Mer tyrosine kinase as a possible link between resolution of inflammation and tissue fibrosis in IgG4-related disease

Abstract

Objectives

IgG4-related disease (IgG4-RD) is a systemic fibro-inflammatory disorder characterized by a dysregulated resolution of inflammation and wound healing response that might develop after an apoptotic insult induced by cytotoxic T lymphocytes (CTLs). Mer receptor tyrosine kinase (MerTK) and its ligand, protein S (ProS1), have a pivotal role in the resolution of inflammation, being implicated in the clearance of apoptotic cells, quenching of the immune response and development of tissue fibrosis. In the present work we aimed to investigate a possible involvement of the MerTK signalling pathway in the pathogenesis of IgG4-RD and development of tissue fibrosis.

Methods

MerTK and ProS1 expression patterns in IgG4-RD lesions were evaluated by immunohistochemistry and immunofluorescence studies. Circulating MerTK⁺ monocytes, soluble Mer and MerTK ligands were measured in the peripheral blood of IgG4-RD patients and healthy controls by flow cytometry and ELISA, respectively.

Results

MerTK was highly expressed by macrophages infiltrating IgG4-RD lesions. MerTK⁺ macrophages were more abundant in IgG4-RD than in Sjögren’s syndrome and interacted with apoptotic cells and ProS1-expressing T and B lymphocytes. Moreover, they expressed the pro-fibrotic cytokine TGF-β and their numbers declined following rituximab-induced disease remission. Circulating MerTK⁺ monocytes, soluble Mer and MerTK ligands were not increased in the peripheral blood of patients with IgG4-RD.

Conclusions

The MerTK–ProS1 axis is activated in IgG4-RD lesions, possibly leading to persistent stimulation of processes involved in the resolution of inflammation and tissue fibrosis.

Rheumatology key messages

• MerTK⁺ macrophages infiltrate IgG4-RD lesions, interact with apoptotic cells and express the pro-fibrotic molecule TGF-β.

• IgG4⁺ and T cytotoxic lymphocytes express MerTK ligand ProS1 and interact with MerTK⁺ cells.

• Tissue infiltrate of MerTK⁺ macrophages in IgG4-RD lesions decreases following B-cell depletion with rituximab.

Introduction

Resolution of inflammation after inflammatory responses requires dampening of immune cell activation, clearance of apoptotic debris and reconstitution of vascular integrity [1]. In addition, modulation of fibroblast proliferation and collagen deposition is equally important to avoid detrimental tissue fibrosis, dysregulated wound healing and subsequent organ dysfunction [2].

The Mer receptor tyrosine kinase (MerTK) is a member of the TAM (Tyro3, Axl and Mertk) family and plays a central role in the wound healing process, being involved in the negative regulation of immune responses, in the phagocytosis of apoptotic cells (‘efferocytosis’) and in tissue fibrosis [3]. MerTK is expressed mainly on dendritic cells and on CD163⁺ ‘alternatively activated’ macrophages upon sensing of intracellular lipoprotein-derived sterols by their nuclear receptors, the transcription factors liver X receptor α and β [4, 5]. MerTK rapidly induces the production of IL-10 and TGF-β following binding to its ligands growth arrest specific factor 6 (Gas6) and protein S (ProS1) on apoptotic cells and on activated T lymphocytes. This leads to gradual quenching of the immune response and to collagen secretion by activated myofibroblasts [6, 7]. Shedding of the membrane-bound MerTK results in the release of a soluble Mer protein (sMer) that acts as a decoy receptor for MerTK ligands, thus preventing excessive stimulation of membrane-bound Mer and regulating apoptotic cell engulfment [3].

Given this crucial role at the crossroads between resolution of inflammation and tissue fibrosis, both impaired and increased MerTK activity have been associated with human diseases [8–10]. Mice lacking MerTK exhibit defective clearance of apoptotic cells, experience a breach in self-tolerance and develop a chronic inflammatory lupus-like phenotype [8, 9]. On the other hand, overexpression of MerTK in liver macrophages increases TGF-β secretion and accelerates liver fibrosis in a mouse model of non-alcoholic steatohepatitis [10].

IgG4-related disease (IgG4-RD), an increasingly recognized immune-mediated condition named for the characteristic infiltrate of IgG4⁺ plasma cells in affected organs, is characterized by the development of tumour-like masses composed of immune cells and fibrosis. IgG4-RD offers a disease state to study the resolution of inflammation and the potential roles of MerTK biology in the context of an immune-mediated fibrotic disease [11–14]. To this end, we have demonstrated the accumulation of cells undergoing apoptotic cell death in tissues affected by IgG4-RD, which associates with the clonal expansion of CD4⁺ and CD8⁺ T lymphocytes with cytotoxic features (CTLs) at the same sites of disease [15–17]. In addition, tissue fibrosis is a histological hallmark and a major determinant of organ damage in IgG4-RD but the link between IgG4⁺ plasma cells, CTLs, and fibroblast activation remains inadequately defined [18–20]. We hypothesized that MerTK⁺ macrophages actively participate in the immune response and fibrogenesis of IgG4-RD through efferocytosis and direct interaction with infiltrating B and T lymphocytes.

Methods

Study population

Serum and peripheral blood mononuclear cells (PBMCs) were obtained from 44 patients with IgG4-RD referred to the rheumatology clinics of San Raffaele Scientific Institute (Milan, Italy) or Massachusetts General Hospital (Boston, MA, USA) between September 2008 and August 2017. Serum and PBMCs from 30 sex- and age-matched healthy subjects recruited from the San Raffaele Blood Donor Centre and stored at the Institutional Biobank were included as experimental controls. Blood samples for the studies performed were drawn during active disease at the time of diagnosis before the institution of immunosuppressive treatment. Disease activity was defined according to the IgG4-RD Responder Index and active disease was considered for values >3 [21]. Submandibular gland (n = 14), pancreatic gland (n = 5), biliary tract (n = 3), skin (n = 1) and lung (n = 1) biopsies from 24 different patients with IgG4-RD from the pathology units of San Raffaele Scientific Institute, Massachusetts General Hospital and Kyushu University Hospital (Fukuoka, Japan) were used for immunohistochemistry and immunofluorescence studies. Submandibular glands from patients with Sjögren’s syndrome referred to the Kyushu University Hospital (Fukuoka, Japan) were used (n = 14) as controls for immunofluorescence studies. Pancreatic tissue from patients with ‘groove pancreatitis’ (n = 3) and a patients with ‘alcohol-induced chronic pancreatitis’ (n = 3) retrieved from the Pathology Unit of San Raffaele Hospital were used as controls for immunohistochemistry studies. IgG4-RD was diagnosed according to the Consensus Statement on the Pathology of IgG4-RD and retrospectively classified according to the 2019 American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) Classification Criteria for IgG4-RD [22, 23]. Sjögren’s syndrome was diagnosed based on the 2016 ACR/EULAR Classification Criteria for primary Sjögren’s syndrome [24], objective evidence of salivary gland involvement, subjective xerostomia, decreased salivary flow rate, abnormal findings on parotid sialography and focal lymphocytic infiltrates in labial salivary glands. This study was approved by the Institutional Review Board of San Raffaele Scientific Institute (Milan, Italy), of Massachusetts General Hospital and of Kyushu University Hospital (Fukuoka, Japan). Informed written consent was obtained from all participants referred to the rheumatology clinics of both hospitals and enrolled in the study.

Flow cytometry

PBMCs were isolated from IgG4-RD patients and healthy blood donors by Ficoll-Paque centrifugation as described in standard protocols, re-suspended in fetal bovine serum containing 10% dimethyl sulfoxide (DMSO) and cryopreserved in vapour phase liquid nitrogen until use. At the time of analysis, PBMCs were thawed, centrifuged at 500 g for 5 min, resuspended in 3 ml of pure medium and incubated for 30 min on ice in dark with 7.5 μl of Fc Receptor Blocking solution (Human TruStain FcX, Biolegend, San Diego, CA, USA). Cells were then washed two times and incubated at 4°C in the dark for 30 min in staining buffer (Cell Stain Buffer, Biolegend) containing optimized concentrations of fluorochrome-conjugated antibodies for detection of surface markers. All antibodies were obtained from Biolegend except where indicated. The following monoclonal antibodies were used for analysis: anti-human CD19-FITC (clone HIB19; dilution 1:100), anti-human CD3-FITC (clone UCHT1; dilution 1:100), antihuman CD56-FITC (clone HCD56; dilution 1:100), anti-human CD66b-FITC (clone G10F5; dilution 1:100), anti-human HLA-DR-APC/Cy7 (clone L243; dilution 1:100), antihuman CD14-Pacific Blue (clone M5E2; dilution 1:20), anti-human CD16-PE (clone 3G8; dilution 1:200), anti-human Mer-APC (clone 125518, R&D Systems, Minneapolis, MN, USA; dilution 1:10). Cells were washed two times and were resuspended in staining buffer, to which 7-amino-actinomycin D viability staining solution (Biolegend; dilution 1:100) was added before flow cytometry analysis. All samples were acquired on a BD LSR II flow cytometer (BD Biosciences, San Jose, CA, USA), using positive and negative beads (VersaComp antibody capture beads, Beckman Coulter, Brea, CA, USA) for compensation, and analysed with FlowJo software (BD Bioscience). Gating of monocyte subsets was performed in accordance with established literature. MerTK surface expression on monocytes was quantified both as a MerTK⁺ cell percentage and by means of median fluorescence intensity (MFI).

ELISA

Plasma concentrations of sMer, Gas6 and ProS1 were measured using quantitative ELISA for sMer (MER Human ELISA Kit, Abcam, Cambridge, MA, USA), Gas6 (Human Gas6 DuoSet ELISA, R&D Systems) and ProS1 (Asserachrom Total Protein S, Diagnostica Stago, Milano, Italy). The assays were calibrated using standards provided by the manufacturer.

Immunofluorescence and immunohistochemistry

For immunofluorescence studies, tissue samples were fixed in formalin, embedded in paraffin and sectioned. Specimens were incubated with primary antibodies anti-human MerTK (clone Y323, Abcam; dilution 1:800), CD68 (clone KP1, Abcam; dilution 1:100), CD163 (clone EPR19518, Abcam; dilution 1:2000), ProS1 (rabbit polyclonal, Sigma-Aldrich, St Louis, MO, USA; dilution 1:100), Gas6 (rabbit polyclonal, Abcam; dilution 1:800), CD4 (clone BC/1F6, Biocare Medical, Pacheco, CA, USA; dilution 1:50), signalling lymphocyte activation molecule family member 7 (SLAMF7; rabbit polyclonal, Sigma-Aldrich; dilution 1:2000), CD19 (clone CD19, Biocare Medical; dilution 1:200), IgG4 (clone EP4420, Abcam; dilution 1:2000), cleaved caspase-3 (clone ASP175-5A1E, Cell Signalling Technology, Danvers, MA, USA; dilution 1:400) followed by incubation with secondary antibody using a SuperPicTure Polymer Detection Kit (Thermo Fisher Scientific, Waltham, MA, USA) and an Opal 3-Plex Kit (fluorescein, Cyanine3, Cyanine5 and AlexaFlour750) (Perkin Elmer, Waltham, MA, USA). Slides were then mounted with Vectashield mounting medium containing 4′,6-diamidino-2-phenylindole (DAPI; Vector Laboratories, Burlingame, CA, USA). For immunohistochemistry studies, paraffin-embedded tissue slides from IgG4-RD patients and healthy controls were stained with human anti-MerTK monoclonal antibody (clone Y323, Abcam; dilution 1:700), CD68 (clone KP1, Cell Marque; prediluted), anti-CD163 (clone MRQ-26, Cell Marque, Rocklin, CA, USA; prediluted), or an anti-cleaved caspase-3 polyclonal antibody (clone ASP175-5A1E, Cell Signalling Technology; dilution 1:400).

Microscopy and quantitative image analysis

Images were acquired with the TissueFAXS platform (TissueGnostics, Wien, Austria). For quantitative analysis, the entire area of the tissue was acquired as digital greyscale images in five channels with filter settings for fluorescein, Cyanine3, Cyanine5, AlexaFlour750 and DAPI. Cells of a given phenotype were identified and quantitated using Tissue-Quest software (TissueGnostics), with cut-off values determined relative to the negative controls. This microscopy-based multicolour tissue cytometry software permits multicolour analysis of single cells within tissue sections similar to flow cytometry. The principle of the method and the algorithms used have been described in detail elsewhere [25]. Cell–cell contact was assessed by manually measuring the distance between the edges of two nearby nuclei belonging to the cells of interest. A nuclear distance of <0.5 µm was considered indicative of cell–cell interactions according to established literature [26].

Statistical analysis

Statistical analysis was performed using GraphPad Prism 8.0 (GraphPad Software, La Jolla, CA, USA). Normal distribution of continuous variables was assessed with the D’Agostino–Pearson normality test. Student’s unpaired t-test was used to perform comparisons between two groups of normally distributed variables. Non-normally distributed variables were compared using the Mann–Whitney U-test. A P-value <0.05 was considered significant. Continuous variables are expressed as median and interquartile range (IQR), unless otherwise specified.

Results

MerTK⁺ macrophages significantly infiltrate IgG4-RD lesions

To investigate the relevance of MerTK in IgG4-RD, we first quantified its expression at the tissue level by immunohistochemistry and multicolour immunofluorescence. MerTK staining was observed in different organs affected by IgG4-RD and largely overlapped with that of the macrophage lineage markers CD68 and CD163 (Fig. 1A) [27]. Specifically, up to 35% of all nucleated cells in IgG4-RD affected tissues expressed MerTK, 33% (IQR 26–41%) of MerTK⁺ cells co-expressed CD68 and CD163, and 30% (IQR 19–41%) expressed CD68 but not CD163. MerTK⁺ cells were equally distributed between CD68⁺CD163⁻ (median 38%, IQR 27–49%) and CD68⁺CD163⁺ (median 44%, IQR 8–66%) macrophage subsets (P = 0.3). A minor proportion of CD68⁻MerTK⁺ cells was also observed, likely representing mesenchymal and/or dendritic cells (Fig. 1B and C) [4]. To further appraise the involvement of MerTK⁺ macrophages in IgG4-RD pathogenesis, we next assessed MerTK expression on CD68⁺ and CD163⁺ cells in salivary glands from patients with Sjögren’s syndrome, a mimicker of IgG4-related sialoadenitis featuring an abundant lymphocytic infiltrate with less prominent tissue fibrosis (Fig. 1D). As shown in Fig. 1E, the proportion of MerTK⁺ macrophages and the total number of CD68⁺ and MerTK⁺ cells were significantly higher in salivary gland tissues affected by IgG4-RD compared with Sjögren’s syndrome (P < 0.0001 for all comparisons). Of note, the ratio of CD68⁺CD163⁺ to CD68⁺CD163⁻ cells did not differ between IgG4-related sialoadenitis and Sjögren’s syndrome (P > 0.05), suggesting a similar distribution of macrophage subsets in the two conditions. Altogether, these findings indicate that MerTK⁺ macrophages preferentially infiltrate organs affected by IgG4-RD and that MerTK expression identifies a macrophage population of potential pathogenic relevance to IgG4-RD.

Fig. 1.

MerTK⁺ macrophages are abundant in IgG4-RD tissues

(A) Immunohistochemical evaluation of MerTK, CD68 and CD163 expression performed on surgical samples of IgG4-RD of the lung and pancreas showing an overlapping distribution of these cell types in tissues. MerTK⁺ macrophages appear to be located mainly within the inflammatory infiltrate surrounding glandular and ductal structures. (B) Multicolour immunofluorescence for CD68 (orange), CD163 (green), MerTK (red) and DAPI (blue) of an IgG4-RD submandibular gland biopsy showing co-localization of CD68 and MerTK, but not CD163 (upper left quadrant), and of CD68, CD163 and MerTK (upper right quadrant). (C) TissueQuest quantification of MerTK, CD68 and CD163 expression in salivary glands (n = 14; G1–G14), pancreas (n = 1; AIP) and skin (n = 1) of IgG4-RD patients. (D) Immunofluorescence study for CD68 (orange), CD163 (green), MerTK (red) and DAPI (blue) of a representative minor salivary gland biopsy from a patient with IgG4-RD and with Sjögren’s syndrome (SS). (E) TissueQuest quantification of CD68⁺ and MerTK⁺ cells/mm² and of the proportion of CD68⁺MerTK⁺ and CD68⁺CD163⁺ out of the total number of CD68⁺ cells in IgG4-RD (n = 14) and SS biopsies (n = 14). DAPI: 4′,6-diamidino-2-phenylindole; AIP: autoimmune pancreatits; IgG4-RD: IgG4-related disease; MerTK: Mer receptor tyrosine kinase.

MerTK⁺ macrophages in IgG4-RD lesions interact with apoptotic cells

MerTK⁺ macrophages are implicated in the phagocytosis of apoptotic cells, a process termed efferocytosis [28]. Given the observed accumulation of MerTK⁺ macrophages in IgG4-RD tissues, we next performed immunohistochemistry studies for cleaved caspase-3 to assess tissue apoptosis in tissues affected by IgG4-RD [29]. As shown in Fig. 2, cleaved caspase-3⁺ cells were consistently observed in the lung, pancreas, biliary tract and salivary glands of IgG4-RD patients, suggesting that apoptosis represents a characteristic feature of IgG4-RD pathology. Moreover, apoptotic phenomena were found to be more frequent in IgG4-RD tissues compared with control tissues from other causes of chronic pancreatitis (Fig. 2B). The distribution of MerTK staining largely overlapped with that of cleaved caspase-3 with a clear correspondence around the ducts in both pancreatic and biliary specimens, reinforcing the notion that MerTK⁺ cells are implicated in the clearance of apoptotic debris at disease sites (Fig. 2A and B).

Fig. 2.

Apoptotic cells accumulate in IgG4-RD tissues and interact with MerTK⁺ macrophages via ProS1

(A) Immunohistochemical evaluation of MerTK and cleaved caspase-3 expression performed on surgical samples of IgG4-RD of the lung and biliary tree showing a similar spatial distribution of MerTK⁺ macrophages and apoptotic cells. (B) Immunohistochemical staining for cleaved caspase-3 on surgical samples from representative cases of IgG4-related chronic pancreatitis (n = 3), groove pancreatitis (n = 3) and alcohol-induced chronic pancreatitis (n = 3) showing an increased number of apoptotic phenomena in IgG4-RD with a periductal distribution. (C) Multicolour immunofluorescence study for MerTK (magenta), cleaved caspase-3 (green), ProS1 (red) and DAPI (blue) expression in IgG4-RD salivary gland biopsies (n = 6), showing contacts between MerTK⁺ scavenger macrophages and apoptotic cells co-expressing ProS1 and cleaved caspase-3. (D) TissueQuest quantification of the proportion of caspase-3⁺ cells expressing ProS1 in IgG4-RD salivary gland biopsies (n = 6). DAPI: 4′,6-diamidino-2-phenylindole; IgG4-RD: IgG4-related disease; MerTK: Mer receptor tyrosine kinase; ProS1: protein S.

To confirm the interaction between MerTK⁺ macrophages and apoptotic cells in IgG4-RD lesions, we performed multicolour immunofluorescence for MerTK and cleaved caspase-3 in salivary gland samples (Fig. 2C). We also studied ProS1 and Gas6, the ligands of MerTK that mediate efferocytosis by binding to phosphatidylserine on apoptotic cells [3]. Gas6 was not detected (data not shown). In contrast, diffuse ProS1 expression was observed in IgG4-RD tissues (Fig. 2C). In particular, 60% (IQR 37–86%) of cleaved caspase-3⁺ cells expressed ProS1 and most of them were in close contact with MerTK⁺ macrophages, suggesting ongoing interactions between scavenger macrophages and apoptotic cells (Fig. 2C and D).

IgG4⁺ plasma cells and CD4⁺ CTLs express ProS1 in IgG4-RD lesions

As shown in Fig. 3A, only 4% (IQR 1–7%) of ProS1⁺ cells expressed the apoptotic marker cleaved caspase-3. Since MerTK⁺ cells were mainly observed within the inflammatory infiltrate in IgG4-RD lesions and ProS1 is known to be expressed by activated lymphocytes, we investigated ProS1 expression on T and B lymphocytes in affected tissues [30]. As shown in Fig. 3A and B, among ProS1⁺ cells, 36% (IQR 30–55%) were CD19⁺ B lymphocytes, and 41% (IQR 27–44%) were CD3⁺ T lymphocytes, indicating that T and B lymphocytes represented the vast majority of ProS1 expressing cell types in IgG4-RD lesions. In particular, ProS1 was found on 75% (IQR 39–90%) of IgG4⁺ plasma cells and on 31% (IQR 20–35%) of CD4⁺ CTLs suggesting that these two lymphocyte sub-populations implicated in IgG4-RD pathogenesis can engage MerTK⁺ cells in affected tissues and, potentially, modulate their activation status through the MerTK–ProS1 axis (Fig. 3A and D). Indeed, physical interactions between ProS1 expressing cells and MerTK⁺ cells were observed in affected tissues as demonstrated by measurement of inter-nuclear distances (Fig. 3E).

Fig. 3.

MerTK ligand ProS1 is provided by apoptotic cells, activated B and T lymphocytes in IgG4-RD lesions

(A) TissueQuest quantification of immunofluorescence studies performed in (B–D) quantifying ProS1 expression in different cell types. Apoptotic cells accounts for a small percentage of ProS1 expressing cells (data derived from Fig. 2C). (B) Multicolour immunofluorescence study for CD3 (red), CD19 (green) and ProS1 (orange) of IgG4-RD salivary gland biopsies (n = 4) showing ProS1 expression on CD3⁺ T lymphocytes and CD19⁺ B lymphocytes. (C) Multicolour immunofluorescence study for CD19 (red), IgG4 (green), ProS1 (orange) and DAPI (blue) staining of IgG4-RD salivary gland biopsies (n = 4), showing ProS1 expression by CD19⁺ B lymphocytes and IgG4⁺ plasma cells. (D) Multicolour immunofluorescence study for CD4 (red), SLAMF7 (green), ProS1 (orange) and DAPI (blue) of IgG4-RD salivary gland biopsies (n = 4) showing ProS1 expression on CD4⁺SLAMF7⁺CTLs. (E) Multicolour immunofluorescence study for MerTK (magenta) and ProS1 (green) in a representative IgG4-RD salivary gland biopsy showing nuclear distances <0.5 µm between MerTK⁺ and ProS1⁺ cells indicating cell–cell contacts. CTL: cytotoxic T lymphocyte; DAPI: 4′,6-diamidino-2-phenylindole; IgG4-RD: IgG4-related disease; MerTK: Mer receptor tyrosine kinase; ProS1: protein S; SLAMF7, signalling lymphocyte activation molecule family member 7.

MerTK⁺ macrophages in IgG4-RD lesions express TGF-β and are reduced in number following B cell depletion therapy

MerTK signalling in macrophages induces TGF-β secretion, collagen synthesis by hepatic stellate cells and liver fibrosis [10]. To test the hypothesis that MerTK⁺ macrophages contribute to TGF-β secretion in IgG4-RD lesions, we performed multicolour immunofluorescence and found that 54% (IQR 40–65%) of MerTK⁺ cells expressed TGF-β (Fig. 4A). To gain further clues about the pathogenic relevance of MerTK⁺ macrophages in IgG4-RD, we took advantage of sequential biopsies from a patient with IgG4-related skin disease obtained before treatment and 4 weeks after rituximab infusion [31, 32]. The same skin biopsies were previously used to demonstrate a significant reduction of myofibroblast activation following B cell depletion therapy [19]. As shown in Fig. 4B, the macrophage infiltrate markedly decreased after treatment. In particular, a 56% reduction in the frequency of CD68⁺CD163⁻MerTK⁺ cells and an 83% reduction in the frequency of CD68⁺CD163⁺MerTK⁺ cells were observed compared with active, untreated disease. These results indicate that MerTK⁺ macrophages quantitatively express the pro-fibrotic cytokine TGF-β in tissues of patients with active, untreated IgG4-RD, and that clinical improvement after rituximab and reversal of fibrosis is accompanied by a decrease of the MerTK⁺ infiltrate.

Fig. 4.

MerTK⁺ macrophages express TGF-β and decrease after successful treatment with rituximab

(A) Immunofluorescence study for TGF-β (red), CD163 (green), MerTK (yellow) and DAPI (blue) of an IgG4-RD salivary gland biopsy, showing co-localization of TGF-β and MerTK staining, but not CD163. (B) TissueQuest quantification of immunofluorescence studies performed in (A) quantifying TGF-β expression in MerTK⁺ cells on IgG4-RD salivary gland biopsies (n = 4). (C) Multicolour immunofluorescence for CD68 (red), CD163 (green), MerTK (orange) and DAPI (blue) of a skin biopsy before and after rituximab treatment, showing a decline in the MerTK⁺ macrophages infiltrate after B-cell depletion therapy with rituximab. DAPI: 4′,6-diamidino-2-phenylindole; IgG4-RD: IgG4-related disease; MerTK: Mer receptor tyrosine kinase; RTX: rituximab.

MerTK⁺ monocytes and MerTK ligands are not increased in the peripheral blood of patients with IgG4-RD

To gain indirect evidence of MerTK–ProS1 axis activation in IgG4-RD, we measured the serum levels of MerTK ligands and decoy receptor, and assessed MerTK⁺ circulating monocyte subpopulations in the blood of 34 patients with active untreated IgG4-RD. As shown in Fig. 5A, the serum levels of sMer, ProS1 and Gas6 were similar between IgG4-RD patients and 20 age- and sex-matched healthy donors. The serum concentration of sMer, ProS1 and Gas6 did not correlate with biomarkers of IgG4-RD activity including serum IgG4, IgG4-RD Responder Index, or plasmablasts (data not shown). Similarly, we did not detect significant differences in MerTK expression on circulating monocyte subsets between IgG4-RD patients and healthy controls (Fig. 5B and C). Of note, MerTK was preferentially expressed by non-classical and intermediate monocytes in both IgG4-RD patients and healthy individuals (Fig. 5C).

Fig. 5.

MerTK soluble ligands and MerTK⁺ circulating monocytes are not increased in the peripheral blood of IgG4-RD patients

(A) Plasma levels of soluble Mer, ProS1 and Gas6 in active untreated IgG4-RD patients (n = 34) and age and sex-matched healthy controls (n = 20; HC). (B) Flow cytometry gating strategy used to determine MerTK expression on circulating monocyte subsets of one representative active IgG4-RD patients and one healthy control (HC). (C) MerTK expression, as defined by the median expression level (MFI) and the percentage of MerTK⁺ cells in healthy controls (n = 20; HC) and IgG4-RD patients (n = 20). Gas6: growth arrest specific factor 6; IgG4-RD: IgG4-related disease; MerTK: Mer receptor tyrosine kinase; ProS1: protein S; sMer: soluble Mer.

Discussion

Tissue fibrosis in IgG4-RD has been linked to the secretion of pro-fibrotic cytokines by antigen specific T and B lymphocytes that presumably drive organ damage [12, 15, 18–20]. This model, however, does not fit well with the most recent view of fibrogenesis as a physiological phenomenon that occurs during resolution of inflammation in concert with the negative regulation of immune responses and with the removal of apoptotic cells [2]. Indeed our previous studies have revealed a likely pathophysiological role for cytotoxic T cell mediated apoptosis of tissue cells in two fibrotic diseases, IgG4-RD and scleroderma [17, 33, 34]. The production of IgG4 antibodies—a pathological hallmark of IgG4-RD—may reflect another host anti-inflammatory strategy aimed at restoring tissue homeostasis, as IgG4 molecules typically dampen rather than incite immune activation, but the role of this molecule in the context of IgG4-RD remains unproven [35]. The hypothesis that mechanisms orchestrating the resolution of inflammation might be altered in IgG4-RD and lead to excessive tissue fibrosis has not been previously addressed.

In the present work we first identify MerTK⁺ macrophages as a novel potential contributor in the physiopathology of IgG4-RD and highlight the role of the MerTK–ProS1 axis as a possible link between the resolution of inflammation and tissue fibrosis in this fibro-inflammatory condition [10, 36–39]. By taking advantage of multicolour immunofluorescence, we demonstrate that MerTK⁺ macrophages are not merely a consequence of local sustained inflammation but are inherently associated with disease pathogenesis since they do not infiltrate inflamed glands of patients with Sjögren’s syndrome. The selective accumulation of the most important MerTK ligand, ProS1, on apoptotic cells and on activated lymphocytes in IgG4-RD lesions might be critical for the sustained local activation of MerTK⁺ macrophages. The latter, in fact, appear to physically interact with cells decorated with ProS1 and to be a source of the pro-fibrotic cytokine TGF-β, further reinforcing their possible causative role in the development of tissue fibrosis. Of note, the MerTK–ProS1 axis seems to be robustly activated primarily at the disease site, since we did not observe changes in MerTK⁺ monocytes or in sMer, ProS1 or Gas6 concentrations in the peripheral blood of patients with IgG4-RD. In light of these results, it is tempting to hypothesize an updated pathogenic model of IgG4-RD whereby activated B and T lymphocytes, as well as apoptotic cells, drive scavenger MerTK⁺ macrophages through ProS1 ligation to secrete pro-fibrotic molecules like TGFβ, resulting in tissue fibrosis (Fig. 6). Efferocytosis of apoptotic cells, in turn, could lead to the release of anti-inflammatory and pro-fibrotic cytokines aimed at resolving tissue inflammation [40, 41]. The binding of MerTK to ProS1 expressed on activated B and T lymphocytes infiltrating IgG4-RD lesions might further amplify this process and possibly provide a tonic or survival signal for MerTK⁺ macrophages. The swift decrease of MerTK⁺ cells that we observed after B cell depletion with rituximab might indeed have been mediated by the depletion of ProS1 expressing B and T cells [30, 42, 43].

Fig. 6.

Updated pathogenetic model of IgG4-RD

Apoptotic phenomena triggered by CD4⁺ CTLs activate scavenger MerTK⁺ macrophages through ProS1 ligation. Efferocytosis of apoptotic cells, in turn, leads to the release of anti-inflammatory and pro-fibrotic cytokines aimed at resolving tissue inflammation. Amplification of these processes might occur due to the high quantity of ProS1 that is either exposed on the cell membrane or directly secreted by the abundant B and T lymphocytic infiltrate. ProS1 expressing lymphocytes, in fact, can directly engage MerTK on macrophage, possibly leading to enhanced wound healing and tissue fibrosis. In this sense, rituximab might interfere with MerTK mediated resolution of inflammation through the removal of ProS1 expressing pathogenic B and T lymphocytes. CTL, cytotoxic T lymphocyte; IgG4-RD: IgG4-related disease; LOXL-2: lysyl oxidase like 2; MerTK: Mer receptor tyrosine kinase; Mϕ: macrophage; PDGF-B: platelet derived growth factor-B; ProS1: protein S.

Previous evidence based on immunohistochemistry and DNA-microarray analysis supported the activation of M2 macrophages as an initiating event in IgG4-RD pathogenesis and main determinant of fibroblast regulation [44–46]. Yet unknown exogenous or endogenous antigens were thought to trigger innate immunity by binding to Toll-like receptors on alternatively activated macrophages and by inducing the secretion of pro-fibrotic molecules such as IL-10, IL-33 and CCL-18 [47–48]. Macrophages from patients with IgG4-RD were also shown to induce IgG4 production by B lymphocytes upon non-specific in vitro stimulation of Toll-like receptors and exposure to B cell-activating factor [49].

Our data fit well with the relevance of macrophages for IgG4-RD pathogenesis, adding two unprecedented pieces of the puzzle. First, MerTK⁺ macrophages seem to link the previously established accumulation of activated T cells, B cells and apoptotic cells in lesional tissues of IgG4-RD with the development of tissue fibrosis. Based on the quantification of ProS1 expressing cells in IgG4-RD affected lesions, in fact, apoptotic cells and the characteristic lymphoplasmacytic infiltrate likely represent the most probable drivers of MerTK⁺ macrophage activation. Second, our results suggest that MerTK expression rather than conventional lineage markers of M1 and ‘alternatively activated’ M2 macrophages identifies a population of cells with actual pathogenic relevance for IgG4-RD. MerTK, in fact, was equally expressed on CD68⁺CD163⁺ and CD68⁺CD163⁻ cells in IgG4-RD lesions confirming that the widely accepted classification into M1 and M2 macrophages is probably not sufficient to functionally characterize these cells [50]. Of note, interactions between MerTK⁺ macrophages and apoptotic cells seen at tissue level are likely not specific of IgG4-RD as similar phenomena are typically observed during physiological germinal centre reactions where tingible body macrophages engulf apoptotic lymphocytes in primary and secondary lymphoid organs [3].

Our study has both limitations and strengths. As in any study relying on human tissues, we recognize that the descriptive nature of these data prevents definitive mechanistic conclusions and only allows association inferences. In particular, these data indicates that cleaved caspase-3⁺ cells are concentrated within the inflammatory infiltrate but do not clarify whether stromal, epithelial or immune cells are undergoing apoptosis [17]. Furthermore, we do not know to what extent the MerTK–ProS1 axis is implicated in the development of the storiform fibrosis observed in IgG4-RD tissues. In this regard, in the absence of available mouse models, co-cultures of MerTK expressing macrophages with ProS1 expressing B and T lymphocytes could unveil whether MerTK inhibition would impact the production of pro-fibrotic molecules and affect fibroblast activation. On the other hand, strengths include the adoption of human samples from three international referral centres for IgG4-RD, thus ensuring accurate patient selection, definitive diagnosis and robust confirmation of our results on a variety of organ manifestations from different study cohorts. We additionally leverage state-of-the-art tissue quantification using multi-colour immunofluorescence and cell–cell interaction measurements to help understand the nature and interactions of immune cells at the sites of disease in IgG4-RD.

In conclusion, this is the first study evaluating the resolution of inflammation in IgG4-RD. Resolution of inflammation is an integrated process that appears hyperactive in this fibrotic condition either because of increased apoptotic phenomena or due to an abundant lymphocytic infiltrate that upregulates the MerTK–ProS1 axis, sustaining the reparative action of MerTK⁺ macrophages. Although we provide quantitative evidence for considering MerTK activation in IgG4-RD pathogenesis as a possible link between lymphocyte activation and tissue fibrosis, further in vitro and in vivo mechanistic studies are warranted to confirm the relevance of our findings and to understand whether the MerTK pathway might be amenable to therapeutic intervention. Results of ongoing clinical trials with small molecules inhibiting the MerTK axis in solid and haematological tumours will soon profile the efficacy and safety of interfering with this pathway in vivo (https://www.clinicaltrials.gov/ct2/results?cond=&term=mertk).