Fig 2: Expression of DPP6 in human islets and EndoC-ßH1 cells and other tissues evaluated by qPCR and histology. (A) Quantitative RT-PCR (qPCR) of DPP6 mRNA expression (detecting a shared sequence among all DPP6 splice variants) in EndoC-ßH1 cells (n = 5) and human pancreatic islets (n = 4) that were exposed or not to cytokines (IL-1ß + IFN-?) for 48 h, as compared to pancreatic exocrine tissue (n = 6), two exocrine cell lines (Capan-2 (n = 3) and PANC (n = 3)), and 14 other non-pathological human tissues (n = 1). (B) Immunoblot of EndoC-ßH1 cells under control conditions or following a 48h exposure to cytokines (IL-1ß and IFN-?), with alpha-tubulin as a reference protein. A representative figure is shown at the top and densitometric analysis at the bottom (n = 5), this figure displays a cropped blot, the full-length version is included in supplementary figure 8; (C–F) Immunocytochemistry of EndoC-ßH1 cells. (C) An overlay with cells stained with an anti-DPP6 monoclonal antibody (mAb, red), co-stained for insulin (green) and Hoechst in blue. The separate channels are displayed in (D) insulin (green) and (E) DPP6 (red) (n = 3). The mostly surface localization of DPP6 (red) can be observed in (F) (n = 4), with blue signals indicating Hoechst staining. The negative staining control of EndoC-ßH1 cells (without the DPP6 antibody) is placed in the top right corner of panel F. White scale bar represents 1 µm. RT-qPCR and the western blot data are presented as means ± SEM. Paired and unpaired two-way ANOVA (indicated with * and $, respectively), and unpaired one-way ANOVA (indicated with #) with Šídák correction for multiple comparisons; *, $ and # p = 0.05 as indicated by bars.

Fig 3: Localization of DPP6 expression in human pancreas. (A–E) A representative human pancreas stained for DPP6 (A, red), insulin (B, white), somatostatin (C, green); overlay of DPP6 (red), insulin (white) and somatostatin (green) (D); overlay of DPP6 (red) and somatostatin (green) (E); the data indicate co-staining of insulin and DPP6, but not somatostatin and DPP6; (F–J) A representative human pancreas stained for glucagon (F, green), DPP6 (G, red), insulin (H, white); DPP6 (red) and glucagon (green) overlay (I); overlay of DPP6 (red), insulin (white) and glucagon (green) (J), the data indicate co-staining of both insulin and glucagon with DPP6; (K) Morphometric quantification of DPP6 area in pancreata from T1D patients as compared to control, non-diabetic individuals (n = 3). (L–P) A representative human pancreas from a subject with long-term type 1 diabetes (16 years of disease) stained for glucagon (L, green), DPP6 (M, red), insulin (N, white); Hoechst (O, blue); (P) overlay of DPP6 (green), glucagon (red), insulin (white) and Hoechst (blue), indicating that in the absence of insulin positive cells, the remaining glucagon positive cells co-stain for DPP6. In total, 3 pancreata from normoglycemic individuals and 3 from type 1 diabetes subjects were analysed. White scale bar represents 20 µm.

Fig 4: Flow cytometry analysis of nanobody cell binding. The cell binding of the 4hD29 nanobody (Nb) was evaluated by flow cytometry. (A–D) 4hD29 (red) recognizes human DPP6 in different cell types (HsDPP6); (A) 4hD29 (red) labelled transiently-transfected CHO cells overexpressing DPP6, where neither the irrelevant control Nb (green) nor the secondary control antibodies only (blue) stained the cells (n = 3); (B) non-transfected CHO cells labelled as in A (n = 3); (C) The Nb 4hD29 (red) recognizes DPP6-positive human Kelly neuroblastoma (n = 4) and (D) EndoC-ßH1 cells (n = 5), whereas the secondary antibodies only (blue) or an irrelevant control Nb (green) do not. (E,F) 4hD29 (red) binds to endocrine (TSQ+/Rh-) (E), but not to exocrine tissue (TSQ-/Rh+) (F) of dissociated human pancreas (n = 4). Background staining with secondary staining control is indicated in blue. (G) Overview of the gating strategy for endocrine (TSQ+/Rh-) and exocrine cells (TSQ-/Rh+) analysed in (E,F). The median fluorescence intensity (MFI) was calculated for Kelly neuroblastoma (H) and EndoC-ßH1 cells (I). Delta MFI values were calculated to compare the endocrine (TSQ+/Rh-) and exocrine populations (TSQ-/Rh+), showing that 4hD29 has an increased binding in endocrine cells as compared to exocrine cells (J). Unpaired (E,F) or paired (J) Student’s t-tests were performed to compare two groups; *p = 0.05, **p = 0.01.