Gene Expression Changes Following Acute Hydrogen Sulfide (H 2 S)-induced Nasal Respiratory Epithelial Injury

Abstract

Hydrogen sulfide (H₂S) is a naturally occurring gas that is also associated with several industries. The potential for widespread human inhalation exposure to this toxic gas is a public health concern. The nasal epithelium is especially susceptible to H₂S-induced pathology. Injury to and regeneration of the nasal respiratory mucosa occurred in animals with ongoing H₂S exposure, suggesting that the regenerated respiratory epithelium under-goes an adaptive response and becomes resistant to further injury. To better understand this response, ten-week-old male Sprague-Dawley rats were exposed nose-only to either air or 200 ppm H₂S for three hours per day for one day or five consecutive days. Nasal respiratory epithelial cells at the site of injury and regeneration were laser capture microdissected, and gene expression profiles were generated at three, six, and twenty-four hours after the initial three-hour exposure and at twenty-four hours after the fifth exposure using the Affymetrix Rat Genome 230 2.0 microarray. Gene ontology enrichment analysis showed that H₂S exposure altered gene expression associated with a variety of biological processes, including cell cycle regulation, protein kinase regulation, and cytoskeletal organization and biogenesis. Surprisingly, our results did not show a significant change in cytochrome oxidase gene expression or bioenergetics.

Introduction

Hydrogen sulfide (H₂S) is a toxic gas with a characteristic rotten egg odor. Human exposure to H₂S is widespread owing to its association with a variety of industrial processes, high-intensity agricultural practices, and natural sources. It has been estimated that each year over 1,000 human exposures are sufficiently severe to warrant reporting to human poison control centers in the United States (ATSDR 1999). The toxic effects of H₂S are characteristically dose related and most notably involve the nervous, cardiovascular, and respiratory systems (ATSDR 1999). Headache, coughing, and throat irritation occur at 2.5–5 ppm, whereas collapse and death have been observed following acute H₂S exposure to 1,000 ppm (ACGIH 1991; Reiffenstein et al. 1992; Schiffman et al. 2001).

Animal studies have shown that the nose is a particularly sensitive respiratory tract target for inhaled H₂S (ACGIH 1998; ACGIH 1991; Deng 1992). The nasal cavity is lined with specialized epithelial cell populations that maintain the normal functions of the nose. These surface epithelial cells have specific roles in conducting airflow and maintaining normal nasal function (Miller 1995) and include squamous, respiratory, transitional, and olfactory epithelium. The nasal respiratory epithelium is pseudo-stratified and ciliated with mucus-secreting cells (Morgan 1995). The olfactory epithelium, located in the more caudal portion of the nasal cavity, consists of three cell types supported by a basal lamina: sustentacular or support cells; neurons, essential for the detection of odorants; and basal cells, which act as progenitor cells for the olfactory neurons (Barrow 1986). Thus, the proximity of the respiratory and olfactory epithelial cells to inspired air makes them a target for toxicant-induced damage (Shusterman 2003).

Acute, repeated inhalation exposure of rats to ≥ 80 ppm H₂S resulted in nasal pathology involving both the respiratory and olfactory epithelium (Brenneman et al. 2002). Brenneman et al. further reported that in response to a three-hour H₂S inhalation, the nasal respiratory mucosa initially underwent epithelial regeneration, consisting of replacement of the respiratory epithelium on the lateral wall of the ventral meatus by a poorly differentiated epithelium. However, morphologically normal respiratory epithelium was present after five consecutive days of H₂S exposure.

The present project was conducted to better understand the initial injury and subsequent adaptive changes that result in the rat nasal respiratory epithelium becoming resistant to further H₂S cytotoxicity. Inhibition of cytochrome oxidase is believed to be the main mechanism of toxicity for H₂S (Beauchamp et al. 1984; Deng 1992), although alternative mechanisms, including activation of ATP-activated potassium channels and alteration in cell signaling pathways, have also been postulated (Szabó 2007). The present project was also intended to improve our understanding of the mode of action of H₂S nasal toxicity. We used laser capture microdissection (LCM) to selectively acquire respiratory epithelial cells from affected sites, and we applied microarray technology to identify the gene expression changes that occurred in these cells of interest following H₂S inhalation in the Sprague-Dawley rat.

Materials and Methods

Animals

Following a two-week acclimation period, twenty-four male Sprague-Dawley rats (Charles River Laboratories, Inc., Raleigh, NC, USA) were assigned to six weight-matched exposure groups (n = 3 or 4 rats/time point). Animals were housed two to a cage in polycarbonate cages with cellulose fiber chip bedding (ALPHI-dri^™, Shepherd Specialty Papers, Kalamazoo, MI, USA). Animal rooms were ventilated with HEPA-filtered air and maintained at 18.5–21.5°C with a relative humidity of 40–70% on a twelve-hour light/dark cycle. Animals were fed a certified NIH-07 pelleted rodent chow (Zeigler Brothers, Gardners, PA) and had access to reverse-osmosis water (Hydro Picosystem and Supplies, Research Triangle Park, NC, USA) ad libitum except during exposures. The study was reviewed by the CIIT Institutional Animal Care and Use Committee and conducted under federal guidelines for the care and use of laboratory animals (NRC 1996).

Atmospheric Generation and Exposure System

Methods of H₂S generation and characterization were similar to those previously described, with few modifications (Struve et al. 2001). Briefly, gas cylinders containing 2,000 ppm H₂S were purchased from Holox Gases (Cary, NC, USA). Exposure concentrations of H₂S were generated by metering known amounts of H₂S into the clean air supply of a 52-exposure-port, Cannon-style, nose-only inhalation system (Lab Products, Maywood, NJ, USA) housed in an 8-m³ chamber. Total air flow in the nose-only units provided approximately 0.5 L/min per animal port and was operated under a slightly negative pressure. Exposure concentrations of H₂S were monitored with a calibrated gas chromatograph (Hewlett Packard Model 6890, Hewlett Packard Co., Palo Alto, CA, USA) equipped with a flame photometric detector and GS-Q (30 m ×0.53 μm) column (Alltech, Deerfield, IL, USA).

Rats were held in individual nose-only, polycarbonate tubes and exposed for three hours per day for either one day or five consecutive days to a target concentration of 200 ppm H₂S. The average analytical concentrations (± SD) of H₂S for each three-hour exposure was 205 ± 5 ppm. The average temperatures and relative humidities during the exposures ranged from 20.5°C to 21.2°C and 42% to 45%, respectively.

Euthanasia

Rats were deeply anesthetized with sodium pentobarbital (60 mg/kg IP) and exsanguinated by severing the abdominal aorta approximately three, six, or twenty-four hours after the end of the first three-hour exposure and twenty-four hours after the end of the fifth three-hour exposure. Immediately after death, the lower jaw and skin were removed from the head, and the nose was isolated for either histopathology or gene microarray.

Tissue Processing for Gene Microarray

The isolated nose was infused with Histo-Prep OCT (Fisher Scientific, Pittsburgh, PA, USA) maintained at 37°C, placed in a cryomold, and covered with OCT. The nose was then quick-frozen in an isopentane/dry ice bath and subsequently stored at −80°C.

Frozen OCT-embedded nasal tissue blocks were serially cut with a cryostat transversely perpendicular to the bridge of the nose sectioning rostrally at level 3, as defined by Méry et al. (1994). Ten-micron tissue sections were placed onto plain, uncharged microscope slides (Esco SuperFrost slides, Sigma, St. Louis, MO, USA), allowed to rapidly thaw onto the microscope slide for adhesion, and processed immediately for LCM. During the staining procedure, the slide-mounted tissue sections were maintained on dry ice and processed for LCM four slides at a time to minimize RNA degradation and ensure efficient laser capture. Frozen sections were submerged sequentially in 75% ethanol; deionized water; HistoGene Staining Solution; and 75%, 95%, and 100% ethanol for thirty seconds each, with final dehydration in xylene for five minutes (HistoGene^™ LCM Frozen Section Staining Kit, Arcturus, Mountain View, CA, USA). To avoid cross-contamination, all solutions were changed after each batch of four slides. The slides were then allowed to air dry for five minutes before proceeding to LCM. Proper RNA technique consisting of the use of clean, disposable gloves and RNase-free instruments was maintained throughout.

Tissue Processing for Histopathology

Histopathology for the twenty-four-hour and five-day time points was provided by Brenneman et al. (2002). Briefly, isolated noses were retrograde-flushed and immersion-fixed with 10% neutral buffered formalin for approximately one week. The fixed noses were then rinsed with water and decalcified for seven to fourteen days in 7.5% formic acid. After decalcification, the noses were re-rinsed with water, retrograde-flushed, and immersed in 70% ethanol until gross-trimmed. The noses were processed routinely, paraffin-embedded, and sectioned transversely perpendicular to the bridge of the nose beginning rostrally at level 3. The 5-μm tissue sections were placed onto slides, stained with hematoxylin and eosin (H&E), and examined by bright-field light microscopy. The Brenneman study (2002) did not evaluate histologic changes seen three or six hours after a single three-hour additional H₂S exposure. In these cases, frozen sections from air and H₂S-exposed rats were collected and thaw-mounted onto positively charged glass slides. Slides were then air-dried to ensure adhesion of the sections. Dried slides were briefly water washed to remove the OCT (Tissue-Tek) mounting material, stained with H&E, and examined by bright-field light microscopy.

Laser Capture Microdissection

Microdissection was carried out using a PixCell II Laser Capture Microdissection system and CapSure HS LCM caps (Arcturus Engineering, Mountain View, CA). Since a single three-hour exposure to 200 ppm H₂S results in regeneration of the respiratory mucosa localized to the ventral meatus (Brenneman et al. 2002), the respiratory epithelium lining the lateral wall of the ventral meatus (level 3) was captured by LCM (Figure 1). Nasal respiratory epithelial samples from the same animal were collected onto 2–6 caps to obtain a minimum of 1,000 LCM pulses needed for RNA microarray analysis. In general, a laser pulse setting of 75 mW of power with a duration of 2.0 ms was employed for a 7.5-μm spot size. These settings were adjusted to optimize sample collection. Once samples were captured onto a cap, the cap was immediately processed for RNA isolation. Figure 2 shows the respiratory epithelium of the ventral meatus before (A) and after LCM (B). To demonstrate that only the cells of interest were collected, a picture of the cap (where the cells of interest are captured) was also obtained (C).

RNA Isolation, Extraction, and Amplification

RNA isolation and extraction was completed using the PicoPure RNA Isolation Kit (Arcturus Engineering, Mountain View, CA, USA). Briefly, the captured cells of interest were incubated for thirty minutes at 42°C in RNA extraction buffer. The RNA extract was centrifuged (800 × g for two minutes), and the cell extract was loaded onto a preconditioned purification column. The column with the cell extract was centrifuged to ensure binding of the RNA to the column, the column was washed, and DNA was removed using Qiagen’s DNase treatment (Qiagen, Valencia, CA, USA). The isolated total cellular RNA was then eluted in a low ionic strength buffer, and the quality of the RNA was verified with standard spectrophotometric analysis and gel electrophoresis using the Agilent 2100 Bioanalyzer (Agilent, Foster City, CA, USA). The samples were kept frozen at −80°C until amplification.

Two rounds of amplification were carried out using the RiboAmp OA RNA Amplification Kit (Arcturus). Briefly, in the first round of amplification, approximately 7 ng of total cellular RNA was reverse-transcribed using a polyT-T7 primer. Double-stranded cDNA was synthesized in the second reaction and then purified using MiraCol purification columns contained within the kit. In vitro transcription was performed using T7 RNA polymerase and the aRNA was column purified. After the first round of amplification, purified aRNA was converted to double-stranded cDNA and column-purified. This double-stranded cDNA was the template for the second round of amplification with simultaneous overnight labeling using the GeneChip Expression 3′ Amplification Reagents for IVT Labeling (Affymetrix, Santa Clara, CA, USA). As a final step, the labeled aRNA was column-purified, and yield was determined by measuring the optical density (OD) of the product at A₂₆₀ and A₂₈₀. Average yield was 117 ± 15 μg frozen at −80°C until processing for microarray analysis.

Gene Expression Profiling by Microarray

Preparation of the probe and hybridization to the microarray was performed in the CIIT Gene Expression Core facility using standard Affymetrix procedures and equipment (Affymetrix, Santa Clara, CA, USA). Briefly, 15 μg of labeled aRNA was fragmented and hybridized to Affymetrix Rat Genome 230 2.0 arrays for sixteen hours at 45°C. This array is composed of more than 31,000 probe sets, analyzing over 30,000 transcripts and variants from over 28,000 well-substantiated rat genes. After hybridization, the arrays were washed with the GeneChip Fluidics Station 450 and scanned with the GeneChip Scanner 3000.

Real-time PCR

For confirmation of the microarray gene expression measurements, real-time PCR was performed using TaqMan probes (Applied Biosystems Inc., Foster City, CA, USA). Complementary DNA from the LCM-derived RNA was generated using the High Capacity cDNA Archive Kit (Applied Biosystems Inc., Foster City, CA, USA). PCR was performed according to the Applied Biosystem protocol with TaqMan Universal PCR Master Mix and the corresponding gene-specific TaqMan Gene Expression assay. TaqMan probes (Applied Biosystems, Foster City, CA, USA) were labeled with FAM (carboxyfluorescein) and were chosen for one gene (cell division cycle 2; Cdc2a; Rn00570728_m1) with increased expression (by microarray) at twenty-four and 144 hours postexposure and another that was not increased (lipidosin; Lpd; Rn00592850_m1) at these times. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Rn99999916_s1) was used for normalization. Real-time PCR measurements were performed on an ABI PRISM 7700 sequence detector with a minimum of three experimental replicates for each gene. Real-time PCR data were normalized according to the ΔΔCt method (Livak and Schmittgen 2001). Fold change was determined by comparison to air-exposed, LCM-captured nasal respiratory epithelium. A comparison of the array values with the values obtained by RT-PCR showed good agreement between the Log₂ fold change data (data not shown).

Gene Ontology Assessment

Analysis of enrichment within gene ontology (GO) categories was performed using NIH DAVID (Dennis et al. 2003). Briefly, Affymetrix probe set identifiers for the genes of interest were uploaded to the DAVID web site and analyzed based on the Affymetrix 230_2 reference list. A hypergeometric test was performed to identify GO categories with significantly enriched gene numbers. The resulting list of GO categories was refined by selecting categories containing at least five genes. There were an insufficient number of genes with decreased expression to identify GO categories with significantly reduced gene numbers.

Statistical Analyses

Expression data were preprocessed using the robust multichip average (RMA) procedure with a log base 2 (log₂) transformation (Irizarry et al. 2003). Statistical analysis of the microarray data was performed in R using the affylmGUI package (Smyth 2004, 2005). Prior to examining the effect of H₂S treatment on gene expression, the two air control groups were examined to determine if they were significantly different. A linear model was fit to the data with a contrast between the twenty-four-hour and the 144-hour air control groups. No genes were significantly altered (p > .5), and the air controls were combined into one treatment group with eight samples. To identify genes with significant changes in expression following H₂S treatment, data were analyzed using a linear model with contrasts between the combined air control group and H₂S-exposed groups at each time point. Probability values were adjusted for multiple comparisons using a false discovery rate of 5% (Reiner et al. 2003). Genes identified as statistically significant were subject to an additional filter by selecting only those genes that exhibited a > 1.5-fold change compared to controls. Normalized gene expression results for all experiments have been deposited in the NCBI Gene Expression Omnibus (Accession # GSE5349).

Results

Nasal Respiratory Epithelial Lesions following Acute H₂S Exposure

A highly localized, bilaterally symmetrical, mild respiratory epithelial injury occurred in all animals after a single three-hour exposure to 200 ppm H₂S. This lesion was localized to the lateral wall of the ventral meatus (Figure 3B), and at three hours postexposure it was characterized by infiltration with inflammatory cells (primarily neutrophils). By six hours post-exposure, there was epithelial sloughing and loss of the basal cellular structure (Figure 3C). Respiratory epithelial regeneration occurred by twenty-four hours postexposure. The regeneration consisted of replacement of the respiratory epithelium by a squamous to low cuboidal, sparsely ciliated, and poorly differentiated epithelium (Figure 3D). Complete recovery from this initial respiratory epithelial injury was noted in all animals after five consecutive days of exposure to H₂S (Figure 3E).

Gene Expression Profiling by Microarray

Figure 4 is a cluster analysis depicting significant changes in gene expression in the nasal respiratory epithelium of rats exposed to H₂S. Early (three hours) H₂S-induced changes in gene expression consistent with cellular defense/inflammation included alterations in expression of activating transcription factor 3 (Atf3), heme oxygenase-1 (Hmox1), tumor necrosis factor 12a (Tnfrsf12a), heat shock protein 1b (Hsp1b), toll-like receptor 4 (Tlr4), and initial matrix remodeling (matrix metal-lopeptidase 3 [Mmp3], annexin, cathespin, microtubule-associated protein 6 [Map6]). By twenty-four hours after the three-hour H₂S exposure, genes involved in cellular proliferation and microtubule-based movement with significantly increased expression included genes encoding for cell division cycle associated proteins, cyclins, beta 5 tubulin (Tubb5), and high mobility group protein 2 (Hmg-2). Three-hour inhalation exposure to H₂S for five consecutive days resulted in an up-regulation in gene expression of myosin, serine protease inhibitor, lipidosin, kinesin, and 15-hydroxyprostaglandin dehydrogenase (15-Pgdh).

Gene Ontology Analysis

Table 1 shows an ontological analysis of significant gene expression changes in the respiratory epithelium following H₂S exposure. Gene ontology enrichment analysis showed that H₂S exposure altered gene expression associated with a variety of biological processes including cell cycle regulation, cellular division, DNA metabolism and repair, protein kinase regulation, and cytoskeletal organization and biogenesis.

Discussion

Our laboratory has previously shown that at concentrations reported with acute accidental human poisoning, exposure to H₂S is associated with nasal cell injury in rodents (ATSDR 1999; Brenneman et al. 2000; Brenneman et al. 2002; Dorman et al. 2002). Acute, repeated exposure of rats to concentrations of 80 ppm resulted in nasal pathology characterized by H₂S ≥ respiratory epithelial metaplasia and full thickness necrosis of the olfactory mucosa (Brenneman et al. 2002). The present study used LCM to selectively acquire H₂S-injured nasal respiratory epithelium and applied microarray technology to identify the gene changes that occur with H₂S-induced nasal pathology, thereby providing a more complete assessment of tissue injury (Knezevic et al. 2001). LCM-procured cells have been used by others and examined by genomic and proteomic approaches, creating the unique ability to ascertain patterns of gene expression in specific targeted cell types (Coco et al. 2005; Jones et al. 2002). Our genomic results are in agreement with the observed pathology induced by H₂S exposure.

Nasal respiratory epithelial injury occurred in the lateral wall of the ventral meatus and was present shortly after a single three-hour exposure to 200 ppm H₂S. The initial tissue response was characterized by infiltration of the respiratory epithelium with inflammatory cells. Trevisani and coworkers (2005) showed that intratracheal instillation of the H₂S donor, sodium hydrogen sulfide (NaHS), to guinea pigs resulted in increased airway inflammation, as evidenced by increased extravasation of Evans blue dye. This inflammatory response was ameliorated by pretreatment with capsaicin and the tachykinin NK1 receptor antagonist SR140333, suggesting that this response is mediated by sensory nerve terminals. It remains unknown whether the early lesions seen in the respiratory epithelium represent a neurogenic inflammatory response.

There was a concomitant early up-regulation of expression of genes involved in defense and inflammation that tracked closely with the observed pathology. Exposure to H₂S was associated with an up-regulation of activating transcription factor (Atf3), heme oxygenase 1 (Hmox1), heat shock protein 1b (Hsp1b), and toll-like receptor 4 (Tlr4). Atf3 is a member of the activating transcription factor/cAMP-responsive element binding protein (ATF/CREB) family of transcription factors. Atf3 is induced in a variety of stressed tissues (Chen et al. 1996), apparently through a p38-mediated pathway (Lu et al. 2007). Its expression is also induced by signals that promote cell proliferation and motility. Thus, the Atf3 gene has been characterized as an early-phase gene needed for cellular adaptation to a variety of extra- and/or intracellular changes (Lu et al. 2007). In addition, the initial inflammatory respiratory tissue injury also resulted in an up-regulation of Hmox1 and Hsp1b. Heat shock protein and Hmox1 are stress response molecules that are induced by exposure to oxidative stress (Creagh et al. 2000; Keyse and Tyrrell 1989), and Hmox1 induction, in particular, is widely regarded as a major protective mechanism against oxidative tissue injury (Ryter and Choi 2005). Intraperitoneal administration of NaHS increased gene and protein expression of Hmox1 in the rat pulmonary vascular smooth muscles under hypoxic conditions (Qingyou et al. 2004). Administration of NaHS also resulted in Hmox1 up-regulation in the central nervous system (Han et al. 2006). Acute exposure to H₂S was also associated with increased expression of toll-like receptor 4 (Tlr4). Tlr4 belongs to a family of genes that code for toll-like receptors that activate intracellular signaling, resulting in the induction of a variety of effector genes and the production of inflammatory cytokines. Tlr4 has been shown to play a role in modulating responses to ozone (Kleeberger et al. 2000), residual oil fly ash (Cho et al. 2005), and butylated hydroxytoluene in mice (Bauer et al. 2005). Metallothionein expression was also increased following H₂S exposure. In addition to being a metal-binding protein, metallothionein also acts as an antioxidant by neutralizing reactive oxygen species (Bauman et al. 1991).

Interestingly, by three hours following H₂S exposure, not only was there an initial inflammatory/defense response, but this response was coupled with an increase in expression of genes involved in matrix remodeling (e.g., laminin gamma 2, microtubule-associated protein 6, fibrinogen-like 2). This response became more pronounced by twenty-four hours after the three-hour when the epithelial phenotype H₂S exposure, was characterized by replacement of the respiratory epithelium with a squamous to low-cuboidal, sparsely ciliated, and poorly differentiated epithelium.

In our experiment, at the genomic level at twenty-four hours after exposure, genes associated with cell proliferation, cell cycle control, cytoskeleton organization, and biogenesis were the most highly expressed. These findings are remarkably similar to the results of several in vitro and in vivo studies, despite the significant inherent differences between the two types of studies. A reduced redox environment, as may be created in the nasal cavity by H₂S inhalation, may act as a proliferative stimulus (Kirlin et al. 1999). Direct evidence comes from experiments conducted by Deplancke and Gaskins (2003) in which nontransformed rat intestinal epithelial cells (IEC-18) were exposed to 1 mM NaHS, a sulfide concentration sufficient to decrease the cellular redox environment. The NaHS-exposed IEC-18 cells underwent rapid proliferation as the result of enhanced cell cycle entry (Deplancke and Gaskins 2003). Microarray analysis of the NaHS-exposed IEC-18 cells also showed increased expression of genes involved in cell growth and proliferation, transcriptional regulation, and stress responses. Therefore, our findings provide a further link between observed metaplastic lesions and the genetic changes necessary for such a metaplastic tissue response.

By the end of the repeated five-day exposure regime, there was no notable nasal pathology. Our results are consistent with Brenneman et al. (2002), who suggested that the respiratory epithelium was rapidly repaired following H₂S exposure and became resistant to further damage despite ongoing exposure to H₂S. Our gene expression analysis showed that after five daily exposures to 200 ppm H₂S (three hours/day) there was an up-regulation of expression in very few genes, including those coding for myosin 5B, serine protease inhibitor B3, lipidosin, a benzodiazepine receptor, 15-hydroxyprostaglandin dehydrogenase, and c-myc binding protein. Lipidosin (also known as acyl-CoA synthetase bubblegum family member 1) plays a role in activating long-chain and very-long-chain fatty acids and influences the biosynthesis of steroid hormone precursors (Pei et al. 2003). Interestingly, rat and human colonocytes exposed in vitro to NaHS show impaired beta-oxidation of short-chain fatty acid (Babidge et al. 1998; Moore et al. 1997). Therefore, responses seen in the respiratory epithelium may represent a compensatory mechanism to overcome this bioenergetics response.

Direct inhibition of cellular enzymes is postulated to be the primary mechanism of toxicity for H₂S (Beauchamp et al. 1984; Deng 1992). In particular, the inhibition of cytochrome oxidase is believed to disrupt the electron transport chain and significantly impair oxidative metabolism, leading to anaerobic metabolism, and severely decreased ATP production with curtailed cellular energy generation. Surprisingly, our results did not show a significant change in cytochrome oxidase gene expression or bioenergetics. However, there is growing evidence that like nitric oxide and carbon monoxide, H₂S also functions as a gaseous mediator/neurotransmitter and may influence a variety of biological responses (Fiorucci et al 2005; Leffler et al. 2006; O’Sullivan 2006; Wang 2002).

In conclusion, using the combination of LCM and gene microarray technologies allowed us to selectively isolate and examine specific sites of cellular injury and repair following H₂S inhalation exposure. Furthermore, by linking genomic changes with resultant pathology, we have increased our understanding of how H₂S may cause acute injury to the nasal respiratory epithelium, allowing researchers to establish more accurate risk assessment measures for this chemical. Surprisingly, our results suggest that gene expression changes in cytochrome oxidase did not occur following H₂S exposure, suggesting that alternative mechanisms occur. The results of our study suggest that alterations in signal transduction, inflammatory/defense response, cell cycle regulation, and response to oxidative stress are more likely involved in the early pathogenesis of this lesion. Whether other irritant gases that induce similar change in the respiratory epithelium produce comparable changes in epithelial gene expression is unknown and warrants further study.