S100A4-neutralizing monoclonal antibody 6B12 counteracts the established experimental skin fibrosis induced by bleomycin

Abstract

Objectives

Our previous studies have demonstrated that the Damage Associated Molecular Pattern (DAMP) protein, S100A4, is overexpressed in the involved skin and peripheral blood of patients with SSc. It is associated with skin and lung involvement, and disease activity. By contrast, lack of S100A4 prevented the development of experimental dermal fibrosis. Herein we aimed to evaluate the effect of murine anti-S100A4 mAb 6B12 in the treatment of preestablished experimental dermal fibrosis.

Methods

The effects of 6B12 were assessed at therapeutic dosages in a modified bleomycin-induced dermal fibrosis mouse model by evaluating fibrotic (dermal thickness, proliferation of myofibroblasts, hydroxyproline content, phosphorylated Smad3-positive cell count) and inflammatory (leukocytes infiltrating the lesional skin, systemic levels of selected cytokines and chemokines) outcomes, and transcriptional profiling (RNA sequencing).

Results

Treatment with 7.5 mg/kg 6B12 attenuated and might even reduce pre-existing dermal fibrosis induced by bleomycin as evidenced by reduction in dermal thickness, myofibroblast count and collagen content. These antifibrotic effects were mediated by the downregulation of TGF-β/Smad signalling and partially by reducing the number of leukocytes infiltrating the lesional skin and decrease in the systemic levels of IL-1α, eotaxin, CCL2 and CCL5. Moreover, transcriptional profiling demonstrated that 7.5 mg/kg 6B12 also modulated several profibrotic and proinflammatory processes relevant to the pathogenesis of SSc.

Conclusion

Targeting S100A4 by the 6B12 mAb demonstrated potent antifibrotic and anti-inflammatory effects on bleomycin-induced dermal fibrosis and provided further evidence for the vital role of S100A4 in the pathophysiology of SSc.

Rheumatology key messages.

• S100A4 is a Damage Associated Molecular Pattern protein that mediates the profibrotic effects of TGF-β.

• Anti-S100A4 mAb (7.5 mg/kg 6B12) prevents progression and induces regression of bleomycin-induced dermal fibrosis.

• These effects are mediated by targeting several profibrotic and proinflammatory pathways relevant for SSc.

Introduction

SSc, also known as scleroderma, is a chronic, immune-mediated, fibrotic disease characterized by dysregulated connective tissue remodelling [1, 2]. Its defining characteristics include fibrosis of the skin and internal organs, dysregulation of the immune system and vasculopathy [1, 3]. Unfortunately, although SSc has the highest mortality of all autoimmune-mediated CTDs, current treatment only tackles organ-specific manifestations due to the lack of a mechanistic understanding of fibrogenesis [3, 4].

Tremendous effort and progress have been made in deciphering the signalling pathways involved in tissue fibrosis; however, the underlying cause for the persistently dysregulated fibrogenesis in SSc remains elusive [5, 6]. One of the explanations that has been garnering increasing attention involves the Damage Associated Molecular Pattern (DAMP) proteins, which potentiate innate immune responses and upregulate cytokines/chemokines associated with SSc [7–9]. Most notably, S100A4 is a DAMP protein implicated in the pathogenesis of SSc, and extracellular S100A4 has been particularly linked to various fibrotic diseases, such as pulmonary, cardiac, renal and liver fibrosis [10]. Further evidence supports the significance of S100A4 in activating proinflammatory responses by binding to pattern recognition receptors, such as toll-like receptor (TLR)-4 and advanced glycosylation end product-specific receptor (RAGE) expressed on epithelial cells, innate and adaptive immune cells, and fibroblasts [11–15]. Therefore, investigation into S100A4’s role in various preclinical models of SSc could help identify bottlenecks and guide potential targeted therapy.

Our previous research demonstrated that S100A4 protein is upregulated in the skin of SSc patients via the TGF-β/small mothers against decapentaplegic homologue (Smad) signalling. We demonstrated a positive feedback loop where S100A4 upregulated TGF-β/Smad signalling, which in turn increased S100A4 levels in dermal fibroblasts. Moreover, S100A4 activated dermal fibroblasts and induced collagen production, whereas its deficit blunted the profibrotic effects of TGF-β. Similarly, S100A4 knockout mice were protected from developing skin fibrosis in both inflammatory and non-inflammatory preclinical models of SSc [16]. Our group has additionally examined systemic S100A4 levels in SSc patients compared with age- and sex-matched healthy controls. We observed increased S100A4 levels, mainly in the dcSSc, as well as in SSc patients with interstitial lung disease (ILD). In particular, elevated systemic S100A4 levels were significantly associated with higher disease activity, more extensive skin involvement and more impaired lung functions [17].

In continuation of our previous work to establish S100A4 as a potential novel molecular therapeutic target for SSc, we aimed to assess the efficacy of anti-S100A4 murine mAb 6B12 in the treatment of preestablished experimental dermal fibrosis induced by bleomycin.

Methods

Anti-S100A4 mAb

In this study, we utilized 6B12, mouse IgG1 mAb which selectively binds to S100A4 with high affinity [18], provided by Arxx Therapeutics (Oslo, Norway).

Treatment of established dermal fibrosis induced by bleomycin

To assess the effect of 6B12 on attenuating fibrosis, we chose the murine model of bleomycin-induced dermal fibrosis in a therapeutic design as previously described [19, 20]. Briefly, to induce robust dermal fibrosis, bleomycin (Bleomedac, Medac GmbH, Wedel, Germany) was administered s.c. to 6-week-old male C57BL/6 mice (Velaz, s.r.o., Prague, Czech Republic) every other day for 6 weeks at a concentration of 0.5 mg/ml. Treatment with 6B12 was commenced after 3 weeks of bleomycin administration at three different doses: 2.5, 7.5 and 12.5 mg/kg intra-peritoneally every third day (n = 8 per group), and continued for the following 3 weeks concomitantly with bleomycin challenge. Control groups were challenged either with 0.9% NaCl s.c. injections every other day for 6 weeks (n = 8; representing no dermal fibrosis), with bleomycin for 3 weeks followed by 0.9% NaCl for 3 weeks (n = 8; representing the pre-treatment level of skin fibrosis), or with bleomycin for 6 weeks (n = 8; representing prolonged fibrosis) (Fig. 1).

Figure 1.

Therapeutic design of 6B12 treatment in bleomycin-induced dermal fibrosis. i.p.: intraperitoneally administered; s.c.: subcutaneously administered

Primary outcomes were analysed 6 weeks after the first bleomycin injection and included the assessment of dermal thickness, myofibroblast count, hydroxyproline content, the number of phosphorylated (p)Smad3-positive cells and infiltrating leukocytes, systemic levels of selected inflammatory cytokines/chemokines, and of gene expression by RNA sequencing.

Ethics statement

The study was conducted according to the guidelines of the Declaration of Helsinki, and was approved by the Ethics Committee of the Institute of Rheumatology in Prague (reference number 5689/2015, approved on 6 June 2015) and the Ministry of Education, Youth and Sports of the Czech Republic (reference number MSMT‐9445/2018‐7, approved on 5 May 2018). Animal experiments were conducted in accordance with relevant Czech legislation (Act No. 246/1992 Coll. and Decree No. 419/2012 Coll.) on the use of animals for research and complied with the commonly accepted 3Rs.

Histological staining

Assessments of dermal thickness and of the number of infiltrating leukocytes were performed on haematoxylin–eosin stained skin sections as previously described [20, 21]. Infiltrating leukocytes were also identified by immunohistochemistry staining using rabbit anti-CD45 mAb (1:300, overnight at 4°C, Abcam, Berlin, Germany), co-stained with haematoxylin. The myofibroblast count was evaluated by immunohistochemistry staining and identification of alpha-smooth muscle actin (aSMA)-positive fibroblasts using mouse anti-aSMA mAb (1:1000, clone 1A4, Sigma‐Aldrich, St Louis, MO, USA), co-stained with haematoxylin as previously specified [20]. Further details on CD45 staining and quantification of myofibroblasts, inflammatory infiltrate and CD45-positive cells are available in Supplementary Material, available at Rheumatology online. Detection of pSmad3-positive cells was performed by immunofluorescence staining using rabbit anti-pSmad3 mAb (1:50, phospho S423 + S425, Abcam, Cambridge, UK) and co-staining of nuclei with 4,6-diamidino-2-phenylindole (DAPI, 1:1000, Invitrogen, Carlsbad, CA, USA) as described elsewhere [20]. The staining was visualized at 100- and 400-fold magnification using appropriate fluorescence filters on a BX53 microscope with a DP80 Colour lens camera using CellSens Standard software v.3.1‐Build 21199 (Olympus, Philadelphia, PA, USA) and assessed in a blinded manner, as previously depicted [20].

Hydroxyproline assay

Collagen content in the lesional skin was determined by hydroxyproline assay, as elaborated elsewhere [20, 22] and briefly described in Supplementary Material, available at Rheumatology online.

Serum levels of inflammatory cytokines and chemokines

Systemic concentrations of selected inflammatory cytokines and chemokines were measured by Bio-Plex Pro™ Mouse Cytokine 23-plex Assay (Bio-Rad, Irvine, CA, USA) according to instructions and evaluated by Luminex BIO-PLEX 200 System (Bio-Rad, Hercules, CA, USA) as previously described [20, 23].

Isolation of RNA, preprocessing and bioinformatics analysis

Isolation of RNA from lesional skin samples of mice from groups 1, 3 and 5 (Fig. 1) and the generation of Cell intensity (CEL) files are described in Supplementary Material, available at Rheumatology online.

Raw data (CEL files) were imported and preprocessed in RStudio (R version 4.0.2) [24]. Principal component analysis was performed to determine potential outliers among samples in each batch. Data were then normalized using Robust Multi-array Average (RMA) algorithm. To identify differentially expressed genes (DEGs), Linear Models for Microarray data (LIMMA) package in R (version 3.44.3) [25] was used to compute both nominal and adjusted P-value, using the Benjamini–Hochberg method. Herein, the experimental batch was considered as a factor along with sample phenotype. Results were visualized as volcano blots with the ‘EnhancedVolcano’ package (version 1.6.0) according to the Log2FC and –Log10adjusted P-value of DEGs. Hierarchical cluster analyses were performed using the ‘pheatmap’ package (version 1.0.12). Biological Process Gene Ontology (GO-BP) pathway enrichment analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Set Enrichment Analysis (GSEA) were performed using the ‘clusterProfiler’ package (version 3.16.1) [26]. Results of GO-BP, KEGG and GSEA were presented as bar plots using the ‘ggplot2’ package (version 3.3.2) with the threshold for adjusted P-value of 0.05. The ‘affy’ package (version 1.67.0) and ‘oligo’ package (version 1.52.1) were used for reading and preprocessing the CEL files. All possible relations among different DEGs lists were represented by Venn Diagram obtained from the BioVenn website. The transcriptomic profile of SSc patients was retrieved from a publicly available dataset of North American SSc patient cohort (NCBI/GEO/GSE59787) consisting of 143 SSc patients and 22 healthy individuals [27]. The six significant lists of DEGs are determined as adjusted P-value ≤0.05 and |Log2FC|≥0.5 for comparing between groups 1 and 3, as well as groups 3 and 5 (Fig. 1).

Statistical analysis

Data were analysed and plotted using GraphPad Prism 5 (v.5.02, GraphPad Software, La Jolla, CA, USA), and presented as mean ± standard error of the mean. The inter-group differences were analysed by one-way analysis of variance with Dunnett’s multiple comparison test, and statistical significance was defined as P < 0.05.

Results

6B12 anti-S100A4 mAb attenuates and might reverse preestablished skin fibrosis induced by bleomycin

To assess the effect of the 6B12 anti-S100A4 mAb on the progression and treatment of preestablished dermal fibrosis, we employed the therapeutic design of bleomycin-induced skin fibrosis.

Treatment with 2.5 mg/kg 6B12 for the last 3 weeks of bleomycin administration following 3 weeks of pre-treatment with bleomycin resulted in a mild reduction of dermal fibrosis, represented by a modest decrease in dermal thickness by 21.8 ± 9.9% (P = 0.0461) and myofibroblast count by 83.3 ± 10.4% (P = 0.0011), whereas no change in collagen content (P > 0.05) was observed compared with the 6-week bleomycin challenge.

Treatment with 7.5 mg/kg 6B12 under the same conditions resulted in a significant reduction of all three outcomes compared with the 6-week bleomycin challenge, i.e. a decrease in dermal thickness by 55.0 ± 8.4% (P < 0.0001), myofibroblast count by 158.5 ± 21.0% (P < 0.0001) and collagen content by 42.9 ± 16.8% (P = 0.0230). Remarkably, compared with the pre-treatment levels of fibrosis represented by the mice challenged with bleomycin for the first 3 weeks followed by NaCl injections for the last 3 weeks, treatment with 7.5 mg/kg 6B12 might also induce regression of preestablished fibrosis with reduced dermal thickening by 22.6 ± 3.7% (P < 0.0001), and myofibroblast count by 37.5 ± 15.5% (P = 0.0301); however, the decrease in collagen content by 24.6 ± 13.6% (P = 0.0919) did not attain statistical significance.

Unexpectedly, treatment with 12.5 mg/kg 6B12 under the same conditions did not render any changes in dermal thickness, myofibroblast count or collagen content compared with the 6-week bleomycin challenge (P > 0.05 for all) (Fig. 2A–D).

Figure 2.

Treatment with 7.5 mg/kg 6B12 anti-S100A4 mAb prevents further progression and may induce regression of bleomycin-induced dermal fibrosis. (A) Representative hematoxylin–eosin-stained skin sections. Vertical bars represent dermal thickness, with a horizontal scale bar of 100μm. Treatment with 7.5 mg/kg 6B12 anti-S100A4 mAb prevents further progression and may induce regression of dermal thickening (B), proliferation of myofibroblasts (C), collagen content (D) and activation of TGF-β/Smad signaling (E) induced by bleomycin. Boxes represent the mean, and whiskers represent the standard error of the mean. n = 8 mice in each group; w, week; NaCl, sodium chloride; BLM, bleomycin; 6B12, murine anti-S100A4 mAb; ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001

Given the suppression of TGF-β/pSmad3 signalling in S100A4 knock-out mice challenged with bleomycin [16], we aimed to assess the effect of 6B12 on the activation of the intracellular TGF-β canonical pathway mediated by pSmad3. Treatment with 7.5 mg/kg 6B12 robustly downregulated TGF-β/pSmad3 signalling, demonstrated by a decrease in pSmad3-positive cell percentage by 108.9 ± 24.1% (P = 0.0005) compared with the 6-week bleomycin challenge and by 26.0 ± 11.5% (P = 0.0399) compared with pre-treatment levels (Fig. 2E).

6B12 anti-S100A4 mAb reduces local and systemic inflammation in bleomycin-induced dermal fibrosis

Given the significance of S100A4 in chemotaxis, mobilizing macrophages and neutrophils, activation of proinflammatory intracellular signalling, and upregulation of proinflammatory cytokines/chemokines [11–15, 28], plus the partial resemblance of the bleomycin-induced dermal fibrosis with the early, inflammatory stages of SSc characterized by perivascular inflammatory infiltrates stimulating the cellular and soluble mediators of fibrogenesis [2, 8, 19, 29], we sought to explore whether the antifibrotic effects of 7.5 mg/kg 6B12 are partially mediated by regulation of local and systemic inflammatory response.

Treatment with 7.5 mg/kg 6B12 significantly decreased infiltrating leukocytes by 81.6 ± 17.5% (P = 0.0004) compared with the 6-week bleomycin challenge, and by 39.5 ± 11.8% (P = 0.0049) compared with pre-treatment levels as assessed by haematoxylin–eosin staining (Fig. 3A, Supplementary Fig. S1A, available at Rheumatology online), and by 113.9 ± 27.5% (P = 0.0010) compared with the 6-week bleomycin challenge as evidenced by the immunohistochemistry staining for CD45, a pan-leucocyte marker (Supplementary Fig. S1B and C, available at Rheumatology online).

Figure 3.

Treatment with 7.5 mg/kg 6B12 anti-S100A4 mAb attenuates local and systemic inflammatory response induced by bleomycin. Treatment with 7.5 mg/kg 6B12 anti-S100A4 mAb reduces the number of leukocytes infiltrating the lesional skin (A), as well as the serum levels of IL-1α (B), eotaxin (C), monocyte chemoattractant protein-1 (MCP-1, CCL2) (D), and regulated on activation/normal T cell expressed and secreted (RANTES, CCL5) (E) induced by bleomycin. Boxes represent the mean, and whiskers represent the standard error of the mean. n = 8 mice in each group; w, week; NaCl, sodium chloride; BLM, bleomycin; 6B12, murine anti-S100A4 mAb; ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001

In addition to the suppression of local inflammatory infiltration, 7.5 mg/kg 6B12 also decreased systemic levels of IL-1α [by 7.1 ± 2.0 pg/ml (P = 0.0037)], eotaxin [by 548.6 ± 264.4 pg/ml (P = 0.0207)], monocyte chemoattractant protein (MCP)-1 (CCL2) [by 115.9 ± 20.6 pg/ml (P < 0.0001)], and regulated on activation/normal T cell expressed and secreted (RANTES, CCL5) [by 110.7 ± 35.5 pg/ml (P = 0.0076)] compared with the 6-week bleomycin challenge (Fig. 3B–E). Further details on the remaining cytokines/chemokines measured by the 23-plex assay are available in Supplementary Table S1 (available at Rheumatology online).

6B12 anti-S100A4 mAb modulates inflammatory and fibrotic pathways in skin fibrosis induced by bleomycin

In order to glean insights into molecular mechanisms underlying the antifibrotic effects of 7.5 mg/kg 6B12, we performed transcriptional profiling of murine skin injected with NaCl or bleomycin for 6 weeks, and mice treated with 7.5 mg/kg 6B12 for the last 3 weeks of a 6-week bleomycin challenge.

We first compared datasets on skin injected with 0.9% NaCl for 6 weeks and skin challenged with bleomycin for 6 weeks (BLM-DEGs). Using cut-offs of adjusted P ≤ 0.05 and |Log2FC| ≥0.5, we observed 462 DEGs, 377 of which were upregulated and 85 of which were downregulated (Fig. 4A and B). Comparison of the BLM-DEGs with DEGs in the skin of SSc patients (NCBI/GEO/GSE59787) [27] demonstrated an overlap of 65 genes, comprising 14% of all BLM-DEGs (Fig. 4C). Further evaluation using GO-BP, KEGG and GSEA demonstrated dysregulation of cellular processes and pathways related to inflammation and fibrosis in bleomycin-challenged murine skin (Fig. 4D–G).

Figure 4.

Inflammation-, immune- and fibrosis-related transcriptional pathways are upregulated in the skin challenged with bleomycin. Volcano plot (A) and hierarchical heatmap (B) of differentially expressed genes (DEGs) with the top 10 upregulated or downregulated genes (highlighted in green) in the lesional skin challenged with bleomycin for 6 weeks compared with NaCl-treated controls for 6 weeks. (C) Venn diagram illustrating the overlap between BLM-DEGs (bleomycin vs NaCl) and SSc-DEGs (scleroderma skin vs healthy skin from the NCBI/GEO/GSE59787 cohort). Functional overrepresentation analysis of BLM-DEGs according to Gene Ontology Biological Process (GO-BP) database (D), Kyoto Encyclopedia of Genes and Genomes (KEGG) (E) and Gene Set Enrichment Analysis (GSEA) (F). (G) GSEA for ‘inflammatory response’. NES, nominal enrichment score

We next compared datasets on the skin challenged with bleomycin for 6 weeks and mice treated with 7.5 mg/kg 6B12 for the last 3 weeks of the 6-week bleomycin challenge (anti-S100A4-DEGs). Using cut-offs of P ≤ 0.05 and |Log2FC| ≥0.2, we observed 684 DEGs, 301 of which were upregulated and 383 of which were downregulated (Fig. 5A and B). Comparison of anti-S100A4-DEGs with BLM-DEGs demonstrated 74 shared genes (Fig. 5C), whereas comparison of anti-S100A4-DEGs with BLM-DEGs and SSc-DEGs revealed nine shared genes, namely ISG15, HAVCR2, SLC15A3, CCL2, PLEK, CCL12, S100A9, OAS3 and VAV1 (Fig. 5D). Further analyses (GO-BP, KEGG and GSEA) on anti-S100A4-DEGs provided additional information. Namely, GO-BP demonstrated statistically significant dysregulation of numerous processes related to inflammation and fibroblast activation upon 7.5 mg/kg 6B12 treatment. These included particularly processes related to T cell and monocyte/macrophage activation, and fibrosis-related processes such as regulation of wound response (Fig. 5E). KEGG and GSEA analyses of the anti-S100A4-DEGs provided further insights, especially in the potential mechanisms of anti-S100A4 mAb, and yielded links to IL-6-JAK-STAT3 signalling and reactive oxygen species signalling (Figures 5F-H).

Figure 5.

Treatment with 7.5 mg/kg 6B12 anti-S100A4 mAb downregulates inflammation- and fibrosis-relevant genes in systemic sclerosis. Volcano plot (A) and hierarchical heatmap (B) of differentially expressed genes (DEGs) with the top 10 upregulated or downregulated genes (highlighted in green) in the lesional skin of mice challenged with bleomycin for 6 weeks compared with the 6-week bleomycin challenge with the concurrent treatment with anti-S100A4 mAb (7.5 mg/kg 6B12) during the last 3 weeks. Venn diagram illustrating the overlap between BLM-DEGs (bleomycin vs NaCl for 6 weeks) and S100A4-DEGs (bleomycin for 6 weeks treated with 7.5 mg/kg 6B12 for the last 3 weeks vs bleomycin for 6 weeks) (C) and with SSc-DEGs (scleroderma skin vs healthy skin from the NCBI/GEO/GSE59787 cohort) (D). Functional overrepresentation analysis of S100A4-DEGs according to Gene Ontology Biological Process (GO-BP) database (E), Kyoto Encyclopedia of Genes and Genomes (KEGG) (F) and Gene Set Enrichment Analysis (GSEA) (G). (H) GSEA for ‘reactive oxygen species pathway’, ‘IL-6-JAK-STAT3 signaling’ and ‘inflammatory response’. NES, nominal enrichment score; JAK, Janus kinase; STAT3, signal transducer and activator of transcription 3

In addition, comparing datasets on mice treated with 7.5 mg/kg 6B12 for the last 3 weeks of the 6-week bleomycin challenge with mice treated with NaCl for 6 weeks, we observed 399 DEGs, 310 of which were upregulated and 89 of which were downregulated. GO-BP demonstrated that several crucial processes (e.g. lymphocyte chemotaxis and migration, T cell chemotaxis and activation, mesenchymal cell differentiation, stress fiber assembly, and regulation of Wnt signalling), which were upregulated in BLM-DEGs, were reversed to normal levels upon treatment with 7.5 mg/kg 6B12, comparable to those of NaCl-treated mice (Supplementary Fig. S2, available at Rheumatology online).

The complete list of all DEGs is available in Supplementary Material (.xlsx file), available at Rheumatology online.

Discussion

We provide the first empirical evidence that 6B12, the murine S100A4-neutralizing mAb, effectively attenuates and might even reduce preestablished, bleomycin-induced dermal fibrosis. Anti-S100A4 mAb exerts antifibrotic effects via direct downregulation of TGF‐β/Smad signalling, indirect dampening of the local and systemic inflammation, and transcriptional changes implicated in the pathogenesis of SSc.

Multiple lines of experimental evidence support the crucial role of S100A4 in tumorigenesis as a well-established biomarker of poor prognosis and greater metastatic potential of breast, bladder, pancreatic and colorectal cancer [30–33]. Under physiological conditions, S100A4 is expressed intracellularly in several cell types, including fibroblasts, macrophages, lymphocytes, vascular cells and bone-marrow derived cells, and regulates angiogenesis, cell proliferation, migration, invasion, differentiation and epithelial/endothelial-mesenchymal transition (EpMT/EnMT). In response to cellular damage or stress, S100A4 is upregulated and released from the cells, subsequently inducing proinflammatory responses [12, 33].

In contrast to the well-established function of S100A4 in tumorigenesis, evidence of its role in tissue fibrosis remains scant [10, 33]. However, many of the above-mentioned cellular processes are also implicated in tissue fibrogenesis and angiogenesis, particularly cell proliferation, differentiation, activation of both TGF-β/Smad- and non-canonical pathways, enhanced EpMT/EnMT, and T cell polarization towards the Th2 phenotype [10, 16]. More recently, S100A4 has been implicated in pulmonary, cardiac, renal and liver fibrosis, as well as pulmonary arterial hypertension [12, 28].

To explore the relevance of these findings specifically for SSc, we previously investigated S100A4 in SSc patients and demonstrated an increased S100A4 expression in the SSc skin via the TGF-β/Smad pathway [16]. Moreover, we observed elevated S100A4 systemic levels in SSc patients compared with healthy controls, especially in dcSSc and SSc-ILD [17]. Furthermore, S100A4 was increased in the bronchoalveolar lavage fluid of idiopathic pulmonary fibrosis, predominantly from macrophages in the fibrotic lung tissue [34]. Conversely, serum S100A4 levels in active SSc-ILD decreased upon treatment with CYC, and baseline levels of serum S100A4 predicted the decline in markers of systemic inflammation upon CYC treatment [17].

Our current study expands on previous findings that S100A4 plays a crucial role in the pathogenesis of SSc, and aims to corroborate S100A4 as a therapeutic target for SSc. Therefore, we assessed the efficacy of S100A4 mAb for treating bleomycin-induced dermal fibrosis.

We observed significant antifibrotic effects in the 7.5 mg/kg group and mild effects in the 2.5 mg/kg group, as corroborated by decreased dermal thickness and myofibroblast count, reduced collagen content and dampened TGF-β/pSmad3 activation. Based on functional overrepresentation analysis of S100A4-DEGs from the GO-BP analysis, we demonstrated a statistically significant downregulation of numerous processes related to inflammation and fibroblast activation upon 7.5 mg/kg 6B12 treatment, especially processes related to fibrogenesis, wound healing, regulation of collagen synthesis, tissue remodelling, and the activation of integrins, T cells, monocytes and macrophages. Meanwhile, several overlapping genes, namely ISG15, HAVCR2, SLC15A3, CCL2, PLEK, CCL12, S100A9, OAS3 and VAV1, were also downregulated after 7.5 mg/kg 6B12 treatment in bleomycin-induced dermal fibrosis, further substantiating the significance of S100A4 in fibrogenesis. Notably, these antifibrotic findings are consistent with other preclinical studies using similar methods of S100A4 inhibition [10, 35].

Unexpectedly, there was an apparent loss of effect with the highest dose tested (12.5 mg/kg 6B12). There are several plausible explanations. One could be that there is a U-shaped response curve for the suppression of S100A4. This is very unlikely, given that knockout of S100A4 is associated with very effective antifibrotic effects in the bleomycin-induced skin fibrosis [16]. Another explanation may be related to antibody–target complex formation that activates innate immunity or complement cascade via Fc-gamma receptor binding [36]. Although speculative, 6B12 is an IgG1 mAb with preserved effector functions. The more recently developed humanized version of 6B12 is an IgG4 mAb (AX-202) with attenuated effector functions, where no unexpected effects were observed even at higher doses in the same preclinical model [37]. Last, play of chance cannot be discounted. Thus, further investigations are needed to validate these hypotheses.

Our data suggest that the potent antifibrotic effect of 7.5 mg/kg 6B12 was not achieved through suppression of fibrogenesis alone; rather, we postulate that anti-inflammatory and other potential effects are also at play. We observed fewer infiltrating leukocytes, decreased serum levels of cytokines (IL-1α) and chemokines (eotaxin, MCP-1, RANTES), and effectively downregulated inflammatory pathways such as T cell activation/proliferation, immune cell migration, chemotaxis, chemokine signalling, cell adhesion and IL-6-JAK-STAT3 signalling. STAT3 plays a pivotal role in fibrogenesis and is involved in both IL-6 and non-canonical TGF-β signalling [38–40]. These findings are novel but not entirely surprising since it has been established that extracellular S100A4 acts like a cytokine, interacting with TLR-4 and RAGE. Consequently, S100A4 activates pathways such as mitogen-activated protein kinases (MAPK) and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB), enhancing the motility of macrophages and neutrophils, and upregulates IL-1β, IL-6, IL-10, TNF and other cytokines/chemokines [11–15]. Importantly, the cytokines/chemokines downregulated by 7.5 mg/kg 6B12 are known to be upregulated locally or systemically in SSc and play a pivotal or potential role in SSc [8, 9, 41–45]. Moreover, GSEA analyses of the anti-S100A4-DEGs revealed potential links of anti-S100A4 mAb to reactive oxygen species signalling, which has been implicated in the pathogenesis of SSc [46, 47].

Of note, there were no overt signs of toxicity in response to 6B12 treatment at any dose in either clinical observation, weight assessment or gross necropsy.

Lastly, this study is limited by the lack of isotype control due to logistical constraints; thus, the potential reduction of the studied outcomes to pre-treatment levels needs to be interpreted with caution, and deserves further validation. Moreover, only one preclinical model for assessing the effect of S100A4 neutralizing mAb on pre-established experimental skin fibrosis has been used herein. Indeed, the murine model of bleomycin-induced dermal fibrosis only recapitulates some aspects of the pathogenesis of SSc, mainly the early inflammatory changes where fibrogenesis is predominantly driven by profibrotic cytokines released by infiltrating leukocytes, thus rendering this model more amenable to anti-inflammatory therapies that may not be effective in human disease [19]. Nevertheless, studies with the novel humanized IgG4 anti-S100A4 mAb (AX-202) in various preclinical in vivo and in vitro models of SSc are currently underway.

Taken together, our findings indicate that inactivating S100A4 by 7.5 mg/kg 6B12 attenuates and might reduce preestablished dermal fibrosis induced by bleomycin, rendering both antifibrotic and anti-inflammatory effects, as well as transcriptional changes relevant to SSc pathogenesis. Therefore, anti-S100A4 mAb exhibits immense therapeutic potential, which deserves further research and validation in other preclinical models of tissue fibrosis, and potentially in clinical trials in SSc. Given our previous findings of increased systemic levels of S100A4 in SSc patients and their associations with disease activity, lung and skin involvement, and the ability to predict the decrease in systemic inflammation of SSc-ILD patients treated with CYC [17], S100A4 could be a useful marker for identifying and stratifying SSc patients suitable for targeted treatment with S100A4-neutralizing mAb.

Supplementary material

Supplementary material is available at Rheumatology online.

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

Contribution statement

H.Š., J.K., R.I.H., J.H., L.Šenolt, J.H.W.D. and M.T. designed the study. X.Š., H.Š., T.T.-M., M.C.T., L.Štorkánová, H.H., S.O., B.H., R.B., K.P., J.V., J.K., R.I.H., J.H., L.Šenolt, J.H.W.D. and M.T. performed the acquisition, analysis or interpretation of the data. X.Š., H.Š., T.T.-M., M.C.T., J.H.W.D. and M.T. performed the statistical analysis. X.Š., H.Š. and M.T. prepared the original draft of the manuscript. All authors critically interpreted the results, reviewed the drafts, and approved the final manuscript.

Funding

This work was supported by the Ministry of Health of the Czech Republic [023728] and Ministry of Education Youth and Sports of the Czech Republic [SVV 260638]. This study was also supported by project-specific funding of Arxx Therapeutics. J.H.W.D. was supported by grants DI 1537/14-1, DI 1537/17-1, DI 1537/20–1, DI 1537/22-1 and DI 1537/23-1, of the German Research Foundation, SFB CRC1181 (project C01) and SFB TR221/project number 324392634 (B04) of the German Research Foundation, grant A79 of the IZKF in Erlangen, grant 2013.056.1 of the Wilhelm-Sander-Foundation, grants 2014_A47 and 2014_A184 of the Else-Kröner-Fresenius-Foundation, BMBF, MASCARA program, TP 2 (01EC1903A) and a Career Support Award of Medicine of the Ernst Jung Foundation.

Disclosure statement: J.H.W.D. has consultancy relationships with AbbVie, Active Biotech, Anamar, ARXX, AstraZeneca, Bayer Pharma, Boehringer Ingelheim, Celgene, Galapagos, GSK, Inventiva, Janssen, Novartis, Pfizer and UCB. J.H.W.D. has received research funding from Anamar, ARXX Therapeutics, BMS, Bayer Pharma, Boehringer Ingelheim, Cantargia, Celgene, CSL Behring, Galapagos, GSK, Inventiva, Kiniksa, Sanofi-Aventis, RedX and UCB. J.H.W.D. is stock owner of 4D Science. The 6B12 anti-S100A4 monoclonal antibody was provided by Arxx Therapeutics.

Patient consent for publication: Not required.

Provenance and peer review: Not commissioned, externally peer-reviewed.