Abstract

Glutamine synthetase, encoded by the gene GLUL, is an enzyme that converts glutamate and ammonia to glutamine. It is expressed by endothelial cells, but surprisingly shows negligible glutamine-synthesizing activity in these cells at physiological glutamine levels. Here we show in mice that genetic deletion of Glul in endothelial cells impairs vessel sprouting during vascular development, whereas pharmacological blockade of glutamine synthetase suppresses angiogenesis in ocular and inflammatory skin disease while only minimally affecting healthy adult quiescent endothelial cells. This relies on the inhibition of endothelial cell migration but not proliferation. Mechanistically we show that in human umbilical vein endothelial cells GLUL knockdown reduces membrane localization and activation of the GTPase RHOJ while activating other Rho GTPases and Rho kinase, thereby inducing actin stress fibres and impeding endothelial cell motility. Inhibition of Rho kinase rescues the defect in endothelial cell migration that is induced by GLUL knockdown. Notably, glutamine synthetase palmitoylates itself and interacts with RHOJ to sustain RHOJ palmitoylation, membrane localization and activation. These findings reveal that, in addition to the known formation of glutamine, the enzyme glutamine synthetase shows unknown activity in endothelial cell migration during pathological angiogenesis through RHOJ palmitoylation.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Data availability

Figures 1, 4, 5 and Extended Data Figs. 1, 7, and 8 have associated raw data (uncropped blots and/or gel pictures) in Supplementary Fig. 1. Figures 1, 2 and Extended Data Figs. 1, 4 have associated raw data (Excel files) for all bar graphs representing data from experiments involving mouse models. For the molecular modelling of palmitoyl-CoA docking into GS, models and trajectories are available on Figshare (doi: 10.6084/m9.figshare.6575438). Any additional information required to interpret, replicate or build upon the Methods or findings reported in the manuscript is available from the corresponding author upon request.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. 1.

    De Bock, K. et al. Role of PFKFB3-driven glycolysis in vessel sprouting. Cell 154, 651–663 (2013).

  2. 2.

    Schoors, S. et al. Fatty acid carbon is essential for dNTP synthesis in endothelial cells. Nature 520, 192–197 (2015).

  3. 3.

    Schoors, S. et al. Partial and transient reduction of glycolysis by PFKFB3 blockade reduces pathological angiogenesis. Cell Metab. 19, 37–48 (2014).

  4. 4.

    Huang, H. et al. Role of glutamine and interlinked asparagine metabolism in vessel formation. EMBO J. 36, 2334–2352 (2017).

  5. 5.

    Abcouwer, S. F., Lukascewicz, G. C., Ryan, U. S. & Souba, W. W. Molecular regulation of lung endothelial glutamine synthetase expression. Surgery 118, 325–335 (1995).

  6. 6.

    He, Y., Hakvoort, T. B., Vermeulen, J. L., Lamers, W. H. & Van Roon, M. A. Glutamine synthetase is essential in early mouse embryogenesis. Dev. Dyn. 236, 1865–1875 (2007).

  7. 7.

    Häberle, J. et al. Congenital glutamine deficiency with glutamine synthetase mutations. N. Engl. J. Med. 353, 1926–1933 (2005).

  8. 8.

    Kung, H. N., Marks, J. R. & Chi, J. T. Glutamine synthetase is a genetic determinant of cell type-specific glutamine independence in breast epithelia. PLoS Genet. 7, e1002229 (2011).

  9. 9.

    Schoors, S. et al. Incomplete and transitory decrease of glycolysis: a new paradigm for anti-angiogenic therapy? Cell Cycle 13, 16–22 (2014).

  10. 10.

    Dadsetan, S. et al. Brain alanine formation as an ammonia-scavenging pathway during hyperammonemia: effects of glutamine synthetase inhibition in rats and astrocyte-neuron co-cultures. J. Cereb. Blood Flow Metab. 33, 1235–1241 (2013).

  11. 11.

    Ridley, A. J. Rho GTPase signalling in cell migration. Curr. Opin. Cell Biol. 36, 103–112 (2015).

  12. 12.

    Yuan, L. et al. RHOJ is an endothelial cell-restricted Rho GTPase that mediates vascular morphogenesis and is regulated by the transcription factor ERG. Blood 118, 1145–1153 (2011).

  13. 13.

    Kim, C. et al. Vascular RHOJ is an effective and selective target for tumor angiogenesis and vascular disruption. Cancer Cell 25, 102–117 (2014).

  14. 14.

    Leszczynska, K., Kaur, S., Wilson, E., Bicknell, R. & Heath, V. L. The role of RHOJ in endothelial cell biology and angiogenesis. Biochem. Soc. Trans. 39, 1606–1611 (2011).

  15. 15.

    de Toledo, M. et al. The GTP/GDP cycling of Rho GTPase TCL is an essential regulator of the early endocytic pathway. Mol. Biol. Cell 14, 4846–4856 (2003).

  16. 16.

    Nethe, M. & Hordijk, P. L. The role of ubiquitylation and degradation in RhoGTPase signalling. J. Cell Sci. 123, 4011–4018 (2010).

  17. 17.

    Kaur, S. et al. RHOJ/TCL regulates endothelial motility and tube formation and modulates actomyosin contractility and focal adhesion numbers. Arterioscler. Thromb. Vasc. Biol. 31, 657–664 (2011).

  18. 18.

    Heasman, S. J. & Ridley, A. J. Multiple roles for RHOA during T cell transendothelial migration. Small GTPases 1, 174–179 (2010).

  19. 19.

    Ackermann, K. L., Florke, R. R., Reyes, S. S., Tader, B. R. & Hamann, M. J. TCL/RHOJ plasma membrane localization and nucleotide exchange is coordinately regulated by amino acids within the N terminus and a distal loop region. J. Biol. Chem. 291, 23604–23617 (2016).

  20. 20.

    Wan, J. et al. Tracking brain palmitoylation change: predominance of glial change in a mouse model of Huntington’s disease. Chem. Biol. 20, 1421–1434 (2013).

  21. 21.

    Levine, R. L., Oliver, C. N., Fulks, R. M. & Stadtman, E. R. Turnover of bacterial glutamine synthetase: oxidative inactivation precedes proteolysis. Proc. Natl Acad. Sci. USA 78, 2120–2124 (1981).

  22. 22.

    Blanc, M. et al. SwissPalm: Protein Palmitoylation database. F1000Res. 4, 261 (2015).

  23. 23.

    Roberts, P. J. et al. Rho family GTPase modification and dependence on CAAX motif-signaled posttranslational modification. J. Biol. Chem. 283, 25150–25163 (2008).

  24. 24.

    Wei, X., Song, H. & Semenkovich, C. F. Insulin-regulated protein palmitoylation impacts endothelial cell function. Arterioscler. Thromb. Vasc. Biol. 34, 346–354 (2014).

  25. 25.

    Zheng, B., Zhu, S. & Wu, X. Clickable analogue of cerulenin as chemical probe to explore protein palmitoylation. ACS Chem. Biol. 10, 115–121 (2015).

  26. 26.

    Martin-Ramirez, J., Hofman, M., van den Biggelaar, M., Hebbel, R. P. & Voorberg, J. Establishment of outgrowth endothelial cells from peripheral blood. Nat. Protoc. 7, 1709–1715 (2012).

  27. 27.

    Van Den Bosch, L., Vandenberghe, W., Klaassen, H., Van Houtte, E. & Robberecht, W. Ca2+-permeable AMPA receptors and selective vulnerability of motor neurons. J. Neurol. Sci. 180, 29–34 (2000).

  28. 28.

    McKinney, S. A., Murphy, C. S., Hazelwood, K. L., Davidson, M. W. & Looger, L. L. A bright and photostable photoconvertible fluorescent protein. Nat. Methods 6, 131–133 (2009).

  29. 29.

    Mertens, N. M. A. C., Remaut, E. R. & Fiers, W. C. Regulatory system for inducible expression of genes with lambdoid promoters. Belgian patent WO 98/48025 (1998).

  30. 30.

    Mertens, N. M. A. C. & Kelly, A. G. Use of caspase enzymes for maturation of engineered recombinant polypeptide fusions. Belgian patent WO 04/074488 (2004).

  31. 31.

    Hamel, L. D., Deschenes, R. J. & Mitchell, D. A. A fluorescence-based assay to monitor autopalmitoylation of zDHHC proteins applicable to high-throughput screening. Anal. Biochem. 460, 1–8 (2014).

  32. 32.

    Kümmel, D., Heinemann, U. & Veit, M. Unique self-palmitoylation activity of the transport protein particle component Bet3: a mechanism required for protein stability. Proc. Natl Acad. Sci. USA 103, 12701–12706 (2006).

  33. 33.

    Fraccaroli, A. et al. Endothelial alpha-parvin controls integrity of developing vasculature and is required for maintenance of cell–cell junctions. Circ. Res. 117, 29–40 (2015).

  34. 34.

    Ho, C. Y., Jaalouk, D. E., Vartiainen, M. K. & Lammerding, J. Lamin A/C and emerin regulate MKL1–SRF activity by modulating actin dynamics. Nature 497, 507–511 (2013).

  35. 35.

    Timmerman, I. et al. A local VE-cadherin and Trio-based signaling complex stabilizes endothelial junctions through Rac1. J. Cell Sci. 128, 3041–3054 (2015).

  36. 36.

    van Buul, J. D. et al. RHOG regulates endothelial apical cup assembly downstream from ICAM1 engagement and is involved in leukocyte trans-endothelial migration. J. Cell Biol. 178, 1279–1293 (2007).

  37. 37.

    Heemskerk, N. et al. F-actin-rich contractile endothelial pores prevent vascular leakage during leukocyte diapedesis through local RHOA signalling. Nat. Commun. 7, 10493 (2016).

  38. 38.

    Dedecker, P., Duwé, S., Neely, R. K. & Zhang, J. Localizer: fast, accurate, open-source, and modular software package for superresolution microscopy. J. Biomed. Opt. 17, 126008 (2012).

  39. 39.

    Persson, F., Lindén, M., Unoson, C. & Elf, J. Extracting intracellular diffusive states and transition rates from single-molecule tracking data. Nat. Methods 10, 265–269 (2013).

  40. 40.

    He, Y. et al. Glutamine synthetase deficiency in murine astrocytes results in neonatal death. Glia 58, 741–754 (2010).

  41. 41.

    Benedito, R. et al. The notch ligands Dll4 and Jagged1 have opposing effects on angiogenesis. Cell 137, 1124–1135 (2009).

  42. 42.

    Claxton, S. et al. Efficient, inducible Cre-recombinase activation in vascular endothelium. Genesis 46, 74–80 (2008).

  43. 43.

    Pitulescu, M. E., Schmidt, I., Benedito, R. & Adams, R. H. Inducible gene targeting in the neonatal vasculature and analysis of retinal angiogenesis in mice. Nat. Protoc. 5, 1518–1534 (2010).

  44. 44.

    Kenyon, B. M. et al. A model of angiogenesis in the mouse cornea. Invest. Ophthalmol. Vis. Sci. 37, 1625–1632 (1996).

  45. 45.

    Abraham, M. J. et al. GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1–2, 19–25 (2015).

  46. 46.

    Maier, J. A. et al. ff14SB: improving the accuracy of protein side chain and backbone parameters from ff99SB. J. Chem. Theory Comput. 11, 3696–3713 (2015).

  47. 47.

    Frisch, M. J. et al. Gaussian 09 Rev. C.01 (Gaussian, 2009).

  48. 48.

    Bussi, G., Donadio, D. & Parrinello, M. Canonical sampling through velocity rescaling. J. Chem. Phys. 126, 014101 (2007).

Download references

Acknowledgements

We acknowledge R. Levine for supplying purified bacterial GS; L. Van Den Bosch and W. Scheveneels for providing primary mouse astrocytes; S. Trenson, I. Crèvecoeur, S. Noppen, L. Van Berckelaer and R. Van Berwaer for technical assistance; S.-M. Fendt, D. Verdegem and C. Ulens for discussions and suggestions; W. Vermaelen, A. Acosta Sanchez, A. Brajic, A. Sobrino and M. Cockx for experimental assistance; L.-C. Conradi and A. Pircher for supplying materials; and E. Wauters, A. Wolthuis and J. Jaekers for providing tissues for EC isolations. HecBioSim and PRACE are acknowledged for allocation of computer time. J.G., A.R.C., C.D., C.L., J.K., F.M.-R., S.R. and S.Va. are supported by the FWO; A.Z. by LE&RN/FDRS; B.C. by IWT; U.B. by a Marie Curie-IEF Fellowship; H.H. by an EMBO Long-Term Fellowship; J.D.v.B. by a LSBR fellowship; R.Cu. by a British Heart Foundation Intermediate Clinical Fellowship; and X.W. by the American Cancer Society RSG, NIH/NCI and NIH/NIDDK. F.C., G.S. and F.L.G. are supported by the EPSRC; X.L. by the State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center at Sun Yat-Sen University, and the National Natural Science Foundation of China (81330021 and 81670855). P.C. is supported by a Federal Government Belgium grant, long-term structural Methusalem funding by the Flemish Government, a Concerted Research Activities Belgium grant, grants from the FWO, Foundation against Cancer and ERC Advanced Research Grant. G.E. and M.Dew. received a Foundation against Cancer grant and G.E. received a FWO ‘Krediet aan navorsers’.

Author information

Author notes

    • Anna Rita Cantelmo

    Present address: Université de Lille, INSERM U1003, Physiologie Cellulaire, Lille, France

    • Ulrike Brüning

    Present address: Max-Delbrück-Center for Molecular Medicine, Berlin, Germany

    • Hongling Huang

    Present address: Immunology Department, St. Jude Children’s Research Hospital, Memphis, TN, USA

    • Christian Lange

    Present address: DFG-Research Center for Regenerative Therapies, Technical University Dresden, Dresden, Germany

    • Leanne Ramer

    Present address: Department of Biomedical Physiology and Kinesiology, Simon Fraser University, Burnaby, British Columbia, Canada

    • Richard Cubbon

    Present address: Division of Cardiovascular and Diabetes Research, Multidisciplinary Cardiovascular Research Centre, University of Leeds, Leeds, UK

  1. These authors contributed equally: Guy Eelen, Charlotte Dubois

Affiliations

  1. Center for Cancer Biology, University of Leuven, Leuven, Belgium

    • Guy Eelen
    • , Charlotte Dubois
    • , Anna Rita Cantelmo
    • , Jermaine Goveia
    • , Ulrike Brüning
    • , Annalisa Zecchin
    • , Hongling Huang
    • , Saar Vandekeere
    • , Joanna Kalucka
    • , Christian Lange
    • , Francisco Morales-Rodriguez
    • , Bert Cruys
    • , Lucas Treps
    • , Leanne Ramer
    • , Stefan Vinckier
    • , Katleen Brepoels
    • , Sabine Wyns
    • , Joris Souffreau
    • , Luc Schoonjans
    • , Richard Cubbon
    • , Mieke Dewerchin
    •  & Peter Carmeliet
  2. State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-Sen University, Guangzhou, China

    • Guy Eelen
    • , Rongyuan Chen
    • , Xuri Li
    •  & Peter Carmeliet
  3. Center for Cancer Biology, VIB, Leuven, Belgium

    • Guy Eelen
    • , Charlotte Dubois
    • , Anna Rita Cantelmo
    • , Jermaine Goveia
    • , Ulrike Brüning
    • , Annalisa Zecchin
    • , Hongling Huang
    • , Saar Vandekeere
    • , Joanna Kalucka
    • , Christian Lange
    • , Francisco Morales-Rodriguez
    • , Bert Cruys
    • , Lucas Treps
    • , Leanne Ramer
    • , Stefan Vinckier
    • , Katleen Brepoels
    • , Sabine Wyns
    • , Joris Souffreau
    • , Luc Schoonjans
    • , Richard Cubbon
    • , Mieke Dewerchin
    •  & Peter Carmeliet
  4. Cutaneous Biology Research Center, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, USA

    • Michael DeRan
    • , Gopala Jarugumilli
    •  & Xu Wu
  5. Department of Molecular Cell Biology, Sanquin Research and Landsteiner Laboratory, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands

    • Jos van Rijssel
    •  & Jaap D. van Buul
  6. Department of Chemistry, University College London, London, UK

    • Giorgio Saladino
    • , Federico Comitani
    •  & Francesco L. Gervasio
  7. Molecular Imaging and Photonics, University of Leuven, Leuven, Belgium

    • Susana Rocha
    •  & Johan Hofkens
  8. Tytgat Institute for Liver and Gastrointestinal Research, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands

    • Wouter H. Lamers
  9. Center for Cell Analyses and Modelling, University of Connecticut Health Centre, Farmington, CT, USA

    • Yi Wu
  10. Inflammation Research Center, VIB, Ghent, Belgium

    • Jurgen Haustraete
  11. Department of Biomedical Molecular Biology, Ghent University, Ghent, Belgium

    • Jurgen Haustraete
  12. Department of Microbiology and Immunology, University of Leuven, Leuven, Belgium

    • Sandra Liekens
  13. Metabolomics Core Facility, Center for Cancer Biology, VIB, Leuven, Belgium

    • Bart Ghesquière
  14. Institute of Structural Molecular Biology, University College London, London, UK

    • Francesco L. Gervasio

Authors

  1. Search for Guy Eelen in:

  2. Search for Charlotte Dubois in:

  3. Search for Anna Rita Cantelmo in:

  4. Search for Jermaine Goveia in:

  5. Search for Ulrike Brüning in:

  6. Search for Michael DeRan in:

  7. Search for Gopala Jarugumilli in:

  8. Search for Jos van Rijssel in:

  9. Search for Giorgio Saladino in:

  10. Search for Federico Comitani in:

  11. Search for Annalisa Zecchin in:

  12. Search for Susana Rocha in:

  13. Search for Rongyuan Chen in:

  14. Search for Hongling Huang in:

  15. Search for Saar Vandekeere in:

  16. Search for Joanna Kalucka in:

  17. Search for Christian Lange in:

  18. Search for Francisco Morales-Rodriguez in:

  19. Search for Bert Cruys in:

  20. Search for Lucas Treps in:

  21. Search for Leanne Ramer in:

  22. Search for Stefan Vinckier in:

  23. Search for Katleen Brepoels in:

  24. Search for Sabine Wyns in:

  25. Search for Joris Souffreau in:

  26. Search for Luc Schoonjans in:

  27. Search for Wouter H. Lamers in:

  28. Search for Yi Wu in:

  29. Search for Jurgen Haustraete in:

  30. Search for Johan Hofkens in:

  31. Search for Sandra Liekens in:

  32. Search for Richard Cubbon in:

  33. Search for Bart Ghesquière in:

  34. Search for Mieke Dewerchin in:

  35. Search for Francesco L. Gervasio in:

  36. Search for Xuri Li in:

  37. Search for Jaap D. van Buul in:

  38. Search for Xu Wu in:

  39. Search for Peter Carmeliet in:

Contributions

P.C. conceived the concept of the study and provided supervision. G.E., P.C., X.L., M.Dew. and L.S. contributed to the execution, support and analysis of experiments, and/or provided advice. G.E., C.D., A.R.C., J.G. and P.C. were involved in the experimental design. G.E., C.D., A.R.C., J.G., U.B., A.Z., H.H., S.Va., J.K., C.L., F.M.-R., B.C., L.R., S.Vi., K.B., S.W., J.S., L.S., S.L., R.Ch., R.Cu. and M.Dew. carried out molecular biology and in vivo experiments. B.G. performed mass spectrometry. J.v.R. and J.D.v.B. carried out RHO activity assays. M.DeR., G.J. and X.W. undertook palmitoylation experiments. G.S., F.C. and F.L.G. performed molecular dynamics simulations. S.R. and J.Ho. performed bimolecular fluorescence complementation and SPT experiments. G.E., C.D., A.R.C., J.G., R.Cu., U.B., C.L., S.R., L.T., B.C., M.Dew., J.Ho., S.L., B.G., F.L.G., J.D.v.B., X.W. and P.C. interpreted the data. W.H.L., Y.W. and J.Ha. provided necessary materials. G.E. and P.C. drafted the manuscript. All authors agreed on the final version of the manuscript.

Competing interests

The authors declare no competing interests.

Corresponding authors

Correspondence to Guy Eelen or Xuri Li or Peter Carmeliet.

Extended data figures and tables

  1. Extended Data Fig. 1 GLUL knockout impairs vessel sprouting.

    a, GLUL mRNA levels in HUVECs (n = 9 donors), lung ECs (n = 5), colon ECs (n = 4), liver ECs (n = 3), human umbilical artery ECs (HUAECs) (n = 2) and human blood outgrowth ECs (BOECs) (n = 2); (mean ± s.e.m.; *P< 0.05 versus HUVEC, Student’s t-test) and in HEPG2 cells (mean ± s.e.m.; n = 3; *P< 0.05 versus HUVEC, Student’s t-test). b, c, Western blot of GS protein levels in HUVECs and HEPG2 cells in medium containing 0.6 mM glutamine (+) or 0.025 mM glutamine (−) (b), and in isolated mouse liver ECs (mLiECs) and mouse astrocytes (c) (representative immunoblots of two independent experiments are shown). d, e, Genomic organization of the loxP-flanked Glul allele before and after Cre-mediated excision (d) and correct recombination of the lox allele (L) in GlulvECKO and GlulpECKO mice upon tamoxifen (Tam) treatment, as assessed by genomic DNA PCR (e; the PCR to amplify the loxP-flanked Glul allele (lox) or to amplify the cre-recombined allele (∆) were run in separate reactions but loaded in the same lane; the gel picture shown is representative of all control, GlulvECKO and GlulpECKO mice used in this study). f, Quantification of branchpoints at the rear of the plexus in GlulvECKO mice (mean ± s.e.m.; n = 10 mice for GlulvECKO and 11 for WT controls from 3 litters; *P< 0.05 versus WT littermates, mixed-models R statistics). g, Pericyte coverage of retinal microvessels in WT and GlulvECKO littermates determined by NG2 staining and shown as the NG2+ area as a percentage of the vessel area (mean ± s.e.m.; n = 4 mice for WT and 3 for GlulvECKO from 1 litter; NS, P > 0.05 versus WT, Student’s t-test). h, Reduced complexity of the retinal vascular front in P5 GlulvECKO compared with WT mice, determined by the number of branches on distal sprouts (mean ± s.e.m.; n = 13 mice for WT and 21 for GlulvECKO from 5 litters; *P< 0.05 versus WT, Student’s t-test). i, Quantification of EdU+ ECs at the rear of the plexus (mean ± s.e.m.; n = 12 mice for WT and 22 for GlulvECKO from 4 litters; NS, P > 0.05 versus WT littermates, Student’s t-test). jm, IB4 staining of P5 retinal vascular plexuses from WT (j) and GlulpECKO (k) mice (representative pictures with magnification shown in the inset; A, artery; V, vein) and quantification of branch points at the front (l) and the rear (m) of the plexus (mean ± s.e.m.; n = 10 mice for WT and 18 for GlulpECKO from 4 litters; *P ≤ 0.05 versus WT littermates, Student’s t-test). nu, IB4 staining of the retinal microvasculature of three-week-old (P21) (n, o) and six-week-old (P42) (r, s) WT and GlulvECKO littermates (A, artery; V, vein). The lower left insets display a higher magnification of the IB4-stained superficial plexus, whereas the lower right insets display a higher magnification of the deep plexus. The corresponding quantification of the vascular area (p, t) and the branch point density (q, u) in the superficial and the deep layer is also shown (mean ± s.e.m.; n = 8 mice for WT and 8 for GlulvECKO at P21, from two litters; n = 10 mice for WT and 14 for GlulvECKO at P42, from four litters; NS, P > 0.05 versus WT, Student’s t-test). vag, Representative micrographs of heart (v, z), liver (w, aa) and kidney (x, ab) sections from WT and GlulvECKO littermates immunostained for the EC marker endoglin and of lung (y, ac) sections immunostained for the EC marker CD34 and corresponding quantifications of endoglin+ (ad, heart; ae, liver; af, kidney) or CD34+ (ag, lung) vascular area (mean ± s.e.m.; n = 5 mice (4 for heart) for WT and 7 (6 for heart) for GlulvECKO, from two litters, NS, P > 0.05 versus WT, Student’s t-test). ahai, Images of flat-mounted retinas from control (ah) and MSO-treated (ai) ROP mice (vaso-obliterated area in white). Images shown are representative of 7 (ah) and 6 (ai) mice. Exact P values: HUVEC versus lung ECs: 0.0278; HUVEC versus colon ECs: 0.1086; HUVEC versus liver ECs: 0.3334; HUVEC versus HEPG2: <0.0001 (a); <0.0001 (f); 0.3491 (g); <0.0001 (h); 0.8247 (i); 0.0012 (l); 0.050 (m); superficial: 0.1218; deep: 0.1720 (p); superficial: 0.9995; deep: 0.4289 (q); superficial: 0.9792; deep: 0.6602 (t); superficial: 0.7979; deep: 0.1275 (u); 0.9021 (ad); 0.2279 (ae); 0.7647 (af); 0.3614 (ag). Scale bars: 200 μm (j, k, n, o, r, s), 20 μm (vac), 1 mm (ahai). For gel source images, see Supplementary Fig. 1. Source data

  2. Extended Data Fig. 2 Effects of silencing and pharmacological inhibition of GS on EC viability and central metabolism.

    a, GLUL mRNA levels in control ECs and ECs transduced with two different non-overlapping shRNAs targeting GLUL (GLULKD1 and GLULKD2; GLULKD1 is used in the experiments in the text and is denoted GLULKD) or transfected with scrambled siRNA (SCR) or siRNA targeting GLUL (siGLUL). Data are expressed as a percentage of the respective control, denoted by the horizontal dotted line (mean ± s.e.m.; n = 28 independent experiments for GLULKD1, n = 3 independent experiments for GLULKD2 and n = 9 independent experiments for siGLUL; *P< 0.05 versus the respective control; one-sample t-test). b, c, Quantification of number of sprouts (b) and total sprout length (c) for spheroid-sprouting assays with GLULKD ECs and GLULKD ECs expressing a shRNA-resistant GLUL mutant (rGLULOE) (mean ± s.e.m.; n = 3 independent experiments; *P< 0.05 and NS, P > 0.05 versus control; one-way ANOVA with Dunnett’s multiple comparison versus control). d, Viability of control (Ctrl) and GLULKD ECs as measured by lactate dehydrogenase (LDH) release assay (mean ± s.e.m.; n = 3 independent experiments; NS, P > 0.05 versus control, one-sample t-test). e, Intracellular levels of reactive oxygen species measured by CM-H2DCFDA staining (mean ± s.e.m.; n = 3 independent experiments; NS, P > 0.05 versus control, Student’s t-test). f, Energy charge measurement (([ATP] + 1/2[ADP]) / ([ATP] + [ADP] + [AMP])) in GLULKD and control ECs (mean ± s.e.m.; n = 3 independent experiments; NS, P > 0.05 versus control, Student’s t-test). g, Ratio of oxidized glutathione (GSSG) over total glutathione levels (GSSG/(GSH + GSSG)) in GLULKD and control ECs (mean ± s.e.m.; n = 4 independent experiments; NS, P > 0.05 versus control, Student’s t-test). h, NADP/NADPH ratio in GLULKD and control ECs (mean ± s.e.m.; n = 5 independent experiments; NS, P > 0.05 versus control, one-sample t-test). ik, Effect of GLUL knockdown on major metabolic fluxes including glycolysis (i), glucose oxidation (j) and glutamine oxidation (k) (mean ± s.e.m.; n = 3 independent experiments for i, n = 5 for j and n = 4 for k; NS, P > 0.05 versus control, one-sample t-test). l, m, Oxygen consumption rate (OCR) in control, MSO-treated and GLULKD ECs in basal state and after injection of oligomycin, FCCP and antimycin A (l) (mean ± s.e.m.; n = 3 independent experiments), and calculation of OCRBAS, OCRATP and maximal respiration (m) (mean ± s.e.m.; n = 3 independent experiments). Exact P values: GLULKD1: <0.0001; GLULKD2: <0.0001; siGLUL: <0.0001 (a); control versus GLULKD: 0.0147; control versus GLULKD + rGLULOE: 0.9824 (b); control versus GLULKD: 0.0083; control versus GLULKD + rGLULOE: 0.6528 (c); 0.5717 (d); 0.8206 (e); 0.3715 (f); 0.4398 (g); 0.9291 (h); 0.4691 (i); 0.6643 (j); 0.6786 (k). CM-DCF, CM-H2DCFDA; OCRBAS, basal oxygen consumption rate; OCRATP, ATP-generating oxygen consumption rate; RFU, relative fluorescence units.

  3. Extended Data Fig. 3 GLUL knockdown reduces EC motility.

    a, Wound closure in control and GLULKD2 EC monolayer scratch assays with or without pretreatment with mitomycin C (mean ± s.e.m.; n = 7 and 5 independent experiments with and without mitomycin C, respectively; *P< 0.05 versus corresponding control; Student’s t-test). b, Quantification of lamellipodial area (as a percentage of total cellular area) in control and GLULKD2 ECs (mean ± s.e.m.; n = 3 independent experiments; *P< 0.05 versus control; Student’s t-test). c, Wound closure in monolayer scratch assays with SCR- and siGLUL-transfected ECs (mean ± s.e.m.; n = 5 independent experiments; *P< 0.05 versus SCR; Student’s t-test). d, Quantification of lamellipodial area (as a percentage of total cellular area) in SCR- and siGLUL-transfected ECs (mean ± s.e.m.; n = 5 independent experiments; *P< 0.05 versus SCR; Student’s t-test). e, Proliferation of SCR- and siGLUL-transfected ECs, as measured by [3H]thymidine incorporation into DNA (mean ± s.e.m.; n = 3 independent experiments; NS, P > 0.05 versus SCR; Student’s t-test). Exact P values: control versus GLULKD2: 0.0290; control versus GLULKD2 + MitoC: 0.0223 (a); 0.0088 (b); 0.0407 (c); 0.0083 (d); 0.4335 (e).

  4. Extended Data Fig. 4 Effects of GLUL silencing on cytoskeleton and barrier function.

    ah, Images of control (a, c, e, g) and GLULKD (b, d, f, h) ECs after staining for α-tubulin (a, b), F-actin (c, d) and nuclear staining (e, f); g and h show merged images. The images shown are representative of 3 independent experiments. ik, Representative images of phalloidin (F-actin) + Hoechst-stained liver ECs 6 h after isolation from control (i) and MSO-treated (j) mice, and corresponding quantification of F-actin levels (k) (mean ± s.e.m.; n = 5 mice per group; *P< 0.05 versus control, Student’s t-test). ln, Representative images of phalloidin-stained confluent monolayer control (l) and GLULKD (m) ECs aligning a scratch wound, and quantification of F-actin levels (n) (mean ± s.e.m.; n = 5 independent experiments; *P< 0.05 versus control, Student’s t-test). o, Quantification of the length of discontinuous and continuous VE-cadherin-stained junctions in control and GLULKD ECs (mean ± s.e.m.; n = 4 independent experiments; *P< 0.05 versus control, Student’s t-test). p, Quantification of VE-cadherin gap size index in control and GLULKD EC monolayers (mean ± s.e.m.; n = 4 independent experiments; *P< 0.05 versus control, Student’s t-test). qv, Corresponding representative images of monolayer control (q, s, u) and GLULKD (r, t, v) ECs stained for VE-cadherin (q, r, u, v) and F-actin (s, t, u, v). Yellow arrows in r point to discontinuous VE-cadherin junctions and yellow asterisks indicate intracellular gaps. w, Quantification of transendothelial electrical resistance in control and GLULKD EC monolayers (mean ± s.e.m.; n = 4 independent experiments; *P< 0.05 versus control, Student’s t-test at each time point). xz, Quantification (x) of Evans blue dye extracted from the ears of control and MSO-treated mice, induced by topical application of mustard oil (n = 4 mice for each condition, *P< 0.05; Student’s t-test), and representative pictures of the leakage of Evans blue dye into the ear tissue in control (y) and MSO-treated (z) mice. Exact P values: 0.0030 (k); 0.0036 (n); continuous control versus GLULKD: 0.0005; discontinuous control versus GLULKD: 0.0005 (o); 0.0356 (p); 0.0181 (w); 0.0002 (x). Scale bars: 20 μm (ah, l, m), 10 μm (i, j, qv). AU, arbitrary units. Source data

  5. Extended Data Fig. 5 Enzymatic activity of GS and its role in EC migration.

    a, Scheme for the 15NH4+ labelling of glutamate and glutamine with unlabelled carbons (blue) and labelled nitrogens (red). b, 15N incorporation into glutamine (measured as the percentage of isotope enrichment in glutamine either as M + 1 (singly labelled) or M + 2 (doubly labelled), 30 min after adding 15NH4+) in medium with dialysed serum and different glutamine concentrations (mean ± s.e.m.; n = 3 independent experiments; one-way ANOVA with Dunnett’s multiple comparisons versus 4 mM; *P< 0.05). c, 15N incorporation into glutamate (measured as the percentage of isotope enrichment in M + 1) and glutamine (measured as the percentage of isotope enrichment in M + 1 (singly labelled) and M + 2 (doubly labelled)), 30 min after adding increasing concentrations of 15NH4Cl (mean ± s.e.m.; n = 3 independent experiments). d, Scheme of glutamine labelling from [U-13C]glutamate with unlabelled nitrogens (blue) and labelled carbons (red). e, Contribution of labelled [U-13C]glutamate to intracellular glutamine at various glutamine concentrations (percentage of isotope enrichment in M + 5 glutamine and glutamate, 30 min after adding the tracer) (mean ± s.e.m.; n = 3 independent experiments; one-way ANOVA with Dunnett’s multiple comparisons versus 4 mM; *P< 0.05). f, Scheme for the contribution of carbons from [U-13C]glucose to glutamine, with labelled carbons (red) and unlabelled carbons (blue). Incorporation is shown after one turn of the tricarboxylic acid (TCA) cycle. g, Total contribution of carbons from [U-13C]glucose to α-ketoglutarate, glutamate and glutamine in ECs in medium with or without glutamine, 48 h after adding the tracer (mean ± s.e.m.; n = 3 independent experiments; *P< 0.05 versus total contribution in glutamine at 0.6 mM external glutamine, one-way ANOVA with Dunnett’s multiple comparisons). h, Incorporation of 15N into glutamine (measured as the percentage of isotope enrichment in M + 1 (singly labelled) and M + 2 (doubly labelled), 30 min after adding 15NH4+) in ECs and HEPG2 cells (mean ± s.e.m.; n = 4 independent experiments. ND, not detected). i, 13C-glutamine uptake kinetics in control, MSO-treated and GLULKD ECs and subsequent conversion to glutamate. See Methods for explanation of the different time points. Data are for the M + 5 isotopomer, as a percentage of the total intracellular glutamine or glutamate pool (mean ± s.e.m.; n = 3 independent experiments, except for 30 min for which n = 1 experiment; no statistical differences between control, MSO-treated and GLULKD ECs were observed for glutamine or for glutamate; one-way ANOVA with Dunnett’s multiple comparison versus control at each time point; no statistical analysis was performed at 30 min). j, 14C-glutamine uptake in control and GLULKD ECs (mean ± s.e.m.; n = 5 independent experiments; NS, P > 0.05 versus control, one-sample t-test). k, Ratio of intracellular glutamine and glutamate levels in control and GLULKD ECs (mean ± s.e.m.; n = 3 independent experiments; NS, P > 0.05 versus control, Student’s t-test). l, Velocity measurement of control and GLULKD ECs at different glutamine concentrations (mean ± s.e.m.; n = 4 independent experiments; *P< 0.05 versus corresponding control, mixed-models R statistics). m, n, Effect of glutamine concentration on sprout number (m) and total sprout length (n) in control and GLULKD spheroids (mean ± s.e.m.; n = 3 independent experiments; *P< 0.05 versus corresponding control, mixed-models R statistics). o, p, Number of sprouts per spheroid (o) and total sprout length (p) in control and MSO-treated EC spheroids (mean ± s.e.m.; n = 3 independent experiments; *P< 0.05 versus control, paired Student’s t-test). qs, Effect of MSO-treatment on EC motility parameters: wound closure of mitomycin C-treated ECs (q) (mean ± s.e.m.; n = 11 independent experiments; *P< 0.05 versus control, Student’s t-test), lamellipodial area (r) (mean ± s.e.m.; n = 10 independent experiments; *P< 0.05 versus control, paired Student’s t-test) and F-actin levels, 1 h after latrunculin wash-out (s) (mean ± s.e.m.; n = 4 independent experiments; *P< 0.05 versus control, one-sample t-test). t, [3H]Thymidine incorporation in control and MSO-treated ECs (mean ± s.e.m.; n = 3 independent experiments; NS, P > 0.05 versus control, one-sample t-test). Exact P values: M + 1 0.025 mM versus M + 1 4 mM: 0.0096; M + 1 0.6 mM versus M + 1 4 mM: 0.1206; M + 2 0.025 mM versus M + 2 4 mM: 0.0839; M + 2 0.6 mM versus M + 2 4 mM: 0.9921 (b); Glu M + 5 0.6 mM versus Glu M + 5 4 mM: 0.9372; Glu M + 5 0.025 mM + MSO versus Glu M + 5 4 mM: 0.0034; Glu M + 5 0.025 mM versus Glu M + 5 4 mM: 0.0215; Gln M + 5 0.6 mM versus Gln M + 5 4 mM: 0.9297; Gln M + 5 0.025 mM + MSO versus Gln M + 5 4 mM: 0.9961; Gln M + 5 0.025 mM versus Gln M + 5 4 mM: 0.0268 (e); α-keto 0.6 mM versus Gln 0.6 mM: 0.0001; Glu 0.6 mM versus Gln 0.6 mM: 0.0001; Gln 0 mM versus Gln 0.6 mM: 0.0285 (g); Gln 0.5 min: control versus MSO: 0.4846; control versus GLULKD: 0.5904; Gln 10 min: control versus MSO: 0.6709; control versus GLULKD: 0.6910; Gln 20 min: control versus MSO: 0.5896; control versus GLULKD: 0.6784; Glu 0.5 min: control versus MSO: 0.9774; control versus GLULKD: 0.8810; Glu 10 min: control versus MSO: 0.0502; control versus GLULKD: 0.9598; Glu 20 min: control versus MSO: 0.9782; control versus GLULKD: 0.7783 (i); 0.6623 (j); 0.6704 (k); control versus GLULKD 0.1 mM: 0.0054; control versus GLULKD 0.6 mM: 0.0247 control versus GLULKD 2 mM: 0.0017 (l); control versus GLULKD 0.6 mM and 10 mM: < 0.0001 (m); control versus GLULKD 0.6 mM and 10 mM: < 0.0001 (n); 0.0313 (o); 0.0075 (p); 0.0019 (q); 0.0116 (r); 0.0091 (s); 0.5110 (t). α-keto, α-ketoglutarate; GDH, glutamate dehydrogenase.

  6. Extended Data Fig. 6 Rescuing the phenotype associated with GLULKD in vitro.

    a, Schematic of the DORA–RHOA–FRET biosensor, depicting from N- to C-terminal the circular permutated RHOA effector protein kinase N (cpPKN), the dimeric circular permutated Venus (dcpVen), the ribosomal protein-based linkers (L9), the dimeric Cerulean3 (dCer3) and RHOA. bm, Representative images of control (bd), MSO-treated (eg), GLULKD (hj) and RHOJKD (km) ECs after staining for F-actin (phalloidin) (b, d, e, g, h, j, k, m) and pMLC (c, d, f, g, i, j, l, m). n, Quantification of the pMLC immunoreactivity (mean ± s.e.m.; n = 5 independent experiments; *P< 0.05 versus control, one-sample t-test). ot, Representative images of control (o, q, s) and GLULKD (p, r, t) EC spheroids treated with vehicle (o, p) or the ROCK inhibitors Y27632 (q, r) or fasudil hydrochloride (Fasu.; s, t). u, v, Quantification of the number of sprouts per spheroid (u) and sprout length (v) (mean ± s.e.m.; n = 3 independent experiments; *P< 0.05 and NS, P > 0.05 versus untreated control, one-way ANOVA with Dunnett’s multiple comparisons versus untreated control). w, Quantification of the lamellipodial area in vehicle- or fasudil-hydrochloride-treated control and GLULKD ECs (mean ± s.e.m.; n = 6 independent experiments; *P< 0.05 and NS, P > 0.05 versus untreated control, one-way ANOVA with Dunnett’s multiple comparisons versus untreated control). x, Quantification of the lamellipodial area in vehicle-, ML7- or peptide-18 (pep.18)-treated GLULKD and control ECs (mean ± s.e.m.; n = 4 independent experiments of which 3 experiments included the ML7-treatment; *P< 0.05 versus untreated control, one-way ANOVA with Dunnett’s multiple comparisons versus untreated control). y, Scratch wound closure in vehicle-, ML7- or peptide-18-treated GLULKD and control ECs (mean ± s.e.m.; n = 3 independent experiments; *P< 0.05 versus untreated control, one-way ANOVA with Dunnett’s multiple comparisons versus untreated control). z, Fold changes (versus untreated control ECs) in F-actin levels from phalloidin-stained vehicle-, ML7- or peptide-18-treated GLULKD ECs (mean ± s.e.m.; n = 4 independent experiments of which 3 included the peptide 18-treatment; *P< 0.05 versus untreated control, one-sample t-test). aa, Fold changes (versus untreated control ECs) in pMLC levels from pMLC-immunostained vehicle-, ML7- or peptide-18-treated GLULKD ECs (mean ± s.e.m.; n = 4 independent experiments of which 3 included the peptide 18-treatment; *P< 0.05 versus untreated control, one-sample t-test. Exact P values: MSO: 0.0372; GLULKD: 0.0060; RHOJKD: 0.0051 (n); GLULKD versus control: 0.0045; Fasu. versus control: 0.9596; GLULKD + Fasu. versus control: 0.8857 (u); GLULKD versus control: 0.0199; Fasu. versus control: 0.8309; GLULKD + Fasu. versus control: 0.9327 (v) GLULKD versus control: 0.0074; Fasu. versus control: 0.5906; GLULKD + Fasu. versus control: 0.9900; (w); GLULKD versus control: 0.0011; GLULKD + ML7 versus control: 0.0079; GLULKD + pep.18 versus control: 0.0017 (x); GLULKD versus control: 0.0034; GLULKD + ML7 versus control: 0.0022; GLULKD + pep.18 versus control: 0.0040 (y); GLULKD: 0.0058; ML7: 0.0072; pep.18: 0.0888 (z); GLULKD: 0.0369; ML7: 0.0021; pep.18: 0.1672 (aa). Scale bars: 20 μm (bm), 100 μm (ot). For gel source images, see Supplementary Fig. 1.

  7. Extended Data Fig. 7 Rho GTPase localization and interaction with GS.

    a, Co-IP assays showing no detectable interaction between GS and RHOA or RHOC (red asterisk indicates a non-specific band (also present in the IgG controls and unaffected by silencing of RHOA or RHOC)). Image shown is representative of 3 independent experiments. b, Co-IP of overexpressed GLUL and RHOJ-eGFP or ΔN20-RHOJ-eGFP in ECs. Quantifications are mean ± s.e.m.; n = 4 independent experiments; *P< 0.05, one-sample t-test. In some of the experiments, the expression of ΔN20-RHOJ–eGFP was lower than the expression of RHOJ–eGFP. To correct for this, densitometric quantification was performed and signals in immunoprecipitation lanes were normalized to input signals. c, Immunoblotting for RHOA and RHOC on cytosolic (c) and membrane (m) fractions of ECs with NaK as membrane marker and GAPDH as cytosolic marker. Image shown is representative of 3 independent experiments. d, BiFC assay with GS coupled to the N-terminal half of eGFP, and RHOJ coupled to the C-terminal half of eGFP. Only when GS and RHOJ are in close proximity do the two eGFP half-sites complement each other and form a functional eGFP. e, Percentage of ECs displaying BiFC upon overexpression of GLUL-eGFP1/2 and RHOJ-eGFP2/2 or GLUL-eGFP1/2 and ΔN20-RHOJ-eGFP2/2. Data are mean ± s.e.m.; n = 3 independent experiments; *P< 0.05; Student’s t-test. f, Schematic of SPT-PALM imaging under TIRF illumination with the plasma membrane depicted at the top. The TIRF region is bright (whereas the part outside the TIRF region is greyed out) and contains the plasma membrane and its immediately adjacent space (not shown at exact relative dimensions). Weight and number of arrowheads represent the velocity of single particles (the photoswitchable fluorescent protein (PSFP) or the PSFP coupled to the protein of interest (here GS)). The PSFP is activated upon entry into the TIRF region and is colour-coded differently inside and outside of the TIRF region. PSFP–GS displays reduced velocity in the TIRF region, presumably because of palmitoylation and membrane association of GS. g, Scheme for the in-cell labelling of proteins with clickable alkyne-containing palmitoylation probes and subsequent biotin-azide clicking. X represents a palmitoylated protein, N3 is the biotin-coupled azide. h, i, Rate of CoA release from palmitoyl-CoA as a readout for recombinant human GS autopalmitoylation while varying either the doses of palmitoyl-CoA (h) or the amounts of recombinant GS (i) (mean ± s.e.m.; n = 4 independent experiments for h and n = 5 for i; *P< 0.05, one-way ANOVA with Dunnett’s multiple comparisons versus 0 µM palmitoyl-CoA or versus 0.5 µg recombinant GS). j, Representative GS immunoblot (of 3 independent experiments) for the binding of recombinant human GS to palmitoyl-CoA agarose. IF, input fraction; FT, flow through; W8, wash fraction 8; SDS is the eluate. km, Representative images of RHOJ–eGFP localization in ECs under vehicle-treatment (k) or treatment with 2BP (a pan-palmitoylation inhibitor) (l). Red arrowheads indicate eGFP signal at membrane ruffles, which was quantified as the percentage of the total cellular area (m) (mean ± s.e.m.; n = 4 independent experiments; *P< 0.05 versus vehicle-treated, paired Student’s t-test). np, Representative images of ECs overexpressing wild-type RHOJ-eGFP (n), RHOJ-eGFP encoding the point mutation C3A (o) or RHOJ-eGFP encoding the point mutation C11A (p). Red arrowheads indicate RHOJ at the plasma membrane. ECs that are not completely in the field of view have been masked out in blue. q, Quantification of the RHOJ–eGFP positive area at the plasma membrane as a percentage of the total cell area. Data are mean ± s.e.m.; n = 5 independent experiments; *P< 0.05; one-way ANOVA with Dunnett’s comparison versus wild-type RHOJ. r, RHOJ immunoblotting on membrane versus cytosolic fractions from ECs overexpressing wild-type RHOJ-eGFP (RHOJWT), RHOJ-eGFP encoding the point mutation C3A (RHOJC3A) or RHOJ-eGFP encoding the point mutation C11A (RHOJC11A), with NaK as membrane marker and GAPDH and α-tubulin as cytosolic markers. s, Densitometric quantification of RHOJ/NaK as determined in r. Data are mean ± s.e.m.; n = 6 independent experiments; *P< 0.05; one-sample t-test. t, RHOJ activity in ECs under treatment with vehicle or 2BP (blots are representative of 3 independent experiments; densitometric quantification in arbitrary units is mean ± s.e.m.; *P< 0.05, paired Student’s t-test versus vehicle-treated). u, RHOJ immunoblotting of control and GLULKD ECs overexpressing RHOJ (RHOJOE) subjected to acyl-resin-assisted capture. The cleaved bound fraction (cBF) represents palmitoylated RHOJ. IF is the input fraction, whereas the cleaved unbound fraction (cUF) and the preserved bound fraction (pBF) are controls showing the depletion of RHOJ from the thioester-cleaving reagent and the near absence of non-specific binding of RHOJ to the resin (see Methods). Densitometric quantification of cBF/IF is shown (mean ± s.e.m.; n = 3 independent experiments; *P< 0.05, one-sample t-test versus control). v, Left, autopalmitoylation enables endothelial GS to interact directly (or indirectly) with the Rho GTPase RHOJ and to sustain the palmitoylation, membrane localization and activity of RHOJ (reflected by GTP binding). RHOJ activity then sustains normal EC migration and lamellipodia formation, and keeps actin stress-fibre formation at levels that promote normal EC migration and vessel branching in vivo. Through mechanisms that are not completely understood, active RHOJ inhibits signalling of the RHOA/B/C–ROCK–(p)MLC pathway (itself known to promote stress-fibre formation). The relative contribution of a direct effect of RHOJ on migration versus the indirect effect through RHOA/B/C–ROCK–(p)MLC is yet to be determined. Reduced opacity of RHOA/B/C, ROCK and (p)MLC indicates reduced signalling of this pathway. Right, loss of endothelial GS renders RHOJ less active (visually reflected by fewer palmitoylated, membrane-bound RHOJ proteins), and reduces the inhibition of the RHOA/B/C–ROCK–(p)MLC pathway. The resulting excessive stress-fibre formation causes ECs to lose migratory capacity and reduces vessel branching in vivo. Dashed lines indicate reduced activity; the red cross indicates GS blockade; the question mark indicates unknown mechanisms. Exact P values are as follows: 0.0153 (b); 0.0334 (e); 2 versus 0 μM: 0.6327; 5 versus 0 μM: 0.2841; 10 versus 0 μM: 0.1090; 20 versus 0 μM: 0.0339; 40 versus 0 μM: 0.0034 (h); 1 versus 0.5 μg: 0.5806; 2 versus 0.5 μg: 0.0319; 4 versus 0.5 μg: 0.0037; 8 versus 0.5 μg: 0.0001; 16 versus 0.5 μg: 0.0001 (i); 0.0313 (m); RHOJ C3A versus RHOJ WT: 0.0001; RHOJ C11A versus RHOJ WT: 0.0001 (q); RHOJ C3A versus RHOJ WT: 0.0015; RHOJ C11A versus RHOJ WT: 0.0007 (s); 0.0051 (t); 0.0461 (u). Scale bar, 200 μm (k, l, np). For gel source images, see Supplementary Fig. 1.

  8. Extended Data Fig. 8 Possible molecular model of GS autopalmitoylation.

    a, Structure of human GS, showing its bifunnel-shaped catalytic site. A schematic of the GS decamer is shown from the top and front views with individual subunits A and B labelled and coloured grey and green, respectively. On the right is a close-up of the bifunnel catalytic site that is formed between subunits A and B. The GS decamer has ten active sites, each located at the interface of two adjacent subunits. ATP enters from the top, whereas glutamate enters from below; manganese ions (Mn2+) are shown as grey spheres. b, Molecular dynamics simulation of palmitoyl-CoA in the catalytic cleft of GS predicts that, whereas the head of palmitoyl-CoA is tightly bound to the adenine-binding site, the tail can point in opposing directions with respect to the principal axis of the protein. The most representative structures of the two alternative conformations (A and B) observed during the long molecular dynamics simulations for palmitoyl-CoA binding to GS (in blue, seen from two different perspectives) are shown in red (A, tail bending upwards) and green (B, tail bending downwards). c, Detailed view of conformation A, which is the main conformation. The sulfur atom of palmitoyl-CoA (which is immediately adjacent to the carbon on which the nucleophilic attack occurs) (coloured yellow) approaches the highly conserved cysteine 209 (also coloured yellow), with an interatomic distance (S–S) that, during the simulations, reversibly fluctuates between 3 and 8 Å. The hydrophobic tail positions itself along grooves characterized by the presence of hydrophobic residues. Colour coding is as follows: carbon, grey; nitrogen, blue; phosphorous, gold; oxygen, red. Cysteines and serines within 5 Å of the palmitoyl tail are highlighted in yellow and orange, respectively. The hydrophobic residues around the tail are shown in green. d, Detailed view of conformation B, in which the tail is found in a buried hydrophobic cleft, with the sulfur at a distance of 5 Å or less from the conserved serines 65 and 75 and the tail occupying the site of the GS inhibitor MSO. Details of the extensive steric clash between MSO and the secondary binding pose (B) observed in palmitoyl-CoA MD simulations are shown. Palmitoyl-CoA is represented as sticks, with standard atomic colours as stated in c. MSO is shown in cyan and its position is taken from entry 2QC8 of the Protein Data Bank. Cysteines and serines within 5 Å of the palmitoyl tail are highlighted in yellow and orange, respectively. The hydrophobic residues around the tail are shown in green. e, GS immunoblotting after streptavidin pull-down of biotin-azide-clicked lysates from 16C-YA (palmitoylation probe) labelled HEK-293T cells overexpressing wild-type GLUL or GLUL with a point mutation in C209. The input shows the level of GS overexpression. A representative blot from 4 independent experiments is shown. f, g, Quantification of total sprout length (f) and number of sprouts per spheroid (g) for control and GLULKD ECs with or without overexpression of shRNA-resistant GLUL encoding the point mutation C209A (rGLULC209A-OE) (mean ± s.e.m.; n = 4 independent experiments; *P< 0.05 versus control, one-way ANOVA with Dunnett’s multiple comparison versus control). h, Schematic of protein autopalmitoylation. Upon binding of palmitoyl-CoA to the protein, free CoA (grey oval) is released and can be detected. i, Recombinant wild-type GS and GS with point mutations R324C and R341C were incubated with different concentrations of palmitoyl-CoA in a cell-free system at physiological pH. The amount of CoA released per minute was determined as a direct readout for protein autopalmitoylation. Data are mean ± s.e.m. of 3 (R324C and R341C) and 4 (WT) independent experiments. NS, P > 0.05; *P< 0.05 according to two-way ANOVA comparing the entire dose–response to the dose–response of wild-type GS. j, Different amounts of recombinant wild-type, R324C and R341C GS were incubated with a fixed concentration of palmitoyl-CoA (40 μM), and the amount of CoA released per minute was determined as readout for autopalmitoylation. Data are mean ± s.e.m. of 4 (R324C and R341C) and 5 (wild-type) independent experiments. NS, P > 0.05; *P< 0.05 according to two-way ANOVA comparing the entire dose–response to the dose–response of wild-type GS. The data for wild-type GS from i and j are also included in Extended Data Fig. 7 as stand-alone data, but are included here for comparison purposes. k, Boyden chamber migration for control, GLULKD, GLULKD + rGLULOE, GLULKD + rGLULR341C-OE and GLULKD + rGLULR324C-OE ECs, all under mitomycin C-treatment (mean ± s.e.m.; n = 3 independent experiments; NS, P > 0.05; *P< 0.05, one-way ANOVA with Dunnett’s multiple comparison versus control). Exact P values: GLULKD versus control: 0.0004; GLULKD + rGLULC209A-OE versus control: 0.0004 (f); GLULKD versus control: 0.0001; GLULKD + rGLULC209A-OE versus control: 0.0001 (g); R324C versus WT: 0.8228; R341C versus WT: 0.7530 (i); R324C versus WT: 0.1331; R341C versus WT: 0.0003 (j); GLULKD versus control: 0.0054; GLULKD + rGLULOE versus control: 0.8152; GLULKD + rGLULR341C-OE versus control: 0.3645; GLULKD + rGLULR324C-OE versus control: 0.2118 (k). For gel source images, see Supplementary Fig. 1.

  9. Extended Data Table 1 Weight, haematological and blood plasma parameters for six-week-old GlulvECKO mice and control littermates
  10. Extended Data Table 2 Alignment of the amino acid sequence encompassing the C209 residue across different species

Supplementary information

  1. Supplementary Discussion

    This file contains Supplementary discussions 1-5, all of which contain additional rationale and discussion of experimental data and were not included in the main manuscript due to space restrictions.

  2. Reporting Summary

  3. Supplementary Figures

    This file contains uncropped gel pictures and blots with size markers

  4. Video 1

    Time lapse imaging of randomly moving control ECs Time lapse video of sparsely seeded control ECs. Video shown is representative of nine independent experiments

  5. Video 2

    Time lapse imaging of randomly moving GLULKD ECs Time lapse video of sparsely seeded GLULKD ECs. Video shown is representative of nine independent experiments

Source data

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/s41586-018-0466-7

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.