- Email: [email protected]
Effects of the loss of triose phosphate isomerase activity on carbon metabolism in Kluyveromyces lactis
Research in Microbiology 153 (2002) 593–598 www.elsevier.com/locate/resmic
Effects of the loss of triose phosphate isomerase activity on carbon metabolism in Kluyveromyces lactis Daniele Capitanio, Annamaria Merico, Bianca Maria Ranzi, Concetta Compagno ∗ Dipartimento di Fisiologia e Biochimica Generali, sez. Biochimica Comparata, Università degli Studi di Milano, via Celoria 26, 20133 Milano, Italy Received 18 April 2002; accepted 12 July 2002 First published online 15 July 2002
Abstract The effect of the loss of triose phosphate isomerase activity on carbon metabolism in Kluyveromyces lactis was studied in batch and in continuous cultures. The Kltpi1 mutant was able to grow on media containing glucose as the sole carbon source both in batch and in continuous culture, unlike the corresponding S. cerevisiae mutant. In K. lactis tpi1 mutant no glycerol production was detected in chemostat cultivations. DHAP accumulation triggers glycerol production only when glucose is the sole carbon source in excess. The analysis of the activities of some key enzymes of carbon metabolism shows that in chemostat cultivations on mixed-substrates the activities of enzymes involved in ethanol assimilation are higher both in K. lactis wild type and mutant strains than in S. cerevisiae. 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved. Keywords: Kluyveromyces lactis; Triose phosphate isomerase; Carbon metabolism
1. Introduction The carbon metabolism of K. lactis has been mainly studied on mutants which displayed the Rag− phenotype. Rag− mutants are unable to grow on glucose when respiration is blocked by antimycin A and are usually affected in genes involved in the utilisation of glucose, either in the uptake, in glycolysis or in the fermentative pathway [1,2,7,13]. Triose phosphate isomerase plays at an important branch point in carbon metabolism that is among glycolysis, gluconeogenesis, pentose phosphate and glycerol pathways. The K. lactis tpi1 null mutant failed to grow on glucose in presence of antimycin A, showing a Rag− phenotype [3]. In S. cerevisiae the absence of triose phosphate isomerase causes a shift in the final product of aerobic glucose catabolism, from ethanol to glycerol [4], by the reduction of the accumulated dihydroxyacetone phosphate (DHAP) to glycerol-3-phosphate in the reaction catalysed by cytosolic NADH-dependent glycerol-3-P-dehydrogenase. In a tpi1 mutant glycerol pro* Correspondence and reprints.
E-mail address: [email protected] (C. Compagno).
duction can then represent a good indicator of the fraction of the carbon source that channels through glycolysis. The S. cerevisiae tpi1 mutant is unable to grow on glucose as the sole carbon source, even in conditions triggering respirative metabolism. This can reflect a NADH/energy shortage [4] caused by the competition for the NADH between cytosolic NADH-dependent-glycerol-3-P dehydrogenase, mitochondrial external NADH dehydrogenase and glycerol-3-P shuttle [10,11] and this, in turn, can promote the production of methylglyoxal, a cytotoxic compound originated from the DHAP not converted in glycerol [5]. The aim of this work is to investigate in detail the effects of the loss of triose phosphate isomerase activity on carbon metabolism in K. lactis. To obtain an accurate insight, we have focused our studies on carbon metabolism in a K. lactis tpi1 mutant and in its isogenic wild type strain both in batch and in carbon source-limited continuous cultures, always under monitored aerobic conditions. Most studies on regulation of carbon metabolism in K. lactis have been in fact performed in shake flask cultures. In this system the environmental conditions change continuously during cell growth. Additionally, the poor oxygen transfer easily triggers a partial anaerobic condition that deeply influences cel-
0923-2508/02/$ – see front matter 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved. PII: S 0 9 2 3 - 2 5 0 8 ( 0 2 ) 0 1 3 6 5 - 7
594
D. Capitanio et al. / Research in Microbiology 153 (2002) 593–598
lular metabolism. This cultivation system is then unsuitable for a detailed characterization of growth and product formation in aerobiosis. Furthermore, the activities of some key enzymes involved in carbon metabolism (pentose phosphate pathway and gluconeogenesis) have been determined for the first time in these growth conditions.
2. Materials and methods 2.1. Yeast strains The Kluyveromyces lactis strains used in this work are the haploid PM6-7A (MATa, uraA1-1, adeT-600,) [7] and its derivative Kltpi1 strain (KlTPI1::KanMX4) [3]. 2.2. Batch and chemostat cultivations Batch cultivations were performed in a Biostat-Q-system (B-Braun). A constant working volume of 0.8 l was maintained. An airflow of 0.8 l/min and a stirrer speed of 800 rpm maintained a dissolved oxygen concentration above 50% of air saturation. The culture pH was maintained at 5.0 by automatic addition of KOH 2 M. Preinoculi were performed on YEPDA medium (adenine 100 µg/ml). The cell biomass was washed and used to inoculate batch cultures on defined mineral medium prepared as reported [9] supplemented with adenine 350 µg/ml and uracil 200 µg/ml and containing the appropriate carbon source (glucose 1% (w/v) and glucose 1% (w/v)/ethanol 0.5% (v/v)). Chemostat cultivations were performed as described for batch cultures. When glucose was exhausted the continuous culture was started. “Steady state” is defined as the situation in which at least five volume changes had passed since the continuous culture started and biomass concentration, oxygen consumption, glucose and ethanol concentration in the medium had kept constant over 2 volume changes. All steady states were repeated after further five volume changes. Mixtures of ethanol and glucose at different ethanol fractions were used as carbon source. The concentration of glucose was fixed at 0.25 C-mol (carbon mol/l). Increased amounts of ethanol were added (from 0.026 C-mol/l to 0.156 C-mol/l) in order to obtain different ethanol fractions (calculated as ethanol C-mol/(glucose C-mol + ethanol C-mol)). For glucose pulse experiments a concentrated, sterile solution of glucose was aseptically added to a steady state culture (at ethanol fraction 0.1) to give an initial glucose concentration of about 50 mM. At appropriate intervals, samples were withdrawn from the culture for the analysis. During the pulse medium feed and effluent removal were continued. The concentrations of carbon dioxide and oxygen in exhausted gas were determined with a BM2001 gas analyser (Bioindustrie Mantovane, Italy). Calculations of specific oxygen consumption and carbon dioxide production were performed as described [12].
2.3. Analysis of intracellular and extracellular metabolites and determination of dry weight Samples were quickly withdrawn from chemostat cultures at steady state. Dihydroxyacetone phosphate was extracted by perchloric acid 1 M, neutralised by KHCO3 2.5 M. For the dihydroxyacetone phosphate assay the reaction mixture contained triethanolamine buffer 0.13 M, NADH 0.25 mM, glycerol-3-P dehydrogenase (GDH) 0.56 U/ml. The concentrations of glucose, glycerol, ethanol, acetate and D-lactate in supernatants were determined with R-biopharm (Roche) enzymatic kits. Washed culture samples were filtered with a 0.45 µm glass microfibre GF/A (Millipore) and dried for 36 h at 85 ◦ C. Parallel samples varied less than 1%. 2.4. Preparation of cell extracts and enzyme assays Cell extracts were prepared essentially as described [12], with the exception that cells were disrupted by agitation with glass beads on vortex (alternating 1 minute on ice and 1 minute on vortex for five times) rather than by sonication. Protein content of cell extracts was determined with Bio-Rad kit # 500-002, using bovine serum albumin as a standard. Enzyme assays were performed as described [3,6].
3. Results 3.1. Batch cultures on glucose and glucose/ethanol media Batch cultivations of the Kltpi1 mutant and its isogenic wild type were performed on media containing glucose or a mixture of glucose and ethanol in fermenter under monitored aerobic conditions (see Materials and methods). During the exponential phase of growth on glucose the mutant strain displayed a lower specific growth rate than the wild type (0.07 h−1 and 0.087 h−1 respectively), moreover an early arrest of the growth occurred in the mutant’s cultures. The analysis of the substrate consumption showed that it was extremely reduced in the Kltpi1 mutant (Fig. 1 A–B). In contrast to the wild type, the Kltpi1 mutant accumulated DHAP after 24 hours of growth (1.9 µg g−1 dry weight in the mutant, not detectable in the wild type) and, as a consequence, glycerol was produced, with a molar yield of 20% of the utilized carbon source (Fig. 1 B). Cultivations on a mixture of glucose and ethanol showed interesting features about carbon source utilization in K. lactis. In the wild type strain the specific glucose consumption rate increased when ethanol concentration in the medium decreased (Table 1). This suggests a preferential use of ethanol as carbon source as well as an inhibition of the ethanol on glucose utilization. On mixed-substrates the mutant strain was able to grow very similarly to the wild type and to utilise both the carbon sources (Fig. 2 A–B) (it is to remember that the mutant strain cannot utilize ethanol as the sole carbon
D. Capitanio et al. / Research in Microbiology 153 (2002) 593–598
595
Fig. 1. Growth curves (2), glucose consumption (") and glycerol production (F) of K. lactis wild type (A) and Kltpi1 mutant (B) in batch cultures on glucose. Cells were grown on synthetic medium supplemented with adenine (350 µg/ml), uracil (200 µg/ml) and glucose 1% (w/v) at pH 5.0. The values reported are the means of three determinations with less than 1% variation.
Fig. 2. Growth curves, substrate consumption and product formation of the K. lactis wild type strain (A) and Kltpi1 mutant (B) in batch cultures on mixed substrates (glucose and ethanol). Cells were grown on synthetic medium supplemented with adenine (350 µg/ml), uracil (200 µg/ml) and a mixture of glucose 1% (w/v) and ethanol 0,5% (v/v) at pH 5.0. Biomass (2); glucose ("); ethanol (P); glycerol (F). The values reported are the means of three determinations with less than 1% variation.
source). In the first 23 hours the Kltpi1 mutant strain showed higher glucose consumption rate than the wild type, which reduced when ethanol in the medium decreased. This indicates a positive effect of ethanol on the growth of the mutant strain. When ethanol was completely exhausted, the intracellular DHAP accumulation in the mutant strain reached a concentration that triggered glycerol production (Fig. 2B and Table 1).
Table 1 Specific substrate consumption rates determination for the wild type and Kltpi1 mutant strain during cultivation in batch (in brackets are reported the growth times at which calculations were made)
3.2. Aerobic carbon-limited continuous cultures
DHAP values were determined during exponential growth. n.d.: not detectable
Aerobic glucose-limited chemostat cultures of the Kltpi1 mutant were performed at a dilution rate of 0.05 hr−1 at first. In this condition a decrease of biomass and an increase of residual glucose consistent with a washout kinetic were obtained, although the dilution rate value was below the maximal specific growth rate found in the exponential
Strain
Carbon source
qglu (mmol g−1 hr−1 )
DHAP (µg g−1 )
wild type
glucose/ethanol
tpi1
glucose/ethanol
0.30 (23 h) 0.83 (30 h) 0.86 (23 h) 0.54 (30 h)
n.d. n.d. 2.013 2.432
growth phase of batch cultures. The steady state was reached only by decreasing the dilution rate at a very low value (0.02 hr−1 ); in this condition no residual glucose was detected in the culture supernatants (Table 2).
596
D. Capitanio et al. / Research in Microbiology 153 (2002) 593–598
Table 2 Steady state amounts of main products in the wild type and tpi1 mutant strains during cultivation in chemostat culture on mixtures of ethanol and glucose at different ethanol fractions Ethanol fraction D (hr−1 ) Initial glucose (mM) Initial ethanol (mM) Final glucose (mM) Final ethanol (mM) Dry weight (g/l) Yield (g/C mole) Glycerol (mM) Acetate (mM) D-lactate (mM) DHAP (µg/g )
0
Kltpi1 0.1
0.4
0
0.02 39.6 0 0 0 2.15 9.07 0 0 0 0.956
0.05 42.57 12.81 0 0 2.74 9.73 0 0 0 –
0.05 42.74 77.81 0 0 3.8 9.25 0 0 0 0.659
0.02 39.2 0 0 0 2.19 9.24 0 0 0 n.d.
K. lactis wild type 0.1 0.4 0.05 42.5 21.81 0 0.39 3.3 11.74 0 0 0 –
0.05 42.74 77.71 0 0 4.09 9.94 0 0 0 –
1
S. cerevisiae tpi1* 0.1 0.38
0.05 0 204 0 0 4.5 10.7 0 0 0 n.d.
0.045 42.16 14.54 11.65 0 1.47 10.43 23.78 0.7 0.14 1.53
0.052 41.44 77.33 0 0 4.47 12.83 18.24 0 0.12 0.74
S. cerevisiae wild type* 0.1 0.35 0.05 42.30 13.46 0 0 4.25 15.1 0 0 0 0.08
0.047 45.39 73.89 0 0 6.23 14.8 0 0 0 0.09
n.d.: not detectable; – : not detected; *data from another source [4].
Mixtures of glucose and ethanol at different ethanol fractions (see Materials and methods) were used to perform aerobic carbon-limited chemostat cultures of the wild type and Kltpi1 mutant strains at a higher dilution rate (0.05 hr−1 ). Data collected when steady states were attained (Table 2) indicated that the most remarkable difference between the two strains was the concentration of intracellular DHAP, which accumulated only in the Kltpi1 cells, even though at a lower level than in batch cultures. In contrast to what we observed in continuous cultures of S. cerevisiae tpi1 mutant [4] (Table 2), glycerol was never detected in the culture supernatants of the Kltpi1 mutant, as well as other extracellular metabolites, like acetate or D-lactate, which are implicated in the carbon metabolism of a tpi1 mutant [4]. Also at the lowest ethanol fraction (0.1) no residual glucose was detected in Kltpi1 cultures, in contrast S. cerevisiae tpi1 cultures at the same conditions (Table 2). 3.3. Exposure of carbon-limited continuous cultures to glucose excess Glucose (50 mM) was pulsed into an aerobic chemostat culture of the Kltpi1 mutant growing at a dilution rate of 0.05 hr−1 and at an ethanol fraction of 0.1. A reduced glucose consumption was detected in the cultures of the Kltpi1 mutant, in comparison to the wild type (Fig. 3). In contrast to the wild type strain, in which glucose was completely converted into biomass and CO2 , the exposure of the mutant to glucose excess caused production of glycerol, with a molar yield of 26%. Small amounts of D-lactate (18 mg/l) were detected in the supernatants of the Kltpi1 mutant after 7 hours. This could indicate a higher carbon flow through the methylglyoxal pathway, to drain the methylglyoxal derived from DHAP and not converted into glycerol. 3.4. Activities of key enzymes of carbon metabolism The activities of some enzymes involved in carbon assimilation were determined in cell extracts from batch
Fig. 3. Physiological effects of a glucose pulse on cultures of K. lactis wild type and Kltpi1 mutant growing in chemostat (dilution rate D = 0.05 hr−1 ). Glucose was pulsed at time T = 0. Glucose consumption in the wild type (") and in the Kltpi1 mutant (!); glycerol production in the Kltpi1 mutant (F). The values reported are the means of two determinations with less than 1% variation.
and chemostat cultures of both K. lactis wild type and Kltpi1 mutant strains (Table 3). The same level of activity of glucose-6-phosphate dehydrogenase (G6PDH) was detected in extracts from wild type and Kltpi1 mutant cells growing in batch cultures. In the wild type strain the activities of this enzyme detected in continuous cultures were always lower than in batch cultures. In chemostat cultures the Kltpi1 mutant strain exhibited higher activities than the wild type strain in all the examined conditions. In mixed-substrate chemostat cultures the activity of fructose-1,6-bisphosphatase (FBP) in cell extracts from the K. lactis wild type was detected at the same level than in the Kltpi1 mutant strain. Phosphoenolpyruvate carboxykinase (PEPCK), the other key enzyme for ethanol assimilation, was five fold higher in cell-free extracts of Kltpi1 mutant than in the wild type at
D. Capitanio et al. / Research in Microbiology 153 (2002) 593–598
597
Table 3 Activities of key enzymes in carbon metabolism detected in cell extracts from wild type and tpi1 mutant cells growing in continuous cultures at the steady states and at the exponential phase of growth in batch cultures Ethanol fraction
Continuous cultures 0
G6PDH (U/mg) FBP (U/mg) PEPCK (U/mg) ICL (U/mg)
1.6 n.d. n.d. n.d.
wild type 0.4 0.7 0.048 0.062 0.027
Batch Kltpi1
1
0
0.1
0.2
0.4
0.7 0.042 0.111 0.023
4 n.d. n.d. n.d.
1.37 n.d. n.d. n.d.
1.02 0.024 0.042 n.d.
1.25 0.05 0.31 0.06
wild type 0 0.4 2.3 – – –
1.36 0.038 0.048 0.02
Kltpi1 0
0.4
2.1 – – –
1.4 – 0.06 –
n.d.: not detectable; – : not detected.
high ethanol fraction (0,4). At a lesser extent (2 fold), the same effect was observed for the activity of isocitrate lyase (ICL).
4. Discussion The lack of triose phosphate isomerase causes an accumulation of DHAP and its conversion to glycerol requires cytosolic NADH in the reaction catalysed by glycerol-3phosphate dehydrogenase. This reaction can then compete with the oxidation of cytosolic NADH by mitochondria and this can lead to the accumulation of DHAP, which in turn can be converted to methylglyoxal, a toxic compound. In contrast to the S. cerevisiae tpi1 mutant, which is unable to grow on glucose, Kltpi1 mutant is able to grow on glucose as the sole carbon source. In the Kltpi1 mutant the loss of triose phosphate isomerase activity exerts a negative influence on glucose utilization. As previously observed in the S. cerevisiae tpi1 mutant [4], also in K. lactis this detrimental effect could be caused by DHAP accumulation, both through methylglyoxal formation and through a NADH/energy shortage. As in S. cerevisiae mutant, also in Kltpi1 mutant this negative effect is alleviated by the addition of a carbon source like ethanol that readily supplies the cells with NADH and energy. In the K. lactis tpi1 mutant, glycerol production was detected when glucose was the sole carbon source in excess, as occurs in batch cultures on glucose or after ethanol exhaustion on mixed-substrates. In K. lactis these conditions can promote a high glycolytic flux and DHAP accumulates at high level, triggering glycerol formation. The analysis of specific substrate consumption rates in the batch cultures of the wild type and Kltpi1 strains on mixed-substrates showed that a simultaneous utilization of glucose and ethanol occurred in both strains and indicated the positive effect that ethanol produce on growth and on glucose utilization in the Kltpi1 mutant, in contrast to the wild type in which ethanol seems to inhibit glucose utilization. Interesting differences in carbon source utilization between K. lactis and S. cerevisiae emerged also through the analysis in continuous cultures. The most relevant one is the absence of glycerol production by the K. lactis tpi1 mutant in chemostat cultivations, both on glucose and on mixed
glucose/ethanol media. This is in contrast to what we observed in S. cerevisiae tpi1 mutant, which produces glycerol also at the highest ethanol fraction, indicating that the flux through the glycolytic pathway is always higher in S. cerevisiae than in K. lactis, also in conditions in which glucose was in limiting amount, as occurs in carbon-source limited continuous cultures, and in presence of ethanol in the growth medium. Glycerol production in continuous cultures was detected in K. lactis tpi1 mutant only when glucose was added as the sole carbon source in excess, a condition very similar to a batch culture and then promoting a high glycolytic flux. in contrast to what occurred in the S. cerevisiae mutant, the complete glucose consumption by the Kltpi1 mutant at the lowest ethanol fraction, is another indication that glucose metabolism is not impaired in this condition and then it suggests a less severe effect of tpi1 mutation in K. lactis than in S. cerevisiae. Further information about the different regulation of the carbon flux can be obtained by the analysis of the activities of some key enzymes involved in carbon metabolism. High activity of glucose-6-phosphate dehydrogenase in Crabtree-negative yeasts such as K. lactis and Candida utilis has been already reported [13] and in fact the pentose phosphate pathway in K. lactis is more active if compared to S. cerevisiae [7,8]. Interestingly, the Kltpi1 mutant showed a 2.5 fold increase of this activity in respect to the wild type strain, after prolonged growth on glucose media in chemostat cultures. In chemostat cultivations on mixedsubstrates we found that the activities of the enzymes involved in ethanol assimilation both in K. lactis wild type and in the Kltpi1 mutant strain are higher than in S. cerevisiae at the same conditions of growth [6]. These data suggest a major role of gluconeogenesis in K. lactis on mixed-substrates cultivations than in S. cerevisiae and this could be particularly important for the growth of the Kltpi1 mutant, in which we found a 5 fold higher activity of phosphoenolpyruvate carboxykinase than in the wild type. Taken together, all these observations on carbon source metabolism in the Kltpi1 mutant indicate that in this yeast a reduced glycolytic flux, due also to a major role of pentose phosphate pathway and of gluconeogenesis, can partially prevent the accumulation of high level of DHAP, reducing methylglyoxal formation and NADH shortage and, in this way, alleviating the negative effect of the tpi1 mutation. This can account for its ability to grow on glucose both in
598
D. Capitanio et al. / Research in Microbiology 153 (2002) 593–598
batch or in continuous culture at a low dilution rate as well as its inability to produce glycerol when glucose is not in excess.
Acknowledgement This work was partially supported by CNR PF “Biotecnologie” Sottoprogetto 4 to B.M.R. and by MURST-Università degli Studi di Milano Cofin 2001 to B.M.R.
References [1] M.M. Bianchi, L. Tizzani, M. Destruelle, L. Frontali, M. WesolowskiLouvel, The “petite negative” yeast Kluyveromyces lactis has a single gene expressing pyruvate decarboxylase activity, Mol. Microbiol. 19 (1996) 27–36. [2] J. Blaisonneau, H. Fukuhara, M. Wesolowski-Louvel, The Kluyveromyces lactis equivalent of casein kinase I is required for the transcription of the gene encoding the low-affinity glucose permease, Mol. Gen. Genet. 253 (1997) 469–477. [3] C. Compagno, F. Boschi, A. Daleffe, D. Porro, B.M. Ranzi, Isolation, nucleotide sequence and physiological relevance of the gene encoding triose phosphate isomerase from Kluyveromyces lactis, Appl. Env. Microbiol. 65 (1999) 4216–4219. [4] C. Compagno, L. Brambilla, D. Capitanio, F. Boschi, B.M. Ranzi, D. Porro, Alterations of the glucose metabolism in a triose phosphate isomerase-negative Saccharomyces cerevisiae mutant, Yeast 18 (2001) 663–670.
[5] R.A. Cooper, Metabolism of methylglyoxal in microorganisms, Ann. Rev. Microbiol. 38 (1984) 49–68. [6] P. de Jong-Gubbels, P. Vanrolleghem, J.J. Heijnen, J.P. van Dijken, J.T. Pronk, Regulation of carbon metabolism in chemostat cultures of Saccharomyces cerevisiae grown on mixtures of glucose and ethanol, Yeast 11 (1995) 407–418. [7] P. Goffrini, M. Wesolowski-Louvel, I. Ferrero I, A phoshoglucose isomerase gene is involved in the Rag phenotype of the yeast Kluyveromyces lactis, Mol. Gen. Genet. 228 (1991) 401–409. [8] J. Jacoby, C.P. Hollenberg, J.J. Heinisch, Transaldolase mutants in the yeast Kluyveromyces lactis provide evidence that glucose can be metabolized through the pentose phosphate pathway, Mol. Microbiol. 10 (1993) 867–876. [9] J. Kiers, A.-M. Zeeman, M. Luttik, C. Thiele, J.I. Castrillo, H.Y. Steensma, J.P. van Dijken, J.T. Pronk, Regulation of alcoholic fermentation in batch and chemostat cultures of Kluyveromyces lactis CBS 2359, Yeast 14 (1993) 459–469. [10] C. Larsson, I.L. Pahlman, R. Ansell, M. Rigoulet, L. Adler, L. Gustafson, The importance of the glycerol 3-phosphate shuttle during aerobic growth of Saccharomyces cerevisiae, Yeast 14 (1998) 347– 357. [11] K.M. Overkamp, B.M. Bakker, P. Kotter, A. van Tuijl, S. de Vries, J.P. van Dijken, J.T. Pronk, In vivo analysis of the mechanisms of oxidation of cytosolic NADH by Saccharomyces cerevisiae mitochondria, J. Bacteriol. 182 (2000) 2823–2830. [12] H. van Urk, P.R. Mak, W.A. Scheffers, J.P. van Dijken, Metabolic response of Saccharomyces cerevisiae CBS8066 and Candida utilis CBS 621 upon transition from glucose limitation to glucose excess, Yeast 4 (1988) 283–291. [13] M. Wesolowski-Louvel, C. Prior, D. Bornecque, H. Fukuhara, Ragmutations involved in glucose metabolism in yeast: Isolation and genetic characterization, Yeast 8 (1992) 711–719.
Effects of the loss of triose phosphate isomerase activity on carbon metabolism in Kluyveromyces lactis Daniele Capitanio, Annamaria Merico, Bianca Maria Ranzi, Concetta Compagno ∗ Dipartimento di Fisiologia e Biochimica Generali, sez. Biochimica Comparata, Università degli Studi di Milano, via Celoria 26, 20133 Milano, Italy Received 18 April 2002; accepted 12 July 2002 First published online 15 July 2002
Abstract The effect of the loss of triose phosphate isomerase activity on carbon metabolism in Kluyveromyces lactis was studied in batch and in continuous cultures. The Kltpi1 mutant was able to grow on media containing glucose as the sole carbon source both in batch and in continuous culture, unlike the corresponding S. cerevisiae mutant. In K. lactis tpi1 mutant no glycerol production was detected in chemostat cultivations. DHAP accumulation triggers glycerol production only when glucose is the sole carbon source in excess. The analysis of the activities of some key enzymes of carbon metabolism shows that in chemostat cultivations on mixed-substrates the activities of enzymes involved in ethanol assimilation are higher both in K. lactis wild type and mutant strains than in S. cerevisiae. 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved. Keywords: Kluyveromyces lactis; Triose phosphate isomerase; Carbon metabolism
1. Introduction The carbon metabolism of K. lactis has been mainly studied on mutants which displayed the Rag− phenotype. Rag− mutants are unable to grow on glucose when respiration is blocked by antimycin A and are usually affected in genes involved in the utilisation of glucose, either in the uptake, in glycolysis or in the fermentative pathway [1,2,7,13]. Triose phosphate isomerase plays at an important branch point in carbon metabolism that is among glycolysis, gluconeogenesis, pentose phosphate and glycerol pathways. The K. lactis tpi1 null mutant failed to grow on glucose in presence of antimycin A, showing a Rag− phenotype [3]. In S. cerevisiae the absence of triose phosphate isomerase causes a shift in the final product of aerobic glucose catabolism, from ethanol to glycerol [4], by the reduction of the accumulated dihydroxyacetone phosphate (DHAP) to glycerol-3-phosphate in the reaction catalysed by cytosolic NADH-dependent glycerol-3-P-dehydrogenase. In a tpi1 mutant glycerol pro* Correspondence and reprints.
E-mail address: [email protected] (C. Compagno).
duction can then represent a good indicator of the fraction of the carbon source that channels through glycolysis. The S. cerevisiae tpi1 mutant is unable to grow on glucose as the sole carbon source, even in conditions triggering respirative metabolism. This can reflect a NADH/energy shortage [4] caused by the competition for the NADH between cytosolic NADH-dependent-glycerol-3-P dehydrogenase, mitochondrial external NADH dehydrogenase and glycerol-3-P shuttle [10,11] and this, in turn, can promote the production of methylglyoxal, a cytotoxic compound originated from the DHAP not converted in glycerol [5]. The aim of this work is to investigate in detail the effects of the loss of triose phosphate isomerase activity on carbon metabolism in K. lactis. To obtain an accurate insight, we have focused our studies on carbon metabolism in a K. lactis tpi1 mutant and in its isogenic wild type strain both in batch and in carbon source-limited continuous cultures, always under monitored aerobic conditions. Most studies on regulation of carbon metabolism in K. lactis have been in fact performed in shake flask cultures. In this system the environmental conditions change continuously during cell growth. Additionally, the poor oxygen transfer easily triggers a partial anaerobic condition that deeply influences cel-
0923-2508/02/$ – see front matter 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved. PII: S 0 9 2 3 - 2 5 0 8 ( 0 2 ) 0 1 3 6 5 - 7
594
D. Capitanio et al. / Research in Microbiology 153 (2002) 593–598
lular metabolism. This cultivation system is then unsuitable for a detailed characterization of growth and product formation in aerobiosis. Furthermore, the activities of some key enzymes involved in carbon metabolism (pentose phosphate pathway and gluconeogenesis) have been determined for the first time in these growth conditions.
2. Materials and methods 2.1. Yeast strains The Kluyveromyces lactis strains used in this work are the haploid PM6-7A (MATa, uraA1-1, adeT-600,) [7] and its derivative Kltpi1 strain (KlTPI1::KanMX4) [3]. 2.2. Batch and chemostat cultivations Batch cultivations were performed in a Biostat-Q-system (B-Braun). A constant working volume of 0.8 l was maintained. An airflow of 0.8 l/min and a stirrer speed of 800 rpm maintained a dissolved oxygen concentration above 50% of air saturation. The culture pH was maintained at 5.0 by automatic addition of KOH 2 M. Preinoculi were performed on YEPDA medium (adenine 100 µg/ml). The cell biomass was washed and used to inoculate batch cultures on defined mineral medium prepared as reported [9] supplemented with adenine 350 µg/ml and uracil 200 µg/ml and containing the appropriate carbon source (glucose 1% (w/v) and glucose 1% (w/v)/ethanol 0.5% (v/v)). Chemostat cultivations were performed as described for batch cultures. When glucose was exhausted the continuous culture was started. “Steady state” is defined as the situation in which at least five volume changes had passed since the continuous culture started and biomass concentration, oxygen consumption, glucose and ethanol concentration in the medium had kept constant over 2 volume changes. All steady states were repeated after further five volume changes. Mixtures of ethanol and glucose at different ethanol fractions were used as carbon source. The concentration of glucose was fixed at 0.25 C-mol (carbon mol/l). Increased amounts of ethanol were added (from 0.026 C-mol/l to 0.156 C-mol/l) in order to obtain different ethanol fractions (calculated as ethanol C-mol/(glucose C-mol + ethanol C-mol)). For glucose pulse experiments a concentrated, sterile solution of glucose was aseptically added to a steady state culture (at ethanol fraction 0.1) to give an initial glucose concentration of about 50 mM. At appropriate intervals, samples were withdrawn from the culture for the analysis. During the pulse medium feed and effluent removal were continued. The concentrations of carbon dioxide and oxygen in exhausted gas were determined with a BM2001 gas analyser (Bioindustrie Mantovane, Italy). Calculations of specific oxygen consumption and carbon dioxide production were performed as described [12].
2.3. Analysis of intracellular and extracellular metabolites and determination of dry weight Samples were quickly withdrawn from chemostat cultures at steady state. Dihydroxyacetone phosphate was extracted by perchloric acid 1 M, neutralised by KHCO3 2.5 M. For the dihydroxyacetone phosphate assay the reaction mixture contained triethanolamine buffer 0.13 M, NADH 0.25 mM, glycerol-3-P dehydrogenase (GDH) 0.56 U/ml. The concentrations of glucose, glycerol, ethanol, acetate and D-lactate in supernatants were determined with R-biopharm (Roche) enzymatic kits. Washed culture samples were filtered with a 0.45 µm glass microfibre GF/A (Millipore) and dried for 36 h at 85 ◦ C. Parallel samples varied less than 1%. 2.4. Preparation of cell extracts and enzyme assays Cell extracts were prepared essentially as described [12], with the exception that cells were disrupted by agitation with glass beads on vortex (alternating 1 minute on ice and 1 minute on vortex for five times) rather than by sonication. Protein content of cell extracts was determined with Bio-Rad kit # 500-002, using bovine serum albumin as a standard. Enzyme assays were performed as described [3,6].
3. Results 3.1. Batch cultures on glucose and glucose/ethanol media Batch cultivations of the Kltpi1 mutant and its isogenic wild type were performed on media containing glucose or a mixture of glucose and ethanol in fermenter under monitored aerobic conditions (see Materials and methods). During the exponential phase of growth on glucose the mutant strain displayed a lower specific growth rate than the wild type (0.07 h−1 and 0.087 h−1 respectively), moreover an early arrest of the growth occurred in the mutant’s cultures. The analysis of the substrate consumption showed that it was extremely reduced in the Kltpi1 mutant (Fig. 1 A–B). In contrast to the wild type, the Kltpi1 mutant accumulated DHAP after 24 hours of growth (1.9 µg g−1 dry weight in the mutant, not detectable in the wild type) and, as a consequence, glycerol was produced, with a molar yield of 20% of the utilized carbon source (Fig. 1 B). Cultivations on a mixture of glucose and ethanol showed interesting features about carbon source utilization in K. lactis. In the wild type strain the specific glucose consumption rate increased when ethanol concentration in the medium decreased (Table 1). This suggests a preferential use of ethanol as carbon source as well as an inhibition of the ethanol on glucose utilization. On mixed-substrates the mutant strain was able to grow very similarly to the wild type and to utilise both the carbon sources (Fig. 2 A–B) (it is to remember that the mutant strain cannot utilize ethanol as the sole carbon
D. Capitanio et al. / Research in Microbiology 153 (2002) 593–598
595
Fig. 1. Growth curves (2), glucose consumption (") and glycerol production (F) of K. lactis wild type (A) and Kltpi1 mutant (B) in batch cultures on glucose. Cells were grown on synthetic medium supplemented with adenine (350 µg/ml), uracil (200 µg/ml) and glucose 1% (w/v) at pH 5.0. The values reported are the means of three determinations with less than 1% variation.
Fig. 2. Growth curves, substrate consumption and product formation of the K. lactis wild type strain (A) and Kltpi1 mutant (B) in batch cultures on mixed substrates (glucose and ethanol). Cells were grown on synthetic medium supplemented with adenine (350 µg/ml), uracil (200 µg/ml) and a mixture of glucose 1% (w/v) and ethanol 0,5% (v/v) at pH 5.0. Biomass (2); glucose ("); ethanol (P); glycerol (F). The values reported are the means of three determinations with less than 1% variation.
source). In the first 23 hours the Kltpi1 mutant strain showed higher glucose consumption rate than the wild type, which reduced when ethanol in the medium decreased. This indicates a positive effect of ethanol on the growth of the mutant strain. When ethanol was completely exhausted, the intracellular DHAP accumulation in the mutant strain reached a concentration that triggered glycerol production (Fig. 2B and Table 1).
Table 1 Specific substrate consumption rates determination for the wild type and Kltpi1 mutant strain during cultivation in batch (in brackets are reported the growth times at which calculations were made)
3.2. Aerobic carbon-limited continuous cultures
DHAP values were determined during exponential growth. n.d.: not detectable
Aerobic glucose-limited chemostat cultures of the Kltpi1 mutant were performed at a dilution rate of 0.05 hr−1 at first. In this condition a decrease of biomass and an increase of residual glucose consistent with a washout kinetic were obtained, although the dilution rate value was below the maximal specific growth rate found in the exponential
Strain
Carbon source
qglu (mmol g−1 hr−1 )
DHAP (µg g−1 )
wild type
glucose/ethanol
tpi1
glucose/ethanol
0.30 (23 h) 0.83 (30 h) 0.86 (23 h) 0.54 (30 h)
n.d. n.d. 2.013 2.432
growth phase of batch cultures. The steady state was reached only by decreasing the dilution rate at a very low value (0.02 hr−1 ); in this condition no residual glucose was detected in the culture supernatants (Table 2).
596
D. Capitanio et al. / Research in Microbiology 153 (2002) 593–598
Table 2 Steady state amounts of main products in the wild type and tpi1 mutant strains during cultivation in chemostat culture on mixtures of ethanol and glucose at different ethanol fractions Ethanol fraction D (hr−1 ) Initial glucose (mM) Initial ethanol (mM) Final glucose (mM) Final ethanol (mM) Dry weight (g/l) Yield (g/C mole) Glycerol (mM) Acetate (mM) D-lactate (mM) DHAP (µg/g )
0
Kltpi1 0.1
0.4
0
0.02 39.6 0 0 0 2.15 9.07 0 0 0 0.956
0.05 42.57 12.81 0 0 2.74 9.73 0 0 0 –
0.05 42.74 77.81 0 0 3.8 9.25 0 0 0 0.659
0.02 39.2 0 0 0 2.19 9.24 0 0 0 n.d.
K. lactis wild type 0.1 0.4 0.05 42.5 21.81 0 0.39 3.3 11.74 0 0 0 –
0.05 42.74 77.71 0 0 4.09 9.94 0 0 0 –
1
S. cerevisiae tpi1* 0.1 0.38
0.05 0 204 0 0 4.5 10.7 0 0 0 n.d.
0.045 42.16 14.54 11.65 0 1.47 10.43 23.78 0.7 0.14 1.53
0.052 41.44 77.33 0 0 4.47 12.83 18.24 0 0.12 0.74
S. cerevisiae wild type* 0.1 0.35 0.05 42.30 13.46 0 0 4.25 15.1 0 0 0 0.08
0.047 45.39 73.89 0 0 6.23 14.8 0 0 0 0.09
n.d.: not detectable; – : not detected; *data from another source [4].
Mixtures of glucose and ethanol at different ethanol fractions (see Materials and methods) were used to perform aerobic carbon-limited chemostat cultures of the wild type and Kltpi1 mutant strains at a higher dilution rate (0.05 hr−1 ). Data collected when steady states were attained (Table 2) indicated that the most remarkable difference between the two strains was the concentration of intracellular DHAP, which accumulated only in the Kltpi1 cells, even though at a lower level than in batch cultures. In contrast to what we observed in continuous cultures of S. cerevisiae tpi1 mutant [4] (Table 2), glycerol was never detected in the culture supernatants of the Kltpi1 mutant, as well as other extracellular metabolites, like acetate or D-lactate, which are implicated in the carbon metabolism of a tpi1 mutant [4]. Also at the lowest ethanol fraction (0.1) no residual glucose was detected in Kltpi1 cultures, in contrast S. cerevisiae tpi1 cultures at the same conditions (Table 2). 3.3. Exposure of carbon-limited continuous cultures to glucose excess Glucose (50 mM) was pulsed into an aerobic chemostat culture of the Kltpi1 mutant growing at a dilution rate of 0.05 hr−1 and at an ethanol fraction of 0.1. A reduced glucose consumption was detected in the cultures of the Kltpi1 mutant, in comparison to the wild type (Fig. 3). In contrast to the wild type strain, in which glucose was completely converted into biomass and CO2 , the exposure of the mutant to glucose excess caused production of glycerol, with a molar yield of 26%. Small amounts of D-lactate (18 mg/l) were detected in the supernatants of the Kltpi1 mutant after 7 hours. This could indicate a higher carbon flow through the methylglyoxal pathway, to drain the methylglyoxal derived from DHAP and not converted into glycerol. 3.4. Activities of key enzymes of carbon metabolism The activities of some enzymes involved in carbon assimilation were determined in cell extracts from batch
Fig. 3. Physiological effects of a glucose pulse on cultures of K. lactis wild type and Kltpi1 mutant growing in chemostat (dilution rate D = 0.05 hr−1 ). Glucose was pulsed at time T = 0. Glucose consumption in the wild type (") and in the Kltpi1 mutant (!); glycerol production in the Kltpi1 mutant (F). The values reported are the means of two determinations with less than 1% variation.
and chemostat cultures of both K. lactis wild type and Kltpi1 mutant strains (Table 3). The same level of activity of glucose-6-phosphate dehydrogenase (G6PDH) was detected in extracts from wild type and Kltpi1 mutant cells growing in batch cultures. In the wild type strain the activities of this enzyme detected in continuous cultures were always lower than in batch cultures. In chemostat cultures the Kltpi1 mutant strain exhibited higher activities than the wild type strain in all the examined conditions. In mixed-substrate chemostat cultures the activity of fructose-1,6-bisphosphatase (FBP) in cell extracts from the K. lactis wild type was detected at the same level than in the Kltpi1 mutant strain. Phosphoenolpyruvate carboxykinase (PEPCK), the other key enzyme for ethanol assimilation, was five fold higher in cell-free extracts of Kltpi1 mutant than in the wild type at
D. Capitanio et al. / Research in Microbiology 153 (2002) 593–598
597
Table 3 Activities of key enzymes in carbon metabolism detected in cell extracts from wild type and tpi1 mutant cells growing in continuous cultures at the steady states and at the exponential phase of growth in batch cultures Ethanol fraction
Continuous cultures 0
G6PDH (U/mg) FBP (U/mg) PEPCK (U/mg) ICL (U/mg)
1.6 n.d. n.d. n.d.
wild type 0.4 0.7 0.048 0.062 0.027
Batch Kltpi1
1
0
0.1
0.2
0.4
0.7 0.042 0.111 0.023
4 n.d. n.d. n.d.
1.37 n.d. n.d. n.d.
1.02 0.024 0.042 n.d.
1.25 0.05 0.31 0.06
wild type 0 0.4 2.3 – – –
1.36 0.038 0.048 0.02
Kltpi1 0
0.4
2.1 – – –
1.4 – 0.06 –
n.d.: not detectable; – : not detected.
high ethanol fraction (0,4). At a lesser extent (2 fold), the same effect was observed for the activity of isocitrate lyase (ICL).
4. Discussion The lack of triose phosphate isomerase causes an accumulation of DHAP and its conversion to glycerol requires cytosolic NADH in the reaction catalysed by glycerol-3phosphate dehydrogenase. This reaction can then compete with the oxidation of cytosolic NADH by mitochondria and this can lead to the accumulation of DHAP, which in turn can be converted to methylglyoxal, a toxic compound. In contrast to the S. cerevisiae tpi1 mutant, which is unable to grow on glucose, Kltpi1 mutant is able to grow on glucose as the sole carbon source. In the Kltpi1 mutant the loss of triose phosphate isomerase activity exerts a negative influence on glucose utilization. As previously observed in the S. cerevisiae tpi1 mutant [4], also in K. lactis this detrimental effect could be caused by DHAP accumulation, both through methylglyoxal formation and through a NADH/energy shortage. As in S. cerevisiae mutant, also in Kltpi1 mutant this negative effect is alleviated by the addition of a carbon source like ethanol that readily supplies the cells with NADH and energy. In the K. lactis tpi1 mutant, glycerol production was detected when glucose was the sole carbon source in excess, as occurs in batch cultures on glucose or after ethanol exhaustion on mixed-substrates. In K. lactis these conditions can promote a high glycolytic flux and DHAP accumulates at high level, triggering glycerol formation. The analysis of specific substrate consumption rates in the batch cultures of the wild type and Kltpi1 strains on mixed-substrates showed that a simultaneous utilization of glucose and ethanol occurred in both strains and indicated the positive effect that ethanol produce on growth and on glucose utilization in the Kltpi1 mutant, in contrast to the wild type in which ethanol seems to inhibit glucose utilization. Interesting differences in carbon source utilization between K. lactis and S. cerevisiae emerged also through the analysis in continuous cultures. The most relevant one is the absence of glycerol production by the K. lactis tpi1 mutant in chemostat cultivations, both on glucose and on mixed
glucose/ethanol media. This is in contrast to what we observed in S. cerevisiae tpi1 mutant, which produces glycerol also at the highest ethanol fraction, indicating that the flux through the glycolytic pathway is always higher in S. cerevisiae than in K. lactis, also in conditions in which glucose was in limiting amount, as occurs in carbon-source limited continuous cultures, and in presence of ethanol in the growth medium. Glycerol production in continuous cultures was detected in K. lactis tpi1 mutant only when glucose was added as the sole carbon source in excess, a condition very similar to a batch culture and then promoting a high glycolytic flux. in contrast to what occurred in the S. cerevisiae mutant, the complete glucose consumption by the Kltpi1 mutant at the lowest ethanol fraction, is another indication that glucose metabolism is not impaired in this condition and then it suggests a less severe effect of tpi1 mutation in K. lactis than in S. cerevisiae. Further information about the different regulation of the carbon flux can be obtained by the analysis of the activities of some key enzymes involved in carbon metabolism. High activity of glucose-6-phosphate dehydrogenase in Crabtree-negative yeasts such as K. lactis and Candida utilis has been already reported [13] and in fact the pentose phosphate pathway in K. lactis is more active if compared to S. cerevisiae [7,8]. Interestingly, the Kltpi1 mutant showed a 2.5 fold increase of this activity in respect to the wild type strain, after prolonged growth on glucose media in chemostat cultures. In chemostat cultivations on mixedsubstrates we found that the activities of the enzymes involved in ethanol assimilation both in K. lactis wild type and in the Kltpi1 mutant strain are higher than in S. cerevisiae at the same conditions of growth [6]. These data suggest a major role of gluconeogenesis in K. lactis on mixed-substrates cultivations than in S. cerevisiae and this could be particularly important for the growth of the Kltpi1 mutant, in which we found a 5 fold higher activity of phosphoenolpyruvate carboxykinase than in the wild type. Taken together, all these observations on carbon source metabolism in the Kltpi1 mutant indicate that in this yeast a reduced glycolytic flux, due also to a major role of pentose phosphate pathway and of gluconeogenesis, can partially prevent the accumulation of high level of DHAP, reducing methylglyoxal formation and NADH shortage and, in this way, alleviating the negative effect of the tpi1 mutation. This can account for its ability to grow on glucose both in
598
D. Capitanio et al. / Research in Microbiology 153 (2002) 593–598
batch or in continuous culture at a low dilution rate as well as its inability to produce glycerol when glucose is not in excess.
Acknowledgement This work was partially supported by CNR PF “Biotecnologie” Sottoprogetto 4 to B.M.R. and by MURST-Università degli Studi di Milano Cofin 2001 to B.M.R.
References [1] M.M. Bianchi, L. Tizzani, M. Destruelle, L. Frontali, M. WesolowskiLouvel, The “petite negative” yeast Kluyveromyces lactis has a single gene expressing pyruvate decarboxylase activity, Mol. Microbiol. 19 (1996) 27–36. [2] J. Blaisonneau, H. Fukuhara, M. Wesolowski-Louvel, The Kluyveromyces lactis equivalent of casein kinase I is required for the transcription of the gene encoding the low-affinity glucose permease, Mol. Gen. Genet. 253 (1997) 469–477. [3] C. Compagno, F. Boschi, A. Daleffe, D. Porro, B.M. Ranzi, Isolation, nucleotide sequence and physiological relevance of the gene encoding triose phosphate isomerase from Kluyveromyces lactis, Appl. Env. Microbiol. 65 (1999) 4216–4219. [4] C. Compagno, L. Brambilla, D. Capitanio, F. Boschi, B.M. Ranzi, D. Porro, Alterations of the glucose metabolism in a triose phosphate isomerase-negative Saccharomyces cerevisiae mutant, Yeast 18 (2001) 663–670.
[5] R.A. Cooper, Metabolism of methylglyoxal in microorganisms, Ann. Rev. Microbiol. 38 (1984) 49–68. [6] P. de Jong-Gubbels, P. Vanrolleghem, J.J. Heijnen, J.P. van Dijken, J.T. Pronk, Regulation of carbon metabolism in chemostat cultures of Saccharomyces cerevisiae grown on mixtures of glucose and ethanol, Yeast 11 (1995) 407–418. [7] P. Goffrini, M. Wesolowski-Louvel, I. Ferrero I, A phoshoglucose isomerase gene is involved in the Rag phenotype of the yeast Kluyveromyces lactis, Mol. Gen. Genet. 228 (1991) 401–409. [8] J. Jacoby, C.P. Hollenberg, J.J. Heinisch, Transaldolase mutants in the yeast Kluyveromyces lactis provide evidence that glucose can be metabolized through the pentose phosphate pathway, Mol. Microbiol. 10 (1993) 867–876. [9] J. Kiers, A.-M. Zeeman, M. Luttik, C. Thiele, J.I. Castrillo, H.Y. Steensma, J.P. van Dijken, J.T. Pronk, Regulation of alcoholic fermentation in batch and chemostat cultures of Kluyveromyces lactis CBS 2359, Yeast 14 (1993) 459–469. [10] C. Larsson, I.L. Pahlman, R. Ansell, M. Rigoulet, L. Adler, L. Gustafson, The importance of the glycerol 3-phosphate shuttle during aerobic growth of Saccharomyces cerevisiae, Yeast 14 (1998) 347– 357. [11] K.M. Overkamp, B.M. Bakker, P. Kotter, A. van Tuijl, S. de Vries, J.P. van Dijken, J.T. Pronk, In vivo analysis of the mechanisms of oxidation of cytosolic NADH by Saccharomyces cerevisiae mitochondria, J. Bacteriol. 182 (2000) 2823–2830. [12] H. van Urk, P.R. Mak, W.A. Scheffers, J.P. van Dijken, Metabolic response of Saccharomyces cerevisiae CBS8066 and Candida utilis CBS 621 upon transition from glucose limitation to glucose excess, Yeast 4 (1988) 283–291. [13] M. Wesolowski-Louvel, C. Prior, D. Bornecque, H. Fukuhara, Ragmutations involved in glucose metabolism in yeast: Isolation and genetic characterization, Yeast 8 (1992) 711–719.