Pro-Resolving Factors Released by Macrophages After Efferocytosis Promote Mucosal Wound Healing in Inflammatory Bowel Disease

Abstract

Nonresolving inflammation is a critical driver of several chronic inflammatory diseases, including inflammatory bowel diseases (IBD). This unresolved inflammation may result from the persistence of an initiating stimulus or from the alteration of the resolution phase of inflammation. Elimination of apoptotic cells by macrophages (a process called efferocytosis) is a critical step in the resolution phase of inflammation. Efferocytosis participates in macrophage reprogramming and favors the release of numerous pro-resolving factors. These pro-resolving factors exert therapeutic effects in experimental autoimmune arthritis. Here, we propose to evaluate the efficacy of pro-resolving factors produced by macrophages after efferocytosis, a secretome called SuperMApo, in two IBD models, namely dextran sodium sulfate (DSS)-induced and T cell transfer-induced colitis. Reintroducing these pro-resolving factors was sufficient to decrease clinical, endoscopic and histological colitis scores in ongoing naive T cell-transfer-induced colitis and in DSS-induced colitis. Mouse primary fibroblasts isolated from the colon demonstrated enhanced healing properties in the presence of SuperMApo, as attested by their increased migratory, proliferative and contractive properties. This was confirmed by the use of human fibroblasts isolated from patients with IBD. Exposure of an intestinal epithelial cell (IEC) line to these pro-resolving factors increased their proliferative properties and IEC acquired the capacity to capture apoptotic cells. The improvement of wound healing properties induced by SuperMApo was confirmed in vivo in a biopsy forceps-wound colonic mucosa model. Further in vivo analysis in naive T cell transfer-induced colitis model demonstrated an improvement of intestinal barrier permeability after administration of SuperMApo, an intestinal cell proliferation and an increase of α-SMA expression by fibroblasts, as well as a reduction of the transcript coding for fibronectin (Fn1). Finally, we identified TGF-β, IGF-I and VEGF among SuperMApo as necessary to favor mucosal healing and confirmed their role both in vitro (using neutralizing antibodies) and in vivo by depleting these factors from efferocytic macrophage secretome using antibody-coated microbeads. These growth factors only explained some of the beneficial effects induced by factors released by efferocytic macrophages. Overall, the administration of pro-resolving factors released by efferocytic macrophages limits intestinal inflammation and enhance tissue repair, which represents an innovative treatment of IBD.

Keywords: colitis; efferocytosis; fibroblasts; macrophages; resolution of inflammation; wound healing.

Figure 1

Pro-resolving factors released by macrophages after efferocytosis promote intestine mucosal healing and limit colitis. Colitis was induced by the transfer of naïve CD4⁺CD25^–CD45RB^hi T cells into Rag2^–/– C57BL/6 mice. (A) Colon examination was performed by colonoscopy and lesion quantification was assessed by the murine endoscopic index of colitis severity (MEICS) score. Mice developing colitis received twice (arrows) pro-resolving factors (SuperMApo) or vehicle (1 mL/mouse, i.p.) 10 days after the induction of colitis (i.e., naïve T cell transfer) when they experimented a MEICS score of 6 to 7 out of 15, and 48 h later. Representative pictures are displayed from day 10 and 19 according to MEICS items. Black arrows identify the MEICS items on each picture. Cumulated data from one representative experiment out of three, expressed as mean ± s.e.m. (5 to 6 mice per groups) were also shown. (B, C) Body weight loss and clinical score changes were assessed on colitis mice. Data are expressed as mean of group ± s.e.m. (5 to 6 animals per groups) from one representative experiment out of three independent experiments; ***p < 0.001 (two-way ANOVA with Bonferroni post-test). (D) Colon length was assessed on colitis mice from a. Representative colon pictures are shown as well as cumulative data from one representative experiment out of three independent experiments (5 to 6 mice per group); *p < 0.05 (nonparametric Mann-Whitney test). (E) Representative HES staining of paraffin-embedded sections of the colonic tissues obtained from colitis mice from (A), and cumulative histological scores (pooled from 3 independent experiments) are shown and expressed as mean of group ± s.e.m. (5 to 6 mice per group); **p < 0.01 (nonparametric Mann-Whitney test).

Figure 2

Pro-resolving factors released by macrophages after efferocytosis promote healing of colon lesions induced by dextran sodium sulfate and limit colitis. (A) C57BL/6 Mice were subjected to three cycles of DSS and received at day 0 and day 2 pro-resolving factors (SuperMApo, 1 mL/mouse; i.p.) or vehicle, and were monitored daily for clinical changes over 50 days. Data are expressed as mean of group + s.e.m. (5 to 6 mice per group) and are representative of two independent experiments; ***p < 0.001 (two-way ANOVA with Bonferroni post-test). (B) Endoscopic assessment according to the MEICS score was also performed on colitis mice from (A) as well as body weight loss (C). Representative pictures of colonoscopy at day 7 and day 49 are shown, as well as cumulated MEICS score from two independent experiments. Data are expressed as mean of group + s.e.m. (5 to 6 mice per group); **p < 0.01 (two-way ANOVA with Bonferroni post-test). (D) Colon length was assessed on the colitis mice from a, and representative colon pictures were shown as well as cumulative data from two independent experiments. Data are expressed as mean of group + s.e.m. (5 to 6 mice per group); **p < 0.01 (nonparametric Mann-Whitney test). (E) Representative HES staining of paraffin-embedded sections of the colonic tissues from colitis mice from a (at x50 and x100 magnifications). Cumulative histological scores from one experiment are shown as mean of group + s.e.m. (5 to 6 mice per group); *p < 0.05 (nonparametric Mann-Whitney test).

Figure 3

Pro-resolving factors released by macrophages after efferocytosis trigger wound healing properties in fibroblasts and intestinal epithelial cells. (A) Expression of vimentin, α-smooth muscle actin (α-SMA) assessed by immunofluorescence and DAPI staining in mouse colon primary fibroblasts cultured in the presence of SuperMApo or control medium for 24 (h) Representative pictures are shown as well as cumulative data representative of α-SMA expression from three independent experiments, as mean ± s.e.m.; ***p < 0.001 (unpaired student’s t test). (B) mRNA expression of genes coding for fibronectin (Fn1), type I (Col1a1) and type III (Col3a1) collagen in fibroblasts from (A). Data are expressed as mean of duplicate per condition ± s.e.m. from three independent experiments pooled together. **p < 0.01, ***p < 0.001 (unpaired student’s t test). (C) Migration of mouse fibroblasts in wound-healing scratch assays in the presence of SuperMApo or control medium. Data are expressed as mean of duplicates per conditions ± s.e.m. from one representative experiment out of three independent experiments. **p < 0.01 (two-way ANOVA with Bonferroni post-test). (D) Contractility of mouse fibroblasts was assessed in floating collagen gels containing medium or SuperMApo. Representative pictures are shown, as well as cumulative data (mean ± s.e.m.) from one representative experiment out of three independent experiments. **p < 0.01 (two-way ANOVA with Bonferroni post-test). (E) Engulfment of CFSE-labelled apoptotic cells by the mouse intestinal epithelial cell line MODE-K assessed by flow cytometry after 6 h of co-culture in presence of control medium or SuperMApo. To confirm that this represents an active mechanism and to appreciate CFSE-labelled apoptotic cells stick to MODE-K cells, the same experiment was performed at +4°C instead of 37°C. Data are expressed as representative dot plots showing the percentage of CFSE⁺ MODE-K cells and as index of phagocytosis ± s.e.m. from one representative experiment out of three independent experiments; ****p < 0.0001 (non-parametric Mann-Whitney test). (F, G) Primary human fibroblasts harvested from biopsies of non-inflamed (Fibro) or inflamed (infFibro) colon tissue were grown in the presence of medium (med) or SuperMApo. (F) Proliferation was determined by MTT assay and (G) wound closure was analyzed using the scratch assay at 0, 24, 48 and 72 h of culture. Data are expressed as mean of group ± s.e.m., and are representative of three independent experiments; *p < 0.05, **p < 0.01, ***p < 0.001 (two-way ANOVA with Bonferroni post-test).

Figure 4

Treatment with pro-resolving factors released by macrophages after efferocytosis maintains intestinal barrier permeability via intestinal epithelial cell proliferation and fibroblast activation. (A) Wild-type mice were monitored daily by video-endoscopy to evaluate healing of the colon mucosa, wounded at day 0 by forceps and receiving at day 0 and 2 SuperMApo or vehicle (1 mL/mouse; i.p.). Representative pictures are displayed from day 0 to day 4, as well as cumulated data (expressed as % of wound healing, right hand side panel). Data are representative of three independent experiments showing similar results, and expressed as mean of group ± s.e.m. (5 to 6 animals per group); *p < 0.05, **p < 0.01 (two-way analysis of variance [ANOVA] with Bonferroni post-test). (B–D) Colitis was induced by transfer of naive CD4⁺CD25^–CD45RB^hi T cells into Rag2^–/– C57BL/6 mice and mice received vehicle or SuperMApo at day 10 and 12 (1 mL/mouse; i.p.). (B) FITC-dextran particle gavage and quantification in the serum was assessed to appreciate intestinal barrier integrity. Data are expressed as mean of group ± s.e.m. (5 to 6 mice per group) and are representative of two independent experiments; *p < 0.05 (two-way ANOVA with Bonferroni post-test). (C) Colonoscopy evaluation of the mice and (D) regenerating islet-derived protein 3-gamma (REG3ɣ) quantification in the serum were also performed. Data are expressed as mean of group ± s.e.m. (5 to 6 mice per group) and representative of four independent experiments; **p < 0.01 (two-way ANOVA with Bonferroni post-test). (E) Cell proliferation in the epithelial crypts and (F) fibroblast differentiation into myofibroblasts were determined by Ki-67 and α-SMA staining, respectively, of colon cuts from colitis mice from (B). Nuclei are counterstained with DAPI and an isotype control antibody was used as control (Iso ctrol). (G) Expression of mRNA of genes coding for extracellular matrix proteins type I [Col1a1] and type III [Col3a1] collagen, fibronectin [Fn1]) and TGF-β (tgf-b) was assessed by RT-qPCR in the colon of the colitis mice from (B) Data are expressed as mean of group ± s.e.m. (5 to 6 mice per group) and are representative of two; *p < 0.05 (Unpaired student t test), **p < 0.01; ****p < 0.0001 (nonparametric Mann-Whitney test).

Figure 5

Pro-resolving factors released by macrophages after efferocytosis include TGF-β, IGF-I and VEGF and their capacity to attract primary fibroblasts and IEC are dependent on these growth factors. (A) EGF, FGF, TGF-β, IGF-I and VEGF were quantified by ELISA in the supernatants of macrophages after efferocytosis (SuperMApo) and in the control supernatants of macrophages (MacroSup) cultured alone or apoptotic cells (ApoSup) cultured alone. Data represented as mean ± s.e.m. are the pool of six independents experiments; *p < 0.05; ***p < 0.001 (one-way ANOVA with Tukey’s multiple comparison post-test). (B) Proliferation of mouse primary fibroblasts was assessed by MTT assay for 4 days with or without neutralizing antibodies to TGF-β, IGF-I or VEGF, or control isotype. Data are from one representative experiment out of three and expressed as mean + s.e.m. of 3 replicates per condition; *p < 0.05; **p < 0.01, ****p < 0.0001 (two-way ANOVA with Bonferroni post-test). (C) Mouse primary fibroblasts or MODE-K IEC were incubated with medium or SuperMApo, with or without neutralizing antibodies to TGF-β, IGF-I or VEGF, or control isotype, and migration was determined using a transwell assay during 24 (h). Data are shown as representative pictures and as pooled data expressed as mean of migrated cells/field of duplicates per group ± s.e.m. from one representative experiment out of three. **p < 0.01, ****p < 0.0001 (one-way ANOVA with tukey’s multiple comparisons post-test).

Figure 6

TGF-β, IGF-I and VEGF found in the SuperMApo supernatant participate in lesion healing. (A, B) Colitis was induced by the transfer of naive CD4⁺CD25^–CD45RB^hi T cells into Rag2^−/− C57BL/6 mice and mice received at day 10 and 12 vehicle (PBS; 1 mL/mouse) or the three following recombinant growth factors: TGF-β (700 pg/mL/mouse), IGF-I (1050 pg/mL/mouse) and VEGF (350 pg/mL/mouse) (red arrows) when they exhibited a MEICS score of 6-7, and were then monitored daily. (A) Endoscopic assessment according to the MEICS score was performed and results are shown as representative pictures of mini-colonoscopy at day 10, 13, 17 and 19, as well as cumulated score. (B) Clinical score changes and body weight loss were also assessed and results are expressed as mean of groups ± s.e.m. (5 to 6 mice per group). Data are from two independent experiments pooled together; ***p < 0.05, p < 0.001 (one-way ANOVA with Bonferroni post-test). (C) Colitis was induced similarly and mice received SuperMApo, or SuperMApo depleted in TGF-β, IGF-I and VEGF (SuperMApo-GF), or SuperMApo treated similarly but using isotype control antibodies instead of antibodies against growth factors (SuperMApo-iso), or vehicle when they exhibited a MEICS score of 6-7 and were then monitored daily. Endoscopic assessment according to the MEICS score was performed and results are shown as representative pictures of mini-colonoscopy at day 10, 13, 17 and 19, as well as cumulated score. (D) Clinical score changes and body weight loss were also assessed and results are expressed as mean of groups ± s.e.m. (5 to 6 mice per group). Data are from three independent experiments pooled together; **p < 0.01 (one-way ANOVA with Bonferroni post-test).