XP-OO2236029
`
`Appl Microbiol Biotechnol (1998) 50: 434439
`
`© Spt‘inger-Verlag 1998
`
`ORIGINAL PAPER =
`
`-
`
`H. Valadi ‘ C. Larsson - L. Gustafsson
`
`
` no. annealing
`
`
`Improved ethanol production
`by glycerol-3—phosphate dehydrogenase mutants
`of Saccharomyces cerevisiae
`
`P. ”Shem
`
`Received: 16 February 1998 I Received revision: 16 March 1998 [Acceptor]: 1 June 1998
`
`Abstract The anaerobic performance of gdeA and
`gdeA mutants of Saccharomyces cerevisiae was char-
`acterized and compared to that of a wild-type strain
`under Well—controlled conditions by using a high~per—
`formance bioreactor. There was a 40% reduction in
`glycerol level in the gpd2A mutant compared to the wild—
`type. Also the gdeA mutant showed a slight decrease in
`glycerol formation but to a much lesser degree. As a
`consequence, ethanol formation in the gpd2A mutant
`was elevated by 13%. In terms of growth, the gdeA
`mutant and the wi1d«type were indistinguishable. The
`gpd2A mutant, on the other hand, displayed an extended
`lag phase as well as a reduced growth rate under the
`exponential phase. Even though glycerol-3-phosphate
`dehydrogenase 2 {GPD2) is the important enzyme under
`anaerobic conditions it can, at least in part, be substi-
`tuted by GPDl. This was indicated by the higher ex-
`pression level of GPDI in the gpd2A mutant compared
`to the wild type. These results also show that the cells are
`able to cope and maintain redox balance under anaer-
`obic conditions even if glycerol formation is substan-
`tialIy reduced, as observed in the gpd2A mutant. One
`obvious way of solving the redox problem would be to
`make a biomass containing less protein, since most of
`the excess NADH originates from amino acid biosyn-
`thesis. However, the gdeA mutant did not show any
`decrease in the protein content of the biomass.
`
`Introduction
`
`The production of glycerol by Snccizaromyces cerevzlsiae
`under anaerobic conditions with glucose as a carbon
`
`H. Valadi - C. Larsson (18%) - L. Gustafsson
`Lundberg Laboratory,
`Department of General and Marine Microbiology,
`Box 462, 5-405 30 Goteborg, Sweden
`Tel.: +46—31—773-2579
`Fax: +46—31-773—2599
`e—mail: Christer.Larsson@gmmguse
`
`source is a necessity for maintaining the intracellular
`redox balance and a sustained conversion of sugar into
`ethanol. This is due to the fact that biosynthesis, and
`especially amino acid Synthesis, results in a net forma-
`tion of NADH (Albers et a]. 1996). In addition, for-
`mation of organic acids, e.g. acetic, pyruvic and succinic
`acid, also results in a surplus of NADH. The main
`pathway used by S. cerevt‘siae for regeneration of
`NAD+ is glycerol production, since ethanol formation
`from glucose is a rcdox-neutrai process. However, for—
`mation of glycerol
`represents an unwanted loss of
`carbon if the aim is to produce maximum amounts of
`ethanol. Therefore, attempts have been made to decrease
`the glycerol yield, e.g. by using dilTerent nitrogen sources
`in order to reduce the surplus of NADH due to amino
`acid synthesis (Albers et a1. 1996), or to use microaerobic
`conditions (Franzen et a1. 1994). An alternative strategy,
`used in this study, would be to minimize glycerol form
`mation by employing mutants in the glycerol-producing
`pathway.
`formed from dihy-
`is
`cerevisiae glycerol
`In S.
`droxyacetone phosphate by the consecutive action of
`glycerol-3-phosphate dehydrogenase (GPD),
`to yield
`g1ycerol-3—phosphate, and glycerol-3-phosphatase (GPP)
`to yield glycerol. There are two isogenes encoding dif—
`ferent forms of GPD, GPDJ first described by (Larsson
`et a]. 1993) and GPD2 described by (Eriksson et a1.
`1995). Even though the two isoenzymes show a strong
`homology, it seems as if they play distinctively different
`roles in the cellular machinery. GPD] is induced under
`osmotic stress (Albertyn el al. 1994; Andre et a1. 1991;
`Ansell et at. 1997) whereas GPD2 is induced under an-
`aerobic conditions and is suggested to be important for
`redox regulation under these conditions (Ansell et al.
`1997). The rationale for osmotic induction is that glyc—
`erol is the main osmoregulator in S. cerevisiae under
`hyperosmotic conditions (Blomberg and Adler I992).
`The g1ycerol-3—phosphatase also exists in two isoforms
`encoded by GPPI and GPP2, with the latter being os—
`moticaily induced (Norbeck et al. 1996). In this study we
`chose to reduce glycerOI production by using GPD
`
`BNSDOClD: <XP
`
`7
`
`..
`
`22360299 ___|_>
`
`BUTAMAX 1027
`
`BUTAMAX 1027
`
`

`

`mutants in order to avoid a possible detrimental accu-
`mulation of glycerol-3~phosphate. It
`is, hOWever, as
`stated above, not possible to eliminate all glycerol pro-
`duction completely, because of redox constraints. This
`has been verified by the inability of a double gpdlgpdE
`mutant to grow under anaerobic conditions while aer-
`obically growth was feasible,
`though at a reduced
`growth rate (Bjorkqvist et a1. 1997). Furthermore, this
`mutant showed an immediate reduction in fermentation
`rate when the conditions changed from aerobic to an—
`aerobic. However, addition of acetoin, which couid
`serve as an acceptor of reducing equivalents by its re—
`duction to butanediol, restored or even increased the
`aerobic fermentation rate (Bjc'irkqvist et al. 1997).
`The amount of excess NADH, and hence the re-
`quirement for glycerol formation for amino acid synthesis
`and formation of organic acids, can vary depending on
`the preference for NAD+ or NADP+ as co-factor
`(Albers et al. 1996). In several metabolic steps NAB+ as
`well as NADP+ can be utilized and there are additional
`biosynthetic pathWays for several of the amino acids.
`However,
`it seems as if the metabolic machinery is
`adjusted to produce a minimum amount ofexcess NADH
`and the amount of glycerol formed is only just enough to
`balance this surplus (Albers et a1. 1996). If this supposi-
`tion is true, it would be difiicult or even impossible to
`reduce glycerol formation and hence improve the ethanol
`yield by means of genetic manipulation of the GPD genes,
`unless the cells can be forced to produce lower amounts of
`organic acids but also, more importantly, to manage with
`a lower protein content. On the other hand, step-change
`experiments
`from aerobic to anaerobic conditions
`shOWed that a gpa'2 mutant did produce less glycerol than
`did the wild type without concomitant reduction in
`protein content (Bjorkqvist et al. 1997).
`The purpose of this investigation was to characterize
`the anaerobic growth performance and fermentation
`properties of gdeA and gdeA strains of S. cerevisiae
`under well-controlled conditions in a bioreactor,
`the
`eventual goal being a reduction of glycerol formation
`and an improved ethanol production.
`
`Materials and methods
`
`Yeast strains and media
`
`The S. cerevisiae strains used were all derived from W303-lA (add-
`I°, hisS-l’], leu2-3, ]]2rrp1—1a, ura3-I, can100°) referred to as wild
`type (ATCC 200060); gpdiA (nrz'cJ—f".
`[mi-H, [cull
`urrfi—j‘.
`canlClO", gpdlA::TRPl); gpd2A (adeZ-I“, iiisS-H, 19142-3, 1121113]-
`Ia, realm)", gdeA::URA3). The medium was a defined CBS me-
`dium (Albers et a1. 1996) with 7.5 g/l (NH4);SO4 as nitrogen source
`and 20 g/l glucose as carbon and energy source. The concentration
`of the required bases and amino acids was 120 nag/l, except for
`leuciue where 240 mg/l was used.
`
`Growth conditions
`
`Cultivations were carried out under anaerobic conditions in a fer—
`mentor {Belach Bioteknik AB, Stockholm, Sweden) with a working
`
`435
`
`volume of 2.5 l. The temperature was 30 “C, stirring rate 400 rpm,
`and the pH was kept constant at 5.0 by automatic addition of l M
`NaOH. To ensure anaerobic conditions, the fer-mentor was con-
`tinuously flushed with nitrogen gas at a rate of 37.5 l/h.
`
`Gas analysis
`
`Carbon dioxide evolution was continuously analysed by a carbon
`dioxide and oxygen monitor (type 1308, Bros] and Kjaer, Naerum,
`Denmark).
`
`Microcalorimetry
`
`The heat production rate (dQ/dr) was measured by a flow-through
`microcalorimeter
`(thermal activity monitor,- Thermometric AB
`Jarfalla, Sweden}. The effective volume of the measuring cell was
`0.52 ml.
`
`Growth determinations
`
`Growth was followod by measuring the absorbanoe of the cultures
`at 610 nm in a Beckman B spectrophotometer.
`
`Measurement of glucose, ethanol, glycerol and acetate
`
`TWo samples (1.5 ml each) were centrifuged for 5 min at 15 000 g
`and the resulting supernatants were frozen (—20 °C) until analysis.
`The concentrations were determined by using enzyme combination
`kits from Boehringer Mannheim (Biochemica test combination;
`Boehringer Mannheim, GmbH, Germany).
`
`Determination of dry weight
`
`Two samples (5 ml each) were centrifuged at 5000 g for 10 min and
`washed twice with water, and subsequently the pellets were kept at
`110 '’C for 24 h before temperature equilibration and weighing.
`
`Protein determination
`
`Two samples (10 ml each) were centrifuged at 5000 g for 5 min and
`washed twice with 0.9% (w/v) NaCl. The pellet was resuspended in
`3 ml NaOH and total protein was determined by a modified biuret
`method (Verduyn er a}. 1991) using bovine serum albumin as a
`standard.
`
`Northern blot analysis
`
`Northern blot analysis of GPDI expression was performed ac-
`cording to (Ansell et a1. 1997).
`
`Results
`
`Growth characteristics
`
`The wild type and gde showed an almost indistin-
`guishable growth pattern with close to identical heat
`production and growth rates (Fig.1, Table l). The
`gdeA strain, on the other hand, had a prolonged
`growth period, resulting in a reduced rate of heat pro—
`duction. This was due to an extended lag phase, but also
`a reduced growth rate was obtained with gdeA com-
`
`BNSDOCID: (XP
`
`_ aaeeoaonfi E >
`
`

`

`436
`
`300
`
`250
`
`200
`
`1 50
`
`dQ/dt(1tw) 100
`
`50
`
`0.8
`0.5
`0.4
`0.2
`
`
`
`Wild type
`gpd‘l A
`
`
`
`
`QPUEA ~—
`
`L09A61Gnm
`
`
`
`Glucose(mM)
`
`
`
`Ethanol(mM)
`
`
`
`Glycerol(mM)
`
`—0.2
`-0.4
`41.6
`—0.8
`-1.0
`120
`
`100
`
`30
`
`60
`
`40
`
`20
`
`
`
`180
`160
`140
`120
`100
`80
`60
`4D
`20
`0
`
`4.14.145)omemmomammo
`
`2.5
`
`Table 1 A comparison of growth and ethanol production rates of
`Saccharomyces cerevisiae wild type, gdeA and gpd2A respectively,
`during anaerobic batch growth on 2% glucose
`
`Strain
`
`Growth rate Ethanol
`
`Ethanol
`
`(11")
`production
`production
`
`(mi-no] g"1 h1)
`Ifmmol l" h)
`22.6 :l: 3.5
`11.0i 0 9
`11. 51:1:01
`17.6 a 1.5
`
`5.2 :: 0.2
`7.4 1: 0.8
`
`0.31 i 0.04
`0.32 a 0.07
`017 i 0.02
`
`Wild type
`gdeA
`gpa'ZA
`
`pared to that of the wild type and gdeA. However, the
`biomass concentration at the end of the growth period
`was very similar, Le. the growth yield was comparable
`for all strains (Table 2).
`
`Substrate consumption and product formation
`
`Depletion of glucose and cessation of growth were nicely
`traced by the rapid decline of the heat production rate
`(Fig. 1). Both the glucose consumption and ethanol
`production were very similar in the wild type and the
`gdeA mutant, whereas gdeA showed a delayed re
`sponse. Not only was the response delayed but the eth-
`anol production rate was also reduced in the gpd2
`mutant compared to that
`in the other two strains
`(Table 2). As a consequence, the overall ethanol pro-
`duction rate obtained from the fermentor was lower
`
`(Table 2). However, the final ethanol concentration was
`higher in the gdeA mutant and the ethanol yield was
`elevated by 8% (relative to the amount of substrate
`consumed) or 13% (relative to the biomass formed)
`compared to the wild type (Table 2). There was a
`slightly lower glycerol production in gdeA than in the
`wild~type, whereas gpd2A Showed a drastic reduction in
`the level of this metabolite (Fig. 1, Table 2). Deletion of
`the GPDZ gene resulted in a decline in glycerol pro-
`duction, relative to the amount of biomass formed, by
`almost 40%. Interestingly,
`the acetate concentrations
`were much lower in the mutants than in the Wild type. In
`the gde mutant, acetate production was almost totally
`abolished but gdeA also showed a drastic reduction in
`acetate concentration and yield (Fig. 1, Table 2).
`It
`might be that a decrease in acetate formation is an ex-
`ample of a metabolic adjustment by the cells to minimize
`the NADH surplus when the glycerol production
`capacity is hampered.
`
`ATP yields
`
`The ATP yields (YATP, g biomass/mot ATP) of 16—17 g]
`mol (Table 2) for the wild type and gptlIA were close to
`the value of £6 g/moi reported for optimal growth of
`S. cerevisiae under anaerobic conditions (Verduyn et a].
`
`ig. 1 Changes in measured parameters during anaerobic batch
`cultures of Sacchnromyces cerevisiae wild type (I), gdeA (O) and
`gpd2A (A) respectively, with 2% glucose as carbon and energy source
`
`d F
`
`2.0
`
`1.0
`
`
`
`Acetate(mM) 3
`
`0.5
`
`00
`
`24
`
`28
`
`2Time (h)20
`
`_BNSDOC[D: <XF’_
`
`, 2236029A,,.| . >
`
`

`

`Table 2 A comparison of dif-
`ferent yields, heat production
`and energy balance of S. cere—
`visiac wild type, gpdjd and
`gdeA respectively, during
`anaerobic batch growth on 2%
`glucose
`
`Parameter
`
`Grewth yield (g biomass/g glucose)
`Ethanol yield (moi EtOl-Ifmol glucose)
`Ethanol yield (mmol EtOl—l/g biomass)
`Glycerol yield (mol glycerol/mo] glucose)
`Glycerol yield (mmol glycerol/g biomass)
`Acetate yield (mol acetate/mo] glucose)
`Acetate yield (mmol acetate/g biomass)
`Heat yield (kag biomass)
`ATP yield (g biomass/mol ATP)
`Carbon balance (%)
`
`Wild type
`
`0.110 .+ 0.000
`1.32 i 0.00
`66.74 :t 0.16
`0.175 :1: 0.002
`8.85 :1: 0.04
`0.019 d: 0.004
`0.98 d: 0.22
`4.81 i 0.61
`17.0 :t 0.1
`
`95.2 :: 0.9
`
`gdeA
`
`0.108
`1.33
`68.57
`0.145
`7.43
`0.005
`0.25'
`4.30
`16.3
`93.7
`
`437
`
`gpd2A
`
`0.110 i 0.013
`1.43 i 0.02
`75.15 i 7.97
`0.106 :1: 0.000
`5.58 i 0.68
`0.002 d: 0.000
`0.11 :b 0.01
`5.16 :t 0.78
`14.0 :t 2.0
`95.7 d: 1.]
`
`1990). It should be remembered, however, that the re—
`quired amino acids and bases that are added to the
`medium represent an additional carbon source and
`hence influence the ATP yields obtained. Owing to the
`higher ethanol and lower glycerol yields, the YATP was
`slightly decreased for the gpa’2 mutant. The small in-
`crease in heat yield obtained for gdeA could also be a
`reflection of the elevated ethanol production in this
`strain. On the other hand, the excess NADH and hence
`the amount of glycerol formed have a drastic influence
`on the heat yield, Le.
`the more glycerol the lower the
`heat yield (Larsson and Gustafsson 1998). This is due to
`the fact that glycerol production from glucose is a re-
`ductive energy-requiring process.
`
`The functiou of the two NAD+—dependent isoforms
`of GPD, GPDI and GPD2
`
`It is clear that GPD2 is the most important enzyme
`during anaerobic growth (Ansell et a1. 1997). This was
`manifested also in this study, since deletion of GPD]
`resulted in a phenotype more or less indistinguishable
`from that of the wild type, the only detectable difference
`being the strong reduction in acetate formation in the
`mutant (Fig. 1). The gpd2A strain, on the other hand,
`was severely affected during anaerobic growth. How—
`ever, glycerol formation also persisted in the gdeA
`strain, even though the production rate was much lower
`than the wild-type level (Fig. 1). This suggests that,
`under anaerobic conditions, GPDZ can, at least in part,
`be substituted by GPD]. A supposition corroborated by
`Northern analysis, showing a higher level of expression
`GPDI in gdeA than in the wild-type strain (Fig. 2).
`
`Protein content and NADH formation
`
`There was no reduction in the protein content of the
`gpd2 mutant. Instead a slightly higher value of 59.9%
`(w/w) protein was obtained for the mutant whereas the
`wild type contained 56.7% protein. The acetate pro-
`duction, and hence the accompanying NADH forma-
`tion, due to this acid was much lower in the mutant.
`According to Albers et a1. (1996) each gram of protein
`synthesized is accompanied by the production of 17.9—
`
`BNSDOClD‘. <XP
`
`2236029Ail ,>
`
`21.2 mmol NADH. The range is due to the existence of
`multiple biosynthetic pathways for some of the amino
`acids as well as uncertainties concerning the use of
`NAD+ or NADP+ as coenzyme. One must also con~
`sider that the strain used in the present study is auxo—
`trophic for leucine, histidine and tryptophan. Hence,
`there is no NADH formation through synthesis of these
`amino acids; they are simply taken up from the medium.
`When a correction has been made for this, by using the
`amino acid composition given by Albers et a1. (1996), a
`range of 14.7—33.0 mmol NADH/g protein could be cal—
`culated for the strain used in the present study. By
`taking this into account,
`together with the amount of
`acetate formed, a surplus of 23.9—11.0 mmolfg NADH
`could be calculated for the gdeA mutant. The observed
`glycerol level can only compensate for a fraction of this
`excess NADH (Table 3). There could also be additional
`NADI—I formed as a result of RNA synthesis and for—
`mation of other metabolites, but
`this contribution is
`usually very Small (Albers et a1. 1996).
`
`Discussion
`
`This study showed that it is indeed possible to obtain a
`drastic reduction in glycerol formation and a concomi—
`tant increase in the production of ethanol by using a
`
`140
`
`4.1.
`
`
`
`GPD1expression1%)
`
`#036301000000
`
`I“:
`
`CC!
`
`Early log Midlog Stationary Early log Midlog Stationary
`phase
`phase
`phase
`phase
`phase
`phase
`
`Fig. 2 Northern blot analysis of glycerol-3—phosphate dehydrogenase
`(GPDI) expression in the wild—type and gpd2A mutant respectively.
`The expression level at early log phase in the wild type is set to 100%.
`Samples were taken at early log, mid 10g and early stationary phase
`
`

`

`438
`
`Tabie 3 Calculation of NADH formation of S. cerevisiae wild type
`phan (see Results). The range of values for production during
`acetate formation is due to the uncertainty whether NAD+ or
`and gpd2A respectively, during anaerobic batch growth on 2%
`NADP+ is used in the aldehyde dehydrogenase step. The final
`glumse. The protein content was 56.7% and 59.9% for the wild
`NADH yield is the balance resulting from NADH formation due to
`type and the mutant respectively. The amount of NADH accom-
`protein and acetate formation and NADH consumption due to
`panying protein synthesis was calculated by using the range 17.9—
`glycerol production
`2l.2 mmol NADH/‘g protein as suggested by Albers et a1. (1996)
`and compensating for the uptake of leucine, histidine and trypto-
`
`
`NADH
`Strain
`Glycerol
`Acetate
`i’rotein
`
`(mmolfg)
`(mmol NADH/g)
`(mmol NADH/g)
`(mmol NADH/g)
`
`Wild type
`gpd2A
`
`—8.85
`—5.58
`
`+0.98~l.96
`+ 0.11—0.22
`
`+ 8.3—102.
`+ 83—103
`
`+0.4—3.3
`4- 3.3—5.4
`
`gpd2A strain (Table l). The decrease in glycerol pro-
`duction was not accompanied by any decline in the
`protein content of the cells. Most of the NADH surplus,
`and hence the need for redox adjustment by glycerol
`formation, originates from amino acid synthesis (Albers
`et a1. 1996). Depending on the co~factor
`involved,
`NAD+ or NADP+, and the existence of multiple bio-
`synthesis pathways for at least some of the amino acids,
`it is possible to minimize the amount of excess NADH
`formed. However, when comparing the amount of
`glycerol formed, it seemed as if the wild—type levels in
`this study (Table 3) as well as in previous observations
`(Albers et a1. 1996) were just enough to keep up‘with the
`redox requirements when NADH formation was mini-
`mized. Nevertheless, this study showed that, by using a
`gde mutant, glycerol formation could be reduced even
`further. How the redox balance in the gpd2A mutant is
`accomplished is not known; part of the explanation is a
`decrease in the amount of acetate formed. There must,
`however, be additional mechanisms as well. One possi-
`bility might be that NADP+ can substitute for NAD+
`as a co—factor to a larger extent than expected and/or
`that additional biosynthetic pathways may exist. A
`theoretical possibility is also that S. cerevisiae has the
`potential of using other reductive pathways apart from
`glycerol formation during anaerobic growth on glucose.
`To summarize, it ought to be impossible for the cells to
`reduce glycerol formation under anaerobic conditions
`since not more than what is absolutely required seems to
`be produced. However, the gdeA mutant in some way
`managed to maintain redox balance and sustained
`metabolic activity in spite of the drastic reduction in
`glycerol
`level, the result being an increase in ethanol
`production and reduction in, not only glycerol, but also
`acetate formation.
`The two different
`
`isoforms of GPD seem to have
`
`specific functions but it also appears that, at least in
`some circumstances, they can replace each other (Ansell
`et al. 1997). GPD2 is expressed under anaerobic condi—
`tions and its regulation is somehow related to the redox
`status of the cells. However, redox regulation is not
`limited to GPDZ. Aerobically GPD] was found to be the
`important enzyme when the cells were challenged with a
`high redox adjustment demand due to growth on a re-
`duced substrate, i.e. ethanol (Larsson et a1. 1998). Under
`such aerobic conditions GPD2 did not show any activ-
`
`ity. This study, on the other hand, confirmed the leading
`role of GPD2 under anaerobic conditions but also
`
`showed the partial ability of GPD! as a replacement in
`the gpd2A mutant. A recent publication by Michnick
`et al. (1997) showed a slightly difierent result with respect
`to GPDl under anaerobic conditions.
`In our study,
`GPD] deletion did not
`influence glycerol production
`while, in their study (Michnik et al. 1997), the absence of
`the GPD] gene product gave an almost 50% reduction in
`glycerol formation compared to the wild type. However,
`it might be that the comparably high glucose concen~
`tration used (10%) triggered osmotic induction of GPD]
`since GPD] is the isoform of GPD induced under 05-
`motic stress conditions (Ansell et a1. 1997).
`A striking observation was also the dramatic reduc-
`tion in acetate production observed in the gdeA mutant
`(Fig. 1). Especially since this was the only parameter
`measured that was significantly different from that of
`the wild type. Even though GPD] is not important for
`glycerol production under anaerobic conditions,
`the
`cells still seem to recognize the reduction in NADH—
`Oxidizing capacity when this protein is lacking. The re-
`sponse being a reduced acetate and hence NADI-I for—
`mation rate.
`
`Acknowledgements Financial support from the Swedish National
`board for Technical Development is gratefully acknowledged.
`
`References
`
`Aibers E, Larsson C, Lidén G, Niklasson C, Gustafsson L (1996)
`Influence of the nitrogen source on Saccliaromyces cerevisiae
`anaerobic growth and product formation. Appl Environ Mi—
`crobiol 62: 3187—3195
`Albertyn J, Holtmann S, Thevelein JM, Prior JA (1994) GPD]
`which encodes glycerol-S-phosphate dehydrogenase, is essential
`for growth under osmotic stress in Saccharomycas cereviriae
`and its expression is regulated by the high-osmolan'ty glycerol
`response pathway. Mol Cell Biol 14: 41354144
`Andre L, Hemming A, Adler L (1991) Osmoregulation in Sac-
`charomyccs cerevisiae. Studies on the osmotic induction of
`glycerol production and giycerol 3-phosphate dehydrogenase
`(NADt). FEBS Lett 286:
`i3—17
`Ansell R, Granath K, Hohmann S, Thevelein JM, Adler L (1997)
`The two isoenzyrnes for yeast NAD+-dependent glycerol 3-
`phosphate dehydrogenase encoded by GPD] and GPD2 have
`distinct roles in osmoadaptation and redox regulation. EMBO J
`16: 2179—2187
`
`BNSDOClD: <XP_..
`
`_ 722360233X I __>
`
`

`

`‘
`
`Bjorkqvist S, Ansell R, Adler L, Lidén G (1997) Physiological re—
`sponse to anaerobicity of glycerol-S-phosphaie dehydrogenase
`mutants of Saccharcmyces cerevit'iae. Appl Environ Microbiol
`63'.
`l28—132
`Blornberg A, Adler L (1992) Physiology of osmotolerance in fungi.
`Adv Microb Physiol 33: 146—212
`Eriksson P, André L, Ansell R, Blomberg A, Adler L (1995)
`Cloning and characterization of GPD2, a second gene encoding
`sn—glycerol 3-phosphate dehydrogenase (NAD+) in Scrcdmro-
`myces cerevisiae and its comparison with GPD]. Mol Microbiol
`17: 95—107
`Franzen C, Lidén G, Niklasson C (1994) An experimental guide to
`the relevant aeration rates in microaerobic bioprocesses, In:
`Galindo E, Ramirez 0T (eds) Advances in Bioproccss Engi-
`neering. Kluwer, London New York, pp 255—265
`Larsson C, Gustafsson L (£998) Calorimetry of microbial pro-
`cesses. Thermoehim Acta (in presg)
`Larsson C, Péhlman I‘L. Ansell R, Rigoulet M, Adler L, Gust-
`afsson L (1998) The importance of the glycerol 3~phosphate
`shuttle during aerobic growth of Saccharamyces cerevz'siae.
`Yeast (in press)
`
`439
`
`Larsson K, Ansell R, Eriksson P, Adler L (I993) A gene encoding
`Sir-glycerol 3-phosphate dehydrogenase (NAD'W complements
`an osmosensitive mutant of Saccharomyces ccrevisz'acl M01
`Microbial l0: llOl—llll
`Michnik S. Roustan J-L, Remize F, Barre P, Dequin S (I997)
`Modulation of glycerol and ethanol yields during alcoholic
`fermentation in Saccharomyces cerevisfae strains overexpressed
`or disrupted for GPDI encoding glycerol 3—phosphate dehy—
`drogenase, Yeast 13: 783—793
`Norbeck J, Pfihlman A-K, Akhtar N, Blomberg A, Adler L (1996)
`Purification and characterization of two isoenzyrnes of D1.—
`glycerol-3—phosphatase from Saccharomyces cerevz'sme. J Biol
`Chem 271; 13875713881
`.
`Verduyn C, Postma E, Schefi‘ers WA, Dijken JP van (1990) Ener-
`getics of Sacclmromyces cerevisiae in anaerobic glucose-limited
`chemostat cuimres. J Gen Microbiol 136: 405—412
`Verduyn C, Stouthamer AH, Scheflers WA, Dijken JP van (1991)
`A theoretical evaluation of growth yields of yeasts‘ Antonie van
`Leeuwenhoek 59: 49—63
`
`BNSDOCID.‘ <XP
`
`2236029A7L>
`
`