7/29/2019 1779.full

1/10

Journal of General Microbiology (1986), 132, 1779-1788. Printed in Great Britain 1779

Influence of the Carbon:Nitrogen Ratio of the Growth Medium on theCellular Composition and the Ability of the Methylotrophic Yeast

Hansenula polymorpha to Utilize Mixed Carbon SourcesBy TH. G L I * * A N D . R . Q U A Y L E 2 t

Swiss Federal Institute for Wa ter Resources and Wa ter Pollution Control, Swiss FederalInstitutes of Technology, CH-8600 Diibendorfl Switzerland2Departmentof Microbiology, The University, She field SlO 2TN, U K(Received 13 January 1986; revised 8 March 1986)

The methylotrophic yeast Hansenula pozymorpha was grown in a chemostat with a mediumcontaining a mixture of glucose (c6) and methanol (C , ) (87.8% c6: 12.2% C1,w/w) as solecarbon source and N H,+ as sole nitrogen source. At a constant growth ra te ( D 0.10 h-l) theinfluence of the carbon :nitrogen ratio (C:N) of the inflowing medium on the cellular andenzymic composition of the cells was studied. Th ree distinct growth regimes were recognized. Amedium with a C :N ratio c 2 resulted in carbon-limited growth (high cellular protein con tent ,low carbohydrate content) and under these conditions glucose and methanol were utilizedsimultaneously. A medium w ith a C : N ratio >3 1 resulted in nitrogen-limited grow th (lowprotein but high carbohydrate content of the cells) and the cells metabolized only glucose. Atransition growth regime was observed during growth o n media w ith intermediate C :N ratios(12 < C : N > 31). When assessed from both substrate consumption and cellular composition,growth was double-substrate (carbon and nitrogen)-limited. In this transition growth regime,changes in carbon metabolism an d the cellular and enzymic composition of the cells were found.With increasing C:N ratios in the growth medium a gradual repression of the synthesis ofmethanol-assimilating an d dissimilating enzymes was found. This effect was most pronounce dfor alcohol oxidase, and as a consequence the cells switched from the utilization of t he ca rbonsubstrate mixture to growth o n glucose alone. Th e data presented suggest tha t the range withinwhich double-substrate-limitedgrowth can be expected is predictable from the composition ofcells grown under single substrate limitation.

I N T R O D U C T I O NIn all ecosystems one of the key factors controlling the rate of cell growth is the availability ofnutrients. In contrast to multicellular organisms, microbes can adapt relatively quickly to

various nutrient limitations by changing their cellular composition. Single nutrient limitationssuch as carbon, nitrogen or inorganic ions have been much studied but little attention has beendirected to the fact that growth, as it occurs in nature, is not necessarily limited by a singlenutrient. M icrobes ar e unlikely to grow over an extended period of time under 'steady-state'conditions but rather under conditions where the supply of nutrients fluctuates considerably.Only recently have microbiologists started to investigate the physiology and growth behaviourof microbes under multi-substrate-limited, gradient or transient growth conditions (e.g.Dijkhuizen, 1979; Gottschal, 1980; Wim penny et al., 1983; Leegwater, 1983).t Present address: University of Bath, Bath BA2 7AY, U K.Abbreviations: C1,methanol; C6, glucose; co, no, concentration of carbon or nitrogen, respectively, in theinflowing medium (gl-I); c, n, concentration of carbon or nitrogen, respectively, in the culture liquid

(gl-l; mg1-I); D , dilution rate (h-l); qc, qc,, q c l , qN, specific rate of carbon, glucose, methanol or nitrogenconsumption, respectively [g substrate (g biomass)-' h-I I; x, biomass concentration (g - ' ) ; Yx,c, xIN,rowthyield with respect to carbon or nitrogen, respectively [g biomass (g substrate)-']; G D H , glutamate dehydrogenase;GS, glutamine synthetase; GOGAT, glutamate synthase.0001-32030 986 SG M

7/29/2019 1779.full

2/10

1780 T H . EGLI A N D J . R . Q U A Y L EDuring double-substrate-limitedgrowth of methylotrophic yeasts with glucose and methanol(Egli et al., 1980, 1982) the simultaneous utilization of glucose and methanol in chemostatcultures was strongly dependent on the nature of the growth limitation. During carbon-limitedgrowth the cells were able to utilize both substrates simultaneously and completely, whereasunder nitrogen limitation the synthesis of enzymes required for methanol utilization wasrepressed, and therefore glucose was the only source of carbon metabolized by the cells. Thekineticsof this repression have also been studied (Egli, 1982). In an experiment where the ratioof C :N in the feed to the chemostat was gradually raised to achieve nitrogen-limited growth, thecells were for some time under conditions where neither residual carbon (glucose and/ormethanol) nor nitrogen (NHZ) could be detected in the culture. This indicates that betweencarbon limitation (nitrogen in excess) and nitrogen limitation (carbon source in excess) anadditional growth regime exists where growth is limited by both nutrients. As the growth ofmicrobes under such transition conditions has only been reported twice (Hueting & Tempest,1979; Grazer, 1985) we decided to study the phenomenon in more detail and to investigate theinfluenceof the C :N ratio in the inflowing medium on the physiologyof a methylotrophic yeast,particularly its ability to utilize mixtures of glucose and methanol simultaneously.

M E T H O D SOrganism and cultivation. Hansenulapolymorpha (CBS 4732) was grown in continuousculture (37 "C, pH 5.0) on asynthetic medium (Egli & Fiechter, 1981). A mixture of glucose (C,) and methanol (C,) of constant composition(87.8%,C, :12.2%C , , w/w) was used as the sole source of carbon and energy; NH$ was the only source of nitrogen.The C :N atio in the inflowing medium was varied by changing the ratio C6/Cl-mixture:NHZ. The C :N ratio ofthe medium indicated on the abscissa in the Figures was always calculated on the basis of total carbon from bothglucose and methanol, whether utilized or not, and of nitrogen from NH$. To avoid wall growth due to methanolaccumulation in the fermenter, the concentration of this substrate in the feed was always kept below 1-2g 1-l.Dissolved oxygen tension was kept above 50 % air saturation and the concentration of dry biomass in thefermenter was always lower than 5.0 g 1-l to ensure that growth of the cells was always limited by either carbon ornitrogen.Substrate concentrations.Glucose was measured enzymically (GOD-Perid test, Boehringer) and methanol wasanalysed by gas chromatography as reported by Egli et al. (1983). The detection limits for glucose and methanolwere 2-0mg1-l and 1.0 mg l-l, respectively. NHZ was measured spectrophotometricallyusing the indophenolmethod of Scheiner (1976). The reproducibility of the measurement of residual NHi-N was < f % atconcentrations>0.10g 1-l; between 0.10 and 0.0005 g 1-1 it was < & 20% and reproducible measurements werenot possible at concentrations lower than 0.0005 g NHZ 1-l. To avoid substrate losses due to substrateconsumption by the cells after sampling, the residual concentrations of glucose, methanol and NHd in the culturemedium were assayed by the filtration echnique described by Egli et al. (1983). With this technique the cells couldbe separated from the medium in less than 0.5 s.Cellular components.For dryweight measurements the cells were collected by centrifugation, washed twice withdistilled water and dried to constant weight at 105"C.Parallel samples varied by < +_ 1%. For the determinationof cellular components, the cells were collected from the chemostat outlet and immediately cooled to 0 "C; theywere then harvested by centrifugation at 4 "C, washed twice with sodium phosphate buffer (50 mM, pH 7.0),freeze-dried and stored. Before the individual analyses, the freeze-dried cells were dried again over phosphoruspentoxide.The content of carbon, hydrogen and nitrogen in freeze-dried cells was measured with a C, H, N-Analyzermodel 185(F& M Scientific).All values reported are the mean of at least three determinations and the deviationsfor C, H and N for separate samples were always < f 4%, +_ 11.9% and f -7%,respectively, of the absolutevalue.

The biuret method, as described by Herbert et al. (1971), was used for the determination of the protein content offreeze-dried cells.

Glycogen was measured by the method of Becker (1978).The content of total carbohydrates was determined by the phenol method described by Herbert et al. (1971).Enzyme assays. Cell-free extracts were prepared as described by Egli et al. (1983). All enzyme assays were done

at 37 "C. The following enzymes were measured according to previously published methods: alcohol oxidase(EC 1.1.3.13; van Dijken et al., 1976), dihydroxyacetone kinase (van Dijken et al., 1978), formaldehydedehydrogenase (EC 1 .2.1 .1; Egli et al. , 1980), formate dehydrogenase (EC 1.2.1.2; Egli et al., 1980), glucose-6-phosphate dehydrogenase (EC 1.1.1.49; Kornberg & Horecker, 1955), glucose-6-phosphate isomerase(EC 5.3.1.9; Egli et al., 1983), 6-phosphogluconate dehydrogenase (EC 1.1.1.44; Egli et al., 1982), ribose-5-phosphate isomerase (EC 5.3 .1 .6 ; Wood, 1970), ribulose-5-phosphate 3-epimerase (EC 5. 1.3.1;Wood, 1970),

7/29/2019 1779.full

3/10

GIN-limited growth of a methylotrophic yea st 1781transaldolase (EC 2.2.1 . 2; Tchola& Horecker, 1966), transketolase (EC 2. 2. 1 - 1 Waites& Quayle, 1981). Theassay for NADP+-dependent glutamate dehydrogenase (EC 1 4.1.2) contained, per ml : potassium phosphatebuffer pH 7.8, 100 pmol; 2-oxoglutarate, 3.3 pmol; NH,Cl, 33.3 pmol; NADPH, 0.50pmol; and cell-free extract.The reaction was started by adding 2-oxoglutarate. The Lowry method was used to determine the proteinconcentrations in cell-free extracts. Bovine serum albumin (Armour) was used as the standard.

Enzyme units. The specific activities of all enzymes are expressed as the rate of substrate consumption (orproduct formation) per unit of protein [clmol min-l (mg protein)-'].

R E S U L T SH .polymorpha was grown in chemo stat culture (D 0.10h-l) on a glucose-methanol mixture ofconstant composition, and with NH,+as sole nitrogen source, as described in Methods. Theexperim ent was started by growing cells on the medium w ith the lowest C :N ratio used in thisstudy. After analysis of the resulting steady-state the ratio w as increased stepwise. The C :Nratios, the com position of the media used and the exact residual concentrations of the substratesof interes t are given in Table 1. On the basis of subs trate utilization by the cells, and their cellularand enzymic composition, three d ifferent grow th regimes could be recognized as the C :N ratioof the supplied medium was chang ed. The most important characteristics of these regimes are

summ arized in Table 2. In all the figures, the transition grow th regime is indicated by a shadedarea.Table 1. Substrate and dry biomass concentrations during cultivation of H . polymorphaH . polymorpha was grown in a chemostat at a constant dilution rate of 0.10 h-l with media of differentC :N ratios. All concentrations are given in g 1- I .

Dry cell weight formedSubstrate concentrations and residual substratein the inflowing medium concentrationsAf A 7 I 3C:N c6 C1 NH2-N Dry wt c6 C1 NH2-N

1 oo2.796.6311.3214.4615.8918.4021.9526.7335.0440.1359.11

4-564.564-284.154.444.344.6 15-536.698.856.995-89

0.670.670.640.6 10.630-620.640.720.9 11.130.900.82

2.0880.7460-2940.1670.1390-1240.1 130.1130.1 130-1130.0780-045

2.7 12.862.702.402-442.6 12-693-203.794.743.081-42-, Not detected (c6 < 2 mg 1 - l ; C, < 1 mg 1-l; NH2N < 0.5 mg 1-l).

Table 2. Characterization of diferent growth regimes observed during the cultivation ofH . polymorphaH. olymorpha was grown in a chemostat at a constant dilution rate of 0.10 h-l with media of differentC :N ratios. All concentrations are given in mg 1-l .

C : N ratioGrowth range ofregime mediaCarbon < 11.32limitationTransition 11.32< C :N < 18.4018-40 1 C3.5 Low High LOW

7/29/2019 1779.full

4/10

1782 TH. E G L I A N D J . R. Q U A Y L E

4 c , -0.020 -0.015 -0.010 -0.005-

0 -

4c0.100

0.0750.050

0.025010 20 30 40 50

C :N ratioFig. 1. Influence of the C :N ratio in the inflowing medium on the utilization of a mixture of glucoseand methanol (87.8%C6 12.2%C,,w/w) by H.polymrpha growing in a chemostat atadilution rate of0.10h-l. (a ) Residual concentration of NHi-N (a), lucose (n) and methanol (0 ) . b) Specificsubstrate consumption rates [g (gdry wt h)-l] for glucose (qc,, A!, methanol(qc,, O),NHa-N (qN,D)and total carbon (qc, +). The shaded area indicates the transition growth regime.

Utilization of carbon and nitrogen sourcesThe residual concentrations of glucose, methanol and NHX, and the specific rates of substrateconsumption, are shown in Fig. 1 as a function of the C :N ratio of the inflowing medium. As canbe seen from the residual substrate concentrations in the culture (Fig. la , Table 1) there werefour different stable physiological states. (1) At low C :N ratios (

7/29/2019 1779.full

5/10

GIN-limited growth of a methylotrophic yeas t 1783

10 20 30 40 50C N ratio

Fig. 2. Elemental and cellular composition of H. polymrpha as a function of the C:N ratio in theinflowing medium. Culture conditions were as specified in Fig. 1 . (a )Cellular content of carbon to),hydrogen (n> and nitrogen (0) . b) Cellular content of protein (a),glycogen (B) and totalcarbohydrate (r). he shaded area indicates the transition growth regime.

The specific rate of nitrogen consumption was constant (6.31 x f:0.03 x h-l) bothas long as the nitrogen source was present in excess, and during nitrogen-limited growth withC : N ratios d 5.04 (qN 2.75 x f 0.35 x h-l). In between carbon-limitation andnitrogen-limitation,q N reflected the supply of nitrogen to the cells and decreased as the C :Nratio of the inflowing medium was increased.

Cellular compositionCells of H.polymorpha grown under the conditions described were analysed to determine theirelemental composition (C , N , H) and their protein, total carbohydrate and glycogen content(Fig. 2). The hydrogen content showed little variation. The cellular carbon content was slightlyhigher in cells grown under carbon-limited conditions (49.1 f: 1.2%) than in cells grown undernitrogen-limited conditions (45.7 f 0.9%). The cellular nitrogen content was high (7.5-8.0%) aslong as the cells were supplied with a medium with a C :N ratio d 14.46 and decreased graduallywith increasingC :N ratios. Growth under nitrogen-limited conditions (C:N 2 35.04) resultedin cells with a nitrogen content of approximately 3.7%.The total carbohydrate, glycogen and protein content (Fig. 2b) followed the expected pattern :low carbohydrate/high protein content during carbon-limited growth and high carbohydrate/low protein content during nitrogen-limited growth. Transition from carbon-limited cellcomposition to nitrogen-limited cell composition was a gradual process; within the range of- 13< C : N < - 0 the cell was able to adjust its composition in response to pertainingenvironmental conditions. It should be noted that the carbohydrate content assay includescontributions from both DNA and RNA.

7/29/2019 1779.full

6/10

1784 TH. E G L I A N D J . R. Q U A Y L E

10 20 30 40 50C:N ratio

Fig. 3. Enzyme specific activities in cell-free extracts of H. olymorphaas a function of the C :N ratio inthe inflowing medium. Culture conditions were as specified in Fig. 1 . (a ) Alcohol oxidase (O),formaldehyde dehydrogenase (a) b ) ormate dehydrogenase (a), ihydroxyacetone kinase (i) (c)ribulose-5-phosphate pimerase(W), ribose-5-phosphate somerase (a). he shaded area indicates thetransition growth regime.Specific activities of enzymes

The specific activities of enzym es involved in the metabo lism of glucose and m ethanol and inthe assim ilation of NH ,+are shown in Figs 3 and 4. As expected from the sub strate utilizationpattern (Fig. l a ) enzymes involved in the metabolism of methanol were synthesized undercarbon-limited growth conditions and their synthesis was repressed when nitrogen was thegrow th-limiting substrate. How ever, two different regulatory patterns were observed. Theperoxisomal enzyme alcohol oxidase (Fig. 3a ) was synthesized at a reduced rate whenthe residual nitrogen concentration in the culture was between 16.1 and 3.5mg1-l(11.32 < C : N < 14.46). Repression was complete when the C : N ratio in the feed was >21.95.This clear onset of repression was not observed for the two me thanol dissimilating enzymesformaldehyde and formate dehydrogenase, nor for the C -assimilating enzyme dihydroxy-acetone kinase (Fig. 3a, b).Ri bose-5-phosphate isomerase, ribulose-5-pho sphateepimerase, transaldolase and transketo-lase take part in bo th the assimilation of C,-units from methanol (dihydrox yacetonepathway)and the metabolism of glucose. Except for ribose-5-phosphate isomerase, their specific activity

7/29/2019 1779.full

7/10

CIN-limited growth of a methylotrophic yeas t 1785

10 20 30 40 50C :N ratio

Fig. 4. Enzyme specific activities in cell-free extractsof H.olymorpha as a function of the C :N ratio inthe inflowing medium. Culture conditions were as specified in Fig. 1 . (a ) Transaldolase (n),transketolase (m) ; b) glucose-6-phosphate dehydrogenase (O), 6-phosphogluconate dehydrogenase(a); (c) glucose-6-phosphate isomerase(a), lutamate dehydrogenase(i). he shaded area indicatesthe transition growth regime.was either approximatelyconstant ordecreased slightly with increasingC :N ratios as long as noresidual glucose was detected in the culture (Figs 3 c , 4 a ) . When residual glucose was presenttheir specific activity was enhanced two to threefold. Glucose-6-phosphatedehydrogenase,6-phosphogluconate dehydrogenase (Fig. 4b) and glucose-6-phosphate isomerase (Fig. 4c), threeenzymes which are primarily involved in glucose metabolism, responded similarly. The specificactivity of ribose-5-phosphate isomerase showed a very similar pattern to those enzymesinvolved in methanol metabolism. This indicates the important role of this enzyme in therearrangement reactions of the dihydroxyacetone cycle of formaldehyde fixation.In cell-free extracts of H . polymorpha the specific activity of glutamate dehydrogenase(GDH), the key enzyme in the assimilation of NH,+n yeasts, increased approximately 50-foldfrom0.057pmol min-' (mg protein)-' to a maximum of 2.575 pmol min-l (mg rotein)-' as theC :N ratio of the medium was increased from 1.00 to 15-89.A further increase in the C :N ratio to30 resulted in a reduction of the specific activity of GDH to approximately 1 pmol min-' (mgprotein)-' ;growth with C :N ratios >30 had little further effect on the specific activity.

7/29/2019 1779.full

8/10

1786 T H . E G L I A N D J . R . Q U A Y L ED I S C U S S I O N

Modes of growth limitationIn our experiment, the classic concept of growth limitation, namely that a linear relationshipexists between biomass produced and substrate concentration (compare e.g. Stephenson, 1949;Tempest, 1969), is valid as long as either the nitrogen or the carbon source is clearly present in

excess. The cellular composition with respect to total protein and carbohydrates is then notinfluenced by changes of the medium C :N ratio. However, when media with 12< C :N > 45are fed to the cells, and both the carbon and the nitrogen sources are completely metabolized, theconcept of single substrate limitation no longer applies. Within this C :N range, the organismresponds to a change in the ratio of the two limiting substrates by altering its cellularcomposition to match the nutrient intake. However, if it is assumed that, instead of a singlesubstrate, the combination of carbon and nitrogen source, at a fixed C :N ratio, is growth-limiting, the concept of growth limitation within the transition growth regime can still beapplied for a fixed growth rate.Only two examples of double-substrate-limited growth of aerobic microbes have beenreported. Hueting & Tempest (1979) studied the pattern of overflow metabolite production byKlebsiella aerogenes during the transition from carbon to either potassium or ammonialimitation. Double substrate-limitation was also observed for a culture of HyphomicrobiumZV620 growing with methanol and ammonia as the source of carbon and nitrogen (Grazer,1985). The phenomenon is probably widespread, occurring whenever organisms are grownunder conditions of substrate limitation in media of different ratios of two substrates. The effectmay be predicted from the known growth yields for the individual substrates as our own dataillustrate. The concentration of biomass in the culture (x) can be calculated from equation (l ),where co and no are the concentrations of carbon and nitrogen in the feed,c and n are the residualconcentrationsof carbon and nitrogen in the culture, and Yxlc and YXINre the growth yields forcarbon and nitrogen, respectively, measured under either carbon-limited or nitrogen-limitedgrowth conditions.

Because both c and n are zero under double-substrate-limitation, equation (1) can berearranged to yield the C :N ratio of the inflowing medium (equation 2). In order to estimate theminimum C :N ratio at which the cells switch from carbon-limited to carbon/nitrogen-limitedgrowth Yxlc nd YXINmeasured under carbon-limited growth conditions (1 930 and 16.0,respectively) are inserted in equation 2. The critical C :N ratio of 12.3 calculated in this waycorresponds closely with the experimentally observed value (Fig. la). Using Yxlc and YXINmeasured for a nitrogen-limited culture the same procedure would yield the critical mediumC :N ratio for the maximum C :N ratio. Equation (1) indicates that in order to obtain double-substrate-limited growth with substrates S1 and Sz, the growth yields Y Xl s l nd/or Y X p , ave tobe different under S1 limited and S,-limited growth conditions. Therefore, the wider the rangewithin which a cell component affected by the substrate(s) studied can vary, the more extendedthe double-substrate-limited growth regime will be.Since the cellular composition does also vary with growth rate, and differences are morepronounced in slow growing cells (Herbert, 1976), one can expect the extension of the transitiongrowth regime to become narrower with increasing growth rates.

Cellular and enzymic cornpositionThe generally observed phenotypic response of microbes brought about by the limitedavailability of the two primary nutrients carbon and nitrogen may be summarized as follows (forreviews see Harder & Dijkhuizen, 1983; Tempest et al., 1983; Harder et al., 1984; Tempest &

Neijssel, 1978). Carbon-limited growth results in cells with low carbohydrate and storagepolymer and high protein contents. Such cells are usually able to use mixtures of carbonsubstrates simultaneously due to derepression of many catabolic enzymes. Cells grown under

7/29/2019 1779.full

9/10

CIN-limited growth of a methylotrophic yeast 1787nitrogen-limitation have a reduced protein content and high carbohydrate and storage polymercontents; surplus carbon is frequently excreted into the medium in the form of an intermediarymetabolite.Two studies with H.polymorpha and Hyphomicrobium ZV620 (Grazer, 1985) clearly documentthe different regulation patterns which can be observed for the individual enzymes as the C :Nratio in the growth medium is gradually increased. In Hyphomicrobium ZV620, where NH,+ sassimilated either via GDH at high residual NHJ concentrations, or the high affinity enzymesystem of glutamine synthetase (GS)/glutamate synthase (GOGAT), an 'all-or-nothing' type ofresponse was observed for the specific activity of GDH. In this bacterium GDH was the mainenzyme responsible for the assimilation of NH,+during growth with C :N ratios

7/29/2019 1779.full

10/10

1788 T H . E G L I A N D J . R . Q U A Y L EVA N DIJKEN,. P., HARDER,., EARDSMORE,. J , &QUAYLE,. R. (1978). Dihydroxyacetone: an inter-mediate in the assimilation of methanol by yeasts?FEMS Microbiology Letters 4, 97-102.DIJKHUIZEN,. (1979). Regulation of autotrophic andheterotrophic metabolism in Pseudomonas oxalaticusOX 1 . PhD thesis, University of Groningen,Holland.EGLI,TH. (1982). Regulation of protein synthesis inmethylotrophic yeasts: repression of methanol dissi-milating enzymes by nitrogen limitation. Archives ofMicrobiology 131, 95-101.EGLI,TH.& FIECHTER,. (1981). Theoretical analysisof media used in the growth of yeast on methanol.Journal of General Microbiology 123, 365-369.EGLI,TH.,AN DIJKEN,.P., VEENHUIS, M., HARDER,W. FIECHTER,A. (1980). Methanol metabolism inyeasts: regulation of the synthesis of catabolicenzymes. Archives of Microbiology 124, 115-121.EGLI, TH., KAPPELI,0. & FIECHTER, . (1982).Regulatory flexibility of methylotrophic yeasts inchemostat cultures: simultaneous assimilation ofglucose and methanol at a fixed dilution rate.Archives of Microbiology 131, 1-7.EGLI,TH., LINDLEY, . D. & QUAYLE,. R. (1983).Regulation of enzyme synthesis and variation ofresidual methanol concentration during carbon-limited growth of Kloeckera sp. 2201 on mixtures ofmethanol and glucose. Journal of General Microbi-GOTTSCHAL,. C. (1980). Mixotrophic growth ofThiobacillus A 2 and its ecological significance. PhDthesis, University of Groningen, Holland.G R ~ E R ,. (1985). Chemostatstudie u e r die Regulationder ammoniumassimilierenden Enzyme in Hyphomicro-bium ZV620.Dissertation no. 7790, ETH, Zurich,Switzerland.HARDER,W. & DIJKHUIZEN,. (1983). Physiologicalresponses to nutrient limitation. Annual Review ofMicrobiology 37, 1-23.HARDER, ., DIJKHUIZBN,. & VELDKAMP,. (1984).Environmental regulation of microbial metabolism.Symposia of the Society for General Microbiology 36,HERBERT, ., PHIPPS,P. J. & STRANGE,. E. (1971).Chemical analysis of microbial cells. Methods inMicrobiology 5B,209-344.HERBERT,. (1976). Stoicheiometric aspects of micro-bial growth. In Continuous Culture 6:Applicationsand New Fields, pp. 1-30. Edited by A. C. R. Dean,D. C. Ellwood, C. G. T. Evans & J. Melling.Chichester: Ellis Horwood.

ology 129, 1269-1 281.

51-95.

HUETING,.& TEMPEST,. W. 1979). Influenceof theglucose input concentration on the kinetics ofmetabolite production by Klebsiella aerogenesNCTC418, growing in chemostat culture in potas-sium-or ammonium-limitedenvironments.Archivesof Microbiology 123, 189-194.KORNBERG,. & HORECKER,. L. (1955). Glucose-6-phosphate dehydrogenase.Meth ods in Enzymology 1,LEEGWATER,. P. M. (1983). Microbial reactivity: itsrelevance to growth in natural and artificial environ-ments. PhD thesis, University of Amsterdam,Holland.SCHEINER,. (1976). Determination of ammonia andKjeldahl nitrogen by indophenol method. WaterResearch 10, 31-36.STEPHENSON,. (1949). Growth and nutrition. InBacterial Metabolism ,3rd edn, pp. 159-178. London:Longman, Green.TCHOLA,.& HORECKER,. L. (1966). Transaldolase.Methods in Enzymology 9, 499-505.TEMPEST, . W. 1969). Quantitative relationshipsbetween inorganic cations and anionic polymers ingrowing bacteria. Symposia of the Society for GeneralMicrobiology 19, 87-1 11.TEMPBST, . W. & NEWEL, 0. M. (1978). Eco-physiological aspects of microbial growth in aerobicnutrient-limited environments. Advances in Micro-bial Ecology 2, 105-153.TEMPEST,. W., NEIJSSEL, . M. & ZEVENBOOM,.(1983). Properties and performance of micro-organisms in laboratory culture: their relevance togrowth in natural ecosystems.Symposia of the Society

for General Microbiology 34 , 119-152.VEENHUIS, ., VA N DIJKEN, . P. & HARDER,W.(1983). The significance of peroxisomes in themetabolism of one-carbon compounds in yeasts.Advances in Microbial Physiology 24, 1-82.WAITES,M. J. & QUAYLE,. R. (1981). The inter-relation between transketolase and dihydroxyace-tone synthase activities in the methylotrophic yeastCandkia boidinii. Journal of General Microbiology 124,WIMPENNY,. W. T., LOVITT, . W. & COOMBS,. P.(1983). Laboratory model systems for the investiga-tion of spatially and temporally organised microbialecosystems. Symposia of the Society fo r GeneralMicrobiology 34 , 67-1 17.WOOD, T. (1970). Spectrophotometric assay for Pribose-5-phosphate ketol-isomerase and for D-ribu-lose-5-phosphate 3-epimerase. Analytical Biochemis-try 33, 297-306.

323-327.

309-3 16.