Fig 2: The proliferation of T follicular helper (Tfh) cells, CD4+ T helper (Th) cells, and the identification of a novel inflammatory subtype of T cells associated with NAMPT in pancreatic lesions of type 1 autoimmune pancreatitis (AIP). A) Left: UMAP plot illustrating the annotation of various T‐cell subgroups identified in the analysis. Right: UMAP plot showcasing the relative proportions of each T‐cell subgroup within the dataset. B) Bar graph representing the proportion of each T‐cell subgroup across the different experimental groups examined. C) Summary of the number of DEGs identified when comparing the AIP group to normal samples. D) Pseudotime analysis displayed on a UMAP plot, illustrating the differentiation trajectory of T cells over time. E,F) Selected GO terms for Tfh cells and CD4+ Th cells in the type 1 AIP group as compared to those in normal pancreatic tissue or the post‐PBMC group, determined through Gene Set Enrichment Analysis (GSEA). G) UMAP plot of the Tfh subgroup depicting the proportion of each subgroup within Tfh cells. H) Volcano plot displaying DEGs between clusters 0 & 2 and cluster 1, with the top 10 DEGs highlighted for clarity. I) UMAP plot visualizing the distributions of the genes IFNG and CD200 within Tfh cells, providing insight into their expression patterns. J) Volcano plot illustrating DEGs between type 1 AIP patients and the other experimental groups, with the top 10 DEGs clearly marked for emphasis.

Fig 3: HIF‐1α‐induced activation of visfatin in macrophages contributes to fibrosis through interactions between macrophages and pancreatic stellate cells in both type 1 and type 2 autoimmune pancreatitis (AIP) patients. A) A summary of the number of interaction pairs between various cell subgroups and pancreatic stellate cells is presented for patients with type 1 AIP (left), type 2 AIP (middle), and normal control groups (right). B) The interaction strength of the visfatin signaling pathway between different cell subgroups and pancreatic stellate cells is depicted for type 1 AIP (left), type 2 AIP (middle), and normal groups (right). C) Relative mRNA expression levels of HIF1A and NAMPT are compared in RAW264.7 cells between scramble and HIF1A‐overexpressing groups. D) Analysis of protein expression levels of collagen‐1 (Col‐1), fibronectin (FN), HIF‐1α, NAMPT, and GAPDH in primary MPSCs after co‐culture with various RAW264.7 groups, including the scramble group, HIF‐1α overexpression group (OE‐HIF1A), and the HIF‐1α overexpression group with exogenous FK866 (OE‐HIF1A+FK866). Statistical significance is indicated as ns: not significant (p > 0.05); *: p < 0.05; **: p < 0.01. E) Visualization of enriched motif binding to HIF‐1α on the NAMPT promoter region, highlighting potential regulatory elements. F) Potential binding sites of HIF‐1α (TFBS1 and TFBS2) on the NAMPT promoter were predicted using the JASPAR database, including sequences for both full‐length and mutated promoter constructs (mut1, mut2). G) Results of dual‐luciferase assays measuring the luciferase activity of NAMPT promoter constructs (full, Mut1, and Mut2). Data are expressed as means ± SDs (n = 3), with statistical significance denoted as ns: not significant (p > 0.05); *: p < 0.05; **: p < 0.01, & p < 0.05 when compared to the h‐NAMPT (full) + pcDNA3.1–3xflag group; ## p < 0.01 when compared to the h‐NAMPT (full) + h‐HIF1A group. H) Representative immunofluorescence and immunohistochemical images displaying the expression of NAMPT, α‐smooth muscle actin (α‐SMA), and collagen‐1 (Col‐1) in untreated control mice (NC) versus AIP mice (AIP). Scale bars indicate 50 µm. I) Quantification of the proportions of NAMPT‐, α‐SMA‐, and Col‐1‐positive areas in untreated mice (NC) compared to AIP‐treated mice (AIP), with n = 5 for each group. Statistical significance is indicated with **** denoting p < 0.0001.

Fig 4: Enrichment of classic HIF‐1α‐positive monocytes in patients with type 1 and type 2 autoimmune pancreatitis (AIP), exhibiting a proinflammatory phenotype. A) UMAP plot illustrating the annotation of various myeloid cell subgroups identified within the dataset. B) UMAP plot showcasing the relative proportions of each myeloid cell subgroup, providing insight into their distribution. C) Bar graph representing the proportion of each myeloid cell subgroup across the different experimental groups, including type 1 AIP, type 2 AIP, post‐PBMC, and normal groups. D) Summary of the number of DEGs identified between the AIP group and normal group, encompassing both type 1 and type 2 AIP, and comparing them to the post‐PBMC group or the normal group. E) UMAP plot (left) alongside bee swarm plots (right) illustrating differential abundances of cell types in the pancreatic tissues of the normal group versus the type 1 AIP group. F) Pseudotime analysis revealing the differentiation trajectory among myeloid cells, indicating their developmental progression over time. G) Pseudotime analysis results for each individual myeloid cell subgroup, highlighting their unique differentiation paths. H) Heatmap representing the correlation analysis between different myeloid cell subgroups, showcasing their interrelationships. I) Identification of DEGs and functional modules derived from pseudotime analysis, along with enriched GO terms (p < 0.01), to elucidate biological processes involved. J) Evaluation of signature scores for M1 and M2 polarization in each myeloid cell subgroup, reflecting their inflammatory states. K) Right: Representative immunofluorescence image depicting NAMPT‐positive inflammatory CD4+ T cells present in the pancreatic lesions of type 1 AIP patients. Scale bars indicate 50 µm. Left: Quantification of the density of TGFB1‐positive macrophages and classic HIF1A‐positive monocytes in the pancreatic tissues of type 1 AIP patients (n = 5), type 2 AIP patients (n = 5), and healthy controls (n = 5), with statistical significance indicated (**p < 0.01, ***p < 0.001).