Induction of Dectin-1 and asthma-associated signal transduction pathways in RAW 264.7 cells by a triple-helical (1, 3)-β-d glucan, curdlan
- First Online:
- Cite this article as:
- Rand, T.G., Robbins, C., Rajaraman, D. et al. Arch Toxicol (2013) 87: 1841. doi:10.1007/s00204-013-1042-4
- 566 Views
People living in damp buildings are typically exposed to spore and mycelial fragments of the fungi that grow on damp building materials. There is experimental evidence that this exposure to triple-helical (1, 3)-β-d glucan and low molecular weight toxins may be associated with non-atopic asthma observed in damp and moldy buildings. However, the mechanisms underlying this response are only partially resolved. Using the pure (1, 3)-β-d glucan, curdlan, and the murine macrophage cell line, RAW 264.7, there were two objectives of this study. The first was to determine whether signal transduction pathways activating asthma-associated cell signaling pathways were stimulated using mouse transduction Pathway Finder® arrays and quantitative real-time (QRT) PCR. The second objective was to evaluate the dose and temporal responses associated with transcriptional changes in asthma-associated cytokines, the signal transduction receptor gene Dectin-1, and various transcription factor genes related to the induction of asthma using customized RT-PCR-based arrays. Compared to controls, the 10−7 M curdlan treatment induced significant changes in gene transcription predominately in the NFkB, TGF-β, p53, JAK/STAT, P13/AKT, phospholipase C, and stress signaling pathways. The 10−8 M curdlan treatment mainly induced NFkB and TGF-β pathways. Compared to controls, curdlan exposures also induced significant dose- and time-dependent changes in the gene translations. We found that that curdlan as a non-allergenic potentiator modulates a network of transduction signaling pathways not only associated with TH-1, TH-2, and TH-3 cell responses including asthma potentiation, but a variety of other cell responses in RAW 264.7 cells. These results help provide mechanistic basis for some of the phenotypic changes associated with asthma that have been observed in in vitro, in vivo, and human studies and open up a hypothesis-building process that could explain the rise of non-atopic asthma associated with fungi.
KeywordsTrichocomaceaeAspergillaceaeMoldsDamp buildingsTriple-helical (1, 3)-β-d glucanCurdlanDectin-1RAW264.7 cellsInflammation-associated genesNon-atopic asthma
People living in damp buildings are typically exposed to spore and mycelial fragments of the toxin-producing, anamorphic Trichocomaceae (e.g., Aspergillus, Penicillium, and related hyphomycetes) that grow on damp building materials. These result in exposure to allergens, various low molecular weight toxins, and (1, 3)-β-d glucan. There is sufficient epidemiological evidence that these exposures are linked to a variety of non-allergenic respiratory health outcomes (Cox-Ganser et al. 2005; Mendell et al. 2011; NIOSH 2012; WHO 2009). Evidence for the non-allergenic responses is supported by recent in vitro and in vivo studies that have demonstrated that low molecular weight compounds and the form of (1, 3)-β-d glucan from fungi that grow in damp buildings are potently inflammatory. At doses that can occur in moldy buildings, these studies also demonstrated that some of these metabolites from several fungi modulate cytokine responses involved in asthma (Miller et al. 2010; Rand et al. 2010, 2011). These results provided a potential mechanistic basis for the rise of non-atopic asthma associated with fungi in damp buildings (see NIOSH 2012; WHO 2009). Fungal chitin also has been postulated to play a role in contributing to the asthma response (Van Dyken et al. 2011; Vega and Kalkum 2012). Wu et al. (2010) reported that environmental exposure to fungi and polymorphisms of several human chitinases appear to play a role in severe asthma.
There are many studies where zymosan or curdlan has been used as models of (1, 3)-β-d glucan exposure. Results of in vitro and in vivo studies employing both particulate and NaOH-solubilized zymosan have demonstrated that this substance is potently pro-inflammatory and exhibits acute pulmonary toxicity (Young et al. 2006). However, its use as a model (1, 3)-β-d glucan to explore the pathophysiological impact of mold exposures on respiratory health in the built environment is problematic. Zymosan is not found in spore and mycelial cell wall fragments of fungi that grow on damp building materials (Cherid et al. 2011), and yeasts are not typically damp building contaminants (Miller et al. 2008). Moreover, it is a crude cell wall mixture comprising (1, 3)-β-d and (1, 6)-β-d glucans (1:1) in triple-helical form, mannans, mannoproteins, and chitin, among others (Martin 2012), all of which constitute pattern recognition receptor (PRR) agonists (Vega and Kalkum 2012). Whether these agonists collaboratively interact to induce respiratory responses atypical of those associated with exposure to wall fragments from anamorphic fungi is unclear. However, what is known is that zymosan provokes cellular and gene transcriptional toxicokinetic responses entirely different from those induced by curdlan (Masuda et al. 2012). Some glucan in air is derived from yeasts, but this is typically a modest percentage (see Foto et al. 2005).
Curdlan, a pure linear (1, 3)-β-d glucan (161 kDa) from Alcaligenesfaecalis, is similar to the dominant form of glucan in the majority of fungi found growing on damp building materials. Studies on the glucans in houses demonstrate that the molecular weight range measured overlaps with that of building molds (Cherid et al. 2011; Foto et al. 2005). Studies with humans in chambers as well as in rodent and in vitro models have demonstrated that this glucan is potently bioactive. Recent studies using mouse models have shown that curdlan targets the Dectin-1 receptor on the surfaces of immuno-sentinel lung cells including alveolar macrophages and bronchiolar epithelium (Rand et al. 2010). They also revealed that exposure to low curdlan concentrations resulted in significant transcriptional modulation of TH1-type cytokines such as Ifng, MIP2, and TNFα; TH2-type cytokines including Ccl5, Ccl11 (=eotaxin), IL-4, IL-6, Il10, Il13, and their receptors in lungs, and hypersecretion of acidic mucus by bronchiolar epithelium (Rand et al. 2010, 2011). Overall, these responses are similar to those from exposure to spores from the facultative pathogenic mold, A. fumigatus (Neveu et al. 2011). Human studies involving experimental exposures also point to the inflammatory potential of curdlan and to its potential to stimulate increased nasal eosinophil concentrations in exposed individuals (Bonlokke et al. 2006; Sigsgaard et al. 2000; Rylander 1993, 1996). Although zymosan and curdlan have similar affinities to glucan receptors (e.g., Tanaka et al. 1991), curdlan inhibits the response of zymosan in rodent glucan receptors (Czop 1986) presumably because of less steric hindrance to the receptor. In the context of the built environment, curdlan or extracts from anamorphic Trichocomaceae are better models for mold exposures.
These TH2-type cytokine changes and bronchiolar epithelium mucus hypersecretion in the in vivo studies as well as the recruitment and increased eosinophil concentrations in human studies are important pathophysiological hallmarks of asthma (Pernis and Rothman 2002; Barnes 2009). The significantly modulated TH-2 cytokines we identified are among those that are recognized to promote cellular inflammation in the asthmatic lung and contribute to the pathogenesis of allergy and lung remodeling in chronic asthma (Mukherjee et al. 2009). These results suggest that (1, 3)-β-d glucan can induce TH2-type transcriptional responses associated with asthma pathophysiology in the lung environment (Rand et al. 2010, 2011). However, the mechanistic details underlying these responses are unclear.
In this study, we investigated whether curdlan can potentiate transcriptional changes TH2- and TH3-type response in a murine cell model of lung disease. Based on the results of the in vivo studies, we hypothesized that exposure to (1, 3)-β-d glucan from indoor-associated fungi contributes to a phenotypic response associated with asthma-related lung disease by activating not only TH-2 and TH-3 genes but also asthma-associated cell signaling pathways transducing these effects such as the JAK/STAT signaling pathway, which control many of the physiologic events associated with asthma (Pernis and Rothman 2002). Using the murine macrophage cell line, RAW 264.7, which is sensitive to curdlan (Kataoka et al. 2002), there were two objectives of this study. The first was to determine whether signal transduction pathways activating asthma-associated cell signaling pathways were stimulated. This was done with mouse transduction Pathway Finder® arrays and quantitative real-time (QRT) PCR. The second objective was to evaluate the dose and temporal responses associated with transcriptional changes in asthma-associated cytokines, the signal transduction receptor gene Dectin-1, and various transcription factor genes related to the induction of asthma using customized QRT-PCR-based arrays.
Materials and methods
Curdlan from A. faecalis (Sigma-Aldrich C7821, lot # 89H4032 ≥99 % purity), which had been chemically characterized by Foto et al. (2005), was used in the experiments. A fresh curdlan stock solution was prepared for each in vitro experiment by dissolving it in 0.3 M sodium hydroxide (NaOH) followed by dilution in pyrogen-free phosphate-buffered saline (PBS) (TekNova) to 10−6 M. Stock curdlan solution was then diluted and administered to the cell cultures as a single dose of 100 in 900 μl of media at final concentrations of 10−7 and 10−8 M which spans plausible environmental exposures (Rand et al. 2011).
The RAW 264.7 murine macrophage cell line was a gift from the Marine Biosciences Institute, National Research Council, Halifax, NS. Cells were cultured in RPMI medium supplemented with 10 % Fetal Bovine Serum and 100 U/ml penicillin and 100 μg/ml streptomycin. Cell cultures were maintained in 6-well culture plates at 37 °C in a 5 % CO2-humidified incubator and supplied with fresh medium every 2 days until confluent. Cell numbers were determined using a hemacytometer.
Aliquots (100 μl) of stock curdlan solution were administered to confluent cell cultures for treatment concentrations of either 10−7 or 10−8 M and for 2, 4, 8, 18, and 24 h postexposure (PE) with pyrogen-free carrier saline as the control. All experiments were performed in quadruplicate. At the end of each exposure time, the reactions were stopped by decanting the medium and rinsing the cells twice with 1 ml of pyrogen-free PBS. The PBS was then removed, 1 ml of Trizol® reagent was added for 5 min, and the cell lysate was transferred into 1.6-ml Eppendorf tubes and stored at −80 °C for up to 1 week for RNA extraction.
Total RNA isolation was performed using TRIzol® technique according to manufacturer’s specifications (Invitrogen). Briefly, cells immersed in TRIzol® reagent and frozen at −80 °C were thawed at room temperature, and 200 μl of chloroform was added and vortexed at maximum throw. The solution was then centrifuged at 4 °C for 15 min at 13,000 rpm and the top aqueous layer transferred into a new tube. RNA was precipitated by the addition of 500 μl of absolute isopropanol and incubated at room temperature for 10 min. The precipitated RNA was centrifuged at 4 °C for 10 min at 13,000 rpm to pellet the RNA. The RNA pellet was washed twice with 1 ml of 75 % ethanol and centrifuged for 5 min at 4 °C. After air-drying for 30 min, the RNA pellet was resuspended in 20 μl of RNase/DNase-free water (Sigma-Aldrich). The concentration and integrity of RNA were then determined using a NanoDrop® ND-1000. Samples with sufficient RNA concentrations (≥5 μg/ml) and with 260/280 nm ratio of ≥1.8 were used for array analysis.
Real-time PCR and array analysis
Real-time (RT) PCRs were carried out using a reaction ready first strand cDNA synthesis kit (SuperArray, Bioscience Corp.) according to manufacturer’s instructions. The responsiveness of the cells to 10−7 and 10−8 M curdlan was first evaluated at 2 h postexposure (PE) using a mouse transduction Pathway Finder® array (SA Biosciences # PAMM-014) and RT-PCR, following the manufacturer’s procedures, and outlined in Rand et al. (2010, 2011). The array profiles the transcription levels of 84 key genes. This included 18 signal transduction pathways associated with Th2/Th3 responses, two housekeeping genes, a mouse genomic DNA contamination control, three reverse transcription controls, and three positive PCR controls. The no-reverse transcription (NRT) control was made through a combination of a 1 in 100 dilution of the original RNA in RNase-free water with PCR master mix and RNase-free water.
To evaluate dose (10−7 and 10−8 M curdlan) and temporal (2, 4, 8 18, and 24 h PE) responses associated with transcriptional changes, customized RT-PCR-based arrays were then used. For these assays, the receptor gene Dectin-1, the TH-1 cytokine Tnfα, three TH-2 cytokines (= IL4, IL10, IL13), the TH-3 cytokine TGF-β, and eight transcription factor genes (including NFkB1, MAPK1, and STAT-1–6) were assayed because all of these genes are considered to have central roles in the orchestration of asthma-related responses. RT-PCR was carried out with cDNA using SYBR Green 1 (Molecular Probes, Eugene, OR) gene-specific primers (supplementary Table 1) designed using Primer 3 and custom synthesized by Integrated DNA Technologies, and an ABI Prism 7,000. Relative gene expression was determined according to the comparative Ct method, with the Act-β housekeeping gene and carrier control references set as the calibrators after Hudson et al. (2005).
For the mouse transduction Pathway Finder® arrays, genes and treatment groups were clustered using the SuperArray web-based hierarchical clustering software package for the PCR Array System (http://www.sabiosciences.com/pcrarraydataanalysis.php). A Shapiro–Wilks test for normality was performed to verify that the data were normally distributed. A two-tailed unpaired t test (p ≤ 0.05) was then performed to determine whether the treatments showed significant gene transcription in comparison with the carrier controls. All data are shown as their mean fold change values. Gene transcription results were considered significant if they had ≥1.5-fold or ≤−1.5-fold change at p ≤ 0.05 probability level. For the customized arrays, two-way ANOVA was employed to test for dose and time fold change responses and performed using SYSTAT version 11.0, and results were considered significant at α ≤ 0.05.
Mouse Transduction Pathway Finder® arrays
Transduction pathway-associated gene changes in RAW264.7 cells exposed to 10−7 and 10−8 M curdlan at 2 h PE
10−7 M p value
10−8M p value
Signaling pathway induced
Bcl2-associated × protein
B cell leukemia/lymphoma 2
Baculoviral IAP repeat-containing 3
Baculoviral IAP repeat-containing 5
Breast cancer 1
Cyclin-dependent kinase 2
Cyclin-dependent kinase inhibitor 1A (P21)
Cyclin-dependent kinase inhibitor 1B
CCAAT/enhancer-binding protein (C/EBP), beta
Colony-stimulating factor 2 (granulocyte–macrophage)
Early growth response 1
Fas (TNF receptor superfamily member 6)
Growth arrest and DNA-damage-inducible 45 alpha
Glycogen synthase 1, muscle
Homeo box A1
Heat shock factor 1
Intercellular adhesion molecule 1
Insulin-like growth factor–binding protein 4
Inhibitor of kappaB kinase beta
Interleukin 1 alpha
Interferon regulatory factor 1
Ngfi-A-binding protein 2
Nuclear factor of kappa light polypeptide gene enhancer in B cell inhibitor, alpha
Nuclear receptor-interacting protein 1
Ornithine decarboxylase, structural 1
Protein kinase C
Patched homolog 1
Prostaglandin-endoperoxide synthase 2
7.40E − 05
Selectin, endothelial cell
Tumor necrosis factor
Transformation-related protein 53
Vascular endothelial growth factor A
Gene transcription modulation in custom arrays
Results of this study showed that RAW 264.7 cells are acutely sensitive (within 2 h PE)) to exposure to a single dose of pure curdlan as a model (1, 3)-β-d glucan from damp building fungi, at concentrations ranging from 10−7 to 10−8M. These concentrations span plausible environmental exposures and are within the range of concentrations determined to induce cytokine responses in in vivo and in vitro models of lung disease (Rand et al. 2010, 2011).
Using a mouse transduction Pathway Finder® and customized array assays, our results revealed that the Dectin-1 receptor and key genes associated with a variety of signaling pathways including phospholipase C, PI3 K/Akt, MAPK, NFAT, and NF-κB were induced in curdlan-exposed RAW 264.7 cells. Induction of the above-listed receptor and signaling pathways in curdlan-exposed RAW264.7 cells further confirms that this pure, NaOH-solubilized glucan, most similar to the dominant form of glucan in the majority of molds found growing on damp building materials, is potently bioactive. These results support previous studies that demonstrated that Dectin-1 and these transduction signaling genes are induced in cells exposed to (1, 3)-β-d glucan (Kingeter and Lin 2012; Martin 2012; Masuda et al. 2012) and that they have an important role in mediating its biological effects (Brown 2006). The new data also support the position that these various signaling pathways are essential components of the adaptive and innate cell responsiveness to this glucan (Drummond et al. 2011; Faro-Trindade et al. 2012; Martin 2012). Importantly, they provide mechanistic bases for some of the pro-inflammatory gene transcription and expression responses to curdlan as reported in the previous in vitro and in vivo studies using this compound (Bonlokke et al. 2006; Masuda et al. 2012; Rand et al. 2010, 2011; Rylander 1993, 1996; Sigsgaard et al. 2000). For example, Dectin-1 induction by (1, 3)-β-d glucans can lead to NFAT activation through spleen tyrosine kinase (Syk)- and CARD9-Bcl10-Malt1-dependent pathway(s) and to NF-κB signaling (Bauer et al. 2008; Chan et al. 2009; Drummond et al. 2011) either singly or in synergistic collaboration with the TLR4 and TLR2/6 receptors (Ferwerda et al. 2008). NFAT signaling can induce the TH-1 cytokines IL-2 and Ifn-γ (Porter and Clipstone 2002), while NF-κB signaling can stimulate the transcription of TH-1-type cytokine genes such as Ifn–γ, MIP2 (= CXC2 = Il-8), and Tnfα and other inflammatory related genes such as Ccl2, GM-CSF, ICAM-1, IL-1β, LTA, iNOS, PTGS2 (=COX2), and Vegfa responses (Tak and Firestein 2001), all of which have documented in the present or previous studies. Induction of these signaling pathways by (1, 3)-β-d glucan helps explain some of the TH-1 type cytokine responses. However, their induction does not appear to directly contribute to the CXC and CC motif cytokines and some TH-2 type interleukin responses which have been consistently demonstrated to be significantly modulated in curdlan-exposed in vitro or in vivo animal models (Masuda et al. 2012; Rand et al. 2010, 2011). This suggests that other receptors and their associated signaling pathways also play important roles in the innate and adaptive cell responsiveness to this glucan.
Recent studies have demonstrated that the scavenger receptors F1 and CD36, and LacCer, CR3, and SIGNR1 receptors are activated on macrophage and other immune cell surfaces by β-glucans. These can result in the downstream induction of phospholipase C, PI3 K/Akt, and MAPK signaling cascades (Chan et al. 2009; Martin 2012; Masuda et al. 2012). Rand et al. (2010) also demonstrated that a variety of other receptors (e.g., CXCR3, CCR1, CCR2, CCR3, CCR6, CCR7, CCR9, CCR10 and IL1RA, IL5RA, IL6RA, Il13RA) are significantly and rapidly (within 4–12 h PE) up-regulated in curdlan-exposed mouse lungs. Interestingly, CCR2, which is associated with eosinophil recruitment and inflammation, is also activated by chitin (Roy et al. 2012), suggesting that this receptor may have a central role in the modulation of cell responses to complex fungal cell wall polysaccharides. While it is unclear from the present study whether all these receptor types are activated on RAW264.7 cell surfaces, the significant transcription of key genes associated with a variety of signaling pathways suggests that curdlan, like other β-glucans, acts not only on Dectin-1 but other multiple receptors on RAW264.7 cell surfaces. Involvement of other receptors helps explain an important feature of the present study. Use of the customized array assays revealed that MAPK, NF-κB, and Tnfα gene transcription was significantly up-regulated starting within 2–4 h PE whereas significant Dectin-1 transcriptional up-regulation was comparatively laggard (8–24 h PE). This result suggested that some of the initial cell responses to curdlan are Dectin-1 independent. A relatively delayed Dectin-1 activation was also observed by Masuda et al. (2012) in mouse peritoneal macrophages exposed to curdlan. Interestingly, these investigators demonstrated that the toxicokinetics of Dectin-1, CR3, and TLR2 receptor activation; p38 MAPK ERK, p38 MAPK, and syk phosphorylation; and transcription of GM-CSF, G-CSF, IL-6, Tnfα, IL-12p40, and IL-10 responses in zymosan- and curdlan-exposed mouse peritoneal macrophages was markedly different. These results help account for the very different immunogenic potency and effects of β-glucan forms (soluble vs. particulate; long chain vs. short chain; pure vs. crude) on in vitro and in vivo models of lung disease reported in the literature.
Importantly, that curdlan initially acts on receptors other than those associated with Dectin-1 also provides a mechanism by which this compound can modulate asthma-associated TH-2 type cytokines including IL-4, IL-6, Il-10, and Il-13 that have been reported in in vitro and in vivo models (Masuda et al. 2012; Rand et al. 2010, 2011), without obvious Dectin-1 involvement. These genes are modulated through the JAK/STAT pathway, which is activated by a variety of ligands including cytokines and growth factors such as CSF-2 (=GM-CSF) and TGF-β (http://www.sabiosciences), some of which (e.g., GM-CSG) may be derived through autocrinal mechanisms (Masuda et al. 2012). Transcriptional induction of Cebpb, Csf2, Irf1, IL-4, IL-10, and IL-13 genes as members of the JAK/STAT signaling pathways in the present study as well as IL-4, IL-10, and IL-13 in our previous in vitro and in vivo studies (Rand and Miller 2010; Rand et al. 2010, 2011) supports our hypothesis that exposure to this (1, 3)-β-d glucan in particulate material from anamorphic Trichocomaceae may have an important role to play in the phenotypic expression of non-atopic asthma in individuals in damp buildings.
The importance of the JAK/STAT signaling cascade in transducing the cytokine-mediated signals associated with asthma has been previously highlighted (Darnell 1997; Pernis and Rothman 2002). This cascade triggers downstream signaling of Stat transcription factors, which in the present study were all modulated in a time-dependent fashion especially in 10−7 M curdlan-exposed RAW 264.7 cells. These transcription factors and most notably Stat-4 and Stat-6 are well known for their involvement in asthma potentiation (Grünebach et al. 2002; Pernis and Rothman 2002; Shah et al. 2009; Xu et al. 2009). Stat-6 activation plays a central role in exerting Il-4- and Il-13-mediated biological responses including control and all the major components that characterize an inflammatory asthmatic response (Pernis and Rothman 2002). Activated Il-4 also interacts with a variety of cytokines including IL1-B, IL4-R, IL-10, Ccl11, and Tnfα, which we have previously reported to be significantly up-regulated in curdlan-exposed mouse lungs (Rand et al. 2010). IL-13 that is produced primarily by activated TH-2 cells is considered critical in the pathophysiological involvement of allergen-induced asthma (Wills-Karp et al. 1998; Pernis and Rothman 2002). It is also involved in the recruitment and activation of eosinophils (Zhu et al. 1999; Pernis and Rothman 2002), which may help explain results of the human curdlan exposure studies (Bonlokke et al. 2006; Sigsgaard et al. 2000; Rylander 1993, 1996). In exploratory immunohistochemical localization experiments, we showed positive IL-13 staining of bronchiolar epithelium and alveolar macrophages in mouse lungs exposed to 10−7 M curdlan but not in carrier control lungs which suggested its involvement in the pathobiology observed in our in vivo experiments (Rand and Miller 2011). Moreover, heterodimerization of the receptor IL-13R by IL-13 also induces activation of the phosphatidylinositol-3 (PI-3) kinase pathway (Kasaian and Miller 2008) and transforming growth factor-beta-1 (TGFB1) (Fichtner-Feigl et al. 2006). TGFB1 was significantly transcribed in the curdlan-exposed RAW 264.7 cells. TGFB1 derived through the TGF signaling cascade by TH-3 type cell responses is a player in lung fibrosis and an important phenotypic manifestation of asthma. Other Stat transcription factors such as Stat-2 also appear to play important roles in asthma potentiation. Induction of Stat-2 in turn results in IL-6 up-regulation that in turn activates Stat-3 (http://www.sabiosciences; Crawford 2010). JAK/STAT transduction pathway activation provides a mechanism that helps explain results of our previous studies (Rand et al. 2010, 2011) showing rapid IL-6 and IL6rα up-regulation induced by curdlan exposures.
In summary, the present study has shown that curdlan as a non-allergenic potentiator can induce a pleomorphic network of transduction signaling pathways not only associated with TH-1, TH-2, and TH-3 cell responses including asthma potentiation but a variety of other phenotypic cell responses in RAW 264.7 cells. These include the modulation of apoptotic, angiogenic, growth, mitogenic, proliferation, and metabolic cell responses, among others. This result helps provide mechanistic basis for many of the phenotypic changes observed in in vitro, in vivo, and human studies. Moreover, this study also demonstrates that curdlan can induce genes through the JAK/STAT pathway which is involved in asthma. Lastly, the results revealed connectivity between signals triggered by different receptors and cytokine (e.g., Tnfα) induction through the MAPK, NF-kB,JAK/STAT, and other signaling pathways. This highlights the complex epistatic signaling interrelations among the various genes induced by curdlan exposure in RAW 264.7 cells and other models of lung disease. This reinforces the position of Weiss (2010) that genes operate in complex networks, not as individual actors in a molecular physiologic play, and to focus on individual genes at the expense of networks is an attempt to avoid what is a much more complicated problem.
We thank Dr. G. Sun for use of the RT-PCR instrument. This work was supported by NSERC operating grants to T.G.R. and an NSERC IRC to J.D.M.
Conflict of interest
The authors declare that they have no conflict of interest.