DEV035204 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. Typical RNAi phenotypes of regenerating legs. Typical morphological changes of regenerating legs induced by rdRNAi against various Ds-Ft signalling genes with leg growth. The regenerating legs after amputation at the third instar were observed for the fourth, fifth, sixth instar nymphs and adult of RNAi animals. A normal regenerating leg is shown at the top as a control. The distal enlargement phenotype of Gb’ft, Gb’ds and Gb’wts rdRNAi nymphs at the fourth instar are indicated by asterisks. Tibial spines and tibial and tarsal spurs are indicated by arrowheads and arrows, respectively. Defects found in tarsal segments of Gb’Pc rdRNAi nymphs at sixth instar and adult are indicated by small arrows. An extra-tarsal joint phenotype of the Gb’e(z) rdRNAi adult is indicated by a small arrow. Scale bars: 5 mm and 1 cm for nymphal and adult legs, respectively.

• Supplemental Figure S2 -

  Fig. S2. Effects of rdRNAi on relative mRNA levels of Gb’ft and Gb’ds in regenerating legs. (A) The relative ratio of Gb’ft mRNA in regenerating legs at 2 dpa for control or Gb’ft rdRNAi nymphs at third and sixth instar. The relative ratios of Gb’ft mRNA of regenerating tibial stumps (n=10) for control and Gb’ft rdRNAi nymphs are shown by white and grey boxes, respectively. (B) To confirm that there was no off-target effect, we examined specificity in depletion of mRNA by rdRNAi. Relative ratios of Gb’ft or Gb’ds in regenerating tibial stumps (n=10) at 2 dpa for control, Gb’ft rdRNAi and Gb’ds rdRNAi nymphs at third instar are shown by white, grey and dark grey boxes, respectively. (C,D) We examined consistency in phenotypes obtained by injection of the different sets of dsRNAs. We observed the same ‘short and thick’ phenotype in rdRNAi regenerated legs after injection of dsRNA encoding an N-terminal cadherin domain (C) or a C-terminal cytoplasmic domain (D) of Gb’ft. (E) A relative ratio of Gb’hpo mRNA in regenerating legs at 2 dpa for control and Gb’hpo rdRNAi nymphs at third instar. The relative ratios of Gb’hpo mRNA of regenerating tibial stumps (n=10) for control and Gb’hpo rdRNAi nymphs are shown by white and grey boxes, respectively.

• Supplemental Figure S3 -

  Fig. S3. Expression patterns of Gb’ft, Gb’ds, Gb’fj and Gb’d during Gryllus embryogenesis. Expression patterns of Gb’ft, Gb’ds, Gb’fj and Gb’d by whole-mount in situ hybridization are shown. Gb’ft was expressed in the distal region of the prothoracic limb and in head and abdominal regions at stages 4-5. At stages 7, 9 and 11, Gb’ft was expressed in the antenna, limb, cercus and in each of the abdominal segments. Gb’ds was expressed in the head region and distal regions of the prothoracic limb and cercus at stages 4-5. At stage 7, Gb’ds was expressed in the distal region of the antenna, limb, cercus and rectum. At stages 9-11, Gb’ds was expressed in the distal region of each segment of the antenna and limb, and in distal regions of the cercus and rectum. Gb’fj was expressed in the central region of the thorax, gnathal regions and the abdominal region at stages 4-5. Gb’fj was expressed in the limb, cercus, rectum and in each segment of the thorax and abdomen at stages 7-11, and additionally in the antenna and labrum at stage 9. Gb’d was expressed in the growth zone at stage 4. At stage 5, Gb’d was expressed in the distal region of the prothoracic limb, head and the abdominal region at stage 5. At stages 7-11, Gb’d was expressed in the limb, cercus and rectum. Gb’ft, Gb’ds, Gb’fj and Gb’d were simultaneously expressed in the developing limb. Gb’ft was expressed with a proximal-to-distal gradient, Gb’ds was expressed only in the most distal region, Gb’fj was expressed only in the proximal region, and Gb’d was expressed in the distal region of each leg segment at stages 7-11.

• Supplemental Figure S4 -

  Fig. S4. Scanning electron microscope images of regenerating legs with ‘thick’ phenotypes in Gb’ft rdRNAi nymphs. Scanning electron microscope (SEM) images of a control (A,A′) and Gb’ft rdRNAi (B,B′) regenerating nymphal legs at the sixth instar. Regenerated legs were fixed in 4% paraformaldehyde and 5% glutaraldehyde in PBS overnight and processed for SEM observation according to a standard procedure. Images were captured with an FE-SEM S-4700 (Hitachi). (A,B) Surface bristles on the legs were oriented from proximal to distal in the control and Gb’ft rdRNAi nymphs. (A′,B′) High-magnification images of A and B. Processes of sclerites on the leg surface were oriented from proximal to distal in the control and Gb’ft rdRNAi nymphs. As Ft and Ds are known to be regulators for PCP in Drosophila, the ‘thick’ phenotype in Gb’ft or Gb’ds rdRNAi legs might be caused by defects in PCP. However, the orientation of bristles (A,B) and processes on sclerites (A′,B′) in Gb’ft rdRNAi nymphs were similar to the orientation observed in the control. These results suggest that even though the thickening might be due to a PCP defect, it must be uncoupled from the bristle PCP.

• Supplemental Figure S5 -

  Fig. S5. EdU incorporation assay for proliferation in regenerating legs with ‘long’ leg phenotypes in Gb’ex and Gb’Mer rdRNAi nymphs. Localization of EdU-incorporated cells in whole regenerating tibiae and blastemal regions at 2 dpa for control, Gb’ex and Gb’Mer rdRNAi nymphs at third instar. EdU-positive cells, nuclei and merged cells are shown in green, magenta and yellow, respectively. EdU-positive cells were localized in the epithelial cells under the wound surface in the control regenerating leg; however, they were localized in whole regenerating tibiae in the Gb’ex or Gb’Mer rdRNAi nymphs. The ratio of EdU-positive cells versus nuclei in distal longitudinal sections of Gb’ex rdRNAi and Gb’Mer rdRNAi regenerating legs were 0.61±0.12 (n=5) and 0.60±0.17 (n=7), respectively, suggesting that cell proliferation in the blastema is increased by rdRNAi against Gb’ex or Gb’Mer in comparison with the control leg 0.22±0.03 (n=5) for the control. EdU-positive cells were predominantly localized in the amputated region in the control regenerating leg; however, EdU-positive cells were observed in the whole tibiae in the Gb’ex and Gb’Mer rdRNAi regenerating legs, suggesting that Gb’Ex and Gb’Mer suppress overproliferation in the stump, except for amputated regions.

• Supplemental Figure S6 -

  Fig. S6. Effects of rdRNAi against Gb’ex or GbMer on intercalary regeneration and cell proliferation at the junction. (A) Effects of rdRNAi against Gb’ex or Gb’Mer on intercalary regeneration. Normal (left) and reverse (middle) intercalation in Gb’ex RNAi, and Gb’Mer RNAi nymphs at the fifth instar. Orientation of surface bristles is indicated by arrows. In the Gb’ex or Gb’Mer rdRNAi nymphs, both normal and reverse intercalary regeneration took place. We observed both normal- and reverse-orientated bristles on the surface of the reverse intercalary regenerates (arrows), indicating that regenerated cells were derived from both distal and proximal pieces. (B) Effects of rdRNAi against Gb’ex or Gb’Mer on cell proliferation at the junction for reverse intercalation: EdU incorporation assay in grafting legs between a distal host and a proximal graft in a control, Gb’ex RNAi, and Gb’Mer RNAi nymphs. A white arrow indicates the junction between the host and graft. Proliferating cells are shown in green (EdU), whereas nuclei are shown in red (PI). Merged signals appear yellow. The ratio of EdU-positive cells versus nuclei in distal longitudinal sections of the host and graft were 40.1±13.9 and 55.0±8.6 in Gb’ex rdRNAi regenerating legs (n=5), 40.0±1.7 and 50.8±13.7 in Gb’Mer rdRNAi regenerating legs (n=3), respectively, in comparison with the case of control leg 31.5±7.4 and 23.4±7.1 for host and graft, respectively, in the control (n=4). EdU-positive cells were distributed throughout both host and donor leg segments in the Gb’ex and Gb’Mer RNAi nymphs, whereas positive signals were observed predominantly in the control host cells. These results indicated that both graft and host cells proliferated in the Gb’ex or Gb’Mer rdRNAi legs, supporting the possibility that Gb’ex and Gb’Mer are involved in the directional contact-dependent inhibition of proliferation leading to a distal to proximal re-specification (distal preponderance). g, distal graft; h, host stump.

• Supplemental Figure S7 -

  Fig. S7. Supernumerary leg formation in control and rdRNAi nymphs. (A) After amputation of the left and right tibiae at the same level, when the left donor is grafted to the right host so as to reverse their anteroposterior (AP) or dorsoventral (DV) orientation, supernumerary legs are formed (Mito et al., 2002; Nakamura et al., 2008a). (B) Supernumerary leg formation after three moults subsequent to the transplantation in the control and Gb’ft, Gb’ds, Gb’fj, Gb’d and Gb’ex rdRNAi nymphs. All nymphs except for Gb’Mer rdRNAi nymphs can form supernumerary legs.

• Supplemental Figure S8 -

  Fig. S8. Schematic illustrations of the Ds-Ft steepness model for leg regeneration (continued from Fig. 7). (A) Morphallaxis-like regeneration in the Gb’ft, Gb’ds or Gb’d rdRNAi nymphs after distal amputation. The dependence of leg size on the site of amputation in Gb’ft rdRNAi legs is in agreement with the previous interpretation for Morphallaxis-like regeneration (Fig. 7D). (B) ‘Long’ phenotype in the Gb’ex rdRNAi leg. Downregulation of contact-dependent inhibition of proliferation results in the formation of a longer leg, due to overproliferation. In this case, the threshold value of the slope of the Ds-Ft gradient for the arrest of growth would be lower than the normal threshold value indicated by a dotted line, leading to formation of a longer leg. (C-F) In the case of the intercalary transplantation, we can assume a steep gradient and a reverse-steep gradient would be presented in the junction in normal intercalary regeneration and reverse intercalary regeneration, respectively. Intercalary growth would be expected to cease when the slope of the linear gradient equals the pre-existing one. (C) Normal intercalary regeneration. When a distal graft is transplanted to a proximal host, intercalary regeneration occurs so as to restore the missing portion (34567). After grafting, a steep Ds-Ft gradient would be formed at the junction. The cells of the graft proliferate to restore the pre-existing slope. (D) In the case of Gb’ft rdRNAi legs, no intercalary growth occurs, and the transplanted tibia becomes shorter. As shown in Fig. 7D, the Ds-Ft gradient would be expected to shift down with the same slope so as to make the continuous gradient in which the most distal position of the grafted tibia is the minimum scalar. (E) Reverse intercalary regeneration. When a proximally amputated graft is transplanted to a distally amputated host, reverse intercalary regeneration occurs so as to maintain positional continuity (765). After grafting, a steep, reverse Ds-Ft gradient would be formed at the junction. The cells of the host proliferate to restore the pre-existing slope with the DP direction. (F) In the case of Gb’ft rdRNAi legs, no intercalary growth occurs, although the transplanted tibia becomes longer. As shown in Fig. 7D,E, the normal Ds-Ft gradient, indicated by a dotted line, would shift up with the same slope so as to make the continuous gradient in which the most distal position of the grafted tibia is the minimum scalar. Thus, the rdRNAi phenotypes we observed during leg regeneration are interpreted consistently with the Ds-Ft steepness model for regeneration.