Transcriptomic profiling of ductal cells during beta cell regeneration and validation in larvae

A-B) Enrichment ratio of selected non-redundant signatures of KEGG pathways overrepresented in ductal cells after beta cells ablation (UP – A and DOWN-B) compared to ductal cells without beta cells ablation. Gene Ontology (GO) terms were identified using ORA analysis by WebGestalt (Liao et al., 2019) using the list of DE genes provided by DESeq. The light color for Notch pathway means p-value = 0.11.

C) List of genes associated with the signature of cellular senescence from A-B. Genes related to CaN pathway are in bold.

D) Calcineurin canonical pathway with genes in green that are modulated in transcriptomic data from A-B.

E) Experimental design for regeneration test in larvae. Briefly, after nifurpirinol treatment from 3 to 4dpf, larvae were fixed and analyzed at 4-7-10 and 14 days post treatment (dpt).

F) Graph representing the mean number of mCherry+ beta cells in the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) at 0-4-7-10 and 14 dpt. The gray spheres represent non-ablated conditions and the pink triangles the ablated condition. Data are presented as mean values ± SEM. One-way ANOVA test with Tukey’s multiple comparison test, ****p-value>0.0005.

G) Whole mount fluorescent immunohistochemistry (GFP and mCherry) of the pancreas of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 14dpt. 3D projection (stack) of one non-ablated and one ablated representative samples. The principal islet (PI) and the pancreatic tail are shown. Arrows point out mCherry+ beta cells in the pancreatic tail. Scale 100µM.

H) Experimental design for EdU assay in larvae. After NFP treatment for 3 to 4dpf, larvae were exposed to EdU at 2dpt before fixation for analysis.

I) Whole mount fluorescent immunohistochemistry (GFP and EdU) of the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 2dpt. 3D projection (stack) of one non-ablated and one ablated representative samples. Arrows point out GFP+ duct cells EdU+ in the pancreatic tail. Scale 50µM.

J) Barplot representing the percentage of GFP+ ductal cells which incorporated EdU+ in non-ablated and ablated conditions. Data are presented as mean values ± SD. T-test. **p-value>0.005.

Calcineurin inhibition with CsA increases the ductal regenerative response

A) Experimental design for regeneration test in larvae with CsA treatment. Briefly, after nifurpirinol treatment from 3 to 4dpf, larvae were treated with CsA from 1 to 3dpt and fixed and analyzed at 4-7-10 and 14 days post treatment (dpt).

B) Graph representing the mean number of mCherry+ beta cells in the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) at 0-4-7-10 and 14 dpt. The gray spheres represent non-ablated condition ; the pink triangles represent the ablated condition ; the black squares CsA condition and inverted green triangles ablated + CsA condition. Data are presented as mean values ± SEM. Two-way ANOVA test with Sidak’s multiple comparisons test, *p-value>0.05.

C) Whole mount fluorescent immunohistochemistry (GFP and mCherry) of the pancreas of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 10dpt. 3D projection (stack) of non-ablated and ablated larvae treated with DMSO or CsA representative samples. The principal islet (PI) and the pancreatic tail are shown. Arrows point out mCherry+ beta cells in the pancreatic tail. Scale 100µM.

D) Barplot representing the number of number of mCherry+ beta cells in the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 10dpt. The gray spheres represent non-ablated condition ; the pink triangles represent the ablated condition ; the black squares CsA condition and inverted green triangles ablated + CsA condition. Data are presented as mean values ± SD. Two-way ANOVA with Tukey’s multiple comparison test, *p-value>0.05.

E) Graph representing the mean number of GFP+ neurod1+ cells in the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(neurod1:GFP) at 0-4-7 and 10 dpt. The gray spheres represent non-ablated condition ; the pink triangles represent the ablated condition ; the black squares CsA condition and inverted green triangles ablated + CsA condition. Data are presented as mean values ± SEM. Two-way ANOVA test with Sidak’s multiple comparisons test, **p-value>0.005; *** p-value>0.0005; **** p-value>0.00005.

F) Graph representing the mean number of GFP+ neurod1 EdU+ cells in the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(neurod1:GFP) at 0-4-7 and 10 dpt. The gray spheres represent non-ablated condition ; the pink triangles represent the ablated condition ; the black squares CsA condition and inverted green triangles ablated + CsA condition. Data are presented as mean values ± SEM.

Transgenic mediated overexpression of calcineurin abolish the ductal regenerative response

A) Experimental design for regeneration test in larvae with heat-shocks. Briefly, after nifurpirinol treatment from 3 to 4dpf, four heat shock were performed from 1 to 3dpt and larvae were fixed and analyzed at 14dpt.

B) Whole mount fluorescent immunohistochemistry (GFP and mCherry) of the pancreas of Tg(hsp70:CaNCA); Tg(ins:NTR-P2A-mCherry); Tg(nkx6.1:GFP) larvae at 14dpt. 3D projection (stack) of one non-ablated and one ablated with or without heat-shock representative samples. The principal islet (PI) and the pancreatic tail are showed. Arrows point out mCherry+ beta cells in the pancreatic tail. Scale 100µM.

C) Barplot representing the mean number of mCherry+ beta cells in the pancreatic tail of Tg(hsp70:CaNCA); Tg(ins:NTR-P2A-mCherry); Tg(nkx6.1:GFP) larvae at 14dpt. The gray spheres represent non-ablated condition ; the pink triangles represent the ablated condition ; the black squares heat-shock condition and inverted green triangles ablated + heat-shock condition. Data are presented as mean values ± SD. Two-way ANOVA with Tukey’s multiple comparisons test, *p-value>0.05, *** p-value>0.0005, ns = non-significant.

D) Whole mount fluorescent immunohistochemistry (GFP and mCherry) of the pancreas of larvae at 14dpt. 3D projection (stack) of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) one non-ablated and one ablated representative control samples and Tg(UAS:CaNCA); Tg(ins:NTR-P2A-mCherry) ; Tg(cftr:gal4) one non-ablated and one ablated representative samples. The principal islet (PI) and the pancreatic tail are showed. Arrows point out mCherry+ beta cells in the pancreatic tail. Scale 100µM.

E) Barplot representing the mean number of mCherry+ beta cells in the pancreatic tail of larvae at 14dpt. The gray spheres represent non-ablated Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) condition ; the pink triangles represent the ablated Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) condition ; the black squares non-ablated Tg(UAS:CaNCA); Tg(ins:NTR-P2A-mCherry) ; Tg(cftr:gal4) condition and inverted green triangles ablated Tg(UAS:CaNCA); Tg(ins:NTR-P2A-mCherry) ; Tg(cftr:gal4) condition. Data are presented as mean values ± SD. Two-way ANOVA with Tukey’s multiple comparison test, *** means p-value>0.0005, **** p-value>0.00005, ns = non-significant.

CaN repression potentializes the effect of Notch inhibition on beta cell formation

A) Experimental design for Notch inhibition test in non-ablated condition. Larvae were treated concomitantly with LY411575 (Notch inhibitor) and CsA from 3 to 4dpf and were fixed and analyzed at 6dpf.

B) Graph representing the mean number of mCherry+ beta cells in the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 6 dpf depending the concentration of LY411575. The blue dots represent LY411575; and purple combination of LY411575 and CsA. Data are presented as mean values ± SEM. Two-way ANOVA test with Sidak’s multiple comparison test, *p-value>0.05, **p-value>0.05, ns = non-significant.

C) Whole mount fluorescent immunohistochemistry (GFP and mCherry) of the pancreas of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 6dpf. 3D projection (stack) of one control (without any treatment) ; one CsA-treated ; one LY411757-treated and one with both CsA and LY411575 treated larvae. The principal islet (PI) and the pancreatic tail are showed. Arrows point out mCherry+ beta cells in the pancreatic tail. Scale 50µM.

D) Barplot representing the mean number of mCherry+ beta cells in the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 6dpf. The black dots represent the control; gray CsA treatment ;blue LY411575; and purple combination of LY411575 and CsA. Data are presented as mean values ± SD. Two-way ANOVA with Tukey multiple comparison test, ***p-value>0.0005, ****p-value>0.00005.

E) Graph representing the mean number of mCherry+ beta cells in the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 5-6-7 dpf. The black dots represent the control; gray CsA treatment ;blue LY411575; and purple combination of LY411575 and CsA. Data are presented as mean values ± SEM. Two-way ANOVA test with Sidak’s multiple comparison test, *p-value>0.05.

CaN repression increases the proportion of duct proliferating cells

A) Experimental design for EdU assay in Notch test. Larvae were treated concomitantly with LY411575 (Notch inhibitor) and CsA from 3 to 4dpf and then briefly treated with EdU before fixation and analysis at 4 dpf or at 6dpf.

B) Whole mount fluorescent immunohistochemistry (GFP and EdU) of the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 4dpf. 3D projection (stack) of one control (without any treatment), one with CsA only, one with LY411575 only and one with both CsA and LY411757 representative samples. Scale 50µM.

C) Barplot representing the percentage of GFP+ ductal cells which incorporated EdU+ in pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae for the Notch test. The black dots represent the control; gray CsA treatment ;blue LY411575; and purple combination of LY411575 and CsA. Data are presented as mean values ± SD. T-test. Two-way ANOVA test with Tukey’s multiple comparisons test, *p-value>0.05 ; ****p-value>0.00005 ; ns = non-significant.

D) Barplot representing the number of GFP+ ductal cells which in pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 6dpf for the Notch test. The black dots represent the control; gray CsA treatment ;blue LY411575; and purple combination of LY411575 and CsA. Data are presented as mean values ± SD. T-test. Two-way ANOVA test with Tukey’s multiple comparisons test, *p-value>0.05 ; ****p-value>0.00005 ; ns = non-significant.

E) Experimental design for EdU assay in regeneration. Larvae were treated with nifurpirinol for beta cell ablation from 3 to 4dpf then with CsA from 4 to 5dpf and then briefly treated with EdU before fixation and analysis.

F) Whole mount fluorescent immunohistochemistry (GFP and EdU) of the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 5dpf. 3D projection (stack) of one representative sample of non-ablated or ablated with or without CsA are shown. Scale 50µM.

G) Barplot representing the percentage of GFP+ ductal cells which incorporated EdU+ in pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 5dpf. The gray spheres represent non-ablated condition ; the pink triangles the ablated condition ; the black squares CsA condition and inverted green triangles ablated + CsA condition. Data are presented as mean values ± SD. Two-way ANOVA test with Tukey’s multiple comparisons test, *p-value>0.05 ; ****p-value>0.00005 ; ns = non-significant.

H) Whole mount fluorescent immunohistochemistry (VenusPest and EdU) of the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(tp1:VenusPest) larvae at 5dpf. 3D projection (stack) of one representative sample of non-ablated or ablated with or without CsA are shown. Scale 50µM.

I) Barplot representing the percentage of GFP+ ductal cells which incorporated EdU+ in pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(tp1:VenusPest) larvae at 5dpf. The gray spheres represent non-ablated condition ; the pink triangles the ablated condition ; the black squares CsA condition and inverted green triangles ablated + CsA condition. Data are presented as mean values ± sd. Two-way ANOVA test with Tukey multiple comparisons test, *p-value>0.05 ; ****p-value>0.00005 ; ns means non-significant.

J) Barplot representing the number of VenusPest+ ductal cells which incorporated EdU+ in pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(tp1:VenusPest) larvae at 5dpf. The gray spheres represent non-ablated condition ; the pink triangles the ablated condition ; the black squares CsA condition and inverted green triangles ablated + CsA condition. Data are presented as mean values ± SD. Two-way ANOVA test with Tukey’s multiple comparisons test, **p-value>0.005.

CaN regulation is important in juveniles/adults and necessary for correct glycemia recovery

A) Whole mount fluorescent immunohistochemistry (GFP and mCherry) of the pancreas of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) 2 months old zebrafish at 7dpt. 3D projection (stack) of non-ablated and ablated larvae treated with DMSO or CsA representative samples. The principal islet (PI) and the pancreatic tail are shown. Arrows point out mCherry+ beta cells in the pancreatic tail. Scale 200µM

B-C) Barplot representing the number of number of mCherry+ small secondary islets (=<5 cells) in the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) 2 months old zebrafish at 7 (B) and 10dpt (C). The gray spheres represent non-ablated condition ; the pink triangles represent the ablated condition ; the black squares CsA condition and inverted green triangles ablated + CsA condition. Data are presented as mean values ± SD. Two-way ANOVA with Tukey’s multiple comparison test, *p-value>0.05.

D-E) Barplot representing the glycemia (mg/dL) of Tg(ins:NTR-P2A-mCherry) ; months old adult zebrafish at 7 (D) and 10dpt (E). The pink triangles represent the ablated condition ; the inverted green triangles ablated + CsA condition; the blue squares Tg(hsp70:CaNCA) after heat shocks ; the orange lozenges Tg(UAS:CaNCA); Tg(cftr:gal4). The grey line represents the mean glycemia of controls (non-ablated) fish. Data are presented as mean values ± SD. One-way ANOVA with Tukey’s multiple comparison test, *p-value>0.05; **p-value>0.005.

Model of CaN action on ductal progenitors to regenerate beta cells

A) Under physiological conditions, the behavior of the ductal progenitors is determined by Notch signaling. Calcineurin is active in these progenitors and enable a proper control between proliferation and differentiation. When CaN in repressed, more ductal progenitors enter in the cell cycle (2dpt) and switch to a mode of proliferation leading to differentiation of the two daughter cells (4dpt), as more pro-endocrine cells are formed. The result is a exhaustion of the progenitors and a premature beta cell differentiation (10dpt).

Transcriptomic profiling of ductal cells during beta cell regeneration and validation in larvae

A) Expression of ppp3cca; ppp3ccb ; nfatc3a ; nfatc3b in acinar, alpha, beta, delta or ductal cells population from the zebrafish pancreas [40].

B) Calcineurin (ppp3ccb and ppp3cca) and NFATc3 (nfatc3a and nfatc3b) expression in ductal cells from zebrafish in non-ablated and ablated conditions.

Calcineurin inhibition with CsA increases the ductal regenerative response

A) Whole mount fluorescent immunohistochemistry (mCherry) of the pancreas of Tg(ins:NTR-P2A-mCherry) larvae at 4-7 and 14dpt. 3D projection (stack) of ablated larvae treated with DMSO or CsA representative samples. The principal islet (PI) and the pancreatic tail are shown. Scale 50 or 100µM.

B-C) Barplot representing the number of mCherry+ beta cells in the principal islet (PI) (B) and the number of secondary islets of mCherry+ beta cells in the pancreatic tail (C) of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 10dpt. Data are presented as mean values ± SD. T-test or Two-way ANOVA with Tukey’s multiple comparison test. **p-value>0.005

D) Experimental design for EdU assay in larvae. After NFP treatment for 3 to 4dpf, larvae were exposed to EdU before fixation for analysis.

E) Whole mount fluorescent immunohistochemistry (GFP and EdU) of the pancreas of Tg(ins:NTR-P2A-mCherry) ; Tg(neurod1:GFP) larvae at 2-4-7 and 10 dpt. 3D projection (stack) of non-ablated and ablated larvae treated with DMSO or CsA representative samples. The principal islet (PI) and the pancreatic tail are shown. Scale 50µM.

F-G) Barplot representing the number of gcg+ alpha cells (F); the number of GFP+ sst1.1 delta cells (G); of Tg(ins:NTR-P2A-mCherry) ; Tg(sst1.1:GFP) larvae at 10dpt. Gcg was detected by IHC. The gray spheres represent non-ablated condition ; the pink triangles represent the ablated condition ; the black squares CsA condition and inverted green triangles ablated + CsA condition. Data are presented as mean values ± SD. Two-way ANOVA with Tukey’s multiple comparison test. **p-value>0.005

CaN repression potentializes the effect of Notch inhibition on beta cell formation

A) Experimental design for Notch inhibition test in non-ablated condition. Larvae were treated concomitantly with LY411575 (Notch inhibitor) and FK506 from 3 to 4dpf and were fixed and analyzed at 5dpf.

B) Barplot representing the mean number of mCherry+ beta cells in the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 5dpf. The black dots represent the control; gray FK506 treatment ;blue LY411575; and purple combination of LY411575 and FK506. Data are presented as mean values ± SD. Two-way ANOVA with Tukey multiple comparison test, *p-value>0.05,**p-value>0.005, ****p-value>0.00005.

C) Experimental design for Notch inhibition test in non-ablated condition. Larvae were heat-shocked and then directly treated with LY411575 (Notch inhibitor) from 3 to 4dpf and were fixed and analyzed at 6dpf.

D) Barplot representing the mean number of mCherry+ beta cells in the pancreatic tail of Tg(hsp70:CaNCA); Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 5dpf. The black dots represent the control; gray heat-shock ;blue LY411575; and purple combination of LY411575 and heat-shock. Data are presented as mean values ± SD. Two-way ANOVA with Tukey multiple comparison test, ****p-value>0.00005.

E) Barplot representing the mean number of mCherry+ beta cells in the pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 6dpf after Notch inhibition (LY411575), CaN inhibition (CsA) and NFATc inhibition (CHIR99021). Data are presented as mean values ± SD. Two-way ANOVA with Tukey multiple comparison test, *p-value>0.05,**p-value>0.005, ****p-value>0.00005.

CaN repression increases the proportion of duct proliferating cells

A-B) Barplot representing the number of GFP+ ductal cells which in pancreatic tail of Tg(ins:NTR-P2A-mCherry) ; Tg(nkx6.1:GFP) larvae at 4dpf for the Notch test. (A) Mild Notch inhibition with LY411575 5µM and (B) stronger Notch inhibition with LY411575 15µM. The black dots represent the control; gray CsA treatment ;blue LY411575; and purple combination of LY411575 and CsA. Data are presented as mean values ± SD. T-test. Two-way ANOVA test with Tukey’s multiple comparisons test, *p-value>0.05 ; ****p-value>0.00005 ; ns = non-significant.

CaN regulation is important in juveniles/adults and necessary for correct glycemia recovery

A) Barplot representing the glycemia (mg/dL) of Tg(ins:NTR-P2A-mCherry) ; months old adult zebrafish at 14dpt. The pink triangles represent the ablated condition ; the inverted green triangles ablated + CsA condition; the blue squares Tg(hsp70:CaNCA) after heat shocks. The grey line represents the mean glycemia of controls (non-ablated) fish. Data are presented as mean values ± SD.

B) Barplot representing the glycemia (mg/dL) of Tg(ins:NTR-P2A-mCherry) ; and Tg(UAS:CaNCA); Tg(cftr:gal4) in non-ablated adult zebrafish. Data are presented as mean values ± SD.