Bioenergetic consequences of microbial adaptation to low-nutrient environments

Journal of Biotechnology 59 (1997) 117 – 126

Bioenergetic consequences of microbial adaptation to low-nutrient environments M.J. Teixeira de Mattos *, O.M. Neijssel E.C. Slater Institute for Biochemistry and Microbiology, Uni6ersity of Amsterdam, Amsterdam, The Netherlands Received 22 September 1996; received in revised form 13 February 1997; accepted 25 July 1997

Abstract A striking property of many prokaryotes is their enormous metabolic flexibility with respect not only to catabolic and anabolic substrates but also with respect to the continuously changing availability of nutrients. The phenotypic responses to low-nutrient growth conditions involve structural changes in the cellular make-up, changes in the specific capacity of the enzyme system(s) involved in uptake and/or assimilation of the limiting nutrient and changes in the affinity of these enzymes. Here the responses of some members of the Enterobacteriaceae to potassium-, ammoniumand energy source-limited conditions will be reviewed. The focus will be on the energetic consequences of these adaptations as reflected by the growth yield value for the energy source (Yenergy source). It will be illustrated that Yenergy source values can be dramatically lowered as a result of incomplete oxidation of the energy source (overflow metabolism), bypassing potential sites of energy conservation (uncoupling) or catabolic cycles that have no other apparent effect than the hydrolysis of ATP (futile cycles). Thus, it is concluded that adaptation to low nutrient conditions aims at maintaining high metabolic fluxes at low nutrient concentrations at the cost of a loss in the energetic efficiency of the overall metabolism. © 1997 Elsevier Science B.V. Keywords: Bioenergy; Microbial adaptation; Low-nutrient environments

1. Introduction Microbial growth is dependent on the presence of both anabolic resources and catabolic, i.e. energy providing substrates. Moreover, whenever one nutrient is present in a subsaturating con* Corresponding author.

centration, the specific rate (q, mmol substrate consumed (g dry weight) − 1 h − 1) at which this nutrient can be consumed by the cell is dependent on its actual concentration. Assuming that the key enzymes in the metabolism of a particular nutrient follow Michaelis-Menten enzyme kinetics: 6= 6max

0168-1656/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved. PII S 0 1 6 8 - 1 6 5 6 ( 9 7 ) 0 0 1 7 4 - 0

s Km + s

(1)

M.J. Teixeira de Mattos, O.M. Neijssel / Journal of Biotechnology 59 (1997) 117–126

118

and that q is proportional to 6 (the enzymes that catalyze the metabolic steps involved being a fraction of the biomass), it follows that the dependency of q on the substrate concentration is hyperbolic: q= qmax

s Ks + s

(2)

where s is the concentration of the limiting nutrient, Ks the cell’s affinity constant for this nutrient and qmax the rate that can be obtained under saturating conditions. In addition, in a first approach (Tempest and Neijssel, 1984) the specific consumption rate is related linearly with the specific growth rate m: m =q · Ysx

(3)

For the purpose of this contribution this equation is useful despite the fact that it is only rarely valid, due to changes in cellular composition with growth rate (if s refers to an anabolic substrate), or due to maintenance requirements (if s refers to a catabolic substrate). (Stouthamer, 1979; Pirt, 1982; Stouthamer and van Verseveld, 1985; Neijssel and Teixeira de Mattos, 1994; Russell and Cook, 1995; Neijssel et al., 1996). It can be seen that the proportionality factor Ysx (the growth yield) is thus defined as the amount of cells formed per mol substrate consumed. In nature, micro-organisms are often faced with a limited availability of nutrients. It is, therefore, not surprising that mechanisms have evolved that serve to adapt to environments in which the availability of a single (or more) nutrient(s) is low (Harder and Dijkhuizen, 1983; Dawes, 1985). Although genotypic adaptation, the arise of a mutant growing faster under a particular nutrientlimited condition, is of tremendous importance (Dykhuizen and Hartl, 1983), this contribution will be limited to phenotypic adaptive mechanisms in which genes already present in the genome are expressed as a result of the growth environment. If it is assumed that adaption is successful when it results in a sustained growth rate despite a lowering of the availability of a nutrient, it follows from the equations that two strategies could fulfil this condition. The rate of consumption

should be kept constant by an increased maximum specific consumption rate (qmax) or by lowering the affinity constant. Alternatively, the Ysx value should be increased, allowing the same growth rate at a decreased consumption rate. The latter strategy is observed often with anabolic substrates as it can be achieved by a decreased biosynthetic demand for that substrate (e.g. the cellular content of phosphorus of Bacillus subtilis is lowered under phosphate-limited conditions due to a change in cell wall constituents (Tempest et al., 1968). However, here we will focus on some examples of adaptation that affect the energetics of microbial growth considerably.

2. Adaptive mechanisms

2.1. Increased affinity Most, if not all micro-organisms that have been studied until today possess two (or more) structurally different uptake systems with a widely different affinity towards a particular nutrient. A well-studied example (Bakker et al., 1987) is the uptake of potassium ions. Potassium is the major cation in the cell and is necessary for a wide range of processes: it serves as a counter-ion for many acidic metabolites, it stabilizes ribosomes, it activates certain enzymes and it is essential in the maintenance of turgor pressure and pH homeostasis (Booth, 1985; Csonka and Epstein, 1996). In Gram-negative bacteria, such as Escherichia coli or Klebsiella pneumoniae, grown in the usual media, [K + ]in is 150–250 mM, but this concentration can become even higher when the osmotic pressure of the medium is increased. In Gram-positive bacteria, such as B. subtilis, [K + ]in is close to 400 mM (Tempest et al., 1968). In E. coli K + can be taken up via two transport systems, called Trk and Kdp. The former system is constitutive and has a low-affinity for K + (Km = 1.5 mM) (Stewart et al., 1985). K + transport in environments with [K + ] \ 1 mM is mediated by this system only, but when the concentration of K + is below this value, the Kdp system is induced which has a very high affinity for K + (Km = 2 mM) (Walderhaug et al., 1987; Siebers and Al-

M.J. Teixeira de Mattos, O.M. Neijssel / Journal of Biotechnology 59 (1997) 117–126

tendorf, 1988). It should be emphasized here that the energy consumption required for K + uptake via Kdp is higher than via the Trk system. Thus, in low potassium environments the above mentioned conditions for an increased affinity are fulfilled at the cost of an extra expenditure of energy. The vastly more dramatic effects of the simultaneous presence of two transport systems (Trk is not repressed under low-potassium conditions) will be discussed below. Much along the same lines, one can interpret the difference in the kinetics of incorporation of NH3 (ammonia) or NH4+ (ammonium ion) between organisms growing ammonium-limited and under ammonium excess conditions (Reitzer, 1996a,b). As it is now generally accepted that NH3 can rapidly diffuse through the membrane of many microbial species (Kleiner, 1985), the adaptation of the cell in this case does not involve transport but rather the kinetics of the first assimilatory intracellular enzymes. In most microbes grown under N-excess conditions with NH3 as the sole N-source, this compound is assimilated via the following reaction that is catalyzed by glutamate dehydrogenase which does not have a low affinity constant for ammonia: 2-oxoglutarate+ NH3 +NAD(P)H“glutamate +NAD(P) + H2O. After glutamate is formed, all other amino acids are formed via transamination reactions. When K. pneumoniae is grown in an ammonialimited chemostat culture, it can be observed that in these cells glutamate dehydrogenase is absent and that the extracellular NH3 levels are very low, which suggests that the assimilation is carried out by another, high affinity enzyme system. Tempest et al. (1970) discovered that the high-affinity route of ammonia assimilation is the following:

119

called GOGAT, for glutamine 2-oxoglutarate aminotransferase). Whereas glutamine synthetase is always present in organisms grown on NH3 as the sole N-source because glutamine is essential for protein synthesis, it is repressed when sufficient NH3 is available (\ 1 mM). From the reaction equations given above it can be seen that under ammonia-limited conditions glutamate formation takes place at the expense of energy. As a last example of variable affinity, it is interesting to draw attention to the use of oxygen by heterotrophic micro-organisms. For many prokaryotic species molecular oxygen is, strictly speaking, not an essential nutrient for growth. Very often oxygen serves only as an electron acceptor reacting with the terminal cytochrome of the respiratory chain. Hence, the rate of oxygen consumption is determined by the rate of supply of reducing equivalents on the one hand and the capacity of the respiratory chain on the other. The respiratory chain of E. coli is not a simple sequence of redox carriers (Gennis and Stewart, 1996). From Fig. 1 it can be seen that various pathways for the electrons to flow to oxygen are possible. Two NADH dehydrogenases can feed electrons to the quinone pool. From there on, electrons flow to either of two terminal oxidases. Cytochrome-bo is a cytochrome with a Km towards O2 of 1.4–2.9 mM. Electron transport from the ubiquinone pool via this chain is supposedly coupled to proton translocation with a stoichiometry of 2H + /e. At lower, growth limiting oxygen tension, the synthesis of cytochrome-bo is repressed and the synthesis of a different cytochrome, cytochrome-bd is stimulated (Iuchi and Lin, 1993; Lynch and Lin, 1996). The affinity

glutamate+NH3 + ATP “ glutamine + ADP + Pi + H2O glutamine + 2-oxoglutarate +NADH “2 glutamate + NAD+ + H2O . 2-oxoglutarate+ ATP + NADH +NH3 “glutamate+ ADP+ Pi + NAD + + 2H2O Glutamine formation is catalyzed by glutamine synthetase, which has a higher affinity towards NH3 than glutamate dehydrogenase. Next, glutamine is converted with 2-oxoglutarate to glutamate via catalysis by glutamate synthase (also

constant of this cytochrome is about 10-fold lower (Km value of 0.23–0.38 mM) than that of cytochrome-bo. Thus at lower oxygen concentrations a high affinity system is induced to scavenge the oxygen from the environment. It is important

120

M.J. Teixeira de Mattos, O.M. Neijssel / Journal of Biotechnology 59 (1997) 117–126

to note that the H + /e stoichiometry of oxygen reduction via cytochrome-bd is lower (approximately 1) than that via cytochrome-bo (Puustinen et al., 1991). Therefore, as with the other examples given above, the price paid for the increased affinity is a loss in energetic efficiency of respiration. In summary, it can be generalized that induction of high affinity uptake systems is dependent on the substrate being present in concentrations too low for the low affinity system to work at Vmax, that is, induction occurs under nutrient-limited conditions. A second conclusion is that involvement of high affinity systems is always accompanied by a loss in energetic efficiency: this may explain the presence of low affinity systems as it allows the cell to economize on the energy costs of nutrient uptake when the nutrient is present in excess of the growth requirement.

2.2. Increased transport capacities K. pneumoniae transports glucose into the cell via the phospho-enol-pyruvate-dependent phos-

Fig. 1. Schematic representation of the respiratory chain of E. coli. NDH1, the translocating NADH dehydrogenase; NDH2, the uncoupled NADH dehydrogenase; bo, bo type (low affinity, high H + /e stoichiometry) cytochrome oxidase; bd, bd type (high affinity, low H + /e stoichiometry) cytochrome oxidase. Adapted from Calhoun et al. (1993).

Fig. 2. Changes in the glucose uptake capacity with growth rate of K. pneumoniae NCTC 418 grown in aerobic, glucoselimited chemostat culture. The capacity is measured in vitro. Adapted from Neijssel et al. (1980).

photransferase system (PTS). Details about the structure of this system can be found elsewhere (Postma et al., 1996), here it is sufficient to note that glucose uptake is accompanied by phosphorylation of the molecule yielding glucose 6-phosphate intracellularly. The activity of the PTS of K. pneumoniae was studied in glucose-limited chemostat cultures of this organism and the results are shown in Fig. 2 (Neijssel et al., 1980). It is clear that the Vmax of the PTS and hence the qmax, is increased at lower dilution rates. Since a lower dilution rate implies a lower glucose concentration in the culture fluid, it is evident that this organism behaves as predicted above. In contrast, the relationship between the amount of PTS in cells of E. coli (another member of the Enterobacteriaceae) and the dilution rate (and therefore the extracellular glucose concentration) was quite different (data not shown): at increased dilution rates an increase in the PTS activity was observed. These results would suggest that in this organism glucose acts as an inducer of the PTS (its activity increases with the extracellular glucose concentration). Assuming that in E. coli the PT-system is solely responsible for glucose transport, this shows that we have to be careful with a generalization of the concepts we have developed above.

M.J. Teixeira de Mattos, O.M. Neijssel / Journal of Biotechnology 59 (1997) 117–126

Nevertheless, if we follow the predictions we are forced to conclude that K. pneumoniae must be better adapted to a glucose-limited environment than E. coli. This has been observed indeed (Leegwater et al., 1982).

3. Bioenergetic consequences Many of the physiological responses to nutrient-limited growth affect the Yenergy source values by mechanisms that can be catagorized as overflow metabolism, uncoupling mechanisms and futile cycles. Here, their relation with the above mentioned examples of adaptation will be illustrated.

3.1. O6erflow metabolism Under aerobic conditions the energy source is usually completely oxidized to CO2 when it is available in growth rate limiting amounts, but this is often not the case under energy source excess conditions. For example, it was found with K. pneumoniae (Neijssel and Tempest, 1979) that with glucose-containing cultures there was a 2 – 3fold difference in Yglucose value between glucose sufficient cultures as compared with those that were glucose-limited (Table 1). With the former Table 1 Specific rates of glucose consumption and of product formation in variously-limited anaerobic chemostat cultures of K. pneumoniae NCTC 418 Limitation

C

N

S

P

K

Formate+CO2 Acetate Ethanol D-lactate 2,3-Butanediol Succinate

11 5.4 5.4 0 0 0

10 3.1 4.0 1.2 1.5 0.3

12 1.3 4.8 3.2 3.2 0.4

12 3.0 5.0 0.3 2.5 0.4

19 7.0 8.6 3.1 2.5 0.6

Yglucose ATP/glucosea

21 3.0

16 2.2

12 2.2

17 2.6

8.7 2.2

81

97

C recovery

100

103

87

D =0.13 h−1; 35°C; pH 6.8. Rates are expressed in mmol (gram dry weight)−1 h−1, carbon recovery in %. a The efficiency is corrected for anabolic glucose consumption. Adapted from Teixeira de Mattos and Tempest (1983).

121

cultures, a number of partially oxidized end products (e.g. acetate, ketoglutarate, pyruvate, ketogluconate, etc) was excreted into the culture fluid. This phenomenon has been named overflow metabolism. Another well-known example is the production of acetate by E. coli when grown under carbon source excess conditions (e.g. ElMansi et al., 1986; Tempest and Neijssel, 1992). Although the energetic consequences of this phenomenon are difficult to quantify, (one reason being the variable efficiency of the respiratory chain (see below)), the amount of reducing equivalents made available to the respiratory chain per amount of energy source dissimilated was lowered as compared to glucose-limited conditions. Under fermentative conditions it could be unequivocally established that a decreased efficiency of ATP synthesis from glucose via substrate level phosphorylation occurred when the energy source was in excess (Teixeira de Mattos and Tempest, 1983). Only under glucose-limited conditions the energy source was fermented solely to acetate and ethanol (plus the concomittant formate/H2 + CO2), yielding the maximal number of 3 mol of ATP per glucose catabolized (Table 1). Under all other nutrient-limited conditions other products, such as 2,3-butanediol, lactate and succinate, were produced and therefore an energetic efficiency lower than 3 ATP/glucose was calculated. In analogy with aerobic overflow metabolism, all products other than acetate and ethanol can be considered as anaerobic overflow products. Similar behaviour is found with lactic acid bacteria: whereas under carbon source excess conditions a major part of the energy source is fermented to lactate, under energy-limited conditions these organisms shift their fermentation to the production of acetate which yields additional ATP (Gottschalk, 1986; Russell and Cook, 1995).

3.2. Uncoupling Under aerobic conditions, the respiratory chain is the major system to conserve energy. Another, equally important function of this system is to dispose of reducing equivalents in order to maintain redox neutrality. If these two functions were tightly coupled, an increased flux through the

122

M.J. Teixeira de Mattos, O.M. Neijssel / Journal of Biotechnology 59 (1997) 117–126

respiratory chain due to an increased capacity and/or affinity of the system would result in an energy conservation rate that would exceed the cell’s actual rate of energy consumption for anabolism, unless mechanisms exist that uncouple the oxidation of NADH from the production of ATP equivalents. Although one would not expect uncoupling mechanisms to operate in organisms that were limited in their growth by the availability of the carbon and energy source, their presence is indicated by the fact that a rapid increase in the glucose and oxygen consumption rate is observed upon the sudden addition of a cell-saturating concentration of glucose to a glucose-limited culture. An explanation for this may be found in the fact that the respiratory chain of E. coli is branched. Above we have argued that the respiratory chain of E. coli is branched at the terminal side in order to scavenge oxygen from the environment under low oxygen conditions. It was mentioned also that the H + /e stoichiometry of the two oxidase is different and their simultaneous presence would constitute a potential uncoupling mechanism. In view of the fact that NDH2 does not contribute to proton translocation (Jaworowski et al., 1981) whereas NDH1 does, it is tempting to postulate that the branching at the level of NADH dehydrogenases (see Fig. 1) is another means to enhance the capacity of the respiratory system in order to counteract an increased flux of NADH into the chain (e.g. upon relief of energy source limited conditions) without the rate of energy conservation being increased. The growth energetics of strains of E. coli in which the genes encoding the two NADH dehydrogenases and the two cytochrome oxidases have been selectively deleted have been studied (Calhoun and Gennis, 1993; Calhoun et al., 1993). Since these mutants differed from the parent organism only in the deletion of a single gene, it is reasonable to assume that all strains required the same amount of energy for cell synthesis and that therefore differences in the specific rates of oxygen consumption reflected differences in the energetic efficiencies of the alternative respiratory branches. By comparing the qO2 values of the various strains grown under identical conditions (Fig. 3),

Fig. 3. Specific rates of oxygen consumption (qO2) of E. coli strains growing in aerobic, glucose-limited chemostat culture: (“) wild type; ( ) ndh. Adapted from Calhoun et al. (1993).

the contribution of each of the alternative branches to the total electron flux could be calculated. Interestingly, in the wild type organism part of the electron transport is catalyzed by NADH dehydrogenase II (the enzyme that does not conserve energy): the ndh mutant (in which the gene encoding NADH dehydrogenase II has been deleted) consumed less oxygen than did the wild type at similar growth rates. In addition, from Fig. 3 it can be seen that the q-D curves for the wild type and the ndh − strain are parallel. This could be interpreted as a consequence of differences in maintenance requirements, but as such a difference is difficult to envisage for these two virtually identical strains, the conclusion must be that at lower growth rates a larger fraction of the total electron flux is funneled through the uncoupled NADH dehydrogenase. Hence, at lower growth rates, i.e. under more stringent nutrient limited conditions, the energetic efficiency is decreased. Once more, this is an example of a mechanism that serves to safeguard high metabolic flexibility at the cost of energetic efficiency. Equally important, it illustrates how carefully one should go about interpreting YO values with regard to the energetics of bacterial growth: a change in these values as a consequence of a change in the growth conditions does not necessarily imply a similar change in the YATP value. Since substrate level phosphorylation is a fully coupled process, under fermentative conditions

M.J. Teixeira de Mattos, O.M. Neijssel / Journal of Biotechnology 59 (1997) 117–126

the cell is not endowed with an energy conservation system with such a high degree of flexibility. Yet, as we have seen, organisms capable of carrying out fermentations with branched routes that differ in ATP yield do have some flexibility (e.g. Enterobacteriaceae and lactic acid bacteria from 2 to 3 ATP/glucose, Clostridium spec. from 2 to (theoretically) 4 ATP/glucose (Gottschalk, 1986)). However, it has been found that K. pneumoniae is able to uncouple glucose catabolism from ATP synthesis to a much larger extent due to the activity of the so-called methylglyoxal (MGO) bypass (Fig. 4) (Teixeira de Mattos et al., 1984). When these organisms are grown anaerobically under glucose-limited conditions glucose is catabolized completely to acetate and ethanol (1:1) and the concomitant amount of formate/H2 +CO2 (3 ATP/glucose). When the limitation was relieved by the sudden addition of glucose, it was taken up instantaneously at a specific rate that is almost double that of the cells growing in the steady state. All extra glucose was fermented to D-lactate (50%) (Fig. 5) and butanediol and succinate (50%)

123

Fig. 5. Uncoupling of anaerobically growing, glucose-limited K. pneumoniae NTCT 418 upon a sudden addition of glucose. The increased rate of glucose consumption is accompanied by production of D-lactate (and succinate and 2,3-butanediol, not shown). No anabolic (growth rate m) change is observed. Assuming all D-lactate is formed via the methylglyoxal bypass, the additional glucose consumption rate is fully uncoupled from ATP synthesis. Adapted from Teixeira de Mattos et al. (1984).

(not shown). Although now unconstrained by the availability of nutrients, no increase in the growth rate was observed during the first 20 min. It was found that at high dihydroxy-acetone-P concentrations, D-lactate formation could have taken place via the MGO bypass. Assuming that this is the case, it could be calculated that the increased qglucose in the transient state did not at all affect the qATP, that is, the additional qglucose was fully uncoupled from ATP synthesis. (Note that lactate formation via MGO consumes ATP).

3.3. Futile cycles

Fig. 4. The methylglyoxal bypass. Glucose is broken down glycolytically to glyceraldehyde-3-phosphate (GAP) and dihydroxyacetone-phosphate (DHAP). DHAP is converted to methylglyoxal and subsequently to D-lactate. The net result is hydrolysis of 1 ATP per lactate formed.

Finally, we will consider the bioenergetic consequence of the simultaneous presence of a low and a high affinity uptake system. It has been proposed that the high rate of energy consumption by potassium-limited cultures of many microbes, including K. pneumoniae and E. coli (Table 1), is only partly due to the elevated energetic demands imposed by the high affinity Kdp system (see above). A significant amount of energy consumption is supposedly involved in two additional mechanisms. Under low potassium conditions the high activity of the Kdp system will result in an

124

M.J. Teixeira de Mattos, O.M. Neijssel / Journal of Biotechnology 59 (1997) 117–126

enormous K + gradient (a few mM outside and 150 mM or more inside). It has been postulated that the low-affinity uptake system (Trk) is not able to withstand such a gradient and will start to work in reverse: it leaks K + ions to the extracellular fluids (Mulder et al., 1986). Since it is likely that this downhill K + efflux generates less energy (if any) per K + than has been spent in the uptake of a K + ion, a futile cycle of K + ions across the cell membrane has been created. Later experiments (Buurman et al., 1991) pointed to a second futile cycle that is quantitatively even more important than the one described above. Thus, it was found that the presence of NH4+ ions in the culture medium led to a high rate of energy dissipation. A more detailed analysis showed that this was caused by the uptake of NH4+ via the Kdp system. The complication is that the ammonia molecule (NH3) will diffuse freely across the cell membrane, which implies that under steady state conditions the intracellular and extracellular NH3 will be almost equal (there will be a small gradient due to the assimilation of NH3), and the gradient of NH4+ ions is determined by the pH gradient across the cell membrane. Active transport of NH4+ into the cell should result in the subsequent dissociation of NH4+ ions into NH3 molecules and protons in the cytoplasm. This leads to a slightly increased intracellular NH3 concentration, which will lead to NH3 efflux (Neijssel et al., 1990). The release of protons will lead to an acidification of the cytoplasm and the cells will presumably compensate this by (energylinked) proton extrusion. A schematic presentation of the cycle is given in Fig. 6.

prokaryotes in biotechnology, but necessitates detailed knowledge of the mechanisms underlying its regulation on the other. In this context, yield values are important growth parameters as they provide us with information on the metabolic demands of an organism growing under a particular condition. Of particular interest is the Yenergy source value with respect to the energy source as it is the resultant of the energetic costs of biosynthesis on the one hand and of the efficiency of energy conservation on the other. From the above examples it can be deduced that the Yenergy source is highly variable not so much as a consequence of variation in the energetics of biosynthesis but rather as a consequence of changes in the efficiency of energy conservation. It seems justified to conclude that adaptation to low nutrient conditions invariably is accompanied by a loss of efficiency. Since in large scale fermentors homogeneity is rarely achieved, one may expect the cells to adapt to local physical and chemical conditions. In addition, conditions may vary during production (e.g. fed batch cultures). This

4. Conclusion Being unicellular, with a high volume/surface ratio, microbes are exposed intensively to physical and chemical variations in their surroundings. Therefore, it is not surprising to find that highly dynamic interactions between the cell and its environment have evolved as these are a prerequisite for survival. This has resulted in an extremely versatile metabolic machinery which on the one hand may be the rationale for the application of

Fig. 6. Futile cycling of ammonium/ammonia and potassium. Both NH4+ and K + are transported by the high affinity K + transporter Kdp at the expense of ATP. K + is exported via Trk and NH4+ dissociates intracellularly, acidifying the cytosol. NH3 diffuses out of the cell. The net result is ATP hydrolysis.

M.J. Teixeira de Mattos, O.M. Neijssel / Journal of Biotechnology 59 (1997) 117–126

poses a serious problem in practice since maximization of yield values in general is an important tool to improve the feasibility of biotechnological processes. Physiological studies have greatly increased our insights in the mechanisms that underlie phenotypic adaptation but we are still far away from quantitative descriptions of their effect on microbial ‘performance’. Unless efforts are undertaken to further quantify the energetic effects of adaptation, the metabolic versatility of prokaryotes that can be so advantageous in biotechnological applications, can turn into a major disadvantage.

References Bakker, E.P., Booth, I.R., Dinnbier, U., Epstein, W., Gajewska, A., 1987. Evidence for multiple K + export systems in E. coli. J. Bacteriol. 174, 7370–7378. Booth, I.R., 1985. Regulation of cytoplasmic pH in bacteria. Microbiol. Rev. 49, 359–378. Buurman, E.T., Teixeira de Mattos, M.J., Neijssel, O.M., 1991. Futile cycling of ammonium ions via the high affinity uptake system (Kdp) of E. coli. Arch. Microbiol. 155, 391 – 395. Calhoun, M.W., Gennis, R.B., 1993. Demonstration of separate genetic loci encoding distinct membrane-bound respiratory NADH dehydrogenases in E. coli. J. Bacteriol. 175, 3013 – 3019. Calhoun, M.W., Oden, K.L., Gennis, R.B., Teixeira de Mattos, M.J., Neijssel, O.M., 1993. Energetic efficiency of E. coli: effects on mutations in components of the aerobic respiratory chain. J. Bacteriol. 175, 3020–3025. Csonka, L.N., Epstein, W., 1996. Osmoregulation. In: Neidhardt, F.C., Curtiss, R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., Reznikoff, W.S., Riley, M., Schaechter, M., Umbarger, H.E. (Eds.), E. coli and Salmonella, Cellular and Molecular Biology, 2nd ed. ASM Press, Washington, DC, pp. 1210–1224. Dawes, E.A., 1985. Starvation, survival and energy sources. In: Floodgate, M.F., Floodgate, G.D. (Eds.), Bacteria in their Natural Environments. Academic Press, London, pp. 43– 79. Dykhuizen, D.E., Hartl, D.L., 1983. Selection in chemostats. Microbiol. Rev. 47, 150–168. El-Mansi, E.M.T., Nimmo, H.G., Holms, W.H., 1986. Pyruvate metabolism and the phosphorylation state of isocitrate dehydrogenase in E. coli. J. Gen. Microbiol. 132, 797–806. Gennis, R.B., Stewart, V., 1996. Respiration. In: Neidhardt, F.C., Curtiss, R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., Reznikoff, W.S., Riley, M., Schaechter, M., Umbarger, H.E. (Eds.), E. coli and Salmonella. Cellular and Molecular Biology, 2nd ed. ASM Press, Washington, DC, pp. 217 – 261.

125

Gottschalk, G., 1986. Bacterial Metabolism, 2nd ed. Springer, New York, pp. 182 – 196. Harder, W., Dijkhuizen, L., 1983. Physiological responses to nutrient limitation. Ann. Rev. Microbiol. 37, 1 – 23. Iuchi, S., Lin, E.C.C., 1993. Adaptation of E. coli to redox environments. Mol. Microbiol. 9, 9 – 15. Jaworowski, A., Mayo, G., Shaw, D.C., Campbell, H.D., Young, I.G., 1981. Characterization of the respiratory NADH dehydrogenase of E. coli and reconstitution of NADH oxidase in ndh mutant membrane vesicles. Biochemistry 20, 3621 – 3628. Kleiner, D., 1985. Bacterial ammonium transport. FEMS Microbiol. Rev. 32, 87 – 100. Leegwater, M.P.M., Neijssel, O.M., Tempest, D.W., 1982. Aspects of microbial physiology in relation to process control. J. Chem. Technol. Biotechnol. 32, 92 – 99. Lynch, A.S., Lin, E.C.C., 1996. Responses to molecular oxygen. In: Neidhardt, F.C., Curtiss, R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., Reznikoff, W.S., Riley, M., Schaechter, M., Umbarger, H.E. (Eds.), E. coli and Salmonella, Cellular and Molecular Biology, 2nd ed. ASM Press, Washington DC, pp. 1526 – 1538. Mulder, M.M., Teixeira de Mattos, M.J., Postma, P.W., van Dam, K., 1986. Energetic consequences of multiple potassium uptake systems in E. coli. Biochim. Biophys. Acta 851, 223 – 228. Neijssel, O.M., Tempest, D.W., 1979. The physiology of metabolite overproduction. In: Bull, A.T., Ellwood, D.C., Ratledge, C. (Eds.), Microbial Technology: Curent State, Future Prospects, 29th Sym. Soc. Gen. Microbiol. Cambridge University Press, Cambridge, pp. 53 – 82. Neijssel, O.M., Teixeira de Mattos, M.J., 1994. The energetics of bacterial growth: a reassessment. Mol. Microbiol. 12, 179 – 182. Neijssel, O.M., Hardy, G.P.M.A., Lansbergen, J.C., Tempest, D.W., O’Brien, R.W., 1980. Influence of growth environment on the phospho-enol-pyruvateglucose phosphotransferase activities of E. coli and K. aerogenes: a comparative study. Arch. Microbiol. 125, 175 – 179. Neijssel, O.M., Buurman, E.T., Teixeira de Mattos, M.J., 1990. The role of futile cycles in the energetics of microbial growth. Biochim. Biophys. Acta 1018, 252 – 255. Neijssel, O.M., Teixeira de Mattos, M.J., Tempest, D.W., 1996. Growth yield and energy distribution. In: Neidhardt, F.C., Curtiss, R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., Reznikoff, W.S., Riley, M., Schaechter, M., Umbarger, H.E. (Eds.), E. coli and Salmonella, Cellular and Molecular Biology, 2nd ed. ASM Press, Washington, DC, pp. 1683 – 1692. Pirt, S.J., 1982. Maintenance energy: a general model for energy-limited and energy sufficient growth. Arch. Microbiol. 133, 300 – 302. Postma, P.W., Lengeler, J.W., Jacobson, G.R., 1996. Phosphoenolpyruvate: carbohydrate PTS. In: Neidhardt, F.C., Curtiss, R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., Reznikoff, W.S., Riley, M., Schaechter, M., Umbarger, H.E. (Eds.), E. coli and Salmonella, Cellular

M.J. Teixeira de Mattos, O.M. Neijssel / Journal of Biotechnology 59 (1997) 117–126

126

and Molecular Biology, 2nd ed. ASM Press, Washington, DC, pp. 1149 – 1174. Puustinen, A.M., Finel, M., Haltia, T., Gennis, R.B., Wikstro¨m, M., 1991. Properties of the two terminal oxidases of E. coli. Biochemistry 30, 3936–3942. Reitzer, L.J., 1996a. Sources of nitrogen and their utilization. In: Neidhardt, F.C., Curtiss, R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., Reznikoff, W.S., Riley, M., Schaechter, M., Umbarger, H.E. (Eds.), E. coli and Salmonella, Cellular and Molecular Biology, 2nd ed. ASM Press, Washington, DC, pp. 380–390. Reitzer, L.J., 1996b. Ammonia assimilation and the biosynthesis of glutamine, glutamate, aspartate, asparagine, Lalanine and D-alanine. In: Neidhardt, F.C., Curtiss, R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., Reznikoff, W.S., Riley, M., Schaechter, M., Umbarger, H.E. (Eds.), E. coli and Salmonella, Cellular and Molecular Biology, 2nd ed. ASM Press, Washington, DC, pp. 391 – 407. Russell, J.B., Cook, G.M., 1995. Energetics of bacterial growth: balance of anabolic and catabolic reactions. Microbiol. Rev. 59, 48 –62. Stewart, L.M.D., Bakker, E.P., Booth, I.R., 1985. Energy coupling to K + uptake via the Trk system in E. coli: the role of ATP. J. Gen. Microbiol. 131, 77–86. Siebers, A., Altendorf, K., 1988. The K + -translocating KdpATPase from E. coli. Eur. J. Biochem. 178, 131–140. Stouthamer, A.H., 1979. The search for correlation between theoretical and experimental growth yields. In: Quayle, R. (Ed.), International Review of Biochemistry and Mi-

.

crobial Biochemistry, vol. 21. University Park Press, Baltimore, pp. 1 – 47. Stouthamer, A.H., van Verseveld, H.W., 1985. Stoichiometry of microbial growth. In: Moo-Young, M. (Ed.), Comprehensive Biotechnology, vol. 1. Pergamon, Oxford, pp. 215 – 238. Teixeira de Mattos, M.J., Tempest, D.W., 1983. Metabolic and energetic aspects of the growth of K. aerogenes NCTC 418 on glucose in anaerobic chemostat culture. Arch. Microbiol. 134, 80 – 85. Teixeira de Mattos, M.J., Streekstra, M.J., Tempest, D.W., 1984. Metabolic uncoupling of substrate-level phosphorylation in anaerobic glucose-limited chemostat cultures of K. aerogenes NCTC 418. Arch. Microbiol. 139, 260 – 264. Tempest, D.W., Neijssel, O.M., 1984. The status of YATP and maintenance energy as biologically interpretable phenomena. Ann. Rev. Microbiol. 38, 459 – 486. Tempest, D.W., Neijssel, O.M., 1992. Physiological and energetic aspects of bacterial metabolite overproduction. FEMS Microbiol. Lett. 100, 169 – 176. Tempest, D.W., Dicks, J.W., Ellwood, D.C., 1968. Influence of growth condition on the concentration of potassium in B. subtilis var. niger and its possible relationship to cellular ribonucleic acid, teichoic acid and teichuronic acid. Biochem. J. 106, 237 – 243. Tempest, D.W., Meers, J.L., Brown, C.M., 1970. Synthesis of glutamate by A. aerogenes by a hitherto unknown route. Biochem. J. 117, 405 – 407. Walderhaug, M.O., Dosch, D.C., Epstein, W., 1987. Potassium transport in bacteria. In: Rosen, B.P., Silver, S. (Eds.), Ion Transport in Procaryotes. Academic Press, New York, pp. 84 – 130.