Growth of the yeast Saccharomyces cerevisiae on a non-fermentablesubstrate: control of energetic yield by the amount of mitochondria

Laurent Dejean, Bertrand Beauvoit, Bernard Guerin, Michel Rigoulet *Institut de Biochimie et Genetique Cellulaires du CNRS, Universite Victor Segalen, Bordeaux II, 1 rue Camille Saint Sae«ns,

33077 Bordeaux cedex, France

Received 12 July 1999; received in revised form 29 November 1999; accepted 10 January 2000

Abstract

The purpose of this study was to investigate the long-term control of ATP synthesis during the course of Saccharomycescerevisiae batch grown on lactate, a purely respiratory substrate. For this, we used a respirometric and on-line calorimetricapproach to analyse the energetic balances and the control of energetic metabolism during growth. Enthalpic growth yieldsassessed by enthalpy balance (taking account of substrate consumption, by-product accumulation, biomass formation andheat dissipation) remained constant during the entire exponential growth. Moreover, at the same time, a parallel decrease inbasal respiratory rate and enthalpy flux occurred. It is shown that the decrease in respiration corresponds to a decrease in theamount of mitochondria per cell but not to a change of steady state of oxidative phosphorylation. Taking into account thepart of energy used for maintenance, it can be concluded that mitochondria by themselves are the major heat dissipativesystem in a fully aerobic metabolism, and that the decrease in the amount of mitochondria when growth rate decreases leadsto an enthalpic growth yield constant. ß 2000 Elsevier Science B.V. All rights reserved.

Keywords: Yeast; Microcalorimetry; Enthalpic growth yield; Mitochondria; Oxidative phosphorylation

1. Introduction

From studies on isolated mitochondria, it has been

proposed that oxidative phosphorylation operatesclose to equilibrium [1], apart from cytochrome oxi-dase which does not [2]. Consequently, mitochon-drial respiration and ATP synthesis are controlledby the phosphate potential, the redox potential stateof NAD� and the e¡ectors of cytochrome oxidase.Application of the metabolic control theory, how-ever, has shown that the control over mitochondrialrespiration and ATP synthesis is shared by a numberof reactions, and that the distribution of control de-pends on the origin of mitochondria and varies ac-cording to the metabolic state [3]. In this respect, thecrucial role of the Pi carrier [4] and some dehydro-genases [5] in yeast mitochondria has been demon-strated.

In vivo studies of mammalian resting cells and

0005-2728 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved.PII: S 0 0 0 5 - 2 7 2 8 ( 0 0 ) 0 0 0 5 3 - 0

Abbreviations: P, heat production rate, expressed as WW/ml;JO, rate of oxygen consumption, expressed as nanomoles of atomO consumed per min and per mg dry weight (nat.O/min/mg dryweight) ; DCIP, 2,6-dichloroindophenol; DTNB, 5,5P-dithiobis-2-nitrobenzoic acid; W, growth rate constant, expressed as h31 ;YX=S, mass yield, expressed as g dry weight/g lactate consumed;vQX, heat yield, expressed as kJ/g dry weight; E.R., energy re-covery; JH, speci¢c enthalpy £ux, expressed as WW/mg dryweight; ClCCP, carbonyl cyanide m-chlorophenyl hydrazone;TET, triethyltin chloride; v~WH�, proton electrochemical potentialdi¡erence across the mitochondrial inner membrane

* Corresponding author. Fax +33 556999010;E-mail : michel.rigoulet@ibgc.u-bordeaux2.fr

BBABIO 44812 15-2-00

Biochimica et Biophysica Acta 1457 (2000) 45^56www.elsevier.com/locate/bba

organs (see [6^8] for reviews) and of resting cells ofthe yeast Saccharomyces cerevisiae [9] have shownthat the overall rate of oxidative phosphorylation isshort-term controlled (i) downstream from the ATPsynthase by the cytosolic ATP turnover, (ii) up-stream from the respiratory chain by reducing equiv-alent availability and allosteric activation of dehy-drogenases, and (iii) by ionic permeability of theinner mitochondrial membrane. More recently, along-term regulation phenomenon has been de-scribed on growing tumour cells. Hence, a decreasein oxygen uptake associated with a decrease in mito-chondria content per cell occurs during the exponen-tial growth phase of various cultured cell lines[10,11]. Moreover, a similar behaviour has beenlinked to a constant cellular ATP/ADP ratio ongrowing glioma cells [12]. Therefore, during growthof mammalian cells, a modulation of mitochondrialcontent seems to be involved in the long-term regu-lation of cellular oxidative phosphorylation. Never-theless, the relationships between mitochondrial con-tent and energetic metabolism regulation of growingcells are not extensively documented.

Therefore, we used a respirometric and on-line cal-orimetric approach to analyse the energetic balancesand the control of energetic metabolism duringgrowth phase transitions of batch grown S. cerevisiaewith limiting amounts of a non-fermentable carbonsource (lactate). Enthalpic growth yields were as-sessed using continuous microcalorimetric monitor-ing associated with the determination of metaboliteconcentrations in the culture medium to establishenthalpy balances in the growth phases ([13]; see[14,15] for reviews). Hence, when the growth-depen-dent ATP demand decreases, a modulation of themitochondrial enzyme content mainly causes a de-crease in respiratory activity, which in turn is respon-sible for a constant oxidative phosphorylation regimeand enthalpic growth yield.

2. Materials and methods

2.1. Yeast strain, culture medium and growthcondition

The yeast strain used for experiments was the hap-loid strain S. cerevisiae W303-1a (ade2-10, his3-11,

112trp1-1a, ura3-1, can1000). Cells were grownaerobically at 28³C on the following medium:0.17% yeast nitrogen base w/o amino acids w/o am-monium sulphate (Difco), 0.1% potassium phos-phate, 0.5% ammonium sulphate at pH 5.5, contain-ing a growth-limiting amount of D,L-lactate (0.2% w/v) (Aldrich). Cultures (200 ml) were performed underconstant bubbling of sterile humidi¢ed air in 1 l£asks connected to the microcalorimeter. The con-centration of auxotrophic requirements was 100mg/l.

2.2. Growth determinations

Growth of W303-1a was measured turbidometri-cally at 600 nm in a Phillips spectrophotometer. Dryweight determinations were performed on samples ofcells harvested throughout the culture period andwashed twice in distilled water. A coe¤cient of tur-bidity (0.17 þ 0.01 mg dry weight/optical density unit(n = 6)) was obtained regardless of the growth phase.

2.3. Calorimetry setup and heat measurement assay

The heat production rate (i.e. P = dQ/dt) was con-tinuously monitored with a multichannel microca-lorimeter (Thermal Activity Monitor, TAM, Ther-mometric, Jarfalla, Sweden) in the £ow-throughmode. For all experiments, growing cell suspensions(200 ml in a 1 l £ask) were incubated in a water bathat 28³C, continuously bubbled with water-saturatedair and stirred by a magnetic stirrer at 300^400 rpm.Using a peristaltic pump (Gilson), cells were trans-ported via te£on tubings to the calorimeter (thermo-statted at 28³C) and back to the £ask. The £ow rateof the pump was 2^3.5 ml/min, resulting in a totaltransport time of 0.5^1 min to avoid hypoxia.Throughout the culture time, 5 ml cell sampleswere collected to measure optical density, metabo-lites in the culture medium, oxygen consumptionand mitochondrial enzyme activities (i.e. D,L-lactatedehydrogenase and citrate synthase activities).

Heat production rate was recorded at samplingintervals of 10 s (Digitam software), considering thee¡ective volume of the £ow cell to be 0.6 ml. Datacorrection was performed with the software (Digi-tam) using the Tian equation and time constants[16].

BBABIO 44812 15-2-00

L. Dejean et al. / Biochimica et Biophysica Acta 1457 (2000) 45^5646

2.4. Respiration assay

The oxygen consumption of cells was measuredpolarographically at 28³C using a Clark oxygen elec-trode in a 2 ml thermostatically controlled chamber(Oroboros Oxygraph, Paar, Graz, Austria). Datawere recorded at sampling intervals of 1 s (DatLabAcquisition software, Oroboros, Innsbruck, Austria).Respiratory rates (JO) were determined from theslope of a plot of O2 concentration versus time.For all assays, 2.5 ml of the growing cell suspensionwere quickly transferred from the £ask connected tothe microcalorimeter to the respirometer cuve.

2.5. Determination of D,L-lactate, pyruvate andacetate contents of culture media

Samples of growing cell suspension were ¢lteredfor metabolite content determination in the culturemedium. Pyruvate was measured enzymatically asdescribed previously [17]. D,L-Lactate and acetatewere measured using enzyme combination kits (Bio-chemica Test Combination; Boehringer Mannheim).

2.6. D,L-Lactate dehydrogenase and citrate synthaseassays

Samples of growing cells were centrifuged for1 min at 10 000Ug. The supernatants were removedand the pellets resuspended and incubated for 30 minat 32³C in a medium containing potassium phos-phate 100 mM (pH 7.4), EDTA 2 mM, zymolyase(1 mg/ml) (ICN Biomedicals), Triton X-100 0.3%(v/v) and an anti-protease mixture (1 tablet for10 ml solution) (complete EDTA-free; BoehringerMannheim).

D-Lactate and L-lactate dehydrogenase (EC 1.1.2.3and 1.1.2.4, respectively) activity was measured usinga modi¢ed method for determining succinate dehy-drogenase activity [18]. Brie£y, it was determined inthe presence of 20 mM of D,L-lactate by monitoringat 578 nm the phenazine methasulphate mediatedreduction of 2,6-dichloroindophenol (DCIP) as afunction of time, in an Eppendorf photometer. Allassays were performed in the presence of antimycinA (0.3 Wg/ml) and potassium cyanide (3 mM) to in-hibit respiratory chain complexes III and IV, respec-tively. The enzyme activity was calculated using an

extinction coe¤cient of 16 500 M31 cm31 at 578 nm.One D,L-lactate dehydrogenase unit was equal to1 Wmole of DCIP reduced per minute per mg dryweight.

Citrate synthase (EC 4.1.3.7) activity was deter-mined by monitoring at 412 nm the oxidation ofcoenzyme A (produced by citrate synthase activity)by 5,5P-dithiobis-2-nitrobenzoic acid (DTNB) as afunction of time, in an Eppendorf photometer [19].The enzyme activity was calculated using an extinc-tion coe¤cient of 13 600 M31 cm31 at 412 nm. Onecitrate synthase unit was equal to 1 Wmole of DTNBreduced per minute per mg dry weight.

2.7. Cytochrome content of growing cells

Cells were harvested in the appropriate phase,washed twice with distilled water and concentratedto obtain 2 ml of a cell suspension of about 50 opti-cal density units at 600 nm. They were placed in adual spectrophotometer (Aminco DW2000) and adi¡erential spectrum (from 500 to 650 nm) was per-formed between 1 ml of cells in the presence of 1 Wlof H2O2 70% (w/v) (oxidised state) and 1 ml of cellsin the presence of a few grains of dithionite (reducedstate). Calculations of cytochrome c+c1 and cyto-chrome b contents were performed using an extinc-tion coe¤cient of 18 000 M31 cm31 for the wave-length pairs 550^540 nm and 561^575 nm,respectively [20]. The calculation of cytochromea+a3 contents was performed using an extinction co-e¤cient of 12 000 M31 cm31 for the 603^630 nminterval [21].

3. Results

3.1. Lactate-limited growth of yeast cells in batchculture

The heat dissipation associated with the growth ofS. cerevisiae in a medium containing a growth-limit-ing amount of D,L-lactate was continuously moni-tored by direct microcalorimetry. Three distinctphases were de¢ned on the basis of the growth curve:(i) in the early exponential phase, biomass increasedwith a constant growth rate (W) of about 0.175 h31 ;(ii) in the late exponential phase, growth rate de-

BBABIO 44812 15-2-00

L. Dejean et al. / Biochimica et Biophysica Acta 1457 (2000) 45^56 47

creased in parallel to lactate concentration in theculture medium; and (iii) the biomass remained con-stant in the stationary phase, which occurred whenlactate had been totally consumed (Fig. 1). The heatproduction rate increased in parallel with biomassproduction in the early exponential phase but theheat dissipation pattern no longer correlated withthe biomass increase in the late exponential phase(Fig. 1). In the stationary phase, the heat productionrate decreased to ¢nally reach a constant value, inagreement with the growth arrest of yeast cells (Fig.1). Accumulation of by-products of the lactate catab-olism was measured in the culture medium duringexponential growth. For instance, at the end of theearly exponential phase, pyruvate and acetate con-centrations reached values of about 2 and 1 mM,respectively. However, in the late exponential phase,acetate was consumed whereas pyruvate was stillproduced (data not shown).

3.2. Enthalpy balance during exponential growth oflactate-limited batch growth

During exponential growth, mass yield (YX=S) andheat yield (vQX) were obtained from the slopes ofplots of the cumulative biomass production versuscumulative substrate consumption, and cumulative

heat production versus the cumulative biomass pro-duction, respectively (Fig. 2) [13]. These plots gave alinear correlation in both the early and the late ex-ponential phase, so these two phases could be con-sidered as constant physiological states (Fig. 2) [13].Interestingly, the mass yield (YX=S) calculated fromlactate consumption was only slightly di¡erent in theearly and late exponential phases (0.25 vs. 0.28 g dryweight/g lactate consumed). Moreover, the energy

Fig. 2. Mass yields and heat yields of W303-1a cells in expo-nential phase. W303-1a cells were grown aerobically in minimalmedium containing limiting amounts of D,L-lactate (0.2% w/v).(A) Mass yields (YX=S), expressed in g dry weight/g lactate con-sumed were determined by plotting the cumulative biomass pro-duction versus the lactate consumption during early (a) andlate exponential phases (b). (B) Heat yields (vQX), expressed inkJ/g dry weight, were determined by plotting the cumulativeheat production versus the cumulative biomass production dur-ing early (a) and late exponential phases (b). Values are fromtwo independent experiments.

Fig. 1. Lactate-limited growth of W303-1a in batch culture(typical experiment). W303-1a cells were grown aerobically inminimal medium containing limiting amounts of D,L-lactate(0.2% w/v). Heat production rate (P) (continuous line), D,L-lac-tate concentration in the culture medium (a) and biomass (b)were measured as described in Section 2. I, II and III representthe early, late exponential and stationary growth phase, respec-tively. Values are from a single experiment.

BBABIO 44812 15-2-00

L. Dejean et al. / Biochimica et Biophysica Acta 1457 (2000) 45^5648

lost as heat during growth (vQX) was approximatelythe same in the early and late exponential phases(25.2 vs. 24.5 kJ/g dry weight).

Energy balances in the early and late exponentialphases were calculated with integrated values of netsubstrate consumption (energy input) and net prod-uct formation (energy output) transformed into en-ergy units (kJ). These calculations were performedusing molar enthalpy variations associated to com-plete oxidation of substrate and by-products in CO2

and H2O (i.e. 31350 kJ/mole, 31110 kJ/mole and3854 kJ/mole for lactate, pyruvate and acetate, re-spectively [22]). These values were ¢nally normalisedto one unit of biomass formed (g dry weight) for theearly and late exponential phases, respectively[13,15]. Energy recoveries (E.R.) between input andoutput close to unity (0.98 and 1.02 in the early andlate exponential phase, respectively) were attained(Fig. 3), indicating that in the balances all majorcomponents of the system were taken into account.The total energy input needed to synthesise 1 g of

biomass was not signi¢cantly di¡erent throughoutthe exponential growth period (361 þ 5 vs. 353 þ 5kJ/g dry weight in the early and late exponentialphase, respectively) and had the same value whencorrected for by-product formation (about 340 kJ/g dry weight). In the early exponential phase, lactateutilisation accounted entirely for this value, whereasin the late growth phase, acetate utilisation ac-counted for about 8% of the total energy input(Fig. 3). The energy input lost as heat was approx-imately the same in the early and late exponentialphases (41% vs. 47%) (Fig. 3). The by-products oflactate catabolism (i.e. pyruvate+acetate) representedabout 27 and 16% of the energy output in the earlyand late exponential phase, respectively. The part ofthe energy output composed of biomass was esti-mated according to an average value of the enthalpyof the combustion of yeast biomass of 321.4 þ 0.4 kJg31 ash-free biomass, which has been shown to beindependent of the carbon source and the growthphase [13,14], and an ash content of 9% of the dryweight [23]. The part of the energy input conservedin biomass, i.e. the enthalpic growth yield, was nearlythe same in the early (32%) and late exponentialphases (37%) (Fig. 3).

Fig. 3. Enthalpy balance during exponential phases of batchgrown W303-1a cells. The energy contents of the substratesconsumed and the products formed (including biomass) perunit of biomass formed (kJ/g dry weight) are abbreviated as:X, biomass; Q, integrated heat yield; Lac, D,L-lactate; Pyr, py-ruvate; Ac, acetate. The combustion enthalpies of substratesand products were calculated from tabulated enthalpies of theformation of the di¡erent species involved in the reactions[13,15,22]. Determinations of D,L-lactate, pyruvate, acetate, heatproduction rate and biomass were performed as described inSection 2. Enthalpy recoveries between input and output (E.R.)close to unity were obtained. I and II represent the early andlate exponential growth phase, respectively. Values are meansof two independent experiments.

Fig. 4. Parallel decrease in heat production £ux and respiratoryrate of batch grown W303-1a as a function of remaining D,L-lactate in the culture medium. W303-1a cells were grownaerobically in minimal medium containing limiting amounts ofD,L-lactate (0.2% w/v). Speci¢c calorimetric (b) and respiromet-ric (E) £uxes were measured throughout the culture time. D,L-Lactate content of culture medium was assayed as described inSection 2. I and II represent the early and late exponentialgrowth phase, respectively. Values are from ¢ve independent ex-periments.

BBABIO 44812 15-2-00

L. Dejean et al. / Biochimica et Biophysica Acta 1457 (2000) 45^56 49

3.3. Heat production and oxygen consumption duringlactate limited batch growth

Using the growth equation (i.e. dX =WUXUdt)and the speci¢c enthalpic £ux (i.e. JH = dQ/dtU1/X), the instantaneous heat yield (i.e. dQ/dX) hasbeen shown to be equal to the speci¢c enthalpy£ux, JH, divided by the growth rate constant W[13,24]. Since the growth rate decreased during thetransition from the early to the late exponentialphase whereas the heat yield remained constant, theenthalpy £ux must decrease during this transition.Fig. 4 represents the speci¢c heat production £uxand the basal oxygen consumption rate as a functionof the remaining D,L-lactate concentration in the cul-ture medium of batch grown cells. During the earlyand late exponential phases, JH and JO simultane-ously decreased as the lactate was consumed in theculture medium. Consequently, the calorimetric-re-spirometric ratio (CR ratio), de¢ned as the ratio be-tween JH and JO [25], remained constant duringthese two phases (3310 þ 50 kJ/mol O2). The largestfall in both £uxes was observed during the transitionfrom the early to the late exponential phase (Fig. 4).The question is therefore raised as to the origin of

the decrease in the respiratory activity of yeast cellsin lactate-limited batch culture.

3.4. Respiratory activity modulation of yeast cells inbatch culture

3.4.1. Control of respiration by extracellular pH andlactate concentration throughout the cultureperiod

Because of the weak bu¡ering capacity (e.g. lowconcentration of inorganic phosphate), the pH of theculture medium increased as the lactate was con-sumed throughout the batch culture period (Fig. 5).The parallel decrease in the basal respiratory ratecould be due to a decrease in the ATP consumingprocess linked to the maintenance of intracellular pH(e.g. plasma membrane H�-ATPase). Then, we com-pared the basal oxygen uptake of growing yeast cellsmeasured in the culture medium or in the same me-dium at pH 7. This increase in extracellular pH onlyslightly decreased the basal respiratory rate (300 þ 15vs. 335 þ 20 nat.O/min/mg dry weight for cells inearly and 180 þ 10 vs. 230 þ 20 nat.O/min/mg dryweight for cells in late exponential phase (n = 2)).Hence, we can conclude that the maintenance ofthe intracellular pH throughout the culture period

Fig. 6. Typical determination of uncoupled respiratory rates(JOClCCP) of growing W303-1a cells. Respiratory rates of grow-ing W303-1a cells were measured in the growth medium in thepresence of various ClCCP concentrations, as described in Sec-tion 2. The respiratory rate stimulation is the ratio (JOClCCP/JObasal)U100, measured for cells in early (b), late exponential (E)and stationary phases (a). Control values (JO basal) were equalto 294, 200 and 46 nat.O/min/mg dry weight in the early, lateexponential and stationary phase, respectively. Values are froma single experiment for each titration.

Fig. 5. Change in pH of culture medium of W303-1a cells as afunction of remaining D,L-lactate. W303-1a cells were grownaerobically in minimal medium containing limiting amounts ofD,L-lactate (0.2% w/v). Throughout the culture time, cell suspen-sion was ¢ltered for pH determination of the culture medium.D,L-Lactate content of culture medium was measured as de-scribed in Section 2. I and II represent the early and late expo-nential growth phase, respectively. Values are from ¢ve inde-pendent experiments.

BBABIO 44812 15-2-00

L. Dejean et al. / Biochimica et Biophysica Acta 1457 (2000) 45^5650

accounted for only 10^20% of the basal oxygen up-take in the exponential phase.

Since the lactate transport system of the yeast plas-ma membrane has been characterised as a proton-lactate symport [26], the decrease in the basal oxygenuptake could also be explained by limitations in thelactate delivery to the respiratory chain due to a risein the external pH (i.e. a decrease in the vpH acrossthe plasma membrane) and to substrate depletion inthe culture medium. Hence, to investigate possiblelimitations in the electron delivery to the respiratorychain in the di¡erent growth phases, the basal andmaximal respiratory activities that can be sustainedby yeast cells were measured in the culture medium,on one hand, and in fresh growth medium (pH 5.5)containing 0.2% D,L-lactate on the other. Maximalrespiratory rate was measured by means of the pro-tonophoric uncoupler, ClCCP, to dissipate the pro-ton electrochemical gradient across the inner mito-chondrial membrane, and thereby to stimulate therespiratory rate to its maximum value (JOClCCP).However, since ClCCP is a weak acid lipophilic mol-ecule, its protonophoric e¡ect may depend on the pHvalue of the incubation medium [27]. Finally, highconcentrations of ClCCP may also induce respiratorychain inhibition [28].

For each ClCCP titration performed on growingyeast cell respiration, bell-shaped curves were ob-tained regardless of the growth phase (Fig. 6). More-over, the ClCCP concentration allowing the maximalstimulation of respiratory rates was higher as theculture medium pH increased (Fig. 6). Finally, itmust be noted that the stimulation of respiratoryrate by ClCCP was about twofold higher in the sta-tionary phase than in the exponential phases (100%vs. 40%). Considering batch cultures in the late ex-ponential and stationary phases, the restoration ofboth the initial pH value (5.5) and the initial lactatecontent (0.2%) by a fresh culture medium onlyslightly increased the basal respiratory £ux, whichremained lower than that measured in the early ex-ponential phase (Fig. 7A). Indeed, such environmen-tal conditions (increasing pH and substrate deple-tion) accounted for less than 10 and 20% of thedecline in the respiratory activity throughout thelate exponential and stationary phase period, respec-tively. Moreover, restoration of both initial pH andinitial lactate concentration in fresh medium did not

signi¢cantly increase the uncoupled respiratory ratesin the late exponential phase (Fig. 7B). Therefore,electron delivery from lactate to the respiratory chainis not a limiting step during this culture period ascompared to the early exponential phase. In the sta-tionary phase, respiratory rate stimulation inducedby ClCCP was abolished for yeast cells suspendedin a fresh culture medium (compare Fig. 7A andB). Nevertheless, pyruvate addition in the fresh me-dium restored this £ux stimulation (Fig. 7B), suggest-ing that the respiratory capacity of yeast cells is

Fig. 7. E¡ect of pH and lactate concentration on basal and un-coupled respiratory rates of growing W303-1a cells. W303-1acells were grown aerobically in minimal medium containing lim-iting amounts of D,L-lactate (0.2% w/v). For each growth phase,basal (A) and uncoupled respiratory rates (B) were determinedin the growth medium (white bars) and in fresh culture medium(pH 5.5; lactate 0.2%) (black bars). For B, striped bars repre-sent addition of 5 mM pyruvate in fresh culture medium. Val-ues are means þ S.D. of ¢ve independent experiments.

BBABIO 44812 15-2-00

L. Dejean et al. / Biochimica et Biophysica Acta 1457 (2000) 45^56 51

mainly due to pyruvate catabolism in the stationaryphase.

3.4.2. Control of respiration by ATP synthesisthroughout the culture period

To estimate the part of oxygen uptake coupled toATP synthesis in the di¡erent growth phases, weused the inhibitor of the F0F1 ATP synthase, tri-ethyltin (TET) [29]. In this case, the TET-insensitiveoxygen uptake (JOTET) represents the respiratoryactivity not associated to mitochondrial ATP synthe-sis.

Like the basal and the uncoupled respiratory rates(see Fig. 7A,B), TET-insensitive oxygen uptake ofcells decreased during the late exponential phase, toattain a minimal value in the stationary phase (Table1). The ratio between the maximal oxygen uptakeand the non-phosphorylating respiratory activity

(JOClCCP vs. JOTET) did not vary signi¢cantly withregard to the growth phase (between 3.3 and 3.5)(Table 1). This indicates that the respiratory chainactivity is controlled by the proton electrochemicalgradient to the same extent throughout the growthperiod, thus making an in situ uncoupling of themitochondria unlikely. Combining the values of thebasal respiratory rate, the TET-insensitive oxygenuptake and the maximal respiratory activity, we cal-culated the percentage of stimulation of the mito-chondrial oxidative phosphorylation compared toits maximal regime (JOClCCP). This parameter, whichrepresents a relative respiratory control of mitochon-dria within cells, was not signi¢cantly di¡erent in theearly and late exponential phases (i.e. about 60 vs.50% in the early and late exponential phase, respec-tively) and was about twice lower in the stationaryphase (i.e. about 30%) (Table 1).

Table 1Oxidative phosphorylation regime of yeast cells during lactate-limited growtha

Growth phase

Early exponential Late exponential Stationary

JO basal (nat.O/min/mg dry wt) 318 þ 22 223 þ 39 63 þ 14JOTET (nat.O/min/mg dry wt) 135 þ 16 99 þ 7 35 þ 3JOClCCP (nat.O/min/mg dry wt) 443 þ 18 345 þ 27 123 þ 12JOClCCP/JOTET 3.3 þ 0.2 3.5 þ 0.2 3.5 þ 0.2(JO basal3JOTET)/(JOClCCP3JOTET)U100 59 þ 4% 50 þ 6% 32 þ 3%aThe respiratory rates were measured in the growth medium without any addition (JO basal) or in the presence of 50 WM of the F0F1

ATP synthase inhibitor, TET (JOTET), as described in Section 2. The basal respiratory capacity of cells (JOClCCP) was measured ingrowth medium after titration with the protonophoric uncoupler, ClCCP. Values are means þ S.D. of ¢ve independent measurements.(JO basal3JOTET)/(JOClCCP3JOTET)U100 is the percentage of stimulation of oxidative phosphorylation compared to the basal respira-tory capacity.

Table 2Mitochondrial activities of yeast cells throughout the culture perioda

Growth phase

Early exponential Late exponential Stationary

JOmax (nat.O/min/mg dry wt) 612 þ 36 434 þ 39 225 þ 14D,L-Lactate dehydrogenase activity (mU) 195 þ 16 130 þ 21 102 þ 10Citrate synthase activity (mU) 318 þ 19 233 þ 45 121 þ 30aThe maximal respiratory capacity of cells (JOmax) was measured in the growth medium supplemented with saturating amounts of res-piratory substrates (lactate 0.2%, ethanol 1% and glucose 20 mM) after titration with the protonophoric uncoupler, ClCCP. Valuesare means þ S.D. of ¢ve independent measurements. D,L-Lactate dehydrogenase and citrate synthase activities were measured as de-scribed in Section 2. Values are means þ S.D. of six independent measurements.

BBABIO 44812 15-2-00

L. Dejean et al. / Biochimica et Biophysica Acta 1457 (2000) 45^5652

3.5. Modulation of mitochondrial content of yeastcells during lactate-limited batch growth

The uncoupled respiration of yeast cells measuredwith lactate 0.2% as respiratory substrate continu-

ously declined during the growth period (Fig. 7B).Moreover, Table 2 shows that the maximal respira-tory capacity (JOmax) measured in the presence ofClCCP and saturating amounts of lactate, glucoseand ethanol (0.2%, 20 mM and 100 mM in the cul-ture medium, respectively [9]) also decreased duringthe transition from the early to the late exponentialphase, as well as to the stationary phase. In this way,this growth phase-dependent modulation of the max-imal respiratory capacity of yeast cells was investi-gated in relation to the mitochondrial enzyme con-tent.

The ¢rst step of D,L-lactate catabolism of S. cere-visiae is the oxidation of lactate in pyruvate by twolactate dehydrogenases, which are connected to therespiratory chain between complexes III and IV (see[30,31] for reviews). This enzyme activity decreasedthroughout the growth period (from 195 þ 16 mU inearly exponential phase to 102 þ 10 mU in stationaryphase) (Table 2). Hence, this twofold decrease inlactate dehydrogenase activity could explain the de-crease in the uncoupled and non-phosphorylatingrespiratory activities (Table 1). Moreover, this is inagreement with the observation that the respiratorycapacity of yeast cells seems to be due mainly topyruvate catabolism in the stationary phase (Fig.7B).

Since the mitochondrial citrate synthase isoenzymehas been shown to account for 90% of the total cit-rate synthase activity [32], we measured the speci¢cactivity of citrate synthase to probe the mitochon-drial matricial enzyme content of yeast cells through-out the culture period. Table 2 shows that the citratesynthase content continuously declined during theentire growth period (from 318 þ 19 mU in early ex-ponential phase to 121 þ 10 mU in stationary phase).Moreover, Fig. 8A shows that the maximal respira-tory capacity of cells decreased proportionally to thecitrate synthase content during the late exponentialand the stationary phase. The contents of cyto-chromes a+a3, c+c1 and b also declined in parallelwith the citrate synthase activity (Fig. 8B). However,the extent of this decrease was about sevenfold forcytochrome a+a3 content whereas it was only two-and threefold for cytochrome c+c1 and b contents,respectively (Fig. 8B). Nevertheless, these results in-dicate that the modulation of the mitochondrial en-zyme content is involved in respiratory activity adap-

Fig. 8. Relationship between respiratory capacity, cytochromecontent and citrate synthase of batch grown W303-1a cells. (A)Respiratory capacity of cells as a function of their citrate syn-thase activity. Respiratory capacity (JOmax) was measured inthe growth medium supplemented with saturating amounts ofrespiratory substrates and in the presence of ClCCP (see Table2). Citrate synthase activity of the same cells was assayed as de-scribed in Section 2. I, II and III represent the early, late expo-nential and stationary growth phase, respectively. Values arefrom six independent experiments. (B) Content of cytochromesc+c1 (b), cytochrome b (a) and cytochromes a+a3 (O) of cellsas a function of the citrate synthase activity. Cytochrome con-tent and citrate synthase activity of cells were measured as de-scribed in Section 2. Values of cytochrome content are from asingle determination of four cultures.

BBABIO 44812 15-2-00

L. Dejean et al. / Biochimica et Biophysica Acta 1457 (2000) 45^56 53

tation during growth on limiting amounts of lactate(see Table 1).

4. Discussion

In this study, we used a microcalorimetric methodallowing construction of an enthalpy balance of yeastcells batch grown with a respiratory substrate [13,15].The main observation was that the part of lactateenergy input conserved as biomass (the enthalpicgrowth yield or energetic yield) was approximatelythe same in the early and the late exponential growthphases (32 þ 2% versus 37 þ 3%) whereas growth rateand respiratory rate continuously decreased duringthis phase transition (Table 3). The question is there-fore raised as to the mechanisms by which the celladapts its energy production to growth rate varia-tions, during phase transition, in such a mannerthat the energetic growth yield remains constant.

As already pointed out [23,36], the growth yield isa function of both the ATP turnover value and thepercentage of ATP synthesis used for the anabolicprocess compared to the maintenance process. Understeady state conditions where ATP synthesis matchesATP consumption, the cellular ATP turnover can beestimated by assuming mitochondrial oxidative phos-phorylation to be the major ATP production systemduring growth on lactate, and by taking into accountthe e¡ective ATP/O ratio of yeast cells. This param-eter is the amount of ATP produced per total oxygenconsumed at any steady state of respiration driven byproton leak and ATP synthesis [33,34]. Therefore,the e¡ective ATP/O ratio varies from zero under

non-phosphorylating conditions to the maximalATP/O that can be sustained in the presence of sat-urating amounts of Pi, ADP and respiratory sub-strate. The e¡ective ATP/O ratio for yeast cells wascalculated: (i) by using the relationship between theATP synthesis rate and the oxygen uptake of isolatedmitochondria [4], (ii) by taking the maximal ATP/Oratio to be 1.4 for lactate+pyruvate oxidation [5,35],and (iii) by considering the regime of oxidative phos-phorylation compared to its maximal regime (see Ta-ble 1). ATP/O ratio values of 1.15, 1.05 and 0.85were obtained for the early exponential, late expo-nential and stationary phase, respectively. The calcu-lated ATP turnover decreased simultaneously to thebasal respiratory rate during the transition from theearly to the late exponential growth phase (Table 3).Hence, the constant enthalpic growth yield of yeastcells during the transition between the early and thelate exponential phase is due to the fact that ATPproduction decreases simultaneously to the growthrate. Moreover, in the stationary phase, when theATP needs for growth are null, the ATP turnoverrepresented about 15% and 23% of that calculatedin the early and late exponential growth phases (Ta-ble 3), thus indicating that the ATP requirements formaintenance are rather small compared to thegrowth-dependent ATP needs, in agreement with[23].

In fully oxidative metabolism, the ATP £uxcoupled to respiratory chain activity in the cell de-pends on the functional steady state of mitochondria.When reduced equivalent delivery is not a limitingstep, this steady state is de¢ned by a number ofcharacteristics, namely the v~WH� value, the phos-

Table 3Growth yields, basal respiratory activity and calculated ATP turnover rate of yeast cells during lactate-limited growtha

Growth phase

Early exponential Late exponential Stationary

YX=S (g dry weight/g lactate) 0.25 þ 0.01 0.28 þ 0.02 ^Enthalpic growth yield (%) 32 þ 2 37 þ 3 ^JO basal (nat.O/min/mg dry wt) 318 þ 22 223 þ 39 63 þ 14JATP (nmol ATP/min/mg dry wt) 366 þ 31 234 þ 38 54 þ 12aGrowth yields were determined by calculating the part of enthalpy equivalent of lactate input conserved as biomass (enthalpic growthyield) (see Fig. 2 and Section 3) or by plotting cumulative biomass production versus cumulative lactate consumption (YX=S) (see Fig.2A). The respiratory rates were measured in the growth medium as described in Section 2. The ATP turnover rate, JATP, was calcu-lated from the basal respiratory rate multiplied by e¡ective ATP/O ratio values of 1.15, 1.05 and 0.85 for early, late exponential andstationary phase, respectively, as described in the text.

BBABIO 44812 15-2-00

L. Dejean et al. / Biochimica et Biophysica Acta 1457 (2000) 45^5654

phate potential and the percentage of stimulation ofthe respiratory rate, and by the mitochondrial en-zyme content. Hence, the energetic yield of growingcells is also a function of the percentage of electron£ux that is e¡ectively coupled to ATP synthesis inthe cell. For instance, with a constant amount ofmitochondria, the decrease in ATP demand by thegrowth processes might lead to an increase in non-phosphorylating respiration (originating from protonleak and redox slipping [5,37]) compared to basaloxygen uptake, thus increasing the part of heat pro-duction compared to biomass production, and inturn decreasing the energetic growth yield. This im-plies that mitochondria by themselves are the majorheat dissipative system when yeast cells are cultivatedwith a purely respiratory substrate. In fact, weclearly show that a long-term adaptation prevailsover the short-term one since the mitochondrial en-zyme content decreased during growth phase transi-tion. Such a parallel decrease in respiratory activitiesand mitochondrial content per cell has already beendescribed during the transition between the early andlate exponential growth phases of mammalian cellculture [10^12].

In conclusion, our study indicates that mitochon-dria by themselves are the major energy dissipativesystem in a fully aerobic metabolism. Moreover, thisenergy dissipation can be regulated by the decrease inmitochondrial enzyme content to maintain the oxi-dative phosphorylation regime and, therefore, the en-ergetic growth yield during the exponential growth ofyeast cells batch grown on a respiratory substrate.

Acknowledgements

The authors wish to thank Dr. R. Cooke for hiscontribution to the editing of the manuscript. Thiswork was supported by grants from the Conseil Re-gional d'Aquitaine and the Programme de rechercheCNRS `Genie des procedes chimiques, physiques etbiotechnologiques'.

References

[1] M. Klingenberg, Biochem. Z. 335 (1961) 231^272.[2] M. Erecinska, D.F. Wilson, J. Membr. Biol. 70 (1982) 1^14.

[3] J.M. Tager, R.J.A. Wanders, A.K. Groen, W. Kunz, R.Bohnensack, U. Ku«ster, G. Letko, G. Bo«hme, J. Duszynski,L. Wojtczack, FEBS Lett. 151 (1983) 1^9.

[4] J.P. Mazat, E. Jean-Bart, M. Rigoulet, B. Guerin, Biochim.Biophys. Acta 849 (1986) 7^15.

[5] V. Fitton, M. Rigoulet, R. Ouhabi, B. Guerin, Biochemistry33 (1994) 9692^9698.

[6] D.G. Nicholls, R.M. Locke, Physiol. Rev. 64 (1984) 1^64.[7] R.S. Balaban, Am. J. Physiol. 258 (1990) C377^C389.[8] G.C. Brown, Biochem. J. 284 (1992) 1^13.[9] B. Beauvoit, M. Rigoulet, O. Bunoust, G. Ra¡ard, P. Can-

ioni, B. Guerin, Eur. J. Biochem. 214 (1993) 163^172.[10] S. Walenta, A. Bredel, U. Karbach, L. Kunz, L. Vollrath,

W. Mueller-Kiesler, Adv. Exp. Med. Biol. 248 (1989) 847^853.

[11] A. Bredel-Geissler, U. Karbach, S. Walenta, L. Volrath, W.Mueller-Klieser, J. Cell. Physiol. 153 (1992) 44^52.

[12] M. Martin, B. Beauvoit, P.J. Voisin, P. Canioni, B. Guerin,M. Rigoulet, J. Bioenerg. Biomembr. 30 (1998) 565^578.

[13] C. Larsson, A. Blomberg, L. Gustafsson, Biotechnol. Bio-eng. 38 (1991) 447^458.

[14] L. Gustafsson, Thermochim. Acta 193 (1991) 145^171.[15] U. von Stockar, L. Gustafsson, C. Larsson, I. Marison, P.

Tissot, E. Gnaiger, Biochim. Biophys. Acta 1183 (1993) 221^240.

[16] S.L. Randzio, J. Suurkuusk, in: A.E. Beezer (Eds.), Biolog-ical Microcalorimetry, Academic Press, London, 1980, pp.311^341.

[17] H.U. Bergmeyer, in: Methods of Enzymatic Analysis, vol.VI, Academic Press, New York, 1974, pp. 570^577.

[18] T.P. Singer, E. Rocca, E.B. Kearney, in: E.C. Slater (Ed.),Flavins and Flavoproteins, Elsevier, Amsterdam, 1966, pp.391^426.

[19] P.A. Srere, Methods Enzymol. 13 (1969) 3^11.[20] T. Ohnishi, A. Kruger, H.W. Held, E. Pfa¡, M. Klingen-

berg, Eur. J. Biochem. 1 (1967) 301^311.[21] W.H. Vanneste, Biochim. Biophys. Acta 113 (1966) 175^

182.[22] Y. Guan, R.B. Kemp, in: H.V. Westerho¡, J.L. Snoep, J.E.

Wijker, F.E. Sluse, B.N. Kholodenko (Eds.), Biothermoki-netics of the Living Cell, Biothermokinetics Press, Amster-dam, 1996, pp. 387^397.

[23] R. Lagunas, Biochim. Biophys. Acta 440 (1976) 661^674.[24] A. Blomberg, C. Larsson, L. Gustafsson, J. Bacteriol. 170

(1988) 4562^4568.[25] E. Gnaiger, R.B. Kemp, Biochim. Biophys. Acta 1016 (1990)

328^332.[26] F. Cassio, C. Leao, N. Van Uden, Appl. Environ. Microbiol.

53 (1987) 509^513.[27] S.G.A. Mac Laughlin, J.P. Dilger, Physiol. Rev. 60 (1980)

825^863.[28] R. Steinlechner-Maran, T. Eberl, M.L. Kunc, R. Margreiter,

E. Gnaiger, Am. J. Physiol. 271 (1996) C2053^C2061.[29] K. Cain, D.E. Gri¤th, Biochem. J. 162 (1977) 575^580.[30] S. de Vries, C.A.M. Marres, Biochim. Biophys. Acta 895

(1987) 205^239.

BBABIO 44812 15-2-00

L. Dejean et al. / Biochimica et Biophysica Acta 1457 (2000) 45^56 55

[31] B. Guerin, in: A.H. Rose, J.S. Harrison (Eds.), The Yeasts,vol. 4, Academic Press, London, 1991, pp. 541^600.

[32] G. Kispal, M. Rosenkrantz, L. Guarente, P.A. Srere, J. Biol.Chem. 263 (1988) 11145^11149.

[33] M.E. Harper, M.D. Brand, J. Biol. Chem. 268 (1993) 14850^14860.

[34] M.D. Brand, M.E. Harper, H.C. Taylor, Biochem. J. 291(1993) 739^748.

[35] N. Averet, V. Fitton, O. Bunoust, M. Rigoulet, B. Guerin,Mol. Cell. Biochem. 184 (1998) 67^79.

[36] A.H. Stouthamer, C. Bettenhaussen, Biochim. Biophys. Acta301 (1973) 53^70.

[37] R. Ouhabi, M. Rigoulet, J.L. Lavie, B. Guerin, Biochim.Biophys. Acta 1060 (1991) 293^298.

BBABIO 44812 15-2-00

L. Dejean et al. / Biochimica et Biophysica Acta 1457 (2000) 45^5656