Blister Fluid Induces MMP-9-Associated M2-Type Macrophages in Bullous Pemphigoid

Abstract

Bullous pemphigoid (BP) is a cutaneous autoimmune disease, characterized by an inflammatory cascade leading to blister formation. Although macrophages were shown to participate in BP pathophysiology, their role in the blister formation process still needs to be investigated. We here addressed the influence of serum and blister fluid (BF) from patients with BP on the polarization status of macrophages with regards to the metalloproteinase-9 (MMP-9) expression. We demonstrated that several markers related to the alternatively activated macrophage phenotype (M2) including IL-10, TARC, arginase, TNFα, and IL-1RA were meaningfully increased in BF of patients with BP. We further showed that BF, but not serum from patients with BP, significantly induced the expression of CD163, CD206, and IL-10 in BP monocyte-derived macrophages (MDMs). Notably IL-10 was the only cytokine to be correlated to the reference clinical score, BP disease activity index (BPDAI), especially to the inflammatory BPDAI subscore evaluating urticarial and erythematous skin lesions (r = 0.57, p = 0.0004). We also found elevated levels of MMP-9 to M2-type macrophages ex vivo and highlighted the presence of CD163⁺ MMP-9⁺ macrophages histologically, at skin lesional site. Finally, we showed that methylprednisolone reduced MMP-9 levels in MDMs without modifying the other M2 markers. All together these results strongly support the presence of M2-phenotype macrophages with pro-inflammatory properties susceptible to favor blister formation in BP.

Keywords: MMP-9; autoimmunity; bullous pemphigoid; inflammation; macrophage polarization.

Figure 1

Detection of macrophage-associated cytokines in serum and blister fluid (BF) of BP patients. The concentrations of a panel of M1-type (A) and M2-type (B) macrophage markers were measured in serum from control subjects (Control sera, n = 8) and in serum (BP sera, n = 12) and BF (n = 40) from patients with BP at time of diagnosis using the bead-based immunoassays LEGENDplex Human Macrophage/Microglia Panel and Human Th Cytokines Panel. Non-parametric unpaired Mann–Whitney's test was used to compare populations (^*p < 0.05; ^**p < 0.01; ^***p < 0.001).

Figure 2

Blister fluid (BF) from patients with BP affected the expression of M1/M2-type macrophage markers in THP-1 derived macrophages. M1-type (A) and M2-type (B) macrophage marker expressions were analyzed by real-time qPCR in THP-1 derived macrophage unstimulated (NS) or stimulated with BF (BF) from patients with BP. A paired T-Test was performed for statistics (^*p < 0.05; ^**p < 0.01; ^***p < 0.001).

Figure 3

Regulation of M1-type and M2-type macrophage marker expression in THP-1- and monocyte-derived macrophages. M1-type (A,C) and M2-type (B,D) macrophage marker expressions were analyzed by real-time qPCR in CTR or BP serum-stimulated macrophages originated either from PMA-derived THP-1 cells (A,B) or from CTR or BP monocytes differentiated with their autologous serum for 7 days (C,D). The error bars denote the mean ± SEM. Non-parametric unpaired Mann–Whitney's test was used to compare populations (^*p < 0.05; ^**p < 0.01; nd, not detected).

Figure 4

BF essentially increased the expression of M2-type macrophage markers in BP monocyte-derived macrophages. M1-type (A) and M2-type (B) macrophage marker expressions were analyzed by real-time qPCR in macrophages issued from BP monocytes differentiated for 7 days with the autologous serum collected at diagnosis (V1) and stimulated with BF for 24 h (V1 + BF). Non-parametric paired Wilcoxon's test was used to compare populations (ns, not significant; ^*p < 0.05; ^**p < 0.01).

Figure 5

The M2-type macrophage marker IL-10 was related to BP disease activity. (A) Correlation between levels of IL-10 and the M2-type macrophage marker Arginase in blister fluid of patients with BP at diagnosis was analyzed using the nonparametric Spearman's correlation test. (B–D) Correlations between total BPDAI score (B), blister/erosion (C), and erythema (D) BPDAI subscores and IL-10 levels measured in BF were analyzed using the non-parametric Spearman's correlation test (n = 35).

Figure 6

Polarization of BP-MdM influenced their capacity to produce the protease MMP-9. MMP-9 secretion (A) and MMP-9 mRNA expression (B) were analyzed, respectively, by gelatin zymography and real-time qPCR in M0, M1 (+IFNγ), and M2 (+IL-10) THP-1 derived macrophages. The error bars denote the mean ± SEM. Paired Student's T-Test was used for statistics (n = 3; ^*p < 0.05; ^**p < 0.01). (C–E) MMP-9 mRNA expression was analyzed by real-time qPCR in both THP-derived (C) and BP monocyte-derived macrophages (D,E) stimulated with either control serum, BP serum or blister fluid from BP patient. The error bars denote the mean ± SEM. Nonparametric unpaired Mann-Whitney's test was used for statistical analysis of (C,D) (n = 8 and n = 5, respectively). Nonparametric paired Wilcoxon's test was used to statically analyze (E) (n = 8; ^*p < 0.05). (F,G) Punch biopsies of lesional skin from patients with BP were subjected to CD163 immuno-staining alone (F) or double immunofluorescence staining for MMP-9 (red) and CD163 (green) (G) with Hoescht counterstain (blue). Negative CTR, negative control where primary antibodies were not added. White islets show lower magnification images of blister area and orange line designed the dermo-epidermal junction. The boxes in white islets match to the area where the high magnification image has been taken.

Figure 7

Treatment by methylprednisolone decreased MMP-9 expression and favored M2-type macrophage markers. MMP-9 (A) and M2-type macrophage markers CD163 (B) and CD206 (C) mRNA expressions were analyzed by real-time qPCR in BP MdM treated (MP) or not (NS) with 10 μM methylprednisolone. The error bars denote the mean ± SEM. Nonparametric paired Wilcoxon's test was used (^*p < 0.05).

Figure 8

BF-induced M2-type macrophages in the auto-inflammatory response associated to BP. Bullous pemphigoid is characterized by the production of autoantibodies against two proteins of the hemidesmosome structure, BP180 and BP230. The binding of BP autoantibodies ① onto their target initiates the activation of the complement pathway ② which then provokes the release of chemokines and cytokines by mast cells ③. These pro-inflammatory molecules trigger subsequent recruitment of inflammatory cells at the dermal-epidermal junction ④. Inflammatory key mediators secreted by these latter cells induce an overexpression of proteases, such as the matrix-metalloproteinases-9 (MMP9) and the neutrophil elastase, which are involved in extracellular matrix (ECM) degradation and blister formation ⑤. We here further demonstrated the ability of blister fluid to induce M2-type macrophages at site of skin lesion by increasing CD163, CD206, and IL-10 expression. BF-induced M2-type macrophages also show pro-inflammatory function by releasing MMP-9 and then participating to pathophysiological process of BP. Under corticotherapy the auto-amplification process associated to pro-inflammatory cell interaction is dismantled. Indeed, steps ③, ④, and ⑤ are inhibited. In addition, we showed that inhibition of MMP-9 production by methylprednisolone was not associated with a switch toward a M1 macrophage phenotype, but instead reinforced their M2-type phenotype promoting tissue regeneration and repair.