UNIVERSIDADE ESTADUAL DE CAMPINAS Faculdade de Engenharia de Alimentos
BRUNO LABATE VALE DA COSTA
THE IMPACT OF OXYGEN AVAILABILITY ON THE PHYSIOLOGY AND NUTRITIONAL REQUIREMENTS OF THE YEAST Saccharomyces cerevisiae
O IMPACTO DA DISPONIBILIDADE DE OXIGÊNIO NA FISIOLOGIA E NOS REQUERIMENTOS NUTRICIONAIS DA LEVEDURA Saccharomyces cerevisiae
CAMPINAS 2019
BRUNO LABATE VALE DA COSTA
THE IMPACT OF OXYGEN AVAILABILITY ON THE PHYSIOLOGY AND NUTRITIONAL REQUIREMENTS OF THE YEAST Saccharomyces cerevisiae
O IMPACTO DA DISPONIBILIDADE DE OXIGÊNIO NA FISIOLOGIA E NOS REQUERIMENTOS NUTRICIONAIS DA LEVEDURA Saccharomyces cerevisiae
Thesis presented to the Faculty of Food Engineering of the University of Campinas in partial fulfillment of the requirements for the degree of Doctor in Science
Tese apresentada à Faculdade de Engenharia de Alimentos da Universidade Estadual de Campinas como parte dos requisitos exigidos para a obtenção do título de Doutor em Ciências
Orientador: Andreas Karoly Gombert
Co-orientador: Thiago Olitta Basso
ESTE EXEMPLAR CORRESPONDE À VERSÃO FINAL DA TESE DEFENDIDA PELO ALUNO BRUNO LABATE VALE DA COSTA E ORIENTADA PELO PROF. DR. ANDREAS KAROLY GOMBERT
CAMPINAS 2019
Ficha catalográfica
Universidade Estadual de Campinas
Biblioteca da Faculdade de Engenharia de Alimentos Claudia Aparecida Romano - CRB 8/5816
Costa, Bruno Labate Vale da,
C823i CosThe impact of oxygen availability on the physiology and nutritional
requirements of the yeast Saccharomyces cerevisiae / Bruno Labate Vale da Costa. – Campinas, SP : [s.n.], 2019.
CosOrientador: Andreas Karoly Gombert. CosCoorientador: Thiago Olitta Basso.
CosTese (doutorado) – Universidade Estadual de Campinas, Faculdade de Engenharia de Alimentos.
Cos1. Anaerobiose. 2. Oxigênio. 3. Saccharomyces cerevisiae. 4. Quimiostato. 5. Fatores de crescimento anaeróbico. I. Gombert, Andreas Karoly. II. Basso, Thiago Olitta. III. Universidade Estadual de Campinas. Faculdade de
Engenharia de Alimentos. IV. Título.
Informações para Biblioteca Digital
Título em outro idioma: O impacto da disponibilidade de oxigênio na fisiologia e nos
requerimentos nutricionais da levedura Saccharomyces cerevisiae
Palavras-chave em inglês:
Anaerobiosis Oxygen
Saccharomyces cerevisiae Chemostat
Anaerobic growth factors
Área de concentração: Bioenergia Titulação: Doutor em Ciências Banca examinadora:
Thiago Olitta Basso [Coorientador] Diogo Ardaillon Simões
Aldo Tonso
Ângela Maria Moraes Everson Alves Miranda
Data de defesa: 12-08-2019
Programa de Pós-Graduação: Bioenergia
Identificação e informações acadêmicas do(a) aluno(a)
- ORCID do autor: https://orcid.org/0000-0002-7055-2762 - Currículo Lattes do autor: http://lattes.cnpq.br/6782652334708832
COMISSÃO EXAMINADORA
Dr. Thiago Olitta Basso
Universidade de São Paulo
Dr. Diogo Ardaillon Simões
Universidade Federal de Pernambuco
Dr. Aldo Tonso
Universidade de São Paulo
Dra. Ângela Maria Moraes
Universidade Estadual de Campinas
Dr. Everson Alves Miranda
Universidade Estadual de Campinas
A Ata da defesa com as respectivas assinaturas dos membros encontra-se no SIGA/Sistema de Fluxo de Dissertação/Tese e na Secretaria do Programa da
To my parents, who always supported my dreams and beliefs.
ACKNOWLEDGEMENTS
First, I would like to thank my supervisors Andreas and Thiago, for the support and guidance throughout this doctorate. To Andreas, for sharing his knowledge, scientific rigor. To Thiago, my friend in the first place, for sharing his knowledge, for having so much patience, and for putting so much energy into this work.
I would like to thank also professor Claudio Oller, for allowing me to carry out all the experiments of this work being carried out in their laboratory. Thanks Anita, Meriellen, Enrique, Larissa, and Mariana for the patience and assistance with the use of lab equipments. A special thanks to Lidiane, for the enourmous help with the quantification of sterols.
In addition, thanks to Sayuri Miyamoto, from the Laboratory for the Research of Modified Lipids, IQ-USP, who opened the doors of her laboratory so that I could work on the quantification of lipids. Thanks to the lipids team: Adriano, Marcos and Maria Fernanda, for the assistance and guidance through lipid research.
To all the professors at LEMeB UNICAMP, Marcus, Rosana, and Chico, thanks for the support and the discussions.
To the friends from Campinas, Felipe Beato, José Valdo, Pamela, Allan, Felipe Ferrari, Suéllen, Daniele, Juliana, Laura, and Dielle, thanks for all the happy hours, chats, and coffee breaks. A special thanks to Vijay, who, in addition to all the above-mentioned socializing moments, took a huge part in this work, and to Wesley, Rafael, and Cristiane for the friendship which helped me to keep an acceptable level of sanity. To professor Aldo Tonso, my master’s supervisor, thanks for the guidance and support. To friends from the Yeast research group in USP, Priscila Cola, Thais, Thiago Araujo, thanks for the support and discussions.
Thanks also to my family and friends, for the support. You were more essential to this work than you can imagine. A special thanks to Priscila for being there for me since the beginning of this doctorate.
I would like to acknowledge the State University of Campinas and the University of São Paulo, for accepting me as their student and researcher in the framework of the Ph.D. Program in Bioenergy, and CAPES for the scholarship granted.
This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001.
I am turned into a sort of machine for observing facts and grinding out conclusions. Charles Darwin
ABSTRACT
It has been debated whether Saccharomyces cerevisiae can grow under complete anaerobiosis only if the anaerobic growth factors (AGF) oleic acid and ergosterol are added to the medium since the biosynthesis of one mole of ergosterol requires 12 moles of molecular oxygen, and each unsaturation in a fatty acid moiety requires one molecule of oxygen. Chemostat cultivations with the Saccharomyces cerevisiae CEN.PK113-7D and PE-2 strains were performed, under the most stringent oxygen-excluded conditions. After a shift from aerobic to anaerobic growth in a defined medium devoid of AGF, with glucose as the limiting nutrient, and at a dilution rate of 0.1 h-1, a
steady-state with growing cells could still be observed. When compared with aerobically grown cells, the unsaturated fatty acid content of anaerobically grown cells decreased significantly. On the other hand, the saturated fraction increased. Regarding sterol composition, extreme oxygen limitation led to an accumulation of squalene and lanosterol at the expense of ergosterol. Contrarily, aerobically grown cells predominantly synthesized ergosterol with very little accumulation of squalene and lanosterol. Although O2 starved, cells diligently adjusted their lipid composition to
accumulate moieties that are not dependent on oxygen for sustaining growth, at expense of cellular fitness. When anaerobically grown cells were exposed to ethanolic or acidic stress conditions, we observed a sharp decrease in cell viability, when compared to aerobic steady-state cells exposed to the same stressors. Overall, this work documents the impact of extreme O2 limitation in continuously-grown S.
RESUMO
Tem sido debatido se a levedura Saccharomyces cerevisiae cresce em plena anaerobiose apenas se os fatores de crescimento anaeróbicos (FCA) ácido oleico e ergosterol forem adicionados ao meio de cultura, visto que a biossíntese de um mol de ergosterol requer 12 moles de oxigênio molecular e cada insaturação em um ácido graxo requer uma molécula de oxigênio. No presente trabalho, foram conduzidos cultivos em quimiostato com as linhagens de Saccharomyces cerevisiae CEN.PK113-7D e PE-2, nas condições de oferta de oxigênio mais restritas possíveis. Após a transição de crescimento aeróbio para anaeróbio, em meio sem oferta de FCA, com glicose como nutriente limitante e a uma vazão específica de 0,1 h-1, um estado
estacionário com crescimento celular ainda pode ser observado. Quando comparadas com células cultivadas em aerobiose, a fração de ácidos graxos insaturados em células cultivadas em anaerobiose foi significativamente menor. Por outro lado, a fração de ácidos graxos saturados, foi maior. Quanto à composição de esteróis, a extrema limitação de oxigênio levou a um acúmulo de esqualeno e lanosterol, em detrimento da síntese de ergosterol. Por outro lado, células cultivadas em aerobiose sintetizaram ergosterol e acumularam muito pouco esqualeno e lanosterol. Apesar da necessidade de oxigênio, as células ajustaram a sua composição lipídica de modo a acumular mais espécies lipídicas não dependentes de oxigênio para sua síntese, de forma a sustentar o crescimento, ao custo de robustez e tolerância a fatores de estresse. Quando células cultivadas em anaerobiose foram expostas a condições de estresse ácido ou etanólico, foi possível observar uma perda de viabilidade mais acentuada, quando comparada a células retiradas de estados estacionários em aerobiose, expostas aos mesmos fatores de estresse. O presente trabalho documenta o impacto da extrema limitação de O2 em células de S. cerevisiae cultivadas em quimiostato.
LIST OF PUBLICATIONS AND AUTHORS' CONTRIBUTIONS
This thesis is based on the following publications, which consist the chapters 2, 3, and 4:
Chapter 2: DA COSTA, B. L. V.*; BASSO, T. O.*; RAGHAVENDRAN, V.; GOMBERT, A. K. G. Anaerobiosis revisited: growth of Saccharomyces cerevisiae under extremely low oxygen availability. Applied Microbiology and
Biotechnology, v. 78, n. 23, p. 8340–16, 2018.
*These authors contributed equally.
The idea of writing this review and the topics approached was first given by prof. Andreas Gombert. I wrote the first draft and structured the review, while Thiago Basso, Vijayendran Raghavendran and Andreas Gombert gave further contributions.
Chapter 3: DA COSTA, B. L. V.*; RAGHAVENDRAN, V.; FRANCO, L. F. M.; CHAVES-FILHO, A. B.; YOSHINAGA, M. Y.; MIYAMOTO, S.; BASSO, T. O.; GOMBERT, A. K. G. Forever panting and forever growing: physiology of Saccharomyces cerevisiae at extremely low oxygen availability in the absence of ergosterol and unsaturated fatty acids. FEMS Yeast
Research, v. 19, n.6, p. foz054, 2019.
Thiago Basso, Andreas Gombert and I designed the study. Thiago Basso and I designed the experimental set-up and I performed all the chemostat cultivations and lipid analysis. The fatty acids and sterol extractions were also performed by me, under supervision of Adriano Chaves-Filho, Marcos Yoshinaga, and Sayuri Miyamoto. Mathematical modeling and simulation were performed by prof. Luís Franco. I analyzed the data and wrote most of the manuscript while prof. Thiago Basso and prof. Andreas Gombert gave further contributions.
Chapter 4: DA COSTA, B. L. V.*; CHAVES-FILHO, A. B.; YOSHINAGA, M. Y.; MIYAMOTO, S.; BASSO, T. O.; GOMBERT, A. K. G. Lipidomics of Saccharomyces cerevisiae grown in steady-state anaerobic chemostats reveals changes in major lipid classes. Manuscript in preparation. I designed the study and performed the chemostat cultivations. The lipidomic analysis was performed by Adriano Chaves-Filho, Marcos Yoshinaga, and Sayuri Miyamoto. I analyzed the data and wrote most of the manuscript while Thiago Basso and Andreas Gombert gave further contributions.
LIST OF FIGURES
Fig. 1 Reported physiological responses to varying O2 saturation levels (ACEITUNO
et al., 2012; DENNY, 1993; WALDBAUER; NEWMAN; SUMMONS, 2011). ... 28
Fig. 2 The dissolved O2 concentration decreases with increasing temperature,
reaching as low as 202 µM at 40 °C (WEISS, 1970). ... 29
Fig. 3 Oxygen permeation from ambient air through a 7.62 m Teflon-PFA tubing that
is flushed with N2. Open circles refer to a N2 pressure of 4.4 atm; black triangles refer
to a N2 pressure of 7.8 atm (GIACOBBE, 1990). ... 30
Fig. 4 Effect of N2 flow rate on the dissolved O2 in the culture medium. Values are
taken from (VISSER et al., 1990). ... 30
Fig. 5 O2 permeability coefficients of commonly employed tubing material in Barrer.
Silicone has a permeability coefficient of 800 Barrer (MASTERFLEX®; SAINT-GOBAIN). ... 31
Fig. 6 Lipid biosynthesis in yeast. Figure adapted from (ROSENFELD; BEAUVOIT,
2003). ... 48
Fig. 7 Wash-out kinetics of K. lactis CBS 2359 after a switch from aerobic to anaerobic
mode (without AGF) in a glucose-limited chemostat culture. Vertical dashed line at t = 0 h indicates the switch from air to nitrogen in the gas supply. Symbols: biomass (solid square); residual glucose (solid triangle); glucose in feed medium vessel (open triangle); CO2 in the off-gas (solid circle); ethanol (open circle); glycerol (multiplication
sign). Dashed lines for biomass, glucose and ethanol shows the modelled kinetics of this experiment, with µ = 0. ... 63
Fig. 8 O2 diffusion measured in the culture medium with a polarographic probe. Slope
Fig. 9 Kinetics of S. cerevisiae CEN.PK113-7D after transition from air to N2 (ultrapure
+ oxygen trap) sparging, in a glucose-limited chemostat at a dilution rate of 0.1 h-1.
Dashed line at t = 0 indicates the gas switch from air to nitrogen. Symbols: biomass (solid square); residual glucose (solid triangle); glucose in feed medium vessel (open triangle); CO2 in the off-gas (solid circle); ethanol (open circle); glycerol (multiplication
sign). Compounds with negligible amounts detected (in g L-1) not shown in the graph:
succinic acid (0.02); lactic acid (0.08); acetic acid (0.09); pyruvate (0.02). ... 66
Fig. 10 Fatty acid content (A) and relative amounts of total fatty acids (B) for S.
cerevisiae CEN.PK113-7D (black bars) and PE-2 (gray bars) strains cultivated under different O2 availabilities. Error bars represent the average deviation of analytical
duplicate. ... 71
Fig. 11 Total fatty acids content in S. cerevisiae CEN.PK113-7D (black bars) and
PE-2 (gray bars) strains cultivated under different O2 availabilities. Error bars represent
the average deviation of analytical duplicate. ... 72
Fig. 12 Neutral lipid composition in yeast cells from glucose-limited chemostat cultures
at different conditions of oxygen and AGF supply. Panel A: absolute amounts of neutral lipids for S. cerevisiae CEN.PK113-7D (black bars) and PE-2 (gray bars) strains cultivated under different O2 availabilities. Panel B: relative amounts of neutral lipids
for S. cerevisiae CEN.PK 113-7D (black bars) and PE-2 (gray bars) strains cultivated under different O2 availabilities. Error bars represent the average deviation of analytical
duplicate. Insert below: structures of the three quantified neutral lipids, with a simplified ergosterol biosynthetic pathway, showing the precursor, the relevant intermediate, and its O2 requirements for synthesis. ... 73
Fig. 13 Loss-of-viability kinetics of cells collected from different steady-states. Control
or absence of stress condition (triangles), ethanol 100 g L-1 stress (squares) and pH
1.5 stress (circles). ... 75
Fig. 14 Relative amounts of major lipid classes in cells of S. cerevisiae during
glucose-limited chemostat cultures. Strains CEN.PK113-7D (black bars) and PE-2 (gray bars) were cultivated under different O2 availabilities (aerobic, anaerobic +AGF and
Cer - Ceramides; CL - Cardiolipins; LysoPC - Lysophosphatidylcholine; LysoPE - Lysophosphatidylethanoamine; LysoPI - Lysophosphatidylinositol; oPC - Plasmanylphosphatidylcholine; PA - Phosphatidic acid; PC - Phosphatidylcholine; PE - Phosphatidylethanolamine; PI - Phosphatidylinositol; PS - Phosphatidylserine. .... 89
LIST OF TABLES
Table 1 The purity of N2 and its cost. ... 31
Table 2 The amount of O2 entering the reactor when sparged with 0.5 L per min of N2
gas of varying purity. ... 32 Table 3 Anaerobic setups reported in the literature. ... 35 Table 4 Physiological parameters of Saccharomyces cerevisiae CEN.PK113-7D and PE-2 strains growing in an aerobic and anaerobic chemostats at a dilution rate of 0.1 h-1 using a synthetic medium with glucose as the limiting nutrient and sole carbon and
energy source. Specific rates (q) are given in mmol gDCM-1 h-1. Results are given as
average values from duplicate experiments ± deviation of the mean, unless otherwise stated ... 68 Table 5 Yeast sterol composition reported in literature ... 95
TABLE OF CONTENTS
1. INTRODUCTION ... 20
2. ANAEROBIOSIS REVISITED: GROWTH OF Saccharomyces cerevisiae UNDER EXTREMELY LOW OXYGEN AVAILABILITY ... 24
Abstract ... 25
Introduction ... 25
Discovery of anaerobic life ... 27
O2 in numbers ... 28
Permeability of tubing material ... 29
Anaerobic cultivation systems ... 32
The role of O2 in the metabolism of S. cerevisiae ... 34
O2 and MembraneLipids ... 38
How anaerobic are our anaerobic laboratory cultivations?... 42
Genetics and regulation of ergosterol and UFA biosynthesis in S. cerevisiae ... 46
Haem Biosynthesis ... 46
Sterol biosynthesis ... 47
UFA biosynthesis ... 49
Final remarks ... 49
Compliance with Ethical Standards ... 50
3. FOREVER PANTING AND FOREVER GROWING: PHYSIOLOGY OF SACCHAROMYCES CEREVISIAE AT EXTREMELY LOW OXYGEN AVAILABILITY IN THE ABSENCE OF ERGOSTEROL AND UNSATURATED FATTY ACIDS ... 51
Abstract ... 52
Introduction ... 52
Yeast strains and preservation ... 54
Composition of media ... 54
Chemostat cultivations ... 55
Nitrogen gas sparging during anaerobic chemostats ... 56
Fitness assays ... 57
Analytical methods ... 57
Determination of biomass and extracellular metabolites concentrations... 58
Elemental composition of cell biomass ... 58
Fatty acids extraction and quantification... 59
Neutral lipids extraction and quantification ... 59
Results ... 62
1. Validating the continuous cultivation set-up ... 62
1.1 K. lactis is washed out during an aerobic to anaerobic switch ... 62
1.2 O2 diffusing into the system from the surroundings becomes significant when the purity of the sparging N2 gas is increased ... 64
2. Saccharomyces cerevisiae does not wash-out after the switch from aerobic to anaerobic conditions in the absence of AGF ... 65
3. Physiology of laboratory and industrial S. cerevisiae strains cultivated under anaerobic conditions is severely altered in the absence of AGF ... 67
4. S. cerevisiae alters its lipid composition under severe O2 limitation ... 70
5. Growth under severe O2 limitation renders both laboratory and industrial S. cerevisiae less tolerant to low pH and ethanol stresses ... 74
Discussion... 75
Acknowledgements ... 81
4. LIPIDOMICS OF Saccharomyces cerevisiae GROWN IN STEADY-STATE ANAEROBIC CHEMOSTATS REVEALS CHANGES IN MAJOR LIPID CLASSES . 82 Abstract ... 83
Methods ... 83
Microbial strains and preservation ... 83
Medium composition... 84
Chemostat cultivations ... 84
Nitrogen gas sparging during anaerobic +AGF and -AGF chemostats ... 85
Lipid analysis ... 86
Results and Discussion ... 87
Acknowledgements ... 92
5. DISCUSSION ... 93
6. CONCLUSION ... 100
7. SUGESTIONS FOR FUTURE WORKS... 101
8. REFERENCES ... 103
20
1. INTRODUCTION
Anaerobiosis is a very important condition for industrial biotechnology (HATTI-KAUL; MATTIASSON, 2016; KÖPKE et al., 2014) since it is commonly achieved in industrial applications, such as in fuel ethanol production, in which large fermenters are used. Anaerobic bacteria and archaea, as well as thermostable enzymes produced by these microorganisms, have been the subject of several studies aiming at their potential application as industrial biocatalysts or for the synthesis of chemicals for the pharmaceutical industry (BRAGGER et al., 1989; LITTLECHILD, 2015). In the area of bulk chemicals, it is worth noting that anaerobic bacteria of the genus Clostridium have been for many years the workhorse for the so-called acetone-butanol-ethanol (ABE) fermentation (VISIOLI et al., 2014). Industrial vats contain in the range of millions of liters of fermentation medium, which easily leads to fully anaerobic environments in non-aerated processes. The low surface area to volume ratios in industrial vessels slow down transport phenomena such as mass and heat transfer. On the other hand, small-scale laboratory cultivations present high surface area to volume ratios, improving transport phenomena and making it difficult to exclude oxygen completely. Fully anaerobic lab-scale experiments are required to enable studies under industry-like conditions. In spite of several efforts, it remains rather difficult and expensive to establish such setups (VERDUYN et al., 1990b; VISSER et al., 1990). Thus, when trying to perform scale-down studies of anaerobic industrial bioprocesses, which are necessary to grow the microorganism of interest under conditions that resemble as much as possible the real industrial conditions, it is important to know how far oxygen is present (or not) and how far this affects overall microbial performance.
The yeast Saccharomyces cerevisiae is classified as a facultative anaerobe (VISSER et al., 1990), preferentially-fermenting microorganism (PRONK; STEENSMA; VAN DIJKEN, 1996), and is widely used in various biotechnological processes, from fuel ethanol (DELLA-BIANCA et al., 2013) to recombinant protein production (CECCARELLI; ROSANO, 2014), due to its culture simplicity, rapid growth, safe status and the possibility of achieving high cell density culture (YIN et al., 2007). In some of these applications, respiratory metabolism is required. Due to its low solubility in culture media, oxygen transfer to industrial bioreactors can be a problem, and
21
sometimes increasing stirring speed and gas flow rate are not enough to prevent O2
limitation (GARCIA-OCHOA et al., 2010). Oxygen transfer also increases production costs. The ability of S. cerevisiae to grow anaerobically, at similar rates when compared to aerobic growth (0.3 h-1 and 0.4 h-1 respectively) (VERDUYN et al., 1992),
is of particular interest for applications in which fermentative metabolism is demanded. The nutritional requirements of S. cerevisiae for optimal anaerobic growth were extensively investigated in the past and this is discussed in the review presented in Chapter 2 of this thesis, which corresponds to an article published in Applied Microbiology and Biotechnology. Also, other topics are addressed in this review, such as the debate on whether or not S. cerevisiae can grow anaerobically without supplementation of lipids in the culture medium, strategies to reach anaerobiosis in the laboratory environment, and lipid synthesis.
The development of a reliable and effective system for the cultivation of anaerobes dates back to the 1940´s. But it was only in 1969 that Robert Hungate published what would be the definitive cultivation system for these organisms, which consisted of a test tube with a thin layer of agar medium uniformly distributed over its internal surface, and flushed with an oxygen-free gas previously to inoculation (HUNGATE, 1969). Hungate's system was then improved by other groups (BALCH; WOLFE, 1976; BRYANT, 1972) and is still commonly used in laboratory practice for the cultivation of strict anaerobes (BÖRNER, 2016). Cultivation systems designed for oxygen-free environment where anaerobes can be handled appropriately has improved thereafter. A commercially available example is the widely used “anaerobic chamber”, where all routine laboratory manipulations can be performed in a confined anoxic environment. In this system, O2 is removed from the chamber by injecting gases such as N2, Ar, or
CO2 (PLUGGE, 2005; SPEERS; COLOGGI; REGUERA, 2009). Thus, anaerobic
chambers are the system of choice for static-anaerobic cultivations and for oxygen-free manipulations in the laboratory. However, although versatile, anaerobic chambers allow cultivations to be made in small shake or static flasks, thus making it impossible to collect data from continuously-grown microorganisms. Chemostats, for instance, are some of the few options available for trully quantitative physiological studies.
22
Although it is reported that S. cerevisiae can grow in the absence of oxygen when sterols and fatty acids are supplemented in the culture medium, not all yeast have the same ability. Some strict aerobic organisms, such as Kluyveromyces lactis, are commonly used as a negative control to ensure that the experimental setup is indeed anaerobic. K. lactis lacks a few genes that are present in S. cerevisiae (SNOEK; STEENSMA, 2006), including genes related to sterol uptake, rendering K. lactis cells incapable of growing without oxygen.
Molecular oxygen is needed not only for aerobic respiration but is actually involved in 48 reactions and accounts for 3% of the total reactions collected in the genome-scale model iTO977, proposed by Österlund et al. (2013). In fact, two molecules, namely ergosterol and oleic acid, are essential to provide functional properties for the cell membrane. When yeasts are subjected to anaerobiosis, oleic acid and ergosterol, the latter requiring 12 molecules of O2 for its synthesis, must be imported from outside the
cell. Therefore, growth under anaerobic conditions requires the exogenous supply of these two molecules (ANDREASEN; STIER, 1953; 1954). Therefore, it is assumed that ergosterol and oleic acid are essential for the growth of S. cerevisiae in the absence of oxygen.
However, no other work so far, to our knowledge, aimed at comparing the physiology and the lipid composition of cells grown in chemostats, under different oxygen availabilities. Thus, physiological aspects, as well as fatty acids and sterol synthesis, are discussed in the article presented in Chapter 3, which was recently submitted for publication. The information collected from chemostat steady states, which allowed the study of aerobic and anaerobic continuously grown yeast cells, was used to establish a cause/effect relationship between anaerobiosis, and yeast cell viability. Two S. cerevisiae strains were studied, with very different backgrounds: CEN.PK113-7D, a laboratorial strain, the workhorse of academic research, and PE-2, a strain isolated from a fuel ethanol mill. The study aimed to fill the gap of information on lipid composition of aerobically and anaerobically grown cells and the viability outcome when these cells are subjected to stressful conditions.
Considerable attention has been given by the scientific community to lipidomics studies, and to the relation between an unbalanced lipid composition of cells and
23
diseases. Lipids play important roles in physiology, serving as reserve of chemical energy in form of reduced molecules, as essential components of cellular membranes, and also as precursors for signalling molecules, vitamins and hormones (GUAN; WENK, 2006). Sterols, which are major membrane constituents, also play a role in stabilizing membranes and protecting cells against stress agents such as temperature (LAHTVEE et al., 2016) and or chemicals, such as ethanol (VANEGAS et al., 2012). Despite recent achievements on the understanding of how the lipid composition changes when cells are subjected to stressful conditions (GUO et al., 2018; LINDBERG et al., 2013) there is still a gap of information regarding aerobic and anaerobic growing yeast cells with respect to their lipidomics. Chapter 4 of this thesis corresponds to a description of the lipidomics of yeast cells collected from aerobic and anaerobic chemostat cultures.
The work presented in this thesis aimed at verifying whether the yeast S. cerevisiae is capable of growing under anaerobic conditions without lipid supplementation, and to examine to which extent the physiology of such cells is affected. Our results demonstrate that a fully anaerobic environment is not only hard to validate, but also difficult to be established in bioreactor cultivations. We also observed that, in spite of being capable of still growing at extremely low oxygen levels, S. cerevisiae cells exposed to these conditions do present decreased stress tolerance.
24
2. ANAEROBIOSIS REVISITED: GROWTH OF Saccharomyces cerevisiae UNDER EXTREMELY LOW OXYGEN AVAILABILITY
Article published in Applied Microbiology and Biotechnology
Bruno Labate Vale da Costa1,2,*, Thiago Olitta Basso2,*,Vijayendran Raghavendran1,3,
Andreas Karoly Gombert1#
1University of Campinas, School of Food Engineering, Rua Monteiro Lobato 80,
13083-862, Campinas-SP, Brazil
2University of Sao Paulo, Department of Chemical Engineering, Av. Prof. Luciano
Gualberto 380, 05508-010, São Paulo-SP, Brazil
3Chalmers University of Technology, Department of Biology and Biological
Engineering, Kemivägen 10, Göteborg, SE-412 96, Sweden
*both authors contributed equally to this work #Corresponding author:
Andreas Karoly Gombert, University of Campinas, School of Food Engineering, Rua Monteiro Lobato 80, 13083-862 Campinas-SP, Brazil, Phone: +55 19 35214031; e-mail: gombert@unicamp.br
Keywords: Anaerobiosis, Oxygen, Saccharomyces cerevisiae, chemostat cultivation, anaerobic growth factors
25
Abstract
The budding yeast Saccharomyces cerevisiae plays an important role in biotechnological applications, ranging from fuel ethanol to recombinant protein production. It is also a model organism for studies on cell physiology and genetic regulation. Its ability to grow under anaerobic conditions is of interest in many industrial applications. Unlike industrial bioreactors with their low surface area relative to volume, ensuring a complete anaerobic atmosphere during microbial cultivations in the laboratory is rather difficult. Tiny amounts of O2 that enter the system can vastly
influence product yields and microbial physiology. A common procedure in the laboratory is to sparge the culture vessel with ultrapure N2 gas; together with the use
of butyl rubber stoppers and norprene tubing, O2 diffusion into the system can be
strongly minimized. With insights from some studies conducted in our laboratory, we explore the question ‘how anaerobic is anaerobiosis?’. We briefly discuss the role of O2 in non-respiratory pathways in S. cerevisiae and provide a systematic survey of the
attempts made thus far to cultivate yeast under anaerobic conditions. We conclude that very few data exist on the physiology of S. cerevisiae under anaerobiosis in the absence of the anaerobic growth factors ergosterol and unsaturated fatty acids. Anaerobicity should be treated as a relative condition since complete anaerobiosis is hardly achievable in the laboratory. Ideally, researchers should provide all the details of their anaerobic setup, to ensure reproducibility of results among different laboratories.
Introduction
During evolution two significant O2 peaks occurred, at 500 Mya and 2000 Mya ago,
which resulted in two distinct growth spurts (PAYNE et al., 2009). Understandably oxygen is thus the most abundant element on earth (in earth’s crust and in air) and is constantly renewed by the photosynthetic action of green plants and algae. Bigger life forms emerged and flourished, owing to the vast amount of energy that was generated in the mitochondria, using O2 as the terminal electron acceptor in the oxidative
26
MARTIN, 2010). To extract the maximum free energy from the energy sources that were consumed, living beings undertook diverse adaptations in O2 poor environments
such as the deep sea (TYACK et al., 2006), high altitudes (HUERTA-SANCHEZ et al., 2014; SCOTT, 2011), or in deep glacial ice crystals (ROHDE; PRICE, 2007).
Although unquestionably O2 has contributed to complexity, microbes from the
pre-oxygenation event (e.g. obligate anaerobes) are of much industrial relevance for biotechnological processes (HATTI-KAUL; MATTIASSON, 2016; KÖPKE et al., 2014). Anaerobic bacteria and archaea are used in the production of thermostable enzymes (amylases, cellulases, lipase, pectinases, proteases, and xylanases) for use as industrial biocatalysts, for the synthesis of chiral compounds for the pharmaceutical industry (BRAGGER et al., 1989; LITTLECHILD, 2015), or for the production of bulk chemicals (acetone-butanol-ethanol, medium-chain carboxylic acids) (JEON et al., 2016; VISIOLI et al., 2014).
Anaerobic processes are likely to be more economical than their aerobic counterparts because of the reduced costs in aeration and mixing (CURRAN; SMITH; HOLMS, 1989; DE BECZE; LIEBMANN, 1944). Industrial fermenters often have very low surface area compared to their volume1 resulting in reduced heat and mass transfer
rates (SIMPSON; SASTRY, 2013). In a typical first-generation sugarcane ethanol plant, as much as 23,000 m3of CO
2 is generated in a reactor of 500 m3. Such large
volume of CO2 not only displaces the dissolved O2 out of the liquid phase but also
ensures a completely anoxic atmosphere. In contrast to industrial practice, complete anaerobiosis is extremely difficult to establish in the laboratory (where surface area to volume ratios, and thus mass transfer, are high), and it is quite a challenge to exclude O2 completely from the cultivation systems. While performing scale-down studies of
anaerobic industrial bioprocesses, it is imperative to grow the microorganism under
1 A cylindrical reactor with height (h) and radius (r) has a lateral surface area of 2πrh and a volume of
πr2h. The surface area to volume is inversely proportional to the radius of the reactor; thus, the larger
the reactor, smaller the surface area to volume. This has profound consequences for heat and mass transfer. Heat transfer is proportional to the surface area, while the metabolic heat generation is proportional to the culture volume. Thus, at very large volumes (& large radius), the available heat transfer area is insufficient to dissipate the heat that is generated. Unlike laboratory reactors which are well mixed, there will be concentration gradients in large scale reactors affecting the mass transfer of O2, as well as other nutrients, vastly affecting the cellular physiology.
27
conditions that closely resemble the real industrial conditions, as the presence of even trace amounts of O2 could affect the overall performance of the microbial cell factories.
In this mini-review, we revisit the much-studied anaerobic fermentation in yeast focusing on the cultivation systems and the role of O2 in non-respiratory pathways. We
illustrate the challenges in mimicking an anaerobic atmosphere in the laboratory and the ways to minimize the leakage of O2 into the system. We provide a chronological
list of all the attempts thus far made to create an anaerobic set-up, as well as the physiology of Saccharomyces cerevisiae under anaerobic conditions with emphasis on lipid composition. And we conclude with some perspectives on future research and the need to exercise caution when one declares a set-up as anaerobic.
Discovery of anaerobic life
The discovery of the so-called “anaerobic bacteria” dates back to 1680 when Antonie van Leeuwenhoek observed ‘a kind of living animalcules’ in a small heat-sealed glass vial which was previously filled with crushed pepper powder and clear/clean rainwater (GEST, 2004). In that condition, the environment inside the glass vessel became anaerobic owing to the depletion of O2 by the aerobic microorganisms. When the
sealed glass vial was opened, an overpressure forced the liquid out (GEST, 2004). The pressure inside the vial was due to the formation of CO2 via fermentation.
Leeuwenhoek's experiment was repeated by Martinus Beijerinck in 1913, who identified the predominant microorganism as Clostridium butyricum. This species, as most Clostridium species, is classified as an obligate anaerobe – absence of growth in the presence of O2, due to their inability to deactivate the reactive oxygen species
(ROS) such as peroxides (O'BRIEN; MORRIS, 1971). However, obligate anaerobes do tolerate micro-oxic conditions when grown in liquid medium (IMLAY, 2008; KATO; FIELD; LETTINGA, 1997; KAWASAKI et al., 2005). Aerobes, on the contrast, produce catalase and superoxide dismutase, two key enzymes that detoxify peroxides and other ROS that cause oxidative damage to DNA, lipid molecules, and proteins (STORZ et al., 1990). Anaerobes produce these enzymes to a certain extent, but they possibly
28
have other ways to reduce the oxidative damage. Thus, excluding O2, either partially
or totally, is a necessary requirement for those investigating obligate anaerobes.
O2 in numbers
Under normal atmospheric conditions, O2 constitutes 20.948 mole percent in air, which
makes the partial pressure2 of O
2 as0.20948 atm or 20.67 kPa. Using Henry's law
(constant = 756.7 atm L mol-1), the concentration of O2 in wateris ~ 280 μM at 25 °C.
Fig. 1 shows some physiological responses to varying levels of O2 saturation in water.
The solubility of O2 in water increases with decreasing temperature, reaching as high
as 400 μM at 0 °C (DENNY, 1993; WEISS, 1970) (Fig. 2).
Fig. 1 Reported physiological responses to varying O2 saturation levels (ACEITUNO et al., 2012; DENNY, 1993; WALDBAUER; NEWMAN; SUMMONS, 2011).
2 Partial pressure of O
29
Fig. 2 The dissolved O2 concentration decreases with increasing temperature, reaching as
low as 202 µM at 40 °C (WEISS, 1970).
Permeability of tubing material
The choice of the tubing material used in the cultivation system influences the amount of O2 that diffuses into the system. Silicone tubing that is regularly used in aerobic
cultivations has a permeability coefficient of 800 Barrer3. For this reason, anaerobic
cultivations are often carried out with norprene tubing which has a 40-fold lower permeability than silicone tubing4. Giacobbe (1990) observed that the diffusion of O
2
through a seven meter long coiled teflon tubing was inversely related to the flow rate of N2 used, as shown in Fig. 3. Visser and co-workers (VISSER et al., 1990) reported
that the dissolved O2 concentration had an asymptotic relationship with respect to the
N2 flow rate employed, as depicted in Fig. 4, even after 35 h of continuous flushing. A
comparison of the permeability coefficients of various tubing materials is given in Fig. 5; cost is a major factor while performing anaerobic experiments, as the price of N2 gas
3 Barrer is a non-SI unit for gas permeability. 1 Barrer = 10-10 ∙ cm3 ∙ cm cm2 ∙ s ∙ cmHg
4 For a tubing of 30 cm length, having an internal diameter of 0.31 cm, the diffusion rate of O2 can be calculated using this relation, for a partial pressure of O2 of 15.6 cm Hg and molar volume of 22,400 cm3: O
2 diffusion rate " µmol
h # =
surface area cm2 ∙ Permeability ∙ P O2 cm Hg thickness cm
The rate of O2 diffusing through a norprene tubing is 1.5 µmol h-1, while it is 59 µmol h-1 with a silicone tubing.
30
increases with increasing purity, as shown in Table 1. The amount of O2 (present as
impurity) entering the system is shown in Table 2.
Fig. 3 Oxygen permeation from ambient air through a 7.62 m Teflon-PFA tubing that is
flushed with N2. Open circles refer to a N2 pressure of 4.4 atm; black triangles refer to a N2
pressure of 7.8 atm (GIACOBBE, 1990).
Fig. 4 Effect of N2 flow rate on the dissolved O2 in the culture medium. Values are taken from
31
Fig. 5 O2 permeability coefficients of commonly employed tubing material in Barrer. Silicone
has a permeability coefficient of 800 Barrer (MASTERFLEX®; SAINT-GOBAIN).
Table 1 The purity of N2 and its cost.
Product Volume of the commercially sold cylinder (m3)
O2 as impurity (ppm) Price (Euros) Nitrogen 10 <50 ca. € 50 Nitrogen 4.6 9 <5 € 108 Nitrogen 5.2 10 <3 € 192 Nitrogen 6.0 9* <0.5 € 302 Oxitrap* - <0.015 € 120
Gas mix (85% N2, 10% CO2, &
5% H2) for anaerobic chamber 8
ѳ <5Ω € 480
A typical chemostat (in triplicate) at a dilution rate of 0.1 h-1 carried out for at least 50 h (five residence times) sparged with 0.5 L of N2 gas per L of medium per min, would consume 4.5 m3 of N2 gas. A batch experiment carried out for 20 h under the same condition would consume 1.8 m3 of N2 gas. Prices converted from Brazilian Reais to Euros.
ѳGas flow in an anaerobic chamber is up to 25 L min-1; H2 gas greater than 5% poses a severe risk of explosion.
*One Oxitrap (Sigma Aldrich) will be able to capture 5 mL of O2, equivalent to One N2 cylinder with 50 ppm O2 as impurity.
32
Table 2 The amount of O2 entering the reactor when sparged with 0.5 L per min of N2 gas of
varying purity.
O2 in a N2/O2 mixture (ppm) O2 entering the reactorð
(µmol. h-1) O2 transfer ¤ rate (µmol. h-1) 210,000 (air) 253,377 99,907 50 60 23.8 10 12 4.8 1 1.2 0.5 0.1 0.12 0.05
ð Calculated using the formula $∗&∗'
(∗)∗*; where f is the fraction of O2 in the incoming gas, V is the flow rate of the gas in L. min-1 at 30 oC at 1 atm; R is the universal gas constant
¤O2 dissolving into the liquid is calculated using this formula +,- = /
01∗ 2∗∗ 30, assuming a kLa of 0.1 s-1, C* is the saturation concentration of O2 at the bubble interface; this can be calculated from Henry’s law 4 = 56∗ 7 where p is the partial pressure of O2, KH is Henry’s constant, and c is the concentration in water.
Anaerobic cultivation systems
Owing to the omnipresence of O2, ensuring a complete anaerobic atmosphere during
microbial cultivation in bioreactors is rather difficult. There are three ways to decrease the O2 tension during cultivation: firstly, biological reduction using symbionts; secondly,
physical reduction by boiling, evacuation or the use of inert gases; thirdly, chemical reduction, such as the catalytic ignition of hydrogen and residual O2 or using chemicals
such as cysteine. However, cysteine is toxic to S. cerevisiae as it interferes with its metabolism (MAW, 1961). Once the O2 tension is reduced, it must be maintained low
(or absent) during the cultivation by sealing off the medium. For further details, the reader is urged to consult the comprehensive review by Hall (1929). As the solubility of O2 is lower at high temperatures, boiling is often the quickest way to dispel the O2
trapped in the liquid. However, this is cumbersome and suitable for complex medium only. Experiments done in a synthetic medium pose additional difficulty, as the addition of filter- sterilized heat labile components inevitably introduce some O2 into the system.
To indicate the O2 present in the cultivation medium, it is common to use a
redox-sensitive dye such as resazurin (TWIGG, 1945) or methylene blue (BREWER; ALLGEIER; MCLAUGHLIN, 1966) which turns colourless in the absence of O2;
33
Wimpenny and Necklen (1971) have used redox potential to measure the O2 tension
in a chemostat cultivation with Escherichia coli and Klebsiella aerogenes and a redox potential of <0 mV was considered as anaerobic.
Respiratory inhibitors such as antimycin A have been regularly used to study the fermentative capacity of yeasts, aiming at mimicking a quasi-anaerobic system. Antimycin A inhibits electron transfer from quinone to cytochrome b. Addition of acetoin or high concentrations of lysine or glutamate restored the growth defect in the presence of antimycin A by acting as a redox sink (MERICO et al., 2007). Thus, the inability to grow is mainly due to redox imbalance as the mitochondrial NADH is not reoxidized to NAD+. Thus, efficient mechanisms to maintain the redox balance are a prerequisite for
growth under low supply of O2, requiring a well-regulated network (MERICO et al.,
2009). Great care must be taken when interpreting the results from such experiments, as traces of O2 can affect the physiology even in the presence of antimycin A.
Although the development of a reliable and effective system for the cultivation of anaerobes dates to the 1940s, it was only in 1969 that Robert Hungate published what would be the definitive cultivation system for anaerobic organisms. It consisted of a test tube with a thin layer of agar medium uniformly distributed over its internal surface and flushed with an O2-free gas prior to inoculation (HUNGATE, 1969). Hungate's
system was then improved by other groups (BALCH; WOLFE, 1976; BRYANT, 1972) and it is still commonly used in laboratory practice for the cultivation of strict anaerobes (BÖRNER, 2016). Cultivation systems designed for an O2-free environment where
anaerobes can be handled appropriately has improved thereafter. A commercially available example (since 1969) is the widely used “anaerobic chamber”, with which all laboratory routine and manipulation can be performed in a confined anoxic environment. In this system, O2 is removed from the chamber by injecting gases such
as N2, H2, and CO2 (THOMAS; HYNES; INGLEDEW, 1998) or N2, Ar, or CO2.
(PLUGGE, 2005; SPEERS; COLOGGI; REGUERA, 2009) and any trace of O2 present
is reduced to water on a palladium catalyst. Thus, anaerobic chambers are the system of choice for static-anaerobic cultivations and for O2-free manipulation in the
laboratory. The various systems developed for anaerobic setup in the laboratory are summarized in Table 3.
34
A common procedure for the anaerobic cultivation of yeast in laboratory bioreactors (in which detailed physiological studies can be carried out) is to constantly flush the culture medium with ultrapure N2 gas (containing less than 5 ppm O2), together with the use
of Viton O-rings and norprene tubing. In this way, O2 diffusion into the system is
minimized (BOENDER et al., 2009; DE KOK et al., 2011).
The role of O2 in the metabolism of S. cerevisiae
The yeast S. cerevisiae is classified as a facultative anaerobe (VISSER et al., 1990), preferentially-fermenting microorganism (BARNETT, 2003; PRONK; STEENSMA; VAN DIJKEN, 1996), and is widely used in various biotechnological processes, from fuel ethanol (DELLA-BIANCA et al., 2013) to recombinant protein production (CECCARELLI; ROSANO, 2014), due to its culture simplicity, rapid growth, safe status and the possibility of achieving a high cell concentration culture (YIN et al., 2007). Some of these applications demand a respiratory metabolism that can be a problem due to the low solubility of O2 in the culture media (HANOTU; KONG; ZIMMERMAN,
2016). Increasing the stirring speed and the gas flow rate can increase the O2 transfer
but does not prevent O2 limitation (GARCIA-OCHOA et al., 2010), with a concomitant
increase in the production costs. The ability of S. cerevisiae to grow anaerobically, at similar rates when compared to aerobic growth (~0.3 h-1 and ~0.4 h-1 respectively)
(VERDUYN et al., 1992), is of particular interest for applications in which fermentative metabolism is demanded. Moreover, S. cerevisiae plays an important role as a model organism for studies on microbial physiology and genetic regulation under various environmental conditions, such as anaerobiosis (JOUHTEN; PENTTILÄ, 2014). Respiration is an energetically efficient process (in terms of ATP generation) in which O2 participates as the final electron acceptor. Electrons are transferred from the energy
source, via reduced coenzymes such as NADH, to O2 in the mitochondrial electron
transfer chain, generating a proton-motive force that enables ATP synthesis by the enzyme ATP synthase (JOUHTEN; PENTTILÄ, 2014). In the absence of O2, free
Table 3 Anaerobic setups reported in the literature.
Anaerobic conditions employed Microbial strain (if applicable) Reference
Excellent review on the various ways to achieve anaerobic conditions Obligate anaerobic bacteria (HALL, 1929) A stream of N2 gas was passed through a solution containing acidified vanadyl
sulphate solution and amalgamated Zinc - (MEITES; MELTES, 1948)
Use of 99.99% pure N2 with 0.01% O2. All glass with mercury traps and three-way cock for running a chemostat cultivation
A distillery type yeast; a strain S. cerevisiae SC-1 (DCL), obtained from
Joseph E Seagram and Sons, Inc., Louisville, Kentucky
(STIER; SCALF; BROCKMANN, 1950)
Use of 99.99% pure N2 passed through a solution of chromium chloride to
decrease O2 to ca. 1 ppm S. cerevisiae SC-1 (DCL)
(ANDREASEN; STIER, 1953; 1954)
Closed flask fitted with a rubber stopper sealed with molten parafilm wax with an airlock and medium flushed (for 45 minutes, after inoculation with O2-free N2 stream (Meites and Meites 1948)
A locally isolated diploid strain of
Saccharomyces cerevisiae
(JOLLOW; KELLERMAN; LINNANE, 1968; WALLACE; HUANG; LINNANE, 1968) Agar medium distributed as a thin layer over the internal surface of test tubes
maintained in an anaerobic atmosphere using CO2 gas Bacteria (HUNGATE, 1969)
Use of copper oven to reduce the last traces of O2 Bacteria (GORDON; DUBOS, 1970)
Coy anaerobic chamber (Patent no: US 61000830) - (COY LAB PRODUCTS,
1969) 2 L round flat-bottomed Pyrex flask with a latex-rubber port on the side and fitted
with a water lock allowing CO2 gas to exit the system. The growth medium was autoclaved and whilst it was warm, flushed with a stream of N2 passed through an
O2 trap (Nilox scrubber)
S. cerevisiae NCYC 366 (ALTERTHUM; ROSE, 1973)
N2 flushing during medium preparation; serum bottle closed with a butyl rubber stopper with a crimped metal seal
Table 3 (continued)
Anaerobic conditions employed Microbial strain (if applicable) Reference
Anaerobic shake flask equipped for continuous gassing and measurement of
culture density Bacteria (DANIELS; ZEIKUS, 1975)
Pressure vessel containing a gas mixture of 80% hydrogen and 20% carbon
dioxide at 2 to 3 atm Bacteria (BALCH; WOLFE, 1976)
Bespoke design of serum vial with in situ cell measurement, and gassing port; Traces of oxygen are removed from N2 of 99.995% purity with the help of an Oxisorb cartridge
S. cerevisiae D 273-10B (BIEGLMAYER; RUIS, 1977)
Use of Resazurin, cysteine, boiling, CO2 flush. Oxygen was removed from the CO2 by passing it through a vertical Pyrex column packed with copper metal turnings heated electrically to approximately 350 oC
The Montrachet strain of wine yeast,
UCD enology No. 522 (MACY; MILLER, 1983)
N2 gas containing less than 5 ppm O2 obtained by passing N2 gas containing less than 100 ppm O2, through a column filled with copper turnings and heated to 350 oC in a bioreactor
S. cerevisiae CBS 8066 (SCHULZE et al., 1996)
N2 gas (<0.5 ppm) is passed through an Oxyclear O2 absorber to reduce its
residual oxygen level to below 50 ppb in a bioreactor Saccharomyces cerevisiae JM43 (BURKE et al., 1998) Anaerobic chamber, with palladium pellets as a catalyst, charged with a gaseous
atmosphere consisting of hydrogen, carbon dioxide, and nitrogen in the ratio of 10 :10 : 80. Use of heated (400 °C) copper turnings to reduce residual oxygen levels from pre-purified N2 gas (< 5 ppm O2)
S. cerevisiae NCYC 1324 along with
several brewing and fuel ethanol industrial strains.
(THOMAS; HYNES; INGLEDEW, 1998)
Use of ultrapure N2 followed by the O2 trap. <15 ppb was achieved by passing the
37
(in the form of ATP) is exclusively conserved from substrate-level phosphorylation, via fermentation. The number of ATP moles generated for each mole of substrate consumed in fermentation - which is 2 when glucose is the energy source - is considerably lower than the corresponding yield observed under respiratory metabolism (this is not a precise number but can be assumed to be roughly 15 times higher than in fermentative metabolicm). In response to this lower yield, and to fulfil the cell’s energetic requirements, a 7.5 times higher glycolytic flux (in terms of glucose uptake rate) (JOUHTEN et al., 2008), and consequently a higher ATP production rate through glycolysis, are observed in fermentative metabolism, when compared to respiratory metabolism. In addition, to maintain the redox balance under anaerobic growth, NAD+ is regenerated by alcoholic fermentation and via the formation of
glycerol.
Bisschops and co-workers (2015) report that anaerobic stationary phase cultures of S. cerevisiae had a shortened chronological lifespan and low robustness (assessed through viability and temperature tolerance, and the adenylate energy charge) compared to aerobic stationary phase cultures. This has implications for cell recycling in industry. Merico and others (2007) conducted an exhaustive work on the fermentative lifestyle of over 40 yeasts belonging to the Saccharomyces complex, which includes 14 distinct clades comprising several genera including e.g. Kluyveromyces, Zygossacharomyces, Kazachstania, Naumovia, among others, to reflect 150 million years of evolutionary history. Most of the yeasts exhibited good fermentation ability but only those lineages (Saccharomyces, Kazachstania, Naumovia, Nakaseomyces and Tetrapisispora) that underwent whole genome duplication could grow in the absence of O2.
Molecular O2 is involved in 48 reactions and accounts for 3% of the total reactions
collected in the genome-scale model iTO977, proposed by Österlund and co-workers (2013), in the yeast S. cerevisiae. Two molecules, namely ergosterol and oleic acid, which are essential to provide functional properties of the cell membrane, require O2
for their biosyntheses, and they must be imported from the extracellular medium to the cell, under full anaerobiosis. Therefore, growth under anaerobic conditions requires the exogenous supply of these two molecules (ANDREASEN; STIER, 1953; 1954). However, there are other non-respiratory pathways in S. cerevisiae, that also require
38
molecular oxygen, like biosynthesis of haem, hemoproteins and several other oxidases (ROSENFELD; BEAUVOIT, 2003).
O2 and MembraneLipids
Investigations regarding the lipid composition of S. cerevisiae’s membrane started with observations made in laboratory scale cultivation systems claimed to be anaerobic, which are known to still allow small amounts of oxygen to be present (more details in the following section). On the other hand, these systems, being sufficiently oxygen-limited, induce different physiological and biochemical responses in S. cerevisiae, when compared to systems without oxygen limitation.
Under anaerobic conditions, S. cerevisiae grows poorly in culture media containing only water-soluble ingredients. However, the addition of non-saponifiable lipids from edible oils, such as wheat germ, into the culture medium stimulated growth of the yeast (STIER; SCALF; PETER, 1950), and this initial observation prompted the search for anaerobic growth requirements.
Andreasen and Stier (1953) used an anaerobic setup (detailed in Table 3) and a synthetic medium to test the effect of several non-lipid compounds (such as nucleic acids, purines, pyrimidines, amines, simple peptides, vitamin B12, and casein hydrolysates) on the growth of S. cerevisiae SC-1 strain under fully anaerobic conditions. Because these compounds did not increase the yield of cells under such conditions, they chose to add ergosterol into the medium, as it was known that ergosterol is an important constituent of yeast biomass. Tween 80, on the other hand, was used only as a surfactant to facilitate solubilisation of ergosterol in the synthetic medium. The effects of Tween 80 alone, and of Tween 80 plus ergosterol additions were assessed, and as a conclusion, ergosterol was found to be an essential anaerobic growth factor for S. cerevisiae. It was found later, that the Tween 80 used to solubilize ergosterol, was also supplying another essential requirement for anaerobic growth, oleic acid (ANDREASEN; STIER, 1954). Tween 20 and Tween 40 (sources of saturated fatty acids) were then tested, as well as a non-lipid surfactant to solubilize ergosterol, to clarify whether another lipid ester would promote the same
39
results. As a conclusion, oleic acid and ergosterol, simultaneously, were found to be growth factors for S. cerevisiae’s anaerobic growth. Later, several groups investigated the effect of trace amounts of O2 on the anaerobic growth of S. cerevisiae in a
chemically-defined medium, without exogenous addition of lipids. Jollow, Wallace and co-workers (JOLLOW; KELLERMAN; LINNANE, 1968; WALLACE; HUANG; LINNANE, 1968) cultivated a diploid strain of S. cerevisiae under several different growth conditions, to assess the occurrence of mitochondrial profiles and other cell membrane systems. Their anaerobic setup consisted of a closed flask fitted with a rubber stopper sealed with molten parafilm wax with an airlock. The medium was flushed (after inoculation, for 45 min) with N2 that had been bubbled through an
acidified solution of vanadyl sulphate reduced with amalgamated zinc to provide an O2-free gas (MEITES; MELTES, 1948). Even though the yeast extract based medium
contains lipids such as ergosterol and UFAs, remarkable differences were observed between cells cultivated in aerobic YPD and anaerobic YPD cultivation (with and without Tween 80 and ergosterol). Although total lipid content varied between the conditions tested, the most pronounced changes were in total fatty acid, sterol fractions, glycerides and phospholipid contents. In general, total lipid, ergosterol, glycerides, and phospholipid contents of cells grown anaerobically were lower, when compared to cells cultivated in aerobiosis. On the other hand, anaerobiosis triggered squalene accumulation, with a reduction in ergosterol levels. Aerobically-grown yeast presented a predominance of mono-UFAs (C16:1 and C18:1). On the other hand, cells cultivated under anaerobiosis without Tween 80 and ergosterol addition presented predominantly saturated fatty acids, comprising: C10:0, C12:0 and C16:0. When ergosterol and Tween 80 were added to the YEG medium, C16:1 and C18:1 species accounted for approximately 70 % of fatty acids, still less than what was synthesized by aerobically grown cells (85%). Thus, it can be observed that the choice of the cultivation medium and conditions are very crucial in understanding microbial physiology.
To expand the range of UFAs that could be incorporated into yeast, Alterthum and Rose (1973) used growth media with different fatty acid compositions (oleic, linoleic, or γ-linolenic acid as a source of UFA) to anaerobically cultivate S. cerevisiae NCYC 366. In all the conditions, cells grew to the same extent and did not differ in their content
40
of total lipids or total phospholipids. However, cells were enriched by approximately 60% with the specific fatty acid that was supplemented. Sterol composition changed according to the availability of O2. These results reveal the flexibility of S. cerevisiae to
incorporate even polyunsaturated fatty acids such as Omega-6 as building blocks for structural and bulk storage lipids. Anaerobically grown S. cerevisiae cells accumulated squalene, which does not require O2 for its synthesis, which is in accordance with
Jollow et al. (1968). The ratio between squalene/ergosterol as a function of O2
availability, although intriguing, was not correlated to any environmental (dis)advantage.
Watson and Rose (1980) followed the studies conducted by Alterthum and Rose (1973) on the extent to which exogenous FAs are incorporated into lipids in the S. cerevisiae NCYC 366 strain. The composition of aerobically/anaerobically grown S. cerevisiae in defined media containing ergosterol and different fatty acids (oleic, linoleic and α-linolenic) was investigated. Fatty-acyl composition of phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylinositol (PI), phosphatidylserine (PS) and triacylglycerols from anaerobically grown cells showed predominance of C18:1, C18:2, and C18:3 species when the medium was supplemented with oleic (C18:1), linoleic (C18:2), and α-linoleic (C18:3) acids, respectively, followed by C16:0 under all the conditions tested. Jollow et al. (1968) also observed the predominance of C18:1 fatty acid when Tween 80 was added to the anaerobic medium formulation, but rather than C16:0, significant amounts of C16:1 were detected. In aerobically grown cells, C16:0 plus C16:1 accounted for the major fraction of the fatty-acyl composition of PC, PE, PI, PS, and triacylglycerols.
Rosenfeld and Beauvoit (2003) reviewed and compiled data of total FA (saturated, unsaturated, and phospholipid fractions) and sterol (squalene and ergosterol fractions), highlighting the accumulation of squalene over ergosterol in cells cultivated in anaerobic conditions. Klose et al. (2012) examined the lipidome at the level of lipid classes, of aerobically grown S. cerevisiae cultivated in complex media with different carbon sources, different temperatures, and growth phases (early and middle logarithmic and early stationary phases) at 95% lipidome coverage. In their study, the flexibility of the yeast's lipidome was confirmed, in face of the variability of triacylglycerol content, and the ratio of unsaturated/saturated glycerophospholipids.
41
On the other hand, a low variability of sphingolipid and ergosterol abundances was observed between the conditions tested. Lindberg et al. (2013) cultivated S. cerevisiae CEN.PK113-7D and Zygosaccharomyces bailli under aerobiosis to investigate the high tolerance of Z. bailli to acetic acid. Upon exposure of Z. bailli cells to acetic acid, they observed an increase of monounsaturated fatty acids, as well as a higher level of complex sphingolipids. This led to a membrane with a lower fluidity that prevented the entry of un-dissociated acetic acid, thereby conferring higher acetic acid tolerance when compared to S. cerevisiae.
In wine fermentations, it is common practice to add O2 in the stationary phase to rescue
sluggish fermentations (FORNAIRON-BONNEFOND et al., 2002; VALERO; MILLÁN; ORTEGA, 2001). The squalene contents of the cells decreased upon O2 addition but
the ergosterol content did not increase proportionately relative to the total sterol levels. In air saturated beer wort fermentation, the sterol content increased rapidly from 1 mg gDCM-1 to 10 mg gDCM-1 and then declined rapidly at the end of the fermentation,
when O2 was depleted (ARIES; KIRSOP, 1977).
Most of the investigations mentioned above studied the macromolecular physiology by measuring the lipid composition using a liquid chromatography system and the focus was more on the anaerobic physiology from a process engineering point of view. However, Waldbauer and co-workers (2011) were interested in identifying the molecular fossil record of archaea and resorted to the microaerobic sterol biosynthesis. They used 13C labelling studies to investigate the role of very low O2 concentrations (1
nM to 1 µM) on the enzymatic biosynthesis of sterols. Three different O2 concentrations
were tested (6.5 μM, 0.6 μM, and 7 nM), including one anaerobic condition (< 0.7 nM). They used unlabeled ergosterol and 13C glucose and followed the incorporation of
carbon label from glucose to steroids using 13C-NMR. Steroid biosynthesis occurred at
each of the three dissolved O2 concentrations tested in their experiments: 6.5 μM, 0.6
μM, and 7 nM, as 13C label was observed in squalene—the last steroid biosynthetic
intermediate that can be produced in the absence of O2, as well as in lanosterol (the
first O2 requiring step in steroid synthesis), demonstrating de novo steroid production.
However, only under anaerobic conditions (< 0.7 nM), lanosterol was not detected, and only unlabeled ergosterol that was taken up by the cell exogenously could be detected. Thus, yeast can still grow and multiply at such ‘anaerobic conditions’.
42
How anaerobic are our anaerobic laboratory cultivations?
Studies on nutritional requirements of S. cerevisiae under anaerobic conditions were initiated in the 1950s, when Andreasen and Stier determined the amount of ergosterol and fatty acid contents necessary in a defined medium to provide for adequate anaerobic growth (ANDREASEN; STIER, 1953; 1954). Their research was based on the evidence that S. cerevisiae grown in a complex medium (yeast extract), under anaerobic conditions, were described as “small” (in terms of cell concentration in the culture medium), while under aerobic conditions (in the presence of O2), the growth
was referred to as being “excellent” (BROCKMANN; STIER, 1947). In their studies, N2
gas with a purity of 99.99 % was employed, and in some cases, the N2 gas was
sparged through a solution of Chromium(II) chloride to further decrease the O2
concentration to ca. 1 ppm. It was concluded that for anaerobic growth of S. cerevisiae in a chemically defined medium, ergosterol and a source of UFA (such as Tween 80) must be added to the culture medium. They also concluded that, for aerobic conditions, the addition of both Tween 80 and ergosterol had a negative effect on growth. Their recommendation as regards to the necessity of adding ergosterol and UFAs for anaerobic growth conditions was then corroborated by several authors during the two subsequent decades (ALTERTHUM; ROSE, 1973; DAVID; KIRSOP, 2013; KOVÁČ et al., 1967; PALTAUF; SCHATZ, 1969).
However, contradicting Andreasen and Stier's recommendations, Macy and Miller (1983), using the Hungate protocol for anaerobic cultivation, did observe S. cerevisiae growth after a 12 h lag phase in a defined medium lacking both ergosterol and fatty acids. The authors also emphasized that previous investigations stated as having been conducted under anaerobic conditions were not carried out under the complete absence of O2. This observation raised the question: "How anaerobic are our
laboratory anaerobic cultivations?", since O2 is necessary for UFAs and sterols
biosyntheses.
The observation made by Macy and Miller prompted Verduyn and co-workers (1990b) to investigate whether S. cerevisiae was able to grow, and if so, to what extent, in a defined medium under anaerobic conditions. Although they did observe growth in the absence of Tween 80 in anaerobic chemostat cultivations at a dilution rate of 0.1 h-1,
