Introduction

The heterolactic bacterium Oenococcus oeni (formerly Leuconostoc oenos) (Dicks et al. 1995) uses the phosphoketolase pathway for conversion of hexoses such as glucose to lactate, CO2 and acetylphosphate (Fig. 1). The excess reducing equivalents from the pathway (2 mol NAD(P)H/mol hexose) are then transferred to acetyl-CoA resulting in ethanol production. O. oeni and related bacteria, however, can use alternative pathways for NAD(P)H reoxidation. In addition, erythrose 4-phosphate, produced from fructose 6-phosphate by phosphoketolase, can be reduced yielding erythritol (Veiga-Da-Cunha et al. 1992; 1993). Further on, fructose is reduced to mannitol, and in a minor pathway, 3-phosphoglyceraldehyde is converted to glycerol. Use of these pathways for NAD(P)H reoxidation is accompanied by the excretion of acetate, since part of the acetylphosphate is no longer required as electron acceptor.

Fig. 1.
figure 1

Heterofermentative lactic acid fermentation of glucose and fructose by Oenococcus oeni: alternative pathways for NAD(P)H reoxidation (erythritol, acetate and mannitol fermentation). The central phosphoketolase pathway (formation of lactate and acetylphosphate or acetyl-SCoA from the hexoses) is printed in black (bold letters), the alternative pathways for NAD(P)H reoxidation in gray and italics

Full size image

During glucose and fructose fermentation, different NAD(P)H reoxidation reactions are preferred. In glucose fermentation, erythritol and ethanol, but not mannitol, are excreted. The erythritol pathway is required in addition to the ethanol pathway owing to the low activity of acetaldehyde dehydrogenase in the ethanol pathway (Veiga-Da-Cunha et al. 1993; Richter et al. 2001). Normally, the activity of acetaldehyde dehydrogenase is just sufficient for NAD(P)H reoxidation. Under pantothenate limitation, however, the bacteria contain low amounts of HSCoA and acetyl-CoA (Richter et al. 2001). This causes a decrease in acetaldehyde dehydrogenase activity below the levels required for NAD(P)H reoxidation in the ethanol pathway, and the additional capacity required is supplied by the erythritol pathway (Richter et al. 2001).

When O. oeni grows on fructose, on the other hand, large amounts of mannitol are produced in addition to lactate and acetate, whereas ethanol and erythritol are found only in low amounts (Salou et al. 1994). About one-third of the fructose is fermented via the phosphoketolase pathway, similar to glucose, whereas the remainder is used as the electron sink for NAD(P)H reoxidation (3 fructose→1 acetate+1 lactate+2 mannitol+2 ATP). Since mixed heterolactic/mannitol fermentation (0.66 ATP/hexose) provides no increased ATP yields or other obvious advantages compared to heterolactic fermentation (1 ATP/hexose), the heterolactic fermentation of fructose in the absence of the mannitol pathway was tested. As shown here, O. oeni is able to grow by pure heterolactic fermentation of fructose. This raises the question why, under most conditions, mannitol or other reduced end-products are formed in addition or instead of ethanol. To answer this question, the capacities of the ethanol, mannitol and erythritol pathways for NAD(P)H reoxidation were determined and compared to the activities required for growth on various sugars. The data showed that fructose fermentation is limited by the activities of NAD(P)H reoxidation pathways. In fructose fermentation, this limitation was overcome by the use of the mannitol pathway for NAD(P)H reoxidation, demonstrating that NAD(P)H reoxidation is a general problem in hexose fermentation by O. oeni.

Materials and methods

Bacterial strain and growth conditions

Oenococcus oeni strain B1 (Richter et al. 2001) was grown in modified tomato-juice medium (250 ml of 8,500xg tomato-juice supernatant/l , 5 g meat-derived peptone/l, 20 gtryptone (from caseine)/l, 5 g yeast extract/l , 1 ml/l Tween 80, pH 5.8) for maintenance, and for all other purposes in modified malolactic differential or MLD medium (Cavin et al. 1989) adjusted to pH 5.8, or medium B (Garvie 1967) with the additions indicated by Richter et al. (2001). For growth experiments, fermentation balances and enzyme assays, the bacteria were propagated under anoxic conditions at 30 °C in rubber-stoppered bottles. Medium B was used without or with d-pantothenate (1 mg/l) as specified for individual experiments. For growth experiments, the bacteria were grown in modified MLD medium, which supplies all essential growth factors. The medium was supplemented with fructose, glucose, ribose, or pyruvate as indicated. To follow culture growth and fermentation, samples were withdrawn aseptically as required, and optical density was measured at 578 nm. The relationship between optical density and dry weight was 300 mg dry weight (dw)−1 (OD unit)−1. The growth rate constant (µ) was determined by plotting the log10 of the optical density against time. For growth in continuous culture, anoxic modified MLD medium was used, and the chemostat (400 ml) and the reservoir were flushed continuously with a stream of N2. The dilution rate was varied between 0.023 and 0.103 (h−1). In the overflow, optical density (OD578) and the concentrations of the substrates and products were measured by HPLC.

Fermentation

To measure fermentation by growing cells, the bacteria were grown in 50 ml modified MLD medium with fructose, glucose, ribose, or pyruvate. At various times, samples were taken aseptically, clarified by centrifugation, and the supernatant was stored at −80 °C. For measurement in cell suspensions, bacteria were harvested at the exponential-growth phase, washed, and resuspended in Mops buffer at OD578=6 under anoxic conditions in rubber-stoppered bottles. Samples (0.5 ml) were withdrawn by syringes as required, and centrifuged at 10 000xg for 5 min. Substrates and products in the supernatant were determined by HPLC using an Aminex HPX 87H column and conditions and procedures described previously (Richter et al. 2001).

Enzyme activities

Subcultures of O. oeni were grown in 10 ml tomato juice medium for 1–2 days at 30 °C under anoxic conditions. Eight ml of the subculture were inoculated into 400 ml of modified MLD medium containing the respective C-source. The bacteria were grown under anoxic conditions to OD578=0.8, harvested at 4 °C, and the wet cells stored at −80 °C. Before use, the bacteria were thawed slowly by incubation at 0 °C, resuspended and washed in 20 ml Mops (0.1 M) at pH 6.5. The cells were then mixed with 10 ml of the same buffer and 20 g glass beads (Zirkona-Silica, 0.1 mm, Roth, Germany) and disrupted in a cell mill (Vibrogen VI4, Bühler) at 4 °C/75 Hz for three cycles (3 min each) with intervals of 0.5 min. The mixture was then centrifuged for 10 min at 10,000xg, and the supernatant (1.0–1.5 mg protein/ml) was used for measurement of enzyme activities. Cell homogenates used for measurement of acetaldehyde dehydrogenase were prepared by the same procedure, but the Mops buffer was replaced by a buffer containing 0.1 M Tris and 5 mM dithiothreitol at pH 7. Mannitol dehydrogenase, erythritol-4-phosphate dehydrogenase, lactate dehydrogenase, and acetaldehyde dehydrogenase were measured photometrically as the NAD(P)H-dependent reduction of fructose (NADPH), erythrose 4-phosphate (NADPH), pyruvate (1 mM) (NADH), and acetyl-CoA (NADH), respectively, at 30 °C and 365 nm (ε 365=3.4 mM−1 cm−1) (adenine dinucleotides in parentheses are the coenzymes) as described (Veiga-Da-Cunha et al. 1993; Richter et al 2001).

Results

Two alternative pathways for fructose fermentation

Oenococcus oeni was grown in batch culture on fructose, and the fermentation products were determined throughout to characterize the fermentation (Fig. 2). During exponential growth, mannitol, lactate and acetate were the major products, small amounts of ethanol and erythritol were formed in addition. In the late-exponential growth phase, the fermentation pattern changed, and lactate, ethanol and erythritol were the major products. The fermentation balance of the second phase was similar to that of heterolactic or glucose fermentation by the same bacteria, yielding lactate, ethanol and CO2 (Table 1). Furthermore, acetate and erythritol were excreted, but only limited amounts of mannitol. During fermentation in the first fermentation phase, by contrast, large amounts of mannitol and lactate were produced, but not ethanol and erythritol. This form of fermentation is characteristic for mixed heterolactic/mannitol fermentation known for Oenococcus and related bacteria (Kandler 1983; Salou et al. 1994), in which fructose is used as the major acceptor for NAD(P)H from the phosphoketolase pathway. In the intermediate phase, products of both fermentation reactions were found. When resting bacteria or cell suspensions were analyzed for fructose fermentation, the mannitol pathway was even more predominant, and essentially no ethanol or erythritol were formed (not shown), presumably due to uncoupling of energy metabolism from cell-mass production in the resting cells.

Fig. 2a, b.
figure 2

Growth of O. oeni strain B1 in modified MLD medium with 20 mM fructose. The cultures were grown under anoxic conditions. a Content of fructose (❚), lactate (▼), ethanol (❍), acetate (◆), mannitol (●), erythritol (▲). b OD578(❍) and pH (◊) for the same growth experiment

Full size image
Table 1. Fructose fermentation during exponential (phase I) and late-exponential (phase II) growth of a batch culture of Oenococcus oeni. The fermentation balances are derived from the experiment shown in Fig. 2 and refer to the substrate consumption and product formation within phase I and phase II. The fermentation balance for phase I is calculated from the substrate consumption and product formation at the end of the phase; the balance of phase II represents the difference in substrate consumption and product formation between the beginning and the end of the phase. The amount of CO2 produced is calculated and is equivalent to the amount of lactate produced. %C Carbon yield, O/R redox balance of the fermentation balance
Full size table

Suppression of mannitol production by pyruvate

Electron acceptors such as pyruvate are known to affect the fermentation of hexoses by O. oeni (Nuraida et al. 1992). Therefore the effect of pyruvate on the formation of mannitol and other reduced fermentation products was studied in more detail (Table 2). In the presence of pyruvate, lactate and acetate were the only fermentation products of glucose, and no ethanol or erythritol were excreted. Thus, only pyruvate was used as electron acceptor for NAD(P)H regeneration, and acetylphosphate was excreted as acetate. Part of the pyruvate was decarboxylated and converted to acetate (Richter and Unden, unpublished data). A similar result was obtained with fructose. The presence of pyruvate prevented mannitol formation completely, and lactate and acetate were the only fermentation products. Therefore, in the presence of external pyruvate, other pathways for NAD(P)H regeneration, such as the ethanol, erythritol and mannitol pathways, are no longer used by growing (not shown) or resting cells of O. oeni. Lactate dehydrogenase from O. oeni has high activity, but uses only NADH and no NAD(P)H as an electron donor for pyruvate reduction (Veiga-Da-Cunha et al. 1993). This might create problems in the reoxidation of NAD(P)H produced in the phosphoketolase pathway. The bacteria contain, however, a cluster of three genes homologous to the pntAA pntAB pntB genes of three-subunit proton-translocating NAD(P) transhydrogenases which could be responsible for NADH/NADPH transhydrogenation (genes 400, 401 and 402 in scaffold 14 of the O. oeni genome; http://www.jgi.doe.gov/JGI_microbial/html/). Overall, this shows that the pathways for NAD(P)H reoxidation are used to largely varying extents depending on the growth conditions and available substrates.

Table 2. Effect of pyruvate on glucose and fructose fermentation by washed O. oeni cells. The bacteria were grown in modified MLD medium with glucose to the late-exponential growth phase, harvested, washed and incubated in buffer containing glucose or fructose (10 mM) and an excess of pyruvate. After consumption of the hexoses, the fermentation balances were determined. The carbon yields were comparable to those given in Table 1
Full size table

Control of fructose fermentation pathways by the growth rate

The transition from heterolactic/mannitol to heterolactic fermentation in batch culture (Fig. 2) was related to a decrease in pH and an increase in the concentrations of organic acids (acetate and lactate). Therefore, it was of interest to determine whether pH and increased concentrations of the organic acids are responsible for the shift in the fermentation pattern. The fermentation was carried out from the beginning at decreased pH (pH 4.5) or in the presence of 10 mM acetate, but neither of these conditions prevented or reduced mannitol excretion (not shown). When growing bacteria in the state of heterolactic fermentation (phase II), were incubated in buffer with fructose, the bacteria converted fructose by mixed heterolactic/mannitol fermentation, even when chloramphenicol was added after growth to prevent de novo protein synthesis. This suggests that for the shift from heterolactic to heterolactic/mannitol fermentation enzyme synthesis is not required and that the shift occurs at the post-translational level.

Other factors that could be related to the shift from heterolactic/mannitol fermentation to heterolactic fermentation are growth rate and growth phase. Using a chemostat, the growth rate of O. oeni growing on fructose was varied by changing the dilution rate, and the fermentation products were determined (Fig. 3). At high dilution rates (µ=0.103 h−1) mannitol was excreted in large amounts, surpassing ethanol excretion by a factor of 4.1. Mannitol production decreased continuously with decreasing growth rate, and at the same time the amounts of ethanol increased. At low growth rates, the ratio mannitol/ethanol was as low as 0.19. The amounts of acetate did not change significantly, whereas erythritol production slightly increased. Thus, at low growth rates heterolactic fermentation dominates. At high growth rates, 57% of the fructose () is converted to mannitol, which is close to the theoretical amount (67%) for mannitol/heterolactic fermentation. It can be concluded that the type of fructose fermentation is related to the growth rate: high growth rates favor mixed heterolactic/ mannitol fermentation, whereas low growth rates favor heterolactic fermentation.

Fig. 3.
figure 3

The effect of carbon (fructose) limitation on the growth rate of a chemostat culture of O. oeni B1. The bacteria were grown under anoxic conditions in the chemostat at various dilution rates due to fructose limitation. After changing the dilution rates, the bacteria were grown until a new equilibrium was obtained. Samples were then withdrawn and the OD578, fructose concentration, and the concentrations of the fermentation products lactate (▼), ethanol (❍), and mannitol (●) were measured by HPLC

Full size image

Activities of the dehydrogenases for NAD(P)H formation and reoxidation

The activities of the dehydrogenases from the NAD(P)H-reoxidizing ethanol, erythritol, and mannitol pathways were measured after growth of O. oeni on various substrates and compared to those from the NAD(P)H-generating enzymes from the phosphoketolase pathway, i.e. glucose-6-P dehydrogenase and 6-P-gluconate dehydrogenase (Table 3). In general, the activities of the glucose-6-P and the 6-P-gluconate dehydrogenases largely exceeded those of the NAD(P)H-reoxidizing enzymes from the ethanol, erythritol and mannitol pathways (at least 2.6-fold). Only lactate dehydrogenase had significantly higher activities. In the ethanol pathway, which contains acetaldehyde and ethanol dehydrogenases, only acetaldehyde dehydrogenase was measured; this enzyme has low activity and is the rate-limiting enzyme of this pathway (Veiga-Da-Cunha et al. 1993; Richter et al. 2001). The activity of ethanol dehydrogenase is two orders of magnitude higher and supposedly not significant for the control of the ethanol pathway. In addition, lactate dehydrogenase, which can be used for reoxidation of NAD(P)H from the phosphoketolase pathway when external pyruvate is supplied, was measured.

Table 3. Dehydrogenase activities (µmol (g dry weight−1) min−1) from the phosphoketolase pathway (glucose-6-P and 6-P-gluconate dehydrogenases), and from the ethanol, erythritol, mannitol and lactate pathways of NAD(P)H reoxidation. The dehydrogenases were measured in cell homogenates after growth of the bacteria on various sugars. G6P DH Glucose-6-P dehydrogenase, 6PG DH, 6-P-gluconate dehydrogenase, ALDH acetaldehyde dehydrogenase, E4P DH erythritol-4-phosphate dehydrogenase, MDH mannitol dehydrogenase, LDH lactate dehydrogenase. ND Not determined
Full size table

Lactate dehydrogenase exceeded the activities of the other dehydrogenases by 1 to 2 orders of magnitude (Table 3). With the exception of acetaldehyde dehydrogenase, the activities of the dehydrogenases were rather constant after growth on the various sugars, varying only by factors of maximally 2.4. Addition of pyruvate or fructose as electron acceptor had no major effects on the enzyme activities. This suggests that erythritol-4-phosphate, mannitol, and lactate dehydrogenase genes are regulated only slightly at the transcriptional or translational level. Acetaldehyde dehydrogenase activity, however, responded strongly to the growth substrates. The activity was maximal after growth on glucose, where the enzyme is required for ethanol production. The activity dropped to very low levels after growth on fructose or in the presence of pyruvate. Under these conditions, very low amounts of ethanol are produced by the bacteria. It is not known whether the activity of acetaldehyde dehydrogenase is regulated at the transcriptional or post-translational level.

Variation of the growth rates by the substrates

When O. oeni was grown on different sugars as the substrate, the resulting growth rates varied considerably. The rate was lowest with glucose, intermediate with fructose, and with ribose about three times higher than with glucose (Table 4). Supplementation of glucose-grown bacteria with pyruvate or fructose greatly increased the growth rate such that it was close to that measured using ribose as substrate. Since fermentation of all sugars requires the phosphoketolase pathway but different NAD(P)H reoxidation pathways, this suggests that the latter are responsible for the low growth rates obtained with the hexoses. Supplementation of bacteria growing on fructose with pyruvate resulted in only a small stimulation of growth (not shown) compared to glucose, which could have been due to the higher growth rates and NAD(P)H reoxidation capacities during growth on fructose.

Table 4. Growth rates and yields (Y Hexose, Y ATP) of O. oeni on sugars with or without electron acceptors (fructose or pyruvate), and kinetic parameters for sugar consumption. The growth rates and yields were estimated from batch cultures in modified MLD medium with the respective substrates (40 mM glucose, 20 mM fructose, 10 mM glucose plus 10 mM pyruvate, or 20 mM ribose). The ATP produced (mol/mol sugar) was derived from the fermentation balances and the ATP producing reactions shown in Fig. 1. Y ATP was calculated from Y Hexose and the ATP (mol/mol sugar) produced in the pathway. The rate of hexose metabolism (a) was calculated from the corresponding growth rates (µ) and the growth yields (Y Hexose) according to a=µ/Y as described (Richter et al. 2001). dw Dry weight. The fermentation balances are given below
Full size table

The cell yields were estimated from cell mass and sugar consumption (Table 4). For growth on glucose or fructose, relatively low yields were found, and addition of pyruvate or fructose to cells growing on glucose increased the yields considerably. The Y ATP values, calculated from Y sugar and the ATP yield, were all similar, which confirms the validity of the experimental values.

Effect of pantothenate on fructose fermentation

In glucose fermentation, the erythritol pathway is required mainly due to the limitation of the ethanol pathway by a low pantothenate content and low acetaldehyde dehydrogenase activity (Richter et al. 2001; Veiga-Da-Cunha et al. 1993). The limitation can be overcome by supplying pantothenate, which increases the amount of cellular HSCoA (Richter et al. 2001). Fructose fermentation by cell suspensions of O. oeni produced low amounts of erythritol (0.09 mol/ mol fructose) in addition to lactate, acetate, and mannitol, but no ethanol (see Table 2). When the bacteria were supplied with pantothenate during growth and incubation, the amounts of ethanol increased significantly to 0.07 mol/mol fructose, and the alternative reduced products (mannitol and erythritol) decreased together by the same amount. Therefore, in fructose fermentation, too, the ethanol branch can be stimulated by pantothenate, although to a lesser extent than in glucose fermentation.

Discussion

Two pathways for fructose fermentation in O. oeni

Fructose can be metabolized by two different pathways in O. oeni, either by heterolactic fermentation or by mixed heterolactic/mannitol fermentation. Heterolactic fermentation is carried out by bacteria in the late-exponential growth phase and in bacteria growing at low growth rates due to C-limitation. Mixed heterolactic/mannitol fermentation is the common growth reaction on fructose during exponential growth and for O. oeni growing at a high rate. Conversion of fructose exclusively to mannitol, on the other hand, is found during cofermentation with glucose. The switch between the two types of fermentation apparently occurs at the metabolic level, and only acetaldehyde dehydrogenase activities change significantly in response to the growth conditions. The shift is related to the growth rate, or to the high metabolic rates present during high growth rates. A relation to fructose concentration is unlikely, since fructose concentration did not significantly increase at the relevant dilution rates in the chemostat; however, regulation by the fructose concentration cannot be completely excluded. At high growth and metabolic rates, the typical ethanol pathway for NAD(P)H reoxidation in heterolactic fermentation becomes limiting and complementary pathways for NAD(P)H regeneration are required. In glucose fermentation the erythritol pathway takes over this role, in fructose fermentation the mannitol pathway.

The need for alternative pathways for NAD(P)H reoxidation in O. oeni

The rate of NAD(P)H production in the phosphoketolase pathway can be calculated from the growth rate and from yields of the bacteria (Table 5); this rate also sets the rate of NAD(P)H reoxidation that has to be achieved by the NAD(P)-reoxidizing pathways. These rates can be compared to the activities of the NAD(P)H-producing and NAD(P)H-regenerating dehydrogenases of the phosphoketolase and of the NAD(P)H-reoxidizing pathways (Table 3). The activities of glucose-6-P and of 6-P-gluconate dehydrogenases exceed the required activities of the pathway by factors of at least 8.7 and 5.1, respectively Table 3). Thus NAD(P)H production should not be limited by the activities of these enzymes. A comparison of the metabolic rates with the activities of the NAD(P)H-oxidizing enzymes shows, however, that these enzymes do not keep up with this task (Table 5). Thus, it is evident that the capacity of the typical pathway for NAD(P)H reoxidation in heterolactic bacteria, the ethanol pathway,is insufficient. For this reason, generally more than one pathway for NAD(P)H reoxidation is required. With glucose, the rates of NAD(P)H reoxidation nearly approach the required activities if sufficient pantothenate is supplied (Richter et al. 2001). Under these conditions, the additional use of the (low-activity) erythritol pathway provides sufficient activity. With fructose, on the other hand, the mannitol pathway is additionally required. Quantitative evaluation of the NAD(P)H-supplying and NAD(P)H-consuming pathways explains why pathways for NAD(P)H reoxidation in addition to the ethanol pathway are necessary.

Table 5. Comparison of experimentally determined dehydrogenase activities for NAD(P)H reoxidation with activities of NAD(P)H reoxidation required for various fermentation reactions. It is assumed that for growth on glucose, fructose, glucose+fructose and glucose+pyruvate, the ethanol+erythritol, ethanol+erythritol+mannitol, ethanol+erythritol+mannitol, and lactate pathways, respectively, are used for NAD(P)H reoxidation (compare fermentation balances of Table 4). The rate of NAD(P)H production was calculated from the rates of hexose metabolism given in Table 4. This value was corrected for the amount of hexose metabolized via the phosphoketolase pathway and the amount of NAD(P)H produced in this pathway (2 NAD(P)H per pathway). The total capacity for NAD(P)H oxidation is calculated from the activities of respective dehydrogenases (compare with Table 3), and the amount of NAD(P)H reoxidized in the pathway (2 in the ethanol pathway, 1 in the erythritol and mannitol pathways). EtOH Ethanol pathway, Eryt erythritol pathway, Man mannitol pathway, Lac lactate pathway for reduction of externally supplied pyruvate
Full size table

Independent experimental support for the limiting NAD(P)H reoxidation capacity of the ethanol pathway is provided by the chemostat experiment. At slow growth rates, and the resulting decreased NAD(P)H production rates, the activities of the ethanol pathway are nearly sufficient resulting in low levels of mannitol excretion. At high growth rates, on the other hand, the high NAD(P)H production rates require NAD(P)H reoxidation by the mannitol pathway resulting in high mannitol/ethanol ratios. The maximal activities of acetaldehyde dehydrogenase (80 µmol (g dry wt)−1 min−1, Richter et al. 2001) would be sufficient for growth rates up to 0.04 h−1 if only the ethanol pathway is used for NAD(P)H reoxidation. For high growth rates (up to µ=0.103 h−1 in the chemostat), NAD(P)H reoxidation by the mannitol pathway is required in addition. Since the activity of acetaldehyde dehydrogenase is decreased also during growth on fructose, this limitation becomes even more serious.

Similarly, Maicas et al. (2002) recently showed that NAD(P)H reoxidation is stimulated if O2 is present as an electron acceptor, although growth and some enzyme activities are inhibited by O2. The alternative NAD(P)H reoxidation pathways (mannitol and erythritol pathways) discussed here stimulate NAD(P)H reoxidation and growth. Thus, the question of NAD(P)H reoxidation is central to heterolactic fermentation, and an important role was attributed to the NAD(P)H/NAD(P) ratio (Maicas et al. 2002). Comparing the NAD(P)H-producing and NAD(P)H-reoxidizing activities demonstrates that low capacities for NAD(P)H reoxidation apparently represent the primary problem in heterolactic fermentation of O. oeni. This limitation would be responsible for the increased NAD(P)H/NAD(P) ratios, and it is possible that the NAD(P)H/NAD(P) ratio has some regulatory effect on metabolism.

Effect of NAD(P)H reoxidation capacity on growth rates of O. oeni

During growth on various sugars, glucose resulted in the lowest, and ribose the highest growth rate. Since for growth on ribose only the phosphoketolase pathway, but no NAD(P)H reoxidizing reactions are required, this suggests that the phosphoketolase pathway allowed growth rates up to 0.101 h−1. On the other substrates, growth is limited by the low capacities for NAD(P)H reoxidation. In accordance with this assumption, the increased growth rate is related to the capacities of the NAD(P)H reoxidation pathway in the presence of the respective sugar. When pyruvate is present in addition to the hexoses, increased growth rates similar to those on ribose are measured. This stimulation is presumably due to the high capacities for NAD(P)H reoxidation by lactate dehydrogenase with pyruvate, although the growth rates are not as high as those on ribose. The experiments suggest that the growth rates of O. oeni are controlled to a large extent by the capacities of NAD(P)H reoxidation. Thus, a quantitative assessment of NAD(P)H reoxidation is an important aspect for understanding fermentation and growth of O. oeni.