Introduction
IgG4-related disease (IgG4-RD) is an autoimmune disease characterized by the infiltration of IgG4-positive plasma cells, obliterative phlebitis, and storiform fibrosis [
1]. Although the etiology of this disease remains largely unknown, recent studies have shed light on the significant roles of T- and B-lymphocytes [
2‐
5]. While there have been notable advances in understanding the involvement of immune cells in the pathology of IgG4-RD, limited insight presently exists regarding the implications of non-immune cells on IgG4-RD pathogenesis, as well as their distinctive characteristics [
6]. Notably, unlike other fibrotic diseases such as systemic sclerosis and idiopathic pulmonary fibrosis, the fibrotic changes observed in IgG4-RD are generally reversible, indicating the existence of distinct profibrotic machinery [
7].
One promising approach for elucidating the pathogenic cell subsets in various diseases is single-cell RNA sequencing (scRNA-seq). This relatively new technique allows for the exploration of novel cell subsets in the affected organs. In the context of IgG4-RD, scRNA-seq has been applied to analyze peripheral blood mononuclear cells, revealing the heterogeneity of monocytes, T-lymphocytes, and B-lymphocytes [
8]. However, only a few reports have utilized scRNA-seq to analyze non-immune cells in the affected tissues [
9], which could potentially provide crucial insights into the pathogenesis of IgG4-RD.
In this study, we conducted scRNA-seq analysis of submandibular gland tissues from patients with IgG4-RD. Our findings unveil the presence of a disease-related cell subset, namely collagen type XV-expressing fibroblasts, which may have distinct functions. Furthermore, we demonstrate that serum levels of collagen type XV could serve as a monitoring tool for IgG4-RD patients.
Methods
Patients
IgG4-RD patients who fulfilled the 2019 American College of Rheumatology/European League Against Rheumatism Classification Criteria for IgG4-Related Disease [
10] were enrolled in this study. As controls for scRNA-seq analyses, patients with oropharyngeal cancer who underwent neck dissection surgery were also included. Healthy volunteers were recruited as controls for serum collagen type XV assessment. Patients with active disease were defined as those who have received systemic immunosuppressive therapy due to severe glandular symptoms leading to a decrease in vision or exhibit visceral lesions. Written informed consent was obtained from all participants in accordance with the Declaration of Helsinki. Tables
1 and
2 provide an overview of the demographic characteristics of the participants. Supplementary Table
S1 provides clinical manifestation of patients and control subject.
Table 1
Clinical features of IgG4-RD patients and control subject analyzed using scRNA-seq and IHC
Age, median (IQR), years | 62 (55.75–74) | 60 (57.5–61) |
Sex, female (%) | 0 | 0 |
number of involved organs, median (IQR) | 2 (1.75–2) | N.D. |
*Serum IgG4, median (IQR), mg/dL | 1333 (925–1700) | N.D. |
**Serum IgG, median (IQR), mg/dL | 2519 (2390–3058) | N.D. |
IgG4 responder index, median (IQR) | 8 (8–8.75) | N.D. |
Table 2
Clinical features of IgG4-RD patients and control participants subjected to serum collagen type XV measurement
Age, median (IQR), years | 69 (59–73) | 69 (66–69) | 65 (64–69.25) |
Sex, female (%) | 40 | 20 | 20 |
number of involved organs, median (IQR) | N.D. | 1 (1–2) | 2 (1.25–2) |
*Serum IgG4, median (IQR), mg/dL | N.D. | 397 (220–480) | 1073 (437.5–1240) |
*Serum IgG4 3 months after treatment, median (IQR), mg/dL | N.D. | N.D. | 173 (93.5–252.75) |
IgG4-RD responder index before treatment, median (IQR) | N.D. | 0 (0–1) | 11.5 (8.5–13.5) |
IgG4-RD responder index 3 months after treatment, median (IQR) | N.D. | N.D. | 1 (0.25–1.75) |
Submandibular gland preparation
Human submandibular gland samples were placed in a C tube (Miltenyi Biotec, North Rhine-Westphalia, German) containing digestion buffer (Multi Tissue Dissociation Kit 1, Miltenyi Biotec), and tissues were dissociated by a gentleMACS™ octo dissociator with heaters (Miltenyi Biotec). Submandibular gland cells in cell suspension were separated using Percoll solution (GE Healthcare, Chicago, IL). Subsequently, cells were stained with anti-human CD45 PE (BD biosciences, Franklin Lakes, NJ) antibodies. Non-immune cells (CD45-) were purified from stained cells using a BD Aria III (BD).
scRNA-seq
Cells from submandibular glands were encapsulated into droplets, and libraries were prepared using Chromium Single Cell 3.1’ Reagent Kits v3 according to the manufacturer’s protocol (10X Genomics, Pleasanton, CA). The generated scRNA-seq libraries were sequenced with a NovaSeq 6000 (Illumina, San Diego, CA). Sequence reads from all samples were processed and aggregated using a Cell Ranger (version 6.0) (10X Genomics). Aggregated data were further analyzed by Seurat [
11]. Specifically, we first removed sex chromosome-related genes from the analyses because of the imbalance for the participant’s sex. We applied SoupX to remove ambient RNA from the data. After this procedure, we log-normalized the expression matrix and regressed the data against the total number of unique molecular identifiers (UMIs) detected per cell. For each sample, poor-quality cells were filtered out using the following criteria: (1) the percentage of UMIs derived from mitochondrial genes above 10%, and (2) the number of detected genes below 700. Cells expressing hematopoietic markers (
PTPRC, CD3E, TRAC, TRDC, NKG7, GZMA, CD79A, IGHG1, IGHG2, IGHG3, IGHG4, IGHM, IGHE, MS4A1, PF4, FCGR3A, MS4A7, FCER3A, and
ITGAX) were also excluded from the analysis. Subsequently, the data were subjected to principal component analysis (PCA), and we used PCA dimensions 1–15 to find clusters on a Uniform Manifold Approximation and Projection (UMAP) analysis. The marker genes of each cluster and differentially expressed genes between samples were identified by FindMarkers in Seurat. Gene set enrichment analysis (GSEA) was performed on differentially expressed genes using clusterProfiler (v4.8.1) [
12].
Immunohistochemistry
Collagen type I expression was determined as an indicator for fibrosis [
13], and collagen type XV was identified as a disease-specific molecule. Fresh frozen submandibular gland samples were subjected to immunohistochemical analysis. The slides were thawed and washed in PBS, followed by blocking with 5% goat serum at room temperature for 1 h in a humid chamber. Subsequently, the slides were treated with the following primary antibodies: anti-collagen I alpha 1 antibody (1:1000, COL-1, Novus Biological, Centennial, CO) and anti-COL15A1 antibody (1:200, polyclonal, Merck, Rahway, NJ). The slides were then incubated overnight at 4 ºC. Following overnight incubation, the slides were washed and incubated with the appropriate secondary antibodies at room temperature for 1 h in the dark. The secondary antibodies used were as follows: Alpaca anti-mouse IgG1 antibody conjugated with Alexa Fluor 488 (1:250, Thermo Fisher Scientific) and goat anti-rabbit IgG antibody conjugated with Alexa Fluor 647 (1:200, Thermo Fisher Scientific). Nuclear DNA was counterstained with DAPI and visualized using an EVOS M7000 (Thermo Fisher Scientific). Collagen type XV-positive cells were defined as being absent in the basement membrane region, having an isolated nucleus, and exhibiting collagen type XV staining in the plasma membrane/cytoplasm. The number of collagen type XV-positive cells between the control and IgG4-RD groups was compared by analyzing five randomly selected fields of view captured using a 20x lens.
Enzyme-linked immunosorbent assay (ELISA)
Serum collagen type XV levels were determined by using a Human Collagen Type XV ELISA kit (Elabscience, cat# E-EL-H0772, Houston, TX) according to the manufacturer’s protocol.
Statistical analyses
Statistical analyses were performed using GraphPad Prism 10 (GraphPad, San Diego, CA). The number of collagen type XV-positive cell were analyzed using an unpaired t-test. Serum levels of type XV collagen were analyzed using One-way ANOVA test followed by Kruskal-Wallis test and paired t-test. P values less than 0.05 were considered statistically significant.
Data availability
RNA sequence data are submitted to the Gene Expression Omnibus (GEO). The data is available under the accession number GSE242778.
Discussion
Fibroblasts are essential cells involved in a range of biological processes, with their prominent role being the production of extracellular matrix. In this study, we identified a specific subset of fibroblast in IgG4-RD patients characterized by the expression of collagen type XV. Notably, collagen type XV-positive fibroblasts exhibited distinctive gene expression patterns. This discovery underscores the significance of comprehensively examining the unique properties and functions of collagen type XV-positive fibroblasts within the context of IgG4-RD.
Genome-wide association study (GWAS) revealed that
HLA-DRB1 is a disease susceptibility gene for IgG4-RD [
17]. This observation indicates an interaction between CD4-positive T cells and antigen-presenting cells in the pathogenesis of IgG4-RD. Additionally, successful treatment with rituximab, aimed at removing B cells, indicates that the antibodies or B cells themselves are pathogenic [
4,
18]. These studies suggest potential abnormalities in the acquired immune system and the presence of autoantigens in IgG4-RD. Interestingly, several reports have documented the deposition of IgG4 on the basement membrane of the pancreatic and bile duct [
19,
20], indicating an immune response specifically targeted against a component of the basement membrane. As mentioned earlier, collagen type XV is mainly expressed on the basement membrane. Notably, collagen type XV expression in fibroblasts found in tissues from patients with IgG4-RD is significantly higher than in the basement membrane in healthy subjects. If collagen type XV is targeted by autoantibodies, it is possible that collagen type XV-positive fibroblasts are implicated in the pathogenesis of the disease through the production of autoantigens. In this context, the quantification of antibodies to collagen type XV in the bloodstream would be of significant interest.
Collagen type XV is known to be associated with pro-inflammatory responses. In a murine kidney injury model, collagen type XV and type XVIII mutant mice exhibit an attenuated influx of leukocytes in the ischemic kidney, suggesting that these collagens are crucial for the migration of leukocytes into the injured tissues [
21]. It has also been shown that collagen type XV – integrin b1 interaction induces focal adhesion kinase (FAK) activation [
22], resulting in endoplasmic reticulum (ER) stress-mediated inflammatory macrophage polarization [
23]. Therefore, it is plausible that collagen type XV could promote tissue inflammation through the recruitment and activation of immune cells. To explore this further, it would be interesting to analyze immune cells infiltrated in the submandibular gland in IgG4-RD patients.
APOD, which we have identified as a characteristic gene of
COL15A1-positive fibroblasts, is well known to be a component of high-density lipoprotein (HDL) [
24]. The expression of ApoD in skin fibroblasts has been reported to be linked with cellular aging [
15]. The process of skin tissue aging is believed to be influenced, in part, by an imbalance between extracellular matrix production and degradation [
25]. Furthermore, tissue fibrosis is also characterized by abnormal matrix deposition and reduced degradation [
26]. Considering this, the elevated expression of ApoD in collagen type XV-positive fibroblasts may be associated with fibrosis, as it may disrupt the extracellular matrix homeostasis within the tissue.
Interestingly, fibroblasts positive for
COL15A1 exhibited lower expression of immune-related genes such as
HLA-DRA,
THY1, and
CXCR13 (Fig.
1H). Previous studies have demonstrated that certain fibroblasts express MHC class II and possess the capability of antigen presentation [
27‐
29]. Additionally, we observed a reduction in the expression of
CXCL13 in
COL15A1-positive fibroblasts. It is worth noting that CXCL13-producing fibroblasts have been recognized as the crucial cells responsible for the formation of lymph follicles [
30]. These findings suggest that fibroblasts expressing collagen type XV may not have a direct impact on the formation of lymphocyte aggregates observed in patients with IgG4-RD. It is evident that multiple phenotypes of fibroblasts exist, including those associated with tissue fibrosis pathology and those contributing to the immune response. Considering the absence of prominent features observed in the pathway analysis (Supplementary Fig. 2), it is possible that collagen type XV-expressing fibroblasts may not constitute an independent fibroblast subset but rather represent a condition associated with IgG4-RD. Further investigation is necessary for precise fibroblast classification and their functional analysis in IgG4-RD.
This study has several limitations. Firstly, the sample size of the enrolled participants was small. While IgG4-RD is considered a relatively rare disease, larger cohort studies are necessary to validate the presence and functional significance of collagen type XV-positive fibroblasts in IgG4-RD. Increasing the sample size would enhance the statistical power and reliability of the findings. Secondly, we focused exclusively on the analysis of non-hematopoietic cells and did not perform scRNA-seq for hematopoietic cells. Considering the emerging importance of cell-cell communication in disease pathology, it would be valuable to conduct scRNA-seq analysis for both hematopoietic and non-hematopoietic cells in the affected organs. Such an approach could provide comprehensive insights into the complex interactions between different cell types and potentially reveal additional factors contributing to the pathogenesis of IgG4-RD. Thirdly, there are some differences in the phenotype of submandibular fibroblasts recently reported, which were also characterized using scRNA-seq [
9]. This discrepancy could potentially arise from the heterogeneity of IgG4-RD pathology and variations in the backgrounds of the control patients employed. Additional insights are anticipated to emerge as more tissue scRNA-seq data are accumulated from individuals affected by IgG4-RD.
Although many questions remain, this study offers valuable insights into the roles of non-hematopoietic cells in IgG4-RD. The identification of collagen type XV-positive fibroblasts as a specific fibroblast population associated with IgG4-RD opens up new avenues for potential therapeutic targeting in patients with IgG4-RD.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.