DEV035014 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. Specificity and efficiency of Tie1-Cre-mediated Rac1 deletion. (A) Representative whole-mount X-Gal staining of E9.5 R26R⁺;Tie1-Cre⁻;Rac1^fl/+ and E9.5 and E10.5 R26R⁺;Tie1-Cre⁺;Rac1^fl/+ embryos. Tie1-driven Cre expression induces β-galactosidase activity (blue), which was observed in the majority of blood vessels of E9.5 R26R⁺;Tie1-Cre⁺;Rac1^fl/+ embryos and in virtually all vessels of E10.5 R26R⁺;Tie1-Cre⁺;Rac1^fl/+ embryos. Ao, aorta; En, endocardium; Isv, intersomitic vessels; Pn, developing perineural plexus. Scale bars: 250 µm. (B) Rac1 (green) and endomucin (red) immunostained sections of E10.5 Tie1-Cre⁻;Rac1^fl/fl and Tie1-Cre⁺;Rac1^fl/fl embryos show dramatic loss of Rac1 in Tie1-Cre⁺;Rac1^fl/fl blood vessels. Asterisks, Rac1-positive surrounding cells; arrow, Rac1-positive/endomucin-positive blood vessel; arrowheads, Rac1-negative/endomucin-positive blood vessel. (C) Representative PCR analysis of DNA isolated from aortae (Ao), yolk sacs (Ys) and tails from E12.5 Tie1-Cre⁻;Rac1^fl/fl and Tie1-Cre⁺;Rac1^fl/fl embryos. Considerably more Rac1-null allele was found in aortae and yolk sacs than in tails from Tie1-Cre⁺;Rac1^fl/fl embryos, whereas it was not detected in any tissue from controls. (D) Representative western blot analysis of extracts from endothelial cell-enriched embryonic cell populations from E12.5 Tie1-Cre⁻;Rac1^fl/fl and Tie1-Cre⁺;Rac1^fl/fl embryos immunoblotted for Rac1. Graph shows Rac1 quantitation relative to Hsc-70, which provided the loading control. (E) Chart showing the percentage survival of Tie1-Cre⁺;Rac1^fl/fl embryos at various gestational stages from progenies of Tie1-Cre⁻;Rac1^fl/fl and Tie1-Cre⁺;Rac1^fl/+ crosses. Tie1-Cre⁺;Rac1^fl/fl embryos die between E13.5 and E17.5. Number of embryos examined: 19 (E9.5), 37 (E10.5), 68 (E11.5), 145 (E12.5), 103 (E13.5), 77 (E15.5) and 21 (E17.5).

• Supplemental Figure S2 -

  Fig. S2. Blood vasculature of E12.5 Tie1-Cre⁺;Rac1^fl/fl embryos appears normal. (A) India ink visualisation of the carotid arteries shows no obvious difference in vascular structure or functionality between E12.5 Tie1-Cre⁺;Rac1^fl/fl and Tie1-Cre⁻;Rac1^fl/fl embryos. (B) Low-magnification micrographs of H&E-stained sections of back skin and limbs shows no obvious haemorrhage or edema in E12.5 Tie1-Cre⁺;Rac1^fl/fl embryos. Endomucin-stained sections of back skin show normal vasculature in both Tie1-Cre⁺;Rac1^fl/fl and Tie1-Cre⁻;Rac1^fl/fl embryos. (C) Aortae of E12.5 Tie1-Cre⁺;Rac1^fl/fl and Tie1-Cre⁻;Rac1^fl/fl embryos show no reproducible differences after immunostaining for laminin (green) and α-SMA (red). (D) Immunostaining for endothelial marker endomucin also revealed no detectable differences in the veins of E12.5 Tie1-Cre⁺;Rac1^fl/fl and Tie1-Cre⁻;Rac1^fl/fl embryos. Scale bars: B, 50 µm in skin H&E, 150 µm in limb H&E, 30 µm in endomucin; 40 µm in C; 50 µm in D.

• Supplemental Figure S3 -

  Fig. S3. Rac1 depletion in endothelial cells does not affect the expression and activity of other Rho-related GTPases. (A) RT-PCR analysis of mRNA expression of Rac isoforms in mouse lung endothelial cells mock, scrambled (Con) or Rac1 siRNA-transfected. RNA isolated from lung (lu), spleen (sp) and brain (br) mouse tissues were used as positive controls for Rac1, Rac2 and Rac3 mRNA expression, respectively. Actin mRNA provided the internal control. L, ladder. Graph represents Rac1 mRNA quantitation relative to actin (mean ± s.e.m), *P<0.003. (B) Western blot analysis of Rac1, Cdc42 and RhoA in mock, Con and Rac1 siRNA-transfected endothelial cells. Hsc-70 provided the loading control. Graph represents Rac1, Cdc42 and RhoA protein quantitation relative to Hsc-70 (mean ± s.e.m), *P<0.001. (C) Rac1 and Cdc42 active levels examined by GST-PAK pull-downs and RhoA active levels analyzed by G-LISA assays. Graphs represent quantitation of active levels of Rac1 and Cdc42 relative to total Rac1 and Cdc42, respectively; expression (mean ± s.e.m; *P<0.01) and quantitation of active levels of RhoA relative to a manufacturer’s positive control (mean ± s.e.m). Mock, endothelial cells transfected with transfection media; Mock+, mock cells stimulated with 10% FCS; positive control, constitutively active RhoA protein. n=3 independent experiments.

• Supplemental Figure S4 -

  Fig. S4. Lymphatic vessels appear congested and ruptured in E13.5 Tie1-Cre⁺;Rac1^fl/fl embryos. (A) E13.5 Tie1-Cre⁻;Rac1^fl/fl embryo section immunostained for Lyve1 shows lymphatic vessels in the skin that are devoid of blood cells. (B) E13.5 Tie1-Cre⁺;Rac1^fl/fl embryo section immunostained for Lyve1 reveals blood-congested lymphatics in the skin. (C) E13.5 Tie1-Cre⁺;Rac1^fl/fl embryo section immunostained for Lyve1 shows blood-congested and enlarged lymphatics in the thoracic region. (D) Evidence that in E13.5 Tie1-Cre⁺;Rac1^fl/fl embryos immunostained for Lyve1, lymphatics can rupture and blood haemorrhages into the surrounding mesenchyme. Serial section stained for endomucin shows that the haemorrhage from the lymphatic is not contained in blood vessels. Large black arrows, lymphatic vessels without red blood cells; large black arrowheads, red blood cell-congested lymphatic vessels; white arrowheads, opening of ruptured lymphatic vessel; small black arrows, red blood cells leaking into the mesenchyme. Scale bars: 100 µm in A,B; 50 µm in C,D.

• Supplemental Figure S5 -

  Fig. S5. Increased Lyve1/CD34 and Vegfr3/CD34 double-positive endothelial cells in Tie1-Cre⁺;Rac1^fl/fl embryos. (A) FACS profile of embryonic CD45⁻/Pecam1⁺ cells at E11.5 identifying the population selected for further analysis (pink). Bar chart shows that the mean number of CD45⁻/Pecam1⁺ cells isolated from E11.5 Tie1-Cre⁺;Rac1^fl/fl and Tie1-Cre⁻;Rac1^fl/fl embryos was the same. (B) FACS analysis of percentage of isolated lymphatic endothelial cells, LECs (Lyve1⁺/CD34⁻); blood endothelial cells, BECs (Lyve1⁻/CD34⁺); and double-positive endothelial cells, DP (Lyve1⁺/CD34⁺) from the CD45⁻/Pecam1⁺ population of cells isolated and sorted from E11.5 Tie1-Cre⁺;Rac1^fl/fl and Tie1-Cre⁻;Rac1^fl/fl embryos. Whereas the proportion of BECs and LECs was normal, the proportion of DP endothelial cells was increased significantly in E11.5 Tie1-Cre⁺;Rac1^fl/fl embryos. (C) Using a second marker, Vegf receptor 3 (Vegfr3), a similar analysis was performed to that in B. FACS analysis of percentage of isolated LECs (Vegfr3⁺/CD34⁻); BECs (Vegfr3⁻/CD34⁺); and double-positive endothelial cells, DP (Vegfr3⁺/CD34⁺) from the CD45⁻/Pecam1⁺ population of cells isolated and sorted from E11.5 Tie1-Cre⁺;Rac1^fl/fl and Tie1-Cre⁻;Rac1^fl/fl embryos. Once again, the proportion of BECs and LECs was normal and a Vegfr3⁺/CD34⁺ double-positive (DP) endothelial cell population was increased significantly in E11.5 Tie1-Cre⁺;Rac1^fl/fl embryos. Graphs in B and C represent mean levels of BEC, LEC or DP cells in Tie1-Cre⁺;Rac1^fl/fl embryos relative to those in Tie1-Cre⁻;Rac1^fl/fl embryos ± s.e.m. *P<0.03-0.05; n=3-4 experimental tests.

• Supplemental Figure S6 -

  Fig. S6. Proliferation of Prox1-positive cells emerging from the cardinal vein and overall numbers of Prox1-positive cells in CD45⁻/Pecam1⁺ populations of E11.5 Tie1-Cre⁺;Rac1^fl/fl embryos is normal. (A) Sections of E11.5 Tie1-Cre⁻;Rac1^fl/fl and Tie1-Cre⁺;Rac1^fl/fl embryos in the region where Prox1-positive cells emerge from the cardinal vein were double stained for Prox1 and Ki67 to identify proliferating cells. (B) Quantitation revealed that there was no significant change in the percentage of Prox1-positive cells that were Ki67-positive in this region of the embryo. (C) CD45⁻/Pecam1⁺ cells isolated and sorted from E11.5 Tie1-Cre⁻;Rac1^fl/fl and Tie1-Cre⁺;Rac1^fl/fl embryos and were cytospun onto glass slides and immunostained for Prox1. Bar chart shows percentage of CD45⁻/Pecam1⁺ cells that were positive for Prox1 and shows no significant difference between genotypes. Values are given as means ± s.e.m. Arrows, Ki67⁺/Prox1⁺ cells; arrowheads, Ki67⁺/Prox1⁻ cells. n=4-9 embryos per genotype.

• Supplemental Figure S7 -

  Fig. S7. Knockdown of Rac1 in endothelial cells reduces Vegf-A164-induced migration. Endothelial cells were transfected with either Rac1-specific siRNA, scrambled control siRNA (con) or with transfection media (mock). (A) Western blot analysis shows a significant degree of Rac1 depletion in Rac1-siRNA treated cells. (B) Cells were grown to confluency and scratched with a P200 tip and either treated with (+) or without (−) Vegf-A164. Scratch wounds were measured at time 0 and 8 hours after wounding. Migration was measured as the percentage of scratch closure/migration relative to 0 hours and showed that Rac1-knockdown inhibits Vegf-A164-mediated migration. Values are given as means ± s.e.m. *P<0.05, n=3 experiments.

• Supplemental Figure S8 -

  Fig. S8. Vegfr3 heterozygous endothelial cells show normal Rac1 activity. Vegf-C156S-stimulated lung endothelial cells isolated from 6- to 8-week-old wild-type (wt) or Vegfr3 heterozygous (Vegfr3^+/−) mice were tested for Vegfr3 and active Rac1 levels. Results show that the Vegfr3 heterozygous cells have dramatically reduced levels of Vegfr3. Hsc-70 acted as a loading control. However, no apparent change in levels of active Rac1 was observed between genotypes. Total Rac1 levels were normal between genotypes. Graphs represent mean relative levels of Vegfr3, active Rac1 (relative to total Rac1), and total Rac1 (relative to Hsc70) ± s.e.m. *P<0.05, n=3 experiments.

• Supplemental Figure S9 -

  Fig. S9. Schematic representation of the role of endothelial Rac1 in lymphatic and blood vessel formation. Up to E11.5, Rac1 deletion in Tie1-Cre endothelial cells does not affect the formation of blood vessels. However, by E11.5, during budding, a specialized subpopulation of cardinal vein endothelial cells (Prox1⁺) that are programmed to migrate away from the cardinal vein and form the lymphatic vasculature (i.e. lymph sacs), is reduced in endothelial Rac1-deficient embryos (Tie1-Cre⁺;Rac1^fl/fl or mutant). Moreover, the number and migratory distribution of Prox1⁺ endothelial cells during this process is compromised by Rac1 deletion. Between E12.5 and E13.5, normally, the newly formed lymph sacs separate and segregate from the parental veins. However, in Tie1-Cre⁺;Rac1^fl/fl embryos this process fails, resulting in the presence of red blood cells (Rbc) in lymph sacs. The abnormal lymphatics are likely to be malfunctioning, as demonstrated by the edema in the mutant embryos. We hypothesize that the presence of excessive blood in the lymphatics, due to their poor segregation from the blood vasculature, may cause lymphatic vessel rupture, which would result in interstitial haemorrhage, capillary collapse, thoracic aorta collapse and finally death of endothelial Rac1-deficient embryos between E13.5 and E17.5.