Files in this Data Supplement:
Fig. S1. Notch signaling reporter lines based on the TP1bglob module. (A) Tg(Tp1:H2BmCherry) drives expression of H2BmCherry, which has a very long half-life, in NRCs. Tg(Tp1:VenusPest) drives expression of a destabilized VenusPEST fluorescent protein, which has a short half-life, in NRCs. (A-C) Tg(Tp1:H2BmCherry); Tg(Tp1:VenusPest) larvae were treated with DMSO or DAPT from 4 to 6 dpf. (B) DMSO-treated larvae are Tg(Tp1:H2BmCherry)+ and Tg(Tp1:VenusPest)+. (C) DAPT-treated larvae are Tg(Tp1:H2BmCherry)+ but strongly downregulate or even lose Tg(Tp1:VenusPest) expression in most tissues. Arrowheads indicate the IPD. (D,D′) An individual Tg(Tp1:GFP) larva was imaged at 3.5 dpf, incubated in DAPT until 5.5 dpf and imaged again. (E,E′) An individual Tg(Tp1:VenusPest) larva was imaged at 3.5 dpf, incubated in DAPT until 5.5 dpf and imaged again. Tg(Tp1:VenusPest) fluorescence is strongly downregulated compared with Tg(Tp1:GFP) fluorescence in D. Arrowheads indicate the IPD. (F) Embryos were obtained from an in-cross of mibta52b heterozygous parents. Wild-type Tg(Tp1:GFP) embryos at 2.5 dpf show strong Tg(Tp1:GFP) expression in multiple organs, whereas mib homozygous mutant embryos, recognized by their curly tail phenotype, failed to express Tg(Tp1:GFP). (G) Tg(Tp1:H2BmCherry) larvae were fixed at 7 dpf and stained for 2F11 (blue) to mark IPD cells (white arrowheads). The Tg(Tp1:H2BmCherry)+ cells (red) in the pancreas correspond to the 2F11+ IPD cells. 2F11+ and Tg(Tp1:H2BmCherry)+ cells ventral to the IPD are gut cells. All images are lateral views, anterior towards the right, ventral towards the bottom, except in F, where anterior is towards the top and ventral towards the right.
Fig. S2. Analysis of Tg(Tp1:VenusPest) expression after a brief period of Notch signaling downregulation. Tg(Tp1:H2BmCherry); Tg(Tp1:VenusPest) larvae were treated with DMSO or LY411575 from 4 to 4.5 dpf. (A) In DMSO-treated larvae, the vast majority of pancreatic NRCs were Tg(Tp1:H2BmCherry)+ and Tg(Tp1:VenusPest)+. (B,B′) Larvae treated with 10 µM LY411575 showed variable donwregulation of Tg(Tp1:VenusPest) expression. (B) Seven out of 10 larvae (Class1) showed minor downregulation of Tg(Tp1:VenusPest) expression. In these larvae, only some Tg(Tp1:H2BmCherry)+ were Tg(Tp1:VenusPest)− (white arrowheads). (B′) Three out of 10 larvae (Class 2) exhibited significant downregulation of Tg(Tp1:VenusPest) expression and multiple Tg(Tp1:H2BmCherry)+ were Tg(Tp1:VenusPest)− compared with DMSO controls (white arrowheads). (C) Larvae treated with 50 µM LY411575 showed consistent downregulation of Tg(Tp1:VenusPest) expression compared with DMSO controls. In these larvae, many Tg(Tp1:H2BmCherry)+ cells were Tg(Tp1:VenusPest)− (white arrowheads). All images are lateral views, anterior towards the top, ventral towards the right; white arrowheads indicate the IPD. The NRCs ventral to the IPD are gut cells. Scale bars: 20 µm.
Fig. S3. Endocrine differentiation of NRCs leads to cell cycle exit; NICD overexpression inhibits cell cycle progression. (A-C) Tg(Tp1:H2BmCherry); Tg(neurod:EGFP) larvae were used to label NRCs (red) and follow endocrine differentiation (green). (A-C) Larvae treated with DMSO or 10 µM LY411575 from 4 to 6.5 dpf (60 hours); EdU was added after a 36-hour delay (5.5 to 6.5 dpf). (A) DMSO-treated control larvae. Yellow arrowheads indicate EdU+ NRCs and white arrowheads indicate one endocrine cell along the IPD (projection of several planes). (B) LY411575-treated larvae show an increase in the number of new endocrine cells (green and red). For clarity, the positions of the endocrine cells are indicated by white outlines. Note that these endocrine cells are EdU−. EdU incorporation, however, can be observed in the NRCs that do not undergo endocrine differentiation (yellow arrowheads) (projection of several planes). (C) Quantification of average proportion of EdU+ cells for A and B. Blue bar: on average 11% of NRCs in DMSO controls were EdU+ (100 cells counted in 5 larvae). Red bar: in LY411575-treated larvae, on average 60% of NRCs not undergoing endocrine conversion were EdU+ (140 cells counted in 7 larvae) (P<0.001). Green bar: only 8% of new endocrine cells were EdU+ (182 cells counted in 4 larvae), indicating that endocrine differentiation leads to a proliferation block. (D) Tg(Tp1:eGFP) larvae were fixed at 4.5 dpf and stained for Pdx1. Pdx1 immunoreactivity is high in the endocrine cells in the PI (yellow arrowhead). Tg(Tp1:eGFP)+ NRCs extending posterior to the PI can be clearly identified based on Pdx1 immunoreactivity (white arrowheads). (E) Larvae obtained from crossing hemizygous Tg(hsp:Gal4) to hemizygous Tg(UAS:myc-Notch1a-intra) parents. Pdx1 immunoreactivity (red) marks NRCs. For clarity, all Myc+ NRCs (red and green) are outlined in green in E′′ and E′′′. Yellow arrowheads indicate several Myc+ NRCs. None of the Myc+ NRCs are EdU+, indicating cell cycle block. By contrast, many neighboring Myc− NRCs are EdU+ (for clarity outlined in white; a white arrowhead indicates several Myc− NRCs) (projection of several planes). We found that 2.5% (118 cells counted, n=9 larvae) of Myc+ NRCs were EdU+ and 49% of the Myc− NRCs were EdU+ (235 cells counted, n=9 larvae). All images are lateral views, anterior towards the top, ventral towards the right. Scale bars: 20 µm. Error bars: s.d.
Fig. S4. Validation of Tg(hsp:ZdnSu(H)-myc; hsp:GFP) as a tool for downregulation of Notch signaling. (A) Embryos obtained from in-crossing (Tg(hsp:ZdnSu(H)-myc; hsp:GFP) parents were heat shocked at 24 hpf, fixed at 30 hpf and stained for Myc. The majority of the cells in the neural tube co-express Myc and GFP (projection of several planes). (B-D) Embryos obtained from crossing Tg(hsp:ZdnSu(H)-myc; hsp:GFP) and Tg(Tp1:H2BmCherry) parents were heat shocked at 12 hpf and examined at 30 hpf for GFP and Tg(Tp1:H2BmCherry) expression. (B) A single plane through the neural tube. GFP+ cells show no or low expression of Tg(Tp1:H2BmCherry), indicating that they had impaired capacity to activate Notch signaling. To visualize the cells better, high-magnification single planes are shown in B′-B′′′. White arrowheads indicate individual Tg(Tp1:H2BmCherry)+ cells. Yellow arrowheads indicate individual GFP+ cells. Note that in most cases, the two cell types are mutually exclusive. Analogous observations were made in the brain (C) and retina (D). The images in A and C are dorsal views, anterior towards the top. The images in B and D are lateral views, anterior towards the top, ventral towards the right. Scale bars: 20 µm.
Fig. S5. Behavior of NRCs under a moderate and a strong downregulation of Notch signaling. (A-C′′′) Tg(Tp1:H2BmCherry); Tg(neurod:EGFP) larvae were treated with (A) DMSO, (B) 1 µM LY411575 and (C) 10 µM LY411575. The chemicals were replaced every 24 hours. Compared with DMSO controls, the larvae treated with 1 µM LY411575 exhibited increased numbers of Tg(Tp1:H2BmCherry)+ cells (yellow arrowheads). A small proportion of the NRCs differentiated into endocrine cells, as assessed by Tg(neurod:EGFP) expression (white arrows). A and B show low-magnification projections corresponding to Fig. 4G,H, respectively. (C) In the larvae treated with 10 µM LY411575, the majority of Tg(Tp1:H2BmCherry)+ cells differentiated into endocrine cells, as assessed by Tg(neurod:EGFP) expression. (C′′) To better visualize these endocrine cells, a high-magnification single plane through one of the secondary islets (white arrow in C and C′) is shown. (C′′′) Orthogonal view (z-axis) through the islet across the green line in C′′. Note that this cell cluster spans several cell-layers. (D) Quantification of the average number of Tg(Tp1:H2BmCherry)+ cells (red bars) and Tg(Tp1:H2BmCherry); Tg(neurod:EGFP) double-positive cells (green bars) from the experiments in A-C. Tg(Tp1:H2BmCherry)+ and Tg(neurod:EGFP)+ cells posterior to the PI were counted. Larvae treated with 1 µM LY411575 exhibited a significant increase in the total number of Tg(Tp1:H2BmCherry)+ posterior to the PI (118 cells, s.d.=21 cells, n=20 larvae) compared with DMSO controls (71 cells, s.d.=19 cells, n=18 larvae) (P<0.0001); and in these larvae, NRC-derived endocrine cells represented on average 17% (s.d.=7.26%, n=19 larvae) of the total number of NRCs posterior to the PI. In larvae treated with 10 µM LY411575, NRC-derived endocrine cells represented on average 79% (s.d.=12%, n=14 larvae) of the total number of Tg(Tp1:H2BmCherry)+ cells posterior to the PI. These larvae exhibited a more modest increase in the average number of Tg(Tp1:H2BmCherry)+ posterior to the PI (101 cells, s.d.=19.5 cells, n=14 larvae) compared with DMSO controls (P<0.001). (E-G) Tg(Tp1:H2BmCherry); Tg(Tp1:VenusPest) larvae were treated with (E) DMSO, (F) 1 µM LY411575 and (G) 10 µM LY411575 from 4 to 6 dpf. The chemicals were replaced after 24 hours. (E) In the DMSO-treated larvae, the vast majority of NRCs are Tg(Tp1:H2BmCherry)+ and Tg(Tp1:VenusPest)+. (F-F′) In 12 out of 20 larvae treated with 1 µM LY411575 (Class 1), a significant proportion of Tg(Tp1:H2BmCherry)+ cells retained Tg(Tp1:VenusPest) expression, albeit at lower levels compared with DMSO controls. In the rest of the larvae (Class 2), only few Tg(Tp1:H2BmCherry)+ cells retained Tg(Tp1:VenusPest) signal. (G) Larvae treated with 10 µM LY411575 showed complete loss of Tg(Tp1:VenusPest) expression in the pancreatic Tg(Tp1:H2BmCherry)+ cells. All images are lateral views, anterior towards the top, ventral towards the right. Two fields of view were combined, as indicated by a white line to display the full length of the pancreas in A,B. Scale bars: 20 µm. Error bars indicate s.d.
Movie 1. Live imaging of Tg(Tp1:H2BmCherry); Tg(neurod:EGFP) larva at 3.5 dpf. Tg(Tp1:H2BmCherry) expression labels NRCs (red) and Tg(neurod:EGFP) expression labels endocrine cells and their direct progenitors (green). An NRC (white arrowhead) upregulates Tg (neurod:EGFP) expression, suggesting onset of endocrine lineage differentiation.
Movie 2. Live imaging of Tg(Tp1:H2BmCherry); Tg(neurod:EGFP) larva at 4.5 dpf. Tg(Tp1:H2BmCherry) expression labels NRCs (red) and Tg(neurod:EGFP) expression labels endocrine cells and their direct progenitors (green). Two endocrine cells move towards each other and coalesce.
Movie 3. Live imaging of Tg(Tp1:H2BmCherry); Tg(neurod:EGFP) larva at 3.5 dpf. Tg(Tp1:H2BmCherry) expression labels NRCs (red) and Tg(neurod:EGFP) expression labels endocrine cells and their direct progenitors (green). Dividing NRCs are indicated by white dots. Division planes are oriented such that both daughter cells remain part of the duct.
Movie 4. Confocal stack of secondary islets stained for glucagon and insulin. Tg(Tp1:H2BmCherry) larvae treated with 50 µM LY411575 from 3.5 to 6.5 dpf and stained for glucagon (blue) and insulin (green) at 7.5 dpf. Tg(Tp1:H2BmCherry) expression labels NRCs. Confocal sections through two secondary islets corresponding to Fig. 2I′′ are shown.
