Hydrogen Sulfide Increases Hypoxia-inducible Factor-1 Activity Independently of von Hippel–Lindau Tumor Suppressor-1 in C. elegans

In C. elegans, hydrogen sulfide (H2S) exposure results in Hif-1 stabilization. hif-1 is required for survival in H2S and constitutive HIF-1 stabilization confers resistance to H2S. H2S-induced HIF-1 reporter activity appears to be independent of VHL-1, whereas VHL-1 is required for hypoxic regulation of HIF-1 reporter activity.

Abstract

Rapid alteration of gene expression in response to environmental changes is essential for normal development and behavior. The transcription factor hypoxia-inducible factor (HIF)-1 is well known to respond to alterations in oxygen availability. In nature, low oxygen environments are often found to contain high levels of hydrogen sulfide (H₂S). Here, we show that Caenorhabditis elegans can have mutually exclusive responses to H₂S and hypoxia, both involving HIF-1. Specifically, H₂S results in HIF-1 activity throughout the hypodermis, whereas hypoxia causes HIF-1 activity in the gut as judged by a reporter for HIF-1 activity. C. elegans require hif-1 to survive in room air containing trace amounts of H₂S. Exposure to H₂S results in HIF-1 nuclear localization and transcription of HIF-1 targets. The effects of H₂S on HIF-1 reporter activity are independent of von Hippel–Lindau tumor suppressor (VHL)-1, whereas VHL-1 is required for hypoxic regulation of HIF-1 reporter activity. Because H₂S is naturally produced by animal cells, our results suggest that endogenous H₂S may influence HIF-1 activity.

INTRODUCTION

In nature, oxygen and hydrogen sulfide (H₂S) together create redox environments in which eukaryotes thrive (Fenchel and Finlay, 1995). In fiords, for example, ciliates and flagellates are most abundant at depths containing chemically reactive mixtures of oxygen and H₂S. In the terrestrial atmosphere, where oxygen is abundant, animal cells produce H₂S (Stipanuk and Beck, 1982). Although the role of endogenous H₂S is unclear, exposure to exogenous H₂S has profound physiological effects including improved outcome after myocardial infarction in mammals (Elrod et al., 2007; Simon et al., 2008) and increased life span in nematodes (Miller and Roth, 2007).

Mice exposed to trace amounts of H₂S consume 10-fold less oxygen and exhibit a corresponding reduction in basal metabolic rate (Blackstone et al., 2005). Treatment with H₂S also improves survival of mice in hypoxia (Blackstone and Roth, 2007). The response to hypoxia is coordinated by the evolutionarily conserved transcription factor hypoxia-inducible factor (HIF)-1 (Semenza, 2004). First identified as the protein responsible for the hypoxia-dependent transcription of erythropoietin (Wang and Semenza, 1993), HIF-1 activity increases as a function of decreasing oxygen levels. The activity of HIF-1 is contingent upon escaping degradation (Huang et al., 1998), nuclear localization (Wang and Semenza, 1993), and coactivator binding (Arany et al., 1996). Degradation of HIF-1 is mediated by an oxygen-dependent hydroxylation event and subsequent ubiquitin-dependent degradation (Epstein et al., 2001). In C. elegans, the enzymes regulating these events are the EGg Laying defective (EGL)-9 prolyl hydroxylase and the von Hippel–Lindau tumor suppressor (VHL)-1 ubiquitin ligase, respectively.

Here, we use C. elegans to study the influence of H₂S on HIF-1. We find that hif-1 is required when nematodes are exposed to H₂S. In addition, an elevated level of HIF-1 activity dramatically increases the maximum tolerable concentration of H₂S. We show both H₂S and hypoxia cause an increase of both HIF-1 protein concentration and nuclear localization throughout the animal. However, H₂S and hypoxia treatments are distinct, with different patterns of HIF-1 transcriptional activity and H₂S can activate HIF-1 in the absence of vhl-1.

MATERIALS AND METHODS

Strains

Wild-type C. elegans (N2 Bristol) and mutant strains were grown as described previously (Brenner, 1974) at room temperature on nematode growth medium (NGM) plates seeded with live Escherichia coli OP50 food. The following mutant strains were obtained from the C. elegans genetic stock center: CB5602, vhl-1(ok161); JT307, egl-9(sa307); ZG31 hif01(ia04); CB6088, egl-9(sa307), hif-1(ia04); and CB6090, vhl-1(ok161), hif-1(ia04). The strain ZG120 nhr-57::gfp(iaIs07) was a gift from Jo Anne Powell-Coffman (Iowa State University) (Shen et al., 2006). Strains bearing the transgene were generated using standard genetic procedures.

Atmospheric Chambers

Atmospheric chambers were constructed as described previously (Nystul and Roth, 2004). In brief, atmospheres were constructed by mixing defined compressed H₂S stock mixtures with room air at a controlled rate to produce the desired final gas concentration. All compressed gas mixtures were obtained from Airgas (Radnor, PA) and were certified standard to within 2% of the indicated concentration. All experiments used continuous flow gas mixtures hydrated by using gas wash bottles. The 5 ppm H₂S atmosphere was constructed by mixing 500 ppm H₂S (balanced with nitrogen gas) with room air. All other H₂S containing atmospheres were constructed by mixing a 5000 ppm H₂S tank with room air. The hypoxic chamber was 5000 ppm O₂ balanced with nitrogen.

Viability Test

For age-dependent viability tests, nematodes were synchronized by hypochlorite treatment and arrested as starved L1 animals in M9 solution. The larvae were released onto NGM plates seeded with OP50 food and placed into the chamber at the time indicated. Survival was scored 24 h after introduction into the atmosphere.

For viability tests of mutant animals, L4 larvae were moved from a mixed stage NGM plate onto a fresh NGM plates seeded with live OP50 E. coli. These were moved into the H₂S containing chamber for 24 h, at which time worms were scored for survival to adulthood or death as L4 larvae. For adaptation experiments, worms were introduced into a chamber containing 50 ppm H₂S and grown for at least one generation in that atmosphere before challenge with a higher concentration of H₂S.

Antibody Production

A plasmid encoding the N terminus (amino acids 103–465) of C. elegans HIF-1 cloned into the pMal vector (New England Biolabs, Ipswich, MA) was a gift from Joanne Powell-Coffman. The plasmid was expressed in XL1 Blue cells and the recombinant protein was purified on an amylose column according to the manufacturer's instructions (New England Biolabs). The purified protein was used to immunize mice and generate the monoclonal antibody (mAb), by using established protocols (Wayner and Carter, 1987).

Immunostaining

Nematode strains were synchronized by hypochlorite treatment and grown to larval stage (L)4 on peptone-enriched NGM plates seeded with live E. coli NA22 food. Nematodes were rinsed in M9 solution until the supernatant looked clear of bacteria. The worms were placed onto unseeded NGM plates and exposed to gas. After 45-min exposure to H₂S or hypoxia, as indicated, the worms were washed off the plate, minced with a razor blade, collected onto glass slides, and frozen on dry ice as described previously (Moore et al., 1999). Fixation was in N,N-dimethylformamide at −20°C for 3.5 min and then rinsed two times with phosphate-buffered saline (PBS) with 1 mM MgCl₂ (PBSM). The specimens were ringed with a PAP pen (Thermo Fisher Scientific, Waltham, MA) and incubated in blocking solution (2% bovine serum albumin [BSA] in PBSM) for 30 min at room temperature. The slides were incubated with the conditioned media diluted 1:5 in blocking solution for 60 min at room temperature. The slides were then washed three times for 5 min in blocking solution and then incubated with Alexa Fluor 488-conjugated goat anti-mouse immunoglobulin (Ig)G (Invitrogen, Carlsbad, CA) secondary antibody diluted 1:2000 with 4′,6-diamidino-3-phenylindole dihydrochloride (DAPI) for 60 min. The slides were then washed with blocking solution four times for 10 min each. Coverslips were mounted onto the slides with 50% glycerol in PBSM with DAPI. The slides were immediately examined on a 510 confocal microscope (Carl Zeiss, Thornwood, NY).

Confocal image stacks were acquired with a 40×/1.3 PlanNeofluar oil immersion objective, on a LSM 510 microscope (Carl Zeiss). Green fluorescent protein (GFP) and Alexa 488 fluorescence was excited with the 488 nm line of an Argon laser and imaged through a 500–550 nm bandpass filter. DAPI was excited in two-photon mode with a pulsed femtosecond Chameleon laser tuned to 780 nm (Coherent, Santa Clara, CA), and detected through a 435–485 nm bandpass filter.

Western Blot Analysis

Nematodes were synchronized by hypochlorite treatment and arrested as starved L1 animals in M9 solution. The worms were counted and 10,000 animals were dispensed onto each of three 15-cm NGM plates seeded with OP50 bacteria (1 plate per treatment). When the worms reach L4, they were exposed to 45 min of gas treatment and then pelleted, brought up in sample buffer (72.5 mM Tris, pH 6.8, 1.8% SDS, 9% glycerol, 9 mM EDTA, 10% 2-Mercaptoethanol, and a trace amount of bromphenol blue), frozen in liquid nitrogen, and stored at −80 C. The time between opening the atmospheric chamber and freezing did not exceed 90 s. This was repeated on three different days to generate replicate samples. Before loading, the samples were heated to 95°C for 15 min, vortexed vigorously, heated for another 15 min, and then centrifuged to remove insoluble matter. Polyacrylamide gel electrophoresis (PAGE) and transfer was performed in an XCell SureLock Mini-Cell and XCell II Blot Module system (Invitrogen). Electrophoresis was performed NuPAGE gel (4–12% Bis-Tris; 1.0 mm × 15 well) with 3-(N-morpholino)propanesulfonic acid buffer, by using 12.5 μl of sample. After electrophoresis, the protein was transferred to nitrocellulose, stained with Ponceau S, rinsed, and incubated with block (PBS/Tween 20 [PBST] with 2% normal goat serum and 5% heat-inactivated nonfat dried milk) for 1 h. The blot was then incubated with conditioned media diluted 1:5 in block for 1 h. The blot was rinsed three times with PBST and then incubated with goat anti-mouse IgG-horseradish peroxidase (HRP) (Southern Biotechnology Associates, Birmingham, AL) diluted 1:20,000 for 1 h. The blot was then washed eight times with PBST, incubated with SuperSignal West Pico Chemiluminescent Substrate (Pierce Chemical, Rockford, IL), and immediately exposed to film (F-BX810; Phenix Research Products, Candler, NC). The membrane was then stripped by placing it in a solution of 50 mM Tris-HCl, pH 6.8, 2% SDS, and 100 mM 2-mercaptoethanol at 60°C for 30 min. The membrane was then rinsed three times and blocked for 30 min in PBST + 2% BSA. The membrane was then incubated with an anti-actin antibody (catalog no. sc-1616-R; Santa Cruz Biotechnology, Santa Cruz, CA) diluted 1:3000 in PBST + 2%BSA for 1 h with rocking. The membrane was washed three times with PBST and then incubated with a secondary antibody-HRP (catalog no. 111-035-003; Jackson ImmunoResearch Laboratories, West Grove, PA) diluted 1:5000 in PBST + 2% BSA for 30 min. This was then washed eight times with PBST and incubated with ECL Western Blotting Substrate (catalog no. 32209; Pierce Chemical, Rockford, IL) and immediately exposed to film. The film was scanned into a computer, and the bands were quantified using the program Quantity One (Bio-Rad Laboratories, Hercules, CA).

mRNA Quantification

Synchronized L4 C. elegans were exposed to air, 60 ppm H₂S, or 0.5% O₂ for 45 min. Then, 1 ml of TRIzol (Invitrogen) was added to 0.1 ml of packed worms and the worms were frozen in liquid nitrogen. RNA was prepared as described in the TRIzol instruction manual. cDNA was synthesized using a Protoscript First Strand Synthesis kit (New England Biolabs) according to the manufacturer's instructions. Quantitative polymerase chain reaction (PCR) was performed on an iCycler IQ system (Bio-Rad Laboratories) by using Platinum SYBR Green QPCR Supermix (Invitrogen) in 0.03-ml reactions. Standard curves were generated with cDNA, and data were analyzed using the Pfaffl analysis method (Pfaffl, 2001). cDNA levels were normalized to sir-2.1 cDNA, which has been shown previously not to change expression in H₂S (Miller and Roth, 2007).

Live Animal Fluorescence Microscopy

Mixed stage NGM plates were sealed in the chamber for the time indicated, after which L4 C. elegans were picked into a drop of 1 mM levamisole in M9. After 5 min, the animals were pipetted onto a pad of 2% agarose in water. The nematodes were visualized on a 510 confocal microscope (Carl Zeiss).

RESULTS

Survival of C. elegans in H₂S

We exposed C. elegans to atmospheres containing H₂S in room air for 24 h. Animals either survived, seemed healthy, and were indistinguishable from untreated animals, or they died; sickly animals were not observed. Survival was dose dependent, with wild-type worms surviving 50 ppm H₂S but not 150 ppm H₂S.

Hif-1 Mediates Survival of Worms in Hydrogen Sulfide

C. elegans has a single conserved homologue of mammalian hypoxia-inducible factor α subunit, hif-1 (Jiang et al., 2001). Animals with the hif-1(ia04) null mutation seem normal and are viable in normal culture conditions, but they display reduced embryonic and larval viability in hypoxia compared with wild-type animals (Jiang et al., 2001; Padilla et al., 2002; Nystul et al., 2003). We observed that hif-1(ia04) worms cannot survive exposure to 15 ppm H₂S (Table 1). In contrast, even at 50 ppm H₂S, wild-type worms survive with high viability and no obvious changes on growth rate or morphology (Miller and Roth, 2007). These results indicate that HIF-1 is necessary for responding appropriately to H₂S.

Table 1.

HIF-1 confers resistance to H₂S

Genotype	% animals surviving after 24 h ± SD (N)
	Room air	15 ppm H2S	50 ppm H2S	150 ppm H2S	500 ppm H2S
Wild type	100* (106)	100* (95)	99 ± 2 (105)	0* (106)	0* (59)
hif-1(ia04)	100* (93)	1 ± 1 (91)	0* (110)	0* (109)	0* (63)
vhl-1(ok161)	100* (89)	100* (86)	100* (99)	99 ± 3 (96)	2 ± 4 (45)
egl-9(sa307)	100* (94)	100* (82)	100* (96)	100* (87)	100* (47)
hif-1(ia04); vhl-1(ok161)	100* (58)	0* (58)
hif-1(ia04); egl-9(sa307)	100* (61)	2 ± 3 (62)
Wild type (H2S acclimated)	100* (49)		100* (68)	100* (47)	100* (38)

L4 worms were exposed to H2S for 24 h and scored for growth to adulthood. All surviving worms were identical to room air controls.

* No deviation was observed.

We wondered whether animals with elevated HIF-1 activity would be resistant to H₂S. HIF-1 protein is a transcription factor that is negatively regulated by EGL-9 and VHL-1 (Semenza, 2007); thus, mutations in either egl-9 or vhl-1 result in an overabundance of HIF-1. We found that animals with a null mutation either in vhl-1(ok161) or egl-9(sa307) were unaffected by 150 ppm H₂S, whereas wild-type worms are incapable of surviving (Table 1). Animals with the egl-9(sa307) tolerated up to 500 ppm, whereas vhl-1(ok161) null animals were only able to tolerate 150 ppm H₂S. To confirm that the H₂S resistance in these mutant animals is dependent on hif-1, we tested these mutant alleles in combination with hif-1 (ia04). Both the egl-9; hif-1 double mutant and the vhl-l; hif-1 double mutant are sensitive to H₂S. Thus, an increase in HIF-1 creates resistance to H₂S.

We hypothesized, given that hif-1 is required for wild-type worms to survive in H₂S, that exposure to H₂S may stabilize and increase HIF-1 activity. In this case, HIF-1 activation may allow worms to survive subsequent exposure to higher levels of H₂S. To test this idea, we raised wild-type worms in low-level H₂S (50 ppm) and then transferred them to higher concentrations (150 and 500 ppm). Consistent with our hypothesis, we found that all naïve worms died in as little as 150 ppm H₂S, whereas acclimated worms survived exposure to 150 ppm and even 500 ppm H₂S (Table 1). We conclude from the above-mentioned information that H₂S may activate HIF-1.

Hydrogen Sulfide Exposure Results in an Increase of HIF-1 Protein Levels and HIF-1 Nuclear Localization

To further explore the possibility that H₂S can activate HIF-1 in vivo, we developed a mAb specific for C. elegans HIF-1. HIF-1 protein is detected at high levels in nuclei of egl-9(sa307) and vhl-1(ok161) mutant animals and is absent in hif-1(ia04) worms (Supplemental Figure 1). The large gut nuclei are most easily visible, but HIF-1 protein is accumulated in nuclei throughout the animal. This is in agreement with previous reports showing that the HIF-1 promoter region is ubiquitously active throughout the worm (Jiang et al., 2001). Western blot analysis of total protein detects two bands (molecular masses 106 and 113 kDa) present in extract prepared from egl-9(sa307) but absent in hif-1(ia04) worm extract, indicating that this antibody is specific for HIF-1 (Supplemental Figure 2).

HIF-1 protein levels increase during exposure to H₂S. As with hypoxia, H₂S exposure results in increased protein concentration in as little as 45 min (Figure 1). We also examined whether H₂S treatment alters the cytological location of HIF-1. Immunostaining reveals HIF-1 accumulation in the nuclei throughout the H₂S-treated worms, compared with untreated controls (Figure 2). We also exposed nematodes to hypoxia and determined that this causes a pattern of nuclear localization indistinguishable from H₂S exposure.

Figure 1.

HIF-1 protein levels increase when worms are exposed to H₂S. (A) A representative Western blot analysis shows HIF-1 protein increases when worms are exposed to 50 ppm H₂S or 0.5% O₂. The blots were stripped and reprobed with an anti-actin antibody to confirm equivalent amounts of total protein. The size markers are shown in kilodaltons. (B) Average increase in signal relative to control ± SD of three independent experiments. *p < 0.05 (Student's t test) compared with air-treated animals.

Figure 2.

H₂S exposure causes HIF-1 nuclear localization. Both H₂S and hypoxia (0.5% O₂) treatment induce HIF-1 nuclear localization throughout the animal. Images are stacks of all nuclei visible with DAPI staining. Animals were stained with anti-Ce-Hif-1 antibody (a, d, and g) and DAPI (b, e, and h) to visualize nuclei. Hypodermal nuclei (tailless arrows) and intestinal nuclei (tailed arrows) are clearly observed in the merged images (c, f, and i). (A) Anterior of the worm. (B) Posterior of the worm. Bars, 50 μm.

Hydrogen Sulfide and Hypoxia Treatments Result in Different Patterns of HIF-1 Activity

To determine whether nuclear localization of HIF-1 in H₂S is associated with transcription of hif-1–responsive genes, we examined expression of nhr-57::gfp(iaIs07), which has HIF-1–dependent expression (Shen et al., 2006). After exposure to hypoxia, NHR-57::GFP was observed in the gut, as has been reported previously (Shen et al., 2006) (Figure 3). In contrast, H₂S treatment resulted in expression of NHR-57::GFP in the hypodermis but not in the gut. We tested various H₂S exposure lengths (1–3 h) to verify that this expression incongruity was unrelated to the amount of time that worms were exposed to H₂S. We found that reporter activity is transient, with maximal fluorescence experienced at 3 h of continuous exposure. We observed bright fluorescence in 100 of 100 worms exposed to H₂S, and we observed fluorescence in 0 of 100 untreated worms. We never observed gut fluorescence upon H₂S exposure, and we conclude that the pattern of HIF-1 activity in hypoxia and H₂S are mutually exclusive. In addition, hif-1(ia04); nhr-57::gfp(iaIs07) worms do not fluoresce when exposed to H₂S, showing that hif-1 is required for H₂S-induced expression.

Figure 3.

H₂S exposure increases expression of NHR-57::GFP. A midfocal plane of a fluorescent confocal photomicrograph and Nomarski micrograph. Fluorescence is not apparent in untreated L4 nhr-57::gfp(iaIs07) transgenic animals. An nhr-57::gfp(iaIs07) animal exposed to hypoxia exhibits HIF-1 activity in the in the gut but absent in the hypodermis. At 1 h into H₂S treatment, fluorescence is apparent in the hypodermis throughout the animal but undetectable in the gut. Fluorescence becomes more intense until 3 h into exposure. nhr-57::gfp(iaIs07); hif-1(ia04) animals fail to show any fluorescence, regardless of H₂S treatment. Bar, 40 μm.

To confirm that HIF-1 activity increase is not an artifact of the transgene, we assayed the relative amounts of known HIF-1 target transcripts in wild-type worms. As with the transgene, H₂S exposure increases nhr-57 mRNA levels (Figure 4A), suggesting that the transgene represents a transcriptional activation present in wild-type worms. To confirm that H₂S-dependent transcription is not limited to only one gene, we tested another known HIF-1 target, K10H10.2 (Figure 4B), and we found that it was also increased during H₂S exposure. Neither of these mRNAs are increased when assayed in a hif-1(ia04) mutant background, showing that hif-1 is required for the induction. These results show that H₂S exposure increases transcription of HIF-1 targets in wild-type C. elegans.

Figure 4.

H₂S exposure increases mRNA levels of HIF-1 target genes. Real-time reverse transcription-PCR was used to quantitate mRNA levels of HIF-1 target genes in wild-type or hif-1(ia04) worms exposed to H₂S or hypoxia. The cDNA of nhr-57 (A) or K10H10.2 (B) is shown as the average increase ± SD relative to sir-2.1 cDNA levels, which do not change during H₂S exposure.

Our previous results demonstrated that egl-9 and vhl-1 have distinct effects on survival, suggesting that they may have distinct effects on HIF-1 activation. To more critically define the relationship between vhl-1 and hif-1, we examined the ability of these mutant worms to express NHR-57::GFP in the presence of H₂S (Figure 5). As expected, egl-9(sa307); nhr-57::gfp(iaIs07) mutant worms exhibit constitutive expression in both the gut and in the hypodermis. In contrast, H₂S exposure of vhl-1(ok161); nhr-57::gfp(iaIs07) worms results in hypodermal fluorescence that is absent in untreated worms. Thus, this H₂S-dependent regulation of HIF-1 is dependent on egl-9 but independent of vhl-l. These results suggest that the activation of HIF-1 by H₂S is necessary and sufficient for survival and is distinct from hypoxia.

Figure 5.

H₂S dependent HIF-1 activity is independent of VHL-1. nhr-57::gfp(iaIs07); egl-9(sa307) exhibits intense fluorescence is apparent throughout the animal, including in the hypodermis and the gut. 100 Untreated nhr-57::gfp(iaIs07); vhl-1(ok161) animals show basal GFP expression in the gut and H₂S treatment induces hypodermal expression. These results are representative of 100 of 100 worms examined for each panel. Bar, 40 μm.

DISCUSSION

Increased H₂S or decreased oxygen both constitute a net change in the redox balance to which cells are exposed. We have shown that nematodes rapidly respond to changes in H₂S by HIF-1 activation. HIF-1 is also activated by decreased oxygen, which suggests that HIF-1 is important in coordinating the response to changes in environmental and organismal redox balance. Our results demonstrate that vhl-1 is not required for H₂S-dependent increase in HIF-1 target gene expression. Previous studies in mammals show that the hypoxic HIF-1 response does require VHL (Iliopoulos et al., 1996), suggesting that cells have multiple ways of activating the HIF responses.

Increased HIF-1 activity is positively correlated with survival of nematodes exposed to H₂S. C. elegans exposed to low levels of H₂S absolutely require hif-1, and increased levels of HIF-1 facilitate quantal survival in otherwise lethal concentrations of H₂S. Hif-1 mutants are sensitive to H₂S concentrations that have no effect on wild type (15 ppm). Furthermore, egl-9 and vhl-1 mutants have increased HIF-1 levels and can survive higher concentrations of hydrogen sulfide than wild type. Egl-9 mutant animals show elevated HIF-1 reporter activity compared with vhl-1 mutants worms, consistent with the finding that egl-9 mutant worms can tolerate higher H₂S concentrations than vhl-1 mutant worms. The increased survival of egl-9 mutant worms in H₂S is reminiscent of previous work which showed that a mutation in egl-9 protects nematodes from cyanide toxicity (Darby et al., 1999; Gallagher and Manoil, 2001). Both sulfide and cyanide disrupt cellular respiration as noncompetitive inhibitors of cytochrome oxidase (Cooper and Brown, 2008). Thus, it is plausible that HIF-1 helps overcome the toxicity by relieving the stress of impeded cytochrome oxidase activity. Indeed, HIF-1 has been shown to increase expression of genes encoding glycolytic enzymes (Chen et al., 2001; Minchenko et al., 2002; Obach et al., 2004) as well as to down-regulate tricarboxylic acid cycle entry and oxygen consumption in mammals (Papandreou et al., 2006).

The H₂S-induced expression of a reporter for HIF-1 activity does not require vhl-1 but does require egl-9. This is in contrast to the increase of HIF-1 protein levels observed in both mutant backgrounds. This suggests that H₂S may be involved in two steps of HIF-1 activation: protein stabilization and transcriptional activation. A recent article demonstrated the existence of an alternative HIF-1 regulatory pathway in C. elegans (Shen et al., 2006). This pathway does not require vhl-1 but does require egl-9 and a novel protein, regulator of hypoxia-inducible factor (rhy-1). The exact function of rhy-1 remains obscure, but perhaps it is important for responding to nonhypoxic impediments of the respiratory chain. In addition, VHL-independent, EGLN1-dependent HIF regulation has been noted in mammalian cells (Ozer et al., 2005; To and Huang, 2005); thus, it will be interesting to investigate the potential for H₂S induction of HIF-1 activity in higher organisms.

We observed that hypoxia and sulfide treatment result in different expression patterns of expression of a reporter for HIF-1 activity. One possibility for the differences observed is that the gut is more easily made hypoxic and the hypodermis is more easily exposed to H₂S. This model would predict that increasing the severity of hypoxia should increase the tissue with detectable HIF-1 activity. We were unable to find any concentrations of O₂, including anoxia, which would result in HIF-1 activity detected in more tissues. Likewise, increasing the concentration of H₂S also failed to increase HIF-1 activity in additional tissues. Also, the near perfect overlap of HIF-1 activity in hypoxia and vhl-1 null animals argues that the observed differences reflect something intrinsic to the tissue, perhaps involving additional transcription factors.

Organisms survive transitions to new environments by changing their physiology to be better suited to the new environment, referred to here as acclimation. Acclimation to nonlethal environmental changes is known to improve the ability to survive otherwise lethal exposure to the same agent. For example, acclimation of nematodes to low concentrations of salt allows them survive subsequent exposure to high salt concentrations that are lethal for naïve animals (Lamitina et al., 2004). We suggest that HIF-1 may be involved in acclimation to H₂S containing environments. Wild-type nematodes acclimated to a sub-lethal concentration of H₂S (50 ppm), are able to withstand a 10-fold increase in H₂S concentration (500 ppm) that would kill all naïve animals. This is concordant with the increase in HIF-1 activity observed and suggests that HIF-1 activity may be essential for the acclimation to H₂S.

Here, we have shown that H₂S exposure increases HIF-1 activity and increases tolerance of further H₂S exposure. This is reminiscent of investigations into hypoxia preconditioning, in which brief exposure to hypoxia increases HIF-1 activity and protects against subsequent hypoxia. We now know that animals produce H₂S in a regulated manner and that the resultant sulfide is physiologically important (Kamoun, 2004). In fact, endogenous H₂S production has been suggested to be important during hypoxia preconditioning (Bian et al., 2006). In addition, several recent studies have described a protective role of H₂S with respect to hypoxia (Blackstone and Roth, 2007), hemorrhage (Morrison et al., 2008), and reperfusion injury (Elrod et al., 2007). It has been shown that overexpression of a sulfide-generating enzyme, cystathionine gamma lyase, results in cardiac protection, suggesting a protective effect of endogenous H₂S (Elrod et al., 2007). Although all of these studies implicate H₂S as a potentially useful therapeutic, the mechanism of action remains ill defined. Other studies have shown that increased HIF-1 levels also provide protection against reperfusion injury, and the data provided here provide a possible link between these phenomena.