{"paper_id":"063e1480-feaf-4fba-a583-615ec6e83ba4","body_text":"A Tale of Two Mice: genetics of model mouse strains suggest a transcriptional basis for risk and resistance in idiopathic pulmonary fibrosis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article A Tale of Two Mice: genetics of model mouse strains suggest a transcriptional basis for risk and resistance in idiopathic pulmonary fibrosis Thea Fennell, Ieva Beržanskytė, Rihab Gam, Wencan Zhu, Minkyung Sung, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5295459/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Idiopathic pulmonary fibrosis (IPF) is a terminal inflammatory lung disease that causes permanent scarring (fibrogenesis). Bleomycin (BLM) is a drug used to induce fibrosis in mouse models, typically C57BL/6. However, meta-analyses show inter-strain heterogeneity in response, e.g. resistance in BALB/c. This study extends transcriptomic analysis of IPF to a resistant strain, qualifying inferences from the standard model and suggesting genetic risk factors to inform clinical research. Methods Transcriptomic datasets were generated from C57BL/6 and BALB/c mice. Test mice were administered BLM – with tissue samples sequenced from control, test, and contralateral lungs at the fibrogenesis stage of the BLM model (7–14 days after injection). Differentially expressed genes (DEGs) were calculated between treatments and strains, followed by gene network and transcription factor (TF) target enrichment analysis of DEGs. Additionally, strain-specific genetic variants were identified in fibrosis-related genes, complemented by analysing human genome-wide association (GWAS) datasets. An in vitro model of TGF𝛽-stimulated stress fibre deposition was used in parallel to confirm transcriptomic findings. Results DEGs calculated between treatment groups were enriched for general fibrosis-related processes across strains. Some fibrogenic processes and functional modules, however, were specifically enriched in C57BL/6, which was orthogonally validated by in vitro TGFβ assays. Conversely, the anti-fibrotic DEG Ctsk was upregulated under fibrosis in BALB/c specifically. TF target enrichment analysis of cross-strain and cross-treatment DEGs, using perturbation data, further identified them as significantly overrepresentative of FOSL1-sensitive genes. Subsequent genetic analysis revealed a non-conservative variant (P170L) located in BALB/c FOSL1. Furthermore, analysis of data from the 100,000 Genomes Project associated human FOSL1 variants with IPF. Conclusions Transcriptional differences in IPF have been characterised for C57BL/6 and BALB/c strains, supporting the consensus on IPF resistance in BALB/c. Analysis of gene set expression within and between strains principally implicates genes sensitive to the TF FOSL1. The significance of this novel finding is amplified by the discovery of a highly non-conservative P170L mutation in the bZIP domain of BALB/c FOSL1. Mechanistic investigation of FOSL1 activity – and potentially other regulators, e.g. Nos2, Il6 – is thus recommended as preclinical IPF research. idiopathic pulmonary fibrosis C57BL/6 BALB/c FOSL1 bleomycin disease resistance mouse model genetics transcriptomics gene regulation Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Background Fibrosis is a multisystem disease that can affect a range of tissues; it is driven by immune cell recruitment and uncontrolled epithelial ‘repair’ processes (Fig. 1 ). Clinically, the condition is characterised by ongoing inflammation and epithelial-to-mesenchymal transition (EMT), e.g. of alveolar cells, accompanied by thickening of the extracellular matrix (ECM); also known as fibrogenesis ( 1 – 3 ). When this disease emerges within human lung interstitium, it is typically terminal and often without obvious cause – idiopathic pulmonary fibrosis (IPF) being the most common type of lung fibrosis ( 3 , 4 ). However, pulmonary injury is a plausible trigger. IPF is typically symmetric in humans, perhaps due to propagation of fibrosis from diseased to contralateral lungs. Indeed, asymmetry is a marker for lower survival rates ( 5 – 8 ). That said, evidence from rodent models also lends credence to the hypothesis of compensatory growth of contralateral – or even diseased – lung volume as a response to fibrogenesis ( 5 , 9 ). Preclinical research into IPF has focused primarily on murine models, with the reference strains – C57BL/6J and closely related strains, e.g. C57BL/6N – being especially popular ( 5 , 10 , 11 ); here collectively termed C57BL/6. However, experimentation in alternative genetic backgrounds has identified divergence within IPF phenotypes. Their relative resistance or susceptibility is defined by the severity of pathological symptoms – unresolved collagenic fibrogenesis (scarring) in particular ( 1 ). Within this schema, C57BL/6 strains are classified as susceptible, whereas BALB/c strains are resistant – despite similar immune responses under fibrosis induction ( 12 ). Conversely, C57BL/6 is resistant to hepatic fibrosis, to which BALB/c is susceptible, illustrating that genetic susceptibility can furthermore be tissue-dependent ( 1 ). Regarding murine pulmonary fibrosis, resistance may actually be the default. For strains where IPF phenotype has been evaluated, only C57BL/6 mice are fibrosis-susceptible – in contrast to BALB/c, DBA/2, Swiss, A/J, and C3H ( 1 ). C57BL/6 has long been known to exhibit a high degree of resistance to radiation-induced pneumonitis ( 13 ). This likely factored into its initial development as an inbred strain for oncology research ( 14 ). Thus the convention of C57BL/6 as ‘standard’ may, paradoxically, be due to its exceptional lung physiology. Unlike humans, most mice are able to recover from fibrosis (Fig. 1 ), albeit in a strain-dependent way. Whilst C57BL/6 strains can resolve scars despite repeated lung injury, DBA/2 strains subjected to the same treatment accumulate fibrotic tissue and develop prolonged disease ( 15 ). Some researchers have posited a negative correlation between T helper 2 (Th2) cell population size and mouse strain ability to resolve lung fibrosis. DBA/2 and BALB/c immunity is biassed towards Th2 cells, whilst C57BL/6 is biassed towards Th1 cells. As the cell types produce distinct cytokine profiles, this immunological divergence has been proposed to differentiate IPF phenotype ( 15 ). Bleomycin (BLM) and Transforming Growth Factor Beta (TGFβ) are agents commonly used to induce fibrosis in IPF models. BLM triggers pathogenesis via DNA damage (Fig. 1 ), whilst TGFβ acts downstream as a central node in the ECM deposition pathway ( 12 , 16 ). TGFβ is one of the key pro-fibrotic cytokines, and has also been used to induce fibrosis in rodent lung tissue in vivo and in vitro ( 12 , 17 , 18 ), as well as in human cell culture in vitro ( 19 ). These distinct modes of fibrosis induction could make research susceptible to model choice, given murine genetic variation. It has been hypothesised that BLM-sensitivity underpins BALB/c resistance to IPF. Early research suggested that BALB/c, in particular, may display a BLM-specific resistance due to higher basal expression of the detoxifying enzyme bleomycin hydrolase (BLMH) ( 20 ); it has further been demonstrated that BALB/c fails to express the fibrosis-characteristic Connective Tissue Growth Factor (CTGF) in response to BLM, but may yet do so in response to TGFβ ( 21 ). If the BLM-specific resistance hypothesis were true, it would be expected that the delivery of fibrogenic factors downstream of injury could induce full fibrosis in this strain (Fig. 1 ). However, not only does CTGF alone fail to induce fibrosis in BALB/c, but combined CTGF + BLM treatment results in full fibrosis ( 22 ). Moreover, recent work on strain-specific responses conflicts with the BLM-sensitivity paradigm, by showing that BALB/c – unlike the control C57BL/6 strain – fails to upregulate the pro-fibrotic collagenase Mmp1 in response to TGFβ treatment ( 12 ). There is thus increasing evidence for a more general resistance mechanism. Much of the challenge in characterising fibrotic susceptibility arises from the disease’s genetic complexity. A recent meta-analysis identified 213 unique factors as pulmonary fibrosis-related genes (PFRGs; Table 1 ). They are enriched for age-related genes, specifically: pro-fibrotic genes were pro-ageing, and anti-fibrotic genes were anti-ageing ( 23 ). It is also likely that this analysis underestimates the contributing genetic factors, as it omits several markers identified elsewhere, including structural proteins and matrix metalloproteinases ( 24 – 26 ). A biological interpretation of genetic susceptibility to fibrosis may benefit from a genomic, rather than a purely gene-level, strategy. Longitudinal observations of transcriptomes in the C57BL/6 BLM-IPF model support this approach, as enrichment analysis of the differential expression profiles – comparing healthy and fibrotic tissue – recapitulates the clinically observed processes of inflammation and ECM over-synthesis ( 10 ). Table 1 Pulmonary fibrosis-related gene sets investigated in the context of strain-specific fibrosis responses Set Members Pulmonary fibrosis-related genes (PFRGs) See Toren et al. (2021) e.g. Ackr2, Zmpste24 Extended PFRGs See Toren et al. (2021), Todd et al. (2020),Blaauboer et al. (2014) & Onursal et al. 2021 e.g. Mmp13 Common PFRGs Timp1, Spp1, Cxcl10, Ccl2, Retnla C57BL/6-specific PFRGs Il10, Il17a, Nos2, Rag1, S100a4 BALB/c-specific PFRGs Klf4, Igf1, Ctsk, Cx2cr1, Hspa1a, Hspa1b C57BL/6-specific interleukins Il10, Il17a, Nos2, Rag1, S100a4 BALB/c-specific interleukins Il6, Il9, Il12b, Il22 Marker matrix proteins See Blaauboer et al. (2014) & Onursal et al. (2021), e.g. Col1a1 In this study, a transcriptome-level analysis is applied to a dataset comprising bulk RNA-seq reads from a BLM-IPF model using two distinct strains: C57BL/6 and BALB/c. As the former is C57-derived – the latter being from Castle’s mice – they are genetically divergent strains, as well as differing in fibrosis susceptibility ( 1 , 27 ). This dataset therefore captures both the genotypic and phenotypic heterogeneity of IPF model mice. Differential expression and enrichment analyses of the data were consistent with the findings of previous research in C57BL/6, with fibrosis in each genotype being characterised by immunological and epithelial processes ( 10 ). However, there was also variation in the identity and functional profiles of the genes differentially expressed under fibrosis in the two genetic backgrounds, consistent with the reported differences in C57BL/6 versus BALB/c IPF risks ( 1 ). These sets of differentially expressed genes (DEGs) included a number of interleukins and other key PFRGs established in the literature (Table 1 ), some of which re-emerged in this study during DEG network analysis. BLM-induced DEGs and cross-strain DEGs – the latter in both control and BLM-treated groups – also showed highly significant enrichment for genes sensitive to one transcription factor (TF): FOSL1. This was in keeping with literature establishing FOSL1 as a major factor in modulating IPF – and it built upon it, suggesting that FOSL1 may play a principal role in differentiating C57BL/6 and BALB/c transcriptomes, including under fibrosis. The hypothesis of FOSL1 forming a transcriptional basis for BALB/c resistance to IPF was further developed by genetic analysis, utilising the Sanger Mouse Project database. These genetic analyses also undermined existing BLM-dependent and Th2-bias hypotheses for BALB/c IPF response ( 15 , 20 ). These analyses were complemented by TGFβ assays in vitro , comparing strain IPF response under this independent experimental model. Methods Bleomycin mouse model for lung fibrosis Adult mice from three strains were administered 1 mg/ml BLM (BI3543, Chemtronica Sweden) in saline as a single 20 µg/mouse dose intratracheally into their left lung to model pulmonary fibrosis: BALB/c, C57BL/6N, and C57BL/6J. Saline-treated (control) mice were also maintained for each population. Doses were delivered as 30 µL of bleomycin solution followed by 40 µL of air. Tissue was sampled from healthy, fibrotic, and contralateral lungs between days 7 and 14; corresponding to expected fibrogenic period of the treated lung tissue. RNA sequencing of murine tissues Collected tissue samples were prepared by snap-freezing and then grinding to a fine powder using a pre-chilled mortar and pestle. 1 ml of Trizol was added per 50–100 mg of tissue and further shredded by passing through a 21G needle. 100 ng RNA was used to prepare an indexed library with the NuGEN Ovation RNA-seq V2 kit (7102, Tecan) and the Ovation Ultra Low kit (S02366A-FG, Tecan) following manufacturer protocol. Libraries were sequenced using HiSeq 4000 (single-end 50 bp). RNA read quality control and alignment Read quality checking was conducted using FastQC (v0.11.9) both before and after trimming with Trimmomatic (v0.39; leading = 3, trailing = 3, slidingwindow = 4:15, minlen = 36). Samples with lower than 10 million total reads were resequenced and the resulting technical replicates concatenated. Subsequently, reads were aligned to the mouse genome; genome annotation was downloaded from the Gencode database (vM23) and chromosomal reference sequence from the UCSC goldenPath database (mm10). Alignment and indexing were both conducted using STAR (v2.7.9a; runTHreadN = 10, sjdbOverhang = 49). Finally, aligned reads were annotated using featureCounts (v2.0.1;) and all steps checked with MultiQC (v1.11). All subsequent analyses of read data, including visualisation of MultiQC output, was conducted within RStudio (v4.1.1 – “Kick Things”). Anomalous samples as identified from hierarchical clustering and dimension reduction were excluded from further analysis: 5 samples were excluded as a result, all of which were C57BL/6N, 3 being BLM-treated and 2 being saline-treated. As this significantly reduced the statistical power of the C57BL/6N treated group (n = 2), C57BL/6J was the focus of downstream analysis on C57BL/6. Linear dimension reduction and clustering rlog-normalised transcriptomes were characterised through principal component analysis, using the default R-package stats (v4.1.1). PCA results were visualised using ggplot2 (v3.3.5). Unsupervised hierarchical clustering and correlation analyses were conducted and visualised using the same software; misclustered samples were excluded as anomalous and all expression analyses rerun without them. Differential expression analysis Read count matrices were filtered for protein-coding autosomal genes and parsed by DESeq2 (v1.34.0), then (a) normalised using the regularised log transformation (rlog) and – independently – (b) used to calculate differential expression within and between each strain. Heatmaps for rlog-transformed absolute expression data were generated using pheatmap (v1.0.12) and box plots using ggplot2 (v3.3.5). Strain-specific DEGs were determined for each strain (BALB/c, C57BL/6J, C57BL/6N) independently. Results for the two C57BL/6 sub-strains were later aggregated to provide a set of C57BL/6-specific DEGs for network analysis. Ontology term enrichment analyses Differential expression data was tested within each combination of strain and treatment condition, using clusterProfiler (v4.2.2) to evaluate enrichment of statistically significant DEGs (p < 0.05; |logFC|>2). Results were visualised with enrichplot (v1.14.2). GO biological process enrichment ( 28 , 29 ) was visualised, using enrichGO (OrgDb = org.Mm.eg.db), for GO terms known to be associated with fibrosis pathology based on Deng et al. (2020) and Wang et al. (2021), where the adjusted p-value was > 0.05 ( 3 , 10 ). KEGG pathway enrichment was visualised for all terms where the adjusted p-value was > 0.0005 ( 30 ). Textual reporting of network cluster enrichment excludes enrichment for categories where the adjusted p-value was > 10 − 4 . Perturbation enrichment analysis Cross-strain, cross-treatment, and strain-specific cross-treatment DEG sets were used as input to transcription factor (TF) perturbation enrichment analysis (enrichR, “TF Perturbations followed by expression”) ( 31 – 33 ). Terms for this analysis are derived from a meta-analysis of studies conducting TF perturbations with follow-up RNA-seq – thus encompassing gene sets sensitive to specific TFs. enrichR calculates significance using Fisher’s exact test assuming a binomial distribution; p-values adjusted for multiple testing corrections (Benjamini-Hochberg) are then calculated. Protein interaction network construction Networks were constructed for protein-protein interactions between the products of defined gene sets, using STRINGdb (v2.6.5; network_type = functional, required_score = 500, network_flavour = confidence) supported by rbioapi (v0.7.4) ( 34 , 35 ). DEG scoring for construction of prioritised protein-protein interaction networks utilised the geneD metric (geneD = -log10(p-value)*|logFC|) to account for both significance and fold-change of expression ( 36 ). Louvain algorithm implementation Network nodes were clustered using the Louvain algorithm, implemented by igraph (v1.3.0). This was conducted using the full set of differentially expressed genes for each network, prior to filtering by geneD score. Strain-specificity of a module for a given strain X (‘parent’ module) was evaluated as the largest proportion of that module’s nodes present together in a single module in the strain-specific network for X (‘child’ module); this corresponds to a single stochastic iteration of the Louvain algorithm. Specificity was considered significant where the conserved node proportion in a child module was > 45%, for a parent module of size > 10 nodes. Subsequent KEGG enrichment analysis was conducted on the parent module. Single nucleotide polymorphism (SNP) calling and variant effect predictions Genetic regions of interest were queried for variants using the online interface for the Mouse Genomes Project (REL-1505; GRCm38), filtering by strain for C57BL/6NJ, BALB/cJ, and DBA/2J ( 37 ). Variant effect prediction was conducted with the Ensembl VEP ( 38 ). Gene sequences required for downstream analysis were extracted using the Multiple Genome Viewer (MGV) from Mouse Genome Informatics (MGI) ( 39 ). Genome-wide association study data mining Genome-wide association study (GWAS) data on familial pulmonary fibrosis (FPF) was extracted from Genomics England’s 100,000 Genomes Project ( 40 ). Variants were filtered by gene to analyse data on Mmp13 , Ctsk , Il6 , Fosl1 , Nos2 , and Nos3 . Association significance of these variants was assessed with respect to age of death, as a metric to capture disease progression. The result was visualised by Manhattan Plot to clearly depict the significance of each SNP across the genome. Data was similarly extracted from an AstraZeneca-funded whole-exome study of IPF ( 41 ). Protein structure predictions and alignment The 3-D structures of FOSL1 were modelled using ColabFold (template_mode = pdb70, num_recycles = 12) ( 42 , 43 ). Output PDB files were visualised and analysed using iCn3D Structure Viewer (v3.11.6) and PyMOL (v2.5.2). In vitro TGFβ-1 model for lung fibrosis Mouse Lung Fibroblasts (MLFs) from BALB/c and C57BL/6 mice were purchased from Generon and cultured up to passage 7 in complete fibroblast medium with supplements (M2267, Generon) on a gelatin-based coated surface (6950, Generon) at 37°C, 5% CO 2 . For fibrosis modelling, cells were plated at 30,000/cm 2 density. After 24h, cells were sub-cultured in serum-free complete fibroblast media, and after another 24h, mouse recombinant TGFβ-1 (ABclonal, RP0116) was added for 72h at 0.1–10 ng/ml concentration. MLFs from both strains were confirmed for the strain-specific FOSL1 mutations using Sanger Sequencing. Quantitative Polymerase Chain Reaction (qPCR) Cells were lysed after several washes with PBS in plate using RLT buffer with 0.1% β-mercaptoethanol. After RNA extraction with an RNeasy kit (Qiagen), cDNA was synthesised using M-MuLV reverse transcriptase (M0253L, NEB). Quantitative real time polymerase chain reaction (qPCR) was performed using Fast SYBR 2x Green Master Mix (ThermoFisher Scientific). Immunocytochemistry Cells were fixed with 4% paraformaldehyde for 15 min at room temperature, and stained with anti-𝛼-sma antibody (ab184675 at 1:200, Abcam)and anti-collagen 1 (ab34710) in 3% bovine serum albumin in 0.1% Triton solution at 4℃ overnight. After PBS washes, 20 min DAPI staining (1:1000 of 1 mg/ml in PBS) was performed. Results Bleomycin-treated lung transcriptomes reflect multicellular fibrotic processes Comparison of transcriptomic data from fibrotic versus healthy murine lung tissue shows a clear and consistent difference in gene expression. Principal component analysis separates bleomycin-treated samples from saline-treated controls along the first component (PC1, 14.9%; Fig. 2 A), whilst the second component separates two distinct genetic strains of mice: BALB/c and C57BL/6. PC1, describing the strain-independent response to fibrosis induction, is enriched for fibrosis-associated processes such as the inflammatory response and extracellular matrix organisation (Fig. 2 B). The set of upregulated genes contributing to PC1 were also enriched for many KEGG terms known to be correlated with fibrosis progression ( Supplementary Fig. 1 ). Many enriched terms matched those recently identified by Wang et al. (2021) including: lysosome, cytokine-cytokine receptor interaction, haematopoietic cell lineage, leishmaniasis, rheumatoid arthritis, osteoclast differentiation, tuberculosis, phagosome, intestinal immune network for IgA production, chemokine signalling pathway, Toll-like receptor signalling pathway, p53 signalling pathway, amoebiasis, and protein digestion and absorption ( 10 ). Whilst both strains upregulate IPF-associated processes, however, their precise overlap is only 48 differentially expressed genes (DEGs). That equates to: <20% of total DEGs for BALB/c, < 10% for C57BL/6. This set includes 5 pulmonary fibrosis related genes (PFRGs); the set does also include genes additionally identified within the extended PFRG list (Table 1 ). That said, a number of fibrotic DEGs are strain-specific in identity, but still contribute towards common or similar biological processes, as defined by GO and KEGG ontologies (Fig. 2 ; Supplementary Fig. 1 ). The majority of transcriptomic variation in contralateral versus healthy tissue is attributable to factors uncorrelated with treatment condition ( Supplementary Fig. 2 ). C57BL/6J samples separate out only along a combination of the third (PC3, 11.8%) and fourth components (PC4, 9.9%) and BALB/c samples along PC3 (15.6%); C57BL/6N samples do not cluster by sample type under PC1-4 ( Supplementary Fig. 2 ). Consequently, the contralateral lung in all strains is not significantly different from the healthy lung within this model. Analysis of this sample type at the level of gene expression was therefore not pursued further, given the lack of clear support for a contralateral pathology. Fibrogenic markers are upregulated in the reference mouse strain relative to BALB/c The second principal component of variation between fibrotic and control transcriptome data (PC2, 14.1%; Fig. 2 A) separates samples along strain, independent of disease status. This is consistent with the observed distinctions in functional enrichment between the upregulated gene sets of each strain (Fig. 2 B). Whilst both strains are indicated to have upregulated immunological processes, there is some strain-differential enrichment for fibrogenic processes (Fig. 2 B). Specifically, C57BL/6 displays upregulation of extracellular matrix organisation and collagen biosynthesis (Fig. 2 B; GO), as well as protein digestion and absorption (KEGG; Supplementary Fig. 1 ); the latter has previously been implicated in late – but not early – stages of BLM-induced murine IPF ( 10 ). In contrast, BALB/c exhibits a more significant inflammatory response, as well as EMT enrichment (Fig. 2 B). However, this analysis does not indicate a corresponding BALB/c-specific elevation of ECM deposition, which the literature would predict as a consequence of elevated level of EMT ( 3 ). The strain dependency of ECM deposition and stress fibre formation was also observed in a TGFβ-based fibrotic assay in vitro. Lung fibrosis was modelled using primary mouse lung fibroblasts cultured from adult C57BL/6 and BALB/c mice, with the cytokine TGFβ-1 applied as the inducer of profibrotic inflammatory signalling. Just as observed in a human lung fibroblast model in the literature ( 19 ), an increasing stress fibre deposition was observed with increasing TGFβ-1 concentration in C57BL/6 (Fig. 3 A), namely fibres of 𝛼-smooth muscle actin (𝛼-sma). BALB/c response to the cytokine was attenuated (Fig. 3 A). Although collagen I fibre deposition was not significantly different between the two strains in the TGFβ-1 assay (Fig. 3 B), C57BL/6 cells did express higher COL1A1 protein levels, consistent with the transcriptomic data ( Fig. 3Ci ). This validation is particularly critical given the C57BL/6-specific upregulation of Tgfb1 (the gene encoding TGFβ-1) observed in the fibrotic transcriptomes ( Fig. 3Cii ) which contributed to the strain-specific enrichment for positive upregulation of collagen biosynthesis (Fig. 2 B). The transcription factor CREB3L1 was the other gene contributing towards this enrichment ( Fig. 3Cii ). Key differentially expressed pulmonary fibrosis-related genes and modules are strain-specific Filtering for PFRGs (Table 1 ) and plotting their transformed expression in fibrotic tissue, it is apparent that many exhibit strain-specificity (Fig. 4 A). Calculating differential expression between fibrotic and healthy tissue separately for each genotype, strain-specific DEGs can further be identified ( Fig. 4Bi-ii ). This supports a model in which strain-specific fibrotic gene expression can be phenomenologically categorised into (“Category 1”) genes that are consistently expressed at different levels across strains in a condition-independent manner and (“Category 2”) genes that are differentially regulated under fibrosis (condition-dependent expression). Of all strain-specific DEGs identified (Category 2), there are 5 PFRGs that are C57BL/6-specific and 6 PFRGs that are BALB/c-specific (Table 1 ). Additionally, several interleukins are Category 2 genes (Table 1 ); Il6, in particular, is the central node of the interaction network for BALB/c-specific DEGs ( Fig. 4Bii and Fig. 4 C). This condition-dependent specificity may contribute to the distinct Th1 versus Th2 immunology of C57BL/6 vs BALB/c, as Il6 is implicated in Th1/Th2 fate determination ( 44 ). Collectively, Category 2 genes can comprise functional modules. Several KEGG pathway-enriched modules appear C57BL/6-specific, associated variously with neuroactive ligand-receptor interaction, chemokine signalling, cytokine-cytokine receptor interaction, ECM-receptor interaction, phototransduction, and TGFβ signalling ( Fig. 4Bi ). It should be noted that some of these terms correlate with late-stage IPF specifically in the literature, i.e. ECM-receptor interaction ( 10 ). This Category 2 enrichment for key fibrogenic functions – i.e. signalling via cytokines/chemokines/TGFβ, ECM remodelling – supports the hypothesis that strain-specific gene regulation can shape fibrosis susceptibility ( 3 , 45 – 47 ). There was no significant KEGG enrichment within the strain-specific BALB/c modules identified ( Fig. 4Bii ). Given the high C57BL/6 fibrosis risk among mice, this finding could indicate that pro-fibrotic processes in C57BL/6 underpin its exceptional susceptibility, rather than the converse for BALB/c ( 1 ). Within geneD-filtered DEGs ( Methods ), Nos2 is the only C57BL/6-specific PFRG ( Fig. 4Bi and Fig. 4 C) ( 23 ). In contrast, all BALB/c-specific PFRGs pass filtering (Table 1 ; Fig. 4Bii ). Of these top-scoring Category 2 PFRGs, only Ctsk is differentially regulated in a direction that might account for observed fibrosis susceptibilities, as an unambiguously anti-fibrotic gene most strongly upregulated in BALB/c (Fig. 4 C). Mmp13 is another Category 2 anti-fibrotic factor upregulated in BALB/c – and a direct interaction partner of Ctsk ( Fig. 4Bii-C ) ( 45 ). Absolute expression of Mmp13 , however, is lower in untreated BALB/c than in C57BL/6 (Fig. 4 C). Thus, whilst both factors fall into Category 2, we hypothesise that: whereas Ctsk may be actively more positively regulated under fibrosis in BALB/c, Mmp13 may instead be relieved from strain-specific basal repression when fibrosis develops in the BALB/c background. Within this framework, strain-specific upregulation of Ctsk could plausibly underlie BALB/c lung fibrosis resistance, but any putative function of Mmp13 is less clear, since it is expressed at similar levels in healthy tissue (Fig. 4 C). Interpretation is further complicated by the conflicting roles reported for Mmp13 , variously promoting and protecting against fibrogenesis ( 25 , 45 , 48 ). FOSL1 targets are overrepresented in cross-strain and cross-treatment differential expression Supplementing the gene-level analysis above, regulatory analysis was conducted in parallel on the expression data. Several DEG sets of interest – including those directly differentially expressed between strains – were tested for enrichment of genes sensitive to a comprehensive range of TF perturbations. The results clearly show that cross-strain DEGs, independent of treatment, were significantly enriched for genes sensitive to perturbation of the TF FOSL1 (control p.adj = 0.001, fibrotic p.adj = 0.01; Fig. 5 A). Moreover, under control conditions, FOSL1-sensitive genes were the only set significantly enriched – suggesting FOSL1 (ENSMUSG00000024912, GRCm38) may be of primary importance in distinguishing C57BL/6 versus BALB/c phenotypes at the transcriptional level. The set (GSE43965) was identified by Rajasekaran et al. (2013) ( 49 ). Applying the same analysis to cross-treatment DEGs, the results again indicate overrepresentation of FOSL1-sensitive genes, with this being the most significantly enriched term in both C57BL/6 (p.adj = 8.1e-18) and BALB/c (p.adj = 8.7e-55) fibrotic DEGs ( Fig. 5Bi ). This suggests FOSL1 might play a role in the transition between healthy and fibrotic lung tissue. However, when the analysis was performed on strain-specific fibrotic DEGs, only BALB/c-specific DEGs remained enriched for FOSL1-sensitive genes ( Fig. 5Bii ). This result would be consistent with an additional or altered role for FOSL1 in BALB/c lung fibrosis pathology. Independently, in vitro TGF𝛽 assays showed that Ctsk – target of the heterodimeric TF AP-1, through which FOSL1 acts ( 49 – 51 ) – was upregulated in BALB/c relative to C57BL/6 ( Fig. 5Ci ), albeit in a Category 1 manner as opposed to the Category 2 behaviour observed in the BLM dataset (Fig. 4 C). This consistency with the in vivo results further supports the idea that a mechanism involving Ctsk contributes to strain-specific IPF resistance in BALB/c. Moreover, established FOSL1 targets curated from the literature ( 52 ) – e.g. Thbs1, Prdm1 and Mmp2 – were differentially expressed between strains ( Fig. 5Cii ). Further analysis additionally indicated tighter regulation of these targets in BALB/c than in C57BL/6 ( data not shown ), which would be consistent with the noisier expression of BALB/c versus C57Bl/6 𝛼-sma previously observed in vitro (Fig. 3 A). As in the BLM dataset Fosl1 was not differentially expressed between strains either under control conditions ( Fig. 5Cii ; Supplementary Fig. 3 ). This provided orthogonal support for a FOSL1-dependent hypothesis of IPF risk, presumably mediated by the post-transcriptional alteration of FOSL1 activity. P170L variant in BALB/c Fosl1 gene may underpin differential regulation of Ctsk and Mmp13 To investigate possible genetic causes for strain-differential IPF prognosis, variant calling was conducted between BALB/c and C57BL/6 genomes. This revealed genetic differences that could underlie the differences in the pathophysiology of – reflected in the transcriptomic responses to – IPF in susceptible versus resistant backgrounds. We focus on both TFs and downstream genes identified from the transcriptomic analysis above – Fosl1 , Ctsk, Mmp13, Il6 , and Nos2 – and on genes implicated by the literature ( 15 , 20 , 21 , 53 , 54 ). The latter includes bleomycin hydrolase ( Blmh ), connective tissue growth factor ( Ctgf ), and Nos3 . A number of missense exonic variants are mechanistically suggestive, especially the single nucleotide polymorphism (SNP) affecting all Fosl1 transcripts recorded by ENSEMBL (Table 1 ). Table 2 Deleterious point substitutions in BALB/c Ctgf , Fosl1 , and Nos3 Gene Chromosome Position SNP C57BL/6 BALB/c DBA/2 Predicted effect SIFT Ctgf 10 24595833 rs8254419 T C T Missense, start codon lost 0.01 Fosl1 19 5450169 rs31137232 C T C Missense 0 Fosl1 19 5500197 rs30851424 C T C Missense 1 Nos3 5 24369876 rs32018659 C C T Missense 0.01 The observed Ctgf start codon variant (Table 2 ) could in theory impair expression, elucidating reports that Ctgf expression is poor in response to BLM alone. This is independent of any BLM-specific hypothesis for fibrosis resistance, as combined BLM and CTGF treatment successfully induces fibrosis ( 22 ). However, isoform analysis indicates that the rs8254419 variant would not affect the canonical CTGF transcript, but rather a truncated alternative spliceoform (ENSMUST00000129142.1). Additionally, a succession of several indels were identified in Blmh from the Sanger Mouse Genome Project, predicted to effect nonsense-mediated transcript decay. This is inconsistent with reports of high Blmh expression in BALB/c ( 20 ), which were themselves here corroborated by a cross-strain comparison within the dataset (logFC = 0.4, p.adj = 0.004). More promisingly, a coding variant (rs31137232) was detected in Fosl1 – the gene encoding FOSL1, which is itself a bZIP (basic-leucine zipper) protein (Table 2 ). This SNP corresponds to a non-conservative leucine zipper-proximal substitution, P170L, indicating a genetic change from a “helix-breaking” to a “helix-inducing” residue ( 55 ). It is thus plausible that the P170L mutation structurally alters the C-terminus of the FOSL1 central α-helix, supporting the hypothesis that this BALB/c-specific variant alters bZIP-mediated interactions. This would be consistent with an altered regulatory function of FOSL1 in BALB/c, as FOSL1 may regulate both Ctsk and Mmp13 through AP-1 dimerization ( 50 , 51 , 56 ). This could account for their Category 2 expression patterns in the BLM dataset (Fig. 4 C). An exploratory alignment was additionally conducted to show the structures of FOSL1 with and without the P170L mutation ( Supplementary Fig. 4 ). Additionally, the Fosl1 gene was amplified and sequenced in the lung cells from each strain used for TGFβ assays, confirming independently that the P170L mutation was present in BALB/c lines. A second coding variant (rs30851424) is also present in BALB/c albeit predicted to be well tolerated by SIFT scoring (Table 2 ). The impact of the L39F mutation is also more difficult to predict, due both to its position in a disordered region and the lack of an empirically determined structure for FOSL1. C-terminal Pro of the C57BL/6 FOSL1 bZIP domain is conserved at the equivalent position (170) in human FOSL1, according to the ENSEMBL database. Whilst a P170L SNP is not recorded in the human population, a P170T substitution is present at 0.001% frequency in the population. This is a similarly non-conservative substitution and therefore might be expected to have correlated or anticorrelated effects to the murine P170L mutation. Independently, it was determined that human FOSL1 has been linked to asthmatic lung disease by genome-wide association studies (GWAS) ( 57 ). Moreover, it is a regulator of MUC2 , a gene not only identified within the same study but also directly implicated in IPF by another GWAS ( 57 – 59 ). In addition to the murine FOSL1 variant, other mutations of interest offer evidence critical to previous work attempting to link mouse strain genotype and phenotype in lung fibrosis. The deleterious DBA/2 Nos3 variant here reported (Table 2 ) maps to all known functional transcripts from the gene. This could account for the documented failure of DBA/2 and Nos3 -deficient C57BL/6 – but not the wildtype – to resolve pulmonary fibrosis in repeated injury experiments ( 15 ). It also weakens the theory that Th2-biased immune systems are responsible for impaired fibrosis resolution, particularly since BALB/c (a) lacks this Nos3 mutation, (b) is Th2-biassed, and (c) possesses fibrosis resistance ( 1 ). Extending these mutational analyses, both Nos3 and Fosl1 were investigated using human familial pulmonary fibrosis (FPF) whole-genome association study data from the 100,000 Genomes Project (Genomics England), as were other genes of interest: Mmp13 , Ctsk , Il6 , and Nos2 . The significant variants identified from this study were predominantly found in Nos3 , consistent with the results in mouse models discussed above ( Supplementary Fig. 5 ). Several significant SNPs were also identified in other genes ( Supplementary Fig. 4, inset; Supplementary Tables 1–5 ). These included Fosl1, wherein 9 FPF-associated SNPs were detected (0.00 < p.val < 0.045; Supplementary Table 1 ). Independently, whole exome data from AstraZeneca was also analysed to identify SNPs correlated with human IPF in these genes. Whilst the adjusted p-values from this data analysis were non-significant, a significant SNP was identified in Fosl1 prior to correction. Discussion Effects of fibrosis are observable in diseased lung transcriptomes Transcriptomic profiles are a rich resource for characterising complex diseases. In this analysis, it is shown that the diagnostic markers of idiopathic pulmonary fibrosis (IPF) can be captured at the cellular level as enrichment for transcripts implicated in inflammation or tissue scarring ( 10 , 60 – 62 ), among genes differentially expressed (DEGs) in lung tissue treated with bleomycin (BLM) (Fig. 2 ; Supplementary Fig. 1 ). This enrichment might further facilitate prognostic inference, as both GO and KEGG analyses highlighted collagenic fibrogenesis/late-stage processes and pathways specific to C57BL/6, but not to BALB/c despite its enrichment for EMT-related GO terms ( 3 , 10 ). These results are consistent with the susceptibility of C57BL/6 strains to progressive fibrogenesis – in contrast to the less severe inflammatory syndromes observed in other strains such as BALB/c ( 1 ). Many fibrotic DEGs interact with each other. These interactions form functional subnetworks which themselves often correspond to pathological processes (Fig. 4 B). An integrative omics approach of this network-oriented type can give a modular readout for research into fibrosis of the lung, as well as other tissues. The application of this framework to strain-specific pathologies further benefits from the distinction between fibrotic DEGs which differ in expression between strains under all conditions (Category 1) and those differentially regulated with regard to fibrosis (Category 2). Intra- but not inter-tissue compensatory mechanisms are detectable In contrast to the above, the contralateral lung does not display any transcriptomic signature distinct from healthy tissue, within this dataset ( Supplementary Fig. 2 ). These findings undermine the possibility that transcriptional mechanisms facilitate compensatory activity as proposed in asymmetric murine models ( 5 , 9 ). Whilst this does not rule out alternative compensation hypotheses within the mouse model – such as those hinging on protein translation or modification – this conjecture has limited applicability to human disease, regardless, given that asymmetry in pulmonary fibrosis is a strong positive predictor of fibrogenic severity ( 6 , 7 ). Anti-fibrotic activity is, however, detected within the diseased lung itself. In particular, strain-specific upregulation of Ctsk and Mmp13 in fibrotic BALB/c tissue (Fig. 4 C) is consistent with the known IPF-resistance of this strain, as both encode collagenases critical to fibrosis resolution ( 1 , 45 , 63 ). It could also suggest why BALB/c was not specifically enriched for ECM deposition, despite strain-differential enrichment for EMT processes (Fig. 2 B). Furthermore, this focus on collagen breakdown fits within the emerging paradigm of epithelial lung disease as a balance between extracellular matrix (ECM) deposition and degradation. This posits a phenotypic spectrum graduated by structural protein turnover. For instance, emphysemic and tubercular symptoms have been linked to ECM degradation (including by Ctsk ), whereas deposition is a defining feature of IPF ( 3 , 63 , 64 ). Prediction of murine fibrotic pathologies is interdependent and hierarchical Whilst Ctsk and Mmp13 are here documented as factors influential in BALB/c resistance to pulmonary fibrosis, Nos2 is conversely a candidate marker for elevated susceptibility as exhibited by the reference strain, C57BL/6 (Fig. 4 B-C); it is known to promote inflammation in early fibrosis, despite a potential role in long-term resolution ( 23 , 53 ). It should be noted that, since these transcripts are selected based on: (a) a meta-analysis of the literature and (b) the significance and magnitude of their differential expression, they do not represent a comprehensive set of predictors for murine fibrosis risk. The expression of multiple C57BL/6-specific modules also highlights the putative uniqueness of C57BL/6 IPF-susceptibility at the gene level ( 1 ). In contrast, from analysis at the regulatory level, the TF FOSL1 emerges as a significant and overlapping influence on both cross-strain and within-strain DEGs (Fig. 5 A-B). The latter is consistent with its documented role in inhibiting pulmonary fibrosis ( 65 ). Highly significant enrichment of FOSL1-sensitive genes among cross-strain DEGs, on the other hand, represents a novel finding. It indicates that FOSL1 may be influential in differentiating C57BL/6 and BALB/c mouse models overall – even outside the context of IPF. Within that context, the additional finding that FOSL1-sensitive genes are significantly overrepresented among BALB/c-specific DEGs under fibrosis is highly suggestive of a role for FOSL1 in modulating strain-specific IPF resistance. Genetic analysis favours a novel FOSL1-modulated hypothesis for IPF risk The novel hypothesis that FOSL1 is a major modulator of murine variation in IPF physiology emerges as the key innovation of this study. This is not only a consequence of its evident significance in modulating the genes characteristic of BALB/c – as opposed to C57BL/6 – IPF pathophysiology (Fig. 5 ), as discussed above. Rather, it is this result in conjunction with the subsequent identification of the nonsynonymous P170L mutation present in BALB/c FOSL1, during analysis of the Sanger Mouse Project database (Table 2 ). The genetic results of this analysis also elucidated some findings of pre-genomic era research, though they were not always consistent with them. For instance, the start codon loss from BALB/c Ctgf (Table 2 ) could play a role in explaining the strain’s impaired Ctgf and collagen expression following BLM treatment ( 22 , 66 ). Comparably, the finding that BALB/c Blmh contains many nonsense-mediated decay variants is apparently paradoxical with the finding – here replicated – that BALB/c expresses elevated levels of Blmh transcript: a proposed basis for IPF-resistance in a specific BLM-dependent manner ( 20 , 21 ). The relation between previous findings and the novel Ctgf and Blmh variants remains to be seen. Nevertheless, the result that BALB/c ECM deposition is similarly impaired in the mouse lung fibroblast (MLF) TGFβ assay as in BLM studies – including within the present dataset (Fig. 3 A) – indicates that BLM-sensitivity is unlikely to be the root cause of variation in fibrosis risk, at least as far as the pulmonary pathology of BALB/c is concerned. This BLM-independent model of risk is orthogonally supported by the strain-differential enrichment of BALB/c for EMT processes (Fig. 2 B), which are characteristically a precursor to ECM accumulation and fibrogenesis ( 2 ). The finding that this enrichment is not here correlated with any BALB/c-specific enrichment in fibrogenesis further implies a resistance mechanism downstream of EMT – such as enhanced ECM degradation, e.g. by the upregulation of collagenases like Ctsk and Mmp13 (Fig. 4 C and Fig. 5 C). Future inquiries into the links between genomic variants and IPF phenotypes would indeed benefit from accounting for non-C57 mouse strains. For instance, the DBA/2 Nos3 mutation identified (Table 2 ) could readily account for previously observed differences in fibrosis resolution between DBA/2 and C57BL/6, questioning prior assertions that this variation was linked to Th1- versus Th2-bias ( 15 ). This theory could have been assessed at the time by including BALB/c (or other Th2-biassed mice). Indeed, whilst the present study represents an initial step in extending transcriptomic fibrosis research to multiple genetic backgrounds, it lacks data from the many mouse strains other than BALB/c that are resistant to IPF, limiting inference. That said, the same P170L Fosl1 variant identified in BALB/c was found, within this study, across other IPF-resistant strains within the Sanger Mouse Project (C3H, A/J). This supports the specific hypothesis that the P170L SNP itself may be responsible for altered FOSL1 activity in resistant strains, which is itself implicated in BALB/c IPF resistance by the transcriptomics analysis discussed. Thus, unlike the BLM-specific and Th2-bias hypotheses regarding BALB/c IPF pathology, the transcriptomic and genetic evidence are consistent with each other regarding FOSL1 – and strengthen the hypothesis that it modulates IPF risk in mouse models. Further research on more diverse strains in the context of fibrosis resistance would be valuable in generalising these findings – as well as in assessing other research previously published using a single strain ( 1 ). It would also be informative to analyse the comparably non-conservative P170T variant observed in human FOSL1, given the multiple threads of evidence linking FOSL1 to IPF in humans as well, including those identified within the present investigation ( Supplementary Fig. 5 ) ( 57 – 59 , 65 ). That said, GWAS studies with a specific focus on fibrotic lung disease have not identified Fosl1 directly as a gene of interest – one factor in this may be the exclusion of FOSL1 variants from single nucleotide polymorphism (SNP) arrays ( 67 ). The mining of unbiased GWAS datasets – e.g. based on genome or exome sequencing – may thus be necessary for the further analysis of human FOSL1 variant activity. In the present study, such analysis has already identified a number of significant FOSL1 SNPs ( Supplementary Table 1 ). A meta-analysis – though beyond the scope of this investigation – could increase the diversity of SNPs identified as well as the statistical power of testing for association. It should also be noted that the largely negative result regarding human IPF-associated FOSL1 variants does not imply that the murine P170L variant is not mechanistically relevant to human IPF. in vitro studies or, ultimately, gene therapy trials would yield empirical data on human applicability. Regulatory factors are promising targets for preclinical testing It has been established in this paper that Fosl1 is a candidate of particular interest for risk modulation in mouse models. Moreover, FOSL1 is already positively implicated in both the resolution of pulmonary fibrosis and induction of hepatic fibrosis ( 65 , 68 ). This coincides precisely with the low and high risks of pulmonary and hepatic fibrosis, respectively, noted for BALB/c relative to C57BL/6 mice ( 1 ). This study thus provides a promising starting point for in vitro or in vivo modulation of murine fibrosis risk through engineering FOSL1, for instance to possess a P170L or comparable colocalised SNP. FOSL1 is also implicated in human fibrogenic mechanisms. Specifically, it has been shown to promote EMT, despite correlating with an overall anti-fibrotic effect in some tissues ( 69 ). Under a hypothesis of FOSL1 gain-of-function in BALB/c, this would be consistent with the enrichment of EMT-related but not fibrogenic processes observed in the dataset (Fig. 2 B). FOSL1 is also the foremost co-regulator of EP300, the central regulator identified for Dupuytren’s disease, a human fibrotic disorder ( 69 ). Moreover, it is plausible that the SNPs here reported for murine Fosl1 (rs30851424 and rs31137232) could alter the equivalent Ep300-Fosl1 interaction, altering key fibrotic processes in the ways observed (Fig. 2 B-C; Table 2 ). This hypothesis of fibrosis resistance through altered Fosl1-Ep300 co-regulation would benefit from structural analysis. More broadly, the Category 2 signatures of not only fibrosis-related DEGs but also entire functional network modules (Fig. 4 B) favours a preclinical focus on pathway regulators, such as TFs, rather than on single downstream gene products like collagenases. Within the network paradigm, these regulators are responsible for ‘switching’ modules relevant to disease, resulting – where regulatory function varies with genetic background – in strain-specific pathology. By this definition, regulators include non-transcriptional and intercellular factors; for example, nitric oxide (NO) and interleukins, both of which mediate immune signalling ( 2 ). Nos2 and Il6 are also highlighted here as regulators exhibiting Category 2 expression under fibrosis. Knockouts of the pro-fibrotic gene Nos2 , in particular, might informatively alter the unusual fibrosis susceptibility of C57BL/6 mice ( Fig. 4Bi ). Whilst Nos2 is considered anti-fibrotic in the long-term, it is non-essential for fibrosis resolution, unlike Nos3 ( 15 , 53 ). The results of this study also indicate the value in further analysis of Il6 , as its role in Th1/Th2 differentiation, known regulation by FOSL1, and centrality to fibrotic DEG interactions in BALB/c ( Fig. 4Bii ) draws on ongoing research into immunotypes and IPF ( 15 , 44 ). Conclusions From these data, a multifactorial view of pulmonary fibrosis in the mouse model emerges, focused on the diseased lung itself – with disease defined by the activation states of various transcriptional subnetworks. A subset of these fibrosis-relevant subnetworks also differ across model mouse strains, correlating with observed differences in disease risk – and highly enriched for genes sensitive to the transcription factor (TF) FOSL1. Combined with the presence of a non-conservative P170L SNP in this TF across IPF-resistant strains, this supports a novel hypothesis that FOSL1 modulates IPF risk in mice. This FOSL1 hypothesis complements the established view of IPF: a genetically complex disease, shaped by the regulation or dysregulation of vital processes ( 2 , 3 , 10 , 23 ). Examination of FOSL1 binding partners (e.g. EP300) and determination of an empirical structure may thus be warranted, to identify the mechanism for FOSL1 modulation of IPF. Overall, our data and analyses highlight the pivotal role of regulators; the TF FOSL1 in particular, but also other signalling proteins such as NOS2 and IL6. The power of this integrated investigation, using both whole lung transcriptomes and mouse strain genetics, lies precisely in the shortlisting of such preclinical targets: particularly, in the identification of existing therapeutic variants, the viability of which has already been tested – by evolution. Abbreviations IPF: idiopathic pulmonary fibrosis BLM: bleomycin DEG: differentially expressed gene GWAS: genome-wide association study CTGF: connective tissue growth factor TGFβ: transforming growth factor beta EMT: epithelial to mesenchymal transition BLMH: bleomycin hydrolase PCA: principal component analysis ECM: extracellular matrix TF: transcription factor PFRG: pulmonary fibrosis-related gene NO: nitric oxide Th2: T helper 2 cells Th1: T helper 1 cells SNP: single nucleotide polymorphism bZIP: basic-leucine zipper Declarations Ethics approval This study was conducted using data from the Blue Sky Collaboration between AstraZeneca and the MRC Laboratory of Molecular Biology. The use of animal tissue and data was approved by the AstraZeneca ethical committee in Gothenburg, Sweden (EA184-2024). The approved site number for in vivo work was 31-5373/11. Consent to participate is not applicable. Consent for publication Not applicable. Availability of data and materials The data that support the findings of this study are available from AstraZeneca but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of AstraZeneca. Competing interests The authors declare no competing interests. Funding This study was supported jointly by the MRC Laboratory of Molecular Biology and AstraZeneca, as part of the Blue Sky Collaboration programme. Authors’ contributions TF conducted all other in silico analysis; they also wrote and revised this manuscript. IB conducted the in vitro work and contributed to the writing and figures of this manuscript. RG conducted processing and RNA sequencing of the lung samples, in collaboration with AstraZeneca (AZ). WZ conducted GWAS data analysis. MS scripted preliminary analysis and visualisation of transcriptomes. HT assisted with protein structure analysis and interpretation. LAM managed the collaboration and provided expert scientific oversight. AZ scientists conducted all in vivo experiments. JG conceived and supervised the project at the LMB. Acknowledgements All figure panels were rendered in BioRender by Thea Fennell, under the use of a Premium license and therefore legal for publication. This includes the use of BioRender graphic assets. References Walkin L, Herrick SE, Summers A, Brenchley PE, Hoff CM, Korstanje R, et al. The role of mouse strain differences in the susceptibility to fibrosis: a systematic review. Fibrogenesis Tissue Repair. 2013;6(1):18. van Geffen C, Deißler A, Quante M, Renz H, Hartl D, Kolahian S. Regulatory Immune Cells in Idiopathic Pulmonary Fibrosis: Friends or Foes? Front Immunol. 2021;12:1161. Deng Z, Fear MW, Suk Choi Y, Wood FM, Allahham A, Mutsaers SE, et al. The extracellular matrix and mechanotransduction in pulmonary fibrosis. Int J Biochem Cell Biol. 2020;126:105802. Moore MW, Herzog EL. Regulatory T Cells in Idiopathic Pulmonary Fibrosis: Too Much of a Good Thing? Am J Pathol. 2016;186(8):1978. Poelmans VVG, De Langhe J, Hillen E, Vanoirbeek A, Himmelreich J. Longitudinal micro-CT provides biomarkers of lung disease that can be used to assess the effect of therapy in preclinical mouse models, and reveal compensatory changes in lung volume. Dis Model Mech. 2016;9(1):91–8. Callahan SJ, Xia M, Murray S, Flaherty KR. Clinical characteristics in patients with asymmetric idiopathic pulmonary fibrosis. Respir Med. 2016;119:96–101. Sokai A, Tanizawa K, Handa T, Kubo T, Hashimoto S, Ikezoe K et al. Asymmetry in acute exacerbation of idiopathic pulmonary fibrosis. ERJ Open Res. 2017;3(2). Baarsma HA, Königshoff M. WNT-er is coming’: WNT signalling in chronic lung diseases. Thorax. 2017;72(8):746–59. Dubaybo BA, Maxwell DL, Sikora CA, Carlson RW. Compensatory Growth in Fibrotic Lung Injury. Am J Med Sci. 1991;302(1):19–22. Wang Y, Hu S, Shen L, Liu S, Wan L, Yang S, et al. Dynamic Observation of Autophagy and Transcriptome Profiles in a Mouse Model of Bleomycin-Induced Pulmonary Fibrosis. Front Mol Biosci. 2021;8:685. Åhlgren J, Voikar V. Experiments done in Black-6 mice: what does it mean? Lab Anim 2019 486. 2019;48(6):171–80. Kolb M, Bonniaud P, Galt T, Sime PJ, Kelly MM, Margetts PJ, et al. Differences in the fibrogenic response after transfer of active transforming growth factor-β 1 gene to lungs of fibrosis-prone and fibrosis-resistant mouse strains. Am J Respir Cell Mol Biol. 2002;27(2):141–50. Down JD. The nature and relevance of late lung pathology following localised irradiation of the thorax in mice and rats. Br J Cancer. 1986;53:330–2. Song HK, Hwang DY. Use of C57BL/6N mice on the variety of immunological researches. Lab Anim Res. 2017;33(2):119–23. Chung MP, Monick MM, Hamzeh NY, Butler NS, Powers LS, Hunninghake GW. Role of repeated lung injury and genetic background in bleomycin-induced fibrosis. Am J Respir Cell Mol Biol. 2003;29(3):375–80. Izbicki G, Segel MJ, Christensen TG, Conner MW, Breuer R. Time course of bleomycin-induced lung fibrosis. Int J Exp Pathol. 2002;83(3):111–9. Kolb M, Margetts PJ, Sime PJ, Gauldie J. Proteoglycans decorin and biglycan differentially modulate TGF-β-mediated fibrotic responses in the lung. Am J Physiol-Lung Cell Mol Physiol. 2001;280(6):L1327–34. Sime PJ, Xing Z, Graham FL, Csaky KG, Gauldie J. Adenovector-mediated gene transfer of active transforming growth factor-beta1 induces prolonged severe fibrosis in rat lung. J Clin Invest. 1997;100(4):768–76. Weigle S, Martin E, Voegtle A, Wahl B, Schuler M. Primary cell-based phenotypic assays to pharmacologically and genetically study fibrotic diseases in vitro. J Biol Methods. 2019;6(2):e115. Filderman AE, Lazo JS. Murine strain differences in pulmonary bleomycin metabolism. Biochem Pharmacol. 1991;42(1):195–8. Lasky J, Ortiz LA, Tonthat B, Hoyle GW, Corti M, Athas G et al. Connective tissue growth factor mRNA expression is upregulated in bleomycin-induced lung fibrosis. Am J Physiol - Lung Cell Mol Physiol. 1998;275(2 19 – 2). Bonniaud P, Martin G, Margetts PJ, Ask K, Robertson J, Gauldie J et al. Connective Tissue Growth Factor Is Crucial to Inducing a Profibrotic Environment in Fibrosis-Resistant Balb/c Mouse Lungs. Httpsdoiorg101165rcmb2004-0158OC. 2012;31(5):510–6. Toren D, Yanai H, Abu Taha R, Bunu G, Ursu E, Ziesche R, et al. Systems biology analysis of lung fibrosis-related genes in the bleomycin mouse model. Sci Rep 2021 111. 2021;11(1):1–13. Blaauboer ME, Boeijen FR, Emson CL, Turner SM, Zandieh-Doulabi B, Hanemaaijer R, et al. Extracellular matrix proteins: A positive feedback loop in lung fibrosis? Matrix Biol. 2014;34:170–8. Todd JL, Vinisko R, Liu Y, Neely ML, Overton R, Flaherty KR, et al. Circulating matrix metalloproteinases and tissue metalloproteinase inhibitors in patients with idiopathic pulmonary fibrosis in the multicenter IPF-PRO Registry cohort. BMC Pulm Med. 2020;20(1):1–12. Onursal C, Dick E, Angelidis I, Schiller HB, Staab-Weijnitz CA. Collagen Biosynthesis, Processing, and Maturation in Lung Ageing. Front Med. 2021;8:495. Beck JA, Lloyd S, Hafezparast M, Lennon-Pierce M, Eppig JT, Festing MFW, et al. Genealogies of mouse inbred strains. Nat Genet 2000 241. 2000;24(1):23–5. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Gene Ontology: tool for the unification of biology. Nat Genet. 2000;25(1):25–9. The Gene Ontology Consortium, Aleksander SA, Balhoff J, Carbon S, Cherry JM, Drabkin HJ, et al. The Gene Ontology knowledgebase in 2023. Genetics. 2023;224(1):iyad031. Kanehisa M, Furumichi M, Sato Y, Kawashima M, Ishiguro-Watanabe M. KEGG for taxonomy-based analysis of pathways and genomes. Nucleic Acids Res. 2023;51(D1):D587–92. Xie Z, Bailey A, Kuleshov MV, Clarke DJB, Evangelista JE, Jenkins SL, et al. Gene Set Knowledge Discovery with Enrichr. Curr Protoc. 2021;1(3):e90. Chen EY, Tan CM, Kou Y, Duan Q, Wang Z, Meirelles GV, et al. Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics. 2013;14:128. Kuleshov MV, Jones MR, Rouillard AD, Fernandez NF, Duan Q, Wang Z, et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 2016;44(W1):W90–97. Szklarczyk D, Franceschini A, Wyder S, Forslund K, Heller D, Huerta-Cepas J, et al. STRING v10: protein-protein interaction networks, integrated over the tree of life. Nucleic Acids Res. 2015;43(Database issue):D447–452. Szklarczyk D, Gable AL, Nastou KC, Lyon D, Kirsch R, Pyysalo S, et al. The STRING database in 2021: Customizable protein-protein networks, and functional characterization of user-uploaded gene/measurement sets. Nucleic Acids Res. 2021;49(D1):D605–12. Rackham OJL, Firas J, Fang H, Oates ME, Holmes ML, Knaupp AS, et al. A predictive computational framework for direct reprogramming between human cell types. Nat Genet. 2016;48(3):331–5. Keane TM, Goodstadt L, Danecek P, White MA, Wong K, Yalcin B, et al. Mouse genomic variation and its effect on phenotypes and gene regulation. Nature. 2011;477(7364):289–94. McLaren W, Gil L, Hunt SE, Riat HS, Ritchie GRS, Thormann A et al. The Ensembl Variant Effect Predictor. Genome Biol [Internet]. 2016 Jun 6 [cited 2022 Jun 30];17(1). /pmc/articles/PMC4893825/ Baldarelli RM, Smith CL, Ringwald M, Richardson JE, Bult CJ, Mouse Genome Informatics Group. Mouse Genome Informatics: an integrated knowledgebase system for the laboratory mouse. Genetics. 2024;227(1):iyae031. Shovlin CL, Morris-Rosendahl DJ, Copeland F, Soyza AD, Hogg C, Jenkins G, et al. S66 Delivering the 100,000 genomes project to establish the functional role of DNA sequence variants in respiratory rare diseases. Thorax. 2019;74(Suppl 2):A44–5. Dhindsa RS, Mattsson J, Nag A, Wang Q, Wain LV, Allen R, et al. Identification of a missense variant in SPDL1 associated with idiopathic pulmonary fibrosis. Commun Biol. 2021;4(1):1–8. Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, et al. Highly accurate protein structure prediction with AlphaFold. Nature. 2021;596(7873):583–9. Mirdita M, Schütze K, Moriwaki Y, Heo L, Ovchinnikov S, Steinegger M. ColabFold: making protein folding accessible to all. Nat Methods. 2022;19(6):679–82. Diehl S, Rincón M. The two faces of IL-6 on Th1/Th2 differentiation. Mol Immunol. 2002;39(9):531–6. Cabrera S, Maciel M, Hernández-Barrientos D, Calyeca J, Gaxiola M, Selman M, et al. Delayed resolution of bleomycin-induced pulmonary fibrosis in absence of MMP13 (collagenase 3). Am J Physiol - Lung Cell Mol Physiol. 2019;316(5):L961–76. Karman J, Wang J, Bodea C, Cao S, Levesque MC. Lung gene expression and single cell analyses reveal two subsets of idiopathic pulmonary fibrosis (IPF) patients associated with different pathogenic mechanisms. PLoS ONE. 2021;16(3 March):e0248889. Hernandez DM, Kang JH, Choudhury M, Andrianifahanana M, Yin X, Limper AH, et al. IPF pathogenesis is dependent upon TGFβ induction of IGF-1. FASEB J. 2020;34(4):5363–88. Nkyimbeng T, Ruppert C, Shiomi T, Dahal B, Lang G, Seeger W, et al. Pivotal Role of Matrix Metalloproteinase 13 in Extracellular Matrix Turnover in Idiopathic Pulmonary Fibrosis. PLoS ONE. 2013;8(9):e73279. Rajasekaran S, Reddy NM, Zhang W, Reddy SP. Expression profiling of genes regulated by Fra-1/AP-1 transcription factor during bleomycin-induced pulmonary fibrosis. BMC Genomics. 2013;14(1):1–13. Pang M, Rodríguez-Gonzalez M, Hernandez M, Recinos CC, Seldeen KL, Troen BR. AP-1 and Mitf interact with NFATc1 to stimulate cathepsin K promoter activity in osteoclast precursors. J Cell Biochem. 2019;120(8):12382–92. Fleischmann A, Hafezi F, Elliott C, Remé CE, Rüther U, Wagner EF. Fra-1 replaces c-Fos-dependent functions in mice. Genes Dev. 2000;14(21):2695. Han H, Cho JW, Lee S, Yun A, Kim H, Bae D, et al. TRRUST v2: an expanded reference database of human and mouse transcriptional regulatory interactions. Nucleic Acids Res. 2018;46(D1):D380–6. Guo C, Atochina-Vasserman E, Abramova H, George B, Manoj V, Scott P, et al. Role of NOS2 in pulmonary injury and repair in response to bleomycin. Free Radic Biol Med. 2016;91:293. Ishida Y, Kimura A, Nosaka M, Kuninaka Y, Hemmi H, Sasaki I et al. Essential involvement of the CX3CL1-CX3CR1 axis in bleomycin-induced pulmonary fibrosis via regulation of fibrocyte and M2 macrophage migration. Sci Rep [Internet]. 2017 Dec 1 [cited 2022 Feb 1];7(1). https://pubmed.ncbi.nlm.nih.gov/29203799/ Chen SSL, Lee SF, Hao HJ, Chuang CK. Mutations in the Leucine Zipper-Like Heptad Repeat Sequence of Human Immunodeficiency Virus Type 1 gp41 Dominantly Interfere with Wild-Type Virus Infectivity. J Virol. 1998;72(6):4765. Meyer MB, Benkusky NA, Pike JW. Selective Distal Enhancer Control of the Mmp13 Gene Identified through Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR) Genomic Deletions. J Biol Chem. 2015;290(17):11093. Olafsdottir TA, Theodors F, Bjarnadottir K, Bjornsdottir US, Agustsdottir AB, Stefansson OA, et al. Eighty-eight variants highlight the role of T cell regulation and airway remodeling in asthma pathogenesis. Nat Commun 2020 111. 2020;11(1):1–11. Fingerlin TE, Murphy E, Zhang W, Peljto AL, Brown KK, Steele MP, et al. Genome-wide association study identifies multiple susceptibility loci for pulmonary fibrosis. Nat Genet 2013 456. 2013;45(6):613–20. Song S, Byrd JC, Mazurek N, Liu K, Koo JS, Bresalier RS. Galectin-3 Modulates MUC2 Mucin Expression in Human Colon Cancer Cells at the Level of Transcription via AP-1 Activation. Gastroenterology. 2005;129(5):1581–91. Perkins TN, Oury TD. The perplexing role of RAGE in pulmonary fibrosis: causality or casualty? https:// doi.org/101177/17534666211016071 . 2021;15. Nakashima T, Liu T, Yu H, Ding L, Ullenbruch M, Hu B, et al. Lung bone marrow-derived hematopoietic progenitor cells enhance pulmonary fibrosis. Am J Respir Crit Care Med. 2013;188(8):976–84. Bendstrup E, Møller J, Kronborg-White S, Prior TS, Hyldgaard C. Interstitial Lung Disease in Rheumatoid Arthritis Remains a Challenge for Clinicians. J Clin Med. 2019;8(12). Bühling F, Röcken C, Brasch F, Hartig R, Yasuda Y, Saftig P, et al. Pivotal Role of Cathepsin K in Lung Fibrosis. Am J Pathol. 2004;164(6):2203–16. Kubler A, Larsson C, Luna B, Andrade BB, Amaral EP, Urbanowski M, et al. Cathepsin K Contributes to Cavitation and Collagen Turnover in Pulmonary Tuberculosis. J Infect Dis. 2016;213(4):618. Rajasekaran S, Vaz M, Reddy SP. Fra-1/AP-1 Transcription Factor Negatively Regulates Pulmonary Fibrosis In Vivo. PLoS ONE. 2012;7(7):41611. Sternlicht MD, Wirkner U, Bickelhaupt S, Lopez Perez R, Tietz A, Lipson KE, et al. Radiation-induced pulmonary gene expression changes are attenuated by the CTGF antibody Pamrevlumab. Respir Res. 2018;19(1):1–16. Allen RJ, Oldham JM, Jenkins DA, Leavy OC, Guillen-Guio B, Melbourne CA, et al. Longitudinal lung function and gas transfer in individuals with idiopathic pulmonary fibrosis: a genome-wide association study. Lancet Respir Med. 2022;11(1):65–73. Kireva T, Erhardt A, Tiegs G, Tilg H, Denk H, Haybaeck J et al. Transcription Factor Fra-1 Induces Cholangitis and Liver Fibrosis. 2011. Williams LM, McCann FE, Cabrita MA, Layton T, Cribbs A, Knezevic B, et al. Identifying collagen VI as a target of fibrotic diseases regulated by CREBBP/EP300. Proc Natl Acad Sci U S A. 2020;117(34):20753–63. Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterial.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {\"props\":{\"pageProps\":{\"initialData\":{\"identity\":\"rs-5295459\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":true,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":374981792,\"identity\":\"99bd4273-12b0-43ff-8ef7-8d9563fecc33\",\"order_by\":0,\"name\":\"Thea Fennell\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABB0lEQVRIiWNgGAWjYDAC5gMg0kKGH8wrYEhASPFg18HDBlYjwSPZAKINSNFicIBYLfZszMceF1RI8BhfO/vwA4PBtjwG9sNPNzDU2DEYnDmAwxa2dOMZZyR4zG6nG0swGNwuZuBJM7vBcCyZweBsA3Yt8j1m0rxtIC1pbMx/DG4nNkgwALWwHWAwOI/LL/zfpHn/AR02O40N6BeQFvZvNxj+4dPCwybN2wD0vjRcC4/ZDca2A7gddozNTJrnmASPxO00ZrBf2Hhyym4k9iXzSOLwPnsb8zNpnhobOf7ZaYwfGCpu5/GzH99248M3Ozm+MwnYXYYB2EBEAs5YGQWjYBSMglFADAAAEe1NmO+xrfUAAAAASUVORK5CYII=\",\"orcid\":\"\",\"institution\":\"MRC Laboratory of Molecular Biology\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Thea\",\"middleName\":\"\",\"lastName\":\"Fennell\",\"suffix\":\"\"},{\"id\":374981793,\"identity\":\"5215f150-29c9-4508-b5da-bbe11ae8b67f\",\"order_by\":1,\"name\":\"Ieva Beržanskytė\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"MRC Laboratory of Molecular Biology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Ieva\",\"middleName\":\"\",\"lastName\":\"Beržanskytė\",\"suffix\":\"\"},{\"id\":374981794,\"identity\":\"b5a86e7a-866b-491e-8877-fb1796c73a38\",\"order_by\":2,\"name\":\"Rihab Gam\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"MRC Laboratory of Molecular Biology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Rihab\",\"middleName\":\"\",\"lastName\":\"Gam\",\"suffix\":\"\"},{\"id\":374981795,\"identity\":\"2bf0bb3c-ac80-4093-9a04-225788755938\",\"order_by\":3,\"name\":\"Wencan Zhu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"MRC Laboratory of Molecular Biology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Wencan\",\"middleName\":\"\",\"lastName\":\"Zhu\",\"suffix\":\"\"},{\"id\":374981796,\"identity\":\"09d25e8a-5620-4f50-82c5-e570ce380658\",\"order_by\":4,\"name\":\"Minkyung Sung\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"MRC Laboratory of Molecular Biology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Minkyung\",\"middleName\":\"\",\"lastName\":\"Sung\",\"suffix\":\"\"},{\"id\":374981797,\"identity\":\"2e435bb1-62b9-4c4c-bb70-551e4076a045\",\"order_by\":5,\"name\":\"Himani Tandon\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"MRC Laboratory of Molecular Biology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Himani\",\"middleName\":\"\",\"lastName\":\"Tandon\",\"suffix\":\"\"},{\"id\":374981798,\"identity\":\"f86647cb-b3aa-481a-9d05-4267fd9f97a2\",\"order_by\":6,\"name\":\"Lynne A. Murray\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"AstraZeneca (United Kingdom)\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Lynne\",\"middleName\":\"A.\",\"lastName\":\"Murray\",\"suffix\":\"\"},{\"id\":374981799,\"identity\":\"0bcf5c1e-3c52-4a9d-aa43-6593f783583e\",\"order_by\":7,\"name\":\"Julian Gough\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"MRC Laboratory of Molecular Biology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Julian\",\"middleName\":\"\",\"lastName\":\"Gough\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2024-10-19 16:23:08\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-5295459/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-5295459/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":68743655,\"identity\":\"a9b7ac43-3642-440e-8ca7-793bd1665c75\",\"added_by\":\"auto\",\"created_at\":\"2024-11-11 14:48:34\",\"extension\":\"jpeg\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":597434,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eBleomycin treatment of mice simulates idiopathic pulmonary fibrosis:\\u003c/strong\\u003e\\u003cem\\u003e \\u003c/em\\u003eBleomycin administration causes damage to DNA (1), resulting in an inflammatory immune response and cellular injury (2). Injury repair processes may form positive feedback loops via mechanotransduction, resulting in excessive deposition of ECM proteins – fibrogenesis (3). This impairs diffusion across alveolar walls, which can lead to death by suffocation if unresolved (4). Resolution is less likely following prolonged treatment with bleomycin (5).\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure1.jpeg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5295459/v1/17b03c4ebc0ffb37ea0a865b.jpeg\"},{\"id\":68743656,\"identity\":\"5599fb09-8764-497a-89b2-eb39fdf540b2\",\"added_by\":\"auto\",\"created_at\":\"2024-11-11 14:48:34\",\"extension\":\"jpeg\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":829579,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eTreated lung tissue transcriptomes are characteristically enriched for fibrotic processes: (A)\\u003c/strong\\u003e\\u003cem\\u003e \\u003c/em\\u003esamples cluster by treatment group and strain under principal component analysis (PCA) of saline- (n=23) and bleomycin-treated (n=11) tissue from each of BALB/c (n=10) and C57BL/6 (n=24), along the first two components. \\u003cstrong\\u003e(B)\\u003c/strong\\u003e GO biological processes associated with fibrosis are enriched in both strains’ upregulated genes, with collagen synthesis being enriched specifically in C57BL/6J (n=12). Dots are coloured by log-transformed significance (p.adj) and their size is scaled by gene ratio.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure2.jpeg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5295459/v1/bcb7f5ebb8fe70bc31acfcfd.jpeg\"},{\"id\":68743654,\"identity\":\"9b0e8a3e-2410-4c7a-92ec-6fcaf4eabaaf\",\"added_by\":\"auto\",\"created_at\":\"2024-11-11 14:48:34\",\"extension\":\"jpeg\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":2156953,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eStress fibre deposition is altered between the two strains: (A)(\\u003c/strong\\u003e\\u003cem\\u003e\\u003cstrong\\u003ei\\u003c/strong\\u003e\\u003c/em\\u003e\\u003cstrong\\u003e) \\u003c/strong\\u003emicrographs used to calculate values in (A) clearly show higher levels of 𝛼-SMA in C57BL/6 than BALB/c at 10ng/ml of TGFβ; scale bar indicates 50 µm. \\u003cstrong\\u003e(\\u003c/strong\\u003e\\u003cem\\u003e\\u003cstrong\\u003eii\\u003c/strong\\u003e\\u003c/em\\u003e\\u003cstrong\\u003e) \\u003c/strong\\u003estress fibre deposition measured by 𝛼-SMA immunostaining. Activated fibres were quantified as area over threshold value normalised to the number of nuclei per frame, at an increasing TGFβ-1 concentration. Fibre area is shown as median (n=3, separate mice lung preparations) and significance calculated using two-way ANOVA with Tukey’s post-hoc test (C57BL/6, 0 vs 10, p\\u003csup\\u003e*\\u003c/sup\\u003e=0.01; for Balb/c, 0 vs 10, p\\u003csup\\u003en.s.\\u003c/sup\\u003e=0.999). \\u003cstrong\\u003e(B)(\\u003c/strong\\u003e\\u003cem\\u003e\\u003cstrong\\u003ei\\u003c/strong\\u003e\\u003c/em\\u003e\\u003cstrong\\u003e) \\u003c/strong\\u003eCOL1A1 fibril formation in cultured adult lung fibroblasts of the two strains exposed to TGFβ-1 \\u003cem\\u003ein vitro.\\u003c/em\\u003e n=1, 3 technical replicates; \\u003cstrong\\u003e(\\u003c/strong\\u003e\\u003cem\\u003e\\u003cstrong\\u003eii\\u003c/strong\\u003e\\u003c/em\\u003e\\u003cstrong\\u003e) \\u003c/strong\\u003emicrographs of COL1A1 immunostaining. Scale bar indicates 50 µm. \\u003cstrong\\u003e(C)(\\u003c/strong\\u003e\\u003cem\\u003e\\u003cstrong\\u003ei\\u003c/strong\\u003e\\u003c/em\\u003e\\u003cstrong\\u003e) \\u003c/strong\\u003eCol1a1 gene expression in the \\u003cem\\u003ein vivo \\u003c/em\\u003eBLM\\u003cem\\u003e \\u003c/em\\u003edataset (BALB/c: n=5 BLM-treated, n=5 saline-treated/CTRL; C57BL/6J: n=4 BLM-treated, n=8 saline-treated/CTRL); \\u003cstrong\\u003e(\\u003c/strong\\u003e\\u003cem\\u003e\\u003cstrong\\u003eii\\u003c/strong\\u003e\\u003c/em\\u003e\\u003cstrong\\u003e) \\u003c/strong\\u003eexpression in the \\u003cem\\u003ein vivo \\u003c/em\\u003eBLM dataset of the C57BL/6J-specific DEGs contributing toward strain-specific enrichment for positive upregulation of collagen biosynthesis: Creb3l1 and Tgfb1 (\\u003cem\\u003eFig. 2B\\u003c/em\\u003e).\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure3.jpeg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5295459/v1/b2d013dfa3ef276a9aeb6394.jpeg\"},{\"id\":68743658,\"identity\":\"b3900d02-4f3c-4a3f-b8e8-d439abbc074b\",\"added_by\":\"auto\",\"created_at\":\"2024-11-11 14:48:34\",\"extension\":\"jpeg\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":1900488,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eTranscriptomic response to IPF varies across strains:\\u003c/strong\\u003e \\u003cstrong\\u003e(A) \\u003c/strong\\u003ethe heatmap of rlog-transformed fibrosis-related gene expression for fibrotic tissue in different strains: BALB/c, C57BL/6J, C57BL/6N. Gene expression is scaled from a minimum of -3 to a maximum of 3. The clustering of red and blue genes has some similarity within a strain but broad differences between strains. \\u003cstrong\\u003e(B) \\u003c/strong\\u003eProtein-protein interaction networks could be generated from the gene sets strain-specifically differentially expressed in fibrotic tissues for both \\u003cstrong\\u003e(\\u003c/strong\\u003e\\u003cem\\u003e\\u003cstrong\\u003ei\\u003c/strong\\u003e\\u003c/em\\u003e\\u003cstrong\\u003e) \\u003c/strong\\u003eC57BL/6 and \\u003cem\\u003e\\u003cstrong\\u003e(ii\\u003c/strong\\u003e\\u003c/em\\u003e\\u003cstrong\\u003e) \\u003c/strong\\u003eBALB/c, using the STRING database to supply edge lists. Networks comprise the interactions between the filtered top 100 most differentially expressed genes specific to each strain, ranked by geneD (\\u003cem\\u003eMethods\\u003c/em\\u003e) score. Each node represents a protein and each edge represents a functional protein interaction. Nodes filled with solid colour represent nodes predicted to constitute a module component by a single stochastic iteration of the Louvain algorithm, when applied to a network generated from the corresponding\\u003cem\\u003e \\u003c/em\\u003eunfiltered gene list. Modules that did not possess a single STRING edge within the final filtered network were discarded. Nodes are coloured by module identity; colour coding is independent between \\u003cem\\u003e\\u003cstrong\\u003e(i) \\u003c/strong\\u003e\\u003c/em\\u003eand \\u003cem\\u003e\\u003cstrong\\u003e(ii\\u003c/strong\\u003e\\u003c/em\\u003e\\u003cstrong\\u003e)\\u003c/strong\\u003e. \\u003cstrong\\u003e(C) \\u003c/strong\\u003eTransformed absolute expression patterns across\\u003cem\\u003e \\u003c/em\\u003etreatment groups and strains vary between key species-specific fibrotic DEGs: \\u003cem\\u003eCtsk\\u003c/em\\u003e, \\u003cem\\u003eIl6\\u003c/em\\u003e, \\u003cem\\u003eMmp13\\u003c/em\\u003e, and \\u003cem\\u003eNos2\\u003c/em\\u003e.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure4.jpeg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5295459/v1/d8dac8e2d4e43e63e8f5e883.jpeg\"},{\"id\":68744560,\"identity\":\"9a569f79-e6e4-4c0b-831e-2d9136c1102f\",\"added_by\":\"auto\",\"created_at\":\"2024-11-11 14:56:34\",\"extension\":\"jpeg\",\"order_by\":5,\"title\":\"Figure 5\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":1956216,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eFOSL1-sensitive genes are very significantly overrepresented in cross- strain and in fibrotic DEG sets: (A) \\u003c/strong\\u003eBar charts used to visualise inter-strain DEG set enrichment for genes known to be sensitive to specific TF perturbation (enrichR). Term names encode TF name, type of perturbation, species, GSE identifier, number of genes in term set, and DE direction. Adjusted p-value (p.adj) is a measure of significance. This analysis was conducted for DEGs between strains (C57BL/6 and BALB/c), for both saline- (green) and bleomycin-treated (red) mice. \\u003cstrong\\u003e(B) \\u003c/strong\\u003eThe same analysis and visualisation as in (A) applied to \\u003cstrong\\u003e(i)\\u003c/strong\\u003e DEG sets distinguishing control and fibrotic states in each strain independently and subsequently \\u003cstrong\\u003e(ii) \\u003c/strong\\u003eto filtered versions of the same sets, containing only strain-specific fibrotic DEGs. \\u003cstrong\\u003e(C)(\\u003c/strong\\u003e\\u003cem\\u003e\\u003cstrong\\u003ei\\u003c/strong\\u003e\\u003c/em\\u003e\\u003cstrong\\u003e) \\u003c/strong\\u003e\\u0026nbsp;\\u003cem\\u003eCtsk \\u003c/em\\u003eexpression measured by qRT-PCR in untreated mouse lung fibroblasts (MLFs), normalised to housekeeper: TATA-binding protein (TBP), n=1, two technical replicates \\u003cstrong\\u003e(\\u003c/strong\\u003e\\u003cem\\u003e\\u003cstrong\\u003eii\\u003c/strong\\u003e\\u003c/em\\u003e\\u003cstrong\\u003e) \\u003c/strong\\u003e\\u003cem\\u003eFosl1\\u003c/em\\u003e, \\u003cem\\u003eThbs1\\u003c/em\\u003e, \\u003cem\\u003eMmp2\\u003c/em\\u003e, and \\u003cem\\u003ePrdm1\\u003c/em\\u003e expression measured by qRT-PCR in untreated BALB/c MLFs relative to C57BL/6, n=1, two technical replicates.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure5.jpeg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5295459/v1/d55594e1a9314ba475cbb7d9.jpeg\"},{\"id\":70593811,\"identity\":\"0ec377be-2549-45bc-ae8e-59fd2c42b624\",\"added_by\":\"auto\",\"created_at\":\"2024-12-04 17:47:03\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":8651561,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5295459/v1/bdbf59da-fe30-48fa-896f-ad9257cff1ba.pdf\"},{\"id\":68743659,\"identity\":\"458561dc-7481-4526-86ee-e63edf9dfca2\",\"added_by\":\"auto\",\"created_at\":\"2024-11-11 14:48:35\",\"extension\":\"docx\",\"order_by\":12,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":2482150,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryMaterial.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5295459/v1/f59e4d8d25f38d147585c2a0.docx\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"A Tale of Two Mice: genetics of model mouse strains suggest a transcriptional basis for risk and resistance in idiopathic pulmonary fibrosis\",\"fulltext\":[{\"header\":\"Background\",\"content\":\"\\u003cp\\u003eFibrosis is a multisystem disease that can affect a range of tissues; it is driven by immune cell recruitment and uncontrolled epithelial ‘repair’ processes (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). Clinically, the condition is characterised by ongoing inflammation and epithelial-to-mesenchymal transition (EMT), e.g. of alveolar cells, accompanied by thickening of the extracellular matrix (ECM); also known as fibrogenesis (\\u003cspan additionalcitationids=\\\"CR2\\\" citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e–\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e). When this disease emerges within human lung interstitium, it is typically terminal and often without obvious cause – idiopathic pulmonary fibrosis (IPF) being the most common type of lung fibrosis (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e4\\u003c/span\\u003e). However, pulmonary injury is a plausible trigger. IPF is typically symmetric in humans, perhaps due to propagation of fibrosis from diseased to contralateral lungs. Indeed, asymmetry is a marker for lower survival rates (\\u003cspan additionalcitationids=\\\"CR6 CR7\\\" citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e–\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e). That said, evidence from rodent models also lends credence to the hypothesis of compensatory growth of contralateral – or even diseased – lung volume as a response to fibrogenesis (\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e9\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003ePreclinical research into IPF has focused primarily on murine models, with the reference strains – C57BL/6J and closely related strains, e.g. C57BL/6N – being especially popular (\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR11\\\" class=\\\"CitationRef\\\"\\u003e11\\u003c/span\\u003e); here collectively termed C57BL/6. However, experimentation in alternative genetic backgrounds has identified divergence within IPF phenotypes. Their relative resistance or susceptibility is defined by the severity of pathological symptoms – unresolved collagenic fibrogenesis (scarring) in particular (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e). Within this schema, C57BL/6 strains are classified as susceptible, whereas BALB/c strains are resistant – despite similar immune responses under fibrosis induction (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e). Conversely, C57BL/6 is resistant to hepatic fibrosis, to which BALB/c is susceptible, illustrating that genetic susceptibility can furthermore be tissue-dependent (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eRegarding murine pulmonary fibrosis, resistance may actually be the default. For strains where IPF phenotype has been evaluated, only C57BL/6 mice are fibrosis-susceptible – in contrast to BALB/c, DBA/2, Swiss, A/J, and C3H (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e). C57BL/6 has long been known to exhibit a high degree of resistance to radiation-induced pneumonitis (\\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e). This likely factored into its initial development as an inbred strain for oncology research (\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e). Thus the convention of C57BL/6 as ‘standard’ may, paradoxically, be due to its exceptional lung physiology.\\u003c/p\\u003e \\u003cp\\u003eUnlike humans, most mice are able to recover from fibrosis (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e), albeit in a strain-dependent way. Whilst C57BL/6 strains can resolve scars despite repeated lung injury, DBA/2 strains subjected to the same treatment accumulate fibrotic tissue and develop prolonged disease (\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e). Some researchers have posited a negative correlation between T helper 2 (Th2) cell population size and mouse strain ability to resolve lung fibrosis. DBA/2 and BALB/c immunity is biassed towards Th2 cells, whilst C57BL/6 is biassed towards Th1 cells. As the cell types produce distinct cytokine profiles, this immunological divergence has been proposed to differentiate IPF phenotype (\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003eBleomycin (BLM) and Transforming Growth Factor Beta (TGFβ) are agents commonly used to induce fibrosis in IPF models. BLM triggers pathogenesis via DNA damage (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e), whilst TGFβ acts downstream as a central node in the ECM deposition pathway (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e). TGFβ is one of the key pro-fibrotic cytokines, and has also been used to induce fibrosis in rodent lung tissue \\u003cem\\u003ein vivo\\u003c/em\\u003e and \\u003cem\\u003ein vitro\\u003c/em\\u003e (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e), as well as in human cell culture \\u003cem\\u003ein vitro\\u003c/em\\u003e (\\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e19\\u003c/span\\u003e). These distinct modes of fibrosis induction could make research susceptible to model choice, given murine genetic variation.\\u003c/p\\u003e \\u003cp\\u003eIt has been hypothesised that BLM-sensitivity underpins BALB/c resistance to IPF. Early research suggested that BALB/c, in particular, may display a BLM-specific resistance due to higher basal expression of the detoxifying enzyme bleomycin hydrolase (BLMH) (\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e); it has further been demonstrated that BALB/c fails to express the fibrosis-characteristic Connective Tissue Growth Factor (CTGF) in response to BLM, but may yet do so in response to TGFβ (\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e). If the BLM-specific resistance hypothesis were true, it would be expected that the delivery of fibrogenic factors downstream of injury could induce full fibrosis in this strain (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). However, not only does CTGF alone fail to induce fibrosis in BALB/c, but combined CTGF + BLM treatment results in full fibrosis (\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e). Moreover, recent work on strain-specific responses conflicts with the BLM-sensitivity paradigm, by showing that BALB/c – unlike the control C57BL/6 strain – fails to upregulate the pro-fibrotic collagenase Mmp1 in response to TGFβ treatment (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e). There is thus increasing evidence for a more general resistance mechanism.\\u003c/p\\u003e \\u003cp\\u003eMuch of the challenge in characterising fibrotic susceptibility arises from the disease’s genetic complexity. A recent meta-analysis identified 213 unique factors as pulmonary fibrosis-related genes (PFRGs; Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). They are enriched for age-related genes, specifically: pro-fibrotic genes were pro-ageing, and anti-fibrotic genes were anti-ageing (\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e). It is also likely that this analysis underestimates the contributing genetic factors, as it omits several markers identified elsewhere, including structural proteins and matrix metalloproteinases (\\u003cspan additionalcitationids=\\\"CR25\\\" citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e–\\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e). A biological interpretation of genetic susceptibility to fibrosis may benefit from a genomic, rather than a purely gene-level, strategy. Longitudinal observations of transcriptomes in the C57BL/6 BLM-IPF model support this approach, as enrichment analysis of the differential expression profiles – comparing healthy and fibrotic tissue – recapitulates the clinically observed processes of inflammation and ECM over-synthesis (\\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e\\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e\\u003ctable float=\\\"Yes\\\" id=\\\"Tab1\\\" border=\\\"1\\\"\\u003e\\u003ccaption language=\\\"En\\\"\\u003e \\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 1\\u003c/div\\u003e \\u003cdiv class=\\\"CaptionContent\\\"\\u003e \\u003cp\\u003ePulmonary fibrosis-related gene sets investigated in the context of strain-specific fibrosis responses\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/caption\\u003e\\u003ccolgroup cols=\\\"2\\\"\\u003e\\u003c/colgroup\\u003e\\u003cthead\\u003e\\u003ctr\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eSet\\u003c/p\\u003e \\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eMembers\\u003c/p\\u003e \\u003c/th\\u003e\\u003c/tr\\u003e\\u003c/thead\\u003e\\u003ctbody\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ePulmonary fibrosis-related genes (PFRGs)\\u003c/p\\u003e \\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eSee Toren et al. (2021) e.g. Ackr2, Zmpste24\\u003c/p\\u003e \\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eExtended PFRGs\\u003c/p\\u003e \\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eSee Toren et al. (2021), Todd et al. (2020),Blaauboer et al. (2014) \\u0026amp; Onursal et al. 2021 e.g. Mmp13\\u003c/p\\u003e \\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCommon PFRGs\\u003c/p\\u003e \\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eTimp1, Spp1, Cxcl10, Ccl2, Retnla\\u003c/p\\u003e \\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eC57BL/6-specific PFRGs\\u003c/p\\u003e \\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eIl10, Il17a, Nos2, Rag1, S100a4\\u003c/p\\u003e \\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eBALB/c-specific PFRGs\\u003c/p\\u003e \\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eKlf4, Igf1, Ctsk, Cx2cr1, Hspa1a, Hspa1b\\u003c/p\\u003e \\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eC57BL/6-specific interleukins\\u003c/p\\u003e \\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eIl10, Il17a, Nos2, Rag1, S100a4\\u003c/p\\u003e \\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eBALB/c-specific interleukins\\u003c/p\\u003e \\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eIl6, Il9, Il12b, Il22\\u003c/p\\u003e \\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eMarker matrix proteins\\u003c/p\\u003e \\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eSee Blaauboer et al. (2014) \\u0026amp; Onursal et al. (2021), e.g. Col1a1\\u003c/p\\u003e \\u003c/td\\u003e\\u003c/tr\\u003e\\u003c/tbody\\u003e\\u003c/table\\u003e\\u003c/div\\u003e \\u003cp\\u003e\\u003c/p\\u003e \\u003cp\\u003eIn this study, a transcriptome-level analysis is applied to a dataset comprising bulk RNA-seq reads from a BLM-IPF model using two distinct strains: C57BL/6 and BALB/c. As the former is C57-derived – the latter being from Castle’s mice – they are genetically divergent strains, as well as differing in fibrosis susceptibility (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e27\\u003c/span\\u003e). This dataset therefore captures both the genotypic and phenotypic heterogeneity of IPF model mice.\\u003c/p\\u003e \\u003cp\\u003eDifferential expression and enrichment analyses of the data were consistent with the findings of previous research in C57BL/6, with fibrosis in each genotype being characterised by immunological and epithelial processes (\\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e). However, there was also variation in the identity and functional profiles of the genes differentially expressed under fibrosis in the two genetic backgrounds, consistent with the reported differences in C57BL/6 versus BALB/c IPF risks (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e). These sets of differentially expressed genes (DEGs) included a number of interleukins and other key PFRGs established in the literature (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e), some of which re-emerged in this study during DEG network analysis.\\u003c/p\\u003e \\u003cp\\u003eBLM-induced DEGs and cross-strain DEGs – the latter in both control and BLM-treated groups – also showed highly significant enrichment for genes sensitive to one transcription factor (TF): FOSL1. This was in keeping with literature establishing FOSL1 as a major factor in modulating IPF – and it built upon it, suggesting that FOSL1 may play a principal role in differentiating C57BL/6 and BALB/c transcriptomes, including under fibrosis. The hypothesis of FOSL1 forming a transcriptional basis for BALB/c resistance to IPF was further developed by genetic analysis, utilising the Sanger Mouse Project database. These genetic analyses also undermined existing BLM-dependent and Th2-bias hypotheses for BALB/c IPF response (\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e). These analyses were complemented by TGFβ assays \\u003cem\\u003ein vitro\\u003c/em\\u003e, comparing strain IPF response under this independent experimental model.\\u003c/p\\u003e \"},{\"header\":\"Methods\",\"content\":\"\\u003ch3\\u003eBleomycin mouse model for lung fibrosis\\u003c/h3\\u003e\\u003cp\\u003eAdult mice from three strains were administered 1 mg/ml BLM (BI3543, Chemtronica Sweden) in saline as a single 20 µg/mouse dose intratracheally into their left lung to model pulmonary fibrosis: BALB/c, C57BL/6N, and C57BL/6J. Saline-treated (control) mice were also maintained for each population. Doses were delivered as 30 µL of bleomycin solution followed by 40 µL of air. Tissue was sampled from healthy, fibrotic, and contralateral lungs between days 7 and 14; corresponding to expected fibrogenic period of the treated lung tissue.\\u003c/p\\u003e\\u003ch2\\u003eRNA sequencing of murine tissues\\u003c/h2\\u003e\\u003cp\\u003eCollected tissue samples were prepared by snap-freezing and then grinding to a fine powder using a pre-chilled mortar and pestle. 1 ml of Trizol was added per 50–100 mg of tissue and further shredded by passing through a 21G needle. 100 ng RNA was used to prepare an indexed library with the NuGEN Ovation RNA-seq V2 kit (7102, Tecan) and the Ovation Ultra Low kit (S02366A-FG, Tecan) following manufacturer protocol. Libraries were sequenced using HiSeq 4000 (single-end 50 bp).\\u003c/p\\u003e\\u003ch3\\u003eRNA read quality control and alignment\\u003c/h3\\u003e\\u003cp\\u003eRead quality checking was conducted using FastQC (v0.11.9) both before and after trimming with Trimmomatic (v0.39; leading = 3, trailing = 3, slidingwindow = 4:15, minlen = 36). Samples with lower than 10\\u0026nbsp;million total reads were resequenced and the resulting technical replicates concatenated.\\u003c/p\\u003e\\u003cp\\u003eSubsequently, reads were aligned to the mouse genome; genome annotation was downloaded from the Gencode database (vM23) and chromosomal reference sequence from the UCSC goldenPath database (mm10). Alignment and indexing were both conducted using STAR (v2.7.9a; runTHreadN = 10, sjdbOverhang = 49).\\u003c/p\\u003e\\u003cp\\u003eFinally, aligned reads were annotated using featureCounts (v2.0.1;) and all steps checked with MultiQC (v1.11). All subsequent analyses of read data, including visualisation of MultiQC output, was conducted within RStudio (v4.1.1 – “Kick Things”). Anomalous samples as identified from hierarchical clustering and dimension reduction were excluded from further analysis: 5 samples were excluded as a result, all of which were C57BL/6N, 3 being BLM-treated and 2 being saline-treated. As this significantly reduced the statistical power of the C57BL/6N treated group (n = 2), C57BL/6J was the focus of downstream analysis on C57BL/6.\\u003c/p\\u003e\\u003ch3\\u003eLinear dimension reduction and clustering\\u003c/h3\\u003e\\u003cp\\u003erlog-normalised transcriptomes were characterised through principal component analysis, using the default R-package stats (v4.1.1). PCA results were visualised using ggplot2 (v3.3.5). Unsupervised hierarchical clustering and correlation analyses were conducted and visualised using the same software; misclustered samples were excluded as anomalous and all expression analyses rerun without them.\\u003c/p\\u003e\\u003ch3\\u003eDifferential expression analysis\\u003c/h3\\u003e\\u003cp\\u003eRead count matrices were filtered for protein-coding autosomal genes and parsed by DESeq2 (v1.34.0), then (a) normalised using the regularised log transformation (rlog) and – independently – (b) used to calculate differential expression within and between each strain. Heatmaps for rlog-transformed absolute expression data were generated using pheatmap (v1.0.12) and box plots using ggplot2 (v3.3.5). Strain-specific DEGs were determined for each strain (BALB/c, C57BL/6J, C57BL/6N) independently. Results for the two C57BL/6 sub-strains were later aggregated to provide a set of C57BL/6-specific DEGs for network analysis.\\u003c/p\\u003e\\u003ch3\\u003eOntology term enrichment analyses\\u003c/h3\\u003e\\u003cp\\u003eDifferential expression data was tested within each combination of strain and treatment condition, using clusterProfiler (v4.2.2) to evaluate enrichment of statistically significant DEGs (p \\u0026lt; 0.05; |logFC|\\u0026gt;2). Results were visualised with enrichplot (v1.14.2). GO biological process enrichment (\\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e) was visualised, using enrichGO (OrgDb = org.Mm.eg.db), for GO terms known to be associated with fibrosis pathology based on Deng et al. (2020) and Wang et al. (2021), where the adjusted p-value was \\u0026gt; 0.05 (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e). KEGG pathway enrichment was visualised for all terms where the adjusted p-value was \\u0026gt; 0.0005 (\\u003cspan citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e). Textual reporting of network cluster enrichment excludes enrichment for categories where the adjusted p-value was \\u0026gt; 10\\u003csup\\u003e− 4\\u003c/sup\\u003e.\\u003c/p\\u003e\\u003ch2\\u003ePerturbation enrichment analysis\\u003c/h2\\u003e\\u003cp\\u003eCross-strain, cross-treatment, and strain-specific cross-treatment DEG sets were used as input to transcription factor (TF) perturbation enrichment analysis (enrichR, “TF Perturbations followed by expression”) (\\u003cspan additionalcitationids=\\\"CR32\\\" citationid=\\\"CR31\\\" class=\\\"CitationRef\\\"\\u003e31\\u003c/span\\u003e–\\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e). Terms for this analysis are derived from a meta-analysis of studies conducting TF perturbations with follow-up RNA-seq – thus encompassing gene sets sensitive to specific TFs. enrichR calculates significance using Fisher’s exact test assuming a binomial distribution; p-values adjusted for multiple testing corrections (Benjamini-Hochberg) are then calculated.\\u003c/p\\u003e\\u003ch3\\u003eProtein interaction network construction\\u003c/h3\\u003e\\u003cp\\u003eNetworks were constructed for protein-protein interactions between the products of defined gene sets, using STRINGdb (v2.6.5; network_type = functional, required_score = 500, network_flavour = confidence) supported by rbioapi (v0.7.4) (\\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e34\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR35\\\" class=\\\"CitationRef\\\"\\u003e35\\u003c/span\\u003e). DEG scoring for construction of prioritised protein-protein interaction networks utilised the geneD metric (geneD = -log10(p-value)*|logFC|) to account for both significance and fold-change of expression (\\u003cspan citationid=\\\"CR36\\\" class=\\\"CitationRef\\\"\\u003e36\\u003c/span\\u003e).\\u003c/p\\u003e\\u003ch3\\u003eLouvain algorithm implementation\\u003c/h3\\u003e\\u003cp\\u003eNetwork nodes were clustered using the Louvain algorithm, implemented by igraph (v1.3.0). This was conducted using the full set of differentially expressed genes for each network, prior to filtering by geneD score. Strain-specificity of a module for a given strain X (‘parent’ module) was evaluated as the largest proportion of that module’s nodes present together in a single module in the strain-specific network for X (‘child’ module); this corresponds to a single stochastic iteration of the Louvain algorithm. Specificity was considered significant where the conserved node proportion in a child module was \\u0026gt; 45%, for a parent module of size \\u0026gt; 10 nodes. Subsequent KEGG enrichment analysis was conducted on the parent module.\\u003c/p\\u003e\\u003ch2\\u003eSingle nucleotide polymorphism (SNP) calling and variant effect predictions\\u003c/h2\\u003e\\u003cp\\u003eGenetic regions of interest were queried for variants using the online interface for the Mouse Genomes Project (REL-1505; GRCm38), filtering by strain for C57BL/6NJ, BALB/cJ, and DBA/2J (\\u003cspan citationid=\\\"CR37\\\" class=\\\"CitationRef\\\"\\u003e37\\u003c/span\\u003e). Variant effect prediction was conducted with the Ensembl VEP (\\u003cspan citationid=\\\"CR38\\\" class=\\\"CitationRef\\\"\\u003e38\\u003c/span\\u003e). Gene sequences required for downstream analysis were extracted using the Multiple Genome Viewer (MGV) from Mouse Genome Informatics (MGI) (\\u003cspan citationid=\\\"CR39\\\" class=\\\"CitationRef\\\"\\u003e39\\u003c/span\\u003e).\\u003c/p\\u003e\\u003ch2\\u003eGenome-wide association study data mining\\u003c/h2\\u003e\\u003cp\\u003eGenome-wide association study (GWAS) data on familial pulmonary fibrosis (FPF) was extracted from Genomics England’s 100,000 Genomes Project (\\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e). Variants were filtered by gene to analyse data on \\u003cem\\u003eMmp13\\u003c/em\\u003e, \\u003cem\\u003eCtsk\\u003c/em\\u003e, \\u003cem\\u003eIl6\\u003c/em\\u003e, \\u003cem\\u003eFosl1\\u003c/em\\u003e, \\u003cem\\u003eNos2\\u003c/em\\u003e, and \\u003cem\\u003eNos3\\u003c/em\\u003e. Association significance of these variants was assessed with respect to age of death, as a metric to capture disease progression. The result was visualised by Manhattan Plot to clearly depict the significance of each SNP across the genome. Data was similarly extracted from an AstraZeneca-funded whole-exome study of IPF (\\u003cspan citationid=\\\"CR41\\\" class=\\\"CitationRef\\\"\\u003e41\\u003c/span\\u003e).\\u003c/p\\u003e\\u003ch2\\u003eProtein structure predictions and alignment\\u003c/h2\\u003e\\u003cp\\u003eThe 3-D structures of FOSL1 were modelled using ColabFold (template_mode = pdb70, num_recycles = 12) (\\u003cspan citationid=\\\"CR42\\\" class=\\\"CitationRef\\\"\\u003e42\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR43\\\" class=\\\"CitationRef\\\"\\u003e43\\u003c/span\\u003e). Output PDB files were visualised and analysed using iCn3D Structure Viewer (v3.11.6) and PyMOL (v2.5.2).\\u003c/p\\u003e\\u003ch2\\u003eIn vitro TGFβ-1 model for lung fibrosis\\u003c/h2\\u003e\\u003cp\\u003eMouse Lung Fibroblasts (MLFs) from BALB/c and C57BL/6 mice were purchased from Generon and cultured up to passage 7 in complete fibroblast medium with supplements (M2267, Generon) on a gelatin-based coated surface (6950, Generon) at 37°C, 5% CO\\u003csub\\u003e2\\u003c/sub\\u003e. For fibrosis modelling, cells were plated at 30,000/cm\\u003csup\\u003e2\\u003c/sup\\u003e density. After 24h, cells were sub-cultured in serum-free complete fibroblast media, and after another 24h, mouse recombinant TGFβ-1 (ABclonal, RP0116) was added for 72h at 0.1–10 ng/ml concentration. MLFs from both strains were confirmed for the strain-specific FOSL1 mutations using Sanger Sequencing.\\u003c/p\\u003e\\u003ch2\\u003eQuantitative Polymerase Chain Reaction (qPCR)\\u003c/h2\\u003e\\u003cp\\u003eCells were lysed after several washes with PBS in plate using RLT buffer with 0.1% β-mercaptoethanol. After RNA extraction with an RNeasy kit (Qiagen), cDNA was synthesised using M-MuLV reverse transcriptase (M0253L, NEB). Quantitative real time polymerase chain reaction (qPCR) was performed using Fast SYBR 2x Green Master Mix (ThermoFisher Scientific).\\u003c/p\\u003e\\u003ch2\\u003eImmunocytochemistry\\u003c/h2\\u003e\\u003cp\\u003eCells were fixed with 4% paraformaldehyde for 15 min at room temperature, and stained with anti-𝛼-sma antibody (ab184675 at 1:200, Abcam)and anti-collagen 1 (ab34710) in 3% bovine serum albumin in 0.1% Triton solution at 4℃ overnight. After PBS washes, 20 min DAPI staining (1:1000 of 1 mg/ml in PBS) was performed.\\u003c/p\\u003e\"},{\"header\":\"Results\",\"content\":\"\\u003cdiv id=\\\"Sec18\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eBleomycin-treated lung transcriptomes reflect multicellular fibrotic processes\\u003c/h2\\u003e \\u003cp\\u003eComparison of transcriptomic data from fibrotic versus healthy murine lung tissue shows a clear and consistent difference in gene expression. Principal component analysis separates bleomycin-treated samples from saline-treated controls along the first component (PC1, 14.9%; Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eA), whilst the second component separates two distinct genetic strains of mice: BALB/c and C57BL/6.\\u003c/p\\u003e \\u003cp\\u003ePC1, describing the strain-independent response to fibrosis induction, is enriched for fibrosis-associated processes such as the inflammatory response and extracellular matrix organisation (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB). The set of upregulated genes contributing to PC1 were also enriched for many KEGG terms known to be correlated with fibrosis progression (\\u003cem\\u003eSupplementary Fig.\\u0026nbsp;1\\u003c/em\\u003e). Many enriched terms matched those recently identified by Wang et al. (2021) including: lysosome, cytokine-cytokine receptor interaction, haematopoietic cell lineage, leishmaniasis, rheumatoid arthritis, osteoclast differentiation, tuberculosis, phagosome, intestinal immune network for IgA production, chemokine signalling pathway, Toll-like receptor signalling pathway, p53 signalling pathway, amoebiasis, and protein digestion and absorption (\\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003eWhilst both strains upregulate IPF-associated processes, however, their precise overlap is only 48 differentially expressed genes (DEGs). That equates to: \\u0026lt;20% of total DEGs for BALB/c, \\u0026lt;\\u0026thinsp;10% for C57BL/6. This set includes 5 pulmonary fibrosis related genes (PFRGs); the set does also include genes additionally identified within the extended PFRG list (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). That said, a number of fibrotic DEGs are strain-specific in identity, but still contribute towards common or similar biological processes, as defined by GO and KEGG ontologies (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e; \\u003cem\\u003eSupplementary Fig.\\u0026nbsp;1\\u003c/em\\u003e).\\u003c/p\\u003e \\u003cp\\u003eThe majority of transcriptomic variation in contralateral versus healthy tissue is attributable to factors uncorrelated with treatment condition (\\u003cem\\u003eSupplementary Fig.\\u0026nbsp;2\\u003c/em\\u003e). C57BL/6J samples separate out only along a combination of the third (PC3, 11.8%) and fourth components (PC4, 9.9%) and BALB/c samples along PC3 (15.6%); C57BL/6N samples do not cluster by sample type under PC1-4 (\\u003cem\\u003eSupplementary Fig.\\u0026nbsp;2\\u003c/em\\u003e). Consequently, the contralateral lung in all strains is not significantly different from the healthy lung within this model. Analysis of this sample type at the level of gene expression was therefore not pursued further, given the lack of clear support for a contralateral pathology.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec19\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eFibrogenic markers are upregulated in the reference mouse strain relative to BALB/c\\u003c/h2\\u003e \\u003cp\\u003eThe second principal component of variation between fibrotic and control transcriptome data (PC2, 14.1%; Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eA) separates samples along strain, independent of disease status. This is consistent with the observed distinctions in functional enrichment between the upregulated gene sets of each strain (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB). Whilst both strains are indicated to have upregulated immunological processes, there is some strain-differential enrichment for fibrogenic processes (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB). Specifically, C57BL/6 displays upregulation of extracellular matrix organisation and collagen biosynthesis (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB; GO), as well as protein digestion and absorption (KEGG; \\u003cem\\u003eSupplementary Fig.\\u0026nbsp;1\\u003c/em\\u003e); the latter has previously been implicated in late \\u0026ndash; but not early \\u0026ndash; stages of BLM-induced murine IPF (\\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e). In contrast, BALB/c exhibits a more significant inflammatory response, as well as EMT enrichment (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB). However, this analysis does not indicate a corresponding BALB/c-specific elevation of ECM deposition, which the literature would predict as a consequence of elevated level of EMT (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eThe strain dependency of ECM deposition and stress fibre formation was also observed in a TGFβ-based fibrotic assay \\u003cem\\u003ein vitro.\\u003c/em\\u003e Lung fibrosis was modelled using primary mouse lung fibroblasts cultured from adult C57BL/6 and BALB/c mice, with the cytokine TGFβ-1 applied as the inducer of profibrotic inflammatory signalling. Just as observed in a human lung fibroblast model in the literature (\\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e19\\u003c/span\\u003e), an increasing stress fibre deposition was observed with increasing TGFβ-1 concentration in C57BL/6 (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eA), namely fibres of \\u0026#120572;-smooth muscle actin (\\u0026#120572;-sma). BALB/c response to the cytokine was attenuated (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eA).\\u003c/p\\u003e \\u003cp\\u003eAlthough collagen I fibre deposition was not significantly different between the two strains in the TGFβ-1 assay (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eB), C57BL/6 cells did express higher COL1A1 protein levels, consistent with the transcriptomic data (\\u003cem\\u003eFig.\\u0026nbsp;3Ci\\u003c/em\\u003e). This validation is particularly critical given the C57BL/6-specific upregulation of Tgfb1 (the gene encoding TGFβ-1) observed in the fibrotic transcriptomes (\\u003cem\\u003eFig.\\u0026nbsp;3Cii\\u003c/em\\u003e) which contributed to the strain-specific enrichment for positive upregulation of collagen biosynthesis (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB). The transcription factor CREB3L1 was the other gene contributing towards this enrichment (\\u003cem\\u003eFig.\\u0026nbsp;3Cii\\u003c/em\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec20\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eKey differentially expressed pulmonary fibrosis-related genes and modules are strain-specific\\u003c/h2\\u003e \\u003cp\\u003eFiltering for PFRGs (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e) and plotting their transformed expression in fibrotic tissue, it is apparent that many exhibit strain-specificity (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eA). Calculating differential expression between fibrotic and healthy tissue separately for each genotype, strain-specific DEGs can further be identified (\\u003cem\\u003eFig.\\u0026nbsp;4Bi-ii\\u003c/em\\u003e). This supports a model in which strain-specific fibrotic gene expression can be phenomenologically categorised into (\\u0026ldquo;Category 1\\u0026rdquo;) genes that are consistently expressed at different levels across strains in a condition-independent manner and (\\u0026ldquo;Category 2\\u0026rdquo;) genes that are differentially regulated under fibrosis (condition-dependent expression).\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003eOf all strain-specific DEGs identified (Category 2), there are 5 PFRGs that are C57BL/6-specific and 6 PFRGs that are BALB/c-specific (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). Additionally, several interleukins are Category 2 genes (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e); Il6, in particular, is the central node of the interaction network for BALB/c-specific DEGs (\\u003cem\\u003eFig.\\u0026nbsp;4Bii\\u003c/em\\u003e and Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eC). This condition-dependent specificity may contribute to the distinct Th1 versus Th2 immunology of C57BL/6 vs BALB/c, as Il6 is implicated in Th1/Th2 fate determination (\\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e44\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eCollectively, Category 2 genes can comprise functional modules. Several KEGG pathway-enriched modules appear C57BL/6-specific, associated variously with neuroactive ligand-receptor interaction, chemokine signalling, cytokine-cytokine receptor interaction, ECM-receptor interaction, phototransduction, and TGFβ signalling (\\u003cem\\u003eFig.\\u0026nbsp;4Bi\\u003c/em\\u003e). It should be noted that some of these terms correlate with late-stage IPF specifically in the literature, i.e. ECM-receptor interaction (\\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eThis Category 2 enrichment for key fibrogenic functions \\u0026ndash; i.e. signalling via cytokines/chemokines/TGFβ, ECM remodelling \\u0026ndash; supports the hypothesis that strain-specific gene regulation can shape fibrosis susceptibility (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan additionalcitationids=\\\"CR46\\\" citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e45\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR47\\\" class=\\\"CitationRef\\\"\\u003e47\\u003c/span\\u003e). There was no significant KEGG enrichment within the strain-specific BALB/c modules identified (\\u003cem\\u003eFig.\\u0026nbsp;4Bii\\u003c/em\\u003e). Given the high C57BL/6 fibrosis risk among mice, this finding could indicate that pro-fibrotic processes in C57BL/6 underpin its exceptional susceptibility, rather than the converse for BALB/c (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eWithin geneD-filtered DEGs (\\u003cem\\u003eMethods\\u003c/em\\u003e), \\u003cem\\u003eNos2\\u003c/em\\u003e is the only C57BL/6-specific PFRG (\\u003cem\\u003eFig.\\u0026nbsp;4Bi\\u003c/em\\u003e and Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eC) (\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e). In contrast, all BALB/c-specific PFRGs pass filtering (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e; \\u003cem\\u003eFig.\\u0026nbsp;4Bii\\u003c/em\\u003e). Of these top-scoring Category 2 PFRGs, only \\u003cem\\u003eCtsk\\u003c/em\\u003e is differentially regulated in a direction that might account for observed fibrosis susceptibilities, as an unambiguously anti-fibrotic gene most strongly upregulated in BALB/c (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eC).\\u003c/p\\u003e \\u003cp\\u003e \\u003cem\\u003eMmp13\\u003c/em\\u003e is another Category 2 anti-fibrotic factor upregulated in BALB/c \\u0026ndash; and a direct interaction partner of \\u003cem\\u003eCtsk\\u003c/em\\u003e (\\u003cem\\u003eFig.\\u0026nbsp;4Bii-C\\u003c/em\\u003e) (\\u003cspan citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e45\\u003c/span\\u003e). Absolute expression of \\u003cem\\u003eMmp13\\u003c/em\\u003e, however, is lower in untreated BALB/c than in C57BL/6 (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eC). Thus, whilst both factors fall into Category 2, we hypothesise that: whereas \\u003cem\\u003eCtsk\\u003c/em\\u003e may be actively more positively regulated under fibrosis in BALB/c, \\u003cem\\u003eMmp13\\u003c/em\\u003e may instead be relieved from strain-specific basal repression when fibrosis develops in the BALB/c background.\\u003c/p\\u003e \\u003cp\\u003eWithin this framework, strain-specific upregulation of \\u003cem\\u003eCtsk\\u003c/em\\u003e could plausibly underlie BALB/c lung fibrosis resistance, but any putative function of \\u003cem\\u003eMmp13\\u003c/em\\u003e is less clear, since it is expressed at similar levels in healthy tissue (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eC). Interpretation is further complicated by the conflicting roles reported for \\u003cem\\u003eMmp13\\u003c/em\\u003e, variously promoting and protecting against fibrogenesis (\\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e45\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR48\\\" class=\\\"CitationRef\\\"\\u003e48\\u003c/span\\u003e).\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec21\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eFOSL1 targets are overrepresented in cross-strain and cross-treatment differential expression\\u003c/h2\\u003e \\u003cp\\u003eSupplementing the gene-level analysis above, regulatory analysis was conducted in parallel on the expression data. Several DEG sets of interest \\u0026ndash; including those directly differentially expressed between strains \\u0026ndash; were tested for enrichment of genes sensitive to a comprehensive range of TF perturbations. The results clearly show that cross-strain DEGs, independent of treatment, were significantly enriched for genes sensitive to perturbation of the TF FOSL1 (control p.adj\\u0026thinsp;=\\u0026thinsp;0.001, fibrotic p.adj\\u0026thinsp;=\\u0026thinsp;0.01; Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eA). Moreover, under control conditions, FOSL1-sensitive genes were the only set significantly enriched \\u0026ndash; suggesting FOSL1 (ENSMUSG00000024912, GRCm38) may be of primary importance in distinguishing C57BL/6 versus BALB/c phenotypes at the transcriptional level. The set (GSE43965) was identified by Rajasekaran et al. (2013) (\\u003cspan citationid=\\\"CR49\\\" class=\\\"CitationRef\\\"\\u003e49\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003eApplying the same analysis to cross-treatment DEGs, the results again indicate overrepresentation of FOSL1-sensitive genes, with this being the most significantly enriched term in both C57BL/6 (p.adj\\u0026thinsp;=\\u0026thinsp;8.1e-18) and BALB/c (p.adj\\u0026thinsp;=\\u0026thinsp;8.7e-55) fibrotic DEGs (\\u003cem\\u003eFig.\\u0026nbsp;5Bi\\u003c/em\\u003e). This suggests FOSL1 might play a role in the transition between healthy and fibrotic lung tissue. However, when the analysis was performed on strain-specific fibrotic DEGs, only BALB/c-specific DEGs remained enriched for FOSL1-sensitive genes (\\u003cem\\u003eFig.\\u0026nbsp;5Bii\\u003c/em\\u003e). This result would be consistent with an additional or altered role for FOSL1 in BALB/c lung fibrosis pathology.\\u003c/p\\u003e \\u003cp\\u003eIndependently, \\u003cem\\u003ein vitro\\u003c/em\\u003e TGF\\u0026#120573; assays showed that \\u003cem\\u003eCtsk\\u003c/em\\u003e \\u0026ndash; target of the heterodimeric TF AP-1, through which FOSL1 acts (\\u003cspan additionalcitationids=\\\"CR50\\\" citationid=\\\"CR49\\\" class=\\\"CitationRef\\\"\\u003e49\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR51\\\" class=\\\"CitationRef\\\"\\u003e51\\u003c/span\\u003e) \\u0026ndash; was upregulated in BALB/c relative to C57BL/6 (\\u003cem\\u003eFig.\\u0026nbsp;5Ci\\u003c/em\\u003e), albeit in a Category 1 manner as opposed to the Category 2 behaviour observed in the BLM dataset (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eC). This consistency with the \\u003cem\\u003ein vivo\\u003c/em\\u003e results further supports the idea that a mechanism involving \\u003cem\\u003eCtsk\\u003c/em\\u003e contributes to strain-specific IPF resistance in BALB/c. Moreover, established FOSL1 targets curated from the literature (\\u003cspan citationid=\\\"CR52\\\" class=\\\"CitationRef\\\"\\u003e52\\u003c/span\\u003e) \\u0026ndash; e.g. \\u003cem\\u003eThbs1, Prdm1\\u003c/em\\u003e and \\u003cem\\u003eMmp2\\u003c/em\\u003e \\u0026ndash; were differentially expressed between strains (\\u003cem\\u003eFig.\\u0026nbsp;5Cii\\u003c/em\\u003e). Further analysis additionally indicated tighter regulation of these targets in BALB/c than in C57BL/6 (\\u003cem\\u003edata not shown\\u003c/em\\u003e), which would be consistent with the noisier expression of BALB/c versus C57Bl/6 \\u0026#120572;-sma previously observed \\u003cem\\u003ein vitro\\u003c/em\\u003e (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eA). As in the BLM dataset \\u003cem\\u003eFosl1\\u003c/em\\u003e was not differentially expressed between strains either under control conditions (\\u003cem\\u003eFig.\\u0026nbsp;5Cii\\u003c/em\\u003e; \\u003cem\\u003eSupplementary Fig.\\u0026nbsp;3\\u003c/em\\u003e). This provided orthogonal support for a FOSL1-dependent hypothesis of IPF risk, presumably mediated by the post-transcriptional alteration of FOSL1 activity.\\u003c/p\\u003e \\u003cp\\u003e \\u003cem\\u003eP170L variant in BALB/c Fosl1 gene may underpin differential regulation of\\u003c/em\\u003e Ctsk \\u003cem\\u003eand\\u003c/em\\u003e Mmp13\\u003c/p\\u003e \\u003cp\\u003eTo investigate possible genetic causes for strain-differential IPF prognosis, variant calling was conducted between BALB/c and C57BL/6 genomes. This revealed genetic differences that could underlie the differences in the pathophysiology of \\u0026ndash; reflected in the transcriptomic responses to \\u0026ndash; IPF in susceptible versus resistant backgrounds. We focus on both TFs and downstream genes identified from the transcriptomic analysis above \\u0026ndash; \\u003cem\\u003eFosl1\\u003c/em\\u003e, \\u003cem\\u003eCtsk, Mmp13, Il6\\u003c/em\\u003e, and \\u003cem\\u003eNos2\\u003c/em\\u003e \\u0026ndash; and on genes implicated by the literature (\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR53\\\" class=\\\"CitationRef\\\"\\u003e53\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR54\\\" class=\\\"CitationRef\\\"\\u003e54\\u003c/span\\u003e). The latter includes bleomycin hydrolase (\\u003cem\\u003eBlmh\\u003c/em\\u003e), connective tissue growth factor (\\u003cem\\u003eCtgf\\u003c/em\\u003e), and \\u003cem\\u003eNos3\\u003c/em\\u003e. A number of missense exonic variants are mechanistically suggestive, especially the single nucleotide polymorphism (SNP) affecting all \\u003cem\\u003eFosl1\\u003c/em\\u003e transcripts recorded by ENSEMBL (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"Yes\\\" id=\\\"Tab2\\\" border=\\\"1\\\"\\u003e \\u003ccaption language=\\\"En\\\"\\u003e \\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 2\\u003c/div\\u003e \\u003cdiv class=\\\"CaptionContent\\\"\\u003e \\u003cp\\u003eDeleterious point substitutions in BALB/c \\u003cem\\u003eCtgf\\u003c/em\\u003e, \\u003cem\\u003eFosl1\\u003c/em\\u003e, and \\u003cem\\u003eNos3\\u003c/em\\u003e\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/caption\\u003e \\u003ccolgroup cols=\\\"9\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c6\\\" colnum=\\\"6\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c7\\\" colnum=\\\"7\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c8\\\" colnum=\\\"8\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c9\\\" colnum=\\\"9\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eGene\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eChromosome\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003ePosition\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003eSNP\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eC57BL/6\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003eBALB/c\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003eDBA/2\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c8\\\"\\u003e \\u003cp\\u003ePredicted effect\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c9\\\"\\u003e \\u003cp\\u003eSIFT\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCtgf\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e10\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e24595833\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003ers8254419\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eT\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003eC\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003eT\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e \\u003cp\\u003eMissense, start codon lost\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c9\\\"\\u003e \\u003cp\\u003e0.01\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eFosl1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e19\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e5450169\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003ers31137232\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eC\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003eT\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003eC\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e \\u003cp\\u003eMissense\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c9\\\"\\u003e \\u003cp\\u003e0\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eFosl1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e19\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e5500197\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003ers30851424\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eC\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003eT\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003eC\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e \\u003cp\\u003eMissense\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c9\\\"\\u003e \\u003cp\\u003e1\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eNos3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e24369876\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003ers32018659\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eC\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e \\u003cp\\u003eC\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e \\u003cp\\u003eT\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e \\u003cp\\u003eMissense\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c9\\\"\\u003e \\u003cp\\u003e0.01\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003cp\\u003eThe observed \\u003cem\\u003eCtgf\\u003c/em\\u003e start codon variant (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e) could in theory impair expression, elucidating reports that \\u003cem\\u003eCtgf\\u003c/em\\u003e expression is poor in response to BLM alone. This is independent of any BLM-specific hypothesis for fibrosis resistance, as combined BLM and CTGF treatment successfully induces fibrosis (\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e). However, isoform analysis indicates that the rs8254419 variant would not affect the canonical CTGF transcript, but rather a truncated alternative spliceoform (ENSMUST00000129142.1). Additionally, a succession of several indels were identified in \\u003cem\\u003eBlmh\\u003c/em\\u003e from the Sanger Mouse Genome Project, predicted to effect nonsense-mediated transcript decay. This is inconsistent with reports of high \\u003cem\\u003eBlmh\\u003c/em\\u003e expression in BALB/c (\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e), which were themselves here corroborated by a cross-strain comparison within the dataset (logFC\\u0026thinsp;=\\u0026thinsp;0.4, p.adj\\u0026thinsp;=\\u0026thinsp;0.004).\\u003c/p\\u003e \\u003cp\\u003eMore promisingly, a coding variant (rs31137232) was detected in \\u003cem\\u003eFosl1\\u003c/em\\u003e \\u0026ndash; the gene encoding FOSL1, which is itself a bZIP (basic-leucine zipper) protein (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e). This SNP corresponds to a non-conservative leucine zipper-proximal substitution, P170L, indicating a genetic change from a \\u0026ldquo;helix-breaking\\u0026rdquo; to a \\u0026ldquo;helix-inducing\\u0026rdquo; residue (\\u003cspan citationid=\\\"CR55\\\" class=\\\"CitationRef\\\"\\u003e55\\u003c/span\\u003e). It is thus plausible that the P170L mutation structurally alters the C-terminus of the FOSL1 central α-helix, supporting the hypothesis that this BALB/c-specific variant alters bZIP-mediated interactions. This would be consistent with an altered regulatory function of FOSL1 in BALB/c, as FOSL1 may regulate both \\u003cem\\u003eCtsk\\u003c/em\\u003e and \\u003cem\\u003eMmp13\\u003c/em\\u003e through AP-1 dimerization (\\u003cspan citationid=\\\"CR50\\\" class=\\\"CitationRef\\\"\\u003e50\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR51\\\" class=\\\"CitationRef\\\"\\u003e51\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR56\\\" class=\\\"CitationRef\\\"\\u003e56\\u003c/span\\u003e). This could account for their Category 2 expression patterns in the BLM dataset (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eC).\\u003c/p\\u003e \\u003cp\\u003eAn exploratory alignment was additionally conducted to show the structures of FOSL1 with and without the P170L mutation (\\u003cem\\u003eSupplementary Fig.\\u0026nbsp;4\\u003c/em\\u003e). Additionally, the \\u003cem\\u003eFosl1\\u003c/em\\u003e gene was amplified and sequenced in the lung cells from each strain used for TGFβ assays, confirming independently that the P170L mutation was present in BALB/c lines. A second coding variant (rs30851424) is also present in BALB/c albeit predicted to be well tolerated by SIFT scoring (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e). The impact of the L39F mutation is also more difficult to predict, due both to its position in a disordered region and the lack of an empirically determined structure for FOSL1.\\u003c/p\\u003e \\u003cp\\u003eC-terminal Pro of the C57BL/6 FOSL1 bZIP domain is conserved at the equivalent position (170) in human FOSL1, according to the ENSEMBL database. Whilst a P170L SNP is not recorded in the human population, a P170T substitution is present at 0.001% frequency in the population. This is a similarly non-conservative substitution and therefore might be expected to have correlated or anticorrelated effects to the murine P170L mutation. Independently, it was determined that human \\u003cem\\u003eFOSL1\\u003c/em\\u003e has been linked to asthmatic lung disease by genome-wide association studies (GWAS) (\\u003cspan citationid=\\\"CR57\\\" class=\\\"CitationRef\\\"\\u003e57\\u003c/span\\u003e). Moreover, it is a regulator of \\u003cem\\u003eMUC2\\u003c/em\\u003e, a gene not only identified within the same study but also directly implicated in IPF by another GWAS (\\u003cspan additionalcitationids=\\\"CR58\\\" citationid=\\\"CR57\\\" class=\\\"CitationRef\\\"\\u003e57\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR59\\\" class=\\\"CitationRef\\\"\\u003e59\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eIn addition to the murine FOSL1 variant, other mutations of interest offer evidence critical to previous work attempting to link mouse strain genotype and phenotype in lung fibrosis. The deleterious DBA/2 \\u003cem\\u003eNos3\\u003c/em\\u003e variant here reported (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e) maps to all known functional transcripts from the gene. This could account for the documented failure of DBA/2 and \\u003cem\\u003eNos3\\u003c/em\\u003e-deficient C57BL/6 \\u0026ndash; but not the wildtype \\u0026ndash; to resolve pulmonary fibrosis in repeated injury experiments (\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e). It also weakens the theory that Th2-biased immune systems are responsible for impaired fibrosis resolution, particularly since BALB/c (a) lacks this \\u003cem\\u003eNos3\\u003c/em\\u003e mutation, (b) is Th2-biassed, and (c) possesses fibrosis resistance (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eExtending these mutational analyses, both \\u003cem\\u003eNos3\\u003c/em\\u003e and \\u003cem\\u003eFosl1\\u003c/em\\u003e were investigated using human familial pulmonary fibrosis (FPF) whole-genome association study data from the 100,000 Genomes Project (Genomics England), as were other genes of interest: \\u003cem\\u003eMmp13\\u003c/em\\u003e, \\u003cem\\u003eCtsk\\u003c/em\\u003e, \\u003cem\\u003eIl6\\u003c/em\\u003e, and \\u003cem\\u003eNos2\\u003c/em\\u003e. The significant variants identified from this study were predominantly found in \\u003cem\\u003eNos3\\u003c/em\\u003e, consistent with the results in mouse models discussed above (\\u003cem\\u003eSupplementary Fig.\\u0026nbsp;5\\u003c/em\\u003e). Several significant SNPs were also identified in other genes (\\u003cem\\u003eSupplementary Fig.\\u0026nbsp;4, inset; Supplementary Tables\\u0026nbsp;1\\u0026ndash;5\\u003c/em\\u003e). These included Fosl1, wherein 9 FPF-associated SNPs were detected (0.00\\u0026thinsp;\\u0026lt;\\u0026thinsp;p.val\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.045; \\u003cem\\u003eSupplementary Table\\u0026nbsp;1\\u003c/em\\u003e). Independently, whole exome data from AstraZeneca was also analysed to identify SNPs correlated with human IPF in these genes. Whilst the adjusted p-values from this data analysis were non-significant, a significant SNP was identified in \\u003cem\\u003eFosl1\\u003c/em\\u003e prior to correction.\\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"Discussion\",\"content\":\"\\u003cdiv id=\\\"Sec23\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eEffects of fibrosis are observable in diseased lung transcriptomes\\u003c/h2\\u003e \\u003cp\\u003eTranscriptomic profiles are a rich resource for characterising complex diseases. In this analysis, it is shown that the diagnostic markers of idiopathic pulmonary fibrosis (IPF) can be captured at the cellular level as enrichment for transcripts implicated in inflammation or tissue scarring (\\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e, \\u003cspan additionalcitationids=\\\"CR61\\\" citationid=\\\"CR60\\\" class=\\\"CitationRef\\\"\\u003e60\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR62\\\" class=\\\"CitationRef\\\"\\u003e62\\u003c/span\\u003e), among genes differentially expressed (DEGs) in lung tissue treated with bleomycin (BLM) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e; \\u003cem\\u003eSupplementary Fig.\\u0026nbsp;1\\u003c/em\\u003e). This enrichment might further facilitate prognostic inference, as both GO and KEGG analyses highlighted collagenic fibrogenesis/late-stage processes and pathways specific to C57BL/6, but not to BALB/c despite its enrichment for EMT-related GO terms (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e). These results are consistent with the susceptibility of C57BL/6 strains to progressive fibrogenesis \\u0026ndash; in contrast to the less severe inflammatory syndromes observed in other strains such as BALB/c (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eMany fibrotic DEGs interact with each other. These interactions form functional subnetworks which themselves often correspond to pathological processes (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eB). An integrative omics approach of this network-oriented type can give a modular readout for research into fibrosis of the lung, as well as other tissues. The application of this framework to strain-specific pathologies further benefits from the distinction between fibrotic DEGs which differ in expression between strains under all conditions (Category 1) and those differentially regulated with regard to fibrosis (Category 2).\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec24\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eIntra- but not inter-tissue compensatory mechanisms are detectable\\u003c/h2\\u003e \\u003cp\\u003eIn contrast to the above, the contralateral lung does not display any transcriptomic signature distinct from healthy tissue, within this dataset (\\u003cem\\u003eSupplementary Fig.\\u0026nbsp;2\\u003c/em\\u003e). These findings undermine the possibility that transcriptional mechanisms facilitate compensatory activity as proposed in asymmetric murine models (\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e9\\u003c/span\\u003e). Whilst this does not rule out alternative compensation hypotheses within the mouse model \\u0026ndash; such as those hinging on protein translation or modification \\u0026ndash; this conjecture has limited applicability to human disease, regardless, given that asymmetry in pulmonary fibrosis is a strong positive predictor of fibrogenic severity (\\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e6\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eAnti-fibrotic activity is, however, detected within the diseased lung itself. In particular, strain-specific upregulation of \\u003cem\\u003eCtsk\\u003c/em\\u003e and \\u003cem\\u003eMmp13\\u003c/em\\u003e in fibrotic BALB/c tissue (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eC) is consistent with the known IPF-resistance of this strain, as both encode collagenases critical to fibrosis resolution (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e45\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR63\\\" class=\\\"CitationRef\\\"\\u003e63\\u003c/span\\u003e). It could also suggest why BALB/c was not specifically enriched for ECM deposition, despite strain-differential enrichment for EMT processes (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB). Furthermore, this focus on collagen breakdown fits within the emerging paradigm of epithelial lung disease as a balance between extracellular matrix (ECM) deposition and degradation. This posits a phenotypic spectrum graduated by structural protein turnover. For instance, emphysemic and tubercular symptoms have been linked to ECM degradation (including by \\u003cem\\u003eCtsk\\u003c/em\\u003e), whereas deposition is a defining feature of IPF (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR63\\\" class=\\\"CitationRef\\\"\\u003e63\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR64\\\" class=\\\"CitationRef\\\"\\u003e64\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cdiv id=\\\"Sec25\\\" class=\\\"Section3\\\"\\u003e \\u003ch2\\u003ePrediction of murine fibrotic pathologies is interdependent and hierarchical\\u003c/h2\\u003e \\u003cp\\u003eWhilst \\u003cem\\u003eCtsk\\u003c/em\\u003e and \\u003cem\\u003eMmp13\\u003c/em\\u003e are here documented as factors influential in BALB/c resistance to pulmonary fibrosis, \\u003cem\\u003eNos2\\u003c/em\\u003e is conversely a candidate marker for elevated susceptibility as exhibited by the reference strain, C57BL/6 (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eB-C); it is known to promote inflammation in early fibrosis, despite a potential role in long-term resolution (\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR53\\\" class=\\\"CitationRef\\\"\\u003e53\\u003c/span\\u003e). It should be noted that, since these transcripts are selected based on: (a) a meta-analysis of the literature and (b) the significance and magnitude of their differential expression, they do not represent a comprehensive set of predictors for murine fibrosis risk. The expression of multiple C57BL/6-specific modules also highlights the putative uniqueness of C57BL/6 IPF-susceptibility at the gene level (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eIn contrast, from analysis at the regulatory level, the TF FOSL1 emerges as a significant and overlapping influence on both cross-strain and within-strain DEGs (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eA-B). The latter is consistent with its documented role in inhibiting pulmonary fibrosis (\\u003cspan citationid=\\\"CR65\\\" class=\\\"CitationRef\\\"\\u003e65\\u003c/span\\u003e). Highly significant enrichment of FOSL1-sensitive genes among cross-strain DEGs, on the other hand, represents a novel finding. It indicates that FOSL1 may be influential in differentiating C57BL/6 and BALB/c mouse models overall \\u0026ndash; even outside the context of IPF. Within that context, the additional finding that FOSL1-sensitive genes are significantly overrepresented among BALB/c-specific DEGs under fibrosis is highly suggestive of a role for FOSL1 in modulating strain-specific IPF resistance.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec26\\\" class=\\\"Section3\\\"\\u003e \\u003ch2\\u003eGenetic analysis favours a novel FOSL1-modulated hypothesis for IPF risk\\u003c/h2\\u003e \\u003cp\\u003eThe novel hypothesis that FOSL1 is a major modulator of murine variation in IPF physiology emerges as the key innovation of this study. This is not only a consequence of its evident significance in modulating the genes characteristic of BALB/c \\u0026ndash; as opposed to C57BL/6 \\u0026ndash; IPF pathophysiology (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e), as discussed above. Rather, it is this result in conjunction with the subsequent identification of the nonsynonymous P170L mutation present in BALB/c FOSL1, during analysis of the Sanger Mouse Project database (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eThe genetic results of this analysis also elucidated some findings of pre-genomic era research, though they were not always consistent with them. For instance, the start codon loss from BALB/c \\u003cem\\u003eCtgf\\u003c/em\\u003e (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e) could play a role in explaining the strain\\u0026rsquo;s impaired \\u003cem\\u003eCtgf\\u003c/em\\u003e and collagen expression following BLM treatment (\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR66\\\" class=\\\"CitationRef\\\"\\u003e66\\u003c/span\\u003e). Comparably, the finding that BALB/c \\u003cem\\u003eBlmh\\u003c/em\\u003e contains many nonsense-mediated decay variants is apparently paradoxical with the finding \\u0026ndash; here replicated \\u0026ndash; that BALB/c expresses elevated levels of \\u003cem\\u003eBlmh\\u003c/em\\u003e transcript: a proposed basis for IPF-resistance in a specific BLM-dependent manner (\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eThe relation between previous findings and the novel \\u003cem\\u003eCtgf\\u003c/em\\u003e and \\u003cem\\u003eBlmh\\u003c/em\\u003e variants remains to be seen. Nevertheless, the result that BALB/c ECM deposition is similarly impaired in the mouse lung fibroblast (MLF) TGFβ assay as in BLM studies \\u0026ndash; including within the present dataset (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eA) \\u0026ndash; indicates that BLM-sensitivity is unlikely to be the root cause of variation in fibrosis risk, at least as far as the pulmonary pathology of BALB/c is concerned. This BLM-independent model of risk is orthogonally supported by the strain-differential enrichment of BALB/c for EMT processes (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB), which are characteristically a precursor to ECM accumulation and fibrogenesis (\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e). The finding that this enrichment is not here correlated with any BALB/c-specific enrichment in fibrogenesis further implies a resistance mechanism downstream of EMT \\u0026ndash; such as enhanced ECM degradation, e.g. by the upregulation of collagenases like \\u003cem\\u003eCtsk\\u003c/em\\u003e and \\u003cem\\u003eMmp13\\u003c/em\\u003e (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eC and Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eC).\\u003c/p\\u003e \\u003cp\\u003eFuture inquiries into the links between genomic variants and IPF phenotypes would indeed benefit from accounting for non-C57 mouse strains. For instance, the DBA/2 \\u003cem\\u003eNos3\\u003c/em\\u003e mutation identified (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e) could readily account for previously observed differences in fibrosis resolution between DBA/2 and C57BL/6, questioning prior assertions that this variation was linked to Th1- versus Th2-bias (\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e). This theory could have been assessed at the time by including BALB/c (or other Th2-biassed mice). Indeed, whilst the present study represents an initial step in extending transcriptomic fibrosis research to multiple genetic backgrounds, it lacks data from the many mouse strains other than BALB/c that are resistant to IPF, limiting inference.\\u003c/p\\u003e \\u003cp\\u003eThat said, the same P170L \\u003cem\\u003eFosl1\\u003c/em\\u003e variant identified in BALB/c was found, within this study, across other IPF-resistant strains within the Sanger Mouse Project (C3H, A/J). This supports the specific hypothesis that the P170L SNP itself may be responsible for altered FOSL1 activity in resistant strains, which is itself implicated in BALB/c IPF resistance by the transcriptomics analysis discussed. Thus, unlike the BLM-specific and Th2-bias hypotheses regarding BALB/c IPF pathology, the transcriptomic and genetic evidence are consistent with each other regarding FOSL1 \\u0026ndash; and strengthen the hypothesis that it modulates IPF risk in mouse models.\\u003c/p\\u003e \\u003cp\\u003eFurther research on more diverse strains in the context of fibrosis resistance would be valuable in generalising these findings \\u0026ndash; as well as in assessing other research previously published using a single strain (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e). It would also be informative to analyse the comparably non-conservative P170T variant observed in human FOSL1, given the multiple threads of evidence linking FOSL1 to IPF in humans as well, including those identified within the present investigation (\\u003cem\\u003eSupplementary Fig.\\u0026nbsp;5\\u003c/em\\u003e) (\\u003cspan additionalcitationids=\\\"CR58\\\" citationid=\\\"CR57\\\" class=\\\"CitationRef\\\"\\u003e57\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR59\\\" class=\\\"CitationRef\\\"\\u003e59\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR65\\\" class=\\\"CitationRef\\\"\\u003e65\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eThat said, GWAS studies with a specific focus on fibrotic lung disease have not identified Fosl1 directly as a gene of interest \\u0026ndash; one factor in this may be the exclusion of FOSL1 variants from single nucleotide polymorphism (SNP) arrays (\\u003cspan citationid=\\\"CR67\\\" class=\\\"CitationRef\\\"\\u003e67\\u003c/span\\u003e). The mining of unbiased GWAS datasets \\u0026ndash; e.g. based on genome or exome sequencing \\u0026ndash; may thus be necessary for the further analysis of human FOSL1 variant activity. In the present study, such analysis has already identified a number of significant FOSL1 SNPs (\\u003cem\\u003eSupplementary Table\\u0026nbsp;1\\u003c/em\\u003e). A meta-analysis \\u0026ndash; though beyond the scope of this investigation \\u0026ndash; could increase the diversity of SNPs identified as well as the statistical power of testing for association. It should also be noted that the largely negative result regarding human IPF-associated FOSL1 variants does not imply that the murine P170L variant is not mechanistically relevant to human IPF. \\u003cem\\u003ein vitro\\u003c/em\\u003e studies or, ultimately, gene therapy trials would yield empirical data on human applicability.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec27\\\" class=\\\"Section3\\\"\\u003e \\u003ch2\\u003eRegulatory factors are promising targets for preclinical testing\\u003c/h2\\u003e \\u003cp\\u003eIt has been established in this paper that \\u003cem\\u003eFosl1\\u003c/em\\u003e is a candidate of particular interest for risk modulation in mouse models. Moreover, FOSL1 is already positively implicated in both the resolution of pulmonary fibrosis and induction of hepatic fibrosis (\\u003cspan citationid=\\\"CR65\\\" class=\\\"CitationRef\\\"\\u003e65\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR68\\\" class=\\\"CitationRef\\\"\\u003e68\\u003c/span\\u003e). This coincides precisely with the low and high risks of pulmonary and hepatic fibrosis, respectively, noted for BALB/c relative to C57BL/6 mice (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e). This study thus provides a promising starting point for \\u003cem\\u003ein vitro\\u003c/em\\u003e or \\u003cem\\u003ein vivo\\u003c/em\\u003e modulation of murine fibrosis risk through engineering FOSL1, for instance to possess a P170L or comparable colocalised SNP.\\u003c/p\\u003e \\u003cp\\u003eFOSL1 is also implicated in human fibrogenic mechanisms. Specifically, it has been shown to promote EMT, despite correlating with an overall anti-fibrotic effect in some tissues (\\u003cspan citationid=\\\"CR69\\\" class=\\\"CitationRef\\\"\\u003e69\\u003c/span\\u003e). Under a hypothesis of FOSL1 gain-of-function in BALB/c, this would be consistent with the enrichment of EMT-related but not fibrogenic processes observed in the dataset (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB). FOSL1 is also the foremost co-regulator of EP300, the central regulator identified for Dupuytren\\u0026rsquo;s disease, a human fibrotic disorder (\\u003cspan citationid=\\\"CR69\\\" class=\\\"CitationRef\\\"\\u003e69\\u003c/span\\u003e). Moreover, it is plausible that the SNPs here reported for murine Fosl1 (rs30851424 and rs31137232) could alter the equivalent Ep300-Fosl1 interaction, altering key fibrotic processes in the ways observed (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB-C; Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e). This hypothesis of fibrosis resistance through altered Fosl1-Ep300 co-regulation would benefit from structural analysis.\\u003c/p\\u003e \\u003cp\\u003eMore broadly, the Category 2 signatures of not only fibrosis-related DEGs but also entire functional network modules (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eB) favours a preclinical focus on pathway regulators, such as TFs, rather than on single downstream gene products like collagenases. Within the network paradigm, these regulators are responsible for \\u0026lsquo;switching\\u0026rsquo; modules relevant to disease, resulting \\u0026ndash; where regulatory function varies with genetic background \\u0026ndash; in strain-specific pathology. By this definition, regulators include non-transcriptional and intercellular factors; for example, nitric oxide (NO) and interleukins, both of which mediate immune signalling (\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003cem\\u003eNos2\\u003c/em\\u003e and \\u003cem\\u003eIl6\\u003c/em\\u003e are also highlighted here as regulators exhibiting Category 2 expression under fibrosis. Knockouts of the pro-fibrotic gene \\u003cem\\u003eNos2\\u003c/em\\u003e, in particular, might informatively alter the unusual fibrosis susceptibility of C57BL/6 mice (\\u003cem\\u003eFig.\\u0026nbsp;4Bi\\u003c/em\\u003e). Whilst \\u003cem\\u003eNos2\\u003c/em\\u003e is considered anti-fibrotic in the long-term, it is non-essential for fibrosis resolution, unlike \\u003cem\\u003eNos3\\u003c/em\\u003e (\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR53\\\" class=\\\"CitationRef\\\"\\u003e53\\u003c/span\\u003e). The results of this study also indicate the value in further analysis of \\u003cem\\u003eIl6\\u003c/em\\u003e, as its role in Th1/Th2 differentiation, known regulation by FOSL1, and centrality to fibrotic DEG interactions in BALB/c (\\u003cem\\u003eFig.\\u0026nbsp;4Bii\\u003c/em\\u003e) draws on ongoing research into immunotypes and IPF (\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e44\\u003c/span\\u003e).\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/div\\u003e\"},{\"header\":\"Conclusions\",\"content\":\"\\u003cp\\u003eFrom these data, a multifactorial view of pulmonary fibrosis in the mouse model emerges, focused on the diseased lung itself \\u0026ndash; with disease defined by the activation states of various transcriptional subnetworks. A subset of these fibrosis-relevant subnetworks also differ across model mouse strains, correlating with observed differences in disease risk \\u0026ndash; and highly enriched for genes sensitive to the transcription factor (TF) FOSL1.\\u003c/p\\u003e \\u003cp\\u003eCombined with the presence of a non-conservative P170L SNP in this TF across IPF-resistant strains, this supports a novel hypothesis that FOSL1 modulates IPF risk in mice. This FOSL1 hypothesis complements the established view of IPF: a genetically complex disease, shaped by the regulation or dysregulation of vital processes (\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e). Examination of FOSL1 binding partners (e.g. EP300) and determination of an empirical structure may thus be warranted, to identify the mechanism for FOSL1 modulation of IPF.\\u003c/p\\u003e \\u003cp\\u003eOverall, our data and analyses highlight the pivotal role of regulators; the TF FOSL1 in particular, but also other signalling proteins such as NOS2 and IL6. The power of this integrated investigation, using both whole lung transcriptomes and mouse strain genetics, lies precisely in the shortlisting of such preclinical targets: particularly, in the identification of existing therapeutic variants, the viability of which has already been tested \\u0026ndash; by evolution.\\u003c/p\\u003e\"},{\"header\":\"Abbreviations\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eIPF:\\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;idiopathic pulmonary fibrosis\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eBLM:\\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;\\u0026nbsp;bleomycin\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eDEG:\\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;differentially expressed gene\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eGWAS:\\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;\\u0026nbsp;genome-wide association study\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCTGF:\\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;connective tissue growth factor\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eTGF\\u0026beta;: \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;\\u0026nbsp;transforming growth factor beta\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eEMT:\\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;\\u0026nbsp;epithelial to mesenchymal transition\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eBLMH: \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;\\u0026nbsp;bleomycin hydrolase\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003ePCA:\\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;principal component analysis\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eECM:\\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;\\u0026nbsp;extracellular matrix\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eTF:\\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;\\u0026nbsp;transcription factor\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003ePFRG:\\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;\\u0026nbsp;pulmonary fibrosis-related gene\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eNO:\\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;\\u0026nbsp;nitric oxide\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eTh2: \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;T helper 2 cells\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eTh1: \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;T helper 1 cells\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eSNP: \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;\\u0026nbsp;single nucleotide polymorphism\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003ebZIP: \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp; \\u0026nbsp;basic-leucine zipper\\u003c/strong\\u003e\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003e\\u003cem\\u003eEthics approval\\u003c/em\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis study was conducted using data from the Blue Sky Collaboration between AstraZeneca and the MRC Laboratory of Molecular Biology. The use of animal tissue and data was approved by the AstraZeneca ethical committee in Gothenburg, Sweden (EA184-2024). The approved site number for \\u003cem\\u003ein vivo\\u003c/em\\u003e work was 31-5373/11. Consent to participate is not applicable.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cem\\u003eConsent for publication\\u003c/em\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eNot applicable.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cem\\u003eAvailability of data and materials\\u003c/em\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe data that support the findings of this study are available from AstraZeneca but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of AstraZeneca.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cem\\u003eCompeting interests\\u003c/em\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe authors declare no competing interests.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cem\\u003eFunding\\u003c/em\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis study was supported jointly by the MRC Laboratory of Molecular Biology and AstraZeneca, as part of the Blue Sky Collaboration programme.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cem\\u003eAuthors\\u0026rsquo; contributions\\u003c/em\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eTF conducted all other \\u003cem\\u003ein silico\\u0026nbsp;\\u003c/em\\u003eanalysis; they also wrote and revised this manuscript. IB conducted the \\u003cem\\u003ein vitro\\u0026nbsp;\\u003c/em\\u003ework and contributed to the writing and figures of this manuscript. RG conducted processing and RNA sequencing of the lung\\u003cem\\u003e\\u0026nbsp;\\u003c/em\\u003esamples, in collaboration with AstraZeneca (AZ). WZ conducted GWAS data analysis. MS scripted preliminary analysis and visualisation of transcriptomes. HT assisted with\\u003cem\\u003e\\u0026nbsp;\\u003c/em\\u003eprotein structure analysis and interpretation. LAM managed the collaboration and provided expert scientific oversight. AZ scientists conducted all \\u003cem\\u003ein vivo\\u0026nbsp;\\u003c/em\\u003eexperiments. JG conceived and supervised the project at the LMB.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cem\\u003eAcknowledgements\\u003c/em\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eAll figure panels were rendered in BioRender by Thea Fennell, under the use of a Premium license and therefore legal for publication. This includes the use of BioRender graphic assets.\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\u003cli\\u003e\\u003cspan\\u003eWalkin L, Herrick SE, Summers A, Brenchley PE, Hoff CM, Korstanje R, et al. The role of mouse strain differences in the susceptibility to fibrosis: a systematic review. Fibrogenesis Tissue Repair. 2013;6(1):18.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003evan Geffen C, Dei\\u0026szlig;ler A, Quante M, Renz H, Hartl D, Kolahian S. Regulatory Immune Cells in Idiopathic Pulmonary Fibrosis: Friends or Foes? Front Immunol. 2021;12:1161.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDeng Z, Fear MW, Suk Choi Y, Wood FM, Allahham A, Mutsaers SE, et al. The extracellular matrix and mechanotransduction in pulmonary fibrosis. Int J Biochem Cell Biol. 2020;126:105802.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMoore MW, Herzog EL. Regulatory T Cells in Idiopathic Pulmonary Fibrosis: Too Much of a Good Thing? Am J Pathol. 2016;186(8):1978.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003ePoelmans VVG, De Langhe J, Hillen E, Vanoirbeek A, Himmelreich J. Longitudinal micro-CT provides biomarkers of lung disease that can be used to assess the effect of therapy in preclinical mouse models, and reveal compensatory changes in lung volume. Dis Model Mech. 2016;9(1):91\\u0026ndash;8.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eCallahan SJ, Xia M, Murray S, Flaherty KR. Clinical characteristics in patients with asymmetric idiopathic pulmonary fibrosis. Respir Med. 2016;119:96\\u0026ndash;101.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSokai A, Tanizawa K, Handa T, Kubo T, Hashimoto S, Ikezoe K et al. Asymmetry in acute exacerbation of idiopathic pulmonary fibrosis. ERJ Open Res. 2017;3(2).\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBaarsma HA, K\\u0026ouml;nigshoff M. WNT-er is coming\\u0026rsquo;: WNT signalling in chronic lung diseases. Thorax. 2017;72(8):746\\u0026ndash;59.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDubaybo BA, Maxwell DL, Sikora CA, Carlson RW. Compensatory Growth in Fibrotic Lung Injury. Am J Med Sci. 1991;302(1):19\\u0026ndash;22.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWang Y, Hu S, Shen L, Liu S, Wan L, Yang S, et al. Dynamic Observation of Autophagy and Transcriptome Profiles in a Mouse Model of Bleomycin-Induced Pulmonary Fibrosis. Front Mol Biosci. 2021;8:685.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003e\\u0026Aring;hlgren J, Voikar V. Experiments done in Black-6 mice: what does it mean? Lab Anim 2019 486. 2019;48(6):171\\u0026ndash;80.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKolb M, Bonniaud P, Galt T, Sime PJ, Kelly MM, Margetts PJ, et al. Differences in the fibrogenic response after transfer of active transforming growth factor-β 1 gene to lungs of fibrosis-prone and fibrosis-resistant mouse strains. Am J Respir Cell Mol Biol. 2002;27(2):141\\u0026ndash;50.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDown JD. The nature and relevance of late lung pathology following localised irradiation of the thorax in mice and rats. Br J Cancer. 1986;53:330\\u0026ndash;2.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSong HK, Hwang DY. Use of C57BL/6N mice on the variety of immunological researches. Lab Anim Res. 2017;33(2):119\\u0026ndash;23.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChung MP, Monick MM, Hamzeh NY, Butler NS, Powers LS, Hunninghake GW. Role of repeated lung injury and genetic background in bleomycin-induced fibrosis. Am J Respir Cell Mol Biol. 2003;29(3):375\\u0026ndash;80.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eIzbicki G, Segel MJ, Christensen TG, Conner MW, Breuer R. Time course of bleomycin-induced lung fibrosis. Int J Exp Pathol. 2002;83(3):111\\u0026ndash;9.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKolb M, Margetts PJ, Sime PJ, Gauldie J. Proteoglycans decorin and biglycan differentially modulate TGF-β-mediated fibrotic responses in the lung. Am J Physiol-Lung Cell Mol Physiol. 2001;280(6):L1327\\u0026ndash;34.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSime PJ, Xing Z, Graham FL, Csaky KG, Gauldie J. Adenovector-mediated gene transfer of active transforming growth factor-beta1 induces prolonged severe fibrosis in rat lung. J Clin Invest. 1997;100(4):768\\u0026ndash;76.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWeigle S, Martin E, Voegtle A, Wahl B, Schuler M. Primary cell-based phenotypic assays to pharmacologically and genetically study fibrotic diseases in vitro. J Biol Methods. 2019;6(2):e115.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eFilderman AE, Lazo JS. Murine strain differences in pulmonary bleomycin metabolism. Biochem Pharmacol. 1991;42(1):195\\u0026ndash;8.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLasky J, Ortiz LA, Tonthat B, Hoyle GW, Corti M, Athas G et al. Connective tissue growth factor mRNA expression is upregulated in bleomycin-induced lung fibrosis. Am J Physiol - Lung Cell Mol Physiol. 1998;275(2 19 \\u0026ndash; 2).\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBonniaud P, Martin G, Margetts PJ, Ask K, Robertson J, Gauldie J et al. Connective Tissue Growth Factor Is Crucial to Inducing a Profibrotic Environment in Fibrosis-Resistant Balb/c Mouse Lungs. Httpsdoiorg101165rcmb2004-0158OC. 2012;31(5):510\\u0026ndash;6.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eToren D, Yanai H, Abu Taha R, Bunu G, Ursu E, Ziesche R, et al. Systems biology analysis of lung fibrosis-related genes in the bleomycin mouse model. Sci Rep 2021 111. 2021;11(1):1\\u0026ndash;13.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBlaauboer ME, Boeijen FR, Emson CL, Turner SM, Zandieh-Doulabi B, Hanemaaijer R, et al. Extracellular matrix proteins: A positive feedback loop in lung fibrosis? Matrix Biol. 2014;34:170\\u0026ndash;8.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eTodd JL, Vinisko R, Liu Y, Neely ML, Overton R, Flaherty KR, et al. Circulating matrix metalloproteinases and tissue metalloproteinase inhibitors in patients with idiopathic pulmonary fibrosis in the multicenter IPF-PRO Registry cohort. BMC Pulm Med. 2020;20(1):1\\u0026ndash;12.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eOnursal C, Dick E, Angelidis I, Schiller HB, Staab-Weijnitz CA. Collagen Biosynthesis, Processing, and Maturation in Lung Ageing. Front Med. 2021;8:495.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBeck JA, Lloyd S, Hafezparast M, Lennon-Pierce M, Eppig JT, Festing MFW, et al. Genealogies of mouse inbred strains. Nat Genet 2000 241. 2000;24(1):23\\u0026ndash;5.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAshburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Gene Ontology: tool for the unification of biology. Nat Genet. 2000;25(1):25\\u0026ndash;9.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eThe Gene Ontology Consortium, Aleksander SA, Balhoff J, Carbon S, Cherry JM, Drabkin HJ, et al. The Gene Ontology knowledgebase in 2023. Genetics. 2023;224(1):iyad031.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKanehisa M, Furumichi M, Sato Y, Kawashima M, Ishiguro-Watanabe M. KEGG for taxonomy-based analysis of pathways and genomes. Nucleic Acids Res. 2023;51(D1):D587\\u0026ndash;92.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eXie Z, Bailey A, Kuleshov MV, Clarke DJB, Evangelista JE, Jenkins SL, et al. Gene Set Knowledge Discovery with Enrichr. Curr Protoc. 2021;1(3):e90.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChen EY, Tan CM, Kou Y, Duan Q, Wang Z, Meirelles GV, et al. Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics. 2013;14:128.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKuleshov MV, Jones MR, Rouillard AD, Fernandez NF, Duan Q, Wang Z, et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 2016;44(W1):W90\\u0026ndash;97.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSzklarczyk D, Franceschini A, Wyder S, Forslund K, Heller D, Huerta-Cepas J, et al. STRING v10: protein-protein interaction networks, integrated over the tree of life. Nucleic Acids Res. 2015;43(Database issue):D447\\u0026ndash;452.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSzklarczyk D, Gable AL, Nastou KC, Lyon D, Kirsch R, Pyysalo S, et al. The STRING database in 2021: Customizable protein-protein networks, and functional characterization of user-uploaded gene/measurement sets. Nucleic Acids Res. 2021;49(D1):D605\\u0026ndash;12.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eRackham OJL, Firas J, Fang H, Oates ME, Holmes ML, Knaupp AS, et al. A predictive computational framework for direct reprogramming between human cell types. Nat Genet. 2016;48(3):331\\u0026ndash;5.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKeane TM, Goodstadt L, Danecek P, White MA, Wong K, Yalcin B, et al. Mouse genomic variation and its effect on phenotypes and gene regulation. Nature. 2011;477(7364):289\\u0026ndash;94.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMcLaren W, Gil L, Hunt SE, Riat HS, Ritchie GRS, Thormann A et al. The Ensembl Variant Effect Predictor. Genome Biol [Internet]. 2016 Jun 6 [cited 2022 Jun 30];17(1). \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e/pmc/articles/PMC4893825/\\u003c/span\\u003e\\u003cspan address=\\\"http:///pmc/articles/PMC4893825/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBaldarelli RM, Smith CL, Ringwald M, Richardson JE, Bult CJ, Mouse Genome Informatics Group. Mouse Genome Informatics: an integrated knowledgebase system for the laboratory mouse. Genetics. 2024;227(1):iyae031.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eShovlin CL, Morris-Rosendahl DJ, Copeland F, Soyza AD, Hogg C, Jenkins G, et al. S66 Delivering the 100,000 genomes project to establish the functional role of DNA sequence variants in respiratory rare diseases. Thorax. 2019;74(Suppl 2):A44\\u0026ndash;5.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDhindsa RS, Mattsson J, Nag A, Wang Q, Wain LV, Allen R, et al. Identification of a missense variant in SPDL1 associated with idiopathic pulmonary fibrosis. Commun Biol. 2021;4(1):1\\u0026ndash;8.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eJumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, et al. Highly accurate protein structure prediction with AlphaFold. Nature. 2021;596(7873):583\\u0026ndash;9.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMirdita M, Sch\\u0026uuml;tze K, Moriwaki Y, Heo L, Ovchinnikov S, Steinegger M. ColabFold: making protein folding accessible to all. Nat Methods. 2022;19(6):679\\u0026ndash;82.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDiehl S, Rinc\\u0026oacute;n M. The two faces of IL-6 on Th1/Th2 differentiation. Mol Immunol. 2002;39(9):531\\u0026ndash;6.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eCabrera S, Maciel M, Hern\\u0026aacute;ndez-Barrientos D, Calyeca J, Gaxiola M, Selman M, et al. Delayed resolution of bleomycin-induced pulmonary fibrosis in absence of MMP13 (collagenase 3). Am J Physiol - Lung Cell Mol Physiol. 2019;316(5):L961\\u0026ndash;76.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKarman J, Wang J, Bodea C, Cao S, Levesque MC. Lung gene expression and single cell analyses reveal two subsets of idiopathic pulmonary fibrosis (IPF) patients associated with different pathogenic mechanisms. PLoS ONE. 2021;16(3 March):e0248889.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHernandez DM, Kang JH, Choudhury M, Andrianifahanana M, Yin X, Limper AH, et al. IPF pathogenesis is dependent upon TGFβ induction of IGF-1. FASEB J. 2020;34(4):5363\\u0026ndash;88.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eNkyimbeng T, Ruppert C, Shiomi T, Dahal B, Lang G, Seeger W, et al. Pivotal Role of Matrix Metalloproteinase 13 in Extracellular Matrix Turnover in Idiopathic Pulmonary Fibrosis. PLoS ONE. 2013;8(9):e73279.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eRajasekaran S, Reddy NM, Zhang W, Reddy SP. Expression profiling of genes regulated by Fra-1/AP-1 transcription factor during bleomycin-induced pulmonary fibrosis. BMC Genomics. 2013;14(1):1\\u0026ndash;13.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003ePang M, Rodr\\u0026iacute;guez-Gonzalez M, Hernandez M, Recinos CC, Seldeen KL, Troen BR. AP-1 and Mitf interact with NFATc1 to stimulate cathepsin K promoter activity in osteoclast precursors. J Cell Biochem. 2019;120(8):12382\\u0026ndash;92.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eFleischmann A, Hafezi F, Elliott C, Rem\\u0026eacute; CE, R\\u0026uuml;ther U, Wagner EF. Fra-1 replaces c-Fos-dependent functions in mice. Genes Dev. 2000;14(21):2695.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHan H, Cho JW, Lee S, Yun A, Kim H, Bae D, et al. TRRUST v2: an expanded reference database of human and mouse transcriptional regulatory interactions. Nucleic Acids Res. 2018;46(D1):D380\\u0026ndash;6.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eGuo C, Atochina-Vasserman E, Abramova H, George B, Manoj V, Scott P, et al. Role of NOS2 in pulmonary injury and repair in response to bleomycin. Free Radic Biol Med. 2016;91:293.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eIshida Y, Kimura A, Nosaka M, Kuninaka Y, Hemmi H, Sasaki I et al. Essential involvement of the CX3CL1-CX3CR1 axis in bleomycin-induced pulmonary fibrosis via regulation of fibrocyte and M2 macrophage migration. Sci Rep [Internet]. 2017 Dec 1 [cited 2022 Feb 1];7(1). \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://pubmed.ncbi.nlm.nih.gov/29203799/\\u003c/span\\u003e\\u003cspan address=\\\"https://pubmed.ncbi.nlm.nih.gov/29203799/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChen SSL, Lee SF, Hao HJ, Chuang CK. Mutations in the Leucine Zipper-Like Heptad Repeat Sequence of Human Immunodeficiency Virus Type 1 gp41 Dominantly Interfere with Wild-Type Virus Infectivity. J Virol. 1998;72(6):4765.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMeyer MB, Benkusky NA, Pike JW. Selective Distal Enhancer Control of the Mmp13 Gene Identified through Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR) Genomic Deletions. J Biol Chem. 2015;290(17):11093.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eOlafsdottir TA, Theodors F, Bjarnadottir K, Bjornsdottir US, Agustsdottir AB, Stefansson OA, et al. Eighty-eight variants highlight the role of T cell regulation and airway remodeling in asthma pathogenesis. Nat Commun 2020 111. 2020;11(1):1\\u0026ndash;11.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eFingerlin TE, Murphy E, Zhang W, Peljto AL, Brown KK, Steele MP, et al. Genome-wide association study identifies multiple susceptibility loci for pulmonary fibrosis. Nat Genet 2013 456. 2013;45(6):613\\u0026ndash;20.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSong S, Byrd JC, Mazurek N, Liu K, Koo JS, Bresalier RS. Galectin-3 Modulates MUC2 Mucin Expression in Human Colon Cancer Cells at the Level of Transcription via AP-1 Activation. Gastroenterology. 2005;129(5):1581\\u0026ndash;91.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003ePerkins TN, Oury TD. The perplexing role of RAGE in pulmonary fibrosis: causality or casualty? https://\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003edoi.org/101177/17534666211016071\\u003c/span\\u003e\\u003cspan address=\\\"http://doi.org/101177/17534666211016071\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e. 2021;15.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eNakashima T, Liu T, Yu H, Ding L, Ullenbruch M, Hu B, et al. Lung bone marrow-derived hematopoietic progenitor cells enhance pulmonary fibrosis. Am J Respir Crit Care Med. 2013;188(8):976\\u0026ndash;84.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBendstrup E, M\\u0026oslash;ller J, Kronborg-White S, Prior TS, Hyldgaard C. Interstitial Lung Disease in Rheumatoid Arthritis Remains a Challenge for Clinicians. J Clin Med. 2019;8(12).\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eB\\u0026uuml;hling F, R\\u0026ouml;cken C, Brasch F, Hartig R, Yasuda Y, Saftig P, et al. Pivotal Role of Cathepsin K in Lung Fibrosis. Am J Pathol. 2004;164(6):2203\\u0026ndash;16.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKubler A, Larsson C, Luna B, Andrade BB, Amaral EP, Urbanowski M, et al. Cathepsin K Contributes to Cavitation and Collagen Turnover in Pulmonary Tuberculosis. J Infect Dis. 2016;213(4):618.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eRajasekaran S, Vaz M, Reddy SP. Fra-1/AP-1 Transcription Factor Negatively Regulates Pulmonary Fibrosis In Vivo. PLoS ONE. 2012;7(7):41611.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSternlicht MD, Wirkner U, Bickelhaupt S, Lopez Perez R, Tietz A, Lipson KE, et al. Radiation-induced pulmonary gene expression changes are attenuated by the CTGF antibody Pamrevlumab. Respir Res. 2018;19(1):1\\u0026ndash;16.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAllen RJ, Oldham JM, Jenkins DA, Leavy OC, Guillen-Guio B, Melbourne CA, et al. Longitudinal lung function and gas transfer in individuals with idiopathic pulmonary fibrosis: a genome-wide association study. Lancet Respir Med. 2022;11(1):65\\u0026ndash;73.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKireva T, Erhardt A, Tiegs G, Tilg H, Denk H, Haybaeck J et al. Transcription Factor Fra-1 Induces Cholangitis and Liver Fibrosis. 2011.\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWilliams LM, McCann FE, Cabrita MA, Layton T, Cribbs A, Knezevic B, et al. Identifying collagen VI as a target of fibrotic diseases regulated by CREBBP/EP300. Proc Natl Acad Sci U S A. 2020;117(34):20753\\u0026ndash;63.\\u003c/span\\u003e\\u003c/li\\u003e\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":false,\"hideJournal\":true,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true},\"keywords\":\"idiopathic pulmonary fibrosis, C57BL/6, BALB/c, FOSL1, bleomycin, disease resistance, mouse model, genetics, transcriptomics, gene regulation\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-5295459/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-5295459/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003ch2\\u003eBackground\\u003c/h2\\u003e \\u003cp\\u003eIdiopathic pulmonary fibrosis (IPF) is a terminal inflammatory lung disease that causes permanent scarring (fibrogenesis). Bleomycin (BLM) is a drug used to induce fibrosis in mouse models, typically C57BL/6. However, meta-analyses show inter-strain heterogeneity in response, e.g. resistance in BALB/c. This study extends transcriptomic analysis of IPF to a resistant strain, qualifying inferences from the standard model and suggesting genetic risk factors to inform clinical research.\\u003c/p\\u003e\\u003ch2\\u003eMethods\\u003c/h2\\u003e \\u003cp\\u003eTranscriptomic datasets were generated from C57BL/6 and BALB/c mice. Test mice were administered BLM \\u0026ndash; with tissue samples sequenced from control, test, and contralateral lungs at the fibrogenesis stage of the BLM model (7\\u0026ndash;14 days after injection). Differentially expressed genes (DEGs) were calculated between treatments and strains, followed by gene network and transcription factor (TF) target enrichment analysis of DEGs. Additionally, strain-specific genetic variants were identified in fibrosis-related genes, complemented by analysing human genome-wide association (GWAS) datasets. An \\u003cem\\u003ein vitro\\u003c/em\\u003e model of TGF\\u0026#120573;-stimulated stress fibre deposition was used in parallel to confirm transcriptomic findings.\\u003c/p\\u003e\\u003ch2\\u003eResults\\u003c/h2\\u003e \\u003cp\\u003eDEGs calculated between treatment groups were enriched for general fibrosis-related processes across strains. Some fibrogenic processes and functional modules, however, were specifically enriched in C57BL/6, which was orthogonally validated by \\u003cem\\u003ein vitro\\u003c/em\\u003e TGFβ assays. Conversely, the anti-fibrotic DEG \\u003cem\\u003eCtsk\\u003c/em\\u003e was upregulated under fibrosis in BALB/c specifically. TF target enrichment analysis of cross-strain and cross-treatment DEGs, using perturbation data, further identified them as significantly overrepresentative of FOSL1-sensitive genes. Subsequent genetic analysis revealed a non-conservative variant (P170L) located in BALB/c FOSL1. Furthermore, analysis of data from the 100,000 Genomes Project associated human FOSL1 variants with IPF.\\u003c/p\\u003e\\u003ch2\\u003eConclusions\\u003c/h2\\u003e \\u003cp\\u003eTranscriptional differences in IPF have been characterised for C57BL/6 and BALB/c strains, supporting the consensus on IPF resistance in BALB/c. Analysis of gene set expression within and between strains principally implicates genes sensitive to the TF FOSL1. The significance of this novel finding is amplified by the discovery of a highly non-conservative P170L mutation in the bZIP domain of BALB/c FOSL1. Mechanistic investigation of FOSL1 activity \\u0026ndash; and potentially other regulators, e.g. \\u003cem\\u003eNos2, Il6\\u003c/em\\u003e \\u0026ndash; is thus recommended as preclinical IPF research.\\u003c/p\\u003e\",\"manuscriptTitle\":\"A Tale of Two Mice: genetics of model mouse strains suggest a transcriptional basis for risk and resistance in idiopathic pulmonary fibrosis\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2024-11-11 14:48:30\",\"doi\":\"10.21203/rs.3.rs-5295459/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"bc0aebd4-5456-4e1b-966b-e85c75c212f3\",\"owner\":[],\"postedDate\":\"November 11th, 2024\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"posted\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2024-12-04T17:38:53+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2024-11-11 14:48:30\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-5295459\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-5295459\",\"identity\":\"rs-5295459\",\"version\":[\"v1\"]},\"buildId\":\"qtupq5eGEP_6zYnWcrvyt\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}