Single‐cell atlas of B cell heterogeneity in lacrimal IgG4‐related disorders: Unraveling the transition from inflammatory exhaustion to clonal malignancy

Rui Liu; Tingting Ren; Liangyuan Xu; Nan Wang; Yufei Zhang; Haihan Yan; Taiyao Wang; Runzi Yang; Jianmin Ma; Jing Li

doi:10.1111/febs.70498

. 2026 Mar 19;293(16):4992–5007. doi: 10.1111/febs.70498

Single‐cell atlas of B cell heterogeneity in lacrimal IgG4‐related disorders: Unraveling the transition from inflammatory exhaustion to clonal malignancy

Rui Liu ¹, Tingting Ren ¹, Liangyuan Xu ¹, Nan Wang ¹, Yufei Zhang ¹, Haihan Yan ¹, Taiyao Wang ¹, Runzi Yang ¹, Jianmin Ma ^1,^✉, Jing Li ^1,^✉

PMCID: PMC13489529 PMID: 41853890

Abstract

This study analyzed B‐cell heterogeneity in lacrimal gland tissues from 4 patients with IgG4‐related ophthalmic disease (IgG4‐ROD) and 2 patients with IgG4‐positive MALT lymphoma using single‐cell transcriptome sequencing (scRNA‐seq). The results revealed that while the B‐cell differentiation trajectories have similarities between the two diseases, critical differences were evident. IgG4‐ROD was predominantly composed of naïve B cells, with memory B cells mainly being the central memory type. These cells were enriched in hormone/innate immunity pathways, and the plasma cells exhibited features of functional exhaustion. Conversely, MALT lymphoma was dominated by memory B cells, particularly enriched in terminally differentiated subtypes, with aberrant activation of oncogenic pathways (BCR/NF‐κB). In MALT lymphoma, naïve B cells showed upregulated expression of immunoglobulin genes (IGHG3/IGHG4) and abnormal activation of EBV/BCR/T‐cell differentiation pathways, despite having suppressed basal metabolism. Germinal center‐like B cells in MALT lymphoma revealed upregulated gene expression enriched in T‐cell activation and PD‐1/PD‐L1 pathways. Plasma cells in MALT lymphoma displayed monoclonal expansion (high expression of IGHG2/IGHG3/IGHG4) and enhanced antibody secretion. Therefore, MALT lymphoma is characterized by abnormal activation and malignant transformation of B cells, whereas IgG4‐ROD manifests as functional exhaustion and metabolic suppression. The differentiation state of memory B cells may represent a critical juncture for malignant transformation.

Keywords: B‐cell heterogeneity, IgG4‐related ophthalmic disease, Immune microenvironment, mucosa‐associated lymphoid tissue lymphoma, single‐cell transcriptome sequencing

Single‐cell RNA sequencing reveals how benign inflammatory eye disease (IgG4‐ROD) can transform into B‐cell lymphoma. Driven by aberrant signaling pathways, B cells undergo a critical transition from functional exhaustion in inflammation to malignant activation in cancer. This provides new insights into the cellular mechanisms of disease progression and potential therapeutic targets.

graphic file with name FEBS-293-4992-g003.webp

Abbreviations

BCR: B cell receptor
DEGs: Differentially expressed genes
EBV: Epstein–Barr virus
ER: endoplasmic reticulum
GC_B: Germinal center B
GC_DZ: GC Dark Zone
GC_LZ: GC Light Zone
GO: Gene Ontology
IgG4‐ROD: immunoglobulin G4‐related ophthalmic disease
KEGG: Kyoto Encyclopedia of Genes and Genomes
MALT: mucosa‐associated lymphoid tissue
Mem B1: Memory B cell 1
Mem B2: Memory B cell 2
PBS: phosphate‐buffered saline
QC: quality control
scRNA‐seq: single‐cell RNA sequencing
TCR: T cell receptor
t‐SNE: t‐distributed Stochastic Neighbor Embedding

Introduction

Immunoglobulin G4‐related ophthalmic disease (IgG4‐ROD) is a fibroinflammatory disorder exclusively affecting ocular tissues. As clinical awareness of IgG4‐ROD grows globally, evidence indicates that although primarily inflammatory, it can progress to lymphoma with diverse pathological subtypes. The most common is mucosa‐associated lymphoid tissue (MALT) lymphoma, followed by diffuse large B‐cell lymphoma and follicular lymphoma [1]. Studies report malignant transformation rates of 10.4–14.3% in IgG4‐ROD patients, accounting for approximately 9% of ocular adnexal lymphomas [2, 3]. The precise etiology and mechanisms underlying lymphomagenesis in IgG4‐ROD remain unclear, posing diagnostic challenges in distinguishing it from lymphoma. Research suggests malignant transformation may involve chronic antigenic stimuli driving lymphoid hyperplasia, where persistent reactive lymphoid follicles create a permissive microenvironment for neoplastic evolution [4]. Asakage et al. using high‐throughput RNA sequencing identified upregulated genes in IgG4‐ROD and MALT lymphoma enriched in immune‐related pathways. These include classical antibody‐mediated complement activation, FCGR activation, complement cascade regulation, and CD22‐mediated BCR regulation [5]. To further elucidate the relationship between IgG4‐ROD and IgG4+ lacrimal gland MALT lymphoma, we employed single‐cell transcriptome sequencing on lesional lacrimal gland tissues. This approach aims to delineate the differential cellular infiltration and mechanistic roles of B cells in these two disease entities.

Results

B‐cell subtyping results

We conducted a comprehensive analysis of B cells in IgG4‐ROD and MALT lymphoma, and the UMAP plot visually demonstrated the distribution differences of B cells (Fig. 1A,B). After the overall dimensionality reduction and clustering of B cells in IgG4‐ROD, they were divided into 9 clusters (Fig. 1C). The overall display of the top marker genes for each cluster is shown in Fig. 1D (Data S1). Cluster 0 represents Naïve B cells, expressing marker genes such as TCL1A, JUNB, IGHD, FCER2, and CD69; Cluster 1 represents memory B1 cells, expressing marker genes such as MARCKS, GPR183, CCDC50, and TNFRSF13B; Cluster 2 represents memory B2 cells, expressing marker genes such as SELL, ITGB1, S100A4, and CRIP1; Cluster 3 represents IgG plasma cells, expressing genes such as IGLC3, IGLC2, IGKC, LCN1, IGHG1, and IGHG4; Cluster 4 represents GC Light Zone (GC_LZ) cells, expressing marker genes such as ACTG1, MARCKSL1, CCDC144A, and BCL2A1; Cluster 5 represents GC Dark Zone (GC_DZ) cells, expressing marker genes such as H4C3, TUBB, HMGB2, CDK1, TOP2A, and BIRC5; Cluster 7 represents IgA plasma cells, expressing genes such as IGHA1, JCHAIN and RRBP1; Cluster 6 and Cluster 8 express genes related to T cells (CD3D, CD3E, CD2, BCL11B) and endothelial cells (CLU, FDCSP, CCL14), respectively, and are thus named T‐like and Endothelial‐like cells. In IgG4‐ROD, naïve B cells are the most numerous, accounting for approximately 58.04%; memory B cells are the second most numerous, with Mem B1 and Mem B2 cells accounting for approximately 15.69% and 6.44% respectively, with Mem B1 cells being the predominant type (Fig. 1E). Due to the small number of Cluster 6 and 8, they can be disregarded in subsequent research and analysis.

In IgG4+ MALT lymphoma, B cells were grouped into six clusters after dimensionality reduction and clustering analysis (Fig. 1F). The top marker genes for each cluster are illustrated in Fig. 1G (Data S2). Cluster 0 corresponds to Memory_MALT B cells, expressing genes such as S100A4, SCIMP and CRIP1. Cluster 1 represents Naïve_MALT B cells, marked by genes including TCL1A, IGHD and CD83. Cluster 2 comprises Plasma_MALT B cells, characterized by IGKC, IGLC3, IGLC2 and IGLV2‐8. Cluster 3 is hypothesized to be Intermediate_MALT memory cells, expressing genes such as CCDC50, GNG11 and TNFRSF13B. Cluster 4 consists of GC_MALT cells, featuring MARCKSL1, TRBC1, HMGB2, CDK1 and STMN1. Cluster 5 corresponds to Plasmablast_MALT cells, expressing RRBP1, FOSB, IGHG2, and JCHAIN. Quantitative analysis revealed that memory B cells (52.26%) are the most abundant in IgG4+ MALT lymphoma, followed by naïve B cells (32.27%), plasma B cells (8.21%), Intermediate_MALT memory cells (3.11%), GC_MALT cells (2.54%), and Plasmablast_MALT cells (1.61%) (Fig. 2H). In IgG4 + MALT lymphoma, we identified different types of B cells. Their functional roles, phenotypes, and gene expressions may be similar to the inflammatory B cells or normal B cells in IgG4‐ROD. Subsequently, we will conduct comparative analyses of B cell populations with similar functions or phenotypes and gene expressions in IgG4‐ROD and IgG4 + MALT lymphoma to further clarify their functional roles and expression differences. B cells and T cells are abundant and co‐clustered in IgG4‐ROD and MALT lymphoma (Fig. 1I,J), with some T‐cell cores surrounded by B cells, and B cells and T cells intermingling, suggesting that some B cells may co‐express T‐cell‐related markers [6]. After verification through immunofluorescence, consultation with experimental technicians and review of relevant literature, we believe that B cells, T cells and endothelial cells are mixed together, which affects the expression of their specific markers. We will discard these cells in the subsequent research.

Fig. 2 — Comparative analysis results of naïve B cells between IgG4 + mucosa‐associated lymphoid tissue (MALT) lymphoma and immunoglobulin G4‐related ophthalmic disease (IgG4‐ROD). (A) Statistical graph showing expression differences of up/downregulated genes (likelihood ratio test, P value ≤ 0.05). (B) Volcano plot of differentially expressed genes (points located towards both ends and top indicate more significant expression differences) (likelihood ratio test, P value ≤ 0.05). (C) Top 20 pathways enriched in upregulated genes by Gene Ontology (GO) analysis (Bonferroni‐corrected p‐value threshold of ≤ 0.05). (D) Top 20 pathways enriched in downregulated genes by GO analysis (Bonferroni‐corrected p‐value threshold of ≤ 0.05); (E) Top 20 pathways enriched in upregulated genes by Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis (Bonferroni‐corrected p‐value threshold of ≤ 0.05). (F). Top 20 pathways enriched in downregulated genes by KEGG analysis (Bonferroni‐corrected p‐value threshold of ≤ 0.05).

Comparison of Naïve_B cells in MALT lymphoma and IgG4‐ROD

We performed a comparative analysis of naïve B cells in IgG4+ MALT lymphoma versus IgG4‐ROD. This analysis identified 135 upregulated and 107 downregulated DEGs (Fig. 2A) (Data S3). Volcano plot visualization highlighted the top upregulated genes as HLA‐DRB5, IGHG3, RPS4Y1, IGHG4, IGHG2, FAM118A, EIF1AY, and DDX3Y (Fig. 2B). The elevated expression of immunoglobulin genes (IGHG3, IGHG4, IGHG2) suggested abnormal differentiation of naïve B cells into plasma cell precursors secreting IgG, particularly the IgG4 subtype. Conversely, the most significantly downregulated genes were XIST and RPS4X. Gene Ontology (GO) analysis revealed 998 significantly enriched upregulated biological processes, primarily associated with immune system regulation, immune response, and B‐cell receptor signaling pathway, among others (Fig. 2C) (Data S4). Conversely, 1083 downregulated processes were enriched, mainly related to protein translation, transcription, localization, and RNA catabolism (Fig. 2D) (Data S5). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis indicated 55 significantly enriched upregulated pathways (Data S6). These were predominantly involved in Epstein–Barr virus infection, TNF signaling pathway, B cell receptor (BCR) signaling pathway, Th1/Th2/Th17 cell differentiation, and Toll‐like receptor signaling pathway (Fig. 2E), suggesting aberrant B cell proliferation/activation and abnormal T cell polarization. Fifty‐nine downregulated pathways were enriched (Data S7), primarily associated with ribosome biogenesis/function and coronavirus disease (Fig. 2F), indicating selective suppression of fundamental biosynthesis and metabolic functions in Naïve B cell.

Comparative analysis of memory B cells

In IgG4‐ROD, analysis between Memory B cell 1 (Mem B1) and Memory B cell 2 (Mem B2) cells identified 193 upregulated genes (top genes: CD69, EEF1B2, RPS2, VPREB3) and 427 downregulated genes (top genes: IGHE, ANXA4, SHISA8, LMO2, ITGAX, SDR16C5, MTSS1, S100A4, TMSB10) (Fig. 3A) (Data S8). In IgG4 + MALT lymphoma, analysis between Intermediate_Memory MALT and Memory_MALT cells revealed 369 upregulated genes (top genes: GNG11, CD1C, MPEG1, CD69, TNFRSF13B, RHOBTB3) and 467 downregulated genes (top genes: IGHE, MEF2C, TMSB10, S100A4, SHISA8) (Fig. 3B) (Data S9). Shared differentiation pathways: Key Mem B1 top genes (CD69, IGHE, SHISA8, S100A4) and characteristic genes (CCDC50, TNFRSF13B) resembled those of Intermediate_Memory cells. Similarly, Mem B2 and Memory_MALT cells showed parallels. This suggests a potentially shared B cell differentiation pathway in both diseases, albeit with different dominant subsets. This divergence may be driven by the disease‐specific microenvironment, positioning Mem B1 cells as analogous to intermediate memory cells. In IgG4‐ROD, Mem B1 cells (intermediate memory phenotype) predominated over Mem B2 cells. Conversely, in IgG4 + MALT lymphoma, Memory_MALT cells (terminal memory phenotype) constituted the major memory subset. This pattern implies a progression from an ‘intermediate state’ to a ‘terminal state’ during the transition from IgG4‐ROD to MALT lymphoma. KEGG analysis revealed that among the top 20 upregulated signaling pathways in IgG4 + MALT lymphoma, tumor‐associated pathways such as the BCR signaling pathway, NF‐κB signaling pathway, Epstein–Barr virus (EBV) infection, pathways in cancer, small cell lung cancer, transcriptional dysregulation in cancer, and Kaposi's sarcoma‐associated herpesvirus infection were primarily enriched in Intermediate_Memory B cell (Fig. 3C) (Data S10). This suggests that abnormal B‐cell receptor (BCR) activation may drive the NF‐κB pathway, synergizing with cancer‐related signaling pathways to promote malignant B‐cell transformation. In contrast, IgG4‐ROD showed enrichment of the estrogen signaling pathway, NOD‐like receptor signaling pathway, PI3K‐Akt signaling pathway, and antigen processing/presentation pathway in Memory B1 cells, indicating potential roles of hormonal microenvironments and innate immunity (NOD receptors) in immune tolerance regulation, which might suppress excessive inflammatory responses in IgG4‐ROD (Fig. 3D) (Data S11). Shared upregulated pathways including ribosome biogenesis, IL‐17 signaling, MAPK signaling, and TNF signaling were observed in both diseases, potentially reflecting their common chronic inflammatory basis involving cellular stress (ribosome), pro‐inflammatory cytokines (IL‐17/TNF), and proliferative signaling (MAPK).

Comparative analysis of memory cells between the IgG4 + MALT group and IgG4‐ROD group revealed 325 upregulated genes (Fig. 3E) (Data S12), with the top five being HLA‐DRB5, RPS4Y1, HLA‐DQA1, MEF2C, and IGKC. These findings suggest abnormal immune activation in the tumor microenvironment. Specifically, the upregulation of MEF2C in lymphoma may promote clonal expansion of malignant B cells and enhanced plasma cell differentiation. Additionally, 253 downregulated genes were identified, with the top five being LCN1, XIST, LTB, RPS4X and RPL10. KEGG pathway analysis demonstrated that upregulated signaling pathways were primarily enriched in disease‐related pathways such as oxidative phosphorylation, prion diseases, chemical carcinogenesis, and diabetic cardiomyopathy, as well as the BCR signaling pathway (Fig. 3F and Data S13). These results indicate that the activation of multiple ‘disease‐associated’ pathways (prion diseases, chemical carcinogenesis, diabetic cardiomyopathy) reflect significant stress responses in the tumor microenvironment (including protein misfolding and oxidative stress), genomic instability, and pro‐fibrotic tendencies. Downregulated signaling pathways were predominantly enriched in ribosomal pathways, apoptosis, microRNAs in cancer, calcium signaling pathway, BCR signaling pathway, and NF‐κB signaling pathway (Fig. 3G) (Data S14).

Comparative analysis of GC_B cells

Comparative analysis of GC_B cells between the IgG4 + MALT and IgG4‐ROD groups identified 682 upregulated genes (including significant differentially expressed genes such as HLA‐DRB5, RPS4Y1, EIF1AY, DDX3Y, USP9Y, CELF2, KDM5D, FYB1, CD2, SELL) and 735 downregulated genes (Fig. 4A,B) (Data S15). GO analysis revealed that upregulated genes were enriched primarily in processes related to biological regulation, immune cell regulation, cell activation, and signal transduction (Fig. 4C) (Data S16). Conversely, downregulated genes were associated mainly with cellular structure‐related processes, such as those involving membrane‐bound organelles, the cytoplasm, and extracellular vesicles (Fig. 4D) (Data S17). KEGG pathway analysis indicated that upregulated pathways were predominantly linked to cell adhesion, Epstein–Barr virus infection, Th17/Th1/Th2 cell differentiation, T‐cell receptor (TCR) signaling, and the PD‐1/PD‐L1 checkpoint pathway in cancer (Fig. 4E) (Data S18). Downregulated pathways were principally correlated with Parkinson's disease, diabetic cardiomyopathy, oxidative phosphorylation, and bacterial invasion of epithelial cells (Fig. 4F) (Data S19).

Fig. 4 — Analysis of Germinal center B (GC_B) cells in IgG4 + mucosa‐associated lymphoid tissue (MALT) lymphoma and immunoglobulin G4‐related ophthalmic disease (IgG4‐ROD). (A) Differential expression analysis of upregulated and downregulated genes in GC_B cells between IgG4 + MALT lymphoma and IgG4‐ROD (likelihood ratio test, P value ≤ 0.05). (B) Volcano plot showing expression differences of the top upregulated and downregulated genes (likelihood ratio test, P value ≤ 0.05). (C) Top 20 pathways enriched in Gene Ontology (GO) analysis of upregulated genes (Bonferroni‐corrected p‐value threshold of ≤ 0.05). (D) Top 20 pathways enriched in GO analysis of downregulated genes (Bonferroni‐corrected p‐value threshold of ≤ 0.05). (E) Top 20 pathways enriched in Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis of upregulated genes (Bonferroni‐corrected p‐value threshold of ≤ 0.05). (F) Top 20 pathways enriched in KEGG analysis of downregulated genes (Bonferroni‐corrected p‐value threshold of ≤ 0.05).

Comparative analysis of plasma cells

In IgG4‐ROD, plasma cells can be classified into IgG and IgA subtypes. Through comparative analysis of these two cell populations, we identified 174 upregulated genes and 1348 downregulated genes. The most significantly downregulated differential genes included NEAT1, ANKRD28, GOLGB1, ERN1, PRDM1, OGT, CREB3L2, MCL1, AKAP9, and GOLGA4 (Fig. 5A) (Data S20). KEGG pathway analysis revealed that the downregulated genes were primarily enriched in endoplasmic reticulum (ER) protein processing, spliceosome, FoxO signaling pathway, TNF signaling pathway, and NOD‐like receptor signaling pathway (Fig. 5B) (Data S21). Notably, IgG plasma cells exhibited a marked predominance of downregulated genes compared to upregulated genes. The downregulated gene list included key functional regulators such as PRDM1 (a master transcription factor for plasma cell differentiation) and MCL1 (an anti‐apoptotic gene). These findings suggest that IgG plasma cells in IgG4‐ROD may undergo severe functional exhaustion and impaired antibody secretion. Furthermore, the observed suppression of ER protein processing pathways and downregulation of FoxO/TNF signaling pathways may indicate a collapse of the unfolded protein response, potentially contributing to disease pathogenesis.

Fig. 5 — Plasma Cell Analysis Results of IgG4 + mucosa‐associated lymphoid tissue (MALT) lymphoma and immunoglobulin G4‐related ophthalmic disease (IgG4‐ROD). (A) Expression differences of top up−/down‐regulated genes in IgG plasma cells vs IgA plasma cells in IgG4‐ROD (likelihood ratio test, P value ≤ 0.05). (B) Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis showing top 20 enriched pathways of downregulated genes in IgG plasma cells compared to IgA plasma cells (Bonferroni‐corrected p‐value threshold of ≤ 0.05). (C) Volcano plot displaying expression differences of top up−/down‐regulated genes between IgG4 + MALT lymphoma and IgG4‐ROD (likelihood ratio test, P value ≤ 0.05). (D) Heatmap showing expression differences of top up−/down‐regulated genes between IgG4 + MALT lymphoma and IgG4‐ROD (likelihood ratio test, P value ≤ 0.05). (E) Gene Ontology (GO) analysis of top 20 enriched pathways for upregulated genes (Bonferroni‐corrected p‐value threshold of ≤ 0.05). (F) GO analysis of top 20 enriched pathways for downregulated genes (Bonferroni‐corrected p‐value threshold of ≤ 0.05). (G) KEGG analysis of top 20 enriched pathways for upregulated genes (Bonferroni‐corrected p‐value threshold of ≤ 0.05); H. KEGG analysis of top 20 enriched pathways for downregulated genes (Bonferroni‐corrected p‐value threshold of ≤ 0.05).

Comparative analysis of plasma cells between the IgG4 + MALT group and IgG4‐ROD revealed 89 upregulated genes (Data S22). The most significantly differentially expressed genes included IGHG3, IGHG2, HLA‐DRB5, SSR4, IGLV2‐8, IGHV1‐2, and IGHG4 (Fig. 5C,D). These findings suggest significant clonal expansion of plasma cells with potential IgG class switching predominantly towards IGHG3 and IGHG2. Notably, the overexpression of SSR4 (an endoplasmic reticulum protein transporter) may enhance antibody secretion to promote tumor growth. Conversely, 128 downregulated genes were identified, predominantly LYZ, LCN1, LTF, OPRPN, LACRT, PRR4, MT‐CO3, IGHA1, and IGHA2 (Fig. 5C,D). GO analysis demonstrated that upregulated genes were primarily enriched in endoplasmic reticulum functions, antigen binding, immune system processes, immunoglobulin complexes, complement activation, and B‐cell activation (Fig. 5E) (Data S23). Downregulated genes showed enrichment in membrane‐bound organelles, cytoplasm, and biological regulation processes (Fig. 5F) (Data S24). KEGG pathway analysis revealed that upregulated signaling pathways were mainly associated with protein processing in the endoplasmic reticulum, protein export, and cell adhesion (Fig. 5G) (Data S25). Conversely, downregulated pathways included ribosome function, coronavirus‐related pathways, antigen processing and presentation, BCR signaling, NF‐κB signaling, and Th17 cell differentiation (Fig. 5H) (Data S26).

Discussion

This study employs single‐cell transcriptome sequencing technology to systematically delineate, for the first time, the heterogeneity and functional states of B cells in IgG4‐positive lacrimal gland MALT lymphoma and IgG4‐ROD. The findings reveal that although both diseases share similar B‐cell differentiation trajectories, they exhibit significant differences in the proportions of key subpopulations, activation levels, differentiation directions, and molecular regulatory networks. These insights highlight potential mechanisms underlying the transition from chronic inflammation to malignant transformation.

The study by Yang et al. found that in MALT lymphoma tissues, the Bmem subpopulation predominated, followed by naïve B‐cell subsets, while GC B‐cell subsets accounted for the lowest proportion. In contrast, IgG4‐ROD tissues exhibited a predominance of GC B‐cell subsets, followed by naïve B cell subsets, with the Bmem subpopulation being the least represented [7]. In this study, multiple B cell subpopulations were also identified in IgG4‐ROD, including naïve B cells, memory B cells, IgG plasma cells, GC_LZ cells, GC_DZ cells, IgA plasma cells, and a small number of T cell‐like and endothelial cell‐like populations. Among these, naïve B cells were the most abundant, followed by memory B cells. In IgG4+ MALT lymphoma, diverse B‐cell subpopulations were similarly observed, but with distinct proportions and characteristics, predominantly memory B cells. The discrepancies between these findings and previous studies may stem from differences in sample size and cell classification methodologies. These results highlight distinct compositional and differentiation pathway differences in B‐cell populations between the two diseases, providing foundational insights for further exploration of their pathogenic mechanisms.

Naïve B cells constitute a certain proportion in both diseases, but in IgG4+ MALT lymphoma, they exhibit a highly activated state with a pronounced tendency toward abnormal differentiation into plasma cells, which contrasts sharply with the relatively subdued inflammatory activation observed in IgG4‐ROD. This aberrant differentiation may be driven by abnormal activation of BCR signaling [8], coupled with EBV infection and activation of Toll‐like receptor and TNF signaling pathways [9, 10, 11], collectively promoting malignant transformation of B cells. These findings reinforce the neoplastic nature of IgG4+ MALT lymphoma and suggest potential therapeutic targets.

Memory B cells also exhibit distinct characteristics in the two diseases. In IgG4‐ROD, there is a higher proportion of Memory B1 (intermediate memory) cells, whereas IgG4 + MALT lymphoma is predominantly composed of Memory MALT (terminal‐state) cells. These differences likely reflect the influence of distinct disease microenvironments on B‐cell differentiation. The enrichment of estrogen signaling and NOD‐like receptor pathways in IgG4‐ROD may suggest hormonal regulation and innate immune involvement in maintaining “intermediate‐state” memory cells [12, 13]. In contrast, the activation of EBV and cancer‐related pathways in MALT lymphoma may drive terminal differentiation accompanied by malignant transformation. Upregulation of oxidative phosphorylation pathways and downregulation of ribosomal pathways indicate that the baseline protein synthesis capacity of IgG4 + MALT lymphoma cells may be suppressed or reprogrammed. Reduced expression of tumor‐suppressive miRNAs in IgG4 + MALT lymphoma could lead to elevated expression of their pro‐oncogenic target genes [14, 15]. Notably, the BCR signaling pathway shows both upregulation and downregulation in memory cells, suggesting complex and heterogeneous regulation of distinct pathway components or downstream branches between the two groups. The downregulation of NF‐κB signaling may indicate the presence of alternative survival signals in IgG4 + MALT lymphoma memory cells, subtype‐ or stage‐specific activation of NF‐κB, or microenvironmental alterations due to glucocorticoid therapy. These findings require further validation to clarify their mechanistic implications.

GC B cells play a crucial role in antibody production and immune memory formation. In this study, we observed aberrant activation of GC B cells in IgG4 + MALT lymphoma, characterized by the high expression of HLA‐II‐related genes. This suggests that abnormal B‐cell activation may promote CD4+ T‐cell activation, which in turn feedback activates B cells, forming a vicious cycle of clonal expansion and plasma cell differentiation. This dysregulated activation may be associated with EBV infection, Th17/Th1 and Th2 cell differentiation, and activation of the TCR signaling pathway [16]. The enrichment of strong immune activation signals (T‐cell differentiation, TCR signaling, cellular activation) and immune checkpoint molecules indicates that GC B cells in IgG4 + MALT reside in a highly immunostimulatory and potentially immunodysregulated microenvironment [17]. Downregulated pathways were primarily involved in maintaining basic cellular structure and energy metabolism (oxidative phosphorylation), suggesting that GC B cells in IgG4‐ROD may exist in a relatively stable or reactive proliferative state, relying more on fundamental cellular processes rather than intense, specific immune activation. Notably, in IgG4 + MALT lymphoma, some GC B cells might fail to differentiate normally into memory B cells or plasma cells, instead remaining arrested in the germinal center stage with proliferative advantages. This differentiation blockade may be related to abnormalities in membrane organelle formation and cytoplasmic compartmentalization during cellular development. These findings provide novel insights into understanding the pathogenesis of IgG4 + MALT lymphoma and exploring new therapeutic strategies.

Plasma cells are the primary cell type responsible for antibody production. In this study, we identified severe functional exhaustion of plasma cells in IgG4‐ROD, manifested by a predominant downregulation of genes in IgG+ plasma cells. Notably, this included critical genes such as PRDM1 (a master regulator of plasma cell differentiation) and the anti‐apoptotic gene MCL1 [18, 19]. This functional exhaustion may impair the ability of plasma cells to efficiently secrete antibodies, thereby compromising their immune function. Additionally, antigens released from dying plasma cells may perpetuate tissue inflammation through recurrent activation, establishing a vicious cycle. These findings provide a novel perspective to explain the paradoxical coexistence of extensive plasma cell infiltration and functional incompetence in IgG4‐ROD. In contrast, plasma cells in IgG4 + MALT lymphoma demonstrate distinct features of monoclonal expansion and IgG class‐switching. These clonally expanded plasma cells secrete large quantities of monoclonal IgG4 antibodies, which mediate tissue destruction through Fc receptor‐dependent inflammatory pathways and complement activation [20]. Concurrently, the downregulation of antimicrobial proteins LYZ/LTF suggests impaired mucosal immune responses, potentially exacerbating local infection risks or chronic inflammation [21]. Collectively, these findings reveal the malignant transformation characteristics of plasma cells in IgG4 + MALT lymphoma and their critical contribution to disease progression.

This study employs scRNA‐seq technology to comprehensively investigate the heterogeneity and functional disparities of B cells in IgG4‐ROD and IgG4‐positive MALT lymphoma. These findings provide novel insights into the pathogenesis of both disorders and reveal potential therapeutic targets for exploring new treatment strategies. However, current limitations include a restricted sample size and insufficient exploration of dynamic changes during disease progression. To address these issues, future research will expand sample cohorts to include patients across diverse disease stages. Additionally, systematic integration of experimental findings with comprehensive multi‐omics analyses will be conducted to holistically elucidate the pathogenic mechanisms and cellular heterogeneity underlying these conditions. Of particular clinical importance, the validation of identified therapeutic targets through translational studies and subsequent clinical trials will constitute a critical research priority.

Materials and methods

Subjects

Between May 2022 and August 2024, we collected samples from 4 patients with histopathologically confirmed IgG4‐ROD (aged 32–72 years, male‐to‐female ratio 1 ∶ 3) and 2 male patients with IgG4+ lacrimal gland MALT lymphoma (both aged 63 years) for single‐cell transcriptome sequencing analysis. Diseased lacrimal gland tissues used in this study were obtained from Beijing Tongren Hospital, Capital Medical University, and subsequently transferred to the laboratory for further processing. Written informed consent was obtained, and all data will be kept strictly confidential. The study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Ethics Committee of Beijing Tongren Hospital, Capital Medical University (TREC2023‐KY061). The inclusion criteria for all patients with IgG4‐ROD strictly followed the new diagnostic criteria for IgG4‐ROD revised by Japanese scholars in 2023 [22]. The characteristic pathological features of IgG4‐ROD and IgG4+ MALT lymphoma were shown in Fig. 6A–D.

Fig. 6 — Pathological manifestations of immunoglobulin G4‐related ophthalmic disease (IgG4‐ROD) and IgG4 + mucosa‐associated lymphoid tissue (MALT) lymphoma. (A) IgG4‐ROD exhibits acinar atrophy (black arrow) and massive lymphocytic infiltration (blue arrow) in the lacrimal gland (Immunohistochemistry staining, n = 3, magnification × 400). (B) Positive immunohistochemical staining for IgG4 in IgG4‐ROD (Immunohistochemistry Staining, magnification × 400, n = 3, black arrow). (C) The replacement of the gland's normal glandular tissue (acini and ducts) by a dense infiltration of malignant B‐cells (Immunohistochemistry staining, n = 3, magnification × 400). (D) Positive immunohistochemical staining for IgG4 in IgG4 + MALT lymphoma (Immunohistochemistry Staining, n = 3, magnification × 400, black arrow).

Tissue specimen collection

Tissues were transported on ice in sterile culture dishes containing 10 mL of 1x Dulbecco's Phosphate‐Buffered Saline (Thermo Fisher, Cat. no. 14190144) to remove residual storage solution and then minced on ice. The minced tissues were digested using a dissociation enzyme mixture containing 0.25% Trypsin (Thermo Fisher, Cat. no. 25200–072, Shanghai, China) and 10 μg·mL⁻¹ DNase I (Sigma, Cat. no. 11284932001, St. Louis, MO, USA) dissolved in PBS supplemented with 5% Fetal Bovine Serum (Thermo Fisher, Cat. no. SV30087.02). Dissociated cells were collected every 20 min to maximize yield and viability. Cell viability was assessed by staining with 0.4% Trypan blue (Thermo Fisher, Cat. no. 15250061) and counted using a Countess® II Automated Cell Counter (Thermo Fisher).

Single‐cell RNA‐sequencing and data analysis

Single‐cell RNA sequencing (scRNA‐seq) was performed using the 10 × Genomics Gem‐Code platform. Sequencing services and subsequent bioinformatic analyses were provided by Majorbio Co., Ltd. (Shanghai, China). Raw reads underwent processing via the Cell Ranger pipeline (v7.1.0) with default recommended parameters. Specifically, Illumina‐derived FASTQ files were aligned to the human reference genome (GRCh38) using the STAR algorithm [23]. Gene‐Barcode matrices were then generated per sample by counting Unique Molecular Identifiers (UMIs) and filtering noncellular barcodes. This aggregated gene‐barcode matrix, containing expression counts for all barcoded cells, served as input for the Seurat R toolkit (v4.1.1) for quality control (QC) and downstream analysis [24]. QC involved filtering low‐quality cells based on three standard metrics: (1) Total transcript count (nUMI), (2) Number of detected genes, and (3) Percentage of reads mapping to mitochondrial genes (calculated using Seurat's PercentageFeatureSet) [25]. Cells exceeding mean ± 2 standard deviations for any metric were excluded. Data normalization was performed (NormalizeData) prior to identifying a subset of highly variable genes, accounting for the mean–variance relationship. Uniform Manifold Approximation and Projection (UMAP) was used for cluster visualization [26]. Subclustering of specific cell populations applied the same workflow (scaling, PCA, Louvain clustering) to subsetted data. Cluster identity assignment leveraged both the SingleR package [25] and known marker genes. Differential gene expression analysis was conducted using Seurat's FindMarkers function, implementing a likelihood ratio test. Comparisons were made between clusters or experimental samples. Differentially expressed genes (DEGs) were defined by an absolute log₂ fold‐change > 0.25 and an adjusted p‐value (Q‐value) ≤ 0.05. Finally, Gene Ontology (GO) functional enrichment analysis of significant DEGs was performed using Goatools (https://github.com/tanghaibao/Goatools), testing against the whole transcriptome background. Enriched terms were identified at a Bonferroni‐corrected p‐value threshold of ≤ 0.05. Unless otherwise noted, all Seurat functions used default parameters.

Immunohistochemical staining (IHC)

The tissue sections underwent deparaffinization in xylene, followed by rehydration using graded ethanol solutions (100, 95, 70, and 50%), and were finally rinsed with distilled water. Subsequently, antigen retrieval was conducted in 0.01 m citrate buffer, and endogenous peroxidase activity was blocked by treating the sections with 3% H₂O₂ in methanol for 10 min. To minimize nonspecific binding, the slides were treated with 10% normal goat serum diluted in phosphate‐buffered saline (PBS) at ambient temperature for 1 h. The primary antibody was then added to the sections and left to incubate overnight at 4 °C, followed by a one‐hour incubation with biotinylated goat antimouse immunoglobulin G (Sigma, MO, USA) at room temperature. After three rinses with PBS, a biotin‐labeled secondary antibody was introduced, and the tissues were incubated at 37 °C for 60 min [27]. A suitable volume of biotin substrate was subsequently applied. Immunoreactivity was visualized using a DAB solution, and the sections were counterstained with hematoxylin. Finally, the tissues were dehydrated and cleared with xylene. The staining outcomes were independently assessed by two experienced pathologists. The primary antibody was as follows: IgG4 (AB109493, Abcam, Cambridge, UK).

Immunofluorescence (IF)

Paraffin‐embedded tissue sections were deparaffinized in xylene and rehydrated through a graded ethanol series (100%, 95%, 85%, 75%), followed by rinsing with PBS. For antigen retrieval, the sections were heated in citrate buffer (pH 6.0) under high pressure and high temperature, then allowed to cool naturally to room temperature [28]. Non‐specific binding sites were blocked by incubating the sections with 5% bovine serum albumin in PBS for 30 min at room temperature. After removing the blocking solution, the sections were incubated with an appropriate dilution of primary antibody overnight at 4 °C in a humidified chamber. The next day, after thorough washing with PBS (three times, 5 min each), the corresponding species‐specific fluorophore‐conjugated secondary antibody was applied and incubated for 1 h at 37 °C in the dark. Following another wash with PBS, the sections were mounted with antifade mounting medium containing DAPI for nuclear counterstaining. Negative controls (with PBS replacing the primary antibody) were included in all staining runs. Fluorescence images were captured using a fluorescence microscope (specify model and manufacturer), and at least five high‐power fields per section were randomly selected for analysis. Primary antibodies were as follows: CD20 (AB64088, Abcam); CD3D (AB103573, Abcam).

Authors' contributions

RL analyzed and wrote the manuscript; TR, LX, NW, and YZ helped collect the data and make the experiments; HY, TW, and RY helped analyze the data; JM and JL read and criticized the manuscript.

Conflict of interest

The authors declare no conflict of interest.

Supporting information

Data S1. Details of B cell expression differences in immunoglobulin G4‐related ophthalmic disease (Marker gene list).

FEBS-293-4992-s010.csv^{(383.2KB, csv)}

Data S2. Details of B cell expression differences in IgG4+ mucosa‐associated lymphoid tissue lymphoma (Marker gene list).

FEBS-293-4992-s008.csv^{(212.8KB, csv)}

Data S3. Comparison of expressed gene differences in Naive B cell expression between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (likelihood ratio test, P value ≤ 0.05).

FEBS-293-4992-s023.csv^{(112.2KB, csv)}

Data S4. The up‐regulated biological processes of Naive_B cells between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (Bonferroni‐correction).

FEBS-293-4992-s006.csv^{(101.8KB, csv)}

Data S5. The down‐regulated biological processes of Naive_B cells between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (Bonferroni‐correction).

FEBS-293-4992-s013.csv^{(109.7KB, csv)}

Data S6. The up‐regulated Kyoto Encyclopedia of Genes and Genomes pathways of Naive_B cells between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (Bonferroni‐correction).

FEBS-293-4992-s014.csv^{(7.6KB, csv)}

Data S7. The down‐regulated Kyoto Encyclopedia of Genes and Genomes pathways of Naive_B cells between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (Bonferroni‐correction).

FEBS-293-4992-s001.csv^{(7.9KB, csv)}

Data S8. The differentially expressed genes between Memory B cell 1 and Memory B cell 2 (likelihood ratio test, P value ≤ 0.05).

FEBS-293-4992-s021.csv^{(265.8KB, csv)}

Data S9. The differentially expressed genes between Intermediate_Memory MALT and Memory_MALT cells (likelihood ratio test, P value ≤ 0.05).

FEBS-293-4992-s009.csv^{(322.6KB, csv)}

Data S10. The Kyoto Encyclopedia of Genes and Genomes pathways analysis between Intermediate_Memory MALT and Memory_MALT cells (Bonferroni‐correction).

FEBS-293-4992-s025.csv^{(33.5KB, csv)}

Data S11. The Kyoto Encyclopedia of Genes and Genomes pathways analysis between Memory B cell 1 and Memory B cell 2 cells (Bonferroni‐correction).

FEBS-293-4992-s020.csv^{(26.7KB, csv)}

Data S12. Comparative analysis of memory cells' differentially expressed genes between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (likelihood ratio test, P value ≤ 0.05).

FEBS-293-4992-s003.csv^{(219KB, csv)}

Data S13. Comparative analysis of memory cells' up‐regulated pathways between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (Bonferroni‐correction).

FEBS-293-4992-s022.csv^{(30.9KB, csv)}

Data S14. Comparative analysis of memory cells' down‐regulated pathways between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (Bonferroni‐correction).

FEBS-293-4992-s007.csv^{(27.8KB, csv)}

Data S15. Comparative analysis of Germinal center B cells' differentially expressed genes between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (likelihood ratio test, P value ≤ 0.05).

FEBS-293-4992-s018.csv^{(461.9KB, csv)}

Data S16. Gene Ontology analysis of Germinal center B cells between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (up‐regulated) (Bonferroni‐correction).

FEBS-293-4992-s005.csv^{(277.7KB, csv)}

Data S17. Gene Ontology analysis of Germinal center B cells between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (down‐regulated) (Bonferroni‐correction).

FEBS-293-4992-s015.csv^{(265.1KB, csv)}

Data S18. Kyoto Encyclopedia of Genes and Genomes analysis of Germinal center B cells between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (up‐regulated) (Bonferroni‐correction).

FEBS-293-4992-s011.csv^{(37KB, csv)}

Data S19. Kyoto Encyclopedia of Genes and Genomes analysis of Germinal center B cells between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (down‐regulated) (Bonferroni‐correction).

FEBS-293-4992-s012.csv^{(41.5KB, csv)}

Data S20. The differentially expressed genes between IgG and IgA plasma cells (likelihood ratio test, P value ≤ 0.05).

FEBS-293-4992-s024.csv^{(421.5KB, csv)}

Data S21. The Kyoto Encyclopedia of Genes and Genomes pathways between IgG and IgA plasma cells (down‐regulated) (Bonferroni‐correction).

FEBS-293-4992-s002.csv^{(38.4KB, csv)}

Data S22. Comparative differentially expressed genes analysis of plasma cells between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (likelihood ratio test, P value ≤ 0.05).

FEBS-293-4992-s019.csv^{(442.7KB, csv)}

Data S23. Comparative Gene Ontology analysis of plasma cells between the immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (up‐regulated) (Bonferroni‐correction).

FEBS-293-4992-s026.csv^{(61.8KB, csv)}

Data S24. Comparative Gene Ontology analysis of plasma cells between the immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (down‐regulated) (Bonferroni‐correction).

FEBS-293-4992-s016.csv^{(355.6KB, csv)}

Data S25. Comparative Kyoto Encyclopedia of Genes and Genomes analysis of plasma cells between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (up‐regulated) (Bonferroni‐correction).

FEBS-293-4992-s004.csv^{(10KB, csv)}

Data S26. Comparative Kyoto Encyclopedia of Genes and Genomes analysis of plasma cells between immunoglobulin G4‐related ophthalmic disease and IgG4 + mucosa‐associated lymphoid tissue lymphoma (down‐regulated) (Bonferroni‐correction).

FEBS-293-4992-s017.csv^{(40.1KB, csv)}

Acknowledgements

Supported by the National Key R&D Program of China (2023YFC2410200), the Internal Seed Grant for Young Reserve Talents from Beijing Tongren Hospital, Capital Medical University (2024‐YJJ‐ZZL‐021), and Beijing Hospitals Authority Clinical Medicine Development of Special Funding Support (ZLRK202503). We thank Shanghai Majorbio Bio‐pharm Technology for providing single‐cell sequencing. Dr. Dong Yachen, Liu weiting, and Jiang yuling from Majorbio Bio‐pharm Technology analyzed single‐cell sequencing data and visualized all sequencing data.

Contributor Information

Jianmin Ma, Email: jmma@sina.com.

Jing Li, Email: ljlsdlyh@126.com.

Data availability statement

The datasets generated during this study are openly available in NCBI (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA1256260). Supplemental materials are provided in the online supplementary files accompanying this publication.

References

1. Cheuk W, Yuen HK, Chan AC, Shih LY, Kuo TT, Ma MW, Lo YF, Chan WK & Chan JK (2008) Ocular adnexal lymphoma associated with IgG4+ chronic sclerosing dacryoadenitis: a previously undescribed complication of IgG4‐related sclerosing disease. Am J Surg Pathol 32, 1159–1167. [DOI] [PubMed] [Google Scholar]
2. Zhang XJ, Zhao PX, Ma MS, Wu H, Liu R, Wang H, Liu MY, Xie F & Ma XM (2023) Association analysis of BclI with benign lymphoepithelial lesions of the lacrimal gland and glucocorticoids resistance. Int J Ophthalmol 16, 1734–1745. [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Liu R, Wang J, Wang N, Li J, Ge X, Zhang J & Ma J (2021) Clinical features and prognoses of IgG4‐positive and IgG4‐negative lacrimal lymphomas. Front Oncol 11, 622847. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Andrew N, Kearney D & Selva D (2013) IgG4‐related orbital disease: a meta‐analysis and review. Acta Ophthalmol 91, 694–700. [DOI] [PubMed] [Google Scholar]
5. Asakage M, Usui Y, Nezu N, Shimizu H, Tsubota K, Umazume K, Yamakawa N, Umezu T, Suwanai H, Kuroda M et al. (2020) Comprehensive gene analysis of IgG4‐related ophthalmic disease using RNA sequencing. J Clin Med 9, 3458. [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Rattner A, Heng JS, Winer BL, Goff LA & Nathans J (2023) Normal and Sjogren's syndrome models of the murine lacrimal gland studied at single‐cell resolution. Proc Natl Acad Sci USA 120, e2311983120. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Yang Y, Wang Y, Jin X & He W (2025) Single‐cell RNA‐sequencing reveals cellular heterogeneity and immune microenvironment characteristics between ocular adnexal mucosa‐associated lymphoid lymphoma and IgG4‐related ophthalmic disease. Front Immunol 16, 1508559. [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Magistri M, Happ LE, Ramdial J, Lu X, Stathias V, Kunkalla K, Agarwal N, Jiang X, Schürer SC, Dubovy SR et al. (2021) The genetic landscape of ocular adnexa MALT lymphoma reveals frequent aberrations in NFAT and MEF2B signaling pathways. Cancer Res Commun 1, 1–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Wehr C, Houet L, Unger S, Kindle G, Goldacker S, Grimbacher B, de Caballero Garcia Oteyza A, Marks R, Pfeifer D, Nieters A et al. (2021) Altered Spectrum of lymphoid neoplasms in a single‐center cohort of common variable immunodeficiency with immune dysregulation. J Clin Immunol 41, 1250–1265. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Fonte E, Agathangelidis A, Reverberi D, Ntoufa S, Scarfò L, Ranghetti P, Cutrona G, Tedeschi A, Xochelli A, Caligaris‐Cappio F et al. (2015) Toll‐like receptor stimulation in splenic marginal zone lymphoma can modulate cell signaling, activation and proliferation. Haematologica 100, 1460–1468. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Kuo SH, Yeh PY, Chen LT, Wu MS, Lin CW, Yeh KH, Tzeng YS, Chen JY, Hsu PN, Lin JT et al. (2008) Overexpression of B cell‐activating factor of TNF family (BAFF) is associated with helicobacter pylori‐independent growth of gastric diffuse large B‐cell lymphoma with histologic evidence of MALT lymphoma. Blood 112, 2927–2934. [DOI] [PubMed] [Google Scholar]
12. Zhang X, Zhao P, Ma M, Wu H, Liu R, Liu Z, Cai Z, Liu M, Xie F & Ma X (2023) Missing link between tissue specific expressing pattern of ERβ and the clinical manifestations in LGBLEL. Front Med 10, 1168977. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Hsieh SC, Shen CY, Liao HT, Chen MH, Wu CH, Li KJ, Lu CS, Kuo YM, Tsai HC, Tsai CY et al. (2020) The cellular and molecular bases of allergy, inflammation and tissue fibrosis in patients with IgG4‐related disease. Int J Mol Sci 21, 5082. [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Wolf J, Reinhard T, Hajdu RI, Schlunck G, Auw‐Haedrich C & Lange C (2023) Transcriptional profiling identifies prognostic gene signatures for conjunctival Extranodal marginal zone lymphoma. Biomolecules 13, 115. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Elhamamsy AR, Metge BJ, Alsheikh HA, Shevde LA & Samant RS (2022) Ribosome biogenesis: a central player in cancer metastasis and therapeutic resistance. Cancer Res 82, 2344–2353. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Lin Y, Wan Z, Liu B, Yao J, Li T, Yang F, Sui J, Zhao Y, Liu W, Zhou X et al. (2024) B cell‐reactive triad of B cells, follicular helper and regulatory T cells at homeostasis. Cell Res 34, 295–308. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Umehara H, Nakajima A, Nakamura T, Kawanami T, Tanaka M, Dong L & Kawano M (2014) IgG4‐related disease and its pathogenesis‐cross‐talk between innate and acquired immunity. Int Immunol 26, 585–595. [DOI] [PMC free article] [PubMed] [Google Scholar]
18. Boi M, Zucca E, Inghirami G & Bertoni F (2015) PRDM1/BLIMP1: a tumor suppressor gene in B and T cell lymphomas. Leuk Lymphoma 56, 1223–1228. [DOI] [PubMed] [Google Scholar]
19. Klanova M, Kazantsev D, Pokorna E, Zikmund T, Karolova J, Behounek M, Renesova N, Sovilj D, Kelemen CD, Helman K et al. (2022) Anti‐apoptotic MCL1 protein represents critical survival molecule for Most Burkitt lymphomas and BCL2‐negative diffuse large B‐cell lymphomas. Mol Cancer Ther 21, 89–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Sugimoto M, Watanabe H, Asano T, Sato S, Takagi T, Kobayashi H & Ohira H (2016) Possible participation of IgG4 in the activation of complement in IgG4‐related disease with hypocomplementemia. Mod Rheumatol 26, 251–258. doi: 10.3109/14397595.2015.1076924 [DOI] [PubMed] [Google Scholar]
21. Huang Y, Wang M, Hong Y, Bu X, Luan G, Wang Y, Li Y, Lou H, Wang C & Zhang L (2021) Reduced expression of antimicrobial protein secretory Leukoprotease inhibitor and Clusterin in chronic rhinosinusitis with nasal polyps. J Immunol Res 2021, 1057186. [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Takahira M, Goto H & Azumi A (2024) The 2023 revised diagnostic criteria for IgG4‐related ophthalmic disease. Jpn J Ophthalmol 68, 293–301. [DOI] [PMC free article] [PubMed] [Google Scholar]
23. Pardoll DM, Papadopoulos N, Kinzler KW, Eshleman JR, Vogelstein B, Anders RA & Diaz LA Jr (2017) Mismatch repair deficiency predicts response of solid tumors to PD‐1 blockade. Science 357, 409–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
24. Satija R, Farrell JA, Gennert D, Schier AF & Regev A (2015) Spatial reconstruction of single‐cell gene expression data. Nat Biotechnol 33, 495–502. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Aran D, Looney AP, Liu L, Wu E, Fong V, Hsu A, Chak S, Naikawadi RP, Wolters PJ, Abate AR et al. (2019) Reference‐based analysis of lung single‐cell sequencing reveals a transitional profibrotic macrophage. Nat Immunol 20, 163–172. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Zhou Y & Sharpee TO (2022) Using global t‐SNE to preserve Intercluster data structure. Neural Comput 34, 1637–1651. [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Li J, Liu R, Sun M, Wang J, Wang N, Zhang X, Ge X & Ma J (2021) The FcεRI signaling pathway is involved in the pathogenesis of lacrimal gland benign lymphoepithelial lesions as shown by transcriptomic analysis. Sci Rep 11, 21853. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Im K, Mareninov S, Diaz MFP & Yong WH (2019) An introduction to performing immunofluorescence staining. Methods Mol Biol 1897, 299–311. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials