Article Text
PDF
Abstract
Objectives To investigate the disease-modifying effects of phosphodiesterase 4 (PDE4) inhibition in preclinical models of systemic sclerosis (SSc).
Methods We studied the effects of PDE4 inhibition in a prevention and a treatment model of bleomycin-induced skin fibrosis, in the topoisomerase mouse model as well as in a model of sclerodermatous chronic graft-versus-host disease. To better understand the mode of action of PDE4 blockade in preclinical models of SSc, we investigated fibrosis-relevant mediators in fibroblasts and macrophages from healthy individuals and patients suffering from diffuse-cutaneous SSc on blockade of PDE4.
Results Specific inhibition of PDE4 by rolipram and apremilast had potent antifibrotic effects in bleomycin-induced skin fibrosis models, in the topoisomerase I mouse model and in murine sclerodermatous chronic graft-versus-host disease. Fibroblasts were not the direct targets of the antifibrotic effects of PDE4 blockade. Reduced leucocyte infiltration in lesional skin on PDE4 blockade suggested an immune-mediated mechanism. Further analysis revealed that PDE4 inhibition decreased the differentiation of M2 macrophages and the release of several profibrotic cytokines, resulting in reduced fibroblast activation and collagen release. Within these profibrotic mediators, interleukin-6 appeared to play a central role.
Conclusions PDE4 inhibition reduces inflammatory cell activity and the release of profibrotic cytokines from M2 macrophages, leading to decreased fibroblast activation and collagen release. Importantly, apremilast is already approved for the treatment of psoriasis and psoriatic arthritis. Therefore, PDE4 inhibitors might be further developed as potential antifibrotic therapies for patients with SSc. Our findings suggest that particularly patients with inflammation-driven fibrosis might benefit from PDE4 blockade.
- Systemic Sclerosis
- Fibroblasts
- Autoimmune Diseases
Statistics from Altmetric.com
Introduction
Fibrosis is a defining characteristic of systemic sclerosis (SSc) as well as a major cause of morbidity and mortality among patients. On a molecular level, fibrosis results from the accumulation of excessive amounts of extracellular matrix proteins released by chronically activated fibroblasts.1 ,2
Particularly in early phases of SSc, leucocytic infiltrates with macrophages, T cells and B cells are a common feature in affected organs. These inflammatory infiltrates are important sources of profibrotic mediators: the release of interleukin-6 (IL-6), transforming growth factor-β (TGF-β) and other profibrotic mediators initiates profibrotic processes through pathological activation of fibroblasts.1 ,3 In a subset of patients, these inflammatory processes persist and further drive the progression of fibrosis.4 ,5
Monocytes and macrophages are among the most abundant cell types in leucocytic infiltrates in SSc. Sclerotic skin of patients with early SSc contains an increased number of CD14+ monocytes/macrophages,6 and the ratio of CD68+ macrophages to T cells is high in sclerotic skin.7 Higashi-Kuwata et al showed an increasing number of cells expressing CD204, a marker for alternatively activated M2 macrophages, in localised scleroderma paralleling the severity of inflammation.8–10 Although macrophage polarisation has not directly been investigated in fibrotic SSc tissue yet, high levels of IL-4 and IL-13 might favour M2 differentiation.8 Of note, research on other fibrotic diseases indicates that M2 macrophages may propel fibrotic processes by releasing profibrotic mediators.11 Molecular profiling of skin biopsies from the FASSCINATE trial has recently suggested that this profibrotic role of M2 macrophages might also hold true in SSc. Gene expression analysis revealed a M2 signature in fibrotic SSc skin, which was downregulated by the IL-6 receptor blocker tocilizumab.12 ,13
Cyclic adenosine monosphosphate (cAMP) is an ubiquitous second messenger molecule that orchestrates physiological responses, such as apoptosis, lipid metabolism and inflammation.14–19 Its homeostasis is controlled by phosphodiesterases, a superfamily of enzymes that catalyse the breakdown of cAMP to monomeric AMP, thereby inactivating the molecule.20 The cAMP-specific phosphodiesterase (PDE) isoenzyme PDE4 is almost exclusively expressed within inflammatory cells.21 ,22 PDE4 inhibition has well-established disease-modifying activity in specific inflammatory diseases, including psoriasis,23 psoriatic arthritis24 ,25 and Behçet's disease.26 In the present study, we evaluated PDE4 inhibition as a novel therapeutic approach in treating fibrosis in SSc. We observed that PDE4 blockade ameliorated experimental fibrosis in different models through downregulating the release of profibrotic mediators from M2 macrophages.
Materials and methods
A detailed description of the methods is provided in the online supplementary file.
supplementary file
Mice and therapeutics
C57/Bl6 and BALB/c mice were purchased from Janvier (Le Genest Saint Isle, France), B10.D2 mice from Jackson Laboratories (Bar Harbor, Maine, USA). (R,S)-Rolipram (LC Laboratories, Woburn, Massachusetts, USA) was dissolved in dimethyl sulfoxide and further diluted in phosphate buffer saline for intraperitoneal application twice daily. Apremilast was diluted in 1% methycellulose and was applied orally twice daily.
Bleomycin-induced dermal fibrosis
Skin fibrosis was induced in C57Bl/6 mice aged 6 weeks by subcutaneous injections of bleomycin as described previously.27–33 After 4 (preventive model) or 6 weeks (therapeutic model) of bleomcyin challenge, mice were sacrificed and the injected skin processed for further analysis.
Topoisomerase I mouse model
Skin fibrosis was induced in C57Bl/6 mice aged 6 weeks by subcutaneous injections of 250 U/mL recombinant DNA topoisomerase I as described previously.34 Controls were injected with 0.9% NaCl. After 8 weeks of topoisomerase I challenge, mice were sacrificed and the injected skin processed for further analysis.
Sclerodermatous, chronic graft-versus-host disease
The B10.D2→Balb/c [H-2(d)] minor histocompatibility antigen-mismatched model was performed as described previously.35–40
Analysis of murine skin
Skin thickness, α-smooth muscle actin (SMA) counts, histomorphometry of fibrotic tissue and inflammatory infiltrates were analysed as described.27–33
Immunofluorescence for F4/80 and arginase and IL-6
The following primary antibodies were used: F4/80 (AdD Serotec, UK), cMAF (Abgent, USA), arginase (Santa Cruz, Germany), IL-6 (Abcam, UK), IgG (Beckton Dickinson, USA).
Human fibroblasts and macrophages
Fibroblasts and peripheral blood were isolated from healthy donors and lesional skin of patients with diffuse-cutaneous SSc. Fibroblasts were cultured as described.27 ,28 ,33 All healthy individuals and patients with SSc provided written informed consent as approved by the institutional ethics committees.
Quantification of collagen protein
The amount of soluble collagen in cell culture supernatants was quantified using the SirCol collagen assay (Biocolor, Belfast, Northern Ireland).
Cell viability and cytotoxicity assays
Cell viability was quantified using the Cell Counting Kit 8 (Dojindo Molecular Technologies, Maryland, USA).41
IL-6 ELISA
IL-6 was determined in the supernatants from the human macrophage experiments with the human IL-6 DuoSet ELISA (R&D Systems, Minneapolis, Minnesota, USA).
Quantitative real-time PCR
Gene expression was quantified by SYBR green real-time PCR on a StepOne System quantitative PCR System (Thermo Fischer Scientific, Waltham, Massachusetts, USA).
Statistical analysis
All data are presented as median with IQR. Differences between the groups were tested for their statistical significance by two-tailed Mann-Whitney U non-parametric test using GraphPad Prism (V.5.03). p Values of <0.05 were considered to be statistically significant.
Results
Inhibition of PDE4 prevents bleomycin-induced dermal fibrosis
We first investigated the effects of PDE4 blockade in bleomycin-induced skin fibrosis. When we treated bleomycin-challenged mice with the PDE4 inhibitor rolipram, a lead compound for the clinically available apremilast, we observed dose-dependent antifibrotic effects as assessed by the reduction of dermal thickening, fibrotic tissue and myofibroblast counts (figure 1A, B). In the group of mice receiving rolipram 5.0 mg/kg twice daily, we found decreases in skin thickening by 64%, in fibrotic tissue by 50% and in myofibroblast counts by 70% (figure 1A, B). Apart from these potent antifibrotic effects, we observed a strong decline in leucocytic infiltrates on PDE4 blockade. Animals receiving 5.0 mg/kg rolipram twice daily showed a reduction of infiltrating leucocytes by 49% (figure 1C). PDE4 inhibition was well tolerated throughout all experiments, as indicated by constant body weight, normal texture of the fur and normal activity.
Inhibition of phosphodiesterase 4 (PDE4) by rolipram inhibits the development of bleomycin-induced skin fibrosis. (A) Representative images of Masson's trichrome with blue staining for collagens (upper pictures) and sirius red with orange staining for collagens (lower pictures). Pictures are shown in 100-fold magnification. Skin thickening as determined by Masson's trichrome stainings. Fibrotic tissue as assessed by histomorphometric measurements. (B) α-Smooth muscle actin-positive myofibroblasts. (C) Inflammatory infiltrates as determined in H&E stainings. (D) F4/80 single positive macrophages. (E) F4/80, cMAF and arginase triple positive macrophages. (F) Tissue interleukin-6 (IL-6) levels as assessed by immunofluorescence staining. (A–F) Animal groups consisted of N≥7 mice each. Statistical description: * for 0.01<p<0.05, ** for 0.001<p<0.01, *** for p<0.0001. bleo, bleomycin-treated mice; HPF, highpower field; veh, vehicle-treated mice; w, weeks.
The antifibrotic effects of PDE4 inhibition are not mediated by direct effects on fibroblasts
Since fibroblasts have been shown to express PDE4,42 we investigated if PDE4 blockade had direct effects on fibroblasts. We observed that rolipram did not inhibit fibroblast proliferation in healthy and SSc fibroblasts until cytotoxicity occurred at high doses of 1000 µM, which by far exceeds clinical relevant concentrations43 ,44 (see online supplementary figure S1). Moreover, PDE4 inhibition left stress fibre formation of healthy and SSc fibroblasts unaffected (figure 2A). Consistently, PDE4 inhibition did neither alter closure time of artificial scratches nor the migration rates of healthy and SSc fibroblasts (figure 2E, F). Finally, we did not detect any inhibitory effects of PDE4 blockade either on COL1A1 and PAI-1 gene transcription (figure 2B, C) or on collagen release (figure 2D) in resting and TGF-β-stimulated fibroblasts. We therefore hypothesised that the antifibrotic effects of PDE4 inhibition might be leucocyte-dependent, which was supported by decreased leucocyte infiltration in the bleomycin model on PDE blockade.
Phosphodiesterase 4 (PDE4) inhibition has no direct effects on fibroblasts. (A) Stress fibre formation in fibroblasts from healthy individuals (upper pictures) and patients with systemic sclerosis (SSc) (lower pictures) as assessed by phalloidin red staining. Nuclei are stained with 4′,6-diamidin-2-phenylindol. Representative stainings are shown in 200-fold magnification. N=3. (B, C) Messenger RNA levels of transforming growth factor-β (TGF-β) target genes COL1A1 and PAI-1 of unstimulated and TGF-β-stimulated dermal fibroblasts from healthy individuals and patients with SSc. (D) Secreted collagen proteins in the supernatant of unstimulated and TGF-β-stimulated dermal fibroblasts from healthy individuals and patients with SSc as assessed by SirCol collagen assay. (E) Representative pictures of scratch assay experiments assessing closure of the scratch after 48 hours. (F) Quantification of scratch closure time and migration rate in dermal fibroblasts from healthy individuals and patients with SSc. (A–F) In all experiments N≥3. Statistical description: * for 0.01<p<0.05, ** for 0.001<p<0.01, *** for p<0.0001. ctr, control.
supplementary figures
Inhibition of PDE4 reduces the release of profibrotic cytokines from alternatively activated macrophages
M2 macrophages are a central source of profibrotic mediators. We hence hypothesised that the antifibrotic effects of PDE4 inhibition in bleomycin-induced skin fibrosis might have resulted from a reduced release of profibrotic cytokines from macrophages. We therefore isolated peripheral blood monocytes from healthy volunteers and patients with diffuse-cutaneous SSc and differentiated them into M1 and M2 macrophages. Increased iNOS mRNA levels confirmed the differentiation into the M1 phenotype, increased ARGINASE mRNA into the M2 phenotype. PDE4 blockade by rolipram inhibited the differentiation of monocytes into M2 macrophages (figure 3A, B), while differentiation into the M1 phenotype remained unaffected (see online supplementary figure S3A, B). In addition, mRNA levels of the profibrotic cytokines IL-6, IL-13, TGF-β1 and TGF-β2 mRNA as well as the secretion of IL-6 were reduced on treatment of M2 macrophages with rolipram (figure 3C–F). Along with our hypothesis that PDE4 blockade particularly affects the M2 macrophages, rolipram treatment did not show significant effects on the expression of IL-6, IL-10, IL-13 and tumour necrosis factor (TNF)-α in M1 macrophages from healthy individuals and patients with SSc (see online supplementary figure S3C,D). Release of IL-6 protein was not affected by PDE4 inhibition in M1 macrophages (see online supplementary figure S3E, F).
Inhibition of phosphodiesterase 4 (PDE4) interferes with the release of profibrotic cytokines from M2 macrophages. (A–F) M2 macrophages from healthy volunteers and patients with diffuse-cutaneous systemic sclerosis (dcSSc). (A–D) Messenger RNA expression of specific ARGINASE as well as interleukin-6 (IL-6), IL-13, transforming growth factor (TGF)-β1 and TGF-β2. (E, F) IL-6 protein levels in the supernatants. (A–F) In all experiments, N≥10. Statistical description: * for 0.01<p<0.05, ** for 0.001<p<0.01, *** for p<0.0001. ctr, control.
We also assessed the effects of rolipram on already differentiated M2 macrophages. Again, mRNA levels of the profibrotic cytokines IL-6, IL-13, TGF-β1 and TGF-β2 as well as protein levels of IL-6 were reduced significantly, suggesting that PDE4 blockade could prevent and reverse the profibrotic cytokine milieu generated by M2 macrophages (see online supplementary figure S4).
To exclude that inhibition of M2 polarisation is due to off-target effects of rolipram, we knocked down PDE4B, the major PDE4 isoform in macrophages by small interfering RNA (siRNA). Consistent with the findings observed with rolipram, siRNA-mediated silencing of PDE4B inhibited alternative activation and M2 polarisation of macrophages. Consistently, knockdown of PDE4B also reduced the mRNA levels of profibrotic mediators such as IL-6, IL-13, TGF-β1 and TGF-β2 (see online supplementary figure S5).
We next investigated whether the inhibitory effects of PDE4 blockade on M2 macrophages were also relevant in vivo. Indeed, we observed a dose-dependent reduction of F4/80 single positive monocytes and F4/80/cMAF/arginase triple positive M2 macrophages in skin sections of bleomycin-challenged mice (figure 1D, E). Furthermore, tissue IL-6 levels were also reduced (figure 1F and see online supplementary figure S2), indicating that PDE4 inhibition blocks the release of profibrotic cytokines from M2 macrophages both in vitro and in vivo.
Inhibition of PDE4 ameliorates established skin fibrosis
So far, we have demonstrated that PDE4 blockade prevents bleomycin-induced fibrosis by interfering with the release of IL-6 and potentially other mediators from M2 macrophages. Next, we wondered if PDE4 inhibition might also be effective in pre-established fibrosis. We therefore took advantage of a modified model of bleomycin-induced fibrosis in which PDE4 blockade was initiated once skin fibrosis had already been established. We used the clinically available PDE4 inhibitor apremilast to demonstrate that the antifibrotic effects were applicable to PDE4 inhibitors in general.
When we compared the treatment groups with the first control group (6 weeks of bleomycin challenge) (see online supplementary figure S6), we observed that skin thickness, morphometric fibrosis assessment and myofibroblast numbers were reduced significantly by both doses of apremilast, suggesting that PDE4 blockade effectively prevented chronic progression of fibrosis. Next, we compared both treatment groups with the second control group (3 weeks of bleomycin challenge followed by 3 weeks of NaCl injections). Intriguingly, we observed that apremilast treatment reduced all fibrosis outcome measures below baseline fibrosis levels, indicating that PDE4 blockade induced regression of fibrosis (figure 4A, B).
Phosphodiesterase 4 (PDE4) inhibition by apremilast induces regression of pre-established bleomycin-induced skin fibrosis. (A) Representative images of Masson's trichrome-stained sections (upper pictures) and sirius red-stained sections (lower pictures) at 100-fold magnification. Skin thickening as determined by Masson's trichrome stainings. Fibrotic tissue as assessed by histomorphometric measurements. (B) α-Smooth muscle actin-positive myofibroblasts. (C) Inflammatory infiltrates as determined in H&E stainings. (D) F4/80 single positive cells. (E) F4/80, cMAF and arginase triple positive macrophages. (F) Interleukin-6 (IL-6) tissue levels assessed by immunofluorescence staining. (A–F) Animal groups consisted of ≥6 mice each. Statistical description: * for 0.01<p<0.05, ** for 0.001<p<0.01, *** for p<0.0001. bleo, bleomycin-treated mice; veh, vehicle-treated mice; w, weeks.
In addition to the antifibrotic effects, leucocytic infiltrates were reduced on treatment with apremilast (figure 4C). Again, we observed a dose-dependent reduction of both F4/80 single positive monocytes and F4/80/cMAF/arginase triple positive M2 macrophages on treatment with apremilast (figure 4D, E). Furthermore, tissue IL-6 levels were dose-dependently reduced by 58% and 73% after treatment with 5.0 and 25.0 mg/kg apremilast twice daily (figure 4F and see online supplementary figure S7). Together, these experiments indicated that PDE4 inhibition by the clinically approved apremilast prevented progression of chronic fibrosis and even reversed establish fibrosis by reducing M2 differentiation and IL-6 release.
PDE4 blockade inhibits dermal fibrosis in the topoisomerase I mouse model
Since autoantibodies play a central role in initiation and progression of fibrosis, we studied the effects of pharmacological PDE4 blockade in mice immunised with the DNA topoisomerase I. Treatment with rolipram reduced skin thickness by 123%, fibrotic tissue by 75% and α-SMA-positive myofibroblasts by 91% (figure 5A, B). In line with our results from the bleomycin models, PDE4 inhibition by 5.0 mg/kg rolipram reduced the inflammatory infiltrates by 84% (figure 5C). F4/80 single positive monocytes were reduced by 67% and F4/80/cMAF/arginase triple positive M2 macrophages by 63% on treatment with rolipram (figure 5D, E). Furthermore, tissue IL-6 levels were reduced by 137% (figure 5F, see online supplementary figure S8).
Phosphodiesterase 4 (PDE4) inhibition by rolipram inhibits dermal fibrosis in the topoisomerase I mouse model. (A) Representative images of Masson's trichrome-stained sections (upper pictures) and sirius red-stained sections (lower pictures) at 100-fold magnification. Skin thickening as determined by Masson's trichrome stainings. Fibrotic tissue as assessed by histomorphometric measurements. (B) α-Smooth muscle actin-positive myofibroblasts. (C) Inflammatory infiltrates as determined in H&E stainings. (D) F4/80 single positive cells. (E) F4/80, cMAF and arginase triple positive macrophages. (F) Interleukin-6 (IL-6) tissue levels assessed by immunofluorescence staining. (A–F) Animal groups consisted of ≥7 mice each. Statistical description: * for 0.01<p<0.05, ** for 0.001<p<0.01, *** for p<0.0001. topo, topoisomerase I-treated mice; veh, vehicle-treated mice.
Inhibition of PDE4 reduces dermal fibrosis caused by chronic graft-versus-host reaction
Since SSc is a systemic disease, we finally investigated the efficacy and tolerability of PDE4 blockade in the sclGvHD mouse model. When we treated transplanted animals with the PDE4 inhibitor rolipram, skin thickening, morphometric fibrosis assessment and myofibroblast counts (figure 6A, B) all indicated strong antifibrotic effects in well-tolerated doses. In line with our results from the bleomycin models, rolipram reduced the inflammatory infiltrates (figure 6C). Again, the antifibrotic effects of PDE4 inhibition were, at least in part, the result of reduced release of profibrotic mediators from M2 macrophages as indicated by a potent decreases in the numbers of M2 macrophages (figure 6D and E) and tissue IL-6 levels in the treatment group (figure 6F and see online supplementary figure S9).
Inhibition of phosphodiesterase 4 (PDE4) ameliorates fibrosis in murine sclerodermatous chronic graft-versus-host disease. (A) Representative images of Massons's trichrome (upper images) and sirius red-stained (lower images) sections at 100-fold magnification. Skin thickening as determined by Masson's trichrome stainings. Fibrotic tissue assessed by histomorphometric measurements. (B) α-Smooth muscle actin-positive myofibroblasts. (C) Inflammatory infiltrates as determined in H&E stainings. (D) F4/80 single positive cells. (E) F4/80, cMAF and arginase triple positive macrophages. (F) Tissue interleukin-6 (IL-6) levels assessed by immunofluorescence staining. (A–F) The groups consisted of ≥6 mice each. Statistical description: * for 0.01<p<0.05, ** for 0.001<p<0.01, *** for p<0.0001. Veh, vehicle-treated mice.
Discussion
Intensive research of the last decades resulted in the clinical development of PDE4 inhibitors for the treatment of psoriasis, psoriatic arthritis and potentially Behçet's disease. The current literature suggests that the disease-modifying effects of PDE4 inhibitors in these diseases mainly result from their anti-inflammatory activity. Since autoimmunity and inflammation drive fibrosis in early stages of SSc and persist in a subset of patients, we initiated the current study to investigate a potential disease-modifying activity of PDE4 inhibition in SSc.
Using the lead compound rolipram and apremilast, the PDE4 inhibitor used in the clinic, we observed potent antifibrotic activity on PDE4 blockade. Although PDE4 is expressed in fibroblasts and although previous studies suggested inhibitory effects of PDE4 blockade on fibroblast contraction and chemotaxis,45 we did not observe any direct effects of PDE4 inhibition on fibroblast activation, migration and collagen release in clinically relevant doses. These observations were consistent with our hypothesis that PDE4 inhibition might have disease-modifying antifibrotic activity through interaction with immune cell activation in SSc.
Both in leucocytic infiltrates in early SSc and in the skin of bleomycin-challenged mice, macrophages represent one of the most abundant cell populations. Accumulating preclinical evidence suggest that subgroups of macrophages, often referred to as M2 macrophages, are key mediators of physiological wound healing and pathological fibrosis.11 These preclinical findings are corroborated by data from the FASSCINATE trial.12 Gene expression analysis revealed a M2-macrophage signature in SSc skin, which was responsive to treatment with tocilizumab. We observed that PDE4 blockade was effective in inhibiting the differentiation of monocytes/macrophages into a M2 phenotype and in reducing the release of profibrotic mediators, including IL-6. Since PDE4 inhibition blocked the expression of several fibrosis relevant mediators in M2 macrophages, we hypothesise that its antifibrotic efficacy may exceed the antifibrotic effects of selective IL-6 inhibition in patients with SSc.
Our study provides first evidence that PDE4 blockade has a profound antifibrotic activity. Gobejishvili et al46 and Udalov et al47 investigated PDE4 blockade in models of cholestatic liver disease and chronic lung injury, respectively, supporting the idea that PDE4 might play a general role in chronic organ damage. In detail, cholestatic liver injury induced by bile duct ligation was accompanied by increased PDE expression. Treatment with rolipram reduced inflammatory and profibrotic cytokine expression. Udalov et al used the PDE4 inhibitor cilomilast to demonstrate reduced lung injury and fibrosis after a single bleomycin challenge. Similar to our results in the skin, treatment with cilomilast reduced the number of macrophages in bronchoalveolar lavage (BAL) fluid, while neutrophils remained unchanged. By contrast to our results, the authors observed decreased TNF-α mRNA but increased IL-6 mRNA BAL levels. This might reflect the early disease stage after injury during which these investigations were performed. At this early stage, inflammation is still dominated by M1 macrophages, whereas M2 macrophages as a predominant source of profibrotic mediators accumulate during later stages of bleomycin-induced injury.11
Since the ‘standard’ bleomycin model is used to investigate prevention of fibrosis, it was crucial to study the effects of PDE4 inhibition in a modified model, in which treatment is initiated, once fibrosis has already been established. PDE4 blockade was effective in preventing progression of chronic fibrosis and reversing established fibrosis. These observations were accompanied by increased counts of M2 macrophages in established fibrosis, which were normalised by PDE blockade. We believe that this finding might indicate that M2 macrophages may contribute to the persistence of fibrotic disease in patients with SSc. Targeting M2 macrophages might therefore re-establish physiological tissue homeostasis and allow reversal of fibrosis.
In addition to its antifibrotic effects in bleomycin-induced fibrosis, we observed that PDE4 blockade was also effective in treating preclinical sclGvHD, a model to mimic systemic fibrosis as seen in patients with diffuse-cutaneous SSc. Although the sclGvHD model was long thought to be T cell-driven, accumulating evidence suggests a central role of M2 macrophages chronic graft-versus-host disease,48 which is in line with our observations. In addition to bleomycin-induced fibrosis and murine sclGvHD, PDE4 inhibition also demonstrated potent antifibrotic effects in topoisomerase-induced fibrosis. Histological analyses indicate that early phases of SSc are characterised by inflammatory infiltrates,3 and genetic profiling studies highlight that SSc may evolve in distinct disease subtypes, including inflammatory subtypes.49 In this context, our current study highlights potent antifibrotic effects of PDE4 inhibition in SSc models reflecting exactly these early stages and inflammatory subtypes. By contrast to its effects in other rheumatic conditions, PDE4 inhibitors act primarily through modulating the release of profibrotic mediators from macrophages expressing a M2 phenotype. Since accumulating evidence suggests that M2 macrophages may play a more general role in several subtypes of fibrosis, including inflammatory and fibroproliferative disease types, our findings might prompt additional experimental studies to investigate a potential role of PDE4 inhibitors in inflammation-independent, fibroproliferative diseases.
The PDE4 inhibitor apremilast is already approved for the treatment of psoriasis and psoriatic arthritis. Apart from minor gastrointestinal side effects during the initiation of therapy, apremilast is very well tolerated and does per se not require routine laboratory testing compared with other disease-modifying agents. Our results suggest that apremilast, as well as other PDE4 inhibitors, might be tested and further developed for the treatment of patients with SSc at early stages or with persistent inflammatory disease.
Acknowledgments
We thank Regina Kleinlein, Katja Dreißigacker and Rossella Mancuso, PhD, and Ina Müller for excellent technical assistance.
References
Footnotes
Handling editor Tore K Kvien
Contributors Design of the study: CM, CB, JHWD, GS. Acquisition of data: CM. Interpretation of data: CM, CB, JHWD. Manuscript preparation: CM, CB, JHWD. Provided essential material: CM, AR, CB,CB, RW, NK.
Funding Grant support was provided by the Erlanger Leistungsbezogene Anschubfinanzierung und Nachwuchsföderung (ELAN), grants J29, A57 and A64 of the Interdisciplinary Center of Clinical Research (IZKF) in Erlangen and grants DI 1537/5–1, DI 1537/7–1, DI 1537/8–1, DI 1537/9–1, DI 1537/11–1, DI 1537/11–1, AK 144/2–1 and BE 5191/1–1 from the Deutsche Forschungsgemeinschaft. In addition, the study was supported by grant 2013.056.1 of the Wilhelm-Sander-Foundation, grant 2014_A47 of the Else-Kröner-Fresenius-Foundation, and a Career Support Award of Medicine of the Ernst Jung Foundation.
Competing interests JHWD has consultancy relationships with Actelion, Active Biotech, Anamar, Bayer Pharma, Boehringer Ingelheim, Celgene, Galapagos, GlaxoSmithKline (GSK), Inventiva, JB Therapeutics, Medac, Pfizer, RuiYi and UCB. JHWD has received research funding from Anamar, Active Biotech, Array Biopharma, BMS, Bayer Pharma, Boehringer Ingelheim, Celgene, GSK, Novartis, Sanofi-Aventis, UCB. JHWD is stock owner of 4D Science.
Ethics approval FAU Erlangen-Nuremberg.
Provenance and peer review Not commissioned; externally peer reviewed.