Developmental Biology

Combined forces of hydrostatic pressure and actin polymerization drive endothelial tip cell migration and sprouting angiogenesis

Igor Kondrychyn
Haymar Wint
Liqun He
Christer Betsholtz
Li-Kun Phng author has email address

Laboratory for Vascular Morphogenesis, RIKEN Center for Biosystems Dynamics Research, Kobe 650-0047, Japan
Department of Immunology, Genetics and Pathology, Rudbeck Laboratory, Uppsala University, 75185 Uppsala, Sweden
Department of Medicine Huddinge, Karolinska Institutet, 14157 Huddinge, Sweden

https://doi.org/10.7554/eLife.98612.1

Open access
Copyright information

Figures and data

Differential expression of aqp1a.1 and aqp8a.1 mRNAs in tip and stalk cells.
(A-F) Detection of aqp1a.1 and aqp8a.1 mRNA by RNAscope in situ hybridization in 20 (A - D) and 22 (E and F) hpf zebrafish embryos. Representative maximum intensity projection of confocal z-stacks of tip and stalk cells sprouting from DA are shown (n = 9 embryos, 2 independent experiments). B and F show magnified region of a tip cell from a 20 and 22 hpf zebrafish, respectively (tip cell and EC nuclei in the DA are outlined). Scale bar, 10 µm (B and F) and 40 µm (A, C - E). Percentage of vascular sprouts with differential expression of aqp1a.1 and aqp8a.1 mRNAs in tip and stalk cells in 22 hpf zebrafish (n=23 sprouts from 8 embryos, 2 independent experiments). (G). qPCR analysis of aqp1a.1 and aqp8a.1 expression in zebrafish embryos treated with 0.01% DMSO or 1 µM ki8751 at 20 hpf for 6 or 24 hours (n = 3 independent experiments). (G) qPCR analysis of AQP1 gene expression in HAECs treated with 0.01% DMSO or 1 nM and 10 nM ki8751 for 6 hr (n = 2 independent experiments). (J) qPCR analysis of aqp1a.1 and aqp8a.1 gene expression in the zebrafish embryos treated with 0.1% DMSO or 10 µM DBZ at 20 hpf for 6 or 24 hr (n = 2 independent experiments). In panels H-J, gene expressions are shown relative to gapdh expression and data are presented as mean ± SD; statistical significance was determined by Brown-Forsythe and Welch ANOVA tests with Dunnett’s multiple comparisons test; ns p>0.05, **p<0.01, ***p<0.001, and ****p<0.0001. DA, dorsal aorta; PCV, posterior cardinal vein. See also Fig. S3.

Aquaporin proteins are enriched at the leading front of migrating cells.
(A and B) Representative maximum intensity projection confocal z-stacks of endothelial tip cells of 25 hpf Tg(fli1ep:aqp1a.1-mEmerald)^rk30;(kdrl:ras-mCherry)^s916(A) and Tg(fli1ep:aqp8a.1-mEmerald)^rk31;(kdrl:ras-mCherry)^s916(B) embryos. Scale bar, 20 µm. (C and D) Still images from time-lapse movies of migrating tip cells from Tg(fli1ep:aqp1a.1-mEmerald)^rk30 (C) and Tg(fli1ep:aqp8a.1-mEmerald)^rk31 (D) embryos. Movies were taken from 24 to 30 hpf. Black arrowheads, Aquaporin protein localization at the leading edge of migrating tip cells. Arrows, Aquaporin localization in filopodia. Time, h:min. Scale bar, 5 µm. See also Movie S1 and S2.

Loss of Aquaporin function leads to defective trunk vessel formation.
(A-D) Representative maximum intensity projection confocal z-stacks of 28 hpf aqp1a.1^+/rk28;aqp8a.1^+/rk29 (A), aqp1a.1^rk28/rk28(B), aqp8a.1^rk29/rk29 (C) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29(D) embryos. (E) Quantification of ISV length in 28 hpf embryos (aqp1a.1^+/rk28;aqp8a.1^+/rk29: n = 21 embryos; aqp1a.1^rk28/rk28: n = 20 embryos; aqp8a.1^rk29/rk29: n = 20 embryos; aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29: n = 19 embryos from 2 independent experiments). (F) Illustration of aISV and vISV connections in the trunk of wild type embryo at 3 dpf. (G-J) Representative maximum intensity projection confocal z-stacks of 3 dpf wild type (G), aqp8a.1^rk29/rk29 (H), aqp1a.1^rk28/rk28(I), and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29 (J) embryos. Asterisk marks incomplete ISVs. (K) Quantification of percentage of wild type and aquaporin mutant embryos at 3 dpf with at least one incomplete ISVs per embryo (wild type, n = 47 embryos; aqp1a.1^rk28/rk28, n = 30 embryos; aqp8a.1^rk29/rk29, n = 38 embryos; aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29, n = 204 embryos). (L) A pie chart showing proportion of different classes of incomplete ISV phenotypes (I-VI) found in aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29 mutant embryos at 3 dpf. DA, dorsal aorta; DLAV, dorsal longitudinal anastomotic vessel; ISV, intersegmental vessel; aISV, arterial ISV; vISV, venous ISV; NC, notochord; NT, neural tube; PCV, posterior cardinal vein. Scale bar, 50 µm (G-J) and 200 µm (A-D). See also Fig. S6, S7 and S8.

Aquaporins promote endothelial tip cell protrusion and migration.
(A) Timing of tip cell emergence from the DA in aqp1a.1^+/rk28;aqp8a.1^+/rk29 (n = 23) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29 (n = 27). (B and C) Still images from time-lapse imaging of migrating tip cells from aqp1a.1^+/rk28;aqp8a.1^+/rk29 (B) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29 (C) embryos from 20 to 30 hpf. White arrows, retracting tip cell. *, secondary sprouting from PCV. Scale bar, 10 µm. (D and E) Stills images from representative time-lapse movies of migrating tip cells in aqp1a.1^+/rk28;aqp8a.1^+/rk29 (D, n = 2) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29 (E, n = 5) embryos. Movies were taken from 25 to 30 hpf (n = 2 independent experiments). Bracket, formation of stable protrusions. Scale bar, 20 µm. (F and G) Representative maximum intensity projection confocal z-stacks of tip cells from wild type (F) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29(G) embryos at 26 hpf. Scale bar, 10 µm. (H) Quantification of filopodia number in tip cells of wild type (n = 36 cells from 10 embryos, 2 independent experiments), aqp1a.1^+/rk28;aqp8a.1^+/rk29 (n = 19 cells from 6 embryos, 2 independent experiments) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29 (n = 28 cells from 9 embryos, 2 independent experiments) embryos. (I) Quantification of filopodia length in tip cells of wild type (n = 24 cells from 7 embryos, 2 independent experiments), aqp1a.1^+/rk28;aqp8a.1^+/rk29 (n = 16 cells from 7 embryos, 2 independent experiments) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29 (n = 11 cells from 6 embryos, 2 independent experiments) embryos. (J) Quantification of tip cell leading edge displacement of aqp1a.1^+/rk28;aqp8a.1^+/rk29 (n = 19 cells from 5 embryos, 3 independent experiments) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29 (n = 47 cells from 13 embryos, 6 independent experiments) embryos at 25 - 30 hpf. (K) Quantification of tip cell nuclei displacement of aqp1a.1^+/rk28;aqp8a.1^+/rk29 (n = 20 cells from 6 embryos, 4 independent experiments) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29(n = 20 cells from 4 embryos, 2 independent experiments) embryos at 21 - 30 hpf. (L) Quantification of tip cell migration velocity in aqp1a.1^+/rk28;aqp8a.1^+/rk29 (n = 19 cells from 5 embryos, 3 independent experiments) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29 (n = 47 cells from 13 embryos, 6 independent experiments) embryos at 25 - 30 hpf. Statistical significance was determined by Brown-Forsythe and Welch ANOVA tests with Dunnett’s (H) or Sidak’s (I) multiple comparisons test, and with unpaired t-test (L). ns, p>0.05, *, p < 0.05, ***, p < 0.001, and ****, p < 0.0001. See also Movie S3-6.

Aquaporin-mediated water influx increases tip cell volume and actin elongation in filopodia.
(A and B) Representative maximum intensity projection confocal z-stacks of tip cell and its 3D-volume rendering of 25 hpf wild type (A) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29(B) labelled with cytosolic mEmerald. Scale bar, 10 µm. (C) Quantification of tip cell volume in 25 hpf wild type (n = 24 cells) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29 (n = 20 cells) embryos from 3 independent experiments. (D) Quantification of growth rate of actin bundles in tip cell filopodia in 25 hpf wild type (n = 12 cells from 7 embryos, 2 independent experiments) and aqp1a.1^rk28;aqp8a.1^rk29(n = 12 cells from 6 embryos, 2 independent experiments) embryos. Statistical significance was determined by unpaired t-test with (C) or without (D) Welch’s correction.

Additive function of actin polymerization and hydrostatic pressure in driving EC migration and sprouting angiogenesis.
(A-D) Representative maximum intensity projection confocal z-stacks of 28 hpf wild type (A and B) and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29 (C and D) embryos treated with 0.1% DMSO (A and C) or 0.8 µg/ml Lat. B (B and D) from 20 to 28 hpf (for each condition: wild type, n = 10 embryos; aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29, n =12 embryos, 2 independent experiments). Scale bar, 20 µm. (E) Quantification of ISV length in 28 hpf wild type and aqp1a.1^rk28/rk28;aqp8a.1^rk29/rk29 embryos treated with 0.1% DMSO or 0.8 µg/ml Lat. B (wild type: n = 60 vessels from 10 control embryos, n = 66 vessels from 10 Lat. B-treated embryos; aqp1a.1^rk28;aqp8a.1^rk29: n = 126 vessels from 12 control embryos, n = 112 vessels from 11 Lat. B-treated embryos, 2 independent experiments). Statistical significance was determined by Brown-Forsythe ANOVA test with Sidak’s multiple comparison test; ns, p,>,0.05, ^**,p < 0.01, and ^****, p < 0.0001. (F) Endothelial tip cells generate membrane protrusions by actin polymerization and increased hydrostatic pressure via Aquaporin-mediated water inflow. In the absence of actin polymerization and presence of Aquaporins, hydrostatic pressure deforms membranes to generate membrane protrusions. When Aquaporin function is lost, membrane protrusions is generated by actin polymerization. Fewer membrane protrusions are formed when only one mechanism is utilized, resulting in slower EC migration. See also Movie S7.

Sign up for email alerts