Aerobic and anaerobic oxidation of hydrogen by acidophilic bacteria



While many prokaryotic species are known to use hydrogen as an electron donor to support their growth, this trait has only previously been reported for two acidophilic bacteria, Hydrogenobaculum acidophilum (in the presence of reduced sulfur) and Acidithiobacillus (At.) ferrooxidans. To test the hypothesis that hydrogen may be utilized more widely by acidophilic bacteria, 38 strains of acidophilic bacteria, including representatives of 20 designated and four proposed species, were screened for their abilities to grow via the dissimilatory oxidation of hydrogen. Growth was demonstrated in several species of acidophiles that also use other inorganic electron donors (ferrous iron and sulfur) but in none of the obligately heterotrophic species tested. Strains of At. ferrooxidans, At. ferridurans and At. caldus, grew chemolithotrophically on hydrogen, though those of At. thiooxidans and At. ferrivorans did not. Growth was also observed with Sulfobacillus acidophilus, Sb. benefaciens and Sb. thermosulfidooxidans, though not with other iron-oxidizing Firmicutes. Similarly, Acidimicrobium ferrooxidans grew on hydrogen, closely related acidophilic actinobacteria did not. Growth yields of At. ferrooxidans and At. ferridurans grown aerobically on hydrogen (c. 1010 cells mL−1) were far greater than typically obtained using other electron donors. Several species also grew anaerobically by coupling hydrogen oxidation to the reduction of ferric iron.


Hydrogen is a widespread substrate and metabolite in both oxic and anoxic environments and can be utilized as source of energy by hydrogen oxidizers, methanogens, acetogens, and sulfate reducers (reviewed in Schwartz et al., 2013). Microorganisms that use hydrogen as an electron donor are widespread in the domains Bacteria and Archaea. The oxidation of hydrogen (Eqn. (1)) is catalyzed by the enzyme hydrogenase, which can also catalyze the reverse reaction of hydrogen production.

display math(1)

Hydrogenases contain metallo-clusters as active sites, through which they can be categorized into three major groups, [NiFe]-, [FeFe]-, and [Fe]-hydrogenases (Vignais et al., 2001; Schwartz et al., 2013). While genes coding for hydrogenases have been identified in the genomes of a number of acidophilic bacteria, their presence does not necessarily imply that these bacteria metabolize hydrogen as the enzymes might have other roles, such as production of hydrogen in conjunction with nitrogen fixation. Some nitrogen-fixing bacteria also possess uptake-hydrogenases that allow them to recover some of the energy used during nitrogenase-mediated hydrogen production (Valdés et al., 2003).

In acidic environments, hydrogen may be formed via a number of processes, including the acid dissolution of metals (e.g. in mine sites) and some minerals and is therefore a potentially widely available electron donor for acidophiles. Several species of thermo-acidophilic crenarchaeotes (Sulfolobus, Acidianus, Metallosphaera, Stygiolobus) have been reported to oxidize hydrogen (Brock et al., 1972; Segerer et al., 1986, 1991; Stetter et al., 1986; Huber et al., 1989). In contrast, only the iron/sulfur-oxidizing mesophilic acidophile Acidithiobacillus (At.) ferrooxidans and the thermo-acidophile Hydrogenobaculum (H.) acidophilum (formerly Hydrogenobacter acidophilus) have been reported to grow by oxidizing hydrogen. Drobner et al. (1990) showed that the type strain (ATCC 23270) and two other strains of At. ferrooxidans could grow aerobically on hydrogen, while Ohmura et al. (2002) found that only one of six strains of iron-oxidizing acidithiobacilli tested could grow aerobically on hydrogen, though three of these could grow by coupling the oxidation of hydrogen to the reduction of ferric iron. Ohmura et al. (2002) were not able to grow At. ferrooxidansT on hydrogen under either aerobic or anaerobic conditions. One strain (JCM 7811) appeared to couple the oxidation of hydrogen to the reduction of elemental sulfur, though cell yields (6.1 × 107 cells mL−1) were relatively small. H. acidophilum is an aerobe that requires either elemental sulfur or thiosulfate for growth on hydrogen (Shima and Suzuki, 1993).

To assess whether hydrogen oxidation coupled to growth is more widespread among acidophilic bacteria than has been previously recognized, we have devised an empirical approach for screening this metabolic trait and tested 38 strains of acidophiles, spanning a number of bacterial orders. Our results have revealed that a number of species that oxidize ferrous iron or sulfur as electron donors can also use hydrogen as sole electron donor to support their growth.

Materials and methods


The acidophilic bacteria used in this study were obtained from national culture collections or from the acidophile culture collection maintained at Bangor University (Table 1). Included were species and strains of obligate and facultative autotrophs, and obligate heterotrophs.

Table 1. Acidophilic strains used in this study (strain designations are shown in parentheses). Strains that are underlined were tested positive for hydrogen metabolism
  1. *Group IV iron-oxidizing acidithiobacilli.

Iron-oxidizing autotrophs
Leptospirillum (L.) ferrooxidansT; Hippe, 2000
L. ferrooxidans (CF12); Johnson, 1995
L. ferriphilumT, Coram & Rawlings, 2002
L. ferriphilum (MT61); Okibe et al., 2003
L. ferrodiazotrophum’, Tyson et al., 2005
Ferrovum myxofaciensT; Hedrich et al., 2011
Sulfur-oxidizing autotrophs
Acidithiobacillus (At.) thiooxidansT; Harrison, 1982
At. thiooxidans (NZ6); D. B. Johnson (unpublished data)
At. caldusT; Hallberg & Lindström, 1994
At. caldus (BRGM3); D. B. Johnson (unpublished data)
Iron/Sulfur-oxidizing autotrophs
At. ferrooxidansT; Kelly & Wood, 2000
At. ferrooxidans (DSM 9465)
At. ferriduransT; Hedrich & Johnson, 2013
At. ferridurans (CC1); Amouric et al., 2011
At. ferrivoransT; Hallberg et al., 2010
At. ferrivorans (CF27); Hallberg et al., 2010
At. ferrooxidans’ (JCM 7812)*, Amouric et al., 2011
‘At. ferrooxidans’ (Malay)*; D. B. Johnson (unpublished data)
Acidiferrobacter thiooxydansT; Hallberg et al., 2011
Iron-oxidizing facultative autotroph
Acidimicrobium ferrooxidansT; Clark & Norris, 1996
Sulfur-oxidizing facultative autotrophs
Acidiphilium (A.) acidophilumT; Guay & Silver, 1974
Thiomonas (Tm.) perometabolisT; London & Rittenberg, 1967
Tm. arsenitoxydansT; Slyemi et al., 2011
Iron/Sulfur-oxidizing facultative autotrophs
Sulfobacillus (Sb.) acidophilusT; Norris et al., 1996
Sb. benefaciensT; Johnson et al., 2008
Sb. thermosulfidooxidansT; Golovacheva & Karavaiko, 1978
Iron-oxidizing heterotrophs
Ferrimicrobium (Fm.) acidiphilumT; Johnson et al., 2009
Fm. acidiphilum (C22B); D. B. Johnson (unpublished data)
Acidithrix ferrooxidans’ (KP1); Kay et al., 2013
Firmicute sp. SLC40; Johnson et al., 2001
Firmicute sp. SLC2; Johnson et al., 2001
Iron-reducing heterotrophs
A. cryptum (SJH); Johnson & McGinness, 1991
A. rubrumT; Wichlacz et al., 1986
Acidiphilium sp. (WJ52); D. B. Johnson (unpublished data)
Acidocella aromaticaT; Jones et al., 2013
Acidocella sp. (M21); Kay et al., 2013
Acidobacterium capsulatumT; Kishimoto et al., 1991
Acidobacterium sp. (PK4); K. C. Coupland (unpublished data)

Screening protocol for the growth of acidophiles via hydrogen oxidation

The acidophilic bacteria were grown on a variety of solid media depending on their physiological characteristics (described in Johnson & Hallberg, 2007). Autotrophic iron oxidizers were grown on an overlay medium containing ferrous iron (iFeo plates), facultatively autotrophic iron/sulfur oxidizers (and At. caldus and At. thiooxidans) were cultivated on overlay medium containing ferrous iron, tetrathionate and tryptone soya broth (TSB; FeSo plates). Heterotrophic and facultatively autotrophic iron oxidizers were grown on overlay plates containing ferrous iron and TSB (Feo plates), while moderately acidophilic Thiomonas spp. were grown on higher pH (c. 3.8) plates containing thiosulfate, ferrous iron and TSB (FeTo plates). Iron-reducing heterotrophic bacteria were grown on nonoverlay plates containing 0.002% yeast extract and 0.5 mM fructose (pH 3.0).

Bacteria were streak-inoculated onto duplicate plates that were then placed in a 2.5-L sealed jar (Oxoid, UK). Hydrogen was provided using H2/CO2 gas-generating kits (Oxoid) which resulted in the atmosphere within the jars containing 43% nitrogen, 39% hydrogen, 10.5% oxygen, and 7.5% carbon dioxide. As a control, a second set of inoculated plates were placed in a jar in which the atmosphere was supplemented only with CO2 (by adding acid to sodium bicarbonate in a 20-mL bottle placed in the jar). This was to confirm that any enhanced growth of autotrophic species was due to utilization of hydrogen rather than to elevated concentrations of CO2. The jars were incubated at 37 °C (the moderate thermophiles, Sulfobacillus (Sb.) spp., At. caldus, Leptospirillum (L.) ferriphilum and Acidimicrobium (Am.) ferrooxidans) or 30 °C (all other strains, which are mesophilic) and plates were checked regularly for growth for up to 4 weeks. In cases where colony sizes were larger or morphologies different when grown in a hydrogen-enriched atmosphere than in the absence of hydrogen, the bacteria were inoculated into a minimal salts liquid medium (supplemented with 100 μM ferrous iron, and 0.01% yeast extract for the facultative autotrophs; pH 2) and re-incubated under H2/CO2- or CO2-enriched atmospheres, as before. The optical densities of cultures were compared after 1–3 weeks.

Aerobic growth of At. ferrooxidans and At. ferridurans on hydrogen, and the effect of pH on growth of At. ferrooxidans on hydrogen

Growth of At. ferriduransT on hydrogen was compared with that of At. ferrooxidansT by incubating duplicate flasks containing minimal salts/100 μM ferrous iron and incubating under H2/CO2- or CO2-enriched atmospheres. Samples were withdrawn at regular intervals to record cell numbers (using a Thoma counting chamber) and optical densities (at 600 nm). The effect of culture pH on the growth of At. ferrooxidansT on hydrogen was tested in liquid media adjusted to different pH values using sulfuric acid.

Hydrogen-coupled ferric iron respiration

Cultures of the type strains of At. ferrooxidans, At. ferridurans, Sb. thermosulfidooxidans and Sb. benefaciens grown aerobically on hydrogen were inoculated into flasks containing the liquid medium described earlier (amended with 0.01% yeast extract for Sulfobacillus spp.) and supplemented with 10 mM ferric sulfate. These were incubated under anaerobic conditions (using the AnaeroGen System; Oxoid) in the presence of hydrogen at 30 °C (Acidithiobacillus spp.) or 37 °C (Sulfobacillus spp.). Cell numbers and ferrous iron concentrations were measured after 2 weeks. A replicate experiment was set up using At. ferrooxidansT and At. ferriduransT in which samples were removed at regular intervals to determine cell numbers and ferrous iron concentrations.

Results and discussion

During the initial screening phase, bacteria were considered positive for growth via hydrogen oxidation if colonies were significantly larger (e.g. Fig 1a) or had very different colony morphologies when grown in a H2-containing than in a H2-free atmosphere. In the case of iron-oxidizing acidophiles, colonies grown on hydrogen were white or gray-colored rather than ferric iron stained (Fig. 1b). In all cases where growth on hydrogen was inferred using solid media, this trait was confirmed using liquid media. Likewise, bacteria that did not appear to grow on hydrogen on plates also did not do so in liquid media. This confirmed the validity of the plate-screening technique for rapidly assessing growth of acidophiles on hydrogen.

Figure 1.

(a) At. caldus grown in a CO2-enriched (top) and CO2/H2-enriched (bottom) atmosphere; (b) single colony of At. ferrooxidans (strain SJ22) grown on a ferrous iron-containing overlay plate under a CO2/H2-enriched atmosphere during the transition between oxidation of ferrous iron (the older ferric iron-stained colony zone) and hydrogen (the younger white-colored biomass).

Interestingly, some Acidithiobacillus spp. grew on hydrogen, while others did not, and this was not related to whether the species could or could not oxidize ferrous iron. Strains of At. caldus grew on hydrogen while those of At. thiooxidans did not (neither species oxidize iron). Likewise, two species that oxidize ferrous iron also grew on hydrogen (At. ferrooxidans and At. ferridurans), while At. ferrivorans and the currently nonclassified ‘Group IV’ iron-oxidizing acidithiobacilli (Amouric et al., 2011) did not. These data clarify the sometimes contrary results of Drobner et al. (1990) and Ohmura et al. (2002). For example, we have confirmed that At. ferrooxidansT can grow both aerobically (as described by Drobner et al., 1990) and anaerobically (via ferric iron respiration) on hydrogen, though this was not found by Ohmura et al. (2002). The variation in hydrogen utilization reported by Ohmura et al. (2002) was probably being due to using strains that are now known to be separate species of iron-oxidizing acidithiobacilli (some of which can, and others of which cannot oxidize hydrogen) but which were, at that time, all categorized as strains of At. ferrooxidans. Interestingly, the only bacterium (IFO 14262) found by Ohmura et al. (2002) to grow aerobically on hydrogen is a strain of At. ferridurans.

Positive growth on hydrogen was also obtained for the three Sulfobacillus spp., but not for the two (currently nonclassified) obligately heterotrophic iron-oxidizing Firmicutes SLC2 and SLC40. Interestingly, cells of Sulfobacillus spp. were much longer than those grown on either ferrous iron or sulfur as electron donor. Variations in cellular morphologies of Sulfobacillus spp. have also been reported by Norris et al. (1996). Differences were also apparent among the iron-oxidizing actinobacteria, with positive growth on hydrogen being found for the facultative authotroph Am. ferrooxidans but not for obligately heterotrophic Ferrimicrobium (Fm.) acidiphilum or ‘Acidithrix ferrooxidans’ KP1. None of the strains of the three Leptospirillum spp. tested were found to grow on hydrogen, confirming that ferrous iron appears to be the sole electron donor used by these acidophiles. The same was the case for the betaproteobacterium ‘Ferrovum myxofaciens’. No growth on hydrogen was observed with Acidiphilium (A.), Acidocella, Acidobacterium or Thiomonas spp. Drobner et al. (1990) had also found that the type strains of At. thiooxidans, L. ferrooxidans and A. acidiphilum (listed at that time as Thiobacillus acidophilus) did not grow on hydrogen (confirmed in the present study). They also reported negative growth on hydrogen for Thiomonas cuprina (listed as Thiobacillus cuprinus) and the halotolerant iron oxidizer Thiobacillus prosperus, neither of which were tested in the present study.

Growth rates and growth yields of hydrogen-oxidizing Acidithiobacillus spp.

The growth rates of both At. ferrooxidansT and At ferriduransT on hydrogen were smaller than those reported for growth on ferrous iron. The equivalent culture doubling times (td's, at pH 2 and 30 °C) were c. 17.5 h for At. ferrooxidansT and c. 12 h for At. ferriduransT, which compares with td values of c. 6 and 4.5 h (respectively) for cultures grown on iron. However, exponential growth on hydrogen was found to be far more protracted than growth on either ferrous iron or sulfur (data not shown), and consequently, cell yields (which exceeded 7 × 109 cells mL−1) were far greater (by up to two orders of magnitude) of those grown on ferrous iron or sulfur. The cell yields achieved in the current work were also far greater than those noted by both Drobner et al. (1990) and Ohmura et al. (2002), possibly due to greater availability of electron donor or acceptor, or carbon dioxide, using the protocols described in this report.

Effect of pH on the growth of At. ferrooxidansT on hydrogen

In contrast to chemolithotrophic growth on ferrous iron or sulfur, the pH of batch cultures of both At. ferrooxidansT and At. ferriduransT remained stable (±0.1 pH unit) during growth on hydrogen. Growth of At. ferrooxidansT was optimum at c. pH 2 (Fig. 2), and this acidophile also grew well on hydrogen at pH 1.3 (similar to growth on ferrous iron) but did not grow on hydrogen at pH 4. These data contrast greatly with those of Drobner et al. (1990) who reported that strains of At. ferrooxidans grew on ferrous iron or sulfur between pH 1 and 6, but on hydrogen only between pH 2.5 and 6 (with a ‘broad optimum’ between pH 3.0 and 5.8).

Figure 2.

Aerobic growth of At. ferrooxidansT on hydrogen at different pH values: (▲) pH 1.3; (●) pH 2.0; (▼) pH 3.0; (■) pH 4.0.

Hydrogen-coupled ferric iron reduction

Acidithiobacillus ferrooxidans, At. ferridurans, Sb. thermosulfidooxidans and Sb. benefaciens all grew anaerobically using hydrogen as electron donor and ferric iron as electron acceptor. Reduction of ferric iron was coupled to the growth of both At. ferrooxidansT and At. ferriduransT using hydrogen as electron donor (Fig. 3). Culture doubling times of both species were similar to those found with aerobic growth on hydrogen (with At. ferriduransT again growing slightly faster than At. ferriduransT) though growth yields were far less, probably due to the more limited availability of the electron donor (ferric iron).

Figure 3.

Anaerobic growth of At. ferrooxidansT and At. ferriduransT by ferric iron respiration, using hydrogen as electron donor. Light-shaded bars, cell numbers At. ferrooxidansT; dark-shaded bars, cell numbers At. ferriduransT; ▲, ferrous iron At. ferrooxidansT; ▼, ferrous iron At. ferriduransT. Data points are mean values of replicate cultures and error bars show the range.

In conclusion, the ability to grow on hydrogen as sole electron donor was demonstrated for three species of Acidithiobacillus and all three Sulfobacillus spp. tested and was also found for Am. ferrooxidans. Using hydrogen as an electron donor has several pragmatic advantages compared to other inorganic growth substrates, including avoiding ferric iron precipitation, medium acidification and the possibility of achieving far greater biomass yields. This in turn facilities some experimental work using chemolithotrophic acidophiles, such as those involving gene transfer.