Segregation of striated and smooth muscle lineages by a Notch-dependent regulatory network
1 Department of Medical Neurobiology, IMRIC and ELSC, Hebrew University of Jerusalem-Hadassah Medical School, Jerusalem 9112102, Israel
2 Present address: Cardiovascular Research Center, MGH Harvard Medical School, Charlestown 02129, MA, USA
BMC Biology 2014, 12:53 doi:10.1186/s12915-014-0053-9Published: 12 July 2014
Additional file 1: Figure S1.:
Inhibition of Notch signaling induces myogenesis. (A) HEK293 cells were co-transfected with a Dll1-encoding plasmid, with the Hes1 reporter to assay for Notch activity and with a RFP plasmid to assess transfection efficiency. Results represent the mean ± SEM of triplicate cultures (see Methods). Dll1 significantly reduced Hes1 promoter activity (*P <0.05). (B-C) Confocal images of somites electroporated with Dll1 (red) and with pTP-1Venus, a CSL-dependent reporter that reflects Notch signaling activity (green). Cells expressing Dll1 do not activate this Notch reporter cell autonomously, but rather activate it in adjacent cells. (D-G) Electroporation of the lateral DM with control GFP (D-D”), dnMAML1 (E), or Dll1 (F). (D) At 40 hours post-electroporation, cells transfected with control GFP were found in myotome (M), sclerotome (Scl) and as SM within blood vessel walls (arrows and higher magnification in D’ and D”, N = 6). (E,F) dnMAML1 and Dll1 abrogate cell migration through the sclerotome and subsequent vascular development while promoting the myotomal lineage (N = 4 and 4, respectively). Lateral DM is outlined by dotted lines. (G) Quantification (mean ± SEM). *P ≤0.05. Bar: (D-F) 50 μm; (D’, D”) 35 μm. CV, cardinal vein; DM, dermomyotome; meso, mesonephros; SEM, standard error of the mean.
Format: JPEG Size: 1.6MB Download file
Additional file 2: Figure S2.:
Late activation of Notch in lateral DM-derived progenitors is not sufficient for upregulating SM markers while in the ventral sclerotome. (A,A’) A Tet-GFP-control plasmid was activated twenty hours post-electroporation to the lateral DM for a period of eight hours. Note that the majority of GFP + cells still reside in the lateral DM and only a few delaminated into the intermediate zone between DM and myotome or into the ventral sclerotome (arrows) (N = 9). (B,B’) A Tet-aN2-GFP plasmid was activated twenty hours post-electroporation to the lateral DM for a period of eight hours. Most cells delaminated from the lateral DM and are located in the myotome as desmin/SMA-negative cells (B’) or in the ventral sclerotome (arrows) (N = 8). In both control and Notch-treated cells located in the ventral sclerotome no ectopic SMA/desmin immunostaining is evident. Hoechst nuclear staining is in grey. The lateral DMs are outlined by a dashed white line. Bar: 50 μm. CV, cardinal vein; DM, dermomyotome; M, myotome; Scl, sclerotome; SM, smooth muscle; SMA, smooth muscle actin.
Format: JPEG Size: 760KB Download file
Additional file 3: Figure S3.:
Expression profiles of Id2, Id3, FoxC2 and Snail1 to developing somites. Id2 (A, E, I), Id3 (B, F, J), FoxC2 (C, G, K) and Snail1 (D, H, L) are expressed in the presumptive lateral domain of epithelial somites at E2.0 (dashed lines and yellow arrows in A-D) and in the lateral DM at E2.5 (E-H, black arrows), but are downregulated at E3.5 in the dissociated DM (I-L, black arrows) although the VLL is still epithelial (dashed lines). Note additional sites of expression of some of the transcripts in sclerotome and in the dorsomedial lip of the DM. Bar: (A-D) 50 μm, (E-L) 100 μm. CV, cardinal vein; DM, dermomyotome; E, embryonic day; M, myotome; VLL, ventro-lateral lip.
Format: JPEG Size: 4.3MB Download file
Additional file 4: Figure S4.:
Quantification of labeled cells in sclerotome. (A) Overexpression of Id2, Id3, Foxc2 and Snail1 enhances the proportion of cells encountered in the sclerotome. Refers to Figure 3. (B) Knockdown of Id2/3 and FoxC2. dsRNA to Id2/3 reduce the proportion of labeled cells in sclerotome. Refers to Figure 4A-D. (C) Knockdown of Snail1 reduces the proportion of labeled cells in sclerotome. Refers to Figure 4E-G. In all cases, GFP + cells in sclerotome were negative for desmin and smooth muscle actin. Results are expressed as mean ± SEM. **P ≤0.01. dsRNA, double-stranded RNA; SEM, standard error of the mean.
Format: JPEG Size: 273KB Download file
Additional file 5: Figure S5.:
Short-term over-expression of Foxc2 or Snail1 enhance delamination of lateral dermomyotome (DM) progenitors. (A) Cells transfected with a control GFP plasmid are mainly located as epithelial cells within the lateral DM and some have translocated into the intermediate zone, between the ventro-lateral lip (VLL) and the myotome (M) 16 hours post-electroporation (N = 12). (B) Foxc2 overexpression. Note depletion of cells from the DM and the presence of labeled cells in an intermediate zone and in the sclerotome (Scl) (N = 17). (C,C’) Control cells marked with GFP reside as epithelial cells within the lateral DM and some have translocated into the myotome (M) 16 hours post-electroporation; note the epithelial morphology of the lateral DM marked by ZO-1 (white bracket) (N = 5). (D,D’) Sixteen hours of enhanced Snail1 activity compromises the epithelial morphology of treated cells, as indicated by the absence of the epithelial marker ZO-1 (white bracket). These cells exit the epithelial sheet but fail to contribute to the myotome (N = 5). Solid and dashed lines demarcate DM and myotome, respectively. (E) Quantifications (mean ± SEM) of over-night exposure to Foxc2 and Snail1. **P-value ≤0.01. Bar: 50 μm. CV, cardinal vein; SEM, standard error of the mean.
Format: JPEG Size: 1.8MB Download file
Additional file 6: Figure S6.:
Attenuation of Id2/3 activities causes precocious differentiation of myocytes in myotome. (A-C) Dorsal views of whole mount segments. (A) Cells treated with control dsRNA against LacZ are still located within the lateral DM (arrows) 16 hours post-electroporation whereas cells that received dsRNA against Id2/3 had already differentiated into myofibers (B, arrows). (C) Co-transfection of Id2/3 along with dsRNAs to Id2/3 rescues the knock-down phenotype, as cells fail to differentiate into fibers (arrows) and instead migrate ventro-laterally (arrowhead). (D-D”) Transverse section 40 hours post-electroporation showing that Id2/3 over-expression rescues the Id2/3 knock-down phenotype (N = 5). (D’,D”) High magnification of the inset in D showing the cardinal vein exhibiting co-localization of a labeled cell with SM markers, desmin and SMA. Endothelial cells were visualized with the Qh1 antibody (blue). (E) Quantification (mean ± SEM) of labeled cell distribution (N = 16 for control; N = 19 for Id2/3; N = 5 for dsRNA Id2/3; N = 5 for dsRNA Id2/3 + Id2/3). *P ≤0.05; **P ≤0.01. Bar: (D) 100 μm; (D’,D”) 12.5 μm. CV, cardinal vein; Des, desmin; dsRNA, double-stranded RNA; M, myotome; SEM, standard error of the mean; SMA, smooth muscle actin.
Format: JPEG Size: 2.1MB Download file
Additional file 7: Figure S7.:
FoxC2 knock-down induces myocyte differentiation at the expense of vascular fates. (A-B) Dorsal view of whole-mount segments electroporated with control-GFP (A) or with FOXC2-engrailed (En) (B) and incubated for 16 hours. Only a few elongating myocytes are observed in the control (arrows, N = 8), whereas FOXC2-En promoted premature fiber formation (arrows in B, N = 6). (C,D) Transverse sections 40 hours post-electroporation with (C) control-GFP or (D) FOXC2-En. In control embryos cells are apparent in both vascular (as SM cells, see C’ and C”) and myotomal domains (N = 4). (C’,C”) High magnification and channel breakdown of the inset showing the cardinal vein (CV) exhibiting co-localization of a GFP-labeled cell with Des/SMA immunoreactivity (arrow). (D) In contrast, knocking-down FoxC2 promotes myotomal colonization while hindering development of SM cells (N = 5). (E) Quantification (mean ± SEM) of labeled cell distribution. (F,G) Dorsal view of whole-mount segments 16 hours post-electroporation of (F) control dsRNA LacZ or (G) dsRNA FoxC2. In control embryos cells are in the lateral DM (arrows, N = 3) whereas in the presence of dsRNA FoxC2 myotomal cells begun differentiating (arrows in G, N = 3). (H) Transverse section showing that co-electroporation of full-length Foxc2 along with dsRNA FoxC2 rescues the FoxC2 knock-down phenotype by inducing cell migration through the sclerotome and SM fate (arrow, see H’, H”) and lack of myotome colonization (compare to Figures 3A,C and 4C). (H’, H”) High magnification and channel breakdown of the inset in H showing the CV exhibiting co-localization of a labeled cell with Des/SMA + immunoreactivity. Qh1 antibody (blue) labels endothelial cells. (I) Quantification (mean ± SEM) of labeled cell distribution (N = 22 for control dsRNA; N = 16 for Foxc2; N = 16 for dsRNA FoxC2; N = 7 for dsRNA FoxC2 + Foxc2). *P ≤0.05; **P ≤0.01. Bar: (C,D) 50 μm; (H) 35 μm; (C’,C”,H’,H”) 17.5 μm. Des, desmin; dsRNA, double-stranded RNA; M, myotome; SEM, standard error of the mean; SMA, smooth muscle actin.
Format: JPEG Size: 1.8MB Download file
Additional file 8: Figure S8.:
Attenuation of Snail1 signaling promotes myocyte differentiation and retention of cells in the lateral epithelium. (A-C) Dorsal view of whole-mount segments. (A) A control scrambled siRNA was electroporated into the lateral somites. At 16 hours post-electroporation few labeled cells have differentiated into myofibers. Other cells, while still residing in the lateral DM have lost their epithelial morphology. (B) In the presence of siRNA to Snail1 both maintenance of transfected epithelial cells within the lateral DM as well as premature myogenesis are apparent. (C) Snail1 over-expression rescues the siRNA-Snail1 knock-down phenotype, as fewer fibers and more scattered cells are observed (arrow). (D-F) Transverse sections 40 hours post-electroporation. (D) Control scrambled siRNA. Labeled cells are distributed in myotome (M), sclerotome and approach the cardinal vein (CV). (E) Knocking-down Snail1 maintains cells within the lateral DM as epithelial cells. (F) Over-expression of Snail1 rescues the effect of siRNA-Snail1 knock-down. The lateral DM is depleted of labeled cells and a notable fraction is able to migrate towards target sites (arrows). (F’,F”) High magnification of the inset in F showing the CV exhibiting co-localization of a labeled cell with SM markers. Endothelial cells were visualized with the Qh1 antibody (blue). (G) Quantification (mean ± SEM) of labeled cell distribution (N = 7 for control; N = 7 for Snail1; N = 6 for siRNA Snail1; N = 3 for Snail1 + siRNA Snail1). *P ≤0.05; **P ≤0.01. Bar: (D-F) 50 μm, (F’-F”) 32 μm. es, desmin; DM, dermomyotome; SEM, standard error of the mean; siRNA, small interfering RNA; SMA, smooth muscle actin.
Format: JPEG Size: 2.1MB Download file
Additional file 9: Table S1.:
Interactions between genes of the regulatory network. Genes depicted in the horizontal axis (bold) were electroporated and genes or reporters depicted along the vertical axis were analyzed for expression by in situ hybridization, immunohistochemistry or activity (see text for details).
Format: PDF Size: 104KB Download file
This file can be viewed with: Adobe Acrobat Reader
Additional file 10: Figure S9.:
Control-GFP to the central sheet does not affect transcription of Id2, Id3, or FoxC2. (A-C) GFP was electroporated into the presumptive central sheet of E2.0 embryos and re-incubated for an additional 10 hours (left panels). No changes in transcription were detected by in situ hybridization with probes against Id2 (A, N = 5), Id3 (B, N = 5) or FoxC2 (C, N = 6) compared to the contralateral sides. Bar: 50 μm. E, embryonic day.
Format: JPEG Size: 2.9MB Download file