Role of the Hya Hydrogenase in Recycling of Anaerobically Produced H₂ in Salmonella enterica Serovar Typhimurium

Abstract

Double and triple uptake-type hydrogenase mutants were used to determine which hydrogenase recycles fermentatively produced hydrogen. The Δhyb Δhya and Δhyd Δhya double mutants evolved H₂ at rates similar to that of the triple mutant strain, so Hya alone oxidizes the bulk of H₂ produced during fermentation. When only Hya was present, no hydrogen production was observed in nutrient-limited medium. H₂ uptake assays showed that Hya can oxidize both exogenously added H₂ and formate hydrogen lyase-evolved H₂ anaerobically. Even after anaerobic growth, all three uptake-type hydrogenases could function in the presence of oxygen, including using O₂ as a terminal acceptor.

Due to the anticipated scarcity of fossil fuels, there has been a surge of interest in H₂ production for alternative energy means. Numerous studies have attempted to engineer H₂-producing organisms, such as photosynthetic bacteria, cyanobacteria, and Escherichia coli, to produce maximal amounts of H₂ while minimizing the H₂-oxidizing capability of the organism (4, 5, 9, 10, 19). Hydrogenase expression and activity are controlled by multiple regulatory pathways and respond to fluctuations in pH, oxygen levels, and availability of metabolites and metal cofactors (17). In addition, the presence of hydrogen uptake hydrogenases decreases the net H₂ yield even under conditions that favor H₂ production. It is therefore important to understand the interactions between H₂-oxidizing enzymes (i.e., respiratory hydrogenases) and H₂-producing enzymes.

Gene sequence analysis has revealed that many enteric bacteria contain the genes necessary for hydrogen production and oxidation. The E. coli hydrogenases have been studied extensively, while Salmonella enterica serovar Typhimurium hydrogenases have been studied to a lesser extent. Both E. coli and Salmonella serovar Typhimurium contain the hydrogen-oxidizing hydrogenases Hya and Hyb. Salmonella serovar Typhimurium also contains Hyd, which is another hydrogen-oxidizing hydrogenase (2, 13, 15). Hyc and Hyf are hydrogen-evolving hydrogenases that are present in both E. coli and Salmonella serovar Typhimurium, although it is unknown whether Hyf is functional (1).

The Salmonella serovar Typhimurium hydrogenases are important for cellular metabolism. Hyc produces H₂ in order to remove excess reductant generated during mixed-acid fermentation. Hyc and formate dehydrogenase constitute the formate hydrogen lyase (FHL) complex (16), which oxidizes formate to produce CO₂ and H₂ (12). The hyb genes in E. coli and Salmonella serovar Typhimurium are expressed at high levels under anaerobic respiration conditions, and Hyb probably contributes to energy conservation (11, 15, 20). Hyb oxidizes H₂ and generates electrons, which are passed through the electron transport chain to terminal acceptors such as fumarate. The protons generated contribute to the proton-motive force. The role of Hya is not as well characterized. Hya may be used to recycle Hyc-produced H₂, since the hya operon is expressed at high levels during fermentative growth, or it may play a role in acid stress resistance (6, 14, 20, 21). The hyb genes are expressed at high levels under aerobic conditions in Salmonella serovar Typhimurium, and Hyb may couple H₂ oxidation to O₂ reduction (20).

Redwood et al. recently examined the roles of uptake-type hydrogenases on net hydrogen production in E. coli (10). Cells were pregrown aerobically or anaerobically with formate and then allowed to ferment in anaerobic bottles. H₂ gas was collected, and other fermentation products were measured. They found that H₂ production increased by 37% in an hya hyb double mutant (compared to that in the wild type) that was grown overnight aerobically with formate. This increase in production was associated with the loss of hyb and not hya. Therefore, in E. coli, Hyb may be responsible for recycling fermentatively produced H₂.

In this study, we measured the effect of uptake-type hydrogenase mutations on H₂ production in Salmonella serovar Typhimurium. We found that the majority of H₂-recycling activity in fermenting cells was dependent on the presence of hya, and having only Hya was sufficient to prevent any detectable H₂ evolution. These results demonstrate yet another difference between H₂ metabolism in E. coli and H₂ metabolism in Salmonella serovar Typhimurium.

Bacterial strains, growth conditions, and reagents.

The strains and plasmids used in this study are listed in Table 1. Strains were maintained in Luria-Bertani (LB) broth or on LB plates with appropriate antibiotics (50 μg ampicillin ml⁻¹ and 25 μg kanamycin ml⁻¹). CR-Hyd medium (15) with glucose (0.4%) and sodium formate (20 mM) was used where indicated. Cells were grown at 37°C either anaerobically or aerobically. Anaerobic conditions were established by sparging sealed 165-ml bottles with N₂ for 15 min and then anaerobic mix (10% H₂, 5% CO₂, and 85% N₂) for 20 min. Although overnight cultures were used for assays (see Table 2), cell densities were no more than 5 × 10⁸ to 6 × 10⁸ cells/ml at harvest.

TABLE 1.

Strains and plasmids

Strain or plasmid	Genotype or description	Source or reference
S. enterica serovar Typhimurium strains
JSG210	14028s (wild type)	ATCC
ALZ7	JSG210 with Δhyb::FRTa	20
ALZ8	JSG210 with Δhyd::FRT	20
ALZ9	JSG210 with Δhya::FRT	20
ALZ36	JSG210 with Δhyb::FRT Δhyd::FRT	This work
ALZ37	JSG210 with Δhyb::FRT Δhya::FRT	This work
ALZ42	JSG210 with Δhyd::FRT Δhya::FRT	This work
ALZ43	JSG210 with Δhyb::FRT Δhyd::FRT Δhya::FRT	This work
Plasmids
pCP20	Ampr; contains flippase gene for λ Red mutagenesis	3
pKD46	Ampr; contains λ Red genes γ, β, and exo	3
pKD4	Kanr; contains kan cassette	3

^aFRT, flippase recombinase recognition target.

TABLE 2.

Hydrogenase activity in wild-type Salmonella serovar Typhimurium or in hydrogenase mutant strains containing a single uptake enzyme

Strain or genotype	Enzyme(s)	Hydrogenase activity (nmol H2/min/109 cells)a
		H2 evolution	H2 uptake with endogenous acceptorsb	H2 uptake with oxygenc
Wild type	Hya, Hyb, Hyd	<0.02d	27 ± 9	47 ± 8
Δhyb Δhyd	Hya	<0.02d	5.2 ± 2.8	18 ± 4
Δhyb Δhya	Hyd	18 ± 8	ND	5.2 ± 1.5
Δhyd Δhya	Hyb	17 ± 7	ND	14 ± 5
Triple mutant	None	14 ± 6	ND	<0.02d

^aCells were grown overnight anaerobically in bottles containing CR-HYD media with 0.4% glucose and 20 mM sodium formate. Results are the means ± standard deviations for at least three independent experiments.

^bH₂ uptake was first assayed anaerobically with overnight cultures in media, with no exogenous electron acceptor added as indicated. In this case (i.e., anaerobic assays), H₂ uptake could not be assessed (ND, not done) for three of the strains due to ample H₂ evolution.

^cH₂ uptake was assayed with low levels (final concentration, 45 to 85 μM) of oxygen added via syringe into the amperometric chamber from a 100% O₂-saturated solution. H₂ (75.4 nmol) from a 100% saturated solution in phosphate-buffered saline was injected into the chamber to initiate H₂ uptake assays.

^dThis level was deemed to be the minimum detectable level by the amperometric assay. H₂ evolution was assayed anaerobically, sometimes after a lag (see the text).

Mutant strain construction.

Mutants with deletions of the hyb, hyd, and hya genes were constructed using the lambda Red system, as reported previously (7, 21). Hydrogenase structural gene deletions made by this system were shown to be nonpolar (see reference 7), and the strains do not contain any antibiotic resistance markers. The double mutants and triple mutant were constructed using each hydrogenase single deletion mutant and a hydrogenase mutant in which the kanamycin resistance cassette replaced the hydrogenase gene (7). Phage P22Htint (J. Gunn, The Ohio State University, Columbus, OH) was used to transduce the antibiotic marker from the Δhyb::Kan, Δhyd::Kan, or Δhya::Kan strain into the hya, hyb, or hyd single mutant, making double mutant strains. The kanamycin resistance cassette was removed by transforming with pCP20 and then growing at 37°C. The triple mutant was constructed similarly by transducing the antibiotic resistance cassette from a hydrogenase single mutant strain into a hydrogenase double mutant. The double mutants and triple mutant were confirmed by PCR. The resulting strains are listed in Table 1.

Amperometric hydrogenase assays.

Wild-type, hya hyb, hyb hyd, or hya hyd double mutant, or hya hyb hyd triple mutant cells were grown overnight under anaerobic conditions in bottles containing CR-Hyd medium with glucose and formate. Cells were removed from the closed bottle with a syringe after overnight growth. The cell suspension was injected into a stirred and sealed chamber containing a Clark type probe. Hydrogen evolution was recorded by amperometry (8). In order to measure hydrogen uptake, 100 μl of H₂-saturated phosphate-buffered saline was added to the chamber and the disappearance of H₂ was recorded.

Effect of hydrogenase deletions on H₂ production.

Redwood et al. found that in E. coli, Hyb is important for recycling Hyc-produced H₂ during fermentative growth (10). Whether Hyb had the capacity for recycling additional (exogenous) H₂ was not addressed. However, it has also been hypothesized that E. coli Hya may function to oxidize H₂ during fermentation (14). Our previous study revealed that in Salmonella serovar Typhimurium, hya gene expression was maximal during fermentative growth (20). We constructed hydrogenase double and triple mutants in Salmonella serovar Typhimurium to determine which hydrogenase was important in recycling H₂ produced via fermentation in this organism.

Wild-type Salmonella serovar Typhimurium and hydrogenase double and triple mutants were grown anaerobically overnight in sealed bottles containing CR-Hyd medium with glucose and formate. This medium was designed for optimal expression and maturation of the FHL complex. Hydrogen evolution was measured by injecting cells into a sealed chamber containing a Clark-type electrode. H₂ production was observed after an approximately 5-min lag time, during which the cells respired any O₂ which may have entered the chamber. The presence of O₂ would inhibit Hyc function as part of the FHL complex (16, 18). We assume that the O₂ exposure placed this H₂-evolving enzyme into the “ready” but oxidatively inactivated state (see reference 18). Consistent with this, we observed that when sodium dithionite (50 μM) was added to the chamber to make it anaerobic, the lag time to reach maximal activities was only about 10 s (the likely time for complete dithionite mixing/oxygen removal).

H₂ production was not observed for wild-type cells (Table 2). This is undoubtedly due to the activity of the H₂-oxidizing enzymes. As Hyc produces molecular hydrogen, the uptake-type hydrogenases oxidize it, resulting in a lack of any observable H₂ uptake or evolution. Interestingly, we did not observe H₂ production in the Δhyb Δhyd mutant, either. This result suggests that Hya has sufficient H₂ uptake activity to use all of the hydrogen produced by Hyc. Alternatively, it is possible that Hya somehow inhibits FHL activity in wild-type cells exposed to H₂ so that when Hya is missing, this negative regulation is relieved. The result could be like that illustrated by the results in Table 2, i.e., the wild type does not produce H₂ but strains lacking Hya do so. However, this possibility is minimal, as when wild-type cells were grown to early exponential phase in medium (LB with glucose [see reference 20]) in which H₂ production is abundant compared to H₂ uptake reactions, both H₂ evolution and H₂ uptake could be detected. In such cells (in which some Hya is still synthesized [20]) assayed anaerobically, 34.5 nmol H₂/min/10⁹ cells was produced (average of two independent determinations), whereas when oxygen was added, H₂ uptake was observed at 19.6 nmol H₂/min/10⁹ cells. This result also shows that Hya is not at sufficient levels to recycle all fermentative H₂ production in some (common) culture conditions.

Hydrogen uptake assays indicated that Hya also has some limited capacity to use exogenous H₂ beyond what is evolved from the FHL system (Table 2). This aspect may be important within the animal tissues where colonically produced molecular hydrogen resides. The Δhyb Δhya and Δhyd Δhya double mutants produced hydrogen at rates comparable to the hydrogen production rates of the triple mutant (Table 2). Whenever hya was absent, hydrogen production was observed. This result indicates that Hyb and Hyd are not important for appreciable anaerobic recycling of H₂ during fermentative growth.

When hydrogenase assays are conducted aerobically, the Hyc-associated hydrogen evolving (FHL) system (16, 18) is inoperative (see the text above). Therefore, we tested the ability of the strains to use H₂ with O₂ present as both an inhibitor of FHL (H₂ evolution) and a terminal respiratory acceptor. Although grown anaerobically, all double mutant strains, each containing a single uptake enzyme, were capable of H₂ oxidation (Table 2, rightmost column) with O₂ present. The strain with Hya had only about three times more uptake activity with O₂ present than the same strain did via the endogenous acceptor-mediated H₂ oxidation activity. This means that the Hya hydrogenase is likely well coupled to acceptors that accumulate anaerobically, but it can in addition carry out some H₂/O₂ respiration. In strains grown anaerobically in LB-plus-glucose medium, in which H₂ evolution is abundant, the same trend shown in Table 2 occurred for the Hyb-only and Hyd-only strains; these strains evolved more (about two times more) H₂ than the Hya strain. Also, the Hyd-only strain oxidized the smallest amount of H₂ in assays conducted aerobically from these anaerobic LB medium-grown cultures, whereas the Hya-only strain was able to take up the most H₂ (of the mutant strains) in the oxygen-coupled assays.

Our previous study showed that hyb and hyd are expressed at low levels during fermentative growth with formate (six times and three times less, respectively, than at their optimal growth conditions. The optimal conditions were anaerobic plus glycerol and fumarate (for hyb) and aerobic plus glucose (for hyd) (20). However, the high rates of hydrogen production in double mutants (that contain only Hyb or Hyd) indicate that either these protein levels are too low to oxidize FHL-produced H₂ or these hydrogenases are not properly localized to couple H₂ oxidation to the FHL system. Curiously, in our previous study, hya expression was repressed when cells were grown with formate compared to expression when cells were grown with glucose alone. Hya gene expression levels were 45.7 ± 3.2 Miller units when cells were grown in CR-Hyd medium with glucose and 9.9 ± 1.6 Miller units when cells were grown with glucose and sodium formate (20). This level of hya expression must be sufficient for the recycling of H₂ under this condition. A previous study by Sawers et al. (15) used rocket immunoelectrophoresis to measure levels of Salmonella serovar Typhimurium hydrogenase isoenzyme under different growth conditions. They found that the cellular content of isoenzyme 1 (presumably Hya) was higher in cells grown with formate than in cells grown with glucose alone (15).

It has been shown that Hyb (but not Hya) is important for hydrogen uptake during fermentation in E. coli (10). In contrast, Hya is responsible for hydrogen recycling under these conditions in Salmonella serovar Typhimurium. Also, Hya can still function aerobically with exogenous H₂. This study further defines the roles of the Salmonella serovar Typhimurium uptake-type hydrogenases. Hya is used during fermentation and has also been shown to be important in acid resistance (21). Nevertheless, it has been shown that hyd is expressed at high levels during aerobic growth and while Salmonella serovar Typhimurium resides in phagocytes (20, 21). Hyd is probably used primarily for efficient O₂-coupled energy conservation/respiration at a time of external stress. On the other hand, the hyb genes are maximally expressed during anaerobic fermentation (20), yet the Hyb enzyme may not be well coupled to anaerobically accumulating acceptors or to the FHL system. The multiple hydrogenases appear to play important roles in increasing the fitness of the bacterium in order to survive a variety of fluctuating growth environments.