or chemostat.

Potential of Raw Glycerol in Fungal Carotenoid Production Although the utilization of raw glycerol as a fermentation feedstock has an economic advantage against the use of pure glycerol, in the majority of studies concerning fungal carotenoid production the pure form has been used. In fact, only in recent works aiming at future industrial applications experimentation involved raw glycerol. This trend is illustrated in the works of Razavi and collaborators (Razavi and Marc, 2006; Razavi et al., 2007) and Mantzouridou et al. (2008). The first group systematically examined the production of carotenoids by S. ruberrimus using technical glycerol in a batch process performed in a 3-L fermentor. When investigating the effects of initial glycerol level of addition, ammonium sulphate, initial pH, temperature, it was demonstrated that S. ruberrimus is capable of high carotenoid production from an inexpensive culture medium containing technical glycerol. The results showed that although glycerol had an important effect on the biomass and lipid content, it did not affect the massive accumulation of carotenoids in cells (as mg/g of biomass dry weight). However, the volumetric carotenoid production increased almost 10 times (70

116

F. Mantzouridou

mg/L of culture medium) by exposing S. ruberrimus cells to the highest glycerol level (75.8 g/L of culture medium). This was due to the significant increase of biomass to a maximum of 99.5 mg/L of culture medium. It was also found that glycerol level of addition did not affect the proportion of individual carotenoids. Thus, in the maximum total carotenoid yield of 3.84 mg/g, torularhodin (3.70 mg/g) and β-carotene (0.14 mg/g) was included. In the work of Mantzouridou and collaborators (2008) the dynamics of two industrial types of glycerol, from biodiesel production and soap manufacture, as a supplementary carbon source to glucose for β-carotene production by B. trispora was investigated in shake flask cultures. B. trispora could not grow in the presence of raw glycerol from a biodiesel production unit. Owing to the fact that raw glycerol consisted of methanol and soaps as the major impurities, an efficient pre-treatment protocol was applied to reduce the levels of these potent growth inhibitors (Figure 4). Briefly, the mixture was first treated with hydrochloric acid to split contaminating soaps into free acids and salt (i.e. sodium chloride). Removal of the lipid phase was achieved using a separation funnel after centrifugation. Methanol was then recovered by vacuum evaporation. At that point the glycerol content was 40% (w/w) whereas the amount of methanol left was less than 2%, w/w.

Mantzouridou et al., 2008. Figure 4. Flow diagram for the pretreatment of raw glycerol from a biodiesel production unit.

After pre-treatment, other nutrients (glucose, nitrogen source, mineral salts etc.) with additional water were added to adjust the nutrient level (including glycerol) to a desired level. Experimental results were compared with those obtained from control experiments conducted using glucose as a sole carbon source and glucose plus pure glycerol as a supplementary carbon source (Table 3). At glycerol level of 60.0 g/L of culture medium, a considerable delay (18 h) of growth stimulation was observed. At the end of the fermentation process, lower amounts of total biomass (20 g/L of culture medium) and cellular lipids (0.15 g/g of dry biomass) were detected when compared with those on pure glycerol (31 g/L and 0.26 g/g, respectively). This was partially ascribed to the inability of the fungus to tolerate high content

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids... 117 of sodium chloride (see Figure 4). However, a complete consumption of glycerol was achieved. The results also showed satisfactory β-carotene yield and high specificity (8.0 mg/g of dry biomass and 89 % of total carotenoid content, respectively). Since these values were much higher than that obtained using only glucose (1.0 mg/g of dry biomass and 35% of total carotenoid content), it was clearly suggested that more carbon was directed towards the synthesis of carotenoids. The kinetic study also revealed a reduction in β-carotene production and maximum volumetric productivity on raw glycerol as compared to pure glycerol (8.0 mg/g and 11.4 mg/Ld versus 15.0 mg/g and 38.8 mg/Ld, respectively). This was explained by the inhibitory effect of sodium chloride on carotenogenesis (Govind et al., 1982), along with the increase in the lag phase that resulted in a longer fermentation time. Non purified raw glycerol solution from a soap manufacturing unit did not inhibit B. trispora growth. The growth characteristics of the culture with glycerol from soap manufacturing unit in the feed presented similarities with those of the culture with pretreated glycerol from biodiesel production unit (Table 3). Because sodium chloride is used to facilitate separation of soap from glycerol by-product, the latter is mixed with high quantities of this salt. Thus, salt stress was considered as one of the causal factors for the lag phase observed. β-Carotene accumulated at 10 mg/g of dry biomass with maximum volumetric productivity of 20.8 mg/Ld and constituted 88.0% of total carotenoid content; these values were almost 11, 20 and 2.5 folds higher than those observed in control medium containing glucose as a sole carbon source. On the other hand, β-carotene production and maximum volumetric productivity were lower than that observed using pure glycerol (15.0 mg/g and 38.8 mg/Ld, respectively). The negative response was also attributed to the inhibitory effect of sodium salt on carotenogenesis (Govind et al., 1982). Still, the relative composition of βcarotene in both cultures was similar within the last stage of cell growth (88 and 82%, respectively). Authors stressed on the need to test different raw glycerol samples (e.g. from various biodiesel plants or physical refining process), containing various impurities due to the production methods, for their suitability to produce carotenoids by B. trispora. Moreover, they pointed out the necessity to employ trials under fed-batch or continuous operations in an effort to increase the productivity of the fermentation process, thereby, decreasing the running costs. From an economic point of view, it is important to point out the growing world carotenoid market that is expected to reach over $1 billion by 2010 as consumers continue to look for natural ingredients (Martín et al., 2008). Natural β-carotene accounts for 15 to 20% of world demand. On the other hand, the cost of the various components of the culture medium, especially the organic-carbon source, can have a profound effect on the overall cost of a microbial process. In view of these, although both methanol and soap were found to negatively influence B. trispora growth and carotenoid production, and these impurities should be removed from biodiesel by-product, the significantly low raw glycerol cost and the high carotenoid price strengthen the financial viability of the proposed bioprocess. Moreover, process economics can include methanol recovery and its recycling into the transesterification operation, whereas the fatty acid waste derived from soap treatment can have a market value in some industrial plants. However, yield optimization and an economic optimum scale-up needs future investigations.

F. Mantzouridou

118

Table 3. Values of the fermentation parameters measured at the time of maximum βcarotene production by B. trispora grown in control medium in the presence/absence of different types of glycerol

Medium formulation

Control medium1

Type of glycerol

Total biomass dry weight (g/L of culture medium)

-

15

+ Pure glycerol2 + Biodiesel byproduct2 + Soap byproduct2

31

β-Carotene (mg/g of biomass dry weight) 1.0

(% of total carotenoids)3

Maximum volumetric productivity

(mg/Ld)

Cellular lipids (g/g of biomass dry weight)

35

1.5

0.10

15.0

82

38.8

0.26

20

8.0

89

11.4

0.15

25

10.0

88

20.8

0.16

Mantzouridou et al. 2008. 1 Control medium contains glucose at 50 g/L of culture medium as a sole carbon source; 2 initial addition level of glycerol 60 g/L of culture medium; 3 total carotenoids produced: lycopene, γ-carotene, βcarotene.

Safety Aspects In the last part of this chapter toxicological aspects of the raw glycerol employed are stressed with regard to chemical hazards originated from biodiesel glycerol production. The chemical composition of raw glycerol varies between different biodiesel plants, due to different oil feedstocks. In general, the by-product contains impurities like methanol, soap, heavy metals and water (Johnson and Taconi, 2007). Reference to the potential hazard from heavy metals is made recently by Pyle et al. (2008) in their study of producing docosahexanoic acid from raw glycerol by the microalga Schizochytrium limacinum. In view of this, the chemical characterization of raw glycerol used as a fermentation feedstock is an important issue. Thompson and He (2006) in a recent research study investigated the elemental composition of raw glycerol derived from different oil feedstocks (e.g. mustard seeds, canola, soybean, and waste vegetable oil). In most cases, calcium, magnesium, phosphorous and sulphur were present in the range of 5-60 ppm. Only in the case of crambe utilization as feedstock, raw glycerol contained greater amounts of the above elements. This information is a good indication for the possible reasons of selecting oil feedstock. Some safety precautions appear also in the work of Pyle et al. (2008), who investigated the “quality” of algal biomass in terms of its elemental composition and nutritional levels. These investigators found that raw glycerol-derived algal biomass had a high level of docosahexanoic acid and a nutritional value similar to that of commercial algal biomass. Taking into account that both the B. trispora biomass and the final crystalline product should comply with an adequate chemical specification (SCF, 2000), the above work may be the basis for further investigation of chemical composition of raw glycerol-derived fungal

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids... 119 biomass and the final product. A lot of work is needed towards successful exploitation of raw glycerol and high carotenoid production with regards to economics and safety.

Conclusion Raw glycerol has the potential to become an important feedstock for the production of high value-added carotenoids by fungi. However, there is a need to improve fermentation strategies so that carotenoid yield and productivity is feasible on an industrial scale. Taking into account current legislative demands on food safety precautions, another important aspect is the toxicological evaluation of raw glycerol employed as well as the investigation of practices that may reduce risk from potential hazards from this source.

Acknowledgment Dr. F. Mantzouridou gratefully acknowledges Associate Professor Dr. M. Z. Tsimidou (Laboratory of Food Chemistry and Technology, School of Chemistry, Aristotle University of Thessaloniki) for her support and her thoughtful and careful editing and formatting work.

Reference List Anderson, RF; Arnold, M; Nelson, GEN; Ciegler, A. Microbiological production of betacarotene in shaken flasks. Journal of Agricultural and Food Chemistry, 1958, 6, 543545. Bekhtereva, MN; Dedyukhina, EG. Oleic acid as a carbon source for the growth of Blakeslea trispora and for biosynthesis of carotenoids and lipids. Mikrobiologiia, 1970, 39, 77-81. Bhosale, P. Environmental and cultural stimulants in the production of carotenoids from microorganisms. Applied Microbiology and Biotechnology, 2004, 63, 351-361. Bhosale, P; Larson, AJ; Bernstein, PS. Factorial analysis of tricarboxylic acid cycle intermediates for optimization of zeaxanthin production from Flavobacterium multivorum. Journal of Applied Microbiology, 2004, 96, 623–629. Chen, HC; Chang, CC. Production of γ-linolenic acid by the fungus Cunninghamella echinulata CCRC 31840. Biotechnology Progress, 1996, 12, 338–41. Choudhari, S; Singhal, R. Media optimization for the production of β-carotene by Blakeslea trispora: A statistical approach. Bioresource Technology, 2008, 99, 722–730. Ciegler, A. Microbial carotenogenesis. Advances in Applied Microbiology, 1965, 7, 1-34. Ciegler, A., Arnold, M., Anderson, R.F. Effect of lipids and related substances on production of beta-carotene. Journal of Agricultural and Food Chemistry, 1959a, 7, 98-101. Ciegler, A; Arnold, M; Anderson, RF. Effect of various grains on production of beta-carotene by mated strains of Blakeslea trispora. Journal of Agricultural and Food Chemistry, 1959b, 7, 94-98.

120

F. Mantzouridou

Ciegler, A; Lagoda, AA; Sohns, VE; Hall, HH; Jackson, RW. Beta-carotene production in 20-liter fermentors. Biotechnology and Bioengineering, 1963a, 5, 109-121. Ciegler, A; Nelson, GEN; Hall, HH. Enhancement of β-carotene synthesis by citrus products. Applied Microbiology, 1963b, 11, 128-131. Ciegler, A; Nelson, NE; Hall, HH. Influence of hydrocarbon on carotenogenesis by mated cultures of Blakelea trispora. Applied Microbiology, 1962, 10, 132-139. Ciegler, A; Nelson, NE; Hall, HH. Microbiological production of carotenoids. Stabilization of β-carotene in dried fermentation solids. Journal of Agricultural and Food Chemistry, 1961, 9, 447-451. Commission directive (EC) No 50/2001 of 3 July 2001 amending Directive 95/45/EC laying down specific purity criteria concerning colours for use in foodstuffs. Official Journal of the European Union, L190/14 of 12.7.2001. Commission Directive 95/45/EC of 26 July 1995 laying down specific purity criteria concerning colours for use in foodstuffs. Official Journal of the European Communities, L226 of 22.09.95. Commission regulation (EC) No 1288/2004 of 14 July 2004 concerning the permanent authorisation of certain additives and the provisional authorisation of a new use of an additive already authorised in feedingstuffs. Official Journal of the European Union, L243/10 of 15.7.2004. Commission regulation (EC) No 721/2006 of 23 October 2006 authorizing the placing on the market of lycopene from Blakeslea trispora as a novel food ingredient under Regulation (EC) No 258/97 of the European Parliament and of the Council (notified under document number C(2006) 4973). Official Journal of the European Union, L296 of 26.10.2006. Courtright, JB. Characteristics of a glycerol utilization mutant of Neurospora crassa. Journal of Bacteriology, 1975a, 124, 497-502. Courtright, JB. Differential rates of synthesis of glycerokinase and glycerophosphate dehydrogenase in Neurospora crassa during induction. Archives of Biochemistry and Biophysics, 1975b, 167, 34-44. Courtrlght, JB. Intracellular localization and properties of glycerokinase and glycerophosphate dehydrogenase in Neurospora crassa. Archives of Biochemistry and Biophysics, 1975c, 167:21-33. Dandekar, S; Modi, VV. Involvment of cyclic AMP in carotenogenesis and cell diffeentiation in Blakeslea trispora. Biochimica et Biophysica Acta, 1980, 628, 398-406. Deacon, J.W. Modern Mycology. 3rd ed. Oxford: Blackwell Science, Ltd., 1997. Desai, HG; Modi, VV. Stimulation of carotenogenesis by penicillin in Blakeslea trispora. Phytochemistry, 1977, 16, 1373-1376. Dholakia, JN; Modi, VV. Fermentative production of β-carotene and extracellular βglucosidase by Blakeslea trispora grown on cellobiose. European Journal of Applied Microbiology and Biotechnology, 1982, 15, 33-35. Dufossé, L. Microbial production of food grade pigments. Food Technology and Biotechnology, 2006, 44, 313–321.

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids... 121 Dufossé, L; Galaupa, P; Yaronb, A; Aradb, SM; Blancc, P; Murthyd, KNC; Ravishankard, GA. Microorganisms and microalgae as sources of pigments for food use: a scientific oddity or an industrial reality? Trends in Food Science and Technology, 2005, 16, 389– 406. European Commision. 2000. Opinion of the scientific committee on food on β-carotene from Blakeslea trispora, SFC/CS/ADD/COL 158. Feofilova, EP; Tereshina, VM; Memorskaya, AS. Heterothallism in mucorous fungi: Physiological and biochemical characterization of the (+) and (-) strains of Blakeslea trispora. Microbiology, 1997, 66, 701-705. Feofilova, EP; Tereshina, VM; Memorskaya, AS; Dul’kin, LM; Goncharov, NG. Fungal lycopene: the biotechnology of its production and prospects for its application in medicine. Microbiology, 2006, 75, 629-633. Flegal, CJ; Schlaible, PJ; Hall, HH. The relative biopotency of fermentation of beta-carotene, crystalline beta-carotene and vitamin A for poultry. Poultry Science, 1971, 50, 349-358. Gartor, GA; Goodwin, TW; Lijinsky, W. Studies in carotenogenesis. 1. General conditions governing β-carotene synthesis by the fungus Phycomyces blakesleeanus Burgeff. Biochemical Journal, 1951, 48, 154-163. Goksungur, Y; Mantzouridou, F; Roukas, T. Optimization of the production of β-carotene from molasses by Blakeslea trispora: a statistical approach. Journal of Chemical Technology and Biotechnology, 2002, 77, 933-943. Goodwin, TW. Fungal Carotenoids. The Botanic Review, 1952, 18, 291-316. Goodwin, TW; Willmer, JS. Studies in carotenogenesis. 4. Nitrogen metabolism and carotene synthesis in Phycomyces blakesleeanus. Biochemical Journal, 1952, 51, 213-217. Govind, NS; Amin, AR; Modi, VV. Stimulation of carotenogenesis in Blakeslea trispora by cupric Hinds, TS; West, WL; Knight, EM. Carotenoids and retinoids: a review of research, clinical and public health applications. Journal of Clinical Pharmacology, 1997, 37, 551558.ions. Phytochemistry, 1982, 21, 1043-1044. Hsu, WJ; Yokoyama, H; Coggins, CW. Carotenoid biosynthesis in Blakeslea trispora. Phytochemistry, 1972, 11, 2985-2990. Iigusa, H; Yoshida, Y; Hasunuma, K. Oxygen and hydrogen peroxide enhance light-induced carotenoid synthesis in Neurospora crassa. FEBS Letters, 2005, 579, 4012–4016. Jacobson, GK; Jolly, SO; Sedmak, JJ; Skatrud, TJ; Wasileski, JM. Astaxanthin overproducing strains of Phaffia rhodozyma, methods for their cultivation, and their use in animal feeds. US Patent 6,015,684, 2000. JECFA. 2001. Joint FAO/WHO Expert Committee on Food Additive. 57th meeting Rome, 514 June. Jin, E; Feth, B; Melis, A. A mutant of the green alga Dunaliella salina constitutively accumulates zeaxanthin under all growth conditions. Biotechnology and Bioengineering, 2003, 8, 115–124. Johnson, D; Taconi, KA. The glycerin glut: Options for the value-added conversion of crude glycerol resulting from biodiesel production. Environmental Progress, 2007, 26, 338348.

122

F. Mantzouridou

Kritsky, MS; Sokolovsky, VY; Belozerskaya, TA; Chernysheva, EK. Relationship between cyclic AMP level and accumulation of carotenoid pigments in Neurospora crassa. Archives of Microbiology, 1982, 133, 206-208. Kusdiyantini, E; Gaudin, P; Goma G; Blanc, PJ. Growth kinetics and astaxanthin production of Phaffia rhodozyma on glycerol as a carbon source during batch fermentation. Biotechnology Letters, 1998, 20, 929–934. Ladygin, VG. Biosynthesis of carotenoids in the chloroplasts of algae and higher plants. Russian Journal of Plant Physiology, 2000, 47, 796–814. Lampila, LE; Wallen, SE; Bullerman, LB. A review of factors affecting biosynthesis of carotenoids by the order Mucorales. Mycopathologia, 1985a, 90, 65-80. Lampila, LE; Wallen, SE; Bullerman, LB; Lowry, SR. The effect of strain and type of whey on the production of β-carotene and other parameters. Lebensmittel Wissenshaft and Technology, 1985b, 18, 366-369. Leathers, TD; Jones, JD; Hohn, TM. System for the sequential, directional cloning of multiple DNA sequences. US patent 6,696,282, 2004. Lopez-Nieto, MJ; Costa, J; Peiro, E; Mendez, E; Rodriguez-Saiz, M; De la Fuente, JL; Cabri, W; Barredo, JL. Biotechnological lycopene production by mated fermentation of Blakeslea trispora. Applied Microbiology and Biotechnology, 2004, 66, 153-159. Lorenz, RT; Cysewski, GR. Commercial potential for Haematococcus microalgae as a natural source of astaxanthin. Trends in Biotechnology, 2000, 18, 160–167. Mantzouridou F; Tsimidou, MZ. Carotenoid pattern in Blakeslea trispora grown on oilenriched substrates with regard to triacylglycerol species accumulation. European Journal of Lipid Science and Technology, 2007, 109, 3-10. Mantzouridou, F; Naziri, E; Tsimidou, MZ. Industrial glycerol as a supplementary carbon source in the production of β-carotene by Blakeslea trispora. Journal of Agricultural and Food Chemistry, 2008, 56, 2668–2675. Mantzouridou, F; Roukas, T; Kotzekidou, P; Liakopoulou, M. Optimization of β-carotene production from synthetic medium by Blakeslea trispora: A mathematical modeling. Applied Biochemistry and Biotechnoogy, 2002, 101, 153-175. Mantzouridou, F; Tsimidou, MZ. Lycopene formation in Blakeslea trispora. Chemical aspects of a bioprocess. Trends in Food Science and Technology, 2008, 19, 363-371. Mantzouridou, F; Tsimidou, MZ; Roukas, T. Performance of crude olive pomace oil and soybean oil during carotenoid production by Blakeslea trispora in submerged fermentation. Journal of Agricultural and Food Chemistry, 2006, 54, 2575-2580. Martelli, H.L., Da Silva, SNO; Pomeroy, D. Glycerol as substrate for biomass and β-carotene production by Rhodotorula lactose. World Journal of Microbiology and Biotechnology, 1992, 8, 635-637. Martin, JF; Gudiña, E; Barredo, JJ. Conversion of β-carotene into astaxanthin: Two separate enzymes or a bifunctional hydroxylase-ketolase protein? Microbial Cell Factories, 2008, 7:3 (doi:10.1186/1475-2859-7-3). May, JW; Marshall, JH; Sloan, J. Glycerol utilization by Schizosaccharomyces pombe: phosphorylation of dihydroxyacetone by a specific kinase as the second step. Journal of General Microbiology, 1982, 128, 1763-1766.

The Potential of Raw Glycerol in the Production of Food Grade Carotenoids... 123 May, JW; Sloan, J. Glycerol utilization by Schizosaccharomyces pombe: dehydrogenation as the initial step. Journal of General Microbiology, 1981, 123, 183-185 Mehta, BJ; Cerdá-Olmedo, E. Mutants of carotene production in Blakeslea trispora. Applied Microbiology and Biotechnology, 1995, 42, 836-838. Miller, TL; Churchill, BW, Substrates for large scale fermentations. In: Mannual of Industrial Microbiology, pp. 122-136. (Editors Demain, AL; Solomons, NA.) Washington: American Society for Microbiology, 1986. Misawa, N; Shimada, H. Metabolic engineering for the production of carotenoids in noncarotenogenic bacteria and yeasts. Journal of Biotechnology, 1998, 59, 169-181. Mosqueda-Cano, G; Gutierrez-Corona, F. Environmental and developmental regulation of carotenogenesis in the dimorphic fungus Mucor rouxii. Current Microbiology, 1995, 31, 141-145. Murillo, F; Cerdá-Olmedo, C. Regulation of carotene synthesis in Phycomyces. Molecular and General Genetics, 1976, 148, 19-24. Murillo, FJ; Calderon, IL; Lopez-Diaz, I; Cerdá-Olmedo, E. Carotene-superproducing strains of Phycomyces. Applied and Environmental Microbiology, 1978, 36, 639–642. Nelis, HJ; De Leenheer, AP. Microbial sources of carotenoid pigments used in foods and feeds. Journal of Applied Bacteriology, 1991, 70, 181-191. Nelis, HJ; De Leenheer, AP. Reinvestigation of Brevibacterium sp. strain KY-4313 as a source of canthaxanthin. Applied and Environmental Microbiology, 1989, 55, 2505– 2510. North, MJ. Cold-induced increase of glycerol kinase activity in Neurospora crassa: rapid inactivation of the enzyme in vivo. Journal of Bacteriology, 1974, 120, 741-747. North, MJ. Cold-induced increase of glycerol kinase in Neurospora crassa. FEBS Letters, 1973, 35, 67-70. Papanikolaou, S; Fakas, S; Fick, M; Chevalot, I; Galiotou-Panayotou, M; Komaitis, M; Marc, I; Aggelis, G. Biotechnological valorisation of raw glycerol discharged after bio-diesel (fatty acid methyl esters) manufacturing process: Production of 1,3-propanediol, citric acid and single cell oil. Biomass and Bioenergy, 2008, 32, 60-71. Pazola, Z; Niewiadomski, H; Salmonowicz, J; Switek, H; Michnikowska, W. Effect of various lipids on the production of carotenoids by mated strains of Blakeslea trispora and on the chemical composition of mycelium fat. Acta Microbiologica Polonica, 1968, 17, 75-81. Pazola, Z; Switek, H; Janicki, J; Michnikowska, W. Microbiological production of carotene using Blakeslea trispora species. II. Use of new lipid sources. Pr. Inst. Lab. Badaw. Przem. Spozyw., 1967, 17, 267-286. Purcell, AE; Walter, WM. Microbial production of carbon 14C-labelled all-trans betacarotene. U.S. Patent 3,579,424, 1971. Pyle, DJ; Garcia, RA; Wen, Z. Producing docosahexanoic acid (DHA)-rich algae from biodiesel-derived crude glycerol: Effects of impurities on DHA production and algal biomass composition. Journal of Agricultural and Food Industry, 2008, 56, 3933-3939. Razavi, SH; Blanchard, F; Marc, I. UV-HPLC/APCI-MS method for separation and identification of the carotenoids produced by Sporobolomyces ruberrimus H110. Iranian Journal of Chemistry and Chemical Engineering, 2006a, 25, 1-10.

124

F. Mantzouridou

Razavi, SH; Marc, I. Effect of temperature and pH on the growth kinetics and carotenoid production by Sporobolomyces ruberrimus H110 using technical glycerol as carbon source. Iranian Journal of Chemistry and Chemical Engineering, 2006b, 25 59-64. Razavi, SH; Mousavi, SM; Yeganeh, HM; Marc, I. Fatty acid and carotenoid production by Sporobolomyces ruberrimus when using technical glycerol and ammonium sulphate. Journal of Microbiology and Biotechnology, 2007, 17, 1591-1597. Roukas, T; Lazarides, HN. Ethanol production from deproteinized whey by β-galactosidase coimmobilized cells of Saccharomyces cerevisiae. Journal of Industrial Microbiology, 1991, 7, 15–18. Roukas, T; Mantzouridou, F; Boumpa, Th; Vafiadou, A; Goksungur, Y. Production of βcarotene from beet molasses and deproteinized whey by Blakeslea trispora. Food Biotechnology, 2003, 17, 203-215. Sajbidor, J; Certik, M; Dobronova, S. Influence of different carbon sources on growth lipid content and fatty acid composition in four strains belonging to Mucorales. Biotechnology Letters, 1998, 10, 347-350. Salikova, MV; Tischenko, AV; Tarasova, MN; Ustinov, IG. 1979. Microbial carotene and yeast from whole distillers' grain in cow feeding. Vest. Sel'Skok. Nauki, 4, 71-76. Shi, XM; Jiang, Y; Chen, F. High-yield production of lutein by the green microalga Chlorella protothecoides in heterotrophic fed-batch culture. Biotechnology Progress, 2002, 18, 723–727. Thompson, JC; He, BB. Characterization of crude glycerol from biodiesel production from multiple feedstocks. Applied Engineering in Agriculture, 2006, 22, 261-265. Tom, GD; Vlswanath-Reddy, M; Howe, HB. Effect of carbon source on enzymes involved in glycerol metabolism in Neurospora crassa. Achieves of Microbiology, 1978, 117, 259263. Velayos, A; Eslava, AP; Iturriaga, EA. A bifunctional enzyme with lycopene cyclase and phytoene synthase activities is encoded by the carRP gene of Mucor circinelloides. European Journal of Biochemistry, 2000, 267, 5509–5519. Velayos, A; Fuentes-Vicente, M; Aguilar-Elena, R; Eslava, AP; Iturriaga, EA. A novel fungal prenyl diphosphate synthase in the dimorphic zygomycete Mucor circinelloides. Current Genetics, 2004, 45, 371-377. Verwaal, R; Wang, J; Meijnen, JP; Visser, H; Sandmann, G; van den Berg, JA; van Ooyen, AJ. High-level production of beta-carotene in Saccharomyces cerevisiae by successive transformation with carotenogenic genes from Xanthophyllomyces dendrorhous. Applied and Environmental Microbiology, 2007, 73, 4342-4350. Wang, SL; Sun, JS; Han, B; WU, XZ. Optimization of β-carotene production by Rhodotorula glutinis using high hydrostatic pressure and response surface methodology. Journal of Food Science, 2007, 72, 325-329. Yazdani, SS; Gonzalez, R. Anaerobic fermentation of glycerol: a path to economic viability for the biofuels industry. Current Opinion in Biotechnology, 2007, 18, 213-219. Zalokar, M. Studies on biosynthesis of carotenoids in Neurospora crassa. Archives in Biochemistry and Biophysics, 1954, 50, 71-80.

In: Microbial Conversions of Raw Glycerol Editor: George Aggelis

ISBN 978-1-60692-392-4 © 2009 Nova Science Publishers, Inc.

Chapter IX

Characterization of Microbial Biomass Production from Glycerin Waste by Various Yeast Strains Piotr Juszczyk∗ and Waldemar Rymowicz Department of Biotechnology and Food Microbiology, Faculty of Food Science, University of Environmental and Life Sciences, Poland

Abstract The conversion of glycerin waste from ethyl ester biodiesel production into a protein source for animal feed was examined. In the study, six yeast strains were used for biomass production: Yarrowia lipolytica ATCC 8661, Y. lipolytica ATCC 8661 UV’1, Y. lipolytica A-101, Y. lipolytica Z, Candida robusta ATCC 60 559, and C. utilis ATCC 60 558. The glycerin waste, a main by-product of biodiesel industry, contained (w/w) 45% of raw glycerol, 44% of fatty acids, 3–4% of ethyl esters, and large quantities of potassium soaps. Since the pH of the substrate was 9.6, glycerin waste was added batchwise during yeast growth, in order to prevent the pH value from exceeding 4.0. All experiments were conducted in a stirred tank reactor, on media containing 30 g/L of glycerin waste as a substrate. The results of the experiments have shown that the strain Y. lipolytica 8661 UV’1 is the most suitable for biomass production from this substrate. During batch cultivation, the strain simultaneously utilized glycerol and fatty acids. The biomass yield and biomass production rate obtained with this strain were the highest at 0.89 g/g and 2.7 g/Lh, respectively. Protein concentration in the biomass varied from 26.5 to 36.5% (w/w), depending on the yeast strain used. The content of essential amino acids was in compliance with the FAO/WHO standards for fodder yeast, valine and isoleucine occurring in the yeast biomass in higher quantities. The nutritional value of the yeast obtained in a submerged culture on glycerin waste ranged between 64 and 70% according to Oser's Essential Amino Acid Index (EAAI). Analysis of technological parameters (biomass yields, ∗

e-mail: [email protected].

126

Piotr Juszczyk and Waldemar Rymowicz volumetric biomass production rate and protein content has also revealed that the strain Y. lipolytica ATCC 8661 UV’1 is the most efficient biomass producer from glycerin waste. The results are very promising, as these findings may lead to a low-cost process of fodder yeast biosynthesis from a waste generated during biodiesel production.

Introduction Produced from vegetable oils, animal fats, or waste cooking oils and fats, biodiesel is a renewable, environment-friendly substitute for petroleum-based diesel fuel and can therefore be used in existing diesel engines without expensive modifications. Biodiesel is classified as a safe, nontoxic and biodegradable product, which has the capability to reduce the emission of many harmful compounds concomitant with the combustion of petroleum-based diesel. Biodiesel fuels are defined as fatty acid methyl or ethyl esters. The major feedstock for today’s biodiesel production is rapeseed oil because of its widespread availability, especially in European countries. The main by-product generated in the manufacture of esters is the glycerin waste. Glycerin waste contains glycerol (approx. 40 to 80%, depending on the biodiesel process applied), fatty acids (about 40%), some esters (ethyl or methyl), soaps, protein and water (Thompson and He, 2006). Crude glycerol is a product of very low value because of the impurities. If crude glycerol is to be converted into a product of commercial grade, it should be treated and refined through filtration, chemical additions, and fractional vacuum distillation. Chemically pure glycerol is a valuable industrial compound widely used in the production of consumer goods such as cosmetics and pharmaceuticals (Ma and Hanna, 1999; Thompson and He, 2006). It is essential to note that in the majority of existing biodiesel producing plants raw glycerol is purified only by the addition of mineral acid, which accounts for the separation of inorganic salts and consequently for the formation of an organic phase containing predominantly fatty acids and their methyl or ethyl esters. The organic phase can be recycled into the process, the inorganic salts can be used as fertilizers, and the glycerol (upon methanol removal) can be marketed in the form of raw glycerol. In the European Union and the United States, the last few years have witnessed an exponential growth in biodiesel production, which has noticeably increased the quantity of glycerin (9 kg of biodiesel produced generate approx. 1 kg of glycerin), and so its purification has become uneconomical (Dasari et al., 2005). Nevertheless, owing to the large quantities produced and the low costs involved, raw glycerol can be efficiently used as a carbon source in biotechnological processes. Leaving aside the fact that almost 150 different microorganisms grow on media containing glycerol, reports are also found regarding the use of glycerol for the synthesis of a variety of useful compounds by algae, bacteria and yeasts. The most obvious target or research dealing with the valorization of raw glycerol by biotechnological means is referred to its biotransformation into 1,3-propanediol by several bacterial strains include species of Klebsiella, Citrobacter, Enterobacter and Clostridium) (Pachauri and He, 2006; Rehman et al., 2008). However crude glycerol has been subjected to biotransformation by bacteria into other useful products for example: succinic acid, hydrogen, dihydroksyacetone, polyesters (Pachauri and He, 2006). Glycerol can be utilized by some Candida species as a non-

Characterization of Microbial Biomass Production from Glycerin…

127

conventional carbon source for synthesizing citric acid (Papanikolau and Aggelis, 2002a); Rymowicz et al., 2006; Imandi et al., 2007; Levinson et al., 2007, Rymowicz et al., 2008), pyruvic acid (Morgunov et al., 2004), single cell oil (Meesters et al., 1996; Papanikolaou et al., 2002b; Papanikolaou et al., 2003). Various yeast species are used raw glycerol for Single Cell Protein (SCP) production. They are obtained via natural selection from the environment, as well as by mutation or gene manipulation. Only few yeast species have been declared fit for human consumption. According to the directives of Food and Drug Administration, only dried cells of Saccharomyces cerevisiae, Candida utilis and Kluyveromyces fragilis can be consumed. Beet or cane molasses, lactose whey and sulphite waste liquors are conventional carbon sources for biomass production. Non-conventional carbon sources such as n-alkane, ethanol, methanol, lipids or starch were used also for SCP production by various yeasts. Single cell protein refers to the dried cells of microorganisms. As yeast biomass contains from 40 to 50% of crude protein, it is used in human foods and animal feeds. The advantages of using microorganisms for SCP production (impressive as compared with the use of conventional sources of protein such as soybeans or meat) are widely acknowledged. However, like any other biological material, SCP contains carbohydrate cell wall material, lipids, minerals and vitamins (B-vitamins, vitamin E and provitamin D). With egg albumin, which is considered a well balanced source of essential amino acids for human nutrition, yeast SCP does compare well, except that it is deficient in sulphur containing amino acids (Boze et al., 1992). Yeast are relatively rich in lysine and threonine with respect to other traditional protein sources origin, such as wheat. However, there is a limit to how much yeast can be fed, because about 20% of the crude protein nitrogen in yeast is in the form of nucleic acids. Nucleic acids can cause problem if over fed, because excessive nucleic acid intake result in elevated uric acid levels in blood. High level of uric acid tend to crystallize in the joints and in man and this can cause gout and arthritis or even renal stones. However, in animal feeding this problem is tolerated to a greater extend without any deteriorative effects (Boze et al., 1992). On the other hand, pure or crude glycerol has a very low mammalian toxicity and is used as a feed component in animal diets as an energy source (pure glycerin contains approximately 4,100 kcal/kg of gross energy). Glycerin waste from biodiesel production is an acceptable feed ingredient for poultry, pigs and ruminants (Cerrate et al., 2006; Lammers et al., 2008; Schröder and Südekum, 1999). Most biodiesel feedstocks are derived from vegetable oils, first-use animal fats and waste greases. Animal fat cannot generally be used as animal feed for ruminants because of BSE-related implications. On the other hand, the key problem with biodiesel glycerin is methanol, salt content and impurities that are introduced when use is made of recycled feedstocks (Tyson et al., 2004). The use of agro-industrial residues in submerged fermentations is cost-effective and has the added advantage of being environment-friendly. Microbial conversion of glycerol to fodder yeasts appears to be an alternative to the utilization of crude glycerol. For economic reasons, crude glycerol should be used as a carbon source without prior purification. In this chapter, the advantages of using glycerin waste from ethyl ester production for microbial biomass biosynthesis are examined. Various yeast strains were compared and investigated for biomass production rates and yields in batch cultivations with glycerin waste, a by-product of biodiesel manufacturing

128

Piotr Juszczyk and Waldemar Rymowicz

industry. The nutritional values of yeast protein and the content of essential amino acids in the protein were also determined.

Material and Methods The following yeast strains were used for biomass production: Yarrowia lipolytica ATCC 8661, Y. lipolytica ATCC 8661 UV’1, Candida utilis ATCC 60 558 and C. robusta ATCC 60 559 from American Type Culture Collection, as well as Y. lipolytica A-101 and Y. lipolytica Z from the Department of Biotechnology and Food Microbiology, Wroclaw University of Environmental and Life Sciences (Poland). The strains were grown on YMagar slants at 30°C for 48 h and then stored in a refrigerator at + 4°C. The glycerol fraction (a main by-product of biodiesel manufacturing industry) that was used as the substrate consisted (w/w) of raw glycerol, 50%; fatty acids, 46%, and ethyl esters, 3 to 4%, as well as large amounts of potassium soaps, and was obtained from a pilot installation operated at the Department of Chemistry, University of Wroclaw, Poland. The cultures were grown in 250 mL Erlenmayer’s shake flasks containing 50 mL of the inoculation medium on a G-10 rotary shaker (New Brunswick Co.), at 160 rpm and 30oC for 72 h. The inoculation medium consisted (per litre) of 20 g fatty acids (a by-product of the biodiesel process), 2.0 g (NH4)2SO4, 0.5 g KH2PO4, 0.3 g MgSO4· 7H2O, 1.0 g yeast extract and 1.0 g bacto peptone in 1 L of tap water, pH 6.0. The process was conducted in a 3.5 L stirred tank bioreactor AK-3 with a working volume of 1.1 L at 30°C. A 0.9 L portion of the production medium was inoculated with 0.2 L of the inoculum cultures. The production medium contained (g/L): glycerol fraction, 30; (NH4)2SO4, 11.3; KH2PO4, 0.5; MgSO4·7H2O, 0.5; yeast extract, 1.0, and bacto peptone, 0.75 in 1 L of tap water. Aeration rate was fixed at 1 vvm. Agitation speed was adjusted to 550 rpm. The pH was automatically maintained at 3.5–4.0 by addition of 30% NaOH. Microbial biomass concentration was measured by dry weight estimation. 10 mL of cell suspension from the culture was extracted twice, using petroleum ether (2 x 2.5 mL), in order to dissolve the fatty fraction. After centrifuging (at 4000 rpm for 5 min) and isolation of the ether fractions, the biomass was separated using a 0.45 μm pore-size membrane filter (Millipore), dried at 105°C until a constant weight was reached, cooled in a desiccator and weighed. Protein content was determined using Stewart's method (1975). The unconsumed lipids were also determined. The ether fractions obtained during biomass determinations were evaporated at 50°C in a weighing bottle, dried again at 105°C until a constant weight was reached, cooled in a desiccator and weighed. Glycerol was detected by an RI detector and its concentration was determined by HPLC on an Aminex HPX87H column. The column was eluted with 20 mM H2SO4 at room temperature and a flow rate of 0.6 mL/min. Amino acids were determined using an AAA 400 automatic amino acid analyzer (INGOS, Czech Republic). The EAAI and chemical score (CS) were calculated for evaluating the nutritional value of protein. Amino acid composition of the whole egg protein was used as a standard (Sujak et al., 2006). Amino acids were determined on the g/16 g N basis, equivalent to g/100 g of protein.

Characterization of Microbial Biomass Production from Glycerin…

129

Results and Discussion The process of fodder yeast increment from glycerin waste involved yeast strains of the genera Yarrowia and Candida, which have frequently been used for the increment of a yeast or intracellular fat biomass on glycerol and lipid substrates (Crueger and Crueger, 1989). Batch cultivations of yeasts were conducted in a stirred tank bioreactor, with glycerin waste as a sole carbon and energy source. The substrate contained 50% (w/w) of glycerol and 46% (w/w) of fatty acids. At the initial stage of biomass increment the substrate was added batchwise because of the high pH (approx. 10). The process of biomass increment with different yeast strains is depicted in Figure 1. 30 Y. lipolytica 8661

25

Y. lipolytica A-101

20

Biomass, Glycerol, Free Fatty Acids [g/L]

15 10 5 0

0 1

30

2 3

Y. lipolytica 8661 UV’1

25

4 5

6 7

8 9 10 11

C. robusta ATCC 60 559

20 15 10 5 0 30 Y. lipolytica Z

25

C. utilis ATCC 60 558

20 15 10 5 0 0 1 2

3 4 5 6 7 8

9 10 11

0

1 2

3

4 5

6 7

8

9 10 11

Time [h] Figure 1. Batch cultures of various yeast strains on the medium containing glycerin waste. Biomass (■), Glycerol (●), Free Fatty Acids (▲). Culture conditions: 30 g/L of glycerin waste, pH 3.5-4.0 at 30˚C.

As shown by these plots, all the strains of the genus Yarrowia have utilized the substrate in the medium almost completely. In the 6th hour of Y. lipolytica 8661 UV’1 cultivation no

Piotr Juszczyk and Waldemar Rymowicz

130

glycerol was detected in the medium. Fatty acids, which still persisted there, were totally exhausted in the 11th hour of cultivation. Cultures conducted with the other strains of the genus Yarrowia followed a similar pattern of substrate utilization, except the culture performed with Y. lipolytica Z, where the content of non-utilized fatty acids in the medium amounted to 5.5 g/L. As for the strains C. utilis ATCC 60 558 and C. robusta ATCC 60 559, the utilization of fatty acids as a carbon source occurred first and was followed by the utilization of glycerol, which proceeded at a very slow rate. Thus, by the end of the 11th hour of cultivation, the quantity of glycerol persisting in the medium was 5.2 g/L and 12.6 g/L, respectively (Figure 1). The values of maximal biomass concentration, biomass yield, and of the kinetic parameters of biomass growth (specific biomass growth rate, volumetric biomass increment rate) that were obtained on the glycerol containing medium differed considerably, according to the yeast strain used (Table 1). Table 1. Kinetic parameters of biomass production from glycerin waste by various yeast strains in the stirred tank reactor Yeast strain

Xmax [g/L]

μmax [h-1]

Qx [g/Lh]

YC [g/g]

P [%]

Y. lipolytica ATTC 8661

24.3

0.31

2.2

0.81

36.5

Y. lipolytica ATCC 8661 UV’1

26.8

0.41

2.7

0.89

36.1

Y. lipolytica Z

17.2

0.18

1.2

0.44

36.2

Y. lipolytica A-101

20.1

0.29

1.8

0.67

35.2

C. robusta ATCC 60 559

18.6

0.28

1.7

0.62

36.0

C. utilis ATCC 60 558

21.3

0.28

2.4

0.71

26.5

Culture conditions see Figure 1. Xmax = maximum dry weight of biomass. μmax = maximum specific biomass growth rate. QX = volumetric biomass production rate. YC = total biomass conversion yield (g biomass produced per g initial concentration of glycerin

waste). P = concentration of protein in yeast biomass.

Biomass increment was the highest in the culture that involved the Y. lipolytica 8661 UV’1 strain (26.8 g/L), the lowest biomass increment being observed in the cultivations of Y. lipolytica Z (17.6 g/L) and C. robusta (18.6 g/L). One of the major parameters characterizing the process of biomass increment is the total conversion biomass yield, which varied from 0.62 to 0.89 g/g (depending on the yeast strain used) and did not very much differ from the values attained in the processes of yeast increment from fat- or carbohydrate-based substrates (Montet et al., 1983; Lee and Kim, 2001; Parajo et al., 1995). The highest total conversion yield of biomass produced per glycerin waste consumed (Yc= 0.89 g/g) was achieved with the

Characterization of Microbial Biomass Production from Glycerin…

131

yeast strain Y. lipolytica 8661 UV’1, which was also characterized by the highest volumetric biomass increment rate (Qx=2.7 g/Lh) and the highest specific growth rate (μ=0.41h-1). Total conversion yield, volumetric biomass increment rate, and specific biomass growth rate were very low (Yc= 0.44 g/g, Qx=1.2 g/Lh, μ=0.18 h-1) when the culture was performed using the strain Y. lipolytica Z. The specific biomass growth rate value obtained (μ=0.18 h-1) is comparable with the values reported by Rymowicz et al., (1997) for the cultivations of yeasts of the species Y. lipolytica on fat-based substrates, where they varied from 0.1 to 0.3 h1 . Notably higher values of specific biomass growth rate (μ= 0.5–0.62 h-1) have been obtained by Boze et al., (1992) in cultures where carbohydrates were used as carbon and energy sources. It is essential to note that the choice of the increment strain for SCP synthesis entails careful consideration not only of the process yield, yeast growth rate and protein content (all of them need to be high), but also of the amino acid composition of the protein. The protein content in the biomass of the yeasts produced varied from 26.6 to 36.5%. These values are lower than the ones recommended for fodder yeasts in the standards established by FAO/WHO, which range between 40 and 52%. The highest protein content was determined in the biomass of the strain Y. lipolytica ATCC 8661, the lowest being found in the biomass of C. utilis ATCC 60 558 (Table 1). Seemingly, the low protein percentage in the biomass of the yeasts grown on glycerin waste is attributable to the insufficient quantity of organic nitrogen in the medium. It is worthy of note, however, that according to other researchers (Thompson and He, 2006) the feedstock contains also low amounts of organic nitrogen apart from glycerol and fatty acids. Rymowicz et al., (1997) suggest that the low protein content in the yeast cells is due to the insufficient quantity of the nitrogen source in the culture medium. On the other hand Papanikolaou et al., (2001) reported, that yeasts grown on media containing fatty acids accumulate fat in their biomass, which would explain the reduced protein content. From the comparison of the protein content in the biomass of the yeasts which have grown on various substrates it is obvious that when the cultures were conducted on media containing carbohydrates, the protein level in the cell was higher (34.0 to 70.4%) (Jwanny et al., 1989; Choi and Park, 2003; Paraskevopoulou et al., 2003) than when cultivation was performed with fat-based substrates (23.7 to 50.9%) (Montet et al., 1983; Lee et al., 1993). The dry weight of the biomass of the strains Y. lipolytica A-101, Y. lipolytica ATCC 8661 UV’1 and C. utilis 60558 was analyzed for the nutritional value of the protein content on the basis of amino acid composition (Table 2). The most advantageous amino acid profile and the highest nutritional value were determined in the protein of Y. lipolytica A-101. Leucine content in the biomass of the strain Y. lipolytca A-101 was found to be increased by approximately 10%, which had an influence on the highest sum of essential amino acids (38.1 g/100g protein). The low content of sulphuric amino acids (methionine+cystine) in the samples examined classifies them as the main amino acids that limit the utilization of the protein they contain (CS, from 33.3 to 35.1). The deficiency of sulphuric amino acids in the biomass of the yeasts examined undoubtedly contributed to the decrease in the EAAI.

Piotr Juszczyk and Waldemar Rymowicz

132

Table 2. Comparison of essential amino acids in yeast biomass (g/100 g protein) and their nutritional values Amino acid

Y. lipolytica A-101

Y. lipolytica 8661UV'1

C. utilis 60 558

Whole egg

Lysine

6.7

6.5

6.7

7.0

Methionine/Cystine Threonine

2.0 4.8

1.9 5.1

2.0 3.5

5.7 4.7

Isoleucine

3.9

3.5

3.7

5.4

Tryptophan

1.1

1.1

0.9

1.7

Valine

4.6

4.6

4.8

6.6

Leucine

6.6

6.0

5.9

8.6

Phenylalanine/Tyrosine

8.4

8.4

8.1

9.3

Protein [%]

35.2

36.1

26.5

100

CS

38.1 Met+Cys; 35.1

37.1 Met+Cys; 33.3

35.6 Met+Cys; 35.1

EAAI

70.0

65.0

64.0

Nutritional values ∑ΕΑΑ

* Yeast cultured on media containing glycerin waste as a carbon source; pH 3.5–4.0. ∑EΑΑ – sum of Essential Amino Acids. CS - Chemical Score (CS = (ai/as) × 100; ai - concentration ratio of the restrictive amino acid; as concentration of this amino acid in the standard. EAAI - Essential Amino Acid Index.

For the yeast strains Y. lipolytica A-101, Y. lipolytica ATCC 8661 UV’1 and C. utilis the EAAI values were 70.0, 65.0 and 64.0, respectively (Table 2). Despite the low content of sulphuric amino acids, the yeast protein has a balanced amino acid composition and a high nutritional value. The available literature contains references which corroborate the finding that sulphuric amino acids exert a limiting effect on the nutritional value of the protein synthesized by the yeast, as their quantity in the protein is generally low (varying from 1.0 to 1.9 g/100 g protein) (Boze et al., 1992; Paul et al., 2002). However, the nutritional value of the yeast protein can be increased by genetic modification. Momose and Gregory (1978) have obtained mutants of Saccharomyces cerevisiae that were capable to synthesize protein of a high methionine content (4.3 g/100 g protein) notably exceeding the level of this amino acid in the whole egg protein standard (2.4 g/100 g protein). Wild strains of yeasts were characterized by a relatively low methionine content (1.8 g/100g protein). On the other hand, the quantity of methionine in the protein can be increased by reducing the oxygenation or modifying the composition of the culture medium (Halasz et al., 1988).

Characterization of Microbial Biomass Production from Glycerin…

133

Conclusion Glycerin waste from biofuel (biodiesel) production is an efficient low-cost substrate for the synthesis of fodder yeasts. The parameters of yeast growth and the rate of biomass production on such a substrate depend primarily on the type of the yeast strain used. Of these, the yeasts of the species Y. lipolytica, which utilize the main carbon sources (i.e. glycerol and fatty acids) at the same time, are best suited for cultivation on that kind of substrate. The Y. lipolytica strains chosen produce a biomass with a total conversion yield approaching 0.89 g/g (ATCC 8661 UV’1). The nutritional value of the yeast protein produced on glycerin waste (determined by the Oser method) was high at 64 to 70%. On the other hand, the low content of sulphuric amino acids (i.e. methionine and cystine) in the yeast protein exerted a limiting effect on the nutritional values of the yeasts of the species Y. lipolytica and C. utilis, as can be inferred from the low CS index value (33.3-35.1). The improvement of the yeast production process by optimizing the composition of the culture medium and the conditions of cultivation can markedly increase not only the protein content in the yeast biomass but also the quantity of essential amino acids. The low price of waste glycerin and the high quality of the yeast protein make it possible to obtain a commercial product which can compete with vegetable proteins, specifically with soybeans.

Reference [1] [2]

[3] [4] [5]

[6] [7]

[8]

Boze, H. and Moulin, G. and Galzy, P. (1992). Production of Food and Fodder Yeasts. Critical Reviews in Biotechnology, 12, 65-86. Cerrate, S. and Yan, F. and Wang, Z. and Coto, C. and Sacakli, P. and Waldroup, P.W. (2006). Evaluation of Glycerine from Biodiesel Production as a Feed Ingrediend for Broilers. International Journal of Poultry Science, 5, 1001-1007. Choi, M.H. and Park, Y.H. (2003). Production of yeast biomass using waste Chinese cabbage. Biomass and Bioenergy, 25, 221-226. Crueger, W. and Crueger, A. (1989). Biotechnology: A Textbook of Industrial Microbiology, 2nd ed., Science Tech Publishers,NY, USA. Dasari, M.A. and Kiatsimkul, P.p. and Sutterlin, W.R. and Suppes, G.J. (2005). Lowpressure hydrogenolysis of glycerol to propylene glycol. Applied Catalysis A: General, 281, 225-231. Halasz, A. and Barath, A. and Matrai, B. (1988). Yeast as human protein source. Acta alimentaria Academiae Scientiarum Hungaricae, 174, 374-375. Imandi, S. B. and Bandaru, V. V. R. and Somalanka, S. R. and Garapati, H. R. (2007). Optimization of medium constituents for the production of citric acid from byproduct glycerol using Doehlert experimental design. Enzyme and Microbial Technology, 40, 1367-1372. Jwanny, E.W. and Rashad, M.M. and Moharib, S.A. (1989). Microbial biomass protein and polysaccharide production from vegetable processing wastes. Journal of Basic Microbiology, 29, 581-586.

134 [9]

[10] [11] [12]

[13] [14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

Piotr Juszczyk and Waldemar Rymowicz Lammers, P.J. and Kerr, B.J. and Weber, T.E. and Dozier III, W.A. and Kidd, M.T. and Bregendahl, K. and Honeyman, M.S. (2008).Digestible and metabolizable energy of crude glycerol for growing pigs. Journal of Animal Science, 86, 602-608. Lee, B-K. and Kim, J.K. (2001). Production of Candida utilis biomass on molasses in different culture types. Aquacultural Engineering, 25, 111-124. Lee, C. and Yamakawa, T. and Kodama, T. (1993). Rapid growth of a thermotolerant yeast on palm oil. World Journal of Microbiology and Biotechnology, 9, 187-190. Levinson, W.E. and Kurtzman, C.P. and Kuo, T.M. (2007). Characterization of Yarrowia lipolytica and related species for citric acid production from glycerol. Enzyme and Microbial Technology, 41, 292-295. Ma, F. and Hanna, M.A. (1999). Biodiesel production: A review. Bioresource Technology, 70, 1-50. Meesters, P.A.E.P. and Huijberts, G.N.M. and Eggink, G. (1996). High-cell-density cultivation of the lipid accumulating yeast Cryptococcus curvatus using glycerol as a carbon source. Applied Microbiology and Biotechnology, 45, 575-579. Momose, H. and Gregory, K.F. (1978) Temperature-sensitive mutants of Saccharomyces cerevisiae variable in the methionine content of their protein. Applied and Environmental Microbiology, 35, 641-645. Montet, D. and Ratomahenina, R. and Ba, A. and Pina, M. and Graille, J. and Galzy, P. (1983). Production of single cell protein from vegetable oils. Journal of Fermentation Technology, 61, 417-420. Morgunov, I. and Kamzolova, S. and Perevoznikova, O. and Shishkanova, N. and Finogenova, T. (2004). Pyruvic acid production by a thiamine auxotroph of Yarrowia lipolytica. Process Biochemistry, 39, 1469–1474. Pachauri, N, and He, B. (2006). Value – added utilization of crude glycerol from biodiesel production: A survey of current research activities. American Society of Agricultural and Biological Engineers ASABE, Paper No 066223, 2-15. Papanikolaou, S. and Chevalot, I. and Komaitis, M. and Aggelis, G. and Marc, I. (2001). Kinetic profile of the cellular lipid composition in an oleaginous Yarrowia lipolytica capable of producing a cocoa-butter substitute from industrial fats. Antonie van Leeuwenhoek, 80, 215–224, 2001. Papanikolaou, S. and Aggelis G. (2002). Lipid production by Yarrowia lipolytica growing on industrial glycerol in a single-stage continuous culture. Bioresource Technology, 82, 43-49. Papanikolaou, S. and Muniglia, L. and Chevalot, I. and Aggelis, G. and Marc, I. (2002). Yarrowia lipolytica as a potential producer of citric acid from raw glycerol. Journal of Applied Microbiology, 92, 737-744. Papanikolaou, S. and Muniglia, L. and Chevalot, I. and Aggelis, G. and Marc, I. (2003). Accumulation of a cocoa-butter-like lipid by Yarrowia lipoytica cultivated on agro-industrial residues. Current Microbiology, 46, 124-130. Parajo, J.C. and Santos, V. and Dominguez, H. and Vбzquez, M. (1995). Protein concentrates from yeast cultured in wood hydrolysates. Food Chemistry, 53, 157-163.

Characterization of Microbial Biomass Production from Glycerin…

135

[24] Paraskevopoulou, A. and Athanasiadis, I. and Kanellaki, M. and Bekatorou, A. and Blekas, G. and Kiosseoglou, V. (2003). Functional properties of single cell protein produced by kefir microflora. Food Research International, 36, 431-438. [25] Paul, D. and Mukhopadhyay, R. and Chatterjee, B.P. and Guha, A.K. (2002). Nutritional profile of food yeast Kluyveromyces fragilis biomass grown on whey. Applied Biochemistry and Biotechnology, 97, 209-218. [26] Rehman, A. and Saman Wijesekara, R.G. and Nomura, N. and Sato, S. and Matsumura, M. (2008). Pre-treatment and utilization of raw glycerol from sunflower oil biodiesel for growth and 1,3-propanediol production by Clostridium butyricum. Journal of Chemical Technology and Biotechnology, 80, 1072 – 1080. [27] Rymowicz, W. and Kinal, S. and Wojtatowicz, M. and Musiał, I. and Bodarski, R. (1997). Characteristics of Yarrowia lipolytica biomass produced on lipid substrates. Biotechnologia, 3, 70-77. (in Polish). [28] Rymowicz, W. and Rywińska, A. and Żarowska, B. and Juszczyk, P. (2006). Citric Acid Production from Raw Glycerol by Acetate Mutants of Yarrowia lipolytica. Chemical Papers, 60, 391-394. [29] Rymowicz, W. and Rywińska, A. and Gładkowski, W. (2008). Simultaneous production of citric acid and erythritol from crude glycerol by Yarrowia lipolytica Wratislavia K1. Chemical Papers, 62, 239-246. [30] Stewart, P.R. (1975). Analytical Methods for Yeasts, p. 122–123. In D.M. Prescot (Ed.). Methods in Cell Biology, Vol. 12. Academic Press, New York. [31] Sujak, A. and Kotlarz, A. and Strobel, W. (2006). Compositional and nutritional evaluation of several lupin seeds. Food Chemistry, 98, 711-719. [32] Schröder, A. and Südekum, K.H. (1999). Glycerol as a by-product of biodiesel production in Diets for ruminants. Paper presented at the 10th International Rapeseed Conference, Canberra, Australia. Available from: http://www.regional.org.au/au/ gcirc/1/241.htm [33] Thompson, J.C. and He, B.B. (2006). Characterization of Crude Glycerol From Biodiesel Production From Multiple Feedstocks. Applied Engineering in Agriculture, 22, 261-265. [34] Tyson, K.S. and Bozell, J. and Wallace, R. and Petersen, E. and Moens, L. (2004). Biomass Oil Analysis: Research Needs and Recommendations. Technical Report National Renewable Energy Laboratory Golden, Colorado, USA, June. Retrieved 22nd March 2007. Available from: http://www.nrel.gov/docs/fy04osti/34796.pdf.

In: Microbial Conversions of Raw Glycerol Editor: George Aggelis

ISBN 978-1-60692-392-4 © 2009 Nova Science Publishers, Inc.

Chapter X

Microbial Conversion of Glycerol into 1,3-propanediol: Glycerol Assimilation, Biochemical Events Related with 1,3Propanediol Biosynthesis and Biochemical Engineering of the Process Seraphim Papanikolaou6 Laboratory of Food Microbiology and Biotechnology, Department of Food Science and Technology, Agricultural University of Athens, Athens – Greece

Abstract The ongoing energy crisis has resulted in increasing demands for renewable fuels in the market, and this has as an inevitable effect bio-diesel production. This situation will soon lead to the accumulation of tremendous quantities of crude – impure glycerol in every country utilizing bio-diesel. Therefore, glycerol valorization should have much to offer in the cost reduction of the overall bio-diesel production process. The most important studies that are related with the conversion of (crude) glycerol into higher added-value chemical compounds are referred to its conversion into 1,3-propanediol, a substance of noticeable importance for the chemical and the textile industry. This conversion is carried out with the aid of various prokaryotic microorganisms (principally strains belonging to the family Enterobacteriaceae, to the lactic acid group and to the genus Clostridium sp.) principally under anaerobic conditions. This chapter presents a brief survey of studies that have been carried out by various research teams (including our team – Department of Food Science and Technology of the Agricultural University of Athens and Department of Biology, Division of Genetics, Cell and Development Biology of the University of Patras) in relation with the assimilation of glycerol by bacterial strains and its conversion into 1,3-propanediol. To this end, physiological approaches related with the anaerobic assimilation of glycerol, biochemical aspects 6

Agricultural University of Athens, 75 Iera Odos, 11855 Athens – Greece e-mail address: [email protected], tel: +30-210-5294700, fax: +30-210-5294700.

138

Seraphim Papanikolaou related with the biosynthesis of 1,3-propanediol and biotechnological aspects concerning the feasibility of the process in various fermentation configurations will be considered and discussed.

Keywords: Clostridium butyricum; Klebsiella pneumoniae; glycerol assimilation; DHAregulon; bio-diesel; 1,3-propanediol; fermentation; kinetics; process; modeling

1. Introduction Bio-diesel (green diesel) fuels, defined as principally methyl-esters and in a lesser extent ethyl- or butyl-esters deriving from low-value vegetable or animal fats, already constitute an alternative type of fuel for various types of diesel engines and heating systems (Hüsing et al., 2003; Hirschmann et al., 2005). Due to the increasing cost of conventional fuels, the application of bio-fuels in a large commercial scale is strongly recommended by various authorities, and this fact could likely result in the generation of tremendous quantities of glycerol deposit in the market in the near future (Zeng and Biebl 2002; Hirschmann et al., 2005). With the EU directive 2003/30/EC, which plans to introduce a quantity of 5.75% (w/w) of bio-fuel in the conventional fuel by 2010, the over-production of glycerol will constantly increase. On 2007 an over-capacity of more than 600000 metric tones of glycerol residue in Europe occurred (Papanikolaou et al., 2008). Currently, in various countries of Western Europe (e. g. Germany), crude glycerin water deriving from various bio-diesel plants is treated as a typical "industrial waste-water" (with a cost of 0 $ per kg – it is, therefore, a waste material) being used directly for biogas production. In other countries (e. g. Greece), production of bio-diesel in industrial scale has just recently become a reality, which however necessitates the discovery of ways of glycerol valorization due to the increment of the quantity of this by-product into the local markets. To give the order of magnitude for the generation of glycerol into the market volume, with the production of 10 kg of bio-diesel from various oils, 1 kg of (pure) glycerol becomes available (Meesters et al., 1996; Papanikolaou et al., 2000). Furthermore, it should be stressed that besides bio-diesel production units, concentrated glycerol-containing waters (glycerin) are also produced in significant quantities as the main by-product (or waste material) from fat saponification and alcoholic beverage fabrication units (Barbirato et al., 1998; Papanikolaou et al., 2000; Hüsing et al., 2003; Hirschmann et al., 2005). For all of these reasons, hence, and principally because of the constantly increasing expansion of bio-diesel fuels, glycerol over-production and disposal is very likely to cause severe environmental problems in the near future. Therefore, conversion of glycerol to various higher-added value products by the means of chemical and/or fermentation technology currently attracts much interest (Biebl et al., 1999; Zeng and Biebl 2002; Johnson and Taconi 2007; Koutinas et al., 2007; Pagliaro et al., 2007). While glycerol has been widely used as starting material for various chemical transformations (for reviews see: Johnson and Taconi 2007; Pagliaro et al., 2007), up-to-date a relatively limited number of reports deals with the transformation of this material into high added-value products by the means of fermentation technology. The most obvious target of

Microbial Conversion of Glycerol into 1,3-propanediol

139

biotechnological glycerol valorization is referred to its biotransformation into 1,3propanediol (PD). This product is a substance of importance for the textile industry, due to its application as monomer for the synthesis of aliphatic polyesters (Biebl et al., 1999; Lee et al., 2004). Plastics based on this monomer, besides their bio-degradability, exhibit better product properties and higher light stability than those produced by 1,2-propanediol, 2,3-butanediol or ethylene glycol (Witt et al., 1994). Additionally, a recent development of a new polyester called "poly-propylene terephalate", presenting unique properties for the fiber industry, necessitated the drastic increase in the production of 1,3-propanediol (Lee et al., 2004). Moreover, besides its utilization as a base-unit for the synthesis of biodegradable plastics, 1,3-propanediol can present various interesting applications in the chemical industry; this material can be efficiently used as a polyglycol-type lubricant and its addition can significantly improve the properties in various solvent systems (increased flexibility in blending ester quats), adhesives, laminates, resins (low intrinsic viscosity) and cosmetology products (long-lasting but not sticky moisturizing effect) (Biebl et al., 1999; Zeng and Biebl 2002). Until recently, 1,3-propanediol was not considered to be an easily chemically amendable product given that relatively low selectivity of the existing processes occurred (Biebl et al., 1999). The continuously increasing demand of this compound has led to the elaboration of new chemical routes implicated with the production of 1,3-propanediol. Generally, 1,3propanediol is industrially produced via chemical synthesis carried out by two distinct processes (Zeng and Biebl 2002): the process of "Degussa", now owned by "Dupont", uses the conventional preparation method starting from acroleïne which is obtained by catalytic oxidation of propylene (Broosmer and Arntz 2000). Acroleïne is hydrated at moderate temperature and pressure to 3-hydroxypropionaldehyde (3-HPA) which, in a second reaction, is hydrogenated to 1,3-propanediol over a rubidium catalyst under high pressure (90 bar) (Broosmer and Arntz 2000). The second process carried out by "Shell" has as starting material the compound of ethylene oxide, which is produced via oxidation of ethylene. Ethylene oxide is transformed into 3-HPA through a so-called "hydroformylation" reaction, but to this end significantly high pressures (e. g. around 150 bar) are required (Zeng and Biebl 2002; Johnson and Taconi 2007). The aldehyde is subsequently extracted from the organic phase with water and subjected to hydrogenation using nickel as a catalyst, again under high pressure (Zeng and Biebl 2002). The conversion yield as regards the final product (1,3-propanediol) in relation with the starting materials is around 65% for the case of utilization of acroleïne and 80% for the case of ethylene oxide (Zeng and Biebl 2002). For both types of chemical synthesis elaborated, though, due to the utilization of starting materials and catalysts that could potentially provoke severe health problems (e.g., acroleïne, rubidium etc) and because of the extreme operating conditions used (significantly high pressure employed in various of the aforementioned chemical steps), alternative eco-friendly ways of synthesis of PD present noticeable importance. The development of biotechnological processes aiming at the production of 1,3-propanediol and operating under mild conditions is a logic and feasible solution that currently presents constantly increasing interest (Biebl et al., 1999; Zeng and Biebl 2002; Johnson and Taconi 2007). The principal way of the biotechnological conversion of raw materials to 1,3-propanediol is referred to transformation of glycerol into PD conducted by a number of prokaryotic microorganisms belonging to the

140

Seraphim Papanikolaou

family Enterobacteriaceae, to the group of lactic acid bacteria and to the genus Clostridium sp. under (principally) anaerobic conditions (Schülz and Radler 1983; Gottschalk and Averhoff 1990; Homann et al., 1990; Zeng et al., 1994; Saint-Amans et al., 1994; 2001; Pflugmacher and Gottschalk 1994; Petitdemange et al., 1995; Abbad-Andaloussi et al., 1995; 1996; Barbirato et al., 1997; Menzel et al., 1997a; b; Papanikolaou et al., 2000; 2004; 2008; Chen et al., 2003a; b; Xiu et al., 2004; Mu et al., 2006; Du et al., 2006; 2007). Due to the increasing interest of glycerol as renewable carbon substrate, in various works that have appeared the last years in the literature, researchers have utilized this substrate (either in its pure form or as raw – industrial feedstock, issued from bio-diesel or fat saponification plants) as starting material for microbial conversions other than the fermentation of 1,3-propanediol. Various types of prokaryotic or eukaryotic microorganisms and fermentation configurations have been used and glycerol has been successfully converted into microbial lipid (Meesters et al., 1996; Papanikolaou and Aggelis 2002; Papanikolaou et al., 2008), biomass and α-amylase (Kim et al., 2000), butanol (Andrade and Vasconcelos 2003), hydrogen and ethanol (Ito et al., 2005), pyruvic acid (Morgunov et al., 2004), succinic acid (Lee et al., 2004) citric acid (Papanikolaou et al., 2002; 2008; Papanikolaou and Aggelis 2003; Rymowicz et al., 2005; 2006; 2008; Imandi et al., 2006) and erythritol (Rymowicz et al., 2008). Although utilization of the so-called "crude" glycerol (unpurified concentrated glycerol-water containing various impurities such as catalysts, salts, organosolv, methanol etc), directly discharged from bio-diesel plants as substrate in the fermentation medium without prior purification, offers a remarkable advantage against the traditional use or pure glycerol as carbon source, a relatively limited number of reports have appeared in the literature so far, using this substrate as sole carbon source (Papanikolaou et al., 2002; 2008; Papanikolaou and Aggelis 2002; 2003; Andrade and Vasconcelos 2003; Ito et al., 2005; Rymowicz et al., 2005; 2006; 2008; Imandi et al., 2006). Surprisingly, in most of the cases conversions other than the classical 1,3-propanediol fermentation have been studied. As for the typical conversion of glycerol into PD, although this type of conversion is known since 19th century (Freund 1881), surprisingly, in few reports in the international literature there is utilization of this residue in order to study the production of 1,3-propanediol (Petitdemange et al., 1995; Barbirato et al., 1998; Himmi et al., 1999; González-Pajuedo et al., 2004; Mu et al., 2006; Rehman et al., 2008). Recent investigations of our research teams have showed that promising results have been achieved with a newly isolated and studied Clostridium butyricum strain (strain F2b) during growth on crude – industrial glycerol, residue deriving from bio-diesel plants, in batch, single-stage continuous and two-stage continuous cultures. (Papanikolaou et al., 2000; 2004; 2008; Papanikolaou and Aggelis 2003). The present chapter aimed at resuming the above investigations, by combining them with recent insights of the fermentation of 1,3-propanediol. Biochemical considerations concerning the assimilation of glycerol inside the microbial cell, physiological events related with the biosynthesis of 1,3propanediol under anaerobic, micro-aerobic and aerobic conditions and technological and kinetic aspects of this very interesting bioprocess are considered and discussed.

Microbial Conversion of Glycerol into 1,3-propanediol

141

2. The Biochemistry of 1,3-propanediol Fermentation 2.1. Assimilation of Glycerol inside the Microbial Prokaryotic Cell In prokaryotic microorganisms capable of growing aerobically or anaerobically on glycerol utilized as the sole carbon source (e. g. strains belonging to the enteric group bacteria) the biochemical and metabolic events related with the uptake of glycerol from the culture medium, are completely different in relation with the presence or absence of oxygen in the fermentation medium. These pathways have been described in details by a number of authors in previous investigations (Forage and Lin 1982; Ruch and Lin 1975; Daniel and Gottschalk 1992; Tong and Cameron 1992; Biebl et al., 1999). Glycerol assimilation under aerobic and/or anaerobic fermentation conditions can be effectuated by facultative anaerobic microorganisms (principally) belonging to the family Enterobacteriaceae (in most of the cases microorganisms of the species Klebsiella pneumoniae have been studied – see i. e. Forage and Lin 1982; Ruch and Lin 1975; Neijssel et al., 1975; Menzel et al., 1997a; b; Biebl et al., 1999; Zeng and Biebl 2002; Chen et al., 2003a; b; Aggelis 2007; Hao et al., 2008). In contrast, microorganisms belonging to the species Clostridium sp. can breakdown glycerol utilized as substrate only under strictly anaerobic conditions (Abbad-Andaloussi et al., 1995; 1996; Papanikolaou et al., 2000; 2004; Aggelis 2007). In the presence of oxygen in the fermentation medium (utilization of prokaryotic microorganisms belonging to Enterobacteriaceae family), glycerol is firstly converted into glycerol-3-phosphate (G3P), reaction catalyzed by the so-called enzyme glycerol kinase (GK), and then glycerol-3phosphate is oxidized to dihydroxyacetone phosphate (DHA-P), reaction catalyzed by an aerobic dehydrogenase (Ruch and Lin 1975). The genes of those enzymes constitute a regulon, the so-called GLP-regulon (Forage and Lin 1982; Ruch and Lin 1975). As previously stated, various strains of the enteric group bacteria (principally K. pneumoniae strains) can also carry out glycerol breakdown under strictly anaerobic fermentation conditions. In various cases, use of the above mentioned biochemical pathway (GLP-regulon) can be implicated under anaerobic conditions (Forage and Lin 1975; Ruch and Lin 1975; Neijssel et al., 1975). This event is effectuated only in the presence of an exogenous electron acceptor (e. g. nitrate, fumarate etc) and, as previously, glycerol is primarily phosphorylated to G3P by an ATP-dependent GK. The above enzyme is subjected to a non-competitive feedback inhibition by fructose-1,6-diphosphate (Forage and Lin 1982; Ruch and Lin 1975). As previously (fermentation under aerobic conditions), G3P is then converted to DHA-P, but in the latter case, an anaerobic instead of an aerobic dehydrogenase is implicated in this reaction (Ruch and Lin 1975; Neijssel et al., 1975). The reactions carried out in the GLPregulon are illustrated in Figure 1. In the case in which there is not any available external electron acceptor, in order for glycerol assimilation to be effectuated there is an internal (intra-cellular) acceptor of electrons that is needed. In the latter case, glycerol assimilation occurs by a completely different biochemical mechanism compared with the one of GLP-regulon, and glycerol itself (by virtue of its dehydrated derivative 3-hydroxypropionaldehyde – 3-HPA) constitutes the final electron acceptor; in general, anaerobic assimilation of glycerol in the absence of

142

Seraphim Papanikolaou

external electron acceptor is accompanied by biosynthesis of 1,3-propanediol (PD), which is the product formed in order for NAD+ compounds to be re-generated inside the microbial cells. Precisely, glycerol assimilation occurs via an "oxidative" and a "reductive" branch. A portion of glycerol is transformed into dihydroxyacetone (DHA) by NAD+-dependent glycerol dehydrogenase (GDH). DHA is then phosphorylated to dihydroxyacetone phosphate (DHAP) by DHA-kinase (DHA-K) to enter glycolysis, resulting to the generation of the same end-fermentation products as in sugar fermentation, such as acetic acid, butyric acid, lactic acid, ethanol etc (Biebl et al., 1999; Zeng and Biebl 2002). The formation of the endfermentation products is strain-dependent, and presents notable differences in relation with the strains implicated in the consumption of glycerol; in the case in which Clostridium sp. strains are implicated in the process, mainly acetic and butyric acid are the sole endfermentation products besides 1,3-propanediol that are detected in non-negligible quantities into the fermentation medium (Petitdemange et al., 1995; Abbad-Andaloussi et al., 1995; 1996; Papanikolaou et al., 2000; 2004; 2008). In the case of enteric group bacteria capable of utilizing glycerol as the sole carbon source, a variety of products (e. g. ethanol, 2,3butanediol, lactic acid etc) besides 1,3-propanediol, are synthesized in non-negligible, but variable in relation with the culture conditions, quantities (Zeng et al., 1994; Menzel et al., 1997a; Biebl et al., 1999; Chen et al., 2003a; b). As for the entrance of the C3 compounds into the typical EMP pathway, in general, a triose-phosphate isomerase catalyzes the transformation of DHAP to glyceraldehyde-3-phosphate, which enters the glycolytic pathway (Saint-Amans et al., 2001).

Reactions described by Tong and Cameron (1992). Figure 1. Reactions of GLP- and DHA-regulon. DHA-regulon: Assimilation of glycerol and formation of 1,3-propanediol in prokaryotic glycerol-consuming microorganisms: Enzymes: GDH: NAD+dependent glycerol dehydrogenase; GDHt: B12-dependent glycerol dehydratase; PDOR: NAD+dependent 1,3-propanediol oxidoreductase; DHA-K: Dihydroxyacetone kinase. GLP-regulon: Assimilation of glycerol in the presence of exogenous electron acceptor. Enzymes: GK: Glycerol kinase; G-3P-DH: Glycerol-3-phosphate dehydrogenase.

Microbial Conversion of Glycerol into 1,3-propanediol

143

The quantities of reduced co-factors (NADH2) that are synthesized can be re-generated via the "reductive" branch of glycerol assimilation pathway. Glycerol, which had not been metabolized, is dehydrated to 3-HPA, reaction catalyzed by a B12-dependent dehydratase (GDHt) and then is reduced in 1,3-propanediol by a NAD+-dependent oxidoreductase (PDOR) (reaction re-generating NAD+) (Biebl et al., 1999; Zeng and Biebl 2002). Kinetic and pathway analysis elaborated in both Klebsiella pneumoniae and Clostridium butyricum strains in continuous cultures, suggested that GDHt was the major rate limiting enzyme for the consumption of glycerol and the subsequent formation of 1,3-propanediol, mainly at high substrate concentrations (Abbad-Andaloussi et al., 1996; Ahrens et al., 1998). It is evident that this second metabolic pathway (pathway glycerol – 1,3-propanediol) maintains the redox balance of the cell and is completely necessary for the microorganisms using glycerol the sole carbon and energy source under anaerobic conditions. The four genes encoding the four enzymes of both directions are encoded in one and the same regulon, the so-called DHAregulon (Biebl et al., 1999). These genes have been cloned and sequenced (Daniel and Gottschalk 1992; Tong and Cameron 1992). The reactions carried out in the DHA-regulon are presented in Figure 1. From all the above presented analysis and also from Figure 1, it can be easily assumed that for all strains capable of growing anaerobically on glycerol without exogenous presence of electron acceptors and, hence, capable of producing 1,3-propanediol, the production of other fermentation products, besides 1,3-propanediol, is an inevitable event; the pathway glycerol – 1,3-propanediol is not at all coupled with ATP production, since it is solely and exclusively utilized for the NAD+ factors re-generation. The pathways of biosynthesis of other metabolic products (e. g. acetic acid) that are related with ATP generation are indispensable for the microbial growth, therefore, sole bioconversion of glycerol into 1,3-propanediol without production of other metabolites is impossible (Zeng and Biebl 2002). Furthermore, concerning glycerol breakdown by the enteric group bacteria, it should be stressed that in addition to the respiratory control of both GLP- and DHAregulons, the existence of other modulating factors was also proposed on the basis of chemostat studies carried out by the strain K. pneumoniae NCIB 418 (Neijssel et al., 1975). It appeared that, whereas under anaerobic conditions the DHA-regulon was induced regardless of the glycerol concentration in the growth medium, under aerobic conditions the DHA system was induced when glycerol concentration was significant and the GLP system was induced when glycerol was limiting (Forage and Lin 1975; Neijssel et al., 1975). Since the Km value for glycerol is 1-2 μM for glycerol kinase and 20-40 mM for glycerol dehydrogenase (Neijssel et al., 1975), a rationalization was made that the preferential synthesis of the kinase under substrate-limiting conditions was a physiological adaptive response to enhance substrate-scavenging power (Forage and Lin 1975). In addition, glycerol assimilation through the DHA system and subsequent 1,3-propanediol (PD) biosynthesis has been reported by various K. pneumoniae strains cultivated under micro-aerobic conditions in glycerol-rich media, with the efficiency of the fermentation (expressed in both absolute PD values in g dm-3 and conversion yield of product synthesized per substrate consumed – YPD/S in g g-1) being significantly influenced by the elaboration of micro-aerobic conditions instead of strict anaerobic ones (Chen et al., 2003a; b) or by differentiations employed in the extracellular oxido-reduction potential of culture (Du et al., 2006; 2007). Additionally, in a recent investigation, various strains belonging to the enteric group bacteria (namely strains

144

Seraphim Papanikolaou

belonging to Citrobacter freundii and K. pneumoniae) have been identified and were revealed capable of assimilating glycerol through the DHA-regulon under strictly aerobic conditions (Hao et al., 2008). One Klebsiella pneumoniae subsp. Pneumoniae strain, designated as strain TUAC01, demonstrated comparable levels of PDOR, GDHt and GDH activities to the anaerobic microorganisms described in the literature (Hao et al., 2008). Accordingly, in larger scales (5-dm3 bioreactor) fed-batch culture of K. pneumoniae TUAC01 strain showed a remarkable PD biosynthesis under aerobic conditions; 60.1 g dm-3 of 1,3propanediol were produced after 42 h incubation in an agitating bioreactor. Moreover, in airlift bioreactor systems, 66.3 g dm-3 of 1,3-propanediol were produced after 58.5 h incubation (Hao et al., 2008). In another approach recently appeared in the literature, in order to investigate the possibility of maximization of 1,3-propanediol production, genetically modified strains derivatives of the wild strain K. pneumoniae DSM 2026 have been created that presented over-expression of genes of the DHA-regulon. GDHt was over-expressed separately or coordinately with PDOR in order to investigate its effect on 1,3-propanediol fermentation (Zheng et al., 2006). Nevertheless, the recombinant created strains PKM13 and PKM15 that were grown anaerobically using glycerol as the sole carbon source in both batch and continuous bioreactor cultures showed instability of the plasmids bearing the genes encoding GDHt and PDOR. Moreover, the putative regulatory gene DHAR of the DHA-regulon was over-expressed (generation of the genetically engineered derivative strain PKM16) and this strain was anaerobically batch-cultured on bioreactor experiment. Despite increased 1,3propanediol oxidoreductase (PDOR) activity and plasmid stability, significant differences in the spectrum of end-fermentation products compared with the wild strain (DSM 2026) were observed; strain PKM16 presented decreased 1,3-propanediol and biomass production and increased lactate production compared with the wild strain. The instability of plasmids for the strains PKM13 and PKM15 as well as the decreased 1,3-propanediol production of strain PKM16 were mainly attributed to imbalanced conversion of glycerol to 1,3-propanediol resulting to accumulation of the toxic 3-HPA intermediate (Zheng et al., 2006).

2.2. Biochemical Events Related with Glycerol Breakdown in Glycerolconsuming Prokaryotic Microorganisms A number of prokaryotic microorganisms can be revealed capable of growing on glycerol and producing in various amounts PD as one of the final fermentation products. The investigations that deal with the production of 1,3-propanediol by prokaryotic microorganisms implicate strains belonging to the genus Clostridim sp. (Abbad-Andaloussi et al., 1995; 1996; Himmi et al., 1999; Papanikolaou et al., 2000; 2004; 2008; Malaoui and Marczak 2001), to the enteric group bacteria (Boenigk et al., 1992; Menzel et al., 1997a; b; Barbirato et al., 1997; 1998; Chen et al., 2003a; b; Mu et al., 2006; Zheng et al., 2006; Hao et al., 2008) and to lactic acid group bacteria (Schülz and Radler 1984; de Valdez et al., 1997; Claisse and Lonvaud-Funel 2000; Garai-Ibade et al., 2008). The biochemical pathways that are involved in glycerol breakdown by all of the above-mentioned microorganisms present noticeable differences; in the case of Clostridium sp. strains or enteric group bacteria,

Microbial Conversion of Glycerol into 1,3-propanediol

145

glycerol can be utilized as the sole carbon and energy source and can be converted into 1,3propanediol (together with various other metabolic products related with the organisms utilized in the conversion). In contrast, in the case of lactic acid bacteria, glycerol cannot be used as the sole carbon source; a supplementary source is needed (e. g. maltose, glucose, fructose etc. in order for the creation of ATP and biomass to be effectuated, whilst glycerol is utilized only as the final electron (NADH2) acceptor of the process. In most of the cases in which glycerol is utilized as a supplementary carbon source together with a fermentable sugar by lactic acid bacteria, sugar catabolism is carried out through the hetero-fermentative biochemical pathway, while 1,3-propanediol (and in a lesser extent 3-HPA) is formed together with the main fermentation products of the process (acetic acid, lactic acid and ethanol) (Schülz and Radler 1984; Veiga-da-Cunha and Foster 1992; de Valdez et al., 1997; El-Ziney et al., 1998; Claisse and Lonvaud-Funel 2000; Garai-Ibade et al., 2008). For the case of lactic acid bacteria capable of assimilating glycerol, the biochemical synthesis of 1,3propanediol presents significant similarities with the DHA-regulon; as previously GDHt requires a B12 vitamin for its function (Veiga-da-Cunha and Foster 1992), and the dehydrated product of glycerol catabolism (3-HPA) is transformed into 1,3-propanediol, reaction catalyzed again by PDOR. This enzyme, presenting significant similarities with the one of enteric group bacteria or Clostridium sp. strains, follows a Michaelis-Menten kinetics presenting a Km value for 1,3-propanediol of 2 mM. Other final products related with the initial electron acceptor that can be synthesized instead of 1,3-propanediol are 1,2propanediol, meso-2,3-butanediol and ethylene-glycol (Schülz and Radler 1983), while 3HPA can, in rare cases, be spontaneously dehydrated into acroleïne (e. g. studies carried out with Lactobacillus sp. NRRL B-1720, see for instance Slininger and Bothast 1985). The biochemical pathways involved in the catabolism of sugar and glycerol in the various lactic acid bacterial strains is illustrated in Figure 2 (according to Schütz and Radler 1984). It should be stressed that, in general, the concentrations of 1,3-propanediol achieved by lactic acid bacteria implicated in this type of conversion are not very high (within the range of 5-15 g dm-3 – see i. e. de Valdez et al., 1997; Claisse and Lonvaud-Funel 2000; Garai-Ibade et al., 2008), while technical (unpurified) glycerol has never been utilized as microorganisms cosubstrate. The most important part of studies dealing with glycerol consumption and concomitant PD biosynthesis elaborated in terms of biochemistry, biochemical engineering and mathematical modeling, have been effectuated by either bacteria belonging to the enteric group or the ones of the species Clostridium sp.. As for the fermentation carried out by strains of the enteric group bacteria (family Enterobacteriaceae) in most of the cases the cultivations have been effectuated by strains of the species Klebsiella pneumoniae (Menzel et al. 1997a; b; Chen et al., 2003a; b; Mu et al., 2006; Zheng et al., 2006; Hao et al., 2008), Klebsiella oxytoca (Yang et al., 2007), Enterobacter agglomerans (Barbirato et al., 1997; 1998) and Citrobacter freundii (Boenigk et al., 1992). The biochemical pathways of glycerol breakdown by the enteric group bacteria present similar end-fermentation products with the ones carried out on sugars utilized as substrate (with, of course, the exemption of 1,3propanediol accumulation into the fermentation medium that is effectuated only by the presence of glycerol as carbon substrate). The biochemical pathways implicated with the catabolism of glycerol in the various enteric group bacterial strains is illustrated in Figure 3

146

Seraphim Papanikolaou

(according to Biebl et al., 1999; Zeng and Biebl 2002; Chen et al., 2003a). From the pathways illustrated above, it can be assumed that a plethora of products, besides PD, can be generated during glycerol fermentation elaborated by the enteric group bacteria (e.g., 2,3butanediol, succinic acid, acetic acid, ethanol, formic acid, lactic acid). In fact, there are two types of fermentations that can be involved by the aforementioned microbial strains: the socalled "butanediol" and "mixed acids" fermentations (Aggelis 2007). The above types of fermentations are strain-dependent processes, which are also influenced by the fermentation conditions elaborated. Whatever happens, and according to the biochemical pathways illustrated in Figure 3, it should be stressed that the pathway glycerol – 1,3-propanediol (as previously stated – see chapter "2.1. Assimilation of glycerol inside the microbial prokaryotic cell") does not generate energy (ATP) for the cell but it is completely indispensable for the microbial strains since it can re-generate NAD+ compounds (Biebl et al., 1999; Zeng and Biebl 2002).

Pathways described by Schütz and Radler (1984) and Aggelis (2007). Figure 2. Catabolic pathways of simultaneous glycerol and sugar catabolism in lactic acid bacteria. 3HPA: 3-hydroxypropionaldehyde; GDHt: Glycerol dehydratase; PDOR: 1,3-propanediol oxidoreductase; G6P-OR: Glucose-6P-oxido-reductase; Glol6P-OR: 6P-gluconate--oxido-reductase; Rib-5PEp: Ribulose-5P 3-epimerase; Xyl5P-Gl-3P-lyase: Xylulose-5P glycerynaldehyde-3P-lyase; ATP-AcP-trans: ATP-acetate transferase; Gl-3P-OR: Glycerynaldehyde-3P-oxido-reductase; Lac-NAD-OR: Lactate-NAD+-oxido-recuctase.

Microbial Conversion of Glycerol into 1,3-propanediol

147

Pathways described by Biebl et al., (1999); Zeng and Biebl (2002); Chen et al., (2003a). Figure 3. Catabolic pathways of glycerol catabolism in enteric group bacteria. 3-HPA: 3hydroxypropionaldehyde; GDHt: glycerol dehydratase; GDH: glycerol dehydrogenase; DHAk: Dihydroxyacetone kinase; PDOR: 1,3-propanediol oxidoreductase; PDH: Pyruvate dehydrogenase; PFL: pyruvate formate lyase; FHL: Formate hydrogenolyase.

From the biochemical pathway illustrated in Figure 3, it must be noted that the fermentation yield of 1,3-propanediol (in terms of both absolute values of PD produced in g dm-3 and relative values of PD formed per glycerol consumed) is critically influenced by the activity of the biochemical pathways other that the one of glycerol – 1,3-propanediol; it is evident that pathways that are used for the re-generation of NAD+ co-factors are in competition with the one of glycerol – 1,3-propanediol, and, therefore, if these pathways present significant activity, the yield of PD decreases. Therefore, in glycerol fermentation conducted by Enterobacteria, the pathway glycerol – acetic acid is an indispensable one for the microbial metabolism since it provides energy (ATP) necessary for the microbial growth, while it is not found in competition with the one of glycerol – 1,3-propanediol since it does not re-consume NADH2 co-factors (see Figure 3). In contrast, the pathways glycerol – lactate, glycerol – succinate and glycerol – ethanol are competitive to that of glycerol – 1,3propanediol, since all of these pathways need NADH2 for their completion. It is interesting, thus, not to direct the microbial metabolism towards these undesirable end-fermentation by-

148

Seraphim Papanikolaou

products (lactate, succinate and ethanol) (Menzel et al., 1997a; b; Biebl et al., 1999). As previously stated, the variety of fermentation by-products generated during glycerol breakdown by enteric group bacteria is a strain dependent process (Biebl et al., 1999; Aggelis 2007). By using fermentation technology approaches, it is possible to study the metabolism of the given microorganism used for the production of PD and identify and quantify the possible undesirable by-products of the metabolism. In the next step, by using classical mutation of genetic engineering techniques, it is possible to "produce" mutant strains lacking of the undesirable pathways, antagonistic to the ones of PD biosynthesis. For instance, Klebsiella oxytoca strain M5al is considered as an excellent 1,3-propanediol producer, but glycerol breakdown results in the undesirable formation of the by-product lactate in significant quantities (sometimes lactic acid quantities are in the range of 20-40 g dm-3), resulting, thus, in decreased fermentation efficiency for PD (Yang et al., 2007). To counteract the disadvantage of significant lactic acid production, four lactate deficient mutants were obtained by knocking out the ldhA gene encoding for the biosynthesis of lactate dehydrogenase (LDH) of K. oxytoca M5al. The LDH activities of the four mutants were from 3.85 to 6.92% of the parental strain. The fed-batch fermentation of 1,3-propanediol by the mutant LDH3 strain, whose LDH activity is the lowest, was studied. The results showed that higher PD yield, in terms of absolute value, volumetric productivity and conversion rate from glycerol was achieved compared to that obtained from the wild type strain, while no lactic acid at all was produced under both anaerobic and micro-aerobic conditions (Yang et al., 2007). As previously stated, the efficiency of the fermentation of PD production elaborated by bacterial strains of the enteric group is critically influenced of the intra-cellular pool of NADH2 co-factors, that can be recycled yielding NAD+ through the pathway glycerol – 1,3propanediol. A non-negligible portion of NADH2 co-factors can be obtained through the decarboxylation of puryvic acid; in general, in strains belonging to Enrerobacteria, pyruvic acid is subjected to a very specific cleavage, which is carried out in microorganisms of this group, resulting in the synthesis of CH3COSCoA and formic acid (HCOOH) (Streekstra et al., 1987). This reaction is catalyzed by the so-called enzyme pyruvate formate lyase (PFL) and by virtue of this reaction acetic acid and formic acid can be generated. Formic acid then can potentially be cleaved into CO2 and H2, reaction catalyzed by formate hydrogeno-lyase (FHL) (Figure 4) (Streekstra et al., 1987; Chen et al., 2003a; Zeng and Biebl 2002). In these conditions, hence, no supplementary NADH2 quantity could be generated in order to be used in the pathway of glycerol – 1,3-propanediol. Recent investigations though, utilizing the strain K. pneumoniae DSM 2026 have indicated that when this microorganism was cultivated anaerobically in continuous cultures at high initial C/N media with glycerol found in excess of more than 50 g dm-3 in the culture medium, PD quantities of 48 g dm-3 with conversion yield of 1,3-propanediol formed per glycerol consumed (YPD/S) of 0.52 g g-1 were achieved. The corresponding values of 33 g dm-3 and 0.38 g g-1 at glycerol-limiting conditions were observed (Menzel et al., 1997a). Analyses of the intra-cellular enzyme activities and coenzymes concentrations indicated a new pathway of pyruvic acid decarboxylation, effectuated anaerobically and in the presence of glycerol found in noticeable excess into the fermentation medium. In this pathway, the implicated enzyme is the so-called pyruvate dehydrogenase (PDH) and by virtue of the catalyzed reaction CH3COSCoA and CO2 are

Microbial Conversion of Glycerol into 1,3-propanediol

149

synthesized since additional generation of NADH2 co-factors is elaborated (which would be re-oxidized through the pathway of PD biosynthesis) (Menzel et al., 1997a; b). It should also be stressed that yield of 1,3-propanediol produced per glycerol consumed (YPD/S) can be further increased (reaching maximum values of 0.88 mol mol-1), if CH3COSCoA deriving from pyruvate cleavage is channeled into the tricarboxylic acid (TCA) cycle for reducing power and adenosine triphosphate (ATP) generation (Zeng and Biebl 2002). Enzymatic assays have shown that a part of the TCA cycle reaction is active in the anaerobic glycerol fermentation (Zeng and Biebl 2002; Chen et al. 2003a). The dual pathway of pyruvic acid decarboxylation in the enteric group bacteria is illustrated in Figure 4.

Reactions described by Zeng and Biebl (2002); Chen et al., (2003a). Figure 4. Reaction of puryvic acid breakdown in bacterial strains belonging to the enteric group.

Besides the biosynthesis of 1,3-propanediol by the enteric group bacteria, a significant number of studies have been elaborated by using various Clostridium sp. strains. In most of the cases, strains of the species Clostridium butyricum have been successfully used in this type of fermentation (Abbad-Andaloussi et al., 1996; 1998; Reimann and Biebl 1996; Biebl et al., 1999; Papanikolaou et al., 2000; 2004; 2008; Papanikolaou and Aggelis 2002). The biochemical pathways involved in the catabolism of glycerol by Clostridium butyricum is illustrated in Figure 5 (according to Zeng and Biebl 2002; Papanikolaou et al., 2008).

150

Seraphim Papanikolaou

Pathways described by Zeng and Biebl (2002); Papanikolaou et al., (2008). Figure 5. Catabolic pathways of glycerol catabolism in Clostridium sp. strains. 3-HPA: 3hydroxypropionaldehyde; GDHt: glycerol dehydratase; GDH: glycerol dehydrogenase; DHAk: Dihydroxyacetone kinase; PDOR: 1,3-propanediol oxidoreductase; FDox (or red): Ferredoxine oxidoreductase.

From the pathways illustrated, it may be assumed that in contrast with the conversions carried out by the enteric group bacteria, in the case of Clostridium sp. strains, the number of end-fermentation products is indeed restricted. In most cases, the only products (besides PD) that are identified into the fermentation medium are the organic acids acetate and butyrate (Abbad-Andaloussi et al., 1996; 1998; Reimann and Biebl 1996; Papanikolaou et al., 2000; 2004; 2008). Under this optic and in a first approach, hence, it may be assumed that the conversion yield YPD/S (in g g-1) is notably higher in the fermentations carried out by Clostridium sp. strains compared with the one of the enteric group bacteria, since in the former case the intra-cellular carbon flow is channeled towards the synthesis of less products compared with the latter one. Therefore, a higher amount of carbon should be channeled

Microbial Conversion of Glycerol into 1,3-propanediol

151

towards the pathway of glycerol – 1,3-propanediol (Papanikolaou et al., 2000; 2004). As in the case of fermentation carried out by Enterobacteria, from the biochemical pathway illustrated in Figure 5, in the conversion carried out by Clostridium sp. strains it can be seen that the fermentation yield of 1,3-propanediol is critically influenced by the activity of the biochemical pathways other that the one of glycerol – 1,3-propanediol; again pathways that are used for the re-generation of NAD+ co-factors are in competition with the one of glycerol – 1,3-propanediol. Therefore, as in the case of enteric group bacteria, the pathway, glycerol – acetic acid is an indispensable one since it provides energy (ATP) necessary for the microbial growth, while it is not found in competition with the pathway glycerol – 1,3-propanediol and, hence, it does not re-consume NADH2 co-factors (see Figure 5). The undesirable fermentation by-product in this process is the one of butyric acid, since glycerol – butyric acid pathway is in competition with the one of glycerol – 1,3-propanediol for the regeneration of NAD+ co-factors (Abbad-Andaloussi et al., 1996; 1998; Papanikolaou et al., 2000; 2004). Recent investigations with a newly isolated C. butyricum strain (strain F2b) have indicated that when this microorganism was cultivated in continuous mode at various inlet glycerol concentration media (initial glycerol concentrations in the chemostat at 30, 60 and 90 g dm-3) and various dilution rates (D) imposed, at steady-state conditions, increment of D imposed significantly increased the specific production rate of acetic acid (qAc), decreasing simultaneously the one of butyric acid (qBut) (Figure 6). Although it is relatively difficult to conclude if, indeed, there was a disassociation of the products from cell growth, it seems that at low and medium specific growth rates, butyrate production was preferred, whereas high dilution rates provoked a metabolic shift in the strain, favoring acetate production (Papanikolaou et al., 2000). Similar results have been reported during the culture of C. butyricum DSM 5431, in which an increment of acetate/butyrate molar ratio from 0.13 to 22.15 was observed when the dilution rate increased from 0.10 to 0.60 h-1 at constant inlet glycerol concentration (Zeng et al., 1994). It appears that at high growth rates, there was a bottleneck for the carbon flux through butyrate biosynthesis in the level of the enzyme thiolase (catalyzing the condensation reaction of acetyl-SCoA to aceto-acetyl-SCoA), resulting, hence, in the direction of the cellular metabolism towards the biosynthesis of the acetic acid (Abbad-Andaloussi et al., 1996). Furthermore, in the strains of Clostridium sp., besides the specific growth rate imposed (μ), an important factor that also should be seriously taken into consideration is the one of the supplementary production of reducing equivalents (NADH2 co-factors) during the oxidative decarboxylation of pyruvic acid. This specific reaction that exists only in microorganisms of the genus Clostridium sp., the so-called phosphoroclastic reaction, results in the generation of 1 mol of CH3SOSCoA per 1 mol of puryvic acid cleaved, with liberation of 1 mol of CO2 and 1 mol of H2 (Petitdemange et al., 1976). Theoretically, during the phosphoroclastic reaction the molar ratio H2/CO2 is equal to 1. However, there is an enzymatic complex, the so-called ferredoxide oxido-reductase that is implicated during the phosphoroclastic reaction, and its activity can alter the above ratio. Therefore, during growth of Clostridium sp. strains on glucose utilized as substrate (fermentation of acetone-butanol-ethanol) the molar ratio H2/CO2 was higher than 1 suggesting that NADH2 co-factors that had been produced during the catabolism of glucose had been transformed into H2 through the reaction catalyzed by ferredoxide oxido-reductase (Petitdemange et al., 1976).

Seraphim Papanikolaou

152

a.

b. Graphs by Papanikolaou et al., 2000. Figure 6. Representation of the specific rate of butyrate production (qBut) (6a) and acetate production (qAc) (6b) versus the dilution rate imposed (filled circles when initial glycerol concentration – S0=90 g dm-3, open circles when S0=60 g dm-3, open squares when S0=30 g dm-3) during an anaerobic singlestage continuous fermentation of industrial glycerol by Clostridium butyricum F2b.

In contrast, during glycerol breakdown by Clostridium sp. strains, the molar ratio H2/CO2 was found to be lower than 1, indicating inverse activity of ferredoxide oxido-reductase and "capture" of H2 molecules in the form of NADH2 co-factors, that could have been regenerated through the pathway of glycerol – 1,3-propanediol (Biebl et al., 1992). The activity of ferredoxide oxido-reductase towards the supplemented production of NADH2 co-factors seemed very increased in Clostridium butyricum cultures with significant excess of glycerol

Microbial Conversion of Glycerol into 1,3-propanediol

153

into the fermentation medium (Zeng and Biebl 2002). Therefore, in the fermentation of glycerol by Clostridium sp. strains, supplementary reducing power could be generated through the phosphoroclastic reaction, increasing, thus, the yield of PD produced. The oxidative decarboxylation of pyruvic acid effectuated in Clostridium sp. strains is illustrated in Figure 7.

Reactions described by Petitdemange et al. (1976). Figure 7. The phosphoroclastic reaction in Clostridium sp. strains.

3. The Process of 1,3-propanediol Fermentation A significant number of natural, mutant or genetically modified bacterial strains have been reported of being capable to produce 1,3-propanediol, in various amounts and fermentation configurations. The most extensively studied microorganisms belong to the species Citrobacter freundii, Klebsiella pneumoniae, Klebsiella oxytoca, Enterobacter agglomerans, Clostridium butyricum and Clostridium acetobutylicum (Homann et al., 1990; Gottschalk and Averhoff 1990; Biebl et al., 1992; Saint-Amans et al., 1994; 2001; Zeng et al., 1994; Petitdemange et al., 1995; Barbirato et al., 1997; 1998; Menzel et al., 1997a; b; Papanikolaou et al., 2000; 2004; 2008; Chen et al., 2003a; b; Hirschman et al., 2005; Du et al., 2006; Mu et al., 2006; Rehman et al., 2008). As previously stated, a noticeable number of studies have been elaborated in relation with the intra-cellular enzyme activities and coenzyme concentrations in order to identify the limiting steps of PD production (AbbadAndaloussi et al., 1996; Barbirato et al., 1997; Menzel et al., 1997b; Ahrens et al., 1998; Saint-Amans et al., 2001), while classical mutation or genetic engineering techniques have been developed in order to "create" robust PD microbial producers (product tolerant, butyrate-negative or lactate-negative mutants see i.e., Abbad-Andaloussi et al., 1995; Reimann and Biebl 1996; Zheng et al., 2006; Yang et al., 2007). Furthermore, Pflugmacher and Gottschalk (1994) have developed an immobilized cell reactor, whereas Günzel et al.,

154

Seraphim Papanikolaou

(1991) have scaled-up the process in reactors of 2000 dm3, with good productivity (2.8 g dm-3 h-1). The microbial production of 1,3-propanediol using glycerol as sole substrate has been elaborated in almost every possible fermentation configuration. In most of the cases pure glycerol has been employed as substrate, but in some investigations raw (technical – unpurified) glycerol was used as carbon source in both Clostridium sp. or Klebsiella sp. strains (Petitdemange et al., 1995; Barbirato et al., 1998; Himmi et al., 1999; Papanikolaou et al., 2000; 2004; 2008; Papanikolaou and Aggelis 2003; González-Pajuedo et al., 2004; Hirschman et al., 2005; Mu et al., 2006; Rehman et al., 2008). Typical values of PD production in batch and fed-batch systems of both Clostridium sp. or Klebsiella sp. strains range between 20-50 g dm-3 regardless of the strain or the culture conditions employed (Homann et al., 1990; Günzel et al., 1991; Petitdemange et al., 1995; Barbirato et al., 1998; Himmi et al., 1999; Chen et al., 2003a; b; Du et al., 2006; 2007; Mu et al., 2006; Papanikolaou et al., 2008; Rehman et al., 2008). In some cases, utilization of C. butyricum mutants in fed-batch fermentation systems resulted in significant PD quantities, of more than 70 g dm-3 (Reimann and Biebl 1996). The highest final PD concentrations reported in the literature so far are in the range of 80.1-87.7 g dm-3, obtained in fed-batch cultures of a newly isolated Clostridium sp. strain (strain IK 124) when pure or raw glycerol was used as the sole substrate (Hirschman et al., 2005). Additionally, Yang et al., (2007) have indicated final PD concentrations of 83.6 g dm-3 during fed-batch cultures of a genetically modified negative to lactate dehydrogenase derivative of the strain K. oxytoca M5a1, cultivated in micro-aerobic conditions. In the latter case, sucrose has been employed as glycerol co-substrate (Yang et al., 2007). This strategy (addition of sugar e. g. glucose into the fermentation medium) has been also employed in some other cases in order to maximize the yield YPD/S by Klebsiella sp. or Clostridium sp. strains, and this resulted in almost stoichiometric transformation of glycerol consumed to PD produced (YPD/S around 0.8-0.9 g g-1 – Biebl and Marten 1996; AbbadAndaloussi et al., 1998; Malaoui and Marczak 2001). As stated above, the highest 1,3-propanediol concentration obtained in batch and fedbatch cultures is around 80-88 g dm-3 (Hirschman et al., 2005; Yang et al., 2007). Maximum volumetric productivities for these types of processes were about 2.5-2.9 g dm-3 h-1 (Günzel et al., 1991; Saint-Amans et al., 1994; Reimann and Biebl 1996). The maximum concentrations of 1,3-propanediol achieved in continuous cultures are within the range of 3548 g dm-3 for both K. pneumoniae and C. butyricum strains (Menzel et al., 1997a; Papanikolaou et al., 2000; 2004; 2008). In the case of K. pneumoniae strains, simultaneously with the high PD production, increased volumetric productivities (5.1-8.8 g dm-3 h-1) were reported, but in this case significant quantities of glycerol remained unconsumed into the fermentation vessel (Menzel et al., 1997a). This is a major problem considering the aspect of downstream processing and recovery of PD synthesized. Concerning the production of PD in continuous cultures by C. butyricum strains, the newly isolated F2b strain was revealed as an excellent PD producer during growth in chemostat experiments (highest PD concentrations achieved within the range 44-48 g dm-3). It should also be pointed out that in contrast with the majority of the previous investigations in which fermentations were elaborated using pure glycerol as substrate, the aforementioned strain was cultivated on industrial glycerol, by-

Microbial Conversion of Glycerol into 1,3-propanediol

155

product of bio-diesel production plants, utilized as sole carbon source (Papanikolaou et al., 2000; 2004). When this microorganism was cultivated in continuous mode at initial glycerol concentrations in the chemostat at 30, 60 and 90 g dm-3 and various dilution rates (D) imposed, 1,3-propanediol was revealed to be growth-associated product (Figure 8), with global fermentation yield of 1,3-propanediol produced per glycerol consumed of 0.55 g g-1 (Figure 9) (Papanikolaou et al., 2000). It is of importance to state that although the production of acetate was favored at high dilution rates (see Figure 6b), yield YPD/S remained practically constant regardless of the D imposed (Papanikolaou et al., 2000). The constancy of the conversion of glycerol into PD has been reported for various C. butyricum strains in batch and continuous fermentations suggesting the presence of mechanisms that precisely partition the carbon flow between biosynthesis of organic acids and formation of 1,3propanediol (Abbad-Andaloussi et al., 1996; 1998; Barbirato et al., 1998; Papanikolaou et al., 2000; 2004). The drawback of the continuous cultures led by C. butyricum F2b was the inability of achieving simultaneously high concentrations of PD and high volumetric productivities, since the enhanced PD quantities were achieved at low D imposed (Papanikolaou et al., 2000). To alleviate this drawback, a continuous two-stage culture was an attractive perspective and it was effectuated; the first stage of the culture conducted with a reduced volume, presenting, hence, a high dilution rate, so that an increased 1,3-propanediol volumetric productivity could have been achieved. The second stage with a higher active volume, presenting, thus, a lower dilution rate, could be mainly used to further increase PD concentration (Papanikolaou et al., 2000). Initial concentration in the inlet medium was at 90 g dm-3, and indeed, high PD concentrations were obtained in the second stage of the culture (41-46 g dm-3), while the global 1,3-propanediol volumetric productivity achieved, although it was not the maximum one of the two-stage culture, presented a relatively satisfactory value (3.4 g dm-3 h-1) (Figure 10) (Papanikolaou et al., 2000). A two-step production of 1,3-propanediol has been also realized by the use of a Citrobacter freundii strain; in the first stage and under glycerol limitation, active biomass was produced. In the second stage, a decrease of the dilution rate was carried out, in order to further increase the 1,3-propanediol production. The maximum productivity obtained was 1.8 g dm-3 h-1 while the highest 1,3-propanediol concentration was 42 g dm-3 (Boenigk et al., 1993). The fermentations of glycerol realized by Clostridium sp. strains generally present a higher yield YPD/S than that obtained during the culture conducted by the Klebsiella sp. (achieved YPD/S yields 0.54±0.02 against 0.45±0.02 g g-1) since in the latter case, the metabolism is directed towards the synthesis of more end-fermentation products compared with the former one, with evident result a decrease of intra-cellular carbon flow towards the pathway glycerol – 1,3-propanediol (Biebl et al., 1999). In Klebsiella sp. cultures, yield YPD/S and volumetric productivity can be substantially increased when differentiations in the culture conditions are employed; as previously stated (see "2.2. Biochemical events related with glycerol breakdown in glycerol-consuming prokaryotic microorganisms"), maximization of YPD/S yield can be achieved during culture under anaerobiosis in glycerol-excess conditions (Menzel et al., 1997a; b).

156

Seraphim Papanikolaou

Graph by Papanikolaou et al., 2000. Figure 8. Representation of the specific rate of 1,3-propanediol production (qPD) versus the dilution rate (filled circles when initial glycerol concentration – S0=90 g dm-3, open circles when S0=60 g dm-3, open squares when S0=30 g dm-3) during an anaerobic single-stage continuous fermentation of industrial glycerol by Clostridium butyricum F2b.

Graph by Papanikolaou et al., 2000. Figure 9. Representation of the specific rate of 1,3-propanediol production (qPD) versus the specific rate of glycerol consumption (qS) (filled circles when initial glycerol concentration – S0=90 g dm-3, open circles when S0=60 g dm-3, open squares when S0=30 g dm-3) during an anaerobic single-stage continuous fermentation of industrial glycerol by Clostridium butyricum F2b.

Microbial Conversion of Glycerol into 1,3-propanediol

157

Graph by Papanikolaou et al., 2000. Figure 10. 1,3-Propanediol volumetric productivity versus flow rate during an anaerobic two-stage continuous fermentation of industrial glycerol by Clostridium butyricum F2b (inlet glycerol concentration S0=90 g dm-3).

Likewise, fed-batch fermentation of K. pneumoniae DSM 2026 under micro-aerobic conditions resulted in almost a 2-fold increase of the volumetric 1,3-propanediol productivity (Chen et al., 2003b) while the spectrum of end-fermentation products and final PD quantity and yield YPD/S were substantially influenced by the employed extra-cellular oxido-reduction potential in batch cultures of K. pneumoniae M5aL (Du et al., 2006). The spectrum of endfermentation products in K. pneumoniae cultures was also critically influenced by the "purity" of the substrate used or the addition of organic compounds (e. g. fumarate) into the medium. Significant product variations were reported when pure glycerol was utilized as the sole substrate compared with the fermentations in which raw glycerol deriving from alkalicatalyzed or lipase-catalyzed hydrolysis was used in fermentations carried out by K. pneumoniae DSM 2026 (Mu et al., 2006). Furthermore, addition of fumarate into the medium (in quantities 5-25 mM) substantially increased GDH, GDHt and PDOR activities, increasing considerably, thus, the final PD quantity produced (cultures with K. pneumoniae strain M5a1, see Lin et al., 2005). In contrast, concerning the fermentation led by Clostridium sp. strains, given that the microorganisms grow under strictly anaerobic conditions, differentiations in the PD production cannot be obtained by introducing micro- aerobic conditions into the reactor, or by changing the employed extra-cellular oxido-reduction potential. In every case, as previously stated, the maximum PD concentrations that can be achieved, regardless of the fermentation conditions, are 50-90 g dm-3. Final PD concentrations of this magnitude or even higher (e. g. up to 130 g dm-3) although considered as extremely high for biological processes and toxic for microbial cells, they are not high enough to ensure satisfactory downstream processing from the fermentation broth (Hao et al., 2006). Maximum achievable concentrations of 1,3-propanediol produced in various culture conditions and fermentation configurations are shown in Table 1.

158

Seraphim Papanikolaou

Table 1. Experimental results of 1,3-propanediol producing microorganisms growing under various fermentation configurations Strain

Substrate

Culture mode

PD (g dm-3)

YPD/S (g g-1)

Reference

Klebsiella pneumoniae DSM 2026 Citrobacter freundii Zu Klebsiella oxytoca Lin Clostridium butyricum DSM 5431

pure glycerol » » pure glycerol

batch » » fed-batch

22.3 28.1 17.0 58.0

0.43 0.53 0.36 0.56

Clostridium butyricum DSM 5431

pure glycerol

batch

56.1

0.51

Biebl et al., (1992)

Citrobacter freundii DSM 30040

pure glycerol

continuous (2 stages)

41.4

0.51

Boenigk et al., (1993)

Clostridium butyricum VPI 3266

pure glycerol

fed-batch

65.0

0.57

Saint-Amans et al., (1994)

Clostridium butyricum E5

raw glycerol

fed-batch

65.6

0.54

Petitdemange et al., (1995)

Clostridium butyricum DSM 5431 Clostridium butyricum mutant 2/2

pure glycerol »

fed-batch »

57.0 70.5

0.53 0.54

Reimann and Biebl (1996)

Klebsiella pneumoniae DSM 2026

pure glycerol

continuous

48.1

0.52

Menzel et al., (1997a)

Klebsiella pneumoniae ATCC 25955 Citrobacter freundii ATCC 8090 Clostridium butyricum CNCM 1211 Clostridium butyricum CNCM 1211 »

pure glycerol » raw glycerol pure glycerol raw glycerol

batch » » batch batch

31.8 31.2 63.4 67.0 65.4

0.45 0.44 0.57 0.52 0.51

Clostridium butyricum F2b

raw glycerol

continuous (1 stage)

48.1

0.55

Papanikolaou et al., (2000)

Clostridium butyricum F2b

raw glycerol

continuous (2 stages)

46.3

0.53

Papanikolaou et al., (2000)

Klebsiella pneumoniae DSM 2026 »

pure glycerol »

fed-batch a »b

56.0 59.5

0.45 0.45

Chen et al., (2003b)

Clostridium butyricum VPI 3266 »

pure glycerol raw glycerol

continuous »

29.7 31.5

0.51 0.50

González-Pajuedo et al., (2004)

Clostridium sp. IK 124 »

pure glycerol raw glycerol

fed-batch »

87.7 80.1

0.54 0.56

Hirschmann et al., (2005)

Clostridium butyricum F2b

raw glycerol

continuous

44.0

0.51

Papanikolaou et al., (2004)

Klebsiella pneumoniae DSM 2026 »

pure glycerol raw glycerol

fed-batch »

61.9 53.0

0.41 0.40

Mu et al., (2006)

Klebsiella pneumoniae DSM 2026 Klebsiella pneumoniae pKM13

pure glycerol »

batch »

49.9 50.0

0.46 0.47

Zheng et al., (2006)

Klebsiella oxytoca LDH3

pure glycerol

fed-batch c

83.6

0.51

Yang et al. (2007)

Homann et al., (1990) Günzel et al., (1991)

Barbirato et al., (1998) Himmi et al., (1999)

Microbial Conversion of Glycerol into 1,3-propanediol

159

Table 1. (Continued)

a

Strain

Substrate

Culture mode

PD (g dm-3)

YPD/S (g g-1)

Reference

Clostridium butyricum F2b

raw glycerol

batch

47.1

0.53

Papanikolaou et al., (2008)

Clostridium butyricum F2b

raw glycerol

continuous (2 stages)

43.5

0.49

Papanikolaou et al., (2008)

Klebsiella pneumoniae TUAC01

pure glycerol

fed-batch

66.3

0.45

Hao et al., (2008)

Clostridium butyricum DSM 5431 »

pure glycerol raw glycerol

batch »

25.6 25.0

0.52 0.50

Rehman et al., (2008)

Anaerobic fermentation; b Micro-aerobic fermentation; c Addition of sucrose as co-substrate.

The process of 1,3-propanediol production in continuous cultures has been successfully described with the aid of models (Zeng et al., 1994; Menzel et al., 1997a). A common kinetic model that described growth and 1,3-propanediol production by both C. butyricum and K. pneumoniae (the most frequently used microorganisms in this type of fermentation) has been proposed by Zeng et al., (1994) as follows: ∗ μmax

μ= 1+

H + KOH + K H H+

CGly ⎛ CGly ⎞ ⎛⎜ CHac ⎞⎟⎛⎜ CHBu ⎞⎟ ⎛⎜ CEtOH ⎞ ⎛⎜ CPD ⎟⎞ ⎜1 − ⎟ 1− 1− 1− ∗ ⎟ 1− ∗ ⎠ ⎝ CPD ⎠ CGly + KS ⎝ C∗Gly ⎠ ⎝ C∗HAc ⎠⎝ C∗HBu ⎠ ⎝ CEtOH

In this equation, the inhibitory effects of substrate (glycerol) and potential metabolic products generated have been assessed (Zeng et al., 1994; Menzel et al., 1997a). In the aforementioned model, C*Gly, C*HAc, C*HBu, C*EtOH, C*PD are the critical inhibitory concentrations of glycerol (substrate) and acetic acid, butyric acid, ethanol and 1,3propanediol (the most frequent products) respectively. The model presented a very satisfactory predictive ability, while a close convergence between the theoretical and the experimental values was observed (Zeng et al., 1994; Menzel et al., 1997a). For pH=7.0 and chemostat experiments of K. pneumoniae, μmax (maximum growth rate) value predicted was 0.67 h-l while KS (saturation constant) for glycerol was 0.25 g dm-3 (Menzel et al., 1997a). The critical inhibitory concentrations for glycerol, 1,3-propanediol, ethanol, acetate and butyrate were 187.6, 60, 16.6, 27.0 and 18.0 g dm-3 respectively (Zeng et al., 1994). Moreover, Papanikolaou and Aggelis (2003) have been attempted to simulate biomass and metabolites production in continuous experiments at constant inlet glycerol concentration (90 g dm-3) and various D imposed by C. butyricum F2b. For the continuous fermentations of C. butyricum F2b, all the above types of product or substrate inhibition models were tested, but their application to the experimental data were revealed unsuccessful (Papanikolaou and Aggelis 2003). It should be mentioned, though, that the microorganism presented noticeable growth without obvious inhibition at media containing glycerol at 90 g dm-3, 1,3-propanediol at 80-84 g dm-3 and butyric acid at 10-12 g dm-3 (Papanikolaou et al., 2000; 2004; 2008),

Seraphim Papanikolaou

160

therefore not any inhibition exerted by the substrate or the products was taken into consideration. Finally, a Contois-type model was chosen (Papanikolaou and Aggelis 2003):

D = Dmax ⋅

S B⋅ X + S

where Dmax was the maximum dilution rate obtained and S and X were the component concentrations (substrate and biomass respectively) at steady-state; B is a constant. The proposed model presented a satisfactory fitting on the experimental data (see graphs in Papanikolaou and Aggelis 2003). Optimized parameter values Dmax and B were determined using non-linear regression analysis and their values obtained were 0.527±0.085 h-1 and 35.627±9.751 g g-1 respectively. A different approach concerning microbial growth of C. butyricum F2b was proposed by Papanikolaou et al., (2004). The microorganism was cultivated in single-stage continuous cultures at a constant dilution rate D (=0.04 h-1) and various inlet glycerol (S0) concentrations (from 20 to 90 g dm-3), and although microbial growth, production of 1,3-propanediol and consumption of glycerol were remarkable in all inlet glycerol concentrations tested, biomass yield on glycerol consumed dropped (from 0.029 to 0.022 g g-1) with increment of S0 (Papanikolaou et al., 2004). In contrast, yields of acetic and butyric acid on glycerol consumed seemed to slightly increase with substrate concentration increase (see graphs in Papanikolaou et al., 2004). Similar biochemical behavior has been observed by the aforementioned strain in continuous cultures and transitory states in which noticeable quantities of PD were introduced into the chemostat (Papanikolaou et al., 2008). Although PD quantities of around 84 g dm-3 were found into the reactor and yield YX/S was somehow decreased, biomass concentration remained constant whereas butyrate and acetate production were slightly stimulated (Papanikolaou et al., 2008). In order to explain the decrease of YX/S with glycerol concentration increase in the inlet medium, it was assumed that at high substrate concentrations (when the 1,3-propanediol pathway was somehow saturated – see Figure 5), the percentage of NAD+ regenerated via the pathway of butyric acid biosynthesis, increased, resulting, thus, to a YX/S decrease, since butyric acid biosynthesis is coupled to carbon losses in CO2 form, through the phosphoroclastic reaction (see Figures 5 and 7). This assumption is supported by the fact that carbon balance calculations conducted without taking into account the CO2 produced, showed the tendency to present somehow lower recoveries at the steady-states with high inlet glycerol amounts in the feed (Papanikolaou et al., 2004). In order to quantify the effect of substrate concentration on the production of biomass and metabolites by C. butyricum, an analytical physiological model based on the release of CO2 through the phosphoroclastic reaction was developed (Papanikolaou et al., 2004). In this model, the coefficients CCO (X) 2

and CCO (p) introduced, represent the yield losses (in the form of CO2) per unit of substrate 2

consumed for the biomass (X) or the metabolites (p) produced, via the phosphoroclastic reaction. Since CO2 formation is stoichiometrically linked with the decarboxylation of pyruvic acid and the formation of CH3COSCoA (see Figures 5 and 7), in the proposed model CCO (X) and CCO (p) are the indices that quantify the relation of biomass and metabolic 2 2 products synthesized with CO2 release at the various inlet glycerol concentrations. Therefore,

Microbial Conversion of Glycerol into 1,3-propanediol

161

the following expression was used to compute the production of biomass at steady-state (Aggelis et al., 1999):

⎡ ⎤ X = ⎢⎛⎜ Y X ⎞⎟ − C CO2 ( X ) ⋅ S 0 ⎥ ⋅ (S 0 − S ) S ⎠ ⎝ max ⎣ ⎦ where X and S are the component concentrations at steady-state, (YX/S)max is the maximum biomass yield obtained and CCO (X) (at g-1) is a coefficient representing the yield losses (in 2

the form of CO2) per unit of S. The specific production rate of the microbial metabolites qp is expressed by the following equation:

qp =

dp 1 p ⋅ D ⋅ = dt X X

where p are the product concentrations (PD, Ac or But) at steady-state. The proposed model was applied in order to quantify the effect of inlet glycerol concentration (S0) on the production of PD, But and Ac, by combining the above equations as follows:

⎤ ⎡ q ⋅ ⎛Y ⎞ ⎢ p ⎜⎝ X S ⎟⎠ max q p ⋅ C CO2 ( p ) ⋅ S 0 ⎥ p=⎢ − ⎥ ⋅ (S 0 − S ) D D ⎥ ⎢ ⎦ ⎣ where CCO (p) 2

(g-1) are the indices that quantify the relation of the metabolites produced

with CO2 release, for the various inlet glycerol concentrations. This equation after reparameterization, yields:

(

p = A ⋅ (S 0 − S ) − B ⋅ S 0 − S 0 ⋅ S 2

)

where

q p ⋅ ⎛⎜ YX ⎞⎟ q p ⋅ CCO ( p ) ⎝ S ⎠ max 2 and B= A= D D Therefore, the production of each metabolic product at the steady-state is affected by S0

(

2

)

concentration; the function − B ⋅ S 0 − S 0 ⋅ S can positively (for the metabolic products positively related to CO2 production) or negatively (for the metabolic products negatively related to CO2 production) affect the production of p. Comparison of predicted and experimentally measured values of biomass (X), 1,3-propanediol (PD), butyrate (But) and acetate (Ac) are given in Figure 11 (a; b) and indicate the predictive ability of the model.

Seraphim Papanikolaou

162

PD exp erimen tal (g /l)

X e xp erimen tal (g /l)

PD p red icted (g /l)

X p red icted (g /l)

3

50

2.5

40 30

PD (g/l)

X (g/l)

2 1.5

20

1

10

0.5 0 20

30

40

50

60

70

0 90 100

80

Inlet glycerol concentration, S (g/l) 0

a. But experimental (g/l)

A c experimental (g/l)

But predicted (g/l)

Ac pred icted (g/l)

15

2

1.5 But (g/l)

Ac (g/l)

10 1 5 0.5

0 20

30

40

50

60

70

0 90 100

80

Inlet glycerol concentration, S (g/l) 0

b. Data by Papanikolaou et al., 2004. Figure 11. Comparison of predicted and experimentally measured values for biomass (X) and 1,3propanediol (PD) (11a) and butyric acid (But) and acetic acid (Ac) (11b) during growth of Clostridium butyricum on raw glycerol in continuous mode at various steady-states. D=0.04 h-1; various inlet glycerol concentrations; pH=7.

Microbial Conversion of Glycerol into 1,3-propanediol

163

4. Conclusions Aspects related with the microbial production of 1,3-propanediol (anaerobic glycerol assimilation, biochemical events related with the 1,3-propanediol biosynthesis, biochemical engineering of the process) were briefly reviewed in this chapter. 1,3-Propanediol is a highvalue added material with a significant number of applications in the chemical and textile industry. Its production in large scale mainly using technical glycerol, waste discharged in tremendous quantities from bio-diesel production plants represents a very important perspective of this interesting microbial conversion (bio-refinery approach). Due to the large number of applications of 1,3-propanediol, future research is very likely to be focused in the discovery – construction of "new – over-producing" strains (mainly with the utilization of genetic engineering techniques), the optimization of cultivation methods for its production with higher yields and productivities and finally the recovery (downstream processing) of 1,3-propanediol from the fermentation medium.

References Abbad-Andaloussi, S; Maginot-Dürr, C; Amine, J; Petitdemange, E; Petitdemange, H. Isolation and charactirization of Clostridium butyricum DSM 5431 mutants with increased resistance to 1,3-propanediol and altered production of acids. Appl. Environ. Microbiol. (1995) 61 4413-4417. Abbad-Andaloussi, S; Guedon, E; Spiesser, E; Petitdemange, H. Glycerol dehydratase activity: the limiting step for 1,3-propanediol production by Clostridium butyricum DSM 5431. Lett. Appl. Microbiol. (1996) 22 311-314. Abbad-Andaloussi, S; Amine, J; Gerard, P; Petitdemange, H. Effect of glucose on glycerol metabolism by Clostridium butyricum DSM 5431, J. Appl. Microbiol. (1998) 84 515522. Aggelis, G; Fakas, S; Melissis, S; Clonis, YD. Growth of Candida boidinii in a methanollimited continuous culture and the formation of methanol-degrading enzymes. J. Biotechnol. (1999) 72 127-139. Aggelis, G. Microbiology and Microbial Technology (in Greek). Eds Stamoulis A. (2007) Ahrens, K; Menzel, K; Zeng, AP; Deckwer, WD. Kinetic, dynamic and pathway studies of glycerol metabolism by Klebsiella pneumoniae in anaerobic continuous culture. III. Enzymes and fluxes of glycerol dissimilation and 1,3-propanediol formation. Biotechnol. Bioeng. (1998) 59 544-552. Andrade, JC; Vasconcelos, I. Continuous cultures of Clostridium acetobutylicum: culture stability and low-grade glycerol utilization. Biotechnol. Lett. (2003) 25 121-125. Barbirato, F; Astruc, S; Soucaille, P; Camarasa, C; Salmon, JM; Bories, A. Anaerobic pathways of glycerol dissimilation by Enterobacter agglomerans CNCM 1210: limitations and regulations. Microbiology (1997) 143 2423-2432. Barbirato, F; Himmi, EH; Conte, T; Bories, A. 1,3-Propanediol production by fermentation: an interesting way to valorize glycerin from the ester and ethanol industries. Ind. Crops Prod. (1998) 7 281-289.

164

Seraphim Papanikolaou

Biebl, H; Marten, S; Hippe, H; Deckwer, WD. Glycerol conversion to 1,3-propanediol by newly isolated clostridia. Appl. Microbiol. Biotechnol. (1992) 36 592-597. Biebl, H; Marten, S. Fermentation of glycerol to 1,3-propanediol: Use of co-substrates. Appl. Microbiol. Biotechnol. (1995) 44 15-19. Biebl, H; Menzel, K; Zeng, AP; Deckwer, WD. Microbial production of 1,3-propanediol. Appl. Microbiol. Biotechnol. (1999) 52 289-297. Boenigk, R; Bowien, S; Gottschalk, G. Fermentation of glycerol in continuous cultures of Citrobacter freundii. Appl. Microbiol. Biotechnol. (1993) 38 453-457. Brossmer, C; Arntz, D. US Patent 6140543 (2000). Chen, X; Xiu, ZL; Wang, JF; Zhang, D; Xu, P. Stoichiometric analysis and experimental investigation of glycerol bioconversion to 1,3-propanediol by Klebsiella pneumoniae under microaerobic conditions. Enzyme Microb. Technol. (2003a) 33 386-394. Chen, X; Zhang, DJ; Qi, WT; Gao, SJ; Xiu, ZL; Xu, P. Microbial fed-batch production of 1,3-propanediol by Klebsiella pneumoniae under micro-aerobic conditions. Appl. Microbiol. Biotechnol. (2003b) 63 143-146. Claisse, O; Lonvaud-Funel, A. Assimilation of glycerol by a strain of Lactobacillus collinoides isolated from cider, Food Microbiol. (2000) 17 513-519. Daniel, R; Gottschalk, G. Growth temperature-dependent activity of glycerol dehydratase in Escherichia coli expressing the Citrobacter freundii dha regulon. F.E.M.S. Microbiol. Lett. (1992) 100 281-286. de Valdez, GF; Ragout, A; Bruno-Bárcena, JM; Diecmann, H; Siňeriz, F. Shifts in pH affect the maltose/glycerol co-fermentation by Lactobacillus reuteri. Biotechnol. Lett. (1997) 19 645-649. Du, C; Yan, H; Zhang, Y; Li, Y; Cao, Z. Use of oxidoreduction potential as an indicator to regulate 1,3-propanediol fermentation by Klebsiella pneumoniae. Appl. Microbiol. Biotechnol. (2006) 69 554-563. Du, C; Zhang, Y; Li, Y; Cao, Z. Novel redox potential-based screening strategy for rapid isolation of Klebsiella pneumoniae mutants with enhanced 1,3-propanediol-producing capability. Appl. Environ. Microbiol. (2007) 73 4515-4521. El-Ziney, MG; Arneborg, N; Uyttendaele, M; Debevere, J; Jakobsen, M. Characterization of growth and metabolite production of Lactobacillus reuteri during glucose/glycerol cofermentation in batch and continuous cultures, Biotechnol. Lett. (1998) 20 913-916. Forage, RG; Lin, ECC. dha systems mediating aerobic and anaerobic dissimilation of glycerol in Klebsiella pneumoniae. NCIB 418. J. Bacteriol. (1982) 151 591-599. Freund, A. Über die Bildung und Darstellung von Trimethylenalkohol aus Glycerin. Monatsh Chem. (1881) 2 636-641 Garai-Ibabe, G; Ibarburu, I; Berregi, I; Claisse, O; Lonvaud-Funel, A; Irastorza, A; Dueñas, MT. Glycerol metabolism and bitterness producing lactic acid bacteria in cidermaking. Internat. J. Food Microbiol. (2008) 121 253-261. Gottsshalk, G; Averhoff, B. European patent EP 0373230 A1 (1990). González-Pajuedo, M; Andrade, JC; Vasconcelos, I. Production of 1,3-propanediol by Clostridium butyricum VPI 3266 using a synthetic medium and raw glycerol. J. Ind. Microbiol. Biotechnol. (2004) 31 442-446.

Microbial Conversion of Glycerol into 1,3-propanediol

165

Günzel, B; Yonsel, S; Deckwer, WD. Fermentative production of 1,3-propanediol from glycerol by Clostridium butyricum up to a scale of 2 m3. Appl. Microbiol. Biotechnol. (1991) 36 289-294. Hao, J; Xu, F; Liu, H; Liu, D. Downstream processing of 1,3-propanediol fermentation broth, J. Chem. Technol. Biotechnol. (2006) 81 102-108. Hao, J; Lin, R; Zheng, Z; Liu, H; Liu, D. Isolation and characterization of microorganisms able to produce 1,3-propanediol under aerobic conditions. W. J. Microbiol. Biotechnol. (2008) 24 1731-1740. Himmi, EH; Bories, A; Barbirato, F. Nutrient requirements for glycerol conversion to 1,3propanediol by Clostridium butyricum. Bioresour. Technol. (1999) 67 123-128. Hirschmann, S; Baganz, K; Koschik, I; Vorlop, KD. Development of an integrated bioconversion process for the production of 1,3-propanediol from raw glycerol waters. Landbauforschung Völkenrode (2005) 55 261-267. Homann, T; Tag, C; Biebl, H; Decker, WD; Schink, B. Fermentation of glycerol to 1,3propanediol by Klebsiella and Citrobacter strains, Appl. Microbiol. Biotechnol. (1990) 33 121-126. Hüsing, B; Angerer, G; Gaisser, S; Marscheider-Weidemann, F. Biotechnologische Herstellung von Wertstoffen unter besonderer Berücksichtigung von Energieträgern und Biopolymeren, Umweltforschungsplan des Bundesministeriums für Umwelt, Naturschutz und Reaktorsicherheit. Texte 64/03, ISSN 0722-185X (2003) (http://www.Umweltbundesamt.org/fpdf-l/2380.pdf) Imandi, SB; Bandaru, VR; Somalanka, SR; Garapati, HR., Optimization of medium constituents for the production of citric acid from byproduct glycerol using Doehlert experimental design. Enzyme Microb. Technol. (2007) 40 1367-1372 Ito, T; Nakashimada, Y; Senba, K; Matsui, T; Nishio, N. Hydrogen and ethanol production from glycerol-containing wastes discharged after biodiesel manufacturing process. J. Biosci. Bioeng. (2005) 100 260-265. Johnson, DT; Taconi, KA. The glycerin glut: Options for the value-added conversion of crude glycerol resulting from biodiesel production. Environmental Prog. (2007) 26 338348. Kim, JW; Park, TJ; Ryu, DDY; Kim, JY. High cell density culture of Yarrowia lipolytica using a one-step feeding process. Biotechnol. Prog. (2000) 16 657-660. Koutinas, AA; Wang, RH; Webb, C. The biochemurgist –Bioconversion of agricultural raw materials for chemical production. Biofuels, Bioproducts and Biorefining (2007) 1 24-38. Lee, SY; Hong, SH; Lee, SH; Park, SJ. Fermentative production of chemicals that can be used for polymer synthesis. Macromol. Biosci. (2004) 4 157-164. Lin, R; Liu, H; Hao, J; Cheng, K; Liu, D. Enhancement of 1,3-propanediol production by Klebsiella pneumoniae with fumarate addition. Biotechnol. Lett. (2005) 27 1755-1759 Malaoui, H; Marczak, R. Influence of glucose on glycerol metabolism by wild-type and mutant strains of Clostridium butyricum grown in chemostat culture. Appl. Microbiol. Biotechnol. (2001) 55 226-233. Meesters, PAEP; Huijberts, GNM; Eggink, G. High cell density cultivation of the lipid accumulating yeast Cryptococcus curvatus using glycerol as a carbon source. Appl. Microbiol. Biotechnol. (1996) 45 575-579.

166

Seraphim Papanikolaou

Menzel K; Zeng, AP; Deckwer, WD. High concentration and productivity of 1,3-propanediol from continuous fermentation of glycerol by Klebsiella pneumoniae. Enzyme Microb. Technol. (1997a) 20 82-86. Menzel K; Zeng, AP; Deckwer, WD. Enzymatic evidence for an involvement of pyruvate dehydrogenase in the anaerobic glycerol metabolism of Klebsiella pneumoniae. J. Biotechnol. (1997b) 56 135-142. Morgunov, I; Kamzolova, S; Perevoznikova, O; Shishkanova, N; Finogenova, T. Pyruvic acid production by a thiamine auxotroph of Yarrowia lipolytica. Proc. Biochem. (2004) 39 1469-1474. Mu, Y; Zhang, D; Teng, H; Wang, W; Xiu, ZL. Microbial production of 1,3-propanediol by Klebsiella pneumoniae using crude glycerol from bio-diesel preparation. Biotechnol. Lett. (2006) 28 1755-1759. Neijssel, OM; Hueting, S; Crabbendam, KJ; Tempest, DW. Dual pathways of glycerol assimilation in Klebsiella aerogenes NCIB 418 ; Their regulation and possible functional significance. Arch. Microbiol. (1975) 104 83-87. Pagliaro, M; Ciriminna, R; Kimura, H; Rossi, M; Della Pina, C. From glycerol to valueadded products. Angewandte Chemie (2007) 46 4434-4440. Papanikolaou, S; Ruiz-Sanchez, P; Pariset, B; Blanchard, F; Fick, M. High production of 1,3propanediol from industrial glycerol by a newly isolated Clostridium butyricum strain. J. Biotechnol. (2000) 77 191-208. Papanikolaou, S; Muniglia, L; Chevalot, I; Aggelis, G; Marc, I. Yarrowia lipolytica as a potential producer of citric acid from raw glycerol, J. Appl. Microbiol. (2002) 92 737744. Papanikolaou, S; Aggelis, G. Lipid production by Yarrowia lipolytica growing on industrial glycerol in a single-stage continuous culture. Bioresour. Technol. (2002) 82 43-49. Papanikolaou, S; Aggelis, G. Modelling aspects of the biotechnological valorization of raw glycerol: production of citric acid by Yarrowia lipolytica and 1,3-propanediol by Clostridium butyricum. J. Chem. Technol. Biotechnol. (2003) 78 542-547. Papanikolaou, S; Fick, M; Aggelis, G. The effect of raw glycerol concentration on the production of 1,3-propanediol by Clostridium butyricum. J. Chem. Technol. Biotechnol. (2004) 79 1189-1196. Papanikolaou, S; Fakas, S; Fick, M; Chevalot, I; Galiotou-Panayotou, M; Komaitis, M; Marc, I; Aggelis, G. Biotechnological valorisation of raw glycerol discharged after bio-diesel (fatty acid methyl-esters) manufacturing process: production of 1,3-propanediol, citric acid and single cell oil. Biomass Bioen. (2008) 32 60-71. Petitdemange, H; Cherrier, C; Raval, G; Gay, R. Regulation of the NADH-ferredoxin oxidoreductase in Clostridia of the butyric group. Biochim. Biophys. Acta (1976) 421 334-347. Petitdemange, E; Dürr, C; Abbad-Andaloussi, S; Raval, G. Fermentation of raw glycerol to 1,3-propanediol by new strains of Clostridium butyricum, J. Ind. Microbiol. (1995) 15 498-502. Pflugmacher, U; Gottschalk, G. Development of an immobilized cell reactor for the production of 1,3-propanediol by Citrobacter freundii. Appl. Microbiol. Biotechnol. (1994) 41 313-316.

Microbial Conversion of Glycerol into 1,3-propanediol

167

Rehman, A; Saman Wijesekara RG; Nomura, N; Sato, S; Matsumura, M. Pre-treatment and utilization of raw glycerol from sunflower oil biodiesel for growth and 1,3-propanediol production by Clostridium butyricum. J. Chem. Technol. Biotechnol. (2008) 83 10721080. Reimann, A; Biebl, H. Production of 1,3-propanediol by Clostridium butyricum DSM 5431 and product tolerant mutants in fed-batch cultures: feeding strategy for glycerol and ammonium. Biotechnol. Lett. (1996) 18 827-832. Ruch, F; Lin, ECC. Independent constitutive expression of the aerobic and anaerobic pathways of glycerol catabolism in Klebsiella aerogenes. J. Bacteriol. (1975) 124 348352. Rymowicz, W; Rywińska, A; Źarowska, B; Juszczyk, P. Citric acid production from raw glycerol by acetate mutants of Yarrowia lipolytica. Chem. Pap. (2006) 60 391-394. Rymowicz, W; Rywińska, A; Źarowska, B. Biosynthesis of citric acid from crude glycerol by Yarrowia lipolytica in repeated-batch cultivations. J. Biotechnol. (2007) 131 S149-S150. Rymowicz, W; Rywińska, A; Gładkowski W. Simultaneous production of citric acid and erythritol from crude glycerol by Yarrowia lipolytica Wratislavia K1. Chem. Pap. (2008) 62 239–246. Saint-Amans, S; Perlot, G; Goma, G; Soucaille, P. High production of 1,3-propanediol from glycerol by Clostridium butyricum VPI 3266 in a simply controlled fed-batch system, Biotechnol. Lett. (1994) 16 831-836. Saint-Amans, S; Girbal, L; Andrade, J; Ahrens, K; Soucaille, P. Regulation of carbon and flow in Clostridium butyricum VPI 3266 grown on glucose–glycerol mixtures. J. Bacteriol. (2001) 183 1748–1754. Schütz, H; Radler, F. Anaerobic reduction of glycerol to propanediol 1,3 by Lactobacillus brevis and Lactobacillus buchneri. System. Appl. Microbiol. (1984) 5 169-178. Slininger, PJ; Bothast, J. Optimizing aerobic conversion of glycerol to 3hydroxypropionaldehyde. Appl. Envir. Microbiol. (1985) 50 1444-1450. Streekstra, H; Teixeira de Mattos, MJ; Neijssel, OM; Tempest DW. Overflow metabolism during anaerobic growth of Klebsiella aerogenes NCTC 418 on glycerol and dihydroxyacetone in chemostat culture. Arch. Microbiol. (1987) 147 268-275. Tong, I; Cameron, D. Enhancement of 1,3-propanediol production by cofermentation in Escherichia coli expressing genes from Klebsiella pneumoniae dha regulon genes. Appl. Biochem. Biotechnol. (1992) 34/35 149-159. Veiga-da-Cunha, M; Foster, MA. 1,3-Propanediol NAD+ oxidoreductases of Lactobacillus brevis and Lactobacillus buchneri. Appl. Environ. Microbiol. (1992) 58 2005-2010. Witt, U; Muller, RJ; Augusta, J; Widdecke, H; Deckwer, WD. Synthesis, properties and biodegradability of polyesters based on 1,3-propanediol. Macromol. Chem. Phys. (1994) 195 793-802. Xiu, ZL; Song, BH; Wang, ZT; Sun, LH; Feng, EM; Zeng, AP. Optimization of dissimilation of glycerol to 1,3-propanediol by Klebsiella pneumoniae in one- and two-stage anaerobic cultures. Biochem. Eng. J. (2004) 19 189-197. Yang, G; Tian J; Li, J. Fermentation of 1,3-propanediol by a lactate deficient mutant of Klebsiella oxytoca under microaerobic conditions. Appl. Microbiol. Biotechnol. (2007) 73 1017–1024.

168

Seraphim Papanikolaou

Zeng, AP; Ross, A; Biebl, H; Tag, C; Deckwer, WD. Multiple product inhibition and growth modeling of Clostridium butyricum and Klebsiella pneumoniae in glycerol fermentation. Biotechnol. Bioeng. (1994) 44 902-911. Zeng, AP; Biebl, H. Bulk chemicals from biotechnology: the case of 1,3-propanediol production and the new trends. Adv. Biochem. Eng. Biotechnol. (2002) 74 239–259. Zheng, P; Wereath, K; Sun, J; van den Heuvel, J; Zeng, AP. Overexpression of genes of the dha regulon and its effects on cell growth, glycerol fermentation to 1,3-propanediol and plasmid stability in Klebsiella pneumoniae. Proc. Biochem. (2006) 41 2160-2169

In: Microbial Conversions of Raw Glycerol Editor: George Aggelis

ISBN 978-1-60692-392-4 © 2009 Nova Science Publishers, Inc.

Chapter XI

Capital and Manufacturing Cost Estimation of a Bioprocess Converting Raw Glycerol to 1,3-Propanediol Anastasia A. Apostolakou1, Ioannis K. Kookos1 and Apostolis A. Koutinas2 1. Department of Chemical Engineering, University of Patras, Patras GR26504, Greece 2. Department of Food Science and Technology, Agricultural University of Athens, Athens, Greece

Abstract The aim of this chapter is to present the calculations of the capital cost and the total manufacturing cost of a bioprocess that is used to convert raw glycerol to 1,3propanediol. To this end a representative process flow diagram is developed based on well known heuristics. Data from the literature are used to solve the material and energy balances. The process equipment is then designed and the fixed capital cost estimated. Finally, the total manufacturing cost is estimated. The results of the economic analysis are particularly helpful in identifying research direction that will improve process economics and can contribute in transforming the results of fundamental research into successful industrial projects.

1. Introduction Process design can be defined as the conceptual work done prior to the construction of a new production facility or the conceptual work done prior to the expansion/retrofitting of an existing process plant. Process synthesis is the selection and arrangement of a set of unit operations (such as reactors, distillation columns, compressors, etc.) in such a way so as to produce the desired products at an acceptable cost and at the same time satisfy quality, safety and environmental constraints. Process analysis, an invaluable tool in process synthesis, is the

170

Anastasia A. Apostolakou, Ioannis K. Kookos and Apostolis A. Koutinas

evaluation and comparison of alternative feasible solutions. Synthesis is usually followed by analysis in order to improve the results of the synthesis procedure and develop better designs. A quantitative criterion that is commonly used in both process synthesis and analysis is process economics. Economic evaluation requires, however, integration of knowledge from many different disciplines and can be carried out at various levels of detail. Order-of-Magnitude estimate of process economics is usually based on historical cost data and it is the less accurate but the least expensive. Project planning estimate, which is based on knowledge of major equipment items, and scope estimate are the most commonly estimates of process economics that are employed in chemical and petrochemical industry. Project planning and scope estimates can be completed within weeks and are normally used to plan research and development and to estimate project budgets. Detailed Engineering and contractor’s estimate studies, which are quite accurate but can be extremely expensive, are usually performed by highly specialized and experienced engineering and construction companies. These highly specialized companies are normally hired in order to construct plants for new products whose economics appears to be promising based on the results of preliminary studies (scope or project planning estimates). Biodiesel production continuous to increase worldwide as a result of the recent unprecedented increases in the crude oil prices and the increased concerns about the environmental impact of the use of transportation fuels. The byproducts of the biodiesel production activities, which are the residues of the oil extraction processes and glycerol, offer new opportunities in fundamental as well as in applied research. The aim is to demonstrate the technical feasibility of new processes and to evaluate their economic potential so as to identify promising alternatives for successful commercialization or to initiate new research projects. Until very recently, purified glycerol (or glycerin) was considered as a high-value chemical with prices as high as $2/kg. Glycerol is produced together with biodiesel in the transesterification reaction that transforms oil and methanol to biodiesel. More specifically, 1 kg of glycerol is produced for every 10 kg of biodiesel produced. As a result biodiesel production has created a glut in the glycerol market causing sharp decrease in the price of glycerol, which is now estimated to be around $0.1/kg. In addition, biodiesel production units are facing increasing production cost due to the fact that glycerol, a major income generator, has become a waste stream. This is due to the fact that the glycerol produced by a biodiesel production facility contains significant amounts of salts, heavy metals and water and its purification cost is estimated larger than current glycerol prices. As biodiesel producers are facing increasing production costs conversion of crude, unrefined glycerol to high value-added products presents a unique opportunity to increase revenues, expand product market and improve significantly their long-term viability. Existing consumers of glycerol such as the food industry and the cosmetics and pharmaceuticals industry cannot absorb the excess in the glycerol production and the development of new market is more pressing than ever. This will, however, necessitate the development of novel research projects and the successful commercialization of their outcomes. 1,3-propanediol is one of the chemical that can be produced from raw glycerol using biotechnological methods. The interest in producing 1,3-propanediol from glycerol was very limited in the past mainly due to high prices of glycerol. However, interest in its production

Capital and Manufacturing Cost Estimation of a Bioprocess…

171

has now been renewed for reasons that have already been mentioned and are related to the availability of glycerol at large quantities and low prices. 1,3-propanediol is one of the main monomers in the production of high performance polyesters such as PTT (poly-trimethylene terephthalate) and an intermediate in the production of polyester, polyether and polyurethane. 1,3-propanediol has also found a number of niche applications. Shell (Corterra ®, industrial fibres and engineering plastics) and Dupont (SoronaTM fibre) have a strong interest in improving production technologies of 1,3-propanediol. Dupont has invested significant resources in a biotechnological method of producing 1,3-propanediol from glucose. 1,3propanediol demand for PTT production only is expected to reach 220,000 t per year with prices well above $1/kg.

2. Biotechnological Conversion of Raw Glycerol to 1,3-Propanediol The dissimilation of raw glycerol by many microorganisms under anaerobic condition for the production of 1,3-propanediol is known for more than 80 years (Mickelson and Werkman, 1940). Deckwer (1995), Zeng (1996) and Menzel et al., (1997) give critical reviews of the early attempts to transform glycerol to 1,3-propanediol together with the analysis of the maximum yields and productivities that can be achieved. The interested reader is referred to these papers for a comprehensive introduction and review of the research area. More recent accounts on the progress of glycerol transformation to 1,3-propanediol is given by Zeng and Biebl (2002) and Papanikolaou et al., (2008). These authors have identified that in order to optimize the process one should (1) prevent by-product formation (such as ethanol, acetic acid, butyric acid, etc.) (2) increase final 1,3-propanediol concentration and (3) increase productivity (as expressed in g of 1,3-propanediol produced per L, per h). Zeng and Biebl (2002) report that a maximum concentration of 60-70 g/L for 1,3propanediol can be achieved using wild-type strains while using fed-batch fermentation the concentration can be increased to more than 85 g/L. For continuous cultures, however, the maximum values that have been reported are 35-48 g/L. Papanikolaou et al., (2000) have reported the results of continuous experiments with an initial glycerol concentration of 90 g/L for several dilution rates ranging from 0.02 h-1 to 0.21 h-1. 1,3-propanediol concentration decreases as dilution rate increases with a maximum at 0.02 h-1 of 48.1 g/L and a yield of 0.55 g/g. The specific rate of 1,3 propanediol production, however, increases as dilution rate increases approaching a maximum 0f 3.2 g/(g⋅L). Improved productivities were observed using a two-stage continuous system. These results are in close agreement with previous results by Menzel et al., (1997) and Reimann et al., (1998). More recent results by Papanikolaou et al., (2004) support previous findings as for an initial glycerol concentration of 90 g/L the 1,3-propanediol concentration achieved is 44 g/L at a dilution rate of 0.04 h-1.

Anastasia A. Apostolakou, Ioannis K. Kookos and Apostolis A. Koutinas

172

3. Methodology for Manufacturing Cost Estimation The preliminary economic evaluation of a process for the producing a biological product involves the estimation of fixed capital cost, estimation of operating costs and profitability analysis. The capital investment for a new plant includes three main items: • • •

fixed capital investment (FCI), working capital, and startup and validation cost.

The FCI for small biotechnology facilities is usually in the range of M$10-100. For preliminary design purposes, the various items of FCI are estimated based on the total F.O.B. equipment purchase cost using several multipliers as it was proposed by Gurthrie (1969). Detailed definitions of the various cost items can be found in traditional process design textbooks (see for example Peters and Timmerhaus and West, 2003). For preliminary cost estimates the FCI of a plant is a multiple of its equipment purchase cost. The equipment F.O.B. purchase cost can be obtained from vendor quotations or can be estimated from published data. Vendor quotations are usually avoided for preliminary estimation as they can be time-consuming. Published data on equipment purchase cost are usually extrapolations of known cost for given characteristic size(s) of the equipment. The simplest and most frequently used equation is known as the scaling law:

⎛X C = C ⎜⎜ b ⎝ Xa 0 b

0 a

⎞ ⎟⎟ ⎠

n

(1)

where C0 is the equipment F.O.B. purchase cost, X the characteristic size and the exponent n obtains values close to 0.6 (normally between 0.2 and 0.9) and is characteristic to each type of equipment. Cost data on logarithmic coordinates tend to fall on straight lines and this fact justifies the use of the scaling law for extrapolation purposes. In many cases data deviate from linearity and more complex, polynomial equations are normally used. As the price of equipment changes with time due to market conditions there is a need to capture the time-dependent variation of equipment cost data and this is achieved through the use of Plant Cost Indices such as the CE index and the M&S index which are published monthly by Chemical Engineering magazine. The M&S index is used, for instance, to update equipment cost data according to the following equation

⎛ M & St 2 C t02 = C t01 ⎜⎜ ⎝ M & St1

⎞ ⎟ ⎟ ⎠

(2)

where t2 and t1 are referred to specific time instances. Other factors that can affect equipment cost are process conditions (pressures above 15 bar or below 0.5 bar or temperatures above approximately 300-400 OC) and material of

Capital and Manufacturing Cost Estimation of a Bioprocess…

173

construction. A stainless steel vessel is 2-3 times more expensive to build than a carbon steel tank of the same dimensions. The cost increase due to these factors are taken into account using specific multipliers for material (Fm), conditions (Fp) or equipment type (Ft). The installed equipment cost or bare module cost is then obtained through equations of the following general form

⎛ M & St ⎞ ⎟ f (Fm , F p , Ft ) C t01 C BM (t , X , m, p, t ) = ⎜⎜ ⎟ ⎝ M & St1 ⎠

⎛ X ⎜⎜ ⎝ Xa

⎞ ⎟⎟ ⎠

n

(3)

For the case of shell-and-tube heat exchangers, for instance, using data supplied by Gurthrie (1969) the following equation can be used to calculate the installed equipment cost

C BM , S &T − HE

⎛ M & St = ⎜⎜ ⎝ 280

0.65 ⎞ A ⎟ ⋅ Fm (F p + Ft ) ⋅ 9000 ⋅ ⎛⎜ ⎞⎟ ⎟ ⎝ 93 ⎠ ⎠

(4)

where A is the heat transfer area in m2 and the correction factors (Fm, Fp, Ft) can be found in Gurthrie (1969). In order to calculate the FCI, in addition to bare module costs, the contingency and fee costs as well as the auxiliary facilities costs must be taken into consideration. Turton et al., (2003) propose using the following equation for calculation the FCI based on the total installed equipment cost (or bare module cost)

⎧1.18∑ C BM ,i , for expansions/alterations ⎪ i FCI = ⎨ ⎪1.68∑ C BM ,i , for new facilities (grass roots) i ⎩

(5)

The operating cost to run a chemical or biochemical plant consists of all costs associated with raw materials, labor, utilities, waste disposal, overhead, etc. Dividing the operating cost by the production rate yields the unit production cost (in $/unit of product). Cost items are divided into fixed and variable. Fixed costs are those that are independent of the production rate and the clearest example of a fixed cost is depreciation, which is part of the equipmentdependent cost. The clearest case of a variable cost is the cost of raw materials. Raw materials cost (CRM) accounts for the cost of all fermentation media, recovery chemicals, and cleaning materials. For commodity biochemicals, such as ethanol, the raw materials cost consists mainly of the cost of fermentation media while for high value products, the buffers used for product recovery and equipment cleaning can be a major part of the raw materials cost. The unit cost of raw material can be obtained from vendor quotations or published data (in specialized publications such as the Chemical Market Reporter). Consumables costs, i.e. the cost of items that need to be replaced periodically (such as membranes, chromatography resins, activated carbon, etc.) are normally considered as part of the raw materials costs.

174

Anastasia A. Apostolakou, Ioannis K. Kookos and Apostolis A. Koutinas

Labor cost (COL) is estimated based on the total number of operators, which in turn is calculated by summing up the operator requirements of the various process units. The need for direct labor can be estimated based on historical data and experience from similar units. Batch processes have increased demand in operators. The cost of waste treatment (CWT) accounts for the treatment of wastewater and the disposal of solid and hazardous materials and is normally among the minor cost for most biotechnological facilities. Utilities cost (CUT) accounts for steam consumption (lps, mps or hps), process water or cooling water and refrigerant consumption, electricity, etc. The required amounts are calculated through the material and energy balance calculations with acceptable accuracy. Fermentors can be major consumers of electricity. Product purification may require significant amounts of steam (ethanol purification). Electricity costs around $0.1/kWh, steam is around $6-10/t, and refrigerants around $0.025-0.25/MJ of heat removed. Clean steam, which is mainly used for sterilizing equipment as part of equipment cleaning, is far more expensive. Purified water used for buffer preparation and equipment cleaning is normally classified as a utility. In order to calculate the total cost of manufacture (COM), all direct costs, indirect costs and general expenses need to be calculated. Direct cost (DC) consists of all direct cost mentioned above as well as supervisory labor, maintenance related costs, laboratory charges and fees paid for royalties and patents. Turton et al., (2003) propose using the following equation for calculation the DC based on the estimation of the fixed capital investment, operating labor cost, cost of utilities and raw materials costs and cost of waste treatment

DC = C RM + CUT + CWT + 1.33C OL + 0.07 FCI + 0.03COM

(6)

They have also proposed similar equations for the calculation of the fixed costs (FC) and the general expenses (GE)

FC = 0.708C OL + 0.17 FCI

(7)

GE = C RM + CUT + CWT + 0.177C OL + 0.009 FCI + 0.16COM

(8)

Using the last three equations the following equation has been derived for calculating the COM

COM = 1.23(C RM + CUT + CWT ) + 2.73C OL + 0.28 FCI

(9)

4. Process Flow Diagram Description The process flow diagram developed in this study is shown in Figure 1 and is based on the continuous experiments of Papanikolaou et al., (2004) where the conversion of raw glycerol to 1,3-propanediol by Clostridium butyricum is investigated. According to the

Capital and Manufacturing Cost Estimation of a Bioprocess…

175

aforementioned publication at a dilution rate of D=0.04 h-1, temperature of 33 OC and an inlet glycerol concentration of 90 g/L an almost complete (96.44%) conversion of glycerol can be achieved in a continuous fermenter. The main products are 1,3 propanediol (44 g/L), butyric acid (11.1 g/L) and acetic acid (1.2 g/L). The biomass concentration at the steady state is 1.95 g/L. The operation of the fermenter in continuous mode is preferred to the batch (or fedbatch) mode due to the fact that the experiments in batch system require 40-48 h for the complete glycerol consumption. As a result the volume of the fermenters will be significantly larger for operation in batch mode.

Figure 1. Process flow diagram for the 1,3-propanediol production from raw glycerol.

It is assumed that 30,000 t/y of raw glycerol in a 80 wt% solution in water is available as a byproduct of the biodiesel production activity. This is a realistic number for Greece as the biodiesel production is expected to reach the amount of 300,000 t/y by the year 2010. We consider a centralized unit that collects at a minimal cost all glycerol produced in Greece as a byproduct of the biodiesel production. According to the experiments contacted by Papanikolaou et al., (2004) the yield to 1,3-propanediol is 0.51 g/g. It follows immediately that the maximum amount of 1,3-propanediol that can be produced is approximately 15,300 t/y.

Anastasia A. Apostolakou, Ioannis K. Kookos and Apostolis A. Koutinas

176

Raw glycerol is stored in tank TT-301 which has a storage capacity of 3000 m3. This figure corresponds to raw glycerol which is required to support production for four weeks. The raw glycerol is pumped and mixed with process water so as to be diluted to 90 kg of glycerol per m3. The combined stream has a volumetric flowrate of approximately 40 m3/h and is split to five streams and fed to five fermenters (R-301 – R-305). In order to achieve a dilution rate of D=0.04 h-1, each fermenter has a total volume of 250 m3 (80% full at nominal operation). The product stream contains all unreacted glycerol, biomass, the required product 1,3-propanediol and byproducts such as butyric and acetic acid. The product stream is fed to a centrifuge (FF-301) so as to remove all insoluble material including cell mass. Soluble material such as soluble proteins are removed in a series of ultrafiltration membrane units (UF-301, UF-302 and UF-303) of decreasing molecular weight cut-off. The permeate is fed to a double-effect evaporation unit (FE-301 and FE-302) where most of the water (80%) is removed. The concentrated product is fed to a distillation column (T-301) that operates under vacuum where lights components such as water are removed as distillate. In the bottoms product stream 1,3-propanediol is the most volatile component and is recovered as the distillate of a second distillation column (T-301) which also operates under vacuum. The 1,3-propanediol purity that is achieved is around 99.9 mole %. The bottom’s product stream contains all heavy components such as residual glycerol. The retentate of the ultrafilters as well as the distillate of column T-301 are considered as waste streams.

5. Equipment Cost Estimation The installed equipment cost is calculated using the following equations (all cost data are obtained from Blanch and Clark, 1997, or Kookos, 2008, unless otherwise stated): •

Crude glycerol storage tank

⎧ ⎫ ⎛ M & St ⎞ Cstor.tank = 2.4 ⎜ ⎟ exp⎨ 9.37 + lnV (0.045355lnV − 0.1045)⎬, in $ at year t (10) ⎝ 600 ⎠ ⎩ ⎭ which can be used for volumes V: 80 m3≤V≤45000 m3. •

Fermenters

⎛ M & St ⎞ 0.74 Cferm = 1.5 ⎜ ⎟ 33183 V , in $ at year t ⎝ 800 ⎠ which can be used for fermenter volumes V: 50 m3≤V≤250 m3. •

Centrifuge

(11)

Capital and Manufacturing Cost Estimation of a Bioprocess…

⎛ M & St ⎞ 0.73 Ccentrf = 1.6 ⎜ ⎟ 10070 V& , in $ at year t 800 ⎝ ⎠

177

(12)

which can be used for inlet volumetric flowrates between 4 and 70 m3/h. •

Ultrafilters (not including membrane)

⎛ M & St ⎞ 0.89 Cultraf = 1.42 ⎜ ⎟ 1468 A , in $ at year t ⎝ 800 ⎠

(13)

which can be used for membrane areas: 10 m2≤A≤1000 m2. •

Forced circulation evaporators (Peters, Timmerhaus and West, 2003)

⎛ M & St ⎞ 0.523 Cultraf = 2.5 ⎜ ⎟ 120000 A , in $ at year t ⎝ 1124 ⎠

(14)

which can be used for heat transfer areas: 10 m2≤A≤1000 m2. •

Distillation columns

⎛ M & St ⎞ 0.8 1.05 Cdistil = 4.23 ⎜ ⎟ 3900 H D , in $ at year t 1163 ⎝ ⎠

(15)

which can be used for columns which have a diameter (D) less than 3 m and a height (H) less than 30 m. The ratio H/D must also be less than 30. •

Heat exchangers

⎛ M & St ⎞ 0.59 Cheat.exch = 3.29 ⎜ ⎟ 1500 A , in $ at year t ⎝ 1000 ⎠

(16)

which can be used for heat transfer areas A: 10 m2≤A≤800 m2. Based on the equation above the total installed equipment cost is estimated to be of the order of M$34 while the fixed capital investment (FCI) is estimated to be M$40. The analysis is presented in Table 1. As can be seen from Table 1 the equipment cost is dominated by the cost of the five fermenters which accounts for the 61.8% of the total installed equipment cost. The centrifuge and two evaporators also contribute significantly to the total equipment cost (10% and 23.8%, respectively).

Anastasia A. Apostolakou, Ioannis K. Kookos and Apostolis A. Koutinas

178

Table 1. Fixed capital investment cost calculation summary UNIT

CHARACTERISTIC SIZE

TT-301 R-301 – R-305 FF-301

Crude glycerol storage tank Fermenters Centrifuge

UF-301 – UF-303 FE-301 – FE-302

Membrane modules (UF) Evaporators

T-301

Distillation column

T-302

Distillation column

E-301, E-302,

E-303 E-304

Heat exchangers - reboilers Heat exchangers condensers

Volume : 3000 m3 Volume : 5×250 m3 Input volumetric flowrate : 39.5 m3/h Membrane area :3×800 m2 Heat transfer area : 70 m2, 140 m2 Column height : 8 m, diameter : 2.8 m Column height : 6.1 m, diameter : 1m Heat transfer area : 50 m2, 3 m2 Heat transfer area : 35 m2, 5 m2 TOTAL INSTALLED EQUIP COST FIXED CAPITAL INVESTMENT (FCI)

BARE MODULE COST (M$) 0.420 21.200 0.330 3.370 8.100 0.280 0.075 0.110 0.020 34

M$ 40

Table 2. Calculation of operators requirements based on the Ulrich’s method Equipment Fermenters Centrifuge Evaporator Membrane system Process vessel Heat exchangers

# 5 1 2 3 2 4

Operators/shift/unit 0.2-0.3 0.1-0.2 0.1 0.2 0.2-0.5 0.1 TOTAL

Operators/shift 1.25 0.2 0.2 0.6 1 0.4 3.75

6. Estimation of the Operating Costs and Cost of Manufacture The operating costs consist of the raw material costs, direct labor cost and supervision and utilities cost. In this study, it has been assumed that the cost of the raw material (glycerol) is negligible (COL≈0). The direct labor cost has been calculated based on the method proposed by Ulrich (Peters et al., 2003). The calculation summary is presented in Table 2. It is calculated that 4 operators are required per shift giving an overall (4·4.5=) 18 operators. A cost of $30000/operator/year has been assumed resulting in an overall direct labor cost (COL) of M$0.54 per year.

Capital and Manufacturing Cost Estimation of a Bioprocess…

179

Table 3. Utilities cost calculation summary Utility

Amount required

Unit cost

High pressure steam Electricity Membrane replacement Cooling water Process water

36.6 t/h 1980 kW 2400 m3 320 t/h 35.6 t/h

16.7 $/t 0.1 $/kWh 400 $/m2 0.015 $/t 0.07 $/t TOTAL (CUT)

Annual cost (M$) 5.07 1.64 0.32 0.04 0.02 7.1

The utilities cost is calculated based on the material and energy balances for the process under investigation. The utilities cost analysis summary is presented in Table 3. As it can be observed from Table 3 the high pressure steam accounts for 71.5% of the overall utilities cost while the cost of electricity accounts for 23.1% of the overall utilities cost. For the waste treatment facilities a fixed amount of M$0.1 per year was assumed (CWT). For calculating the cost of manufacture (COM) or total annual cost (TAC) Turton et al., (2003) propose the following equation

TAC = 1.23(CUT + CWT + CRM ) + 2.73COL + 0.28 FCI Application of the last equation gives a total annual cost of M$21.53 or a cost of $1.4/kg of 1,3-propanediol produced. It is important to emphasize that this is an optimistic estimation of the actual cost as only the major parts of the equipment and utilities have been considered. The unit cost of $1.4/kg of 1,3-propanediol is clearly large and prohibitive for the successful commercialization of the underlying research ideas. However, the analysis presented can be quite useful in identifying the areas where potential research breakthroughs will have a significant impact on the overall process economics. First of all it is important to observe that the cost of the fermenters contribute to the unit cost more than $0.45/kg of 1,3-propanediol. The second major contribution is that of the high pressure steam cost which accounts for $0.41/kg of 1,3-propanediol. Electricity cost accounts for $0.13/kg of 1,3-propanediol. These three element together are responsible for more than 70% of the overall unit cost of producing 1,3-propanediol from raw glycerol. It is therefore obvious that research should be directed towards decreasing the cost elements associated with the fermentation units and energy consumption. This can be achieved by increasing the initial glycerol concentration, or equivalently increasing the concentration of 1,3-propanediol in the continuous fermenters. If, for instance, an initial concentration of 130 g/L of glycerol is used in the calculations summarized in this chapter a significant decrease in the production cost is observed which is of the order of 34%. In this case the unit cost of 1,3-propanediol will be around 0$.94/kg which will make the commercialization of the process attractive considering the estimated size of the 1,3-propanediol market in the near future. In addition, classical process system optimization, and more particular process integration, can reduce the cost significantly by thermal integration. This is considered feasible as a significant part of the unit production cost has been attributed to utilities consumption and more particular steam

180

Anastasia A. Apostolakou, Ioannis K. Kookos and Apostolis A. Koutinas

and electricity consumption. This direction will be the main focus of our research efforts in the area.

7. Conclusion In this chapter the economic analysis of a continuous process for 1,3-propanediol production has been presented in detail. A skeleton process flow diagram has been developed and the equipment has been dimensioned and its installed cost has been estimated. The unit cost of production of 15000 t/y 1,3-propanediol has been estimated to $1.4/kg which is considered prohibitive for the successful commercialization of the process. However, future direction of research, which can reduce the unit production significantly, have been identified and analyzed.

References Blanch, HW.; Clark, D.S. Biochemical Engineering, NY:Marcel Dekker, Inc.; 1997. Deckwer, WD. Microbial conversion of glycerol to 1,3-propanediol, FEMS Microbiology Reviews, 1995, 16, 143-149. Guthrie, KM. Capital Cost Estimation, Chemical Engineering, 1969, 76,114-137. Johnson, DT; Taconi, KA. The Glycerin glut: Options for the value-added conversion of crude glycerol resulting from biodiesel production, Environmental Progress, 2007, 26(4), 338-348. Kookos, IK. Introduction to Process Synthesis and Design, Thessalonika: Tziola Publications; 2008. (in Greek) Menzel, K; Zeng, AP; Deckwer, WD, High concentration and productivity of 1,3propanediol from continuous fermentatiobn of glycerol by Klebsiella pneumoniae, Enzyme and Microbial Technology, 20, 82-86, 1997. Mickelson, MN; Werkman, CH. The dissimilation of glycerol by coli-aerogenes intermediates, Journal of Bacteriology, 1940, 39(6), 709-715. Papanikolaou, S; Ruiz-Sanchez, P; Pariset, B; Blanchard, F; Fick, M. High production of 1,3propanediol from industrial glycerol by a newly isolated Clostridium butyricum strain, Journal of Biotechnology, 2000, 77, 191-208. Papanikolaou, S; Fick, M; Aggelis, G. The effect of raw glycerol concentration on the production of 1,3-propanediol by Clostridium butyricum, Journal of Chemical Technology and Biotechnology, 2004, 79, 1189-1196. Papanikolaou, S; Fakas, S; Fick, M; Chevalot, I; Galiotou-Panayotou, M; Komaitis, M; Marc, I; and Aggelis G. Biotechnological valorization of raw glycerol discharged after biodiesel manufacturing process: Production of 1,3-propanediol, citric acid and single cell oil., Biomass and Bioenergy, 2008, 32, 60-71. Peters, MS; Timmerhaus, KD; West, RE. Plant Design and Economics for Chemical Engineers, 5th ed., NY: McGraw Hill, 2003.

Capital and Manufacturing Cost Estimation of a Bioprocess…

181

Reimann, A; Biebl, H; Deckwer, WD. Production of 1,3-propanediol by Clostridium butyricum in continuous culture with cell recycling, Applied Microbiology and Biotechnology, 1998, 49, 359-363. Turton, R; Bailie, RC; Whiting, WB; Shaeiwitz, JA. Analysis, Synthesis, and Design of Chemical Processes, NJ: Prentice Hall International Series, 2003. Zeng AP; Biebl, H. Bulk chemicals from biotechnology: The case of 1,3-propanediol production and the new trends, Advances in Biochemical Engineering/Biotechnology, 2002, 74, 239-259.

Index A Aβ, 177 acceptor, 141, 142, 145 acceptors, 143 accounting, 47 accuracy, 174 acetate, viii, 5, 7, 19, 20, 22, 30, 32, 43, 53, 81, 105, 112, 113, 146, 150, 152, 155, 159, 160, 161, 167 acetic acid, 73, 76, 142, 143, 145, 146, 147, 148, 151, 159, 162, 171, 175, 176 acetone, 77, 89, 151 acidic, 68 activated carbon, 173 activation, 15, 62 activators, 15 active transport, 10, 11, 15, 17, 28 adaptability, 54 adaptation, 114 additives, 45, 102, 120 adenine, 101 adenosine, 101, 149 adenosine triphosphate, 101, 149 adhesion, 68 adhesions, 82 adhesives, 139 adjustment, 4 ADP, 101 aerobic, 10, 12, 13, 29, 70, 105, 111, 140, 141, 143, 148, 154, 157, 159, 164, 165, 167 agar, 33, 54, 88, 128 agent, 43, 75, 76, 78, 79, 105 agents, 63, 77, 78, 82 aging, 69

aging process, 69 agricultural, 75, 82, 110, 165 agriculture, 32, 71 aid, x, 105, 137, 159 air, 144 albumin, 127 alcohol, 2, 26, 33, 68 alcohols, 20, 26, 39, 48, 71, 77 algae, 7, 26, 43, 44, 45, 47, 49, 51, 52, 53, 55, 62, 63, 86, 103, 122, 123, 126 Algal, 46, 47, 48, 51, 52, 53 algorithm, 90 alkali, 2, 3, 46, 157 alkaline, 78 alkane, 127 alpha, 17 alternative, vii, viii, 1, 5, 12, 26, 39, 41, 42, 44, 59, 60, 78, 80, 81, 87, 96, 111, 127, 138, 139, 170 alternatives, 66, 170 American Heart Association, 42, 60 amines, 105 amino, x, 29, 110, 125, 127, 128, 131, 132, 133 amino acids, x, 29, 110, 125, 127, 128, 131, 132, 133 ammonia, 67, 72, 75 ammonium, 53, 75, 88, 110, 115, 124, 167 ammonium salts, 110 ammonium sulphate, 115, 124 amorphous, 70, 80 amylase, 140 anaerobes, 10 anaerobic, x, 1, 4, 6, 70, 137, 140, 141, 143, 148, 149, 152, 156, 157, 163, 164, 166, 167, 171 anaerobic bacteria, 70 animal tissues, 98

Index

184 animals, 4, 43, 45 annealing, 69 antagonistic, 148 anticancer, 87 antioxidative potential, 102 application, 18, 76, 114, 121, 138, 139, 159 applied research, 170 aquaculture, 43, 61, 103 arachidonic acid, 60, 87 Aristotle, 101, 119 arsenic, 3 arthritis, 127 ash, 40 Aspergillus niger, viii, 20, 31, 33, 39, 40 assessment, 68 assimilation, ix, x, 9, 10, 13, 15, 16, 101, 113, 137, 138, 140, 141, 143, 163, 166 assumptions, 15 Athens, x, 88, 137, 169 atherosclerosis, 43 atmosphere, 89 atoms, 67, 77 ATP, 12, 86, 87, 101, 111, 141, 143, 145, 146, 147, 149, 151 Australia, 7, 135 autotrophic, 45 availability, 70, 87, 126, 171

B Bacillus, 66, 78 bacteria, ix, 10, 11, 13, 14, 17, 26, 30, 65, 66, 67, 69, 70, 71, 72, 76, 77, 78, 80, 81, 84, 123, 126, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 164 bacterial, x, 29, 66, 72, 77, 78, 80, 83, 84, 126, 137, 145, 148, 149, 153 bacterial cells, 77, 78 bacterial strains, x, 126, 137, 145, 148, 149, 153 bacterium, 69, 70, 76, 84, 103 barley, 110 barrier, 68 beef, 110 beet molasses, 32, 97, 124 behavior, 113, 160 beneficial effect, 42, 102 benefits, viii, 41 beta-carotene, 119, 121, 123, 124 beverages, 20, 102, 107 binary blends, 82 binding, 51

biochemistry, 99, 145 biocompatibility, 67, 70 biocompatible, 80 bioconversion, 66, 143, 164, 165 biodegradability, 67, 68, 78, 167 biodegradable, ix, 33, 65, 66, 80, 81, 82, 102, 126, 139 biodegradation, 70 biodiesel, vii, viii, ix, x, 1, 2, 3, 5, 6, 7, 9, 23, 28, 30, 31, 32, 33, 34, 39, 40, 41, 42, 45, 46, 52, 53, 54, 59, 60, 61, 65, 72, 80, 87, 96, 102, 116, 117, 118, 121, 123, 124, 125, 126, 127, 128, 133, 134, 135, 165, 167, 170, 175, 180 bioengineering, 81 biofuel, 20, 33, 133 biofuels, 7, 124 biogas, 138 biological processes, 157 biological systems, 43 biomass, ix, x, 6, 7, 21, 22, 30, 33, 34, 35, 36, 37, 38, 45, 47, 48, 50, 51, 53, 55, 58, 59, 60, 66, 72, 73, 77, 83, 85, 86, 88, 89, 90, 91, 92, 94, 95, 103, 104, 105, 108, 112, 113, 114, 115, 116, 117, 118, 122, 123, 125, 127, 128, 129, 130, 131, 132, 133, 134, 135, 140, 144, 145, 155, 159, 160, 161, 162, 175, 176 biomass growth, 130, 131 biomedical applications, 68, 70 biopolymers, 66, 80, 81, 83, 84 108 bioreactor, 21, 128, 129, 144 biorefinery, 66, 75, 82 biosynthesis, viii, x, 19, 30, 32, 33, 34, 35, 39, 40, 69, 71, 87, 97, 102, 104, 114, 115, 119, 121, 122, 124, 126, 127, 138, 140, 142, 143, 145, 148, 149, 151, 155, 160, 163 biotechnological, x, 32, 39, 62, 80, 81, 126, 138, 139, 166, 170, 174 biotechnologies, 39 biotechnology, 18, 29, 60, 82, 121, 168, 172, 181 biotransformation, 126, 139 blends, 83, 84 blocks, 87 blood, 6, 68, 127 bottleneck, 151 bovine, 110 brain, 43 Brazil, 83 breakdown, 141, 143, 144, 145, 148, 149, 152, 155

Index breast cancer, 98 breeding, 66 brevis, 167 broilers, 6 bubble, 62 buffer, 46, 174 building blocks, 87 butyric, 5, 74, 142, 151, 159, 160, 162, 166, 171, 175, 176 by-products, viii, 20, 24, 26, 31, 33, 34, 35, 39, 75, 97, 102, 148

C Ca2+, 77 cabbage, 133 calcium, 3, 118 Canada, 42, 61 cancer, 87, 102 cancer treatment, 87 Candida, ix, 16, 20, 28, 29, 30, 40, 95, 104, 111, 125, 126, 128, 129, 134, 163 capillary, 47 capital cost, x, 169, 172 carbohydrates, 35, 38, 107, 108, 127, 130, 131 carbon atoms, 67, 77 carbon dioxide, 70 cardiovascular disease, 42, 43, 61 carotene, 102, 104, 105, 106, 107, 108, 110, 112, 113, 114, 116, 117, 118, 119, 120, 121, 122, 123, 124 carotenoids, ix, 101, 102, 103, 104, 106, 107, 108, 109, 111, 113, 114, 115, 117, 118, 119, 120, 122, 123, 124 carrier, 89 catabolic, vii, 9, 10, 16 catabolism, 10, 12, 14, 16, 17, 18, 111, 145, 146, 147, 149, 150, 151, 167 catalyst, 2, 3, 5, 6, 40, 139 catalytic properties, 18 C-C, 105 celery, 26 cell, ix, 11, 13, 15, 22, 29, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 61, 62, 66, 68, 70, 72, 73, 74, 75, 77, 78, 85, 86, 87, 96, 98, 108, 110, 113, 115, 117, 120, 123, 127, 128, 131, 134, 135, 140, 143, 146, 151, 153, 165, 166, 168, 176, 180, 181 cell adhesion, 68 cell growth, 22, 46, 50, 51, 53, 54, 55, 56, 58, 108, 110, 113, 117, 151, 168

185

cell membranes, 11, 87, 96 cell metabolism, 70 cellulose, 33, 75 cellulosic, 2 cellulosic ethanol, 2 chain termination, 76 chain transfer, 76, 82 cheese, 87, 110 chemical industry, 139 chemical structures, 42 chemicals, 66, 102, 165, 168, 173, 181 chicken, 4, 78 chiral, 68 chloride, 77, 116, 117 chloroform, 77, 81, 89 chloroplasts, 122 chromatography, 68, 79, 89, 173 chronic disease, 62 circulation, 177 cis, 15 citrus, 110, 120 classes, 67, 96, 97, 111 classical, 140, 148, 153, 179 cleaning, 173, 174 cleavage, 148 cloning, 18, 122 closure, 82 CNS, 61 CO2, ix, 65, 71, 148, 151, 152, 160, 161 coal, ix, 65, 71 coatings, 68 cobalt, 46 cocoa, 30, 134 coding, viii, 9, 16, 71 coenzyme, 70, 101 Colorado, 88, 135 combustion, 1, 4, 126 commercialization, 59, 102, 103, 170, 179, 180 commodity, 68, 173 competition, 81, 147, 151 competitiveness, 80 compliance, x, 125 components, 39, 46, 55, 77, 92, 96, 117, 176 composites, 68, 83 composition, ix, 3, 7, 11, 14, 18, 21, 46, 47, 49, 51, 57, 58, 61, 63, 65, 67, 69, 70, 75, 79, 85, 88, 92, 93, 94, 95, 96, 98, 99, 108, 110, 117, 118, 123, 124, 128, 131, 132, 133, 134 composting, 1, 4

186 compounds, x, 3, 68, 82, 105, 110, 126, 137, 142, 146 concentrates, 102, 134 concentration, viii, x, 7, 11, 19, 21, 22, 23, 24, 25, 26, 28, 30, 31, 33, 34, 35, 37, 38, 39, 40, 47, 49, 50, 53, 55, 56, 57, 72, 74, 75, 76, 78, 84, 86, 88, 104, 110, 112, 114, 115, 125, 128, 130, 132, 143, 151, 152, 154, 155, 156, 157, 159, 160, 161, 166, 171, 175, 179, 180 condensation, 107, 151 configuration, 67, 154 conformational stability, 11 conservation, ix, 65, 71 constraints, 169 construction, 3, 163, 169, 170, 173 construction materials, 3 consumer goods, 126 consumers, 42, 104, 117, 170, 174 consumption, 19, 32, 35, 36, 37, 44, 48, 49, 55, 56, 60, 61, 72, 75, 81, 88, 91, 117, 127, 142, 143, 145, 156, 160, 174, 175, 179 consumption rates, 72 contamination, viii, 41, 43, 72 contingency, 173 control, 5, 22, 23, 60, 113, 116, 117, 118, 143 control group, 5 convergence, 159 conversion, ix, x, 3, 6, 20, 38, 94, 102, 121, 125, 127, 130, 131, 133, 137, 138, 139, 140, 143, 144, 145, 148, 150, 155, 163, 164, 165, 167, 170, 174, 180 conversion rate, 20, 148 cooking, 126 cooling, 174 copolymer, 75, 76 copolymers, ix, 65, 69, 72, 73, 75, 77, 79, 80, 84 copper, 18 corn, 5, 32, 53, 110 coronary heart disease, 43 correction factors, 173 correlation, 75 correlations, 75 cosmetics, vii, 1, 20, 102, 126, 170 cost-effective, vii, 1, 127 costs, 32, 39, 45, 71, 102, 117, 126, 170, 172, 173, 174, 178 cotton, 59, 110 coupling, 11 cows, 5, 6 CRM, 173

Index crude oil, 66, 170 Cryptococcus, 134, 165 crystalline, 70, 105, 118, 121 crystallinity, 69 cultivation, x, 21, 28, 36, 37, 38, 61, 75, 83, 88, 92, 95, 103, 105, 112, 121, 125, 129, 131, 133, 134, 163, 165 cultivation conditions, 105 culture conditions, viii, 20, 41, 45, 52, 53, 55, 60, 95, 105, 107, 142, 154, 155, 157 culture media, 108 cyclic AMP, 120, 122 cystine, 131, 133 cytoplasm, 70 cytoplasmic membrane, 12 cytosol, 86 cytosolic, 13, 14, 32 Czech Republic, 128

D dairy, 5, 6 Dallas, 60 de novo, 43, 44 deficiency, 30, 131 definition, 86 degradation, 68, 70, 81, 82, 83, 84 degradation rate, 68, 70 degrading, 163 Degussa, 139 dehydrogenase, 12, 13, 14, 15, 16, 17, 18, 29, 86, 111, 112, 120, 141, 142, 143, 147, 148, 150, 154, 166 dehydrogenation, 123 delivery, 68 denaturation, 78 density, 21, 72, 73, 134, 165 depreciation, 173 depressive disorder, 62 derivatives, 17, 144 detergents, 77, 102 deviation, 53 dextrose, 14, 33, 88 diacylglycerol, 92 dichloroethane, 77 diesel, x, 32, 98, 123, 126, 137, 138, 140, 155, 163, 166, 180 diesel engines, 126, 138 diesel fuel, 126, 138 diet, 105 dietary, 42, 43, 61, 102, 103

Index dietary fat, 42 diets, 5, 7, 127 differentiation, 68 diffusion, 10, 11, 14, 15, 16, 17 digestion, 1, 4, 6, 78, 84 dimorphism, 104 direct cost, 174 direct costs, 174 directives, 127 dissolved oxygen, 59 distillation, 126, 169, 176 distilled water, 21, 46, 47, 55, 88, 115 distribution, 97, 102 DNA, 15, 78, 122 docosahexaenoic acid, viii, 7, 41, 42, 45, 61, 62, 63 double bonds, 42 dressings, 68 drug delivery, 68 drug delivery systems, 68 drying, 33 DSM, 24, 107, 144, 148, 151, 157, 158, 159, 163, 167

E E. coli, 5 economics, ix, x, 101, 117, 119, 169, 170, 179 Education, 28 egg, 62, 78, 127, 128, 132 eicosanoids, 43, 60 eicosapentaenoic acid, viii, 41, 60, 62, 63 elaboration, 139, 143 elasticity, 69 elastomers, 67 electricity, 174, 179, 180 electron, 66, 67, 141, 142, 143, 145 electrons, 141 elongation, 43, 77 emission, ix, 33, 65, 71, 126 EMP, 142 employment, ix, 101, 102 encoding, 17, 71, 143, 144, 148 energy, viii, x, 1, 5, 6, 11, 19, 21, 32, 33, 39, 42, 66, 70, 81, 87, 96, 114, 127, 129, 131, 134, 137, 143, 145, 146, 147, 151, 169, 174, 179 energy consumption, 32, 179 Energy Independence and Security Act, 2 engines, 126, 138 environment, 26, 70, 126, 127 environmental factors, 104

187

environmental impact, 68, 170 Environmental Protection Agency (EPA), viii, 41, 42, 43, 45, 54, 55, 57, 58, 59, 60, 61, 63, 97 enzymatic, ix, 18, 21, 65, 68, 78, 79, 84, 108, 112, 151 enzymes, vii, 2, 9, 10, 14, 16, 18, 43, 67, 69, 70, 71, 77, 78, 81, 83, 97, 113, 122, 124, 141, 143, 148, 163 equilibrium, 107 Escherichia coli, 6, 11, 16, 17, 18, 71, 82, 164, 167 essential fatty acids, 62 ester, viii, ix, 19, 21, 33, 67, 125, 127, 139, 163 esters, ix, 20, 33, 47, 72, 77, 89, 98, 123, 125, 126, 128, 138, 166 ethanol, 2, 5, 15, 16, 20, 26, 28, 33, 34, 39, 48, 61, 77, 127, 140, 142, 145, 146, 147, 151, 159, 163, 165, 171, 173, 174 Ethanol, 124 ethanolamine, 93 ethyl acetate, 105 ethyl alcohol, 34 ethylene, 139, 145 ethylene glycol, 139 ethylene oxide, 139 eukaryotes, 10, 12, 111 Europe, 66, 104, 138 European Parliament, 120 European Union (EU), 39, 66, 104, 120, 126, 138 evaporation, 49, 89, 116, 176 experimental condition, 46 experimental design, 29, 53, 133, 165 exploitation, 102, 104, 119 exposure, 20, 26 extraction, 40, 62, 63, 77, 82, 83, 84, 106, 170 extraction process, 170 extrapolation, 172 extrusion, 12

F fabrication, 138 facilitators, 11 FAD, 12 family, x, 15, 16, 107, 137, 140, 141, 145 FAO, x, 62, 121, 125, 131 farming, 62 fat, 24, 29, 60, 97, 98, 99, 123, 127, 129, 130, 131, 138, 140 fats, 2, 102, 126, 127, 134, 138

188 fatty acids, viii, ix, x, 3, 4, 20, 31, 32, 33, 34, 35, 36, 37, 38, 39, 41, 42, 43, 44, 45, 46, 47, 51, 52, 55, 58, 59, 60, 61, 62, 63, 72, 87, 96, 97, 102, 108, 125, 126, 128, 129, 130, 131, 133 fee, 173 feed additives, 45, 102 feedback, 141 feedback inhibition, 141 feeding, ix, 22, 65, 67, 72, 73, 75, 124, 127, 165, 167 feedstock, ix, 3, 5, 9, 65, 72, 80, 96, 115, 118, 119, 126, 131, 140 fees, 174 fermentation broth, 157, 165 fermentation technology, 138, 148 fertilizers, 126 fiber, 66, 107, 139 fibers, 68 FID, 89 films, 68 filters, 21 filtration, 33, 77, 78, 88, 126 fish, 42, 43, 44, 45, 60, 61, 62 fish meal, 43 fish oil, 43, 44, 61, 62 fisheries, 62 fission, 18 fixed costs, 174 flame, 47 flavor, 20 flexibility, 139 flow, x, 21, 29, 34, 47, 88, 114, 128, 150, 155, 157, 167, 169, 174, 175, 180 flow rate, 21, 34, 47, 88, 128, 157 fluid, 46, 55 focusing, 75 food, vii, ix, 1, 32, 54, 59, 60, 62, 66, 68, 87, 101, 102, 103, 104, 105, 106, 110, 119, 120, 121, 135, 170 Food and Drug Administration, 127 food industry, 32, 170 food safety, 102, 119 foodstuffs, 105, 106, 120 fossil, ix, 65, 66, 71 fossil fuel, 66 freeze-dried, 47, 55, 77 fructose, 26, 76, 83, 108, 110, 141, 145 fruits, 40 fuel, 39, 66, 138 fumarate, 141, 157, 165

Index fumaric, 21, 22, 23, 24, 25, 26 fungal, viii, ix, 40, 41, 54, 55, 58, 59, 60, 101, 102, 104, 105, 107, 108, 111, 112, 115, 118, 124 fungi, ix, 10, 11, 14, 15, 16, 20, 26, 42, 43, 44, 55, 58, 59, 62, 70, 98, 101, 102, 104, 111, 119, 121 fungus, viii, 14, 20, 41, 45, 55, 59, 62, 97, 103, 104, 108, 111, 112, 113, 114, 116, 119, 121, 123 Fusarium, 11, 14, 16, 104 Fusarium oxysporum, 11, 14, 16

G Gamma, v, 85, 87, 98 gas, 47, 68, 71, 79, 89 gas chromatograph, 47, 79 gene, 10, 15, 17, 18, 32, 40, 124, 127, 144, 148 gene expression, 10 general expenses, 174 generation, 111, 138, 142, 143, 144, 147, 149, 151 genes, viii, 9, 10, 15, 16, 17, 71, 104, 124, 141, 143, 144, 167, 168 genetics, 84, 104 genome, 17 Germany, 21, 65, 78, 80, 83, 84, 138 Gibbs, 40 glass, 69, 77 gluconeogenesis, 111 glucose, 5, 12, 14, 15, 16, 20, 21, 24, 26, 28, 29, 30, 32, 35, 39, 40, 45, 46, 47, 48, 49, 54, 55, 56, 57, 58, 59, 95, 97, 98, 107, 108, 110, 111, 112, 113, 114, 116, 117, 118, 145, 151, 154, 163, 164, 165, 167, 171 glycerin, viii, ix, x, 6, 31, 33, 34, 35, 36, 37, 38, 39, 121, 125, 126, 127, 129, 130, 131, 132, 133, 138, 163, 165, 170 glycerine, 6 Glycerine, 88, 133 glycine, 11 glycol, 5, 6, 133, 139, 145 glycolipids, ix, 14, 85, 87, 89, 92 glycolysis, 32, 111, 142 gout, 127 government, iv grain, 124 grains, 97, 110, 119 gram-negative bacteria, 77, 78 granules, 66, 67, 70, 77, 81, 83

Index grapes, 26 grasses, 26 Greece, 9, 85, 88, 101, 137, 138, 169, 175 groups, 32, 45, 87 growth inhibition, 115 growth rate, 15, 47, 49, 61, 86, 91, 130, 131, 151, 159

189

hydrolyzed, 108, 110 hydrophilic, 114 Hydrophobic, 29 hydrostatic pressure, 124 hydroxide, 5 hydroxyapatite, 77

I

H H2, 21, 34, 47, 88, 128, 148, 151, 152 halogenated, 77, 82 harvesting, 33 hazardous materials, 174 hazards, 118, 119 health, 42, 62, 102, 121, 139 health problems, 139 heart, 42, 43, 60, 68 heart disease, 43 heart valves, 68 heat, 77, 78, 173, 174, 177 heat exchangers, 177, 178 heat transfer, 173, 177 heating, 89, 138 heavy metal, viii, 31, 40, 41, 42, 43, 88, 110, 118, 170 heavy metals, 31, 40, 88, 110, 118, 170 height, 177, 178 helical conformation, 70, 80 helium, 89 heterogeneous, 59 heterotrophic, 45, 61, 124 hexane, 110 high pressure, 78, 139, 179 Honda, 45, 61, 62, 63 hormone, 43 HPA, 139, 141, 143, 144, 145, 146, 147, 150 HPLC, 21, 34, 47, 88, 105, 123, 128 human, viii, 41, 42, 54, 68, 70, 102, 127, 133 hydro, 6, 20, 77, 108, 114 hydrocarbon, 2, 108, 120 hydrocarbons, 6, 20, 77, 108 hydrochloric acid, 46, 116 hydroformylation, 139 hydrogen, 5, 12, 71, 77, 121, 126, 140 hydrogen gas, 71 hydrogen peroxide, 12, 77, 121 hydrogenation, 139 hydrolysates, 134 hydrolysis, 12, 70, 108, 157 hydrolytic stability, 68

identification, vii, 9, 10, 123 Illinois, 98 immune system, 102 immunoglobulins, 110 implants, 69 implementation, 60 impurities, vii, viii, 1, 3, 7, 19, 22, 33, 41, 49, 54, 55, 77, 88, 116, 117, 118, 123, 126, 127, 140 in vivo, 70, 80, 123 inactivation, 14, 112, 123 incentives, 66 incidence, 68 income, 170 incubation, 114, 144 incubation period, 114 indication, 118 indices, 160, 161 induction, 10, 13, 14, 16, 120 industrial, x, 5, 6, 20, 28, 30, 32, 45, 60, 78, 80, 87, 97, 98, 102, 104, 105, 115, 116, 117, 119, 121, 126, 127, 134, 138, 140, 152, 154, 156, 157, 166, 169, 171, 180 industrial application, 115 industrial production, 6, 104 industrial wastes, 20 industry, vii, viii, ix, x, 1, 7, 9, 20, 32, 41, 42, 43, 53, 59, 65, 66, 68, 76, 102, 110, 124, 125, 128, 137, 139, 163, 170 infancy, viii, 9, 61 infants, 43 inflammatory, 43, 68 infrared, 6 inhibition, 63, 105, 114, 115, 141, 159, 168 inhibitors, 11, 116 inhibitory, viii, 41, 52, 54, 59, 112, 115, 117, 159 inhibitory effect, viii, 42, 52, 54, 59, 112, 115, 117, 159 inhomogeneities, 78 injection, 47 inoculation, 33, 90, 91, 128 inoculum, 21, 33, 46, 128 inorganic, 33, 110, 126

Index

190 inorganic salts, 33, 126 insects, 43, 60 instability, 69, 144 integration, 90, 170, 179 interaction, 52 interdependence, 108 intrinsic, 139 intrinsic viscosity, 139 inventions, 77 investment, 172, 174, 177, 178 ionization, 47 ions, 20, 54, 121 iron, 39, 46, 53 isolation, ix, 65, 98, 128, 164 isoleucine, x, 125 isomorphism, 80 isoprenoid, 112 isotactic polypropylene, 69 Italy, 62

J joints, 127

K kaolinite, 39 kerosene, 108 ketones, 77 kinase, 11, 12, 13, 14, 15, 16, 17, 18, 111, 112, 122, 123, 141, 142, 143, 147, 150 kinase activity, 14, 123 kinetic model, 159 kinetic parameters, 26, 28, 130 kinetics, 15, 38, 47, 122, 124, 138, 145 KOH, 2, 33, 72, 74, 75

L labor, 173, 174, 178 lactate dehydrogenase, 148, 154 lactation, 6 lactic acid, x, 30, 68, 82, 137, 140, 142, 144, 146, 148, 164 lactic acid bacteria, 30, 140, 145, 146, 164 Lactobacillus, 145, 164, 167 lactose, 32, 35, 36, 108, 110, 122, 127 large-scale, 28, 59, 81, 102, 104 latex, 68 law, 2, 172 LDH, 148 leaching, 31, 40

Lebanon, 32 legislation, 2 leukotrienes, 43 liberation, 151 life cycle, 68, 83 lignin, 88 limitation, 28, 54, 72, 75, 86, 114, 115, 155 limitations, viii, 41, 59, 163 linear, 160 linear regression, 160 linoleic acid, ix, 43, 85 linolenic acid, ix, 43, 85, 87, 97, 98, 119 lipase, 2, 157 lipids, 12, 16, 32, 33, 34, 40, 60, 62, 67, 77, 86, 87, 88, 92, 93, 95, 96, 98, 108, 111, 113, 114, 116, 118, 119, 123, 127, 128 lipopolysaccharides, 77 liquid chromatography, 101 liquid water, 6 liquor, 53, 110 localization, 16, 120 logarithmic coordinates, 172 losses, 160 low molecular weight, 76 low-temperature, 62 lutein, 102, 124 lycopene, 102, 104, 105, 106, 107, 108, 113, 118, 120, 121, 122, 124 lysine, 127 lysis, 77 lysozyme, ix, 65, 77, 78, 79

M magnesium, 3, 118 maintenance, 174 major depressive disorder, 62 malic, 21, 22, 23, 24, 25, 26 malt extract, 21 maltose, 15, 108, 145, 164 management, 39, 96 manganese, 46 manipulation, 127 mannitol, 21, 22, 23, 24, 25 manufacturing, x, 5, 60, 98, 117, 123, 127, 128, 165, 166, 169, 180 marine environment, 81 market, vii, viii, x, 1, 2, 19, 41, 43, 66, 81, 102, 104, 117, 120, 137, 138, 170, 172, 179 market value, 117 markets, 66, 138

Index maximum specific growth rate, 115 meals, 43 meat, 127 mechanical properties, 73, 81 media, ix, 21, 24, 26, 28, 32, 33, 34, 35, 36, 38, 49, 55, 73, 87, 94, 95, 98, 107, 108, 110, 112, 113, 114, 125, 126, 131, 132, 143, 148, 151, 159, 173 medicine, 84, 121 medium composition, viii, 32, 41, 55 melons, 26 melt, 69 melting, 69, 77 melting temperature, 69 membrane permeability, 14 membranes, 10, 34, 68, 95, 96, 173 mercury, 3, 60, 63 metabolic, 6, 34, 70, 71, 80, 84, 97, 104, 111, 113, 141, 143, 145, 151, 159, 160, 161 metabolic shift, 151 metabolism, ix, 10, 13, 16, 17, 18, 65, 66, 67, 70, 71, 80, 82, 83, 84, 111, 112, 114, 121, 124, 147, 151, 155, 163, 164, 165, 166, 167 metabolite, 17, 107, 164 metabolites, 6, 22, 26, 34, 104, 143, 159, 160, 161 metabolizing, 18 metal ions, 20 metals, 31, 40, 88, 110, 118, 170 methane, 70, 71 methanol, viii, 2, 3, 5, 20, 33, 34, 39, 40, 41, 46, 48, 49, 50, 52, 53, 55, 72, 77, 79, 83, 88, 89, 116, 117, 118, 126, 127, 140, 163, 170 methionine, 131, 132, 133, 134 methylation, 47, 63, 89 methylene, 77 methylene chloride, 77 Methylobacterium, 66, 67, 69, 83, 84 metric, 138 Mg2+, 77 MgSO4, 21, 33, 46, 53, 88, 128 microalgae, 42, 45, 48, 60, 63, 121, 122 microbial cells, 10, 142, 157 microflora, 135 Micronesia, 45 micronutrients, 55 microorganism, 10, 27, 32, 86, 148, 151, 155, 159, 160 microorganisms, x, 10, 11, 12, 26, 32, 43, 44, 45, 58, 86, 94, 96, 97, 99, 102, 119, 126, 127, 137,

191

139, 140, 141, 142, 143, 144, 148, 151, 153, 155, 157, 158, 159, 165, 171 milk, 39, 40, 110 minerals, 31, 127 mining, 40 mitochondria, 43, 86 mitochondrial, 14, 15, 17, 18, 112 mitochondrial membrane, 14 modeling, 122, 138, 145, 168 models, 159 modules, 178 modulus, 69, 77 moisture, 70 molar ratio, 151, 152 molasses, 20, 39, 40, 87, 94, 97, 98, 110, 121, 127, 134 mole, 12, 83, 176 molecular biology, vii, 9, 16, 99, 104 molecular mass, 76, 84 molecular oxygen, 12 molecular weight, ix, 65, 67, 75, 76, 77, 78, 176 molecules, 11, 42, 70, 107, 152 molybdenum, 46 monomer, ix, 65, 67, 69, 72, 79, 84, 139 monomeric, 70 monomers, 67, 68, 70, 72, 171 morphogenesis, 81 morphological, 59, 60, 82 morphology, viii, 34, 42, 115 Moscow, 16 MPM, 84 multiplication, ix, 65, 75 municipal solid waste, 40 municipal solid waste (MSW), 40 mushrooms, 26 mutant, viii, 5, 14, 17, 19, 20, 23, 40, 120, 121, 148, 153, 158, 165, 167 mutants, 7, 15, 18, 22, 28, 30, 132, 134, 148, 153, 154, 163, 164, 167 mutation, 104, 127, 148, 153 MVA, 107 mycelium, ix, 33, 34, 55, 85, 96, 104, 123 myelin, 43

N Na2SO4, 89 NaCl, 21, 46, 53, 72, 73 NAD, 12, 71, 101, 111, 142, 143, 146, 147, 148, 151, 160, 167 NADH, 70, 166

Index

192 nanocomposites, 68, 82 natural, 26, 68, 80, 83, 102, 111, 117, 122, 127, 153 natural environment, 26 natural selection, 127 nerve, 69, 81 neutral lipids, ix, 85, 87, 89, 99 New York, 17, 60, 62, 81, 135 nickel, 139 nicotinamide, 101 Nielsen, 4, 6 Niger, v, 31 nitrate, 33, 141 nitrogen, 22, 28, 30, 32, 34, 39, 40, 47, 56, 72, 73, 75, 85, 86, 89, 90, 91, 95, 97, 107, 110, 113, 116, 127, 131 nitrogen compounds, 110 nitrogen oxides, 39 NMR, 72 nontoxic, 126 normal, 66 nucleic acid, 127 nutraceutical, 87, 107 nutrient, 20, 42, 59, 67, 73, 102, 110, 116 nutrient transfer, 59 nutrients, ix, 47, 54, 55, 61, 65, 68, 107, 116 nutrition, 127

O oat, 11 observations, 115 oil, viii, ix, 1, 2, 3, 5, 20, 26, 28, 30, 32, 33, 34, 35, 36, 41, 42, 43, 44, 46, 54, 60, 61, 62, 65, 66, 71, 80, 85, 93, 98, 105, 108, 110, 118, 122, 123, 126, 127, 134, 135, 166, 167, 170, 180 oils, 2, 20, 62, 86, 87, 102, 107, 108, 110, 114, 126, 127, 134, 138 oilseed, 3 oligomers, 70 olive, 108, 122 olives, 26 omega-3, viii, 41, 42, 44, 45, 58, 59, 60, 61, 62, 63 omega-6, 45, 62 operator, 174, 178 optimization, 95, 105, 112, 117, 119, 163, 179 organic, 19, 34, 39, 71, 88, 97, 110, 111, 117, 126, 131, 139, 150, 155, 157 organic compounds, 157 organism, 5, 45

osmotic, 26, 53, 67 osmotic pressure, 53 oxalate, viii, 31, 32, 35, 40 oxalic, viii, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40 oxalic acid, viii, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40 oxidation, 12, 111, 139 oxidative, 10, 12, 16, 111, 142, 151, 153 oxide, 139 oxygen, 12, 29, 30, 59, 72, 75, 88, 121, 141 oxygen consumption, 88 oxygen consumption rate, 88 oxygenation, 132

P Pacific, 61 packaging, 66, 68, 80 palm oil, 134 Pap, 167 paraffins, 30 parameter, 36, 78, 90, 160 Paris, 82, 97 partition, 155 passive, 10, 11, 14, 16, 17 patents, 174 pathogenic, 105 pathways, vii, 9, 10, 12, 13, 44, 111, 141, 143, 144, 145, 146, 147, 149, 150, 163, 166, 167 patients, 43 PCR, 17 PCT, 82, 83 PDC, 112, 113 pears, 26 Pediococcus, 18 penicillin, 120 PEP, 10, 14 pericardial, 68, 82 pericardium, 82 permeability, 14, 34 permeation, 16 permit, 68 Petri dish, 104 petrochemical, 66, 68, 83, 170 petroleum, 32, 34, 66, 126, 128 pH, ix, 3, 4, 5, 21, 22, 23, 24, 25, 27, 28, 32, 33, 35, 36, 37, 38, 40, 46, 53, 54, 70, 72, 73, 74, 75, 77, 78, 79, 90, 92, 115, 124, 125, 128, 129, 132, 159, 162, 164 pharmaceutical, vii, 1, 68, 87, 107 pharmaceuticals, 20, 102, 126, 170

Index PHB, ix, 65, 66, 67, 68, 69, 70, 71, 72, 73, 75, 77, 79, 80, 81, 82, 83 phenotype, 16 Phenylalanine, 132 phosphate, 11, 13, 14, 15, 17, 18, 67, 93, 101, 111, 141, 142 phosphatidylcholine, 93, 96 phosphatidylethanolamine, 96 phosphoenolpyruvate, 16 phospholipids, ix, 14, 70, 85, 87, 89, 92 phosphorous, 3, 118 phosphorus, 32, 34, 40 phosphorylation, 10, 11, 12, 13, 14, 15, 16, 122 photobioreactors, 45, 62 photoreceptor, 43 photoreceptor cells, 43 physical and mechanical properties, 81 physical properties, 67, 69, 77 physiological, x, 11, 17, 66, 87, 88, 96, 115, 137, 140, 143, 160 physiology, 17, 43, 87, 96 piezoelectric, 69 piezoelectric properties, 69 pigments, 102, 120, 121, 122, 123 pigs, 4, 6, 127, 134 placebo, 62 planning, 170 plants, 26, 43, 66, 102, 117, 118, 122, 126, 138, 140, 155, 163, 170 plasma, 60 plasmid, 144, 168 plasmids, 144 plasmolysis, 115 plastic, 70 plastics, ix, 65, 66, 80, 82, 83, 102, 139, 171 play, 87, 112 Poland, 19, 20, 21, 28, 31, 33, 125, 128 polarity, 87 pollutant, 110 poly(3-hydroxybutyrate), 80, 81, 82, 83, 84 polyester, 66, 70, 139, 171 polyesters, 66, 67, 80, 81, 82, 126, 139, 167, 171 polyether, 171 polyhydroxybutyrate, 82 polymer, ix, 65, 66, 68, 69, 70, 72, 74, 75, 76, 77, 78, 79, 80, 82, 165 polymer chains, 70 polymer industry, ix, 65, 76 polymer properties, ix, 65, 75 polymer synthesis, 165

193

polymerase, 67 polymerization, 76 polymers, ix, 65, 67, 68, 69, 70, 72, 75, 77, 81 polynomial, 172 polypropylene, ix, 65, 69 polysaccharides, 111, 133 polyunsaturated fat, viii, 41, 42, 59, 60, 61, 87, 97 polyunsaturated fatty acids, viii, 41, 42, 59, 60, 61, 87, 97 polyurethane, 32, 171 polyurethane foam, 32 pomace, 108, 122 poor, 54, 77, 87, 108, 110, 112, 113 population, 70 pore, 33, 128 porous, 68 postoperative, 82 potassium, ix, 88, 125, 128 potato, 32, 33, 88, 108 poultry, 103, 121, 127 powder, 89 powders, 107 power, 143, 149, 153 precipitation, 77 pressure, 3, 6, 53, 78, 101, 124, 133, 139, 179 prevention, 42 prices, 1, 2, 20, 66, 170 producers, 1, 3, 5, 19, 22, 42, 43, 44, 45, 60, 69, 86, 99, 104, 153, 170 product market, 170 production costs, 39, 170 productivity, viii, ix, 26, 27, 30, 31, 38, 39, 47, 48, 49, 50, 53, 55, 56, 57, 58, 59, 101, 113, 114, 117, 118, 119, 148, 154, 155, 157, 166, 171, 180 profit, 102 profitability, 172 prokaryotic, x, 137, 139, 140, 141, 142, 144, 146, 155 prokaryotic cell, 146 promoter, 15 propionic acid, 72 propylene, 5, 6, 133, 139 prostaglandins, 43 proteases, 77 protein, ix, x, 14, 15, 16, 122, 125, 126, 127, 128, 130, 131, 132, 133, 134, 135 proteins, 11, 12, 14, 15, 17, 67, 68, 70, 71, 78, 83, 110, 133, 176

Index

194 protocol, 48, 49, 59, 116 Pseudomonas, 13, 17, 29, 81 Pseudomonas aeruginosa, 17 PSI, 78 PTT, 171 public, 1, 121 public health, 121 public support, 1 PUFA, 52, 87, 96 PUFAs, 42, 43, 44, 87, 93 purification, 3, 79, 82, 87, 98, 106, 126, 127, 140, 170, 174 pyrophosphate, 107 pyruvate, 12, 32, 111, 112, 147, 148, 166 pyruvic, 12, 127, 140, 148, 151, 153, 160

R radiation, 20 radiolabeled, 52 Raman, 60 range, 3, 4, 28, 32, 34, 37, 50, 56, 66, 67, 68, 107, 112, 118, 131, 145, 148, 154, 172 rape, viii, ix, 19, 20, 65, 72 raw material, 20, 34, 66, 139, 165, 173, 174, 178 reaction temperature, 78 reaction time, 3 reading, 15 reagent, 61 reality, 121, 138 recovery, 77, 78, 79, 80, 81, 82, 105, 106, 117, 154, 163, 173 recycling, 29, 81, 117, 181 redox, 143, 164 reductases, 83 refining, 32, 35, 36, 102, 117 refractive index, 47 refrigerant, 174 regeneration, 68 regional, 135 regression, 160 regression analysis, 160 regular, 67 regulation, 10, 13, 14, 15, 16, 17, 83, 97, 104, 105, 106, 113, 120, 123, 166 regulations, 163 regulators, 17 Reimann, 149, 150, 153, 154, 158, 167, 171, 181 relationship, 111 renal, 127 renewable energy, 1, 81

renewable resource, 20, 80 repair, 69, 81 reparation, 166, 174 repression, 14, 15, 16, 108, 112 research and development, 170 reserves, 66 residues, 20, 30, 34, 75, 102, 127, 134, 170 resins, 139, 173 resistance, 69, 163 resources, ix, 20, 65, 71, 80, 171 respiratory, 143 retention, 47 retina, 43 retinoids, 121 RFA, 2, 7 rice, 110 rings, 107 risk, 102, 119 Rome, 62, 121 room temperature, 21, 34, 89, 128 royalties, 174 RP-HPLC, 101, 108 rubidium, 139 Russia, 107 Russian, 122 rye, 110

S Saccharomyces cerevisiae, 11, 14, 15, 16, 17, 18, 96, 97, 99, 104, 124, 127, 132, 134 safety, 102, 118, 119, 169 salmon, 105 salt, 22, 46, 53, 75, 78, 116, 117, 127 salts, 22, 33, 47, 72, 75, 88, 110, 116, 126, 140, 170 sample, 47, 88 sampling, 92 saturation, 15, 88, 159 scaffold, 68 scaffolds, 68 scaling, 172 scaling law, 172 schizophrenia, 42, 61, 62 SCP, 127, 131 SDS, 79 search, 87, 90 seawater, 46, 53, 70 secretion, 28, 29, 34 seed, viii, 3, 19, 20, 105, 110 seeds, 118, 135

Index selecting, 118 selectivity, ix, 70, 101, 114, 139 separation, 4, 116, 117, 123, 126 series, 176 serum, 18, 60, 110 serum albumin, 110 sewage, 70, 78 shear, 59 Shell, 139, 171 shellfish, 63 shortage, viii, 41 sites, 15 skeleton, 107, 180 Sm, 85, 90 sodium, 3, 5, 77, 78, 79, 81, 88, 116, 117 sodium hydroxide, 5 soil, 70 soils, 82 solid state, 40 solid waste, 31, 40 solubility, 77 solvent, 34, 62, 69, 77, 83, 84, 89, 139 solvents, 77, 82, 105 soy, 3, 110 soybean, 2, 3, 5, 46, 54, 60, 108, 110, 118, 122 soybeans, 127, 133 Spain, 107 species, ix, 20, 29, 42, 43, 45, 48, 52, 54, 60, 86, 104, 122, 123, 126, 131, 133, 134, 141, 145, 149, 153 specificity, 11, 114, 117 spectrum, 144, 157 speed, 21, 33, 35, 36, 37, 38, 59, 128 sphingolipids, ix, 85, 87, 89, 92, 93 spore, 33, 54, 88 stability, 11, 68, 73, 87, 105, 139, 144, 163, 168 stages, 26, 33, 66, 158, 159 stainless steel, 173 standard deviation, 50, 51, 52 standards, x, 47, 89, 125, 131 starch, 95, 108, 110, 127 starvation, 96 steady state, 175 steel, 173 sterilization, 88 stiffness, 69 stoichiometry, 12 storage, ix, 65, 70, 86, 87, 90, 91, 96, 176, 178 strategies, 119 streams, 4, 48, 60, 102, 176

195

strength, 69, 77 stress, 26, 58, 59, 117 structural gene, 71 substances, 43, 119 substitution, 95, 112 substrates, ix, 9, 11, 14, 20, 32, 33, 34, 35, 36, 38, 40, 47, 54, 65, 67, 71, 72, 73, 74, 75, 87, 95, 108, 110, 111, 114, 122, 129, 130, 131, 135, 164 subtraction, 88 Succinic, 6 sucrose, 20, 29, 32, 34, 39, 54, 108, 110, 154, 159 sugar, 10, 14, 15, 16, 18, 26, 39, 61, 87, 94, 95, 98, 110, 142, 145, 146, 154 sugar beet, 39 sugar industry, 110 sugars, ix, 9, 14, 20, 65, 71, 108, 110, 111, 145 sulfur, 3 sulphur, 118, 127 Sun, 124, 167, 168 sunflower, 2, 30, 135, 167 supervision, 178 supplements, viii, 41, 102 supply, 42, 44, 72, 75 surface area, 70 surface layer, 70 surfactant, 83 surfactants, 52, 63 surgical, 69 surplus, 82 surprise, 15 suspensions, 107 sustainability, 43 switching, 58 Switzerland, 66, 81 synaptic vesicles, 43 synthesis, ix, 16, 26, 43, 58, 59, 65, 67, 68, 69, 70, 71, 72, 73, 74, 75, 80, 82, 83, 84, 87, 90, 92, 95, 97, 102, 105, 107, 108, 111, 112, 117, 120, 121, 123, 126, 131, 133, 139, 143, 145, 148, 150, 155, 165, 169 systems, 10, 11, 15, 18, 43, 68, 82, 89, 104, 138, 139, 144, 154, 164

T tamoxifen, 98 taste, viii, 41, 42, 43 taxonomic, 45 technology, 5, 32, 66

196 temperature, viii, 3, 21, 34, 35, 36, 37, 38, 41, 47, 53, 55, 56, 58, 59, 62, 69, 70, 77, 88, 89, 90, 92, 115, 124, 128, 139, 164, 175 tensile, 69, 77 tensile strength, 69, 77 Texas, 62 textbooks, 172 textile, x, 66, 137, 139, 163 textile industry, x, 137, 139, 163 therapeutic benefits, viii, 41 therapy, 98 thermal properties, 69 thermal stability, 73 thermoplastic, 68, 81 Thessaloniki, 101, 119 threonine, 127 threshold, 86, 89, 94 thrombosis, 43 thromboxanes, 43 time, 22, 23, 24, 25, 34, 36, 38, 78, 88, 90, 92, 93, 112, 113, 117, 118, 133, 169, 172 tissue, 68 tissue engineering, 68 tolerance, 20 tomato, 95, 97 toughness, 69 toxic, 144, 157 toxicity, 127 toxicological, 102, 104, 118, 119 toys, 66 transcriptional, 15, 17 transesterification, 2, 3, 33, 46, 117, 170 transesterification reaction, 170 transfer, 59, 76, 82, 173, 177, 178 transformation, 29, 124, 138, 139, 142, 154, 171 transformations, 138 transition, 6, 69, 77 transition temperature, 69, 77 transmission, 66 transport, 9, 10, 11, 14, 15, 16, 17, 18, 28 transportation, 170 trial, 62 tricarboxylic acid, 119, 149 tricarboxylic acid cycle, 119 triglyceride, 2, 3 triglycerides, 2, 18, 33 trout, 105 Tryptophan, 132 turnover, 97, 98 Tyrosine, 132

Index

U Ukraine, 107 uniform, 59 unit cost, 173, 179, 180 United States, vii, 1, 2, 126 urea, 11 uric acid, 127 uric acid levels, 127 USEPA, 43, 63 UV, 20, 21, 34, 68, 69, 123 UV radiation, 20

V vacuum, 88, 116, 126, 176 Valdez, 144, 164 validation, 172 valine, x, 125 values, 3, 23, 26, 28, 36, 38, 53, 75, 90, 92, 94, 117, 128, 130, 131, 132, 133, 143, 147, 148, 154, 159, 160, 161, 162, 171, 172 variation, 172 vascular grafts, 68 vegetable oil, 2, 3, 33, 105, 108, 114, 118, 126, 127, 134 viscosity, 46, 78 vision, 43 visual acuity, 43 vitamin A, 121 vitamin B1, 46 vitamin B12, 46 vitamin E, 127 vitamins, 110, 127

W waste disposal, 173 waste products, 39, 68 waste treatment, 174, 179 waste water, 78 wastes, 20, 87, 94, 102, 165 wastewater, 78, 174 water, 20, 21, 33, 39, 46, 47, 54, 55, 70, 77, 78, 81, 88, 107, 110, 115, 116, 118, 126, 128, 138, 139, 140, 170, 174, 175, 176, 179 waxes, 102 weathering, 69 weight reduction, 77 Western Europe, 138 wheat, 110, 127

Index whey, 32, 35, 39, 40, 87, 95, 97, 110, 122, 124, 127, 135 WHO, x, 121, 125, 131 wild type, 14, 148 wine, 18 WM, 123 wood, 134 workers, 78

Y yeast, viii, ix, x, 5, 10, 11, 17, 19, 20, 21, 22, 23, 26, 28, 29, 30, 46, 54, 55, 57, 86, 88, 94, 97, 98, 99, 103, 104, 110, 115, 124, 125, 127, 128, 129, 130, 131, 132, 133, 134, 135, 165 yield, viii, ix, x, 2, 3, 5, 12, 19, 23, 25, 27, 28, 30, 31, 32, 33, 35, 38, 40, 41, 46, 47, 48, 49, 52, 53, 55, 56, 57, 58, 59, 72, 75, 77, 78, 85, 86, 89, 93, 95, 101, 103, 104, 105, 108, 110, 111, 112, 114, 115, 116, 117, 119, 124, 125, 130, 131, 133, 139, 143, 147, 148, 150, 153, 154, 155, 157, 160, 161, 171, 175 yield loss, 160, 161 yolk, 62

Z zinc, 46

197