Fig 2: Treg cells suppress a systemic type 2 immune response.CP was induced in DEREG mice by repetitive caerulein i.p. injection over 4 weeks in Treg-depleted animals. Control mice received PBS instead of DT. a An anti-GFP antibody was used for immunofluorescence staining of GFP+ Treg cells in lymph node, scale bars represent 10 µm. b Determination of GFP+ and CD25+ splenocytes was performed by flow cytometry (p < 0.0001, PBS n = 13/DT n = 13). c T-effector cell activation was measured by detection of CD69 on CD4+ T cells (p = 0.0008, PBS n = 13/DT n = 13). d–f Furthermore, the T effector differentiation of CD4+ splenocytes were quantified by flow cytometry. The T effector immune response was measured by intracellular staining of GATA3 (p = 0.0002, PBS n = 6/DT n = 7), TBET (PBS n = 6/DT n = 7) and ROR?t (PBS n = 6/DT n = 6) in CD4+ cells. Detected levels of the transcription factors TBET and ROR?t were not significantly different between both groups. In contrast to the Th1 and Th17 immune responses, the Th2 immune response, as hallmarked by the presence of GATA3, was increased by depletion of regulatory T cells during CP. g Measurement of serum cytokines by a cytometric bead array showed an increase of type 2 cytokines like IL-10 (p = 0.0271, PBS n = 12/DT n = 11) and IL-4 (p = 0.0127, PBS n = 12/DT n = 12), whereas typical type 1cytokines like IL-6 (PBS n = 6/DT n = 6) were not affected. h Analysis of the surface marker CD206 on CD11b+ cells in the spleen by flow cytometry indicates that not only the Th2 response was increased but also the number of alternatively activated anti-inflammatory macrophages (p = 0.0186, PBS n = 13/DT n = 13). i The number of CD8a+ T cells is also significantly elevated in the absence of Treg cells during CP (p = 0.0206, PBS n = 13/DT n = 12). All data were presented as means ± SEM, statistically significant differences were tested by unpaired two-tailed students t-test for independent samples and significance levels of p < 0.05 are marked by an asterisk (c, d, g, h, i). Source data are provided as a Source Data file.

Fig 3: mPTCL overexpresses GATA3 and Th2 genes, resembling molecular defined PTCL-GATA3 FACS-sorted TCRß+ CD8- TCRvß8mid ICOShi mPTCL cells and C57bl/6J CD4+ T-cells were RNA sequenced. Differential expression of T-cell subset-associated genes were analyzed (n = 6). GSEA analysis of 23 gene signatures, representative of Th1, Th2, Th17, Tfh, Treg, T effector memory, and T central memory cells. Significantly enriched signatures are plotted (where q < 0.05—indicated by the vertical line). FDR Q-values can be found in Dataset EV2.Gene enrichment plot for the Th2_Bindea gene signature.Expression levels of GATA3 in wildtype CD4+ T-cells versus CD45.2+ gated PTCL cells expanded in CD45.1 congenic hosts (n = 4). P = 7.19E-05. Representative histograms are shown against an isotype-stained mPTCL control (dashed line). Dotted line represents the isotype control stain for CD4 T-cells. Data are representative of three experiments. Bars represent SEM.Average log2 expression levels of Th2 (Gata3, Ccr4), Th1 (Tbx21), and Treg (Foxp3) genes in purified TCRvß8+ mPTCL cells versus wildtype C57bl/6J CD4+ T-cells (n = 6). Adjusted P = 5.08E-07 (Gata3), P = 9.38E-07 (Ccr4), P = 2.87E-08 (Tbx21), P = 6.75E-10 (FoxP3).The transcriptome of mPTCL spleens (n = 6) were analyzed by gene set variance analysis (GSVA) for enrichment of representative genes of molecular defined human PTCL subgroups (Iqbal et al 2014). A positive score indicates enrichment in a single sample of mPTCL-infiltrated spleen. Scores are shown in truncated violin plots in which vertical lines represent the 1st quartile, the median, and the 3rd quartile of the data.Log2-normalized expression of Tfh immunophenotypic markers in 51 cases of human PTCL from previously published datasets, with hierarchical clustering by PTCL subtype. “PTCL-Tfh” represents “nodal PTCL with Tfh phenotype.”The GATA3 and BCL6 expression status of 86 cases of human PTCL-NOS as assessed by IHC using data published in Watatani et al (2019). ***P < 0.001.

Fig 4: Blood and pancreatic tissue samples of CP patients contain increased numbers of Treg cells.T cells were isolated from EDTA blood samples of patients (CP) and blood doners (BD) and were analyzed by flow cytometry. a The T cell activation marker CD25 (p = 0.0317, BD n = 10/CP n = 7) was significantly elevated in CP patients compared to BD, whereas the elevation of the surface expression of CD69 (BD n = 10/CP n = 7) did not reach significance. T cell differentiation was analyzed by intracellular transcription factor staining of FOXP3, GATA3 and TBET. b FOXP3+CD25+ Treg cells were significantly elevated in the blood of CP patients (p = 0.0345, BD n = 10/CP n = 7), c in the same manner as GATA3+ Th2 cells (p = 0.0026, BD n = 10/CP n = 7). d Interestingly also TBET+ Th1 cells were significantly increased (p = 0.0478, BD n = 10/CP n = 7). e Representative immunofluorescence labeling images of CD4 and FOXP3 illustrate a high number of CD4+FOXP3+ Treg cells in human CP tissue. f Summary illustration showing how regulatory T cells control organ fibrosis during chronic pancreatitis by suppressing Th2/ILC2-mediated alternative activation of macrophages via IL-4/IL-13. Alternatively activated macrophages in turn induce activation of pancreatic stellate cells (PSC) which contribute to extracellular matrix production and fibrosis. All data were presented as means ± SEM, statistically significant differences were tested by unpaired two-tailed students t-test for independent samples and significance levels of p < 0.05 are marked by an asterisk (a–d). Source data are provided as a Source Data file.

Fig 5: Chronic pancreatitis leads to pancreatic tissue remodeling.a Flow cytometry analysis of isolated leukocytes from pancreatic tissue showed a significant increase of CCR2+ cells (p = 0.0032, con n = 7/CP n = 15). b CD11b+ labeling identified that also cells of the innate immune system infiltrate into the damaged pancreas (p < 0.0001, con n = 7/CP n = 15). c Most of these cells were positive for CD206 (p < 0.0001, con n = 7/CP n = 15) or CD163 (p < 0.0001, con n = 7/CP n = 15), which characterized them as alternatively activated macrophages. d Labeling of CP tissue for CD206 illustrates their increase in pancreatic tissue of mice. A quantative analysis showed that ~9% of all cells were positive for CD206 in CP tissue (p < 0.0001, con n = 5/CP n = 9), scale bars represent 20 µm. e IL-33 expression of CD206+ macrophages was demonstrated by immunofluorescent labeling, scale bars represent 20 µm. f This cytokine is a strong inducer of ILC2s (lin-CD45+CD127+CD90+GATA3+) which were also significantly elevated in the pancreas of CP mice (p = 0.0002, con n = 7/CP n = 15). g The replacement of exocrine pancreas by fibrotic tissue in CP was illustrated by azan blue staining of the pancreas, scale bars represent 50 µm. h Flow cytometry analysis of isolated pancreatic cells for PSC marker proteins CD271 (p < 0.0001, con n = 7/CP n = 15) and GFAP (p < 0.0001, con n = 7/CP n = 15) identified a significant increase. i Serum collagen levels were significantly elevated in CP (p = 0.0433, con n = 7/CP n = 13) and collagen deposition and tissue fibrosis were observed in histology. j Masson Goldner trichrome staining was utilized to evaluate the tissue remodeling during CP, scale bars represent 50 µm. k Area quantification (in%) by pattern quant software illustrates the loss of exocrine and endocrine tissue (red) and the increase of fibrotic area (green) and of free space, which reflects fatty tissue and edema (yellow). l Area quantification of exocrine and endocrine active tissue (p < 0.0001, con n = 6/CP n = 10) versus fibrosis (p < 0.0001, con n = 6/CP n = 10) demonstrated the significant changes in organ architecture. All data were presented as means ± SEM, statistically significant differences were tested by unpaired two-tailed students t-test for independent samples and significance levels of p < 0.05 are marked by an asterisk (a–f, h, i, l). Source data are provided as a Source Data file.