ToxSci Advance Access originally published online on July 26, 2006
Toxicological Sciences 2006 93(2):422-431; doi:10.1093/toxsci/kfl071
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The Kinetics of Transcriptomic Changes Induced by Cigarette Smoke in Rat Lungs Reveals a Specific Program of Defense, Inflammation, and Circadian Clock Gene Expression
* Philip Morris Research Laboratories GmbH, D-51149 Köln, Germany; Miltenyi Biotec GmbH, D-50829 Köln, Germany; Fraunhofer-Institut für Toxikologie und Experimentelle Medizin, D-30625 Hannover, Germany; and Philip Morris Research Laboratories bvba, B-3001 Leuven, Belgium
1 To whom correspondence should be addressed at Philip Morris Research Laboratories GmbH, Fuggerstr. 3, D-51149 Köln, Germany. Fax: +49 2203-303-362. E-mail: thomas.mueller{at}pmintl.com.
Received May 23, 2006; accepted July 23, 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Gene expression profiling in animal models exposed to cigarette mainstream smoke (CS) shapes up as a promising tool for investigating the molecular mechanisms involved in the onset and development of CS-related disease and may aid in the identification of disease candidate genes. Here we report on differential gene expression in lungs of rats exposed for 2, 7, and 13 weeks to 300 and 600 µg total particulate matter/l CS with sacrifice 2, 6, or 20 h after the last exposure. Regarding antioxidant and xenobiotic-metabolizing (phase I/II) enzymes, a stereotypic, mostly transient, expression pattern of differentially expressed genes was observed after each exposure period. The expression patterns were generally dose dependent for antioxidant and phase II genes and not dose dependent for phase I genes at the CS concentrations tested. However, with increasing length of exposure, there was a distinct, mostly sustained and dose-sensitive, expression of genes implicated in innate and adaptive immune responses, clearly pointing to an emerging inflammatory response. Notably, this inflammatory response included the expression of lung disease–related genes not yet linked to CS exposure, such as galectin-3, arginase 1, and chitinase, as well as genes encoding proteolytic enzymes. Finally, our experiments also revealed a CS exposure–dependent shift in the cyclical expression of genes involved in controlling the circadian rhythm. Altogether, these results provide further insight into the molecular mechanisms of CS-dependent disease onset and development and thus may also be useful for defining CS-specific molecular biomarkers of disease.
Key Words: cigarette smoke; circadian rhythm; inflammation; oxidative stress; rat lung; gene expression.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Cigarette mainstream smoke (CS) is causally linked to diseases known to develop in a chronic inflammatory environment, such as chronic obstructive pulmonary disease (COPD), cardiovascular disease, and cancer (Peto et al., 1992
; US Department of Health and Human Services, 1989a,b). Results from mode-of-action investigations focusing mainly on the mechanism of acute CS-induced cellular and target organ molecular effects have uncovered a distinct pattern of (oxidative) stress-related effects in exposed cells and tissues as reflected by, for example, genotoxic insults, membrane and cytosolic damage, alteration of gene expression, and epigenetic events (summarized in a series of reviews, see Müller and Gebel, 2006). During chronic CS inhalation, repeated exposure to these pleiotropic detrimental conditions is assumed to result in accumulating cellular and tissue damage gradually creating a chronic inflammatory environment, which eventually favors the development of CS-related diseases (summarized for COPD in Rahman, 2005).
Using DNA microarray techniques, we recently reported on the differential gene expression in vitro in cultured cells (Bosio et al., 2002) and in vivo in tissues of the respiratory tract of rats exposed either acutely (once) or short term (3 weeks) to CS (Gebel et al., 2004). Generally, we observed the upregulation of a specific subset of antioxidant and phase II–related genes known to be widely controlled by nuclear factor-E2–related factor 2 (Nrf2) (for review, see Kobayashi and Yamamoto, 2005), thus confirming recent mechanistic investigations showing that Nrf2 activation is a major protective cellular feature in the context of CS exposure in vitro (Knörr-Wittmann et al., 2005) and in vivo (Rangasamy et al., 2004). In CS-exposed nasal epithelium and lungs, genes encoding phase I xenobiotic-metabolizing enzymes, which are mainly controlled by the aryl hydrocarbon receptor (AhR) (Nebert et al., 2004), were found to be induced in parallel to phase II genes (Gebel et al., 2004). Interestingly, a general feature of these investigations was the tight kinetic regulation of almost all differentially expressed genes, i.e., gene expression patterns similar to controls were observed in rat lungs 20 h after exposure, even after a 3-week exposure period. Finally, the lack of persistent changes in gene expression was paralleled by a nearly complete lack of inflammatory response under these exposure conditions (Gebel et al., 2004).
In contrast to investigations using animal models, most studies in humans are done on tumor tissue or material from patients with advanced cases of a given disease; therefore, little is known about early changes in gene expression in normal lung tissue of smokers or the impact of CS on initiation and early development of the different diseases (Müller and Gebel, 2006). Using cellular material obtained from human donors during bronchoscopy, Spira et al. (2004b) recently compared the transcriptomes of chronic smokers, never smokers, and former smokers and, regarding the expression of xenobiotic-metabolizing and antioxidant genes, described expression profiles for the human smoking situation that were similar to those observed in rodent smoking models (Gebel et al., 2004). Notably, the human study (Spira et al., 2004b) detected a group of 13 genes in former smokers, which instead of returning to normal, as was seen for most of the CS-dependently regulated genes after more than 2 years of smoking cessation, retained the expression behavior specific to current smokers, thus indicating a CS exposure–specific imprint on gene expression acquired during chronic CS exposure.
In a first attempt to delineate the transcriptome described for human chronic smoking (Spira et al., 2004b) in a rodent model and to gain further insight into the early steps of CS-related disease development, we exposed rats for up to 13 weeks to two CS concentrations, with special emphasis on the kinetics of gene expression and the identification of disease-related genes. The results revealed a distinct, exposure-length–dependent signature in the differential expression of genes mainly encoding antioxidant and xenobiotic-metabolizing enzymes, inflammatory proteins, and factors involved in regulating circadian rhythm.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Cigarette and cigarette smoke generation.
The Reference Cigarette 2R4F was obtained from the Tobacco and Health Institute at the University of Kentucky. The cigarettes were conditioned and smoked according to International Standards Organization standards as described (Gebel et al., 2004
). Mainstream smoke was diluted with filtered, conditioned air to target concentrations of 300 µg total particulate matter (TPM)/l and 600 µg TPM/l. To monitor the stability and reproducibility of smoke generation, TPM, CO, nicotine, aldehydes (formaldehyde, acetaldehyde, and acrolein), and particle size distribution were determined (Haussmann et al., 1998) (supplementary data Table S1). Temperature and relative humidity in the exposure chambers were within standard limits.
Animals and exposure.
Outbred male Sprague-Dawley rats (Crl: CDBR) were obtained from Charles River (Portage, MI). The rats were 6–11 weeks old at the start of exposure. Histopathological evaluation and serological screening confirmed the good health status of the rats at the beginning of the study. Care and use of the rats was in conformity with the American Association for Laboratory Animal Science Policy on the Humane Care and Use of Laboratory Animals (http://www.aalas.org/association/about/policy-HumaneCare.htm). Animal experiments were approved by the Institutional Animal Care and Use Committee. Four rats per group were nose-only exposed in flow-past exposure chambers (type FPC1-132) to diluted CS (300 or 600 µg TPM/l) or to conditioned fresh air (control) for 2, 7, or 13 weeks (two times 1 h/day with a 30-min fresh air break in between, 5 days/week) and sacrificed at 2, 6, or 20 h after the last exposure. To monitor CS uptake and exposure, carboxyhemoglobin (HbCO) concentrations in blood of additional animals were determined. The mean HbCO concentrations (± standard error [SE]) in the CS-exposed groups were 23.4% (± 0.26) for the 300-µg TPM/l groups and 37.6% (± 1.02) for 600-µg TPM/l groups. HbCO was determined during the last 30 min of exposure on study day 36. Individual rats were removed from the exposure tubes and immediately anesthetized and bled from the retro-orbital venous plexus. No CS exposure–dependent mortality was observed. Significant reduction in body weight gain was seen in the CS-exposed groups by the end of the 13-week exposure (Fig. S1).
RNA preparation.
RNA preparation and modification for microarray hybridization were essentially performed as previously described (Gebel et al., 2004). In brief, whole lung tissue, used for total RNA preparation, was frozen in liquid N2 immediately after dissection and stored at –70°C. After quantification and checking for integrity (using an Agilent 2100 Bioanalyzer according to the supplier's instructions), equal amounts of RNA from the four animals per group were pooled, and, for hybridization on PIQOR microarrays, subjected to linear amplification and labeling. Two micrograms of RNA was labeled by reverse transcription via Cy5-dCTP (deoxycytidine triphosphate) incorporation (CS-exposed tissue) and Cy3-dCTP incorporation (unexposed tissue). Each labeled sample was then divided in half and hybridized on two customized PIQOR microarrays.
PIQOR complementary DNA microarray.
Microarray production was done as described previously (Bosio et al., 2002). A detailed description of the microarray platform and the generated data is deposited in the National Center for Biotechnology Information Gene Expression Omnibus (GEO, accession no. GSE4644). In total, 36 PIQOR microarray experiments were performed. The first PIQOR microarray consisted of 3000 rat genes, the second PIQOR microarray consisted of 2937 rat genes spotted as two replicates on different positions on the array. In addition, 20 stress-relevant genes were spotted as six-fold replicates on the second PIQOR microarray. The six-fold replicates were performed to further strengthen data consistency among spot replicates. As a qualitative measurement for the validity of the data and to check for the uniformity of the hybridization process, the coefficient of variation of the ratio for each gene was calculated (Table S2).
Array hybridization and data analysis.
Hybridization, scanning, and data analysis were performed as described elsewhere (Gebel et al., 2004). Briefly, image capture and signal quantification of hybridized PIQOR microarrays were done with the ScanArrayLite (Packard Bioscience, Billerica, MA) and ImaGene software Version 4.1 (BioDiscovery, Los Angeles, CA). Local background was subtracted from the signal to obtain the net signal intensity and the ratio of Cy5/Cy3. Subsequently, the mean of the ratios of spots for each complementary DNA (cDNA) was computed. Unflagged primary data were globally centralized using the median of averaged spot ratios. For centralization, only those spots were used for which the fluorescent intensity in one of the two channels was twice the mean background of all unflagged spots. In order to exclude unreliable ratios derived from noisy signals, only genes displaying a net signal intensity > 2-fold than the mean background were used for further analysis. From this data set, only those genes that showed a 
2-fold differential gene expression in at least one experiment were considered in the subsequent interpretations and used in the cluster analysis (a complete list of these genes is shown in Table S7). Cluster analysis was carried out using the TMEV program, a component of the TIGR TM4 software package (Saeed et al., 2003). One-dimensional hierarchical clustering was applied using the Pearson correlation as a comparison metric.
Affymetrix GeneChip analysis.
Transcriptome analysis was done following the manufacturer's recommendation in the Affymetrix Gene Chip Expression Analysis Technical Manual (Santa Clara, CA). After synthesis and cleavage of double-stranded cDNA, the degree of fragmentation and the length distribution of the fragmented biotinylated cRNA was checked by capillary electrophoresis using the Agilent 2100 Bioanalyzer. Ten micrograms of biotinylated fragmented cRNA was hybridized to each GeneChip Rat Expression Array 230A and B for 16 h at 60 rpm in a 45°C GeneChip Hybridization Oven 640 (Affymetrix). The arrays were washed and stained on the GeneChip Fluidics Station 400 (Affymetrix) followed by antibody signal amplification, washing, and staining using streptavidin R-phycoerythrin (SAPE; Molecular Probes, Eugene, OR). To amplify staining, SAPE solution was added twice with a biotinylated anti-streptavidin antibody (Vector Laboratories, CA) staining step in between. All procedures were carried out according to Affymetrix protocols and recommendations. The arrays were scanned using the Agilent GeneArray Scanner 3000. Scanned image files were visually inspected for artifacts and then analyzed. Each image was scaled to the same all probe set intensity for comparison between chips. The GeneChip Operating Software was used to control the fluidics station and the scanner, to capture probe array data and analyze hybridization intensity data, applying the default parameters provided in the Affymetrix data analysis software package.
Data analysis.
For the comparison analysis, all intensity data were scaled to the same probe set intensity. Using MS Access, differentially regulated genes were further extracted by applying additional criteria. Induced genes had to have a signal value above 100 and a "present-call" (P) in the treated experiment. In addition, the signal log ratio (SLR) had to be 1, which equals a fold change (FC) 2. For repressed genes, the signal value had to be higher than 100 in the control experiment and the SLR had to be 
–1, which corresponds to an FC –2. Only genes extracted applying these criteria in at least one experiment were considered in the subsequent interpretations of the results (a complete list of these genes is shown in Table S9).
Reverse transcription real-time quantitative PCR experiments.
For each RNA sample, 2 µg was reverse transcribed and 20 ng of the reverse transcription (RT)-reaction product was used as template for further analysis. Transcript levels were measured by reverse transcription real-time quantitative PCR (RT qPCR) using the Perkin Elmer Applied Biosystems (Foster City, CA) prism model 7000 sequence detection system (PE ABI 7000 SDS). The sequences of forward and reverse primers as designed by Primer Express (PE Applied Biosystems) for NAD(P)H dehydrogenase quinone 1 (nqo1), heme oxygenase-1 (hmox1), aldehyde dehydrogenase 3A1 (aldh3A1), cxcl1, chitinase, arginase 1, and gapdh (used for normalization) are listed under supplementary data (Table S3). All RT qPCR experiments were performed in triplicate and repeated at least once using an independently performed reverse transcription reaction. Mean induction values of two independent experiments are shown.
Determination of enzyme activity and Western analysis.
A detailed description of the methods and references can be found under supplementary data. In brief, cytoplasmatic cell extracts (10,000 x g supernatant; determination of NQO1, ALDH-3, and chitinase enzyme activities) and microsomal fractions (determination of ethoxyresorufin O-deethylase [EROD] activity) were prepared from aliquots of the same frozen lungs that were used for RNA preparation. Protein concentrations were determined by a modified Lowry assay. NQO1 activity was determined spectrophotometrically by measuring enzyme-dependent reduction of 2,6-dichlorophenolindophenol. ALDH-3 activity was determined spectrophotometrically using benzaldehyde as substrate by measuring the enzyme-dependent increase of NADPH. Acidic mammalian chitinase activity was measured fluorometrically using the fluorogenic substrate 4-methylumbelliferyl ß-D-N,N'-diacetylchitobiose. EROD activity reflecting lung CYP1A1 activity was also determined fluorometrically using the fluorogenic substrate 7-ethoxyresorufin. Results of all enzyme assays are expressed as mean value ± SE for three to four animals per group. Western analysis was performed with the cytoplasmatic lung tissue extracts. Equal amounts of the individual cytoplasmic protein extract preparations from the lungs of three to four animals per group were pooled. Ten microgram protein each was separated by a 15% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and further analyzed in duplicate as described (Knörr-Wittmann et al., 2005) using a NQO1-specific antibody (SP7251P, Acris, Hiddenhausen, Germany).
For statistical analysis, analysis of variance followed by the Dunnett post hoc test was performed. In some groups, standard deviations increased with increasing means; therefore, log-transformed data were used for statistical testing of all groups. Because several of the measurements were below the detection limit in the controls of the EROD groups, the statistical analysis for the EROD groups had to be performed with rank-transformed data.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Study Concept and General Observations
Male Sprague-Dawley rats (four rats per group) were nose-only exposed to mainstream CS (University of Kentucky Standard Reference Cigarette 2R4F) at TPM concentrations of 300 or 600 µg/l, for 13 weeks. Exposure was 2 x 1 h/day (with 30-min break between the 2 h to avoid toxic concentrations of HbCO), 5 days/week. Daily CS doses were comparable to doses used in other animal studies (e.g., Vanscheeuwijk et al., 2002
) and were up to 10-fold higher (for the high dose group) than the calculated daily dose in the normal adult human smoking situation (approximately 20 cigarettes per day) when referred to body weight and two-fold higher in relation to body surface area (see supplementary data). The typical histopathological findings in the lungs of rats exposed to these dose levels and periods are confined to goblet cell hyperplasia and the appearance of pigmented alveolar macrophages (Vanscheeuwijk et al., 2002). In the present study, no CS exposure–dependent mortality was observed. Reductions in body weight gain commonly observed in CS-exposed rats was within the range described in other studies (Vanscheeuwijk et al., 2002, see Fig. S1).
To characterize alterations in differential gene expression in the context of exposure duration, intermediate dissections at 2 and 7 weeks were performed. To monitor the kinetics of differentially expressed genes at the end of each exposure period, rats were sacrificed 2, 6, or 20 h after final exposure. Whole lung tissue from all groups was analyzed for differential gene expression by cDNA (PIQOR) microarray analysis, using two research glass microarrays covered with a total of 5937 cDNA probes (3000 and 2937, see Materials and Methods). In order to obtain a comprehensive image of CS-induced differential gene expression, the lung transcriptome of the rats exposed for 13 weeks was additionally screened by a genome-wide (Affymetrix) approach. A detailed description of both microarray platforms and the data generated are deposited in the National Center for Biotechnology Information GEO (http://www.ncbi.nlm.nih.gov/geo, accession no. GSE4516: Affymetrix, GSE4644: PIQOR). A detailed quantitative overview of the results is given in Table S4 (supplementary data). In general, a high correlation, in both qualitative and quantitative terms, was observed for those differentially expressed genes explored by both technologies after 13 weeks of exposure, thus principally confirming CS-specific differential gene expression by two independent technology platforms (R2 = 0.85, PIQOR vs. Affymetrix; supplementary material, Fig. S2, Table S5). The validity of the microarray studies was further checked for selected genes by RT qPCR, which, in quantitative terms, generally confirmed the data obtained by microarray analysis (Table S6). In quantitative terms, 237 CS-dependently regulated genes were identified in this study by cDNA microarray analysis according to the criterion of a 2-fold differential regulation in at least one of the 18 experiments (Table S7). Cluster analysis, using a hierarchical clustering algorithm, revealed a dendrogram portraying the expression profiles specific to CS-exposed lung tissue over the whole exposure period (Fig. S3).
Antioxidant Response and Xenobiotic Metabolism
The induction of genes related to antioxidant response and xenobiotic metabolism (phase I/II), seen in our first study and in lung epithelia of human smokers (Spira et al., 2004b), was also observed in the present study (Fig. 1, clustered genes from "FMO3" to "FMO2"). As is most obvious from the profiles exhibited by cytochrome p450 1A1 (cyp1A1), nqo1, aldh3A1, and hmox1, genes in this group are stereotypically unregulated and downregulated during each smoke/smoke-free period.
|
Furthermore, antioxidant genes showed an apparent dose-dependent effect (Table S8) and a remarkable increase in the number of
2-fold upregulated genes after longer exposure periods (i.e., five genes after 2 weeks vs. 10 genes after 13 weeks of exposure). Notably, this intensification of the cellular defense against increasing oxidative burden includes the dose-dependent upregulation of the genes coding for metallothionein 1a and 2 (MT1a, MT2) as well as the gene coding for superoxide dismutase-2, which becomes evident after 7 weeks of exposure (Fig. 1, Table S8). In sharp contrast to all other genes in this functional background, these three genes remain upregulated even after 20 h postexposure, thus showing expression profiles that resemble those exhibited by numerous other genes, mostly of inflammatory/immune regulatory nature (see below).
The activation of mt1a and mt2 in CS-exposed lung tissue is especially intriguing because Spira et al. (2004b) showed that three MT isogenes are repressed in current chronic smokers and, importantly, are part of a group of 13 genes that remain decreased in former smokers, even in individuals who had not smoked for 20–30 years. Upregulation of MT genes has been reported to provide cytoprotection in response to various forms of stress or injury (Theocharis et al., 2003), especially regarding heavy metal ions such as the CS-related carcinogen cadmium. Thus, it would be interesting to know whether there is also selection pressure on the ablation of mt1/mt2 expression during chronic CS exposure of rodent models and, in order to estimate the relevance of MT expression in CS exposure, whether mt1/mt2 knock-out models (Klaassen and Liu, 1998) are more prone to CS-related disease development than corresponding wild-type animals.
It is also of note that, nqo1 proved to be the most strongly induced of the antioxidant and phase II–related genes, thus also corresponding to the data from human smokers (Spira et al., 2004b) and our previous study (Gebel et al., 2004). In fact, the strong transcriptional induction of nqo1 was reflected at the protein level, as demonstrated by Western analysis and on the functional level by determining NQO1-specific enzyme activity (Fig. 2A). Beyond its enzymatic function of reducing quinones to hydroquinones, NQO1 has also been shown to stabilize the tumor suppressor protein p53, especially under oxidative stress conditions (Anwar et al., 2003; Asher et al., 2002). Thus, substantiation of enhanced NQO1 expression in CS-exposed lung tissue may add to the hypothesis that, in addition to its basic antioxidant function, activation of NQO1 is instrumental in the efficient stabilization of p53.
|
The phase I response of CS-exposed rat lung tissue (Fig. 1 and Table S8) is characterized essentially by the pronounced and transient induction of cyp1A1 and aldh3A1, independent of the dose and length of exposure. Determination of the corresponding enzyme activities (Fig. 2B) also confirmed this finding on the functional level, thus indicating that both activities are maximally induced within 2 weeks of exposure to 300 µg TPM/l. Interestingly, CYP1A1, unlike ALDH3, generally shows a significant decrease in enzyme activity with increasing length of postexposure periods, thus confirming that CYP1A1 is subject to rapid turnover. Finally, it is of note that the spectrum of phase I–related genes induced by CS also includes the activation of two genes supposedly involved in nicotine metabolism, i.e., flavin monooxygenase 2 and 3 (Hukkanen et al., 2005
).
Inflammatory and Immune Response
The specific expression profile, as demonstrated by antioxidant and xenobiotic-detoxifying (phase I/II) genes, is contrasted by several genes that are not induced or that even tend to be suppressed after short-term exposure (2 weeks), but which become progressively active with increasing length of CS exposure (Figs. 1 and 3, Table S8). The expression profiles exhibited by the genes forming this cluster are generally characterized by an emerging constitutive expression with increasing length of exposure periods, while the expression ratios show a clear concentration dependency. Importantly, most of the genes represented in Figure 3 share an inflammation-related background. While earlier effects cannot be excluded, the first indication of an inflammatory response in CS-exposed rat lungs was observed as early as 2 weeks after the start of exposure by the dose-dependent and sustained expression of the genes encoding the complement component C3 (Carroll, 2004) and lipocalin 2 (lcn2), a potent bacteriostatic protein involved in innate immune reactions (Goetz et al., 2002). In principle, lcn2 and C3 expression may originate from cells of the target tissue itself and/or from immune cells that invade inflammatory sites in CS-exposed lungs. In fact, in addition to resident macrophages, the presence of neutrophils was found to be significantly elevated in the bronchoalveolar lavage fluid (BALF) of rats exposed for only 2 weeks to both doses of CS (B. Friedrichs et al., 2006 and personal communication).
|
Neutrophils and macrophages are activated by damaged cells and tissues through the expression of chemotactic cytokines, which in CS-exposed rat lungs are potentially represented by cxcl1 (scyb1), a putative functional homologue of human interleukin (IL)-8, mip1
(ccl3), and mip1
(ccl9) (for review, see Kunkel and Strieter, 2005) (Fig. 3). Notably, with increasing length of CS exposure, these signals become intensified in a dose-dependent manner in terms of both strength and duration of expression (Table S8). As a potential consequence of chemokine activation, the expression of interleukin receptors IL8rb and IL1r2 as well as of macrophage- and monocyte-specific cell surface antigens cd68 and cd33 was observed after 7–13 weeks of exposure. The notion of an increasing infiltration of CS-exposed rat lungs with proinflammatory immune cells is further reflected by the dose-dependent upregulation of the myeloid-related genes S100a9 and S100a8, the protein products of which are known to direct neutrophil migration to sites of inflammation (Vandal et al., 2003). Importantly, elevated plasma levels of these proteins were reported in chronic bronchitis (Roth et al., 1992).
The inflammatory processes in CS-exposed lungs may be fueled through the potential cell survival activities provided by galectin-3 (lgals3) and osteopontin (opn/spp1) (Figs. 1 and 3, Table S8), which both function through activation of the antiapoptotic phosphatidyl inositol 3-kinase/protein kinase B (Akt) pathway (Lin and Yang-Yen, 2001; Oka et al., 2005; Zhu et al., 2005). While opn expression was recently demonstrated in alveolar macrophages from human smokers (Woodruff et al., 2005), the expression of lgals3 is described here, to our knowledge, for the first time in the context of CS exposure. Importantly, Lgals3, a general inhibitor of Fas-induced apoptosis (Yang et al., 1996), has also been described as a receptor for advanced glycation end products (Zhu et al., 2001), which have been shown in the plasma of smokers by the proposed reaction of nornicotine, a CS constituent and metabolite of nicotine, with suitable protein residues (Dickerson and Janda, 2002). Equally important, Opn was found to be upregulated under hypoxic conditions (Zhu et al., 2005) and has been implicated in the development and progression of several lung diseases (for review, see O'Regan, 2003).
Extending the characterization of early inflammatory events in CS-exposed rat lungs by genome-wide screening of the transcriptome expressed by rats for 13 weeks (Table 9S) uncovered the upregulation of additional genes known to be involved in proinflammatory processes, the most notable of which are acidic chitinase and arginase 1. Reevaluation of all tissue samples by RT qPCR revealed that both genes are induced as early as 2 weeks after start of exposure with slightly increasing expression ratios still seen at later time points (Fig. 4A). Both genes are regulated in a CS dose-dependent manner and, although exhibiting a transient shape of expression, remain induced for at least 20 h. Notably, as demonstrated for chitinase, this specific expression pattern is reflected almost entirely at the functional (enzymatic) level (Fig. 4B).
|
Both chitinase and arginase 1 have been implicated in allergic asthma (Zhu et al., 2004
; Zimmermann et al., 2003). Recently, the pronounced expression of acidic mammalian chitinase was demonstrated in airway epithelia and pulmonary macrophages in a mouse model of asthma and in allergic asthma in humans (Zhu et al., 2004). Although the involvement of chitinase in asthma is still something of an enigma, the contribution of the enzyme to disease expression has been clearly demonstrated through experiments showing that chitinase neutralization with either antibodies or the chitinase inhibitor allosamidin ameliorated airway inflammation and airway hyperresponsiveness (Zhu et al., 2004). In mechanistic terms, neutralization of chitinase was paralleled by the partial inhibition of IL-13 pathway activation and cytokine induction, which were also shown to be the driving force of arginase 1 expression (summarized in King et al., 2004), thus suggesting that both enzymes are induced by similar routes of activation. The source of IL-13 as well as of IL-4 and IL-5 in asthma is a subset of CD4-T lymphocytes known as type-2 helper (TH2) cells (for review, see Fireman, 2003). TH2-dependent cytokines are key mediators in asthma development because they direct the fundamental features of the disease, i.e., infiltration of eosinophils into the lung, airway hyperresponsiveness, and enhanced mucus production. It is worth noting, that although there was an increase in mucus production, there were no clear signs of an asthmatic phenotype in the lungs of rats exposed for 13 weeks to CS. This conclusion is based on observations at either the histopathological level, as expressed, for example, by a lack of lung hyperinflation and smooth muscle hypertrophy (H. Weiler, personal communication), or by BALF analysis, as indicated by a lack of pulmonary eosinophilia (data not shown). Hence, the meaning of arginase 1 and chitinase expression in CS-exposed rat lungs remains to be elucidated, e.g., by specifically analyzing the cytokine pattern in BALF combined with experiments on the pulmonary function of CS-exposed rats. Finally, the expression of arginase 1 may pertain to the paradoxical observation that smokers exhale lower amounts of NO compared to nonsmokers (Kharitonov et al., 1995), which has been explained in part by the finding that CS is inhibitory to the expression of inducible NO synthase (iNOS) in lung epithelial cells (Hoyt et al., 2003). In fact, arginase 1 may also contribute to this effect by cocompeting for arginine, the basic substrate of NO formation by iNOS (Vercelli, 2003).
In addition to arginase 1 and chitinase, the genome-wide approach revealed the expression of cathepsin K and matrix metallopeptidase 12 (mmp12) (also seen on the cDNA array) (Fig. 3, Table S8 and S9), although this was mainly confined to the lungs of rats exposed to 600 µg TPM/l. Since the expression of both cathepsin peptidase and matrix metallopeptidases has been shown to be involved in COPD formation (for review, see Barnes et al., 2003), this finding further adds to the overall image of early proinflammatory changes induced by CS exposure, which are characteristically dose dependent.
Effects of CS Exposure on Circadian Clock Genes
A striking observation made during these experiments is the obvious impact of CS exposure on the expression of genes involved in regulating the circadian rhythm. In contrast to the different expression profiles described above for the vast majority of differentially expressed genes, a small cluster of genes shows a distinct pattern of cyclic expression (Fig. 5). Intriguingly, this group of genes includes bmal1 (arntl), the protein product of which, together with the proteins encoded either by the noncircadian clock or npas2 gene, functions as a key transcription factor in a transcriptional feedback system controlling the circadian rhythm (for review, see Hirayama and Sassone-Corsi, 2005).
|
The cyclical nature of bmal1 expression is reflected by the activation of the gene 6 h after exposure and by the consistent repression of the gene 2 and 20 h after exposure, while two other genes, i.e., D-site–binding protein (dbp) and orphan nuclear receptor NR1D2 (nr1d2), oscillate in an anticyclic manner (Fig. 5A). At least dbp has been identified as a direct target gene of BMAL1::CLOCK in a mouse liver model (Oishi et al., 2003
). The impact of CS on circadian clock genes becomes most obvious when the intensities of single channel signals are analyzed, allowing the approximate comparison between smoke-exposed animals and unexposed controls. As demonstrated for bmal1 (Fig. 5B), the lungs of CS-exposed rats (13 weeks) and controls show circadian oscillations, but display opposite characteristics, indicating at least a phase shift in the circadian cycle induced by CS exposure. Finally, identification of the differential expression of the circadian clock genes per1 and per2, also anticyclic to bmal1, by genome-wide screening of the rat lung transcriptome (Table 9S) further emphasizes the impact of CS exposure on the circadian rhythm.
Previous investigations have clearly demonstrated that the molecular machinery regulating the circadian clock is susceptible to environmental influences such as light, food, activity, and temperature, thereby adapting the intrinsic clock to the actual environmental situation (for review, see Pittendrigh, 1993). Because nicotine does not significantly influence circadian activities (Benowitz et al., 2002; Liu and Gillette, 1996), the meaning of the CS-dependent effects on the system as described here remains to be elucidated. Importantly, data from a recent report demonstrated that the DNA-binding activity of both BMAL1::CLOCK and BMAL1::NPAS2 is dependent on the redox ratio of NAD(P)+/NAD(P)H with enhanced DNA binding observed at high concentrations of reduced cofactors (Rutter et al., 2001). Thus, it is tempting to speculate that CS exposure, by altering the redox potential of cells and tissues, alters the DNA binding activity and consequently the transcriptional activity of BMAL1-containing transcription factors, thus resulting in CS-dependent entrainment. Moreover, in the context of CS-mediated pulmonary inflammation, the observation that inflammatory changes are associated with circadian variation in pulmonary function as seen in subjects with mild asthma (Kelly et al., 2004) is even more intriguing. However, for a general consideration it must also be taken into account that the rats were exposed during their normal physiological resting periods, which could clearly compromise the quality of recovery and therefore interfere with general mechanisms controlling the circadian clock.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Building on our previous work (Gebel et al., 2004
), we have characterized the progression of differential gene expression induced by CS exposure in whole lungs of rats exposed for up to 13 weeks. The data reveal a specific CS-dependent signature of gene expression, mainly implicating genes whose protein products serve to metabolize and/or detoxify xenobiotics and to combat oxidant stress (phases I and II), as well as inflammation-related genes encoding proteins that regulate and execute innate and adaptive immune-specific responses, and, surprisingly, genes involved in controlling the circadian rhythm. As exemplified for some specific candidate genes, the transcriptional activity could be traced to the functional (enzymatic) level. In general, genes involved in phase I/II response followed a repetitive pattern of upregulation and downregulation, independent of the duration of exposure, suggesting that the regulation of these genes is subject to general principles presumably orchestrated by Nrf2 and AhR. In contrast to the stereotypical antioxidant/xenobiotic-detoxifying response, a gradual progression in the activation of genes implicated in innate and adaptive immune–specific responses is observed, as reflected by the gradual increase in the expression of genes coding for chemokines and their receptors. However, it remains to be elucidated how the CS-triggered phase I/II response affects the inflammatory response as recent findings indicate, for example, that constitutive activation of the AhR results in chronic tissue inflammation in a conditional transgenic mouse model (Tauchi et al., 2005). The CS-provoked inflammatory response in rat lungs also unraveled the upregulation of genes, which, although implicated in different lung diseases, such as galectin-3, chitinase, and arginase 1, have not yet been described in the context of CS exposure. Future studies should reveal the relevance of these genes in the context of human smoking as they potentially may provide the opportunity to serve as critical biomarkers of CS-induced potential harm. Moreover, although the data described here generally correlate with corresponding results reported for human smokers, especially with regard to the expression of phase I/II and inflammation-related genes (Spira et al., 2004a,b), there are on the other hand some crucial differences in gene expression observed between the human chronic smoking situation and the rat subchronic smoking model described in this manuscript. These differences, which are reflected, for example, by the pronounced CS-dependent upregulation of MT genes in the current subchronic smoking model and their consistent repression in chronic current and former smokers (Spira et al., 2004b), may help identify critical molecular markers involved in CS-induced disease development and progression. |
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Supplementary data are available online at http://toxsci.oxfordjournals.org/.
|
ACKNOWLEDGMENTS |
|---|
We thank K. Hofmann (Miltenyi Biotech) for bioinformatics analysis, L. Conroy (Philip Morris Research Laboratories) for expert editorial support; V. Böhm S. Lufen, M. Grabow-Caspari (all Philip Morris Research Laboratories), and G. Grosshauser, F. Huebel, and S. Rueberg (all Miltenyi Biotech) for skillful technical assistance. This work was funded by Philip Morris, USA. Miltenyi Biotec GmbH and the Fraunhofer Institut (Hannover) have received payments from Philip Morris Research Laboratories GmbH for production of the PIQOR cDNA and Affymetrix microarrays and execution of the hybridization experiments, respectively.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Anwar, A., Dehn, D., Siegel, D., Kepa, J. K., Tang, L. J., Pietenpol, J. A., and Ross, D. (2003). Interaction of Human NAD(P)H:quinone oxidoreductase 1 (NQO1) with the tumor suppressor protein p53 in cells and cell-free systems. J. Biol. Chem. 278(12), 10368–10373.
Asher, G., Lotem, J., Kama, R., Sachs, L., and Shaul, Y. (2002). NQO1 stabilizes p53 through a distinct pathway. Proc. Natl. Acad. Sci. U.S.A. 99(5), 3099–3104.
Barnes, P. J., Shapiro, S. D., and Pauwels, R. A. (2003). Chronic obstructive pulmonary disease: Molecular and cellular mechanisms. Eur. Respir. J. 22(4), 672–688.
Benowitz, N. L., Hansson, A., and Jacob, P., III. (2002). Cardiovascular effects of nasal and transdermal nicotine and cigarette smoking. Hypertension 39(6), 1107–1112.
Bosio, A., Knörr, C., Janssen, U., Gebel, S., Haussmann, H. J., and Müller, T. (2002). Kinetics of gene expression profiling in Swiss 3T3 cells exposed to aqueous extracts of cigarette smoke. Carcinogenesis 23(5), 741–748.
Carroll, M. C. (2004). The complement system in regulation of adaptive immunity. Nat. Immunol. 5(10), 981–986.[CrossRef][Web of Science][Medline]
Dickerson, T. J., and Janda, K. D. (2002). A previously undescribed chemical link between smoking and metabolic disease. Proc. Natl. Acad. Sci. U.S.A. 99(23), 15084–15088.
Fireman, P. (2003). Understanding asthma pathophysiology. Allergy Asthma Proc. 24(2), 79–83.[Web of Science][Medline]
Friedrichs, B., Miert, E. V., and Vanscheeuwijk, P. M. (2006). Lung inflammation in rats following subchronic exposure to cigarette mainstream smoke. Exp. Lung Res. 32, 1–29.[CrossRef][Web of Science][Medline]
Gebel, S., Gerstmayer, B., Bosio, A., Haussmann, H. J., Miert, E. V., and Müller, T. (2004). Gene expression profiling in respiratory tissues from rats exposed to mainstream cigarette smoke. Carcinogenesis 25(2), 169–178.
Goetz, D. H., Holmes, M. A., Borregaard, N., Bluhm, M. E., Raymond, K. N., and Strong, R. K. (2002). The neutrophil lipocalin NGAL is a bacteriostatic agent that interferes with siderophore-mediated iron acquisition. Mol. Cell 10(5), 1033–1043.[CrossRef][Web of Science][Medline]
Haussmann, H. J., Anskeit, E., Becker, D., Kuhl, P., Stinn, W., Teredesai, A., Voncken, P., and Walk, R. A. (1998). Comparison of fresh and room-aged cigarette sidestream smoke in a subchronic inhalation study on rats. Toxicol. Sci. 41(1), 100–116.
Hirayama, J., and Sassone-Corsi, P. (2005). Structural and functional features of transcription factors controlling the circadian clock. Curr. Opin. Gen. Dev. 15(5), 548–556.[CrossRef][Web of Science][Medline]
Hoyt, J. C., Robbins, R. A., Habib, M., Springall, D. R., Buttery, L. D. K., Polak, J. M., and Barnes, P. J. (2003). Cigarette smoke decreases inducible nitric oxide synthase in lung epithelial cells. Exp. Lung Res. 29(1), 17–28.[CrossRef][Web of Science][Medline]
Hukkanen, J., Jacob, P., III, and Benowitz, N. L. (2005). Metabolism and disposition kinetics of nicotine. Pharmacol. Rev. 57(1), 79–115.
Kelly, E. A. B., Houtman, J. J., and Jarjour, N. N. (2004). Inflammatory changes associated with circadian variation in pulmonary function in subjects with mild asthma. Clin. Exp. Allergy 34(2), 227–233.[CrossRef][Web of Science][Medline]
Kharitonov, S. A., Robbins, R. A., Yates, D., Keatings, V., and Barnes, P. J. (1995). Acute and chronic effects of cigarette smoking on exhaled nitric oxide. Am. J. Respir. Crit. Care Med. 152(2), 609–612.[Abstract]
King, N. E., Rothenberg, M. E., and Zimmermann, N. (2004). Arginine in asthma and lung inflammation. J. Nutr. 134(10), 2830S–2836S.
Klaassen, C. D., and Liu, J. (1998). Metallothionein transgenic and knock-out mouse models in the study of cadmium toxicity. J. Toxicol. Sci. 23(Suppl. 2), 97–102.
Knörr-Wittmann, C., Hengstermann, A., Gebel, S., Alam, J., and Müller, T. (2005). Characterization of Nrf2 activation and heme oxygenase-1 expression in NIH3T3 cells exposed to aqueous extracts of cigarette smoke. Free Radic. Biol. Med. 39(11), 1438–1448.[CrossRef][Web of Science][Medline]
Kobayashi, M., and Yamamoto, M. (2005). Molecular mechanisms activating the Nrf2-Keap1 pathway of antioxidant gene regulation. Antioxid. Redox Signal. 7(3–4), 385–394.[CrossRef][Web of Science][Medline]
Kunkel, S. L., and Strieter, R. M. (2005). Cytokines and chemokines in lung inflammation and injury. In Fishman's Pulmonary Diseases and Disorders (A. P. Fishman and J. A. Elias, Eds.), 3rd ed., pp. 315–324. McGraw-Hill, New York.
Lin, Y. H., and Yang-Yen, H. F. (2001). The osteopontin-CD44 survival signal involves activation of the phosphatidylinositol 3-kinase/Akt signaling pathway. J. Biol. Chem. 276(49), 46024–46030.
Liu, C., and Gillette, M. U. (1996). Cholinergic regulation of the suprachiasmatic nucleus circadian rhythm via a muscarinic mechanism at night. J. Neurosci. 16(2), 744–751.
Müller, T., and Gebel, S. (2006). Cigarette smoke-induced redox signaling and gene expression in in vitro and in vivo models. In Cigarette Smoke & Oxidative Stress (B. B. Halliwell and H. E. Poulsen, Eds.), pp. 75–111. Springer, Heidelberg.
Nebert, D. W., Dalton, T. P., Okey, A. B., and Gonzalez, F. J. (2004). Role of aryl hydrocarbon receptor-mediated induction of the CYP1 enzymes in environmental toxicity and cancer. J. Biol. Chem. 279(23), 23847–23850.
Oishi, K., Miyazaki, K., Mesaki, M., Kobayashi, H., Iitaka, C., Kudo, T., Shimizu, Y., Ishida, N., Kadota, K., Kikuno, R., et al. (2003). Genome-wide expression analysis of mouse liver reveals CLOCK-regulated circadian output genes. J. Biol. Chem. 278(42), 41519–41527.
Oka, N., Nakahara, S., Takenaka, Y., Fukumori, T., Hogan, V., Kanayama, H., Yanagawa, T., and Raz, A. (2005). Galectin-3 inhibits tumor necrosis factor-related apoptosis-inducing ligand-induced apoptosis by activating Akt in human bladder carcinoma cells. Cancer Res. 65(17), 7546–7553.
O'Regan, A. (2003). The role of osteopontin in lung disease. Cytokine & Growth Factor Reviews 14(6), 479–488.[CrossRef][Web of Science][Medline]
Peto, R., Lopez, A. D., Boreham, J., Thun, M., and Heath, C., Jr. (1992). Mortality from tobacco in developed countries: Indirect estimation from national vital statistics. Lancet 339(8804), 1268–1278.[CrossRef][Web of Science][Medline]
Pittendrigh, C. S. (1993). Temporal organization: Reflections of a Darwinian clock-watcher. Annu. Rev. Physiol. 55(1), 17–54.[CrossRef]
Rahman, I. (2005). Smoking-induced inflammation, injury, and disease—Molecular mechanisms. In Oxygen/Nitrogen Radicals—Lung Injury and Disease (V. Vallyathan, V. Castranova, and X. Shi, Eds.), pp. 211–255. Marcel Dekker, New York.
Rangasamy, T., Cho, C. Y., Thimmulappa, R. K., Zhen, L., Srisuma, S. S., Kensler, T. W., Yamamoto, M., Petrache, I., Tuder, R. M., and Biswal, S. (2004). Genetic ablation of Nrf2 enhances susceptibility to cigarette smoke-induced emphysema in mice. J. Clin. Invest. 114(9), 1248–1259.[CrossRef][Web of Science][Medline]
Roth, J., Teigelkamp, S., Wilke, M., Grun, L., Tummler, B., and Sorg, C. (1992). Complex pattern of the myelo-monocytic differentiation antigens MRP8 and MRP14 during chronic airway inflammation. Immunobiology 186(3–4), 304–314.[Web of Science][Medline]
Rutter, J., Reick, M., Wu, L. C., and McKnight, S. L. (2001). Regulation of clock and NPAS2 DNA binding by the redox state of NAD cofactors. Science 293(5529), 510–514.
Saeed, A. I., Sharov, V., White, J., Li, J., Liang, W., Bhagabati, N., Braisted, J., Klapa, M., Currier, T., Thiagarajan, M., et al. (2003). TM4: A free, open-source system for microarray data management and analysis. Biotechniques 34(2), 374–378.[Web of Science][Medline]
Spira, A., Beane, J., Pinto-Plata, V., Kadar, A., Liu, G., Shah, V., Celli, B., and Brody, J. S. (2004a). Gene expression profiling of human lung tissue from smokers with severe emphysema. Am. J. Respir. Cell Mol. Biol. 31(6), 601–610.
Spira, A., Beane, J., Shah, V., Liu, G., Schembri, F., Yang, X., Palma, J., and Brody, J. S. (2004b). Effects of cigarette smoke on the human airway epithelial cell transcriptome. Proc. Natl. Acad. Sci. U.S.A. 101(27), 10143–10148.
Tauchi, M., Hida, A., Negishi, T., Katsuoka, F., Noda, S., Mimura, J., Hosoya, T., Yanaka, A., Aburatani, H., Fujii-Kuriyama, Y., et al. (2005). Constitutive expression of aryl hydrocarbon receptor in keratinocytes causes inflammatory skin lesions. Mol. Cell. Biol. 25(21), 9360–9368.
Theocharis, S. E., Margeli, A. P., and Koutselinis, A. (2003). Metallothionein: A multifunctional protein from toxicity to cancer. Int. J. Biol. Markers 18(3), 162–169.[Web of Science][Medline]
US Department of Health and Human Services. (1989a). Chronic Obstructive Pulmonary Disease (US Department of Health and Human Services, Ed.), pp. 66–68. Public Health Service, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion Office on Smoking and Health, Rockville, MD. DHHS Publication No. (CDC) 89-8411.
US Department of Health and Human Services. (1989b). Coronary Heart Disease. (US Department of Health and Human Services, Ed.), pp. 58–65. Public Health Service, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion Office on Smoking and Health, Rockville, MD. DHHS Publication No. (CDC) 89-8411.
Vandal, K., Rouleau, P., Boivin, A., Ryckman, C., Talbot, M., and Tessier, P. A. (2003). Blockade of S100A8 and S100A9 suppresses neutrophil migration in response to lipopolysaccharide. J. Immunol. 171(5), 2602–2609.
Vanscheeuwijk, P. M., Teredesai, A., Terpstra, P. M., Verbeek, J., Kuhl, P., Gerstenberg, B., Gebel, S., and Carmines, E. L. (2002). Evaluation of the potential effects of ingredients added to cigarettes. Part 4: Subchronic inhalation toxicity. Food Chem. Toxicol. 40, 113–131.[CrossRef][Web of Science][Medline]
Vercelli, D. (2003). Arginase: Marker, effector, or candidate gene for asthma? J. Clin. Invest. 111(12), 1815–1817.[CrossRef][Web of Science][Medline]
Woodruff, P. G., Koth, L. L., Yang, Y. H., Rodriguez, M. W., Favoreto, S., Dolganov, G. M., Paquet, A. C., and Erle, D. J. (2005). A distinctive alveolar macrophage activation state induced by cigarette smoking. Am. J. Respir. Crit. Care Med. 172(11), 1383–1392.
Yang, R. Y., Hsu, D. K., and Liu, F. T. (1996). Expression of galectin-3 modulates T-cell growth and apoptosis. Proc. Natl. Acad. Sci. U.S.A. 93(13), 6737–6742.
Zhu, W., Sano, H., Nagai, R., Fukuhara, K., Miyazaki, A., and Horiuchi, S. (2001). The role of galectin-3 in endocytosis of advanced glycation end products and modified low density lipoproteins. Biochem. Biophys. Res. Commun. 280(4), 1183–1188.[CrossRef][Web of Science][Medline]
Zhu, Y., Denhardt, D. T., Cao, H., Sutphin, P. D., Koong, A. C., Giaccia, A. J., and Le, Q. T. (2005). Hypoxia upregulates osteopontin expression in NIH-3T3 cells via a Ras-activated enhancer. Oncogene 24(43), 6555–6563.[Web of Science][Medline]
Zhu, Z., Zheng, T., Homer, R. J., Kim, Y. K., Chen, N. Y., Cohn, L., Hamid, Q., and Elias, J. A. (2004). Acidic mammalian chitinase in asthmatic Th2 inflammation and IL-13 pathway activation. Science 304(5677), 1678–1682.
Zimmermann, N., King, N. E., Laporte, J., Yang, M., Mishra, A., Pope, S. M., Muntel, E. E., Witte, D. P., Pegg, A. A., Foster, P. S., et al. (2003). Dissection of experimental asthma with DNA microarray analysis identifies arginase in asthma pathogenesis. J. Clin. Invest. 111(12), 1863–1874.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Sato, Y. Hoshino, T. Hara, S. Muro, H. Nakamura, M. Mishima, and J. Yodoi Thioredoxin-1 Ameliorates Cigarette Smoke-Induced Lung Inflammation and Emphysema in Mice J. Pharmacol. Exp. Ther., May 1, 2008; 325(2): 380 - 388. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-J. Haussmann Smoking and Lung Cancer: Future Research Directions International Journal of Toxicology, July 1, 2007; 26(4): 353 - 364. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||