CD163 interacts with TWEAK to regulate tissue regeneration after ischaemic injury

Abstract

Macrophages are an essential component of the immune response to ischaemic injury and play an important role in promoting inflammation and its resolution, which is necessary for tissue repair. The type I transmembrane glycoprotein CD163 is exclusively expressed on macrophages, where it acts as a receptor for haemoglobin:haptoglobin complexes. An extracellular portion of CD163 circulates in the blood as a soluble protein, for which no physiological function has so far been described. Here we show that during ischaemia, soluble CD163 functions as a decoy receptor for TWEAK, a secreted pro-inflammatory cytokine of the tumour necrosis factor family, to regulate TWEAK-induced activation of canonical nuclear factor-κB (NF-κB) and Notch signalling necessary for myogenic progenitor cell proliferation. Mice with deletion of CD163 have transiently elevated levels of TWEAK, which stimulate muscle satellite cell proliferation and tissue regeneration in their ischaemic and non-ischaemic limbs. These results reveal a role for soluble CD163 in regulating muscle regeneration after ischaemic injury.

Figure 1. CD163^−/− mice exhibit enhanced skeletal muscle repair in both ischaemic and NILs in response to femoral ligation.

(a) Laser Doppler analysis with perfusion quantitation 1, 4, 7, 14, 21 and 28 days after femoral ligation (n=10 per group). (b,c) Immunostaining of NIL and IL for VE-cadherin and β-dystroglycan (red) at 14 and 28 days after femoral ligation. The graphs below show the number of VE-cadherin positive cells/myofibre and the average muscle fibre area. (n=5 per each group). Scale bars, 100 μm. (d) Immunoblotting of IL at 14 days after femoral ligation (n=5 per group) with quantitation of densitometry for MHC and myogenin. (e) Immunoblotting of IL at 28 days after femoral ligation (n=5 per group) with quantitation of densitometry for MHC and myogenin. (f) Immunostaining of IL for β-dystroglycan, Pax7 (blue) and Brdu (red) at 14 days after femoral ligation with quantitation of average number of Pax7 Brdu positive cells (white arrows) per 10 high power field on right (n=5 per each group). (g) Weight of NIL and IL after femoral ligation (n=5 per group). Comparisons between two groups were achieved using a two-sided student's t-test. All bars show mean±s.e.m. *P<0.05 versus WT.

Figure 2. Notch activation in CD163^−/− mice after ischaemic injury directs muscle regeneration.

(a) Immunoblotting of IL at 14 days after femoral ligation with quantitation of densitometry for NICD and DLL-4 (n=5 per group). (b) Immunoblotting of IL at 28 days after femoral ligation with quantitation of densitometry for NICD and DLL-4 (n=5 per group). (c) LASER doppler analysis of CD163^−/− mice 3, 7, 14, 21 and 28 days after femoral ligation with or without administration of Notch inhibitor DAPT (n=5 per group). (d) Immunostaining for VE-cadherin and β-dystroglycan (red) in CD163^−/− IL at 14 days after femoral ligation with or without DAPT administration. The graphs show the number of VE-cadherin positive cells/myofibre and the average muscle fibre area. Scale bars, 100 μm. (e) Immunostaining for β-dystroglycan (red) and TOTO-3 (blue) in CD163^−/− IL 14 days after femoral ligation with or without DAPT administration. Scale bars, 20 um. The graph shows quantitation of regenerating myotubes in CD163^−/− ILs with or without DAPT administration (n=5 per group). All bars show mean±s.e.m. *P<0.05 versus WT in a. versus CD163^−/− con in c,d,e. Comparisons between groups were achieved using a two-sided student's t-test.

Figure 3. Canonical NF-κB signalling mediates Notch activation and the response of CD163^−/− mice to HLI.

(a) Immunoblotting of IL 14 days after femoral ligation (n=5 per group) with graphs showing quantitation of densitometry for phosphorylated (p-p65)/total p65 (T-p65) and p52. (b) Immunoblotting of IL 28 days after femoral ligation (n=5 per group) with graphs showing quantitation of densitometry for phosphorylated (p-p65)/total p65 (T-p65) and p52. (c) Immunostaining for Mac-3 (red) of IL of WT and CD163^−/− mice 3 and 5 days after femoral ligation. Scale bars, 100 μm. The graph shows the timecourse of Mac-3 staining (n=5 per group). (d) Laser Doppler analysis of CD163^−/− mice 3, 7, 14, 21 and 28 days after femoral ligation with administration of scrambled control peptide (control) or NF-κB inhibitor NBD (n=5 per group). (e) Immunoblotting of CD163^−/− IL at 14 days after femoral ligation with administration of scrambled control peptide (control) or NF-κB inhibitor (NBD) (n=5 per group) with bar graphs showing quantitation of densitometry for phosphorylated/total p65, p52 and NICD on right. All bars show mean±s.e.m. *P<0.05 versus WT in a–c; versus CD163^−/− control in d,e. Comparisons between groups were achieved using a two-sided student's t-test.

Figure 4. TWEAK activates canonical NF-κB signalling and Notch activation and mediates the response of CD163^−/− mice to HLI.

(a) Analysis of plasma from WT and CD163^−/− mice before and after femoral ligation for TWEAK and sCD163 levels by ELISA (n=5 per group). (b) Laser doppler analysis of CD163^−/− mice 3, 7, 14, 21 and 28 days after femoral ligation administered an isotype control or TWEAK blocking antibody (n=5 per group). (c) Immunoblotting of CD163^−/− IL at 14 days after femoral ligation with administration of isotype control or TWEAK blocking antibody (n=5 per group) with graphs showing quantitation of densitometry for phosphorylated/total p65 and NICD below. (d) Immunostaining for VE-cadherin (green) and β-dystroglycan (red) in CD163^−/− limbs with 14 day administration of TWEAK (5 or 25 μg kg⁻¹ day⁻¹) or saline (control) (n=5 per group). Scale bars, 100 μm. The graphs show the number of VE-cadherin positive cells/myofibre and the average muscle fibre area. (e) Immunoblotting of CD163^−/− limbs with 14 day administration of TWEAK (5 or 25 μg kg⁻¹ day⁻¹) or saline (control) (n=5 per group) with graphs showing quantitation of densitometry for phosphorylated p65/total p65, p52 and NICD. All bars show mean±s.e.m. *P<0.05 versus WT in a; versus CD163^−/− control in b,c and e. P<0.05 versus other groups in d,e. Comparisons between groups were achieved using a two-sided student's t-test. For multiple group comparisons, we utilized a one-way ANOVA. If the variance ratio test (F-test) was significant, a more detailed post hoc analysis of differences between groups was made using a Tukey–Kramer honest significance difference test.

Figure 5. High-dose TWEAK activates canonical NF-κB and Notch signalling which stimulates cell proliferation in cultured endothelial and myoblast cells.

(a) Immunoblotting of MDEC with or without administration of TWEAK (10 or 150 ng ml⁻¹) (n=4 per group) for the indicated proteins. (b) Immunoblotting of C2C12 cells with or without administration of TWEAK (10 or 100 ng ml⁻¹) (n=4 per group) for the indicated proteins. (c) Immunoblotting of MDEC transfected with scramble or p65 siRNA and treated with nothing or TWEAK (150 ng ml⁻¹), (n=4 per group) for NICD. (d) Immunoblotting of C2C12 cells transfected with scramble or p65 siRNA and treated with nothing or TWEAK (100 ng ml⁻¹) (n=4 per group) for NICD. (e) BrdU proliferation assay of MDEC treated with 150 ng ml⁻¹ TWEAK (n=at least four per experiment repeated twice). (f) BrdU proliferation assay of C2C12 treated with 100 ng ml⁻¹ TWEAK (n=at least four per experimental group repeated twice). (g) BrdU proliferation assay of MDEC treated with TWEAK (150 ng ml⁻¹) in the presence of scramble siRNA or siRNAs against p65 or RBPJ (n=at least four per experimental group repeated twice). (h) BrdU proliferation assay of C2C12 treated with TWEAK (100 ng ml⁻¹) in the presence of scramble siRNA or siRNAs against p65 or RBPJ (n=at least four per experimental group repeated twice). All bars show mean±s.e.m. *P<0.05 versus other groups. Comparisons between groups were achieved using a two-sided student's t-test. For multiple group comparisons, we utilized a one-way ANOVA. If the variance ratio test (F-test) was significant, a more detailed post hoc analysis of differences between groups was made using a Tukey–Kramer honest significance difference test.

Figure 6. sCD163 inhibits the effect of TWEAK on tissue regeneration.

(a) TWEAK (100 ng ml⁻¹) was added to VEGFR1, Fn14 or CD163-coated wells. Bound protein was measured by ELISA using an antibody to detect TWEAK. (b) Fn14 or CD163 were coated onto 96-well plates and incubated with increasing doses of untagged TWEAK. Bound protein was measured by ELISA. (c) Immunoblotting of MDEC stimulated with TWEAK at 200 ng ml⁻¹ (which activates both canonical and non-canonical NF-κB signalling) and increasing doses of recombinant mouse sCD163 for the indicated proteins. (d) Immunoprecipitation of serum before and after femoral ligation using anti-TWEAK antibody with immunoblotting for CD163. (The figures show a representative blot with a total of four experiments conducted.) (e) Immunostaining for VE-cadherin and β-dystroglycan (red) in limbs of CD163^−/− mice treated for 14 days with saline (control), TWEAK administration (25 μg kg⁻¹ day⁻¹ administered via Alzet pump) alone or in combination with sCD163 administration (25 μg kg⁻¹ day⁻¹ administered by intraperitoneal injection) (n=5 per group). The graphs show the number of VE-cadherin positive cells per myofibre and the average muscle fibre area per group. (f) Immunblotting of CD163^−/− limb after 14 days of TWEAK administration (25 μg kg⁻¹ day⁻¹ administered via Alzet pump) alone or in combination with sCD163 administration (25 μg kg⁻¹ day⁻¹ administered by intraperitoneal injection) (n=5 per group). O.D., optical density. Bar graphs show quantitation of densitometry for phosphorylated to total p65, p52 and NICD. *P<0.05 versus other groups. For multiple group comparisons, we utilized a one-way ANOVA. If the variance ratio test (F-test) was significant, a more detailed post hoc analysis of differences between groups was made using a Tukey–Kramer honest significance difference test. For binding experiments, average of at least four well per group shown with each experiment repeated at least two times with representative results shown.

Figure 7. CD163-TWEAK interaction controls skeletal muscle recovery after ischaemic injury.

After ischaemic injury, macrophages secrete TWEAK which induces tissue canonical NF-KB/Notch mediated myogenic progenitor cell proliferation. TWEAK's biological activity is tightly regulated by sCD163 binding which limits serum TWEAK levels and leads to resolution of canonical NF-κB/Notch activation at the site of injury, allowing for cell differentiation. In the absence of sCD163 (as seen in CD163-deficient mice), TWEAK levels are systemically elevated and TWEAK's biological effect is unopposed leading to prolonged activation of NF-κB and Notch signalling which causes greater myogenic progenitor cell proliferation-induced myogenesis not limited to the site of injury.