Background

Pulmonary fibrosis is a chronic, progressive, and usually untreatable group of chronic disorders and appears to be regulated by complex cellular processes [1]. In animal models, a single intratracheal administration of bleomycin induces an inflammatory response that is characterized by leukocyte infiltration, apoptosis, fibroblast proliferation, matrix metalloproteinase (MMP)/tissue inhibitor of metalloproteinase (TIMP) imbalance, and an increase in interstitial collagen content [2, 3] that can culminate in the development of pulmonary lesions similar to those observed in human interstitial pulmonary fibrosis (IPF) [4]. However, the exact mechanisms underlying pulmonary fibrosis remain unclear.

AP-1 is a dimeric transcription factor, mainly comprised of the Jun (c-Jun, Jun-B, Jun-D), Fos (c-Fos, Fos-B, Fra-1, Fra-2), and ATF (ATF1-4) families of b-ZIP transcription factors. AP-1 binds to the TPA response element (TRE, also known as the AP-1 site) and regulates target gene expression in response to various pro-oxidants and toxicants. These gene products mediate (mitigate or promote) oxidative stress and inflammatory responses, as well as cell growth and tumorigenesis [5]. The promoters of many inflammatory response genes, especially those encoding cytokines and chemokines, have functional AP-1 binding sites [6]. Fra-1 regulates gene expression involved in various processes such as cell growth and cell death and regulates the expression of genes controlling tissue/cell remodeling, such as MMP-1, MMP-2, and MMP-9, mainly at the transcriptional level [7–11]. We have recently shown that Fra-1-deficient (Fra-1∆/∆) mice are more susceptible than wild-type (Fra-1+/+) mice to bleomycin-induced fibrosis [12], suggesting that this transcription factor is involved in the regulation of complex genetic networks to maintain cellular homeostasis during bleomycin-induced lung inflammation, injury, and repair processes. Based on these results, we hypothesized that accelerated inflammation and fibrosis observed in Fra-1∆/∆ mice are caused by enhanced inflammatory and fibrotic gene expression. To test this hypothesis and to better understand the mechanisms by which the Fra-1 transcription factor confers pulmonary protection, we have performed microarray analysis to examine the changes in gene expression in the lungs of Fra-1∆/∆ mice after treatment with bleomycin. In the present study, we have evaluated changes in early inflammatory and pro-fibrotic gene expression after 5 days of bleomycin treatment. Our mRNA expression profiling demonstrated increased expression of genes involved in inflammation and immune responses and decreased levels of apoptotic genes in Fra-1∆/∆ mice, suggesting that the Fra-1 transcription factor dampens the development of late fibrotic injury by modulating the early pro-fibrotic responses.

Results and discussion

Genes that encode cytokines, chemokines, and their receptors

The set of genes that was differentially expressed between PBS-treated Fra-1+/+ and Fra-1∆/∆ mice was studied in order to identify those genes for which a genotypic difference in expression exists. We found that the loss of Fra-1 led to an up-regulation of chemokine (C-X-C motif) ligand 13 (Cxcl13, 1.7-fold) and interleukin 1 alpha (Il1a, 1.7-fold) expression. Similarly, we found a down-regulation of chemokine (C-C motif) ligand 5 (Ccl5, -1.6-fold) and chemokine (C-X-C motif) ligand 9 (Cxcl9, -2.7-fold) in Fra-1∆/∆ mice when compared to Fra-1+/+ mice (Table 1). Next, we compared the differentially up-regulated genes between bleomycin-treated Fra-1∆/∆ and Fra-1+/+ mice (Table 2). The genes that showed a -fold change of ≥1.7 were selected for analysis. Interestingly, our data suggested that the lack of Fra-1 leads to up-regulation of cytokines and chemokines in response to bleomycin, including interleukin-1 alpha (Il1a, 1.87-fold), interleukin-2 receptor alpha chain (Il2ra, 2.15-fold), interleukin 2 receptor, beta chain (Il2rb, 1.69-fold), interleukin 6 (Il6, 2.23-fold), chemokine (C-C motif) ligand 8 (Ccl8, 2.27-fold), and C-X-C motif chemokine 10 (Cxcl10, 2.54-fold), whereas Fra-1+/+ mice showed an up-regulation of interleukin 1 receptor, type II (Il1r2, 2.27-fold). These results suggest that Fra-1 signaling controls the expression of some of the genes that are involved in fibrosis. For instance, expression of interleukin 6, a cytokine that promotes greater inflammation and fibrosis [13], was significantly higher in bleomycin-treated Fra-1∆/∆ mice (2.23-fold) than in Fra-1+/+ mice (Table 2). Also, we noted that Fra-1+/+ mice showed an increased expression of Il1r2 (Il1r2, 2.27-fold) in response to bleomycin as compared to Fra-1∆/∆ mice. Interleukin 1 (Il1), a principal pro-inflammatory cytokine that includes two ligands (Il1α and Il1β) and two cell surface receptors namely Il1r1 and Il1r2. Several reports suggest that binding of Il1 to Il1r1 ultimately leads to the activation of several genes, including those encoding cyclooxygenase, nitricoxide synthase, cell adhesion molecules and cytokines and chemokines [14, 15]. More importantly, in mouse models, exogenous administration of recombinant Il1β induced high degree of bleomycin-induced fibrosis, and specific blockade of Il1r1 markedly reduced bleomycin-induced inflammation [16, 17]. Due to the lack of a cytoplasmic-signaling domain for Il1r2, this receptor mainly acts as a decoy receptor to prevent Il1-mediated biological responses [18]. Many anti-inflammatory mediators enhance the expression and release of Il1r2 to induce anti-Il1 pathway [19–21]. The increased Il1r2 expression in Fra-1+/+ but not in Fra-1∆/∆ mice suggests that Fra-1 controls bleomycin-induced inflammation by augmenting the expression of anti-inflammatory genes. In the bleomycin-induced fibrosis model, extensive neovascularization has often been observed to follow the airways and sites of injury [22]. Presence/absence of ERL motif in CXC chemokines dictates their angiogenic property [23]. The amount of Cxcl10 chemokine in the lungs has been directly correlated with the degree of fibrosis [24]. Administration of Cxcl10 to bleomycin-treated mice attenuates pulmonary fibrosis in part due to reduced angiogenesis [25]. However, we found that bleomycin-induced Fra-1∆/∆ mice showed a 2.54-fold increase in Cxcl10 when compared to Fra-1+/+ mice (Table 2).

Table 1 Differentially expressed genes in lung tissues of Fra-1 ∆/∆ and Fra-1 +/+ mice
Full size table
Table 2 Differentially expressed genes induced by bleomycin in lung tissues of Fra-1 +/+ and Fra-1 ∆/∆ mice
Full size table

We then analyzed genes that are uniquely expressed in either Fra-1+/+ (see Additional file 1: Table S1) or Fra-1∆/∆ mice (see Additional file 2: Table S2) after bleomycin treatment. The Fra-1∆/∆ mice showed an up-regulation of chemokine (C-C motif) ligand 19 (Ccl19, 2.07-fold), macrophage-derived chemokine (Ccl22, 2.42-fold), chemokine (C motif) ligand 1 (Xcl1, 2.15-fold), chemokine (C-X-C motif) ligand 11 (Cxcl11, 2.53-fold), chemokine (C-C motif) receptor 4 (Ccr4, 1.75-fold), chemokine (C-C motif) receptor 8 (Ccr8, 2.13-fold) and chemokine (C-C motif) receptor 9 (Ccr9, 2.63-fold) (see Additional file 2: Table S2), whereas Fra-1+/+ mice showed an up-regulation of chemokine (C-X-C motif) receptor 2 (Cxcr2, 2.83-fold), interleukin 1 family, member 9 (Il1f9, 2.43-fold), chemokine (C-X-C motif) ligand 2 (Cxcl2, 3.78-fold), colony stimulating factor 2 receptor, alpha, low-affinity (Csf2ra, 2.02-fold) and interleukin 1 receptor-like 2 (Il1rl2, 1.85-fold) (see Additional file 1: Table S1). Of particular interest is the Ccl22 chemokine, which has been recognized as a Th2 chemokine, and its involvement in the pathophysiology of pulmonary fibrosis has been documented. Belperio et al. have demonstrated that Ccl22 and its receptor, Ccr4, are overexpressed in a mouse model of belomycin-induced fibrosis [26]. Ccl22 and Ccr4 levels are also increased in patients with IPF, and their expression has been detected on epithelium and macrophages, respectively [27]. Neutralization of Ccl22 and Ccr4 has been shown to lead to a significant reduction in lung inflammation during bleomycin-induced fibrosis [26, 28]. Interestingly, our results here showed that the expression of Ccl22 and Ccr4 was significantly increased (2.42- and 1.75-fold, respectively) in bleomycin-treated Fra-1∆/∆ mice when compared to Fra-1+/+ mice.

Next, we analyzed genes that are uniquely down-regulated in bleomycin-treated Fra-1∆/∆ mice (see Additional file 2: Table S2), which included chemokine (C-C motif) receptor-like 2 (Ccrl2, -1.75-fold), bone morphogenetic protein 2 (Bmp2, -1.78-fold), and bone morphogenetic protein 3 (Bmp3, -1.78-fold). Similarly, Fra-1+/+ also showed uniquely down-regulated genes, including chemokine (C-X-C motif) receptor 5 (Cxcr5, -2.38-fold), chemokine (C-C motif) ligand 21A (Ccl21a, -1.72-fold), chemokine (C-C motif) ligand 27a (Ccl27a, -1.75-fold) and ciliary neurotrophic factor receptor (Cntfr, -1.75-fold) (see Additional file 1: Table S1). Taken together, the results of the present study have revealed that genetic disruption of Fra-1 differentially regulates a number of cytokines and chemokines in response to bleomycin, indicating a potential role for Fra-1 in cytokine and chemokine signaling during bleomycin-induced acute lung injury.

Genes encoding proteins that are involved in the inflammatory response

In Fra-1∆/∆ mice treated with vehicle, we found up-regulation of some genes involved in the inflammatory response (Table 1) when compared to their Fra-1+/+ counterparts. These genes included regenerating islet-derived 3 gamma (Reg3g, 7.00-fold), cathepsin E (ctse, 3.61-fold), serine peptidase inhibitor, clade A, member 1B (Serpina1b, 4.61-fold) and serum amyloid A3 (saa3, 1.82-fold). On the other hand, the lack of Fra-1 led to the down-regulation of the expression of a few genes, including serine peptidase inhibitor, clade A, member 3 M (Serpina3m, -2.32-fold) and NLR family, CARD domain containing 5 (Nlrc5, -1.78-fold). Matrix metalloproteases (MMPs) play key roles in tissue repair and remodelling; but recent studies indicate a prominent role for the lysosomal proteinases, such as cathepsins, in the extracellular remodelling [29, 30]. BAL fluid and extracellular space contain secreted cathepsins and their activities are controlled by endogenous inhibitors of cathepsins [31, 32]. Likewise, serine peptidases are implicated in various biological processes such as wound healing and they also contribute to the development of pulmonary fibrosis and acute lung injury [33, 34]. Endogenous serine peptidase inhibitors regulate the activities of serine peptidases. Imbalance in the activities of proteinases/peptidases and endogenous proteinases/peptidases inhibitors may contribute to deregulated protein degradation and resulting in the initiation of lung injury [35]. Serum amyloid A is an acute phase protein, induced by several inflammatory mediators and its serum level is elevated in various conditions like COPD [36], bronchial carcinoma [37], and cardiovascular disease [38]. Thus, it is likely that Fra-1 distinctly regulates proteinases/peptidases and their inhibitor’s expression to maintain lung homeostasis.

When we compared the gene expression patterns of Fra-1∆/∆ mice and their Fra-1+/+ littermates following bleomycin treatment, it was evident that there were many more up-regulated genes than down-regulated genes involved in inflammation in the lungs of Fra-1∆/∆ mice when compared to their Fra-1+/+counterparts (Table 2). We identified differential expression of genes that belong to the immunoglobulin family, specifically immunoglobulin joining chain (Igj, 17.25-fold) and immunoglobulin heavy constant mu (Igh-6, 7.43-fold) in Fra-1∆/∆ mice treated with bleomycin (Table 2). It has previously been shown that immunoglobulin concentrations are increased in immune disorders, such as rheumatoid arthritis [39], inflammatory bowel disease [40], and some respiratory disorders including asthma [41], cystic fibrosis [42], and idiopathic pulmonary fibrosis [43]. The up-regulation of immunoglobulin genes was also accompanied by an up-regulation of several genes involved in antigen presentation and antigen binding. This was true for some of the major histocompatibility genes, including histocompatibility 28 (H28, 3.3-fold), histocompatibility 2, class II, locus Mb2 (H2-DMb2, 1.67-fold), histocompatibility 2, T region locus 24 (H2-T24, 1.81-fold), and histocompatibility 2, Q region locus 8 (H2-Q8, 1.69-fold) (Table 2). Complement components such as complement factor D (CFD, 3.09-fold), and antigens such as CD79A antigen (Cd79a, 2.11-fold), CD2 antigen (CD2, 1.85-fold), and CD4 antigen (CD4, 1.70-fold) were differentially up-regulated in Fra-1∆/∆ mice (Table 2). The expression of other inflammatory genes, including regenerating islet-derived 3 gamma (Reg3g, 4.61-fold) and serum amyloid A 4 (Saa4, 2.81-fold), was also differentially up-regulated in Fra-1∆/∆ mice treated with bleomycin (Table 2). The Reg3g and Saa4 genes have now been consistently associated with pulmonary fibrosis and chronic inflammation [36, 44]. On the other hand, we also noticed down-regulation of some genes in Fra-1∆/∆ mice, such as solute carrier family 11, member 1 (Slc11a1, -1.85-fold), S100 calcium binding protein A8 (S100a8, -2.63-fold), and cathepsin E (Ctse, -2.77-fold) (Table 2), when compared to Fra-1+/+ mice. Again, we analyzed genes that are uniquely expressed in either Fra-1∆/∆ (see Additional file 2: Table S2) or Fra-1+/+ (see Additional file 1: Table S1) mice in response to bleomycin. The results showed that major histocompatibility molecules such as histocompatibility 2, Q region locus 6 (H2-Q6, 2.35-fold), MHC class I like protein GS10 (H2-gs10, 2.12-fold), and complement components such as complement component 4B (C4b, 1.78-fold) were uniquely up-regulated in Fra-1∆/∆ mice treated with bleomycin (see Additional file 2: Table S2). In contrast, we found down-regulation of immunoglobulin heavy constant mu (Igh-6, -8.33-fold), histocompatibility 2, O region beta locus (H2-Ob, -3.12-fold), Cd226 antigen (Cd226, -1.75-fold), and complement receptor 2 (Cr2, -2.22-fold) in Fra-1+/+ mice after bleomycin treatment. Our previous study demonstrated that Fra-1∆/∆ mice showed increased levels of inflammation after bleomycin treatment [12]. Thus, our present data suggest that deregulation of the expression of immune response genes in Fra-1∆/∆ mice is the likely cause of the increased lung inflammation in Fra-1∆/∆ mice.

Genes that encode extracellular matrix and cell adhesion molecules

The unique gene expression pattern in Fra-1+/+ mice (see Additional file 1: Table S1) treated with bleomycin suggested an increase in the expression of genes that encode extracellular matrix, such as collagen, type IV, alpha 1 (Col4a1, 1.96-fold), collagen, type IV, alpha 2 (Col4a2, 1.76-fold), collagen, type VI, alpha 1 (Col6a1, 1.70-fold), collagen, type VI, alpha 2 (Col6a2, 1.85-fold), collagen, type VI, alpha 3 (Col6a3, 1.98-fold), collagen, type XV, alpha 1 (Col15a1, 1.79-fold), a disintegrin-like and metallopeptidase (reprolysin type) with thrombospondin type 1 motif, 1 (Adamts1, 1.94-fold), and a disintegrin-like and metallopeptidase (reprolysin type) with thrombospondin type 1 motif, 2 (Adamts2, 2.07-fold). In contrast, Fra-1∆/∆ mice showed increased expression levels of laminin, alpha 1 (Lama1, 1.75-fold).

Among the various collagens, types I and III collagens are the most widely distributed in both airways and parenchymal structures [45]. To maintain normal structural properties of lung, the controlled distribution of these proteins is vital and their inappropriate accumulation in fibrotic lungs has been reported [46, 47]. In contrast, the physiologic functions and abnormal deposition patterns of other collagens in the lung fibrosis are poorly understood. Our recent study showed an increased expression of the TGF-β1 and type-1-collagen genes in response to bleomycin at the end of 14 and 31 days of bleomycin treatment and demonstrated the presence of increased fibrosis in Fra-1∆/∆ mice [12]. However, we did not observe any differences in the expression levels of type I collagen or type III collagen, nor did we observe altered TGF-β1 gene expression in either genotype at the end of 5 days bleomycin treatment. It has been reported that excessive synthesis and deposition of ECM proteins is a general tissue response to an unresolved chronic inflammation [48]. Hence, we speculate that the persistence of increased inflammation in Fra-1∆/∆ mice is driven by the loss of Fra-1, while higher levels of fibrotic gene expression in the fibrotic stage (14 days) may contribute to the excessive deposition of ECM and disease severity seen in Fra-1∆/∆ mice. This point needs to be addressed in order to better understand the mechanisms underlying the increased fibrosis in Fra-1∆/∆ mice.

We found that a few genes known to be involved in general cell adhesion are affected by bleomycin treatment in Fra-1∆/∆ mice (Table 2). Glycosylation-dependent cell adhesion molecule 1 (Glycam1, 1.98-fold), cd2 antigen (Cd2, 1.85-fold), sialic acid binding Ig-like lectin E (Siglece, -1.78-fold), and integrin alpha M (Itgam, -2.04-fold) were differentially expressed in Fra-1∆/∆ mice treated with bleomycin when compared to Fra-1+/+ mice. In general, the expression of intercellular adhesion molecules is increased by inflammatory signals, which facilitates lymphocytes for higher adhesion and permeation into inflamed tissues. It has been shown that the glycam1 molecule is strongly expressed in inflammatory processes in order to modulate leukocyte trafficking [49]. Sialic acids present on the surface of all mammalian cells and play important roles in physiological and pathological processes, and their expression has been reported to decrease during immune cell activation [50]. Consistent with these findings, we also noted an up-regulation in the expression of glycam1 and a loss of siglece in Fra-1∆/∆ mice when compared to Fra-1+/+mice.

We then analyzed the genes that are uniquely expressed in Fra-1∆/∆ (see Additional file 2: Table S2) or Fra-1+/+ mice (see Additional file 1: Table S1). Fra-1∆/∆ mice showed unique expression of genes that included major histocompatibility molecules such as histocompatibility 2, Q region locus 6 (H2-Q6, 2.35-fold), histocompatibility 2, T region locus 22 (H2-T22, 1.81-fold), histocompatibility 2, K1 region (H2-K1, 1.7-fold), laminin alpha 1 (Lama 1, 1.75-fold), and heparanase (Hpse, 1.98-fold). Fra-1+/+ mice showed unique expression of mesothelin (Msln, 2.16-fold), neuron-glia-CAM-related cell adhesion molecule (Nrcam, 2.1-fold), and a disintegrin and metallopeptidase domain 12 (Adam 12, 2.1-fold). Heparanase is an endoglucuronidase that cleaves heparan sulfate (HS) chains, resulting in HS fragments of 10 to 20 sugar units [51]. Overexpression of heparanase has been reported in numerous tumors, where it regulates angiogenesis and metastasis [52]. Furthermore, the ADAM gene family is associated with proteolytic, cell-cell, and cell-matrix interaction-promoting activities. Several investigations have shown a functional role for ADAMS in collagen deposition in cells and in lungs in which the ADAM gene was knocked down, thus revealing the functional dysregulation of this gene family in the lung fibrosis [53–55]. Thus, an alteration in the expression of cell adhesion molecules may represent another potential mechanism by which more fibrosis can occur in Fra-1∆/∆ mice.

Fra-1-regulated genes involved in programmed cell death

The process of programmed cell death is known to play a major role in maintaining many biological processes, and inappropriate apoptosis can lead to disease conditions, either because cells experience an inappropriately prolonged survival or they die prematurely [56, 57]. Studies using genetic models have demonstrated both cooperative and antagonistic roles of AP-1 family of proteins in modulating cell death in response to a variety of pro-apoptotic stimuli. For example, c-Jun-/- mouse embryonic fibroblasts (MEFs) and liver cells show increased levels of oxidative stress and apoptosis [58]. Likewise, c-Fos also participates in both pro- and anti-apoptotic activities. For example, c-Fos-/- MEFs undergo apoptosis when cultured in vitro and also display an increased susceptibility to UV-induced cell death [59]. Overexpression of Fra-1 also inhibits proliferation, induces apoptosis, and reduces the tumorigenicity of c6 glioma cells [60]. Consistent with a role for Fra-1 in apoptosis, we recently found that mouse embryonic fibroblasts lacking Fra-1 show an increased resistance to oxidant-induced cell death [61]. Fra-1 appears to uniquely up-regulate some genes modulating apoptosis in Fra-1+/+ mice (see Additional file 1: Table S1), including paternally expressed 3 (Peg3, 2.55-fold), the tumor necrosis factor receptor superfamily, member 10b (Tnfrsf10b, 2.03-fold), AXL receptor tyrosine kinase (Axl, 1.73-fold), Eph receptor A2 (Epha2, 1.7-fold), zinc finger matrin type 3 (Zmat 3, 1.76-fold), solute carrier family 40 (iron-regulated transporter, member 1) (Slc40a1, 1.70-fold), and EGL nine homolog 3 (Egln3, 1.81-fold). Similarly, Fra-1∆/∆ mice showed (see Additional file 2: Table S2) up-regulation of glutamate-cysteine ligase, catalytic subunit (GCLC, 1.70-fold) and down-regulation of lectin, galactose binding, soluble 12 (Lgals 12, -1.78-fold), Eph receptor A7 (Epha 7, -1.72-fold), and arachidonate 12-lipoxygenase (Alox12, -1.69-fold). It has been reported that kaempferol exerts an anti-oxidative and anti-apoptotic effects in HEI-OC1 cells treated with cisplatin through enhancing GCLC expression [62]. Consistent with this observation, we found that GCLC was induced in Fra-1∆/∆ mice. Our gene expression results from Fra-1∆/∆ mice are therefore in good agreement with the observation that inappropriate apoptosis can lead to exaggerated lung fibrosis.

Validation of microarray data

Amongst several genes that were significantly affected by bleomycin, we randomly selected 17 genes according to the microarray results to confirm their differential expression by qRT-PCR. We confirmed that bleomycin treatment significantly induced the expression of Il1a, Irf4, Reg3g, and Ccr4 and reduced the expression of S100a8 in Fra-1∆/∆ mice when compared to similarly treated Fra-1+/+ mice (Table 3). These results confirmed the expression patterns of the microarrays. Next, we analyzed the genes that were uniquely expressed in both genotypes. The results revealed that Fbln2 expression was significantly higher in Fra-1+/+ mice treated with bleomycin than in the vehicle-treated control group (Table 4); however, there was no difference observed in Fra-1∆/∆ mice. Similarly, Fra-1∆/∆ mice treated with bleomycin also showed significantly increased expression of Hpse, Gclc, Runx3, Xcl1, and Cxcl11 when compared to the vehicle-treated group, but no differences were observed with their wild-type counterparts. These results further confirmed the expression patterns of the microarrays. According to the qRT-PCR results, there was a tendency for increased expression of Ccr8 and Ccl22 in the Fra-1∆/∆ mice treated with bleomycin as compared to the vehicle-treated mice, but the differences were not significant. Similarly, wild-type mice treated with bleomycin showed greater expression of Col4a1 and Tnfrsf10b than did the vehicle controls. While these qRT-PCR results agree in general direction with the trends measured by microarray, the results did not reach statistical significance. Comparison of microarray and qRT-PCR results revealed discordance in 2 of the 17 genes selected. In the case of Marco, microarray analysis revealed no change in Marco expression in Fra-1+/+ mice but a significant increase in bleomycin-treated (3.08 ± 0.11) Fra-1+/+ mice when compared to vehicle-treated mice, as assayed by qRT-PCR. In the second case, the microarray results showed no change in the expression of Snai2 in Fra-1+/+ mice, but the real-time RT-PCR results indicated a significant decrease (0.37± 0.15) in expression in these mice. The discordance in these two genes may be explained by the lower sensitivity of the microarrays.

Table 3 Validation of differentially expressed genes induced by bleomycin in lung tissues from both Fra-1 +/+ and Fra-1 ∆/∆ mice
Full size table
Table 4 Validation of uniquely expressed genes induced by bleomycin in lung tissues from both Fra-1 +/+ and Fra-1 ∆/∆ mice
Full size table

Analysis of selected microarray genes at different time points after bleomycin treatment

To identify the time course of gene induction by bleomycin, we analyzed samples at different time points for selected genes. All the genes that were used for microarray validation were also used for our analysis of temporal patterns of gene expression (Figure 1). The results revealed that most of the genes continued to show no significant differences between Fra-1+/+ and Fra-1∆/∆ at 7 and 14 days after bleomycin treatment. However, some of the genes were predominantly up-regulated (Xcl1 and S100a8) or down-regulated (Cxcl11 and Ccl22) in Fra-1∆/∆ mice only at 7 days following bleomycin treatment, when compared to the vehicle-treated control and Fra-1+/+ mice. We were further interested in the delayed response of some of the genes involved in lung fibrosis that did not exhibit significant differences in microarray results at 5 days (Figure 2). The results revealed a significant difference in the induction of C-C chemokines such as Ccl2, Ccl3, and Ccl8 between Fra-1+/+ and Fra-1∆/∆ mice: Fra-1∆/∆ mice showed significantly elevated expression at 7 days following bleomycin treatment when compared to wild-type mice. Interestingly, excessive production of Ccl2, Ccl3, and Ccl8 has been shown to aid the development of pulmonary fibrosis [63, 64]. In addition to the up-regulation of Ccl2, Ccl3, and Ccl8, we observed a marked down-regulation of Il10 and Il33 in bleomycin-treated Fra-1∆/∆ mice. Previous studies have reported that in vivo IL-10 gene delivery before [65] and after bleomycin [66] administration suppresses the development of pulmonary fibrosis. Finally, fibroblast transdifferentiation has been shown to contribute to the pathology of pulmonary fibrosis. Therefore, we analyzed the expression of α-SMA, a marker for myofibroblasts. The results revealed that Fra-1∆/∆ mice treated with bleomycin had a significantly higher expression of α-SMA than did vehicle-treated control or Fra-1+/+ mice at 14 days. This result further supports the enhanced susceptibility of Fra-1-null mice to bleomycin-induced lung fibrosis.

Figure 1
figure 1

Validation of the differences in mRNA expression between Fra-1+/+and Fra-1Δ/Δmice for selected microarray genes at different time points after bleomycin treatment. Lung mRNA abundance was determined by quantitative real-time RT-PCR. The graphs represent the fold change over vehicle treated Fra-1+/+ controls after normalization with the expression of GAPDH. Results are mean ± SD for 3-4 mice in each group. *p<0.05, PBS vs bleomycin; †p<0.05, Fra-1Δ/Δ vs Fra-1+/+ mice.

Full size image
Figure 2
figure 2

Analysis of some gene expression between Fra-1+/+and Fra-1Δ/Δmice at different time points after bleomycin treatment. Lung mRNA abundance was determined by quantitative real-time RT-PCR. The graphs represent the fold change over vehicle treated Fra-1+/+ controls after normalization with the expression of GAPDH. Results are mean ± SD for 3-4 mice in each group. *p<0.05, PBS vs bleomycin; †p<0.05, Fra-1Δ/Δ vs Fra-1+/+ mice.

Full size image

Conclusion

The factors that contribute to the pathogenesis of pulmonary fibrosis include persistent inflammation, generation of pro-inflammatory, pro-fibrotic and angiogenic mediators, alveolar epithelial cell injury, fibroblast differentiation, and poor apoptotic activity of the myofibroblasts. These deregulated cellular processes eventually lead to excessive deposition of extracellular collagen and pathological fibrosis [48, 67, 68]. The present mRNA expression profiling analysis has revealed an important role for Fra-1 in regulating components of complex regulatory networks controlling the lung injury and fibrosis. We found that Fra-1∆/∆ mice displayed some of the factors that contribute to pulmonary fibrosis, such as increased expression of pro-inflammatory genes and decreased expression of genes involving in apoptotic process during bleomycin treatment. Thus, we propose that strategies enhancing Fra-1 functions may represent a promising approach to mitigate pulmonary fibrosis.

Methods

Mice

Conventional deletion of Fra-1 is embryonic lethal due to extra-embryonic tissue defects [69]. The mice bearing Fra-1 “floxed” allele [69] (hereafter referred as Fra-1FF mice) were obtained from Erwin F. Wagner (Spanish National Cancer Research Centre, Madrid, Spain). These mice are maintained on a mixed (C57BL6/129) background. Meox2 (Sox2)-Cre transgenic mice (C57BL6/129), in which Cre expression specifically restricted in embryo but not in extra-embryonic tissues, were obtained from the Jackson Labs. Meox2 Cre mice were crossed to Fra-1F/F mice [70], in order to obtain Fra-1F/F-Meox2-Cre mice as described earlier [12]. Fra-1F/F mice with and without Cre are hereafter referred to as Fra-1∆/∆ and Fra-1+/+ genotypes, respectively.

Bleomycin treatment

Bleomycin (0.075U) (APP Pharmaceuticals, LLC, Schaumburg, IL, USA) diluted in 30 μL of PBS was intratracheally administered to mice (n=3) (10-14 weeks old) as described previously [71]. Mice treated with PBS (n=3) served as controls. All experiments were conducted under a protocol approved by the institutional animal care use committee of the University of Illinois at Chicago. At the end of 5 days treatment, the left lungs were frozen immediately in RNAlater (Ambion) for subsequent microarray and qRT-PCR analysis.

RNA isolation and labeling

Total RNA was isolated from Fra-1+/+ and Fra-1∆/∆ mice administered with PBS and bleomycin using Qiagen RNeasy micro kit (Cat no. 74004). RNA concentration and purity was determined before gene expression profiling using the Affymetrix MoGene 1.0ST v1 Array (gene array) (Affymetrix, Inc., Santa Clara, California). The microarray labeling, hybridization and processing was performed at the University of Illinois Research Resource Center according to the manufacturer’s protocol. Microarray data have been deposited in the National Center for Biotechnology Information Gene Expression Omnibus database (Accession number: GSE43695).

Microarray data analysis

The raw probe signal intensities were quantile normalized over all samples, summarized with the robust multi-array average (RMA) algorithm [72] and log2 transformed with a median polish, using the Affymetrix Power Tools. We considered a transcript cluster (gene-level) to be reliably expressed in a sample if the Affymetrix implemented DABG (detection above ground) p-value was less than 0.05. We used local-pooled-error (LPE) estimates and robust statistical tests [73] for evaluating significance of each gene’s differential expression in a comparison (e.g., wild-type vs. wild-type with bleomycin treatment). The LPE estimation was shown to be powerful and effective in case of a small number of replicate arrays [73]. False discovery rate (FDR) was controlled at 1% using the LPE library for the R Statistical Package [74].

Pathway analysis

We searched for any enriched pathways and biological processes among the differential genes in each comparison relative to the genes covered on the gene expression profiling platform using the NIH/DAVID (The Database for Annotation, Visualization and Integrated Discovery) [75]. Particularly, the following databases were interrogated: KEGG (Kyoto Encyclopedia of Genes and Genomes) [76] and GO (Gene Ontology) [77]. A minimum of 5 genes and the Benjamini corrected p-value less than 0.01 were used to call significantly enriched pathways or biological processes. There are distinct temporal phases during bleomycin-induced lung injury and fibrosis. To dissect the differential gene expression during bleomycin-induced initial lung injury, we have analyzed the gene expression profiles in Fra-1+/+ and Fra-1∆/∆ mice given PBS or bleomycin. We then compared the gene expression profiles in various categories: (1) differentially expressed genes in the lung tissue of Fra-1∆/∆ mice vs. Fra-1+/+ mice (Table 1), (2) differentially expressed genes induced by bleomycin in the lung tissue of Fra-1+/+ vs. Fra-1∆/∆ (Table 2), (3) unique gene expression induced by bleomycin in the lung tissue of Fra-1+/+ mice (see Additional file 1: Table S1), and (4) unique gene expression induced by bleomycin in the lung tissue of Fra-1∆/∆ mice (see Additional file 2: Table S2). The resulting gene lists were divided into several categories based on functional analysis in order to dissect Fra-1-dependent and -independent transcriptional programs.

Validation of microarray analysis

Total RNA (1 μg) was reverse transcribed using qScript cDNA super mix (Quanta Biosciences, Inc. Cat no. 95048-100). qRT-PCR was performed using fluorogenic SYBR Green and detection system (Applied Biosystems). PCR was performed using primers listed in (Additional file 3: Table S3). For Gclc, Marco and α-SMA, TaqMan gene expression assays were purchased from Applied Biosystems (Foster City, CA). The cycle threshold (CT) values for each gene were normalized to that of GAPDH, and the relative value for PBS treated Fra-1+/+ was set as one arbitrary unit (AU). Values are shown as mean ± SD, with n=3-4 for each experimental condition. Student’s T test was used and p ≤ 0.05 was considered significant.