Department of Microbiology and Immunology, University of Melbourne, Parkville, Vic. 3052, Australia.
Efficiency of hydrogen utilization during unitrophic and mixotrophic growth of Acetobacterium woodii on hydrogen and lactate in the chemostat
Article first published online: 17 JAN 2006
FEMS Microbiology Ecology
Volume 26, Issue 4, pages 317–324, August 1998
How to Cite
Peters, V., Janssen, P.H. and Conrad, R. (1998), Efficiency of hydrogen utilization during unitrophic and mixotrophic growth of Acetobacterium woodii on hydrogen and lactate in the chemostat. FEMS Microbiology Ecology, 26: 317–324. doi: 10.1111/j.1574-6941.1998.tb00516.x
- Issue published online: 17 JAN 2006
- Article first published online: 17 JAN 2006
- Received 12 March 1998, Revised 28 May 1998, Accepted 2 June 1998
- Growth yield;
Acetobacterium woodii is able to grow chemolithoautotrophically on H2 plus CO2 or heterotrophically on lactate by forming acetate as sole product (homoacetogenesis). In order to investigate the effect of a second substrate on the utilization of H2, the bacteria were grown under substrate limitation in chemostat culture using H2/CO2 or lactate for unitrophic and H2/CO2+lactate for mixotrophic growth. The chemostat was run at different dilution rates (0.007–0.035 h−1) until steady state was reached. Substrate consumption was balanced by production of acetate and biomass (96–115% recovery). Growth yields increased with increasing dilution rates giving maximum values of 7.7, 9.6, and 9.6 g-dw bacteria per mol acetate produced for growth on H2/CO2, lactate, and H2/CO2+lactate, respectively. The maintenance coefficients (expressed as acetate production) were 0.4, 0.08 and 0.17 mmol g-dw−1 h−1, respectively. Residual concentrations of lactate were usually below the detection limit (5 μM). However, H2 partial pressures could always be analyzed and generally increased with increasing dilution rate. It is noteworthy that steady state H2 concentrations (11–20 Pa) were also detected in lactate-grown chemostats demonstrating that H2 was produced. During growth on H2/CO2 residual H2 partial pressures were much higher (50–2450 Pa, depending on dilution rate) than on lactate. Mixotrophic growth, on the other hand, resulted in intermediate H2 partial pressures (25–160 Pa, depending on dilution rate). A similar pattern of H2 partial pressures was obtained when the bacteria were grown at 25°C instead 30°C. Growth yields and H2 partial pressures were not affected by the concentration of lactate (0.1–1.0 mM) under both unitrophic and mixotrophic conditions. The H2 partial pressures at the half maximum growth rate on lactate, lactate+H2/CO2, and H2/CO2 were 16, 42, and 94 Pa, respectively. These results demonstrate that A. woodii is able to utilize H2 down to lower partial pressures when a second heterotrophic substrate is available. However, the residual H2 partial pressures were still too high to allow successful competition with H2-utilizing methanogens.
Hydrogen is an important intermediate in the anaerobic degradation of organic matter. In the absence of electron acceptors other than CO2 consumption of H2 is only possible by methanogenic archaea and homoacetogenic bacteria. Conversion of 4 H2+CO2 to CH4+2 H2O (G°′=−130.7 kJ) is thermodynamically more favorable than conversion of 4 H2+2 CO2 to acetic acid+2 H2O (G°′=−94.9 kJ) . This thermodynamic advantage seems to be the reason that methanogens are able to utilize H2 to about tenfold lower threshold concentrations than homoacetogenic bacteria [2–4]. Lower H2 concentrations are also found when chemostat cocultures of H2-producing and H2-consuming bacteria are prepared using methanogens instead of homoacetogenic bacteria as the H2-consuming partner . Whereas the H2 concentrations in homoacetogenic cultures are typically in a range of 340–620 nM (48.9–89.2 Pa at 25°C), those in methanogenic cultures are 16–65 nM (2.3–9.3 Pa) [6, 7].
Most methanogenic environments exhibit H2 concentrations that are close to the threshold of methanogenic bacteria [6, 7]. There, H2-dependent homoacetogenesis is thermodynamically not possible. For instance, in methanogenic paddy soil H2-dependent homoacetogenesis was usually below the thermodynamic threshold [8, 9] and CO2 reduction contributed only a small percentage to total acetate production [10, 11]. Nevertheless, some H2 consumption by homoacetogens seems to take place although competing methanogens are present . It is presently not understood how homoacetogenic H2 consumption can take place under these thermodynamically restrictive conditions. The situation is even more extreme in the termite hindgut where H2 consumption is dominated by homoacetogenesis although competing methanogens are also active [13, 14]. Homoacetogenesis from H2/CO2 was also found to be active in mammalian hindguts [15–18], although it seems to be less important in those individuals who contain high numbers of methanogens [19, 20].
It has been hypothesized that homoacetogenic bacteria may be able to utilize H2 more efficiently if they are growing mixotrophically with a second heterotrophic substrate . Under these conditions they would not be restricted thermodynamically even at very low H2 concentrations. Indeed, homoacetogenic bacteria are able to utilize organic substrates simultaneously with H2[21, 22]. However, the threshold concentrations of H2 under these conditions are unknown.
Chemostat experiments with combinations of growth-limiting organic substrates showed that the steady state concentration of one substrate decreased when the proportion of the second substrate was increased [23–25]. In analogy, we therefore conducted chemostat experiments with A. woodii growing on either lactate or H2/CO2 or combinations of both.
2Materials and methods
Acetobacterium woodii (DSM 1030) was obtained from the Deutsche Sammlung von Mikroorganismen und Zellkulturen, Braunschweig, Germany. The bacteria were grown on the bicarbonate-buffered sulfide-reduced medium described by Tschech and Pfennig  at pH 7.2 and 25°C. The medium contained a trace element solution, a vitamin solution and a selenite/tungstate solution . The bacteria were routinely cultured in 125 ml serum containing 50 ml mineral medium plus 5 mM sodium l-lactate under a gas phase of N2+CO2 (4:1) for heterotrophic growth, and 50 ml mineral medium under a gas phase of H2+CO2 (4:1) for chemolithoautotrophic growth. These cultures were used as inoculum for chemostat cultures.
For chemostat cultures the pH-controlled system described by Cypionka  was used, but without sulfide regulation, similar to that used by Seitz et al. . The pH was regulated at pH 7 (pH/mV-Regler M8832 N, Mostec, Liestal, Switzerland), the temperature was kept at either 30°C or 25°C using a thermo-regulated water bath. The culture vessel had a volume of 1100 ml which was almost completely filled with medium so that the gaseous headspace was only about 20 ml. The bacterial suspension was magnetically stirred at 300 rpm. All tubing and fittings were made from iso-versinic or butyl rubber. The growth medium was supplied from a reservoir (10 l) which was kept under a gas atmosphere of N2+CO2 (4:1) if lactate was the only substrate, or H2+CO2 (4:1) if H2 was a substrate. The dissolved H2 concentration in the reservoir was about 550 μM which was almost the value expected from Henry's law (525 and 580 μM at 30°C and 25°C, respectively). The dissolved substrates were then supplied together with the mineral salt medium (and with the lactate in case of mixotrophic growth) to the culture vessel by means of a peristaltic pump (Gilson Minipuls 3, Abimed, Düsseldorf, Germany) and resulted in a steady state concentration that was given by the dilution rate and the bacterial consumption. The existence of steady state for H2 was checked by frequently measuring the H2 concentration in the liquid phase as well as the H2 partial pressure in the headspace of the culture vessel and making sure that gas and liquid phase were in equilibrium according to Henry's law (±12%). Steady state was reached after 10–15 volume changes.
Gas samples (0.2 ml) were taken from the headspace of the culture vessel with gas-tight syringes (Dynatech, Baton Rouge, LA, USA) and analyzed for H2 by gas chromatography using a thermal conductivity detector or an RGD2 detector for H2 partial pressures above and below 9 Pa, respectively . Liquid samples were taken from the culture vessel with syringes and analyzed for dissolved compounds and bacterial biomass . Dissolved H2 was analyzed after extraction of the 1-ml liquid sample with 60 ml H2-free N2. Fatty acids were analyzed in membrane-filtered (regenerated cellulose, 0.2 μm; Sartorius, Göttingen, Germany) liquid samples (0.1 ml) by ion exclusion high-pressure chromatography . The detection limit for lactate, acetate and formate was 5 μM. However, formate (or other fatty acids, such as butyrate, propionate etc.) was never detected.
Growth of bacteria was followed by measuring the optical density at 400 nm. The biomass conversion factor was determined in triplicate by growing the bacteria on each substrate. The cells were harvested in the logarithmic phase, washed 3 times by centrifugation (8000×g, 15 min, 4°C) and resuspended each time in 50 mM ammonium acetate buffer (pH 5.3). The dry weight of the washed cells was then determined after drying at 105°C until constant weight was reached. Bacterial suspensions at an optical density of 1.0 had a dry weight per ml (mean±S.E.) of 110±5 mg, for bacteria grown on either lactate or H2/CO2.
The specific growth yields (Yac) were calculated from the gram dry weight (g-dw) cells produced per mole of acetate formed. The specific growth yields per mole of substrate utilized (Y) were calculated from the stoichiometric formation of acetate from either lactate (2 lactate=3 acetate) or H2 (4 H2=1 acetate). The carbon balance was calculated by assuming that 20.6 μmol acetate are required for the synthesis of 1 mg-dw cells with a composition of C4H7O3. The maximum growth rates (μmax) were determined by dilution experiments. The dilution rate (D) of the chemostat was adjusted to a value higher than μmax and the decrease of the initial bacterial biomass (X0) was followed with time, resulting in X=X0 exp(μmax−D).
Acetobacterium woodii was grown in the chemostat with limiting concentrations of either lactate or H2 or with a combination of both substrates. A typical experiment is shown in Fig. 1 using lactate as growth-limiting substrate. After adjusting the dilution rate to D= 0.017 h−1 the bacterial density, measured as optical density, and the concentration of acetate, the product of lactate fermentation, reached steady state after about 5 weeks, corresponding to approximately 15 volume changes. Although acetate was the only stoichiometric product, H2 was produced in trace amounts so that eventually a constant H2 partial pressure established at steady state.
The results of the different chemostat experiments with unitrophically and mixotrophically growing A. woodii, with different dilution rates, different concentrations of lactate in the medium reservoir, and two different temperatures (30 and 25°C) are summarized in Table 1. As expected from the chemostat theory, the steady state concentrations of the substrates (i.e. lactate, H2) increased while the steady state concentrations of the products (i.e. acetate, bacterial cells) decreased with increasing dilution (growth) rates. The conversion of substrates to products was generally well balanced (96–115% recovery of the substrates as acetate plus bacterial cells). Bacterial growth yields (Yac) were calculated for each steady state using the common product acetate as basis for the calculation (Table 1). These values were used to calculate specific acetate production rates (qac) by qac=μ/Yac. Values of qac increased linearly with the dilution rate (=μ) (Fig. 2), thus allowing the determination of the maximum growth yields (Yac max) and the maintenance coefficients (m) . These values are summarized in Table 2.
|Substrate concentration in reservoir||Dilution rate D (h−1)||Steady state concentration||Growth yield Yac (g-dw mol−1 acetate)|
|Lactate (μM)||H2 (Pa)||Acetate (μM)||Cells (mg-dw l−1)|
|Lactate, 1000 μM||0.012||64||10||1310||10.7||8.2|
|H2, 550 μM||0.012||0||38||127||0.77||6.1|
|Lactate+H2, 1000+550 μM||0.012||62||24||1425||11.4||8.0|
|Substrate||μmax (h−1)||Yac max (g-dw mol−1 acetate)||Ymax (g-dw mol−1 substrate)||m (mmol acetate g-dw−1 h−1)|
Lactate was utilized below the detection limit (5 μM) of our analytical HPLC system except for dilution rates larger than μmax/2. However, the detection limit (1 mPa) of the H2-analytical system was low enough to detect H2 under all growth conditions examined. The steady state partial pressures of H2 were lowest on lactate, highest on H2, and intermediate on mixtures of lactate+H2 (Fig. 3). This relation was also true when A. woodii was cultured at 25°C instead of 30°C, but the absolute values were lower (Table 1). This temperature effect is in agreement with the thermodynamic theory . Under mixotrophic conditions the H2 steady state partial pressure was not affected by the relative amount of lactate present in the medium reservoir (Table 1, Fig. 4).
Our results confirm and extend the observation that homoacetogenic bacteria can grow mixotrophically utilizing H2/CO2 simultaneously with an organic substrate [21, 22]. They also show that maximum growth rates were higher under mixotrophic than under unitrophic conditions and that cellular yields were additive for the two substrates resulting in the same maximum growth yield with respect to acetate produced. The growth parameters for unitrophic growth were consistent with the literature on homoacetogens, in particular Acetobacterium species [26, 31–34]. Additive growth yields have also been observed in Sporomusa termitida growing on mixtures of H2+methanol or H2+lactate .
For the first time, we demonstrate that mixotrophic conditions significantly decreased the steady state H2 partial pressures compared to growth on H2 alone. This result was not quite unexpected, since decreased thresholds or steady state concentrations by a second substrate had already been demonstrated for the mixotrophic utilization of organic substrates [23–25, 35]. Lendenmann et al.  showed that during growth of Escherichia coli on mixtures of different sugars the steady state concentration of each sugar decreased with its proportion in the sugar mixture. This principle was extended to mixtures of glucose plus 3-phenylpropionic acid , and seems to be valid in general, provided that the different substrates fulfill the same physiological functions as sources of carbon and energy and that all enzyme systems involved are fully induced and operate at non-saturated conditions [23, 25]. These prerequisites were apparently not valid in the case of A. woodii growing on mixtures of lactate plus H2/CO2. Although the H2 partial pressure was much lower in the presence than in the absence of lactate, it did not linearly increase with the proportion of H2 to the total substrate concentration (H2/(H2+lactate)), but was more or less constant over a range of at least 0.2–0.7 electrons proportionally derived from H2. A similar pattern was reported for methanol when a yeast, Kloeckera sp. 2201, was grown on mixtures of glucose plus methanol, i.e. methanol concentrations were constant for methanol proportions of <50% and then increased non-linearly [23, 36]. This pattern was explained by the observation that the methanol-degrading enzymes operated under saturated conditions when the proportion of methanol was >50%[23, 36]. In analogy, we speculate that in A. woodii some step in the metabolism of H2/CO2 to acetate became limiting when the proportion of H2 in the feed of substrates was increased from 70 to 100%.
Lactate and H2 also did not fulfill the same physiological functions. While lactate could serve as a source for energy, electrons and carbon, H2 could only serve as a source for energy and electrons. Bicarbonate (28 and 30 mM, at 30°C and 25°C, respectively) was not limiting in the growth medium. At a given dilution rate, the cell dry mass formed during mixotrophic growth was only insignificantly higher than during unitrophic growth on lactate alone. The small cell yields on H2/CO2 in the chemostat did not allow us to assess whether the growth yields on the individual substrates were additive. However, additive growth yields could be demonstrated in batch cultures (publication in preparation). Therefore, we conclude that H2 served during mixotrophic growth as a source for both energy and electrons.
It has been shown by Seitz et al.  that A. woodii requires for growth a minimum H2 concentration (H2 threshold) that is equivalent to a Gibbs free energy of −6.3 kJ mol−1 H2. This energy was required when A. woodii was growing with H2 as sole energy source. A further decrease of the H2 concentration, as observed under mixotrophic conditions, results in a decreased requirement of energy. In fact, the Gibbs free energy calculated for acetate production from H2 was only −5 kJ mol−1 H2 under mixotrophic steady state conditions (D= 0.017 h−1; 35 Pa H2; 1500 μM acetate; 0.2 bar CO2; pH 7; 30°C). Extrapolation of the H2 steady state concentrations in mixotrophic cultures to zero growth would result in about 10 Pa H2, equivalent to a Gibbs free energy of about −2 kJ mol−1 H2. Thus, the minimum energy and the equivalent H2 threshold required for homoacetogenesis from H2/CO2 seems to be much lower under mixotrophic than under unitrophic conditions.
It should be noted that unitrophic utilization of lactate resulted in the production of trace amounts of H2. The production of traces of H2 during acetate production from organic substrates has been observed before in cultures of homoacetogenic bacteria [7, 37]. This observation suggests that the catabolism of organic substrates also involves H2 metabolism to some extent. An analogous phenomenon is found in the H2 metabolism that is involved in acetoclastic methanogenesis [38, 39]. Unitrophic growth on lactate resulted in the lowest steady state H2 concentrations that were observed in the chemostat cultures of A. woodii. The steady state H2 partial pressures at the half maximum growth rate were found to be 16, 42, and 94 for unitrophic growth on lactate, mixotrophic growth on lactate+H2/CO2, and unitrophic growth on H2/CO2, respectively. Hence, although lactate increased the efficiency of H2 utilization, H2 steady state concentrations were still higher than those on lactate alone.
The H2 concentrations found in unitrophic or mixotrophic cultures of A. woodii were generally higher than the H2 concentrations that are typically found in anaerobic environments where methanogens prevail (2.3–9.3 Pa) [6, 7]. Extrapolation of the H2 steady state concentrations in mixotrophic cultures to zero growth would still result in >10 Pa H2. Hence, it is unlikely that A. woodii would mixotrophically utilize H2 to concentrations that were lower than the threshold concentrations of H2-utilizing methanogens. Although the patterns may be different for other heterotrophic substrates and other homoacetogens, mixotrophic utilization of H2 by A. woodii, and possibly by homoacetogens in general, probably does not allow them to compete successfully with methanogens for limiting H2. Therefore, other reasons must be found to explain the numerous observations (see Section 1) that H2-consuming homoacetogens persist in environments where methanogens are also present. One possible reason may be the non-homogeneous distribution of these two physiotypes at different sites of the same environment, as recently suggested for the termite hindgut [40, 41]. Under such non-homogeneous conditions, a direct competition for H2 would not take place.
We thank Dr. O. Kappler, Marburg, for technical advice during the set-up of the chemostat and Dr. T. Egli, EAWAG, Switzerland, for helpful discussion.
- 6Conrad, R. (1989) Control of methane production in terrestrial ecosystems. In: Exchange of Trace Gases between Terrestrial Ecosystems and the Atmosphere, Dahlem Konferenzen (Andreae, M.O. and Schimel, D.S., Eds.), pp. 39–58. Wiley, Chichester.
- 7Conrad, R. (1996) Anaerobic hydrogen metabolism in aquatic sediments. In: Cycling of Reduced Gases in the Hydrosphere, Mitt. Internat. Verein. Limnol., Vol. 25 (Adams, D.D., Seitzinger, S.P. and Crill, P.M., Eds.), pp. 15–24. Schweitzerbart'sche Verlagsbuchhandlung, Stuttgart.
- 14Breznak, J.A. (1994) Acetogenesis from carbon dioxide in termite guts. In: Acetogenesis (Drake, H.L., Ed.), pp. 303–330. Chapman and Hall, New York, NY.
- 37Cord-Ruwisch, R. (1987) Contribution à létude du métabolisme de H2 par les bactéries anaérobies. PhD Thesis, Université de Provence Aix-Marseille I, Marseille.