JPET Assistant Professor of Medicine (Clinician-Educator)

Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
QUICK SEARCH:   [advanced]


     

Journal of Pharmacology And Experimental Therapeutics Fast Forward
First published on December 23, 2005; DOI: 10.1124/jpet.105.097014

0022-3565/06/3171-76-87$20.00
JPET 317:76-87, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow All Versions of this Article:
jpet.105.097014v1
317/1/76    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dillman, J. F.
Right arrow Articles by Sabourin, C. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dillman, J. F., III
Right arrow Articles by Sabourin, C. L.

TOXICOLOGY

Microarray Analysis of Mouse Ear Tissue Exposed to Bis-(2-chloroethyl) Sulfide: Gene Expression Profiles Correlate with Treatment Efficacy and An Established Clinical EndpointFormula

James F. Dillman, III, Alison I. Hege, Christopher S. Phillips, Linda D. Orzolek, Albert J. Sylvester, Carol Bossone, Claudia Henemyre-Harris, Robyn C. Kiser, Young W. Choi, John J. Schlager1, and Carol L. Sabourin

Cell and Molecular Biology Branch (J.F.D., A.I.H., C.S.P., L.D.O., A.J.S., J.J.S.), Analytical Toxicology Division (C.B.), and Physiology and Immunology Branch (C.H.-H.), United States Army Medical Research Institute of Chemical Defense, Aberdeen Proving Ground, Maryland; and Medical Research and Evaluation Facility, Battelle Memorial Institute, Columbus, Ohio (R.C.K., Y.W.C., C.L.S.)

Received October 25, 2005; accepted December 22, 2005.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Bis-(2-chloroethyl) sulfide (sulfur mustard; SM) is a potent alkylating agent. Three treatment compounds have been shown to limit SM damage in the mouse ear vesicant model: dimercaprol, octyl homovanillamide, and indomethacin. Microarrays were used to determine gene expression profiles of biopsies taken from mouse ears after exposure to SM in the presence or absence of treatment compounds. Mouse ears were topically exposed to SM alone or were pretreated for 15 min with a treatment compound and then exposed to SM. Ear tissue was harvested 24 h after exposure for ear weight determination, the endpoint used to evaluate treatment compound efficacy. RNA extracted from the tissues was used to generate microarray probes for gene expression profiling of therapeutic responses. Principal component analysis of the gene expression data revealed partitioning of the samples based on treatment compound and SM exposure. Patterns of gene responses to the treatment compounds were indicative of exposure condition and were phenotypically anchored to ear weight. Pretreatment with indomethacin, the least effective treatment compound, produced ear weights close to those treated with SM alone. Ear weights from animals pretreated with dimercaprol or octyl homovanillamide were more closely associated with exposure to vehicle alone. Correlation coefficients between gene expression level and ear weight revealed genes involved in mediating responses to both SM exposure and treatment compounds. These data provide a basis for elucidating the mechanisms of response to SM and drug treatment and also provide a basis for developing strategies to accelerate development of effective SM medical countermeasures.

Bis-(2-chloroethyl) sulfide (sulfur mustard; SM) is a potent bifunctional alkylating agent capable of modifying and cross-linking cellular macromolecules such as DNA and protein by nucleophilic attack (Papirmeister et al., 1991Go). SM exposure can produce debilitating pulmonary, ocular, and cutaneous injuries. After cutaneous exposure to SM, there is a dose-dependent latent phase of 8 to 24 h that precedes clinical expression of tissue damage. Erythema occurs initially and is followed by vesication because of separation at the epidermal-dermal junction. This results in large fluid-filled lesions that are long-lasting and slow to heal (Papirmeister et al., 1991Go; Petrali and Oglesby-Megee, 1997Go). The formation of blisters is accompanied by a potent inflammatory response, observed as increased production of inflammatory mediators and infiltration of the exposure area by activated immune cells (Rikimaru et al., 1991Go; Tsuruta et al., 1996Go; Ricketts et al., 2000Go; Sabourin et al., 2000Go, 2002Go).

The mouse ear vesicant model was developed to rapidly screen for potential SM treatment compounds (Casillas et al., 1997Go). In this animal model, ear weight was determined as a measure of edema and inflammation and was used as a rapid, cost-effective endpoint to determine compound efficacy. Three potential antivesicant treatment compounds were identified using this screening paradigm: 2,3-dimercapto-1-propanol (dimercaprol, British anti-Lewisite), 2-(4-hydroxy-3-methoxy-phenyl)-N-octyl-ethanamide (octyl homovanillamide; OHV), and 2-[1-(4-chlorobenzoyl)-5-methoxy-2-methyl-indole-3-yl]acetic acid (indomethacin) (Casillas et al., 2000Go). Dimercaprol has been used as a treatment for exposure to the chemical warfare agent Lewisite (another vesicant) and also for heavy metal poisoning. Octyl homovanillamide is a vanilloid receptor agonist, and indomethacin is a classic nonsteroidal anti-inflammatory compound.

The mechanism of toxicity of SM is not well characterized, and previous studies have used DNA microarray technology to gain greater insight into the molecular pathways perturbed by SM exposure (Rogers et al., 2004Go; Sabourin et al., 2004aGo). Based on these studies, we hypothesized that gene expression profiling of mouse ear skin exposed to SM alone or pretreated with one of these treatment compounds would provide important insight into the mechanism of cutaneous toxicity of SM and might identify genes and biological pathways involved in the response to SM exposure.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
SM Exposure and Drug Treatments. Outbred male CD-1 mice (28–33 g; Charles River Laboratories, Portage, MI) were anesthetized with a combination of ketamine (60 mg/kg) and xylazine (12 mg/kg) administered i.p. A single 5-µl application of SM (0.08 mg; 0.5 µmol) in methylene chloride was applied topically to the inner surface of the right ear. The left ear (vehicle control) was exposed to 5 µl of methylene chloride only. For drug treatment, animals were administered 10 µl of dimercaprol (6.25 mg; 50.3 µmol), 30 µl of OHV (0.585 mg; 1.995 µmol), or 20 µl of indomethacin (1.34 mg; 3.74 µmol) in ethanol 15 min before SM challenge. Each drug compound was tested at the maximum tolerated dose. A group of animals received drug in ethanol on the right ear and an equal volume of ethanol (drug vehicle) on the left ear. A group of untreated, unexposed animals served as naive controls. At 24 h postexposure, animals were euthanized, and an 8-mm-diameter ear punch biopsy was obtained. The ear punch biopsy was weighed and immediately frozen in liquid nitrogen. In conducting the research described in this report, the investigators adhered to the Guide for the Care and Use of Laboratory Animals by the Institute of Laboratory Animal Resources, National Research Council, in accordance with the stipulations mandated for an Association for Assessment and Accreditation of Laboratory Animal Care International-accredited facility.

Microarray Procedures. All microarray experiments were performed using Affymetrix Mouse 430A oligonucleotide arrays, as described at http://www.affymetrix.com/support/technical/datasheets/mouse430_datasheet.pdf (Affymetrix, Santa Clara, CA). RNA was isolated from frozen mouse ear biopsies using the RNeasy mini kit (QIAGEN, Valencia, CA) according to the manufacturer's instructions. The quality and amount of RNA was monitored throughout processing with an Agilent bioanalyzer (Agilent Technologies, Palo Alto, CA) and a NanoDrop ND-1000 UV-Vis spectrophotometer (Nanodrop Technologies, Rockland, DE). Purified RNA was used to prepare biotinylated target RNA, with minor modifications from the manufacturer's recommendations (http://www.affymetrix.com/support/technical/manual/expression_manual.affx). In brief, 10 µg of total RNA was used to generate first-strand cDNA by using a T7-linked oligo(dT) primer. After second-strand synthesis, in vitro transcription was performed with biotinylated nucleotides (Enzo kits; Affymetrix), resulting in approximately 100-fold amplification of cRNA. The target cRNA generated from each sample was processed as per manufacturer's recommendation using an Affymetrix GeneChip instrument system (http://www.affymetrix.com/support/technical/manual/expression_manual.affx). In brief, spiked controls were added to 15 µg of fragmented cRNA before overnight hybridization using 10 µg of cRNA. Arrays were then washed and stained with streptavidin-phycoerythrin before being scanned on an Agilent GeneArray scanner. After scanning, array images were assessed by eye to confirm scanner alignment and the absence of significant bubbles or scratches on the chip surface. The 3'/5' ratios for GAPDH were between 0.8 and 1.72, and the ratios for beta-actin were between 1.07 and 2.71. BioB spike controls were found to be present on 44 of 64 chips that were scanned (75.0%; four were called marginal and not included in the calculation), with BioC, BioD, and CreX also present in increasing intensity. When scaled to a target intensity of 150 (using Affymetrix Microarray Suite 5.0 array analysis software), scaling factors for all arrays were between 0.559 and 2.243.

Microarray Data Analysis. Sample sizes for each treatment group are indicated in the figure legends. Scanned output files from each array were inspected for quality control as described above. Raw signal intensities were normalized using the robust multiarray averaging (RMA) algorithm (Irizarry et al., 2003Go). The RMA-normalized data were imported into Partek Pro 6.0 (Partek, St. Charles, MO) and analyzed by principal component analysis (PCA) to determine the significant sources of variability in the data. A correlation coefficient (r) between signal intensities and ear weight was calculated for each gene, and a p value was determined. A set of genes with r ≥ 0.90 were used to determine gene pathways and molecular networks highly correlated with ear weight (a measure of drug efficacy). Onto-Express was used to screen for significant pathways modulated by SM exposure (Khatri et al., 2002Go).

Quantitative Real-Time PCR. All quantitative real-time PCR (Q-PCR) was performed with Taq-Man PCR reagents and analyzed using the ABI 7500 Sequence Detection system (Applied Biosystems, Foster City, CA). All primers and probes used for Q-PCR analysis were designed using ABI Prism Primer Express version 2.0 (Applied Biosystems) and are listed in Supplemental Table 1. Primers and probes for each gene were optimized individually for maximum amplification efficiency. A validation experiment was performed to demonstrate that each target gene and endogenous control in a multiplex reaction maintained equal efficiencies (data not shown). Total RNA was purified as described above and DNase I treated on a purification column according to the manufacturer's protocol (QIAGEN). The reverse transcription reaction was carried out using 1 µg of total RNA (final concentration 50 ng/µl) using SuperScript II reverse transcriptase (Invitrogen, Carlsbad, CA). After completion of cDNA synthesis, all reactions were diluted to a final RNA input concentration of 5 ng/µl. For each gene analyzed, the experimental samples being tested (three biological replicates for each treatment group) were run in triplicate (three technical replicates) along with the corresponding no-template control and no-amplification control. The primer and probe pair concentrations used for each gene are as follows: GAPDH (endogenous control), 50 nM forward primer, 50 nM reverse primer, 200 nM VIC probe; Fgfr-3, 800 nM forward primer, 800 nM reverse primer, 100 nM Fam probe; Krt1-17, 300 nM forward primer, 50 nM reverse primer, 100 nM Fam probe; and L-myc, 300 nM forward primer, 300 nM reverse primer, 100 nM Fam probe. Amplification reactions were carried out using the instrument default cycle conditions. GAPDH was used as our internal reference gene to calculate the {Delta}Ct for each sample assayed. The {Delta}{Delta}Ct was then calculated based on the average {Delta}Ct of the naive control samples. The fold change in gene expression was determined as 2–{Delta}{Delta}Ct (Applied Biosystems User Bulletin #2, ABI Prism 7700 Sequence Detection system). Dixon's outlier test (extreme value test) was applied to all -fold change values for each gene investigated.


View this table:
[in this window]
[in a new window]
  		TABLE 1 Genes showing a positive correlation between increased ear weight and expression level (r ≥ 0.90)

 


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
Evaluation of Treatment Compounds in the Mouse Ear Vesicant Model. Mouse ears were exposed to SM or pretreated with test compounds before SM exposure as described under Materials and Methods. Ear weights measured from 8-mm punch biopsies revealed that SM exposure significantly increases mouse ear weight after 24 h compared with control ear weights (Fig. 1). Ear weights after pretreatment with indomethacin were decreased compared with SM alone, but they were still significantly greater than the unexposed controls. In contrast, the weights of ears pretreated with OHV or dimercaprol before SM exposure were not significantly different from the control ear weights. Treatment compounds and vehicles alone did not significantly affect ear weight 24 h after exposure.


Figure 1
View larger version (27K):
[in this window]
[in a new window]
  		Fig. 1. Pretreatment with compounds attenuated the increase in mouse ear weight induced by SM exposure. Mouse ears were pretreated with 6.25 mg of dimercaprol, 1.34 mg of indomethacin, or 0.6 mg of OHV in ethanol and exposed 15 min later to 0.08 mg of SM in methylene chloride. After 24 h, the weight of an 8-mm ear biopsy was determined for each sample. The SM-exposed group and the SM + indomethacin groups had significantly greater ear weights than all other groups (p < 0.05, based on a general linear model analysis). Sample sizes were unexposed (naive), n = 5; SM exposed, n = 5; SM + indomethacin, n = 6; SM + OHV, n = 5; SM + dimercaprol, n = 6; indomethacin alone, n = 4; OHV alone, n = 5; dimercaprol alone, n = 4; ethanol alone, n = 3; and methylene chloride alone, n = 21. The error bars represent the mean ± S.D. *, p < 0.05.

 
Gene Expression Profiling of Mouse Ears Pretreated with Candidate Compounds and Exposed to Sulfur Mustard. Treated and exposed mouse ear biopsies were processed, after obtaining ear weights, for analysis using oligonucleotide microarrays. Data from probed and scanned arrays were normalized using the RMA algorithm and then analyzed by PCA (Fig. 2, A and B). PCA reduces the complexity of high-dimensional data and simplifies the task of identifying patterns and sources of variability in a large data set. The samples are represented by the points in the three-dimensional plot. The distance between any pair of points is related to the similarity between the two observations in high-dimensional space. Samples that are near each other in the plot are similar in a large number of variables (i.e., expression level of individual genes). Conversely, samples that are far apart in the plot are different in a large number of variables. The PCA revealed that the drug-only controls, naive controls, and ethanol vehicle controls partitioned away from the SM-exposed samples and the methylene chloride controls. In addition, SM-exposed samples partitioned away from all the controls. Of the three treatment compounds evaluated, the mouse ears pretreated with indomethacin before SM exposure partitioned closest to the SM-exposed samples. In contrast, the mouse ears pretreated with OHV or dimercaprol before SM exposure partitioned closer to the methylene chloride control samples. This partitioning of the pretreated samples was also observed when looking at ear weights. The SM-exposed ears weighed more 24 h after exposure compared with controls (Fig. 1). Ears pretreated with indomethacin also showed a statistically significant increase in ear weight compared with control ear weights (Fig. 1) and partitioned closer to the SM-exposed ears. In contrast, OHV- and dimercaprol-pretreated ear weights are not significantly different 24 h after SM exposure compared with control ear weights (Fig. 1), and they partition closer to methylene chloride controls. When visualized using ellipsoids representing a Euclidean space ±2 S.D. from the mean of each group, considerable overlap between the SM-exposed samples and the indomethacin-pretreated samples was observed (Fig. 2B). The OHV- and dimercaprol-pretreated samples do not overlap with the SM-exposed samples to the same extent, but they do overlap with the methylene chloride-exposed control group (Fig. 2B).


Figure 2
View larger version (41K):
[in this window]
[in a new window]
  		Fig. 2. PCA of gene expression profiles reveals partitioning of samples based on treatment group. Mouse ears were treated as described in text, ear weights were determined for an 8-mm biopsy of each sample, and the biopsy tissue was processed for analysis with oligonucleotide microarrays. A, the gene expression values were normalized using the robust multiarray averaging algorithm and then visualized using a PCA. Each sphere represents the gene expression profile of an individual sample. Point color corresponds to treatment group, and point size indicates ear weight (three groups with weight cutoffs indicated in the legend). B, PCA showing ellipsoids encompassing an area around each treatment group corresponding to mean ± 2 S.D. of the group. The axes correspond to principal component 1 (PC1; x-axis; 41.3% of information content), PC2 (y-axis; 12.8% information content), and PC3 (z-axis; 5.5% information content).

 

Identification of Genes Whose Expression Profile Correlates with Ear Weight. The gene expression profiles were analyzed and a correlation coefficient between signal intensity (representing gene expression level) and ear weight was determined for each gene. Genes showing a positive correlation or a negative correlation with ear weight to an r ≥ 0.90 are reported in Tables 1 and 2, respectively. To determine the molecular functions and biological processes that are highly correlated with ear weight, the genes with r ≥ 0.90 were mapped to the Gene Ontology (The Gene Ontology Consortium, 2000Go) using the Web-based search engine Onto-Express (Khatri et al., 2002Go). Onto-Express translates lists of differentially regulated genes identified in high-throughput gene expression experiments into functional profiles based on the gene ontology, and a statistical significance value is calculated. Table 3 summarizes genes that represent the molecular functions most highly correlated with ear weight, and Table 4 summarizes the genes that represent the biological processes most highly correlated with ear weight. An examination of the biological processes most highly correlated with ear weight (based on genes with an r ≥ 0.90) reveals several major categories of biological processes: cell cycle regulation, inflammation, signal transduction, and cytoskeletal and cell adhesion processes. The gene expression profiles of genes that are classified in each of these biological processes are shown in Figs. 3 (cell cycle regulation), 4 (inflammation), 5 (signal transduction), and 6 (cytoskeletal and cell adhesion processes).


View this table:
[in this window]
[in a new window]
  		TABLE 2 Genes showing a negative correlation between increased ear weight and expression level (r ≥ 0.90)

 

View this table:
[in this window]
[in a new window]
  		TABLE 3 Molecular functions most highly correlated with ear weight

All genes with an r ≥ 0.90 (see Tables 1 and 2) were uploaded into OntoExpress. Gene ontology categories containing less that two genes were excluded from this table.

 

View this table:
[in this window]
[in a new window]
  		TABLE 4 Biological processes most highly correlated with ear weight

All genes with an r ≥ 0.90 (see Tables 1 and 2) were uploaded into OntoExpress. Gene ontology categories containing less that two genes were excluded from this table.

 

Figure 3
View larger version (35K):
[in this window]
[in a new window]
  		Fig. 3. Gene expression profiles of individual cell cycle control and cellular growth genes that are highly correlated with SM-induced ear weight gain. Genes with expression patterns that were highly correlated (r ≥ 0.9) with ear weight were identified by gene expression profiling. Gene ontology-based analysis was used to determine significantly altered biological processes. Individual genes were graphed according to treatment group and gene expression signal intensity. Error bars represent mean ± S.D. The data were analyzed for statistical significance using a one-way ANOVA with a Dunnett's post test to look for treatment groups significantly different from the unexposed control group. *, p < 0.05; ***, p < 0.001.

 

Figure 4
View larger version (22K):
[in this window]
[in a new window]
  		Fig. 4. Gene expression profiles of individual inflammation and immune response genes that are highly correlated with SM-induced ear weight gain. Genes with expression patterns that were highly correlated (r ≥ 0.9) with ear weight were identified by gene expression profiling. Gene ontology-based analysis was used to determine significantly altered biological processes. Individual genes were graphed according to treatment group and gene expression signal intensity. Error bars represent mean ± S.D. The data were analyzed for statistical significance using a one-way ANOVA with a Dunnett's post test to look for treatment groups significantly different from the unexposed control group. *, p < 0.05; ***, p < 0.001.

 


Figure 5
View larger version (29K):
[in this window]
[in a new window]
  		Fig. 5. Gene expression profiles of individual signal transduction genes that are highly correlated with SM-induced ear weight gain. Genes with expression patterns that were highly correlated (r ≥ 0.9) with ear weight were identified by gene expression profiling. Gene ontology-based analysis was used to determine significantly altered biological processes. Individual genes were graphed according to treatment group and gene expression signal intensity. Error bars represent mean ± S.D. The data were analyzed for statistical significance using a one-way ANOVA with a Dunnett's post test to look for treatment groups significantly different from the unexposed control group. *, p < 0.05; ***, p < 0.001.

 
Validation of Selected Microarray Results by Q-PCR. The reliability of our microarray data was confirmed using Q-PCR analysis of several genes whose expression levels were highly correlated with ear weight. As shown in Supplemental Fig. 1, the relative expression levels of Krt1-17, Fgfr3, and L-myc based on Q-PCR analysis were consistent with the expression profiles determined by microarray analysis.


   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
To gain insight into the molecular mechanism of SM toxicity, we used oligonucleotide microarrays to determine the gene expression profile of mouse ear skin exposed to SM. These results were compared with gene expression profiles determined from ears that were pretreated with one of three treatment compounds (dimercaprol, OHV, or indomethacin). Principal component analysis revealed that the gene expression profiles of dimercaprol- and OHV-pretreated tissues were similar to profiles from tissues exposed to vehicle alone (methylene chloride). In contrast, the gene expression profile of indomethacin-pretreated tissues was similar to that of tissues exposed only to SM. These profiles complement the ear weight data, which show that indomethacin is the least effective treatment compound for attenuating postexposure increases in mouse ear weight. Statistical analysis of the gene expression profiles revealed specific sets of genes that were highly correlated with SM-induced changes in ear weight (r ≥ 0.90). Gene ontology mapping of these gene sets revealed several major biological processes that were affected concomitantly with increased ear weight. These were cell cycle regulation, inflammation, signal transduction, and cytoskeletal and cell adhesion processes.


Figure 6
View larger version (43K):
[in this window]
[in a new window]
  		Fig. 6. Gene expression profiles of individual cytoskeletal and cell adhesion genes that are highly correlated with SM-induced ear weight gain. Genes with expression patterns that were highly correlated (r ≥ 0.9) with ear weight were identified by gene expression profiling. Gene ontology-based analysis was used to determine significantly altered biological processes. Individual genes were graphed according to treatment group and gene expression signal intensity. Error bars represent mean ± S.D. The data were analyzed for statistical significance using a one-way ANOVA with a Dunnett's post test to look for treatment groups significantly different from the unexposed control group. *, p < 0.05; ***, p < 0.001.

 
Previous work has examined changes in gene expression after exposure to SM. The systems studied include mouse ear (Sabourin et al., 2000Go, 2004aGo; Rogers et al., 2004Go), pig skin (Sabourin et al., 2002Go), cultured human keratinocytes (Schlager et al., 2002Go; Platteborze, 2003Go), HepG2 cells transfected with various stress gene reporter constructs (Schlager and Hart, 2000Go), and Jurkat cells exposed to the mustard-related compound chloroethyl ethyl sulfide (CEES) (Zhang et al., 2002Go). Our previous studies show that vanilloids reduce skin edema and cytokine mRNA expression after SM exposure in a mouse model (Sabourin et al., 2003Go, 2004bGo). This is the first study to characterize SM-induced changes in gene expression in relation to a characterized endpoint of SM exposure (ear weight) and also to examine the effect of treatment compounds on SM-induced gene expression using a global gene expression approach. These results provide important details that were previously lacking, and indicate that gene expression profiles may help to uncover the mechanisms of an observed phenotypic response (e.g., increased ear weight). Interestingly, few of the genes found to be highly correlated with increased ear weight showed what would normally be considered significant changes in gene expression (>2-fold change). Our results suggest that many of the genes highly correlated with a clinical endpoint may not show large -fold changes in gene expression. Further investigation with a broader time course and dose response may reveal more dynamic changes in the expression levels of these genes.

Cell cycle regulation has been previously implicated as a biological process affected by SM exposure. This has been attributed to the ability of SM to cross-link DNA by alkylation at the N-7 of guanine (Papirmeister et al., 1991Go). The cell cycle has been shown to stop at the G1/S transition after SM exposure (Smith et al., 1993Go). p53 is an important component of the G1/S cell cycle checkpoint. Rosenthal et al. (1998Go) showed that p53 accumulates after SM exposure in cultured human keratinocytes. Schlager and Hart (2000Go) showed increases in activity of the p53 promoter response element in HepG2 cells transfected with reporter constructs. We recently showed that rat pulmonary tissue exposed to SM via an intravenous route elicits a robust p53 response (Dillman et al., 2005Go). We also reported that p53 is phosphorylated on serine 15 within 15 min of SM exposure in cultured human keratinocytes (Minsavage and Dillman, 2005Go). Interestingly, it has been shown that there is a high incidence of lung cancer in former mustard gas workers, and there are p53 mutations in many of these cancers (Manning et al., 1981Go; Tokuoka et al., 1986Go; Easton et al., 1988Go; Nishimoto et al., 1988Go; Takeshima et al., 1994Go). The relationship between p53 activation, the cell cycle, and the genes identified in SM-exposed mouse skin and rat lung remains to be determined.

Up-regulation of cytokines and chemokines after SM exposure has been well characterized in a number of recent studies (Arroyo et al., 1999Go; Lardot et al., 1999Go; Sabourin et al., 2000Go, 2002Go, 2003Go, 2004aGo). Related to this, a number of cell signaling pathways have been identified that are activated or altered after SM exposure. Pharmacological inhibition of the p38 mitogen-activated protein kinase (MAPK14) has been reported to block up-regulation of cytokine production in response to SM exposure in cultured keratinocytes (Dillman et al., 2004Go). In addition the transcription factor nuclear factor-{kappa}B, which is involved in both inflammation and apoptosis, is activated after SM exposure in cultured cells (Atkins et al., 2000Go; Schlager and Hart, 2000Go). In other systems, nuclear factor-{kappa}B has been shown to be activated by CEES in guinea pig lung (Chatterjee et al., 2003Go), and the Akt pathway is perturbed by CEES exposure in Jurkat cells (Zhang et al., 2002Go). It is not clear how CEES exposure relates to SM exposure, but it is clear that signaling pathways associated with inflammatory response are activated by SM exposure. These signaling pathways may be potential therapeutic targets for attenuating cytokine production and the subsequent recruitment of activated immune cells to the site of injury.

The cytoskeleton is an important target of SM exposure. Several studies have identified keratin proteins as direct targets of SM alkylation (van der Schans et al., 2002Go; Dillman et al., 2003Go). In addition, cross-linking of keratin filaments by SM, but not by other alkylating agents (e.g., CEES), has been observed (Dillman et al., 2003Go). This alkylation results in collapse of the keratin intermediate filament network and subsequent adverse changes in cellular morphology (Werrlein and Madren-Whalley, 2000Go, 2003Go). Furthermore, exposure to SM results in changes in the keratin proteins that are expressed in cultured human keratinocytes (Rosenthal et al., 1998Go). Western blotting of keratinocytes exposed to SM revealed that keratin-5 (K5) and keratin-14 (K14) protein levels decrease, and keratin-1 (K1) and keratin-10 (K10) protein levels increase. K5/K14 are associated with proliferating cells in the basal layer of the epidermis, whereas K1/K10 are associated with terminally differentiating cells in the suprabasal layers of the epidermis. SM exposure also results in perturbation of the actin cytoskeleton (Werrlein et al., 2005Go). Our data regarding altered gene expression are in agreement with these observations that cytoskeletal proteins are adversely affected by SM exposure. Although direct effects of SM on the microtubule network have not been reported to date, we observe changes in the mRNA levels of microtubule-based motor proteins from the kinesin family (Fig. 6), suggesting that changes in microtubule-based motility may result from SM exposure.

In conclusion, our results demonstrate that gene expression profiling can provide additional insight into the pathways important in SM-induced injury and which pathways may be potential targets for future development of antivesicant therapeutics. Our results suggest that gene expression profiling can be predictive of drug efficacy and imply that this could be an important tool in predictive toxicology and drug development. These approaches are of particular importance in the discovery and testing of pharmacological treatments for toxicant exposure, since in these cases traditional clinical trials for regulatory approval are not an option.


   Acknowledgements
 
We thank Robyn Lee for statistical consultation, Rich Sweeney for help with data management, and Gary Minsavage and Robert Werrlein for critical reading of the manuscript.


   Footnotes
 
In conducting the research described in this report, the investigators adhered to the Guide for the Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council. The data discussed in this publication have been deposited in the National Center for Biotechnology Information's Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/geo/) and are accessible through GEO Series accession number GSE2950 [NCBI GEO] .

doi:10.1124/jpet.105.097014.

ABBREVIATIONS: SM, sulfur mustard; OHV, octyl homovanillamide; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; RMA, robust multiarray averaging; PCA, principal component analysis; Q-PCR, quantitative real-time polymerase chain reaction; CEES, chloroethyl ethyl sulfide; ANOVA, analysis of variance.

Formula The online version of this article (available at http://jpet.aspetjournals.org) contains supplemental material. Back

1 Current affiliation: Biosciences and Protection Division, Human Effectiveness Directorate, Wright-Patterson AFB, Ohio. Back

Address correspondence to: Dr. James F. Dillman, III, Cell and Molecular Biology Branch, U.S. Army Medical Research Institute of Chemical Defense, 3100 Ricketts Point Rd., Aberdeen Proving Ground, MD 21010-5400. E-mail: james.dillman{at}us.army.mil


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

Arroyo CM, Schafer RJ, Kurt EM, Broomfield CA, and Carmichael AJ (1999) Response of normal human keratinocytes to sulfur mustard (HD): cytokine release using a non-enzymatic detachment procedure. Hum Exp Toxicol 18: 1–11.[Abstract/Free Full Text]

Atkins KB, Lodhi IJ, Hurley LL, and Hinshaw DB (2000) N-Acetylcysteine and endothelial cell injury by sulfur mustard. J Appl Toxicol 20 (Suppl 1): S125–S128.[CrossRef][Medline]

Casillas RP, Kiser RC, Tuxall JA, Singer AW, Shumaker SM, Neimuth NA, Ricketts KM, Mitcheltree LW, Castrejon LR, and Blank JA (2000) Therapeutic approaches to dermatotoxicity by sulfur mustard I. Modulation of sulfur mustard-induced cutaneous injury in the mouse ear vesicant model. J Appl Toxicol 20: S145–S151.[CrossRef][Medline]

Casillas RP, Mitcheltree LW, and Stemler FW (1997) The mouse ear model of cutaneous sulfur mustard injury. Toxicol Methods 7: 381–397.[CrossRef]

Chatterjee D, Mukherjee S, Smith MG, and Das SK (2003) Signal transduction events in lung injury induced by 2-chloroethyl ethyl sulfide, a mustard analog. J Biochem Mol Toxicol 17: 114–121.[CrossRef][Medline]

Dillman JF III, McGary KL, and Schlager JJ (2003) Sulfur mustard induces the formation of keratin aggregates in human epidermal keratinocytes. Toxicol Appl Pharmacol 193: 228–236.[CrossRef][Medline]

Dillman JF III, McGary KL, and Schlager JJ (2004) An inhibitor of p38 MAP kinase downregulates cytokine release induced by sulfur mustard exposure in human epidermal keratinocytes. Toxicol In Vitro 18: 593–599.[CrossRef][Medline]

Dillman JF III, Phillips CS, Dorsch LM, Croxton MD, Hege AI, Sylvester AJ, Moran TS, and Sciuto AM (2005) Genomic analysis of rodent pulmonary tissue following bis (2-chloroethyl) sulfide exposure. Chem Res Toxicol 18: 28–34.[CrossRef][Medline]

Easton DF, Peto J, and Dol R (1988) Cancers of the respiratory tract in mustard gas workers. Br J Industry Med 45: 652–659.

Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, and Speed TP (2003) Exploration, normalization and summaries of high density oligonucleotide array probe level data. Biostatistics 4: 249–264.[Abstract]

Khatri P, Draghici S, Ostermeier GC, and Krawetz SA (2002) Profiling gene expression using Onto-Express. Genomics 79: 266–270.[CrossRef][Medline]

Lardot C, Dubois V, and Lison D (1999) Sulfur mustard upregulates the expression of interleukin-8 in cultured human keratinocytes. Toxicol Lett 110: 29–33.[CrossRef][Medline]

Manning KP, Skegg DCG, Stell PM, and Doll R (1981) Cancer of the larynx and other occupational hazards of mustard gas workers. Clin Otolaryngol 6: 165–170.[Medline]

Minsavage GD and Dillman JF III (2005) Altered signal transduction in human keratinocytes following exposure to bi-functional alkylating agents. Toxicol Sci 84: 38.[Abstract/Free Full Text]

Nishimoto Y, Yamakido M, Ishioka S, Shigenobu T, and Yukutake M (1988) Epidemiological studies of lung cancer in Japanese mustard gas workers, in Unusual Occurrences as Clues to Cancer Etiology (Miller RW and Watanabe S eds) pp 95–101, Japan Scientific Society Press/Taylor and Francis, Ltd., Tokyo.

Papirmeister B, Feister AJ, Robinson SI, and Ford RD (1991) Medical Defense against Mustard Gas: Toxic Mechanisms and Pharmacological Implications, CRC Press, Boca Raton, FL.

Petrali JP and Oglesby-Megee S (1997) Toxicity of mustard gas skin lesions. Microsc Res Tech 37: 221–228.[CrossRef][Medline]

Platteborze PL (2003) Effects of sulfur mustard on transcription in human epidermal keratinocytes: analysis by mRNA differential display. J Appl Toxicol 23: 249–254.[CrossRef][Medline]

Ricketts KM, Santai CT, France JA, Graziosi AM, Doyel TD, Gazaway MY, and Casillas RP (2000) Inflammatory cytokine response in sulfur mustard-exposed mouse skin. J Appl Toxicol 20: S73–S76.[CrossRef][Medline]

Rikimaru T, Nakamura M, Yano T, Beck G, Habicht GS, Rennie LL, Widra M, Hirshman CA, Boulay MG, Spannhake EW, et al. (1991) Mediators, initiating the inflammatory response, released in organ culture by full-thickness human skin explants exposed to the irritant, sulfur mustard. J Investig Dermatol 96: 888–897.[CrossRef][Medline]

Rogers JV, Choi YW, Kiser RC, Babin MC, Casillas RP, Schlager JJ, and Sabourin CLK (2004) Microarray analysis of gene expression in murine skin exposed to sulfur mustard. J Biochem Mol Toxicol 18: 289–299.[CrossRef][Medline]

Rosenthal DS, Simbulan-Rosenthal CM, Iyer S, Spoonde A, Smith W, Ray R, and Smulson ME (1998) Sulfur mustard induces markers of terminal differentiation and apoptosis in keratinocytes via a Ca2+-calmodulin and caspase-dependent pathway. J Investig Dermatol 111: 64–71.[CrossRef][Medline]

Sabourin CL, Danne MM, Buxton KL, Casillas RP, and Schlager JJ (2002) Cytokine, chemokine and matrix metalloproteinase response after sulfur mustard injury to weanling pig skin. J Biochem Mol Toxicol 16: 263–272.[CrossRef][Medline]

Sabourin CL, Danne MM, Buxton KL, Rogers JV, Niemuth NA, Blank JA, Babin MC, and Casillas RP (2003) Modulation of sulfur mustard-induced inflammation and gene expression by olvanil in the hairless mouse vesicant model. J Toxicol Cutaneous Ocul Toxicol 22: 125–136.[CrossRef]

Sabourin CLK, Rogers JV, Choi YW, Kiser RC, Casillas RP, Babin MC, and Schlager JJ (2004a) Time- and dose-dependent analysis of gene expression using microarrays in sulfur mustard-exposed mice. J Biochem Mol Toxicol 18: 300–312.[CrossRef][Medline]

Sabourin CLK, Rogers JV, Stonerock MK, Niemuth NA, Kiser RC, Casbohm SL, Babin MC, Schlager JJ, and Casillas RP (2004b) Alterations of gene expression in sulfur mustard-exposed skin topically treated with vanilloids. J Toxicol Cutan Ocul Toxicol 23: 321–328.[CrossRef]

Sabourin CL, Petrali JP, and Casillas RP (2000) Alterations in inflammatory cytokine gene expression in sulfur mustard-exposed mouse skin. J Biochem Mol Toxicol 14: 291–302.[CrossRef][Medline]

Schlager JJ, Benjamin HR, Ali K, Dodds AD, Hoffman CM, McGary KL, Yentzer BA, Phillips CS, Braue CR, Sistrunk JE, et al. (2002) Inflammatory transcript and protein production by cultured human keratinocytes following sulfur mustard (SM) exposure. Toxicol Sci 66: 356.[Free Full Text]

Schlager JJ and Hart BW (2000) Stress gene activity in HepG2 cells after sulfur mustard exposure. J Appl Toxicol 20: 395–405.[CrossRef][Medline]

Smith WJ, Saunders KM, Ruddle SE, and Gross CL (1993) Cytometric analysis of DNA changes induced by sulfur mustard. J Toxicol Cutan Ocul Toxicol 12: 337–343.[CrossRef]

The Gene Ontology Consortium (2000) Gene Ontology: tool for the unification of biology. Nat Genet 25: 25–29.[CrossRef][Medline]

Takeshima Y, Inai K, Bennett WP, Metcalf RA, Welsh JA, Yonehara S, Hayashi Y, Fujihara M, Yamakido M, Akiyama M, et al. (1994) p53 mutations in lung cancers from Japanese mustard gas workers. Carcinogenesis 15: 2075–2079.[Abstract/Free Full Text]

Tsuruta J, Sugisaki K, Dannenberg AM, Yoshimura T, Abe Y, and Mounts P (1996) The cytokines NAP-1(IL-8), MCP-1, IL-1beta and GRO in rabbit inflammatory skin lesions produced by the chemical irritant sulfur mustard. Inflammation 20: 293–318.[CrossRef][Medline]

Tokuoka S, Hayashi Y, Inai K, Egawa H, Aoki Y, Akamizu, H, Eto R, Nishida T, Ohe K, Kobuke T, et al. (1986) Early cancer and related lesions in the bronchial epithelium in former workers of mustard gas factory. Acta Pathol Jpn 36: 533–542.[Medline]

van der Schans GP, Noort D, Mars-Groenendijk RH, Fidder A, Chau LF, de Jong LP, and Benschop HP (2002) Immunochemical detection of sulfur mustard adducts with keratins in the stratum corneum of human skin. Chem Res Toxicol 15: 21–25.[CrossRef][Medline]

Werrlein RJ, Braue CR, and Dillman JF 3rd (2005) Multiphoton imaging the disruptive nature of sulfur mustard lesions, in Multiphoton Microscopy in the Biomedical Sciences V (Proceedings of SPIE) (Periasamy A and So PTC eds) vol 5700, pp 240–248, SPIE, Bellingham, WA.

Werrlein RJ and Madren-Whalley JS (2000) Effects of sulfur mustard on the basal cell adhesion complex. J Appl Toxicol 1 (Suppl 20): S115–S123.[CrossRef]

Werrlein RJ and Madren-Whalley JS (2003) Multiphoton microscopy: an optical approach to understanding and resolving sulfur mustard lesions. J Biomed Opt 8: 396–409.[CrossRef][Medline]

Zhang P, Ng P, Caridha D, Leach RA, Asher LV, Novak MJ, Smith WJ, Zeichner SL, and Chiang PK (2002) Gene expressions in Jurkat cells poisoned by a sulphur mustard vesicant and the induction of apoptosis. Br J Pharmacol 137: 245–252.[CrossRef][Medline]



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow All Versions of this Article:
jpet.105.097014v1
317/1/76    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dillman, J. F.
Right arrow Articles by Sabourin, C. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dillman, J. F., III
Right arrow Articles by Sabourin, C. L.

Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
All ASPET Journals Molecular Pharmacology Pharmacological Reviews
Molecular Interventions Drug Metabolism and Disposition