Supplementary information:
VEGF-INDUCED ENDOTHELIAL CELL MIGRATION REQUIRES
UROKINASE RECEPTOR (uPAR)-DEPENDENT INTEGRIN
REDISTRIBUTION
Revu Ann Alexander1,†, Gerald W. Prager1,3, †, Judit Mihaly-Bison1, Pavel Uhrin1, Stefan
Sunzenauer4, Bernd R. Binder1*, Gerhard J. Schütz4, Michael Freissmuth2, and Johannes M. Breuss1¶
Materials:
VEGF165 was from Promocell GmbH (Heidelberg, Germany); VEGF-E and PIGF were from
RELIATech
(Wolfenbüttel,
Germany);
recombinant
mouse
VEGF164
and
(2R)-2-[(4-
biphenylylsulfonyl)amino]-3-phenylpropionic acid (matrix metalloprotease-2/-9 = MMP-2/MMP-9
inhibitor) from Merck (Darmstadt, Germany); wortmannin was from Sigma Chemicals (St. Louis,
MO). The rat receptor-associated protein (RAP, 39kD Protein) was made as a recombinant fusion
protein with glutathione S-transferase as previously described.1 The synthetic human uPAR inhibitory
peptides were from Genosphere Biotechnologies (Paris, France). The sequences of the peptides are:
peptide 243-251 TASMCQHAH and scrambled peptide 244-252 SCAAMHQHT. The murine uPARderived peptide m.P243-251 (TASWCQGSH) was purchased from piChem (Graz, Austria). Matrigel
solution was from Becton-Dickinson (San Jose, CA).DIVAA kit was purchased from Trevigen
(Gaithersburg, MD). The predesigned uPAR siRNA and the scrambled siRNA was from Ambion
(LifeTech, Austria).
Antibodies: uPAR - 3937 monoclonal was from American Diagnostica (Greenwich, CT); the anti
uPAR monoclonal antibody R2 was a kind gift of Gunilla Hoyer-Hansen (Finsen Laboratory,
Copenhagen, Denmark)2; the P4C10 mouse monoclonal antibody against β1 integrin was from Sigma
(St.Louis, MO) a biotinylated version (MEM-101A) was from LifeSpan Biosciences (Seattle, WA);
alternatively we also used the I2G10 from Novus Biologicals (Littleton, CO). The rat monoclonal
antibody against mouse integrin subunit beta-1 was from Becton Dickinson GmbH (Heidelberg,
Germany).
The antibodies P1D6 and P1B5 against the α5 integrin and α3 integrin subunits were also from
Chemicon International; a biotinylated version against the integrin α5 subunit (X-6) was from Acris
(Rancho Santa Margarita, CA). The Clathrin heavy chain rabbit monoclonal antibody and rabbit
polyclonal antibody against P-FAK (Tyr397) was from Cell Signaling Technology (Beverly,MA). The
rabbit polyclonal antibody against Beta3 was from Millipore (Billerica, MA) and FAK rabbit
polyclonal antibody was from Calbiochem (Darmstadt, Germany). The biotinylated anti mouse
antibody from Sigma (St.Louis, MO) and rat monoclonal antibody against mouse CD31 antibody from
Dianova (Hamburg, Germany)
Rabbit antiserum 7 raised against amino acids 8 to 23 of the G protein 1/2-subunits was used as
loading control for surface biotinylation experiments. 3 Alexa Fluor 488 or 568-conjugated anti-mouse
or anti-rabbit IgG antibodies and the Zenon One Mouse IgG1 Labeling Kit were purchased from
Molecular Probes (Leiden, The Netherlands). Alexa Fluor 488 conjugated Fab-fragments (Zenon kit)
were from Molecular Probes (Leiden, The Netherlands); Streptavidin 647 from Molecular Probes
(Leiden, The Netherlands). Goat serum was from (Dako Diagnostics, Vienna, Austria).
M199 (Sigma-Aldrich, St. Louis, MO), Bovine endothelial cell growth supplement (Technoclone,
Vienna, Austria), Dulbecco modified Eagle medium (Invitrogen, Carlsbad, CA), heparin (PromoCell;
Heidelberg, Germany), Triton X-100 (Sigma, MO), sulfo-NHS-LC-biotin (Pierce, Rockford, IL),
Complete™ protease inhibitors, Roche Diagnostics GmbH, Vienna, Austria), streptavidin-coated
Sepharose 4B beads (Zymed Laboratories, San Francisco, CA), Femto-ECL reagents for enhanced
chemiluminescence from Pierce (Rockford, IL), Adhesive silicone masks from (Secure Seal, Grace
Biolabs, Germany), Transwell™ inserts from Corning LifeScienes (Lowell, MA), X-tremeGENE
siRNA transfection reagent from Roche, (Diagnostics GmbH, Vienna, Austria).
Methods:
Genotyping of uPAR-/- mice
Urokinase receptor-deficient mice were kindly provided in 25% 129S/v: 75% C57BL/6 background
by Dr. Peter Carmeliet, Center for Transgene Technology and Gene Therapy, Leuven, Belgium. They
were subsequently backcrossed in our laboratory for 6 generations into C57BL/6 background.
Genotyping was made by PCR amplification using an Eppendorf mastercycler epgradient S (Perkin
Elmer, Austria) at initial 94°C for 2 min, followed by 35 cycles consisting of 94°C for 35 sec, 60°C
for 35 sec and 72°C for 35 sec, with last step extended to 7 min in the last cycle. Knockout allele was
amplified using a transgene specific primer 5'-CAT CGC CTT CTA TCG CCT TCT TGA C-3' in
combination with wild-type specific reverse primer: 5'-TTG ATG AGA GAC GCC TCT TCG GAG
G-3', resulting in a product of approximate size of 0.97 kb. Wild-type allele was detected using
forward primer 5'-TTA CCT CGA GTG TGC GTC CTG CAC-3' in combination with reverse primer
5'-CCA CCA TTG CAG TGG GTG TAG TTG-3', resulting in a 0.78 kb product.
RIPA buffer composition
0.1% sodium dodecyl sulfate (SDS), 1% triton X-100, 1% deoxycholate, 0.15 M NaCl and 0.1 M TrisHCl, pH 7.6 and Complete™ protease inhibitors
DIVAA in vivo angiogenesis assay
For the DIVAA in vivo angiogenesis assay, angioreactors of 1.0 cm length and 0.15 cm diameter were
filled with basement membrane extract either alone or containing 2 µg/ml VEGF164 and 1mg/ml
heparin with or without 250µM of the mouse inhibitory peptide at 4oC. One reactor on each flank was
implanted subcutaneously into 6-8 week old BL6 mice anesthetized as in the matrigel assay. The
reactors were excised on the 11th day after sacrificing the mice. The reactors were frozen in liqN2.
Sections were stained with rat anti-mouse CD31 antibody and Hoechst, photographed and analysed
using the CellP® imaging software (Olympus).
Supplementary Figures:
Supplementary Figure 1. Dose-dependency of VEGF-induced internalization of integrin subunit-1.
HUVECs were stimulated with VEGF165 (1-50 ng/ml) for 60 min. Samples were analyzed for integrin
subunit 1 surface expression by flow cytometry. Line graph summarizes data from 3 experiments.
Supplementary Figure 2.
Long term VEGF-treatment of HUVECs induces cleavage of uPAR-
domain-1. HUVECs were stimulated with either VEGF165 (50ng/mL) or phorbol-12-myristate-13acetate (PMA) (10µM) for the indicated time durations. Thereafter, cells were lysed in sample buffer
containing 4% SDS and -mercaptoethanol. Proteins were separated on SDS polyacrylamide gels,
transferred to PVDF membrane and probed for uPAR (antibody R2) and actin. The graph summarize
three experiments (means±S.E.) plotted as normalized values (to actin) of full length uPAR or D2+D3.
Representative blot of an experiment is shown. VEGF induces cleavage of uPAR-domain-1 (D1) after
36 hours (-■- in graph). Cleavage of uPAR removes the D1-domain and gives rise to a cell attached
D2-D3-stump.4 The cleavage may be accounted for by VEGF-induced activation of metalloproteases
that cause cleavage of surface molecules.4 Total uPAR levels, however, did not decline but modestly
increased over several hours in VEGF-stimulated endothelial cells (-▲- in graph). In the presence of
the PMA, which was used as positive control, D2-D3-fragments accumulated already after 6 h (blot,
lane 3). In contrast, VEGF-induced cleavage was modest after 6 h but increased up to 36 h. These
observations support the interpretation that the observed (Fig.1 and Fig.5) loss of surface-uPAR upon
VEGF-treatment of endothelial cells is caused by internalization of uPAR rather than by GPI-anchor
cleavage and thus release from the cell surface.
Supplementary Figure 3.
VEGF reduces colocalization of 1-integrins with phosphorylated focal
adhesion kinase. Triple immunostaining of HUVECs for uPAR (red), integrin subunit 1(green) and
pFAK (blue) was done following VEGF (50ng/mL) stimulation for 60 min. In unstimulated cells
pFAK colocalised with integrin subunit 1-staining in point contacts (blue arrows) and in occasional
focal adhesions (blue arrow head). Upon VEGF-stimulation, colocalization of β1-integrins and
phosphorylated FAK was reduced. VEGF promoted co-localization of uPAR and β1-integrins in
vesicular structures(orange arrows) and increased colocalization of uPAR/pFAK in focal adhesions
(pink arrows); scale bars 10µm
Supplementary Figure 4.
1-integrins colocalize with uPAR in polarized endothelial cells upon
VEGF-stimulation. A scratch wound was created on a confluent HUVEC-monolayer. Cells were
allowed to migrate overnight into the scratch area and thus become polarized. Cells were either
pretreated with or without RAP (200nM) for 15 min followed by stimulation with VEGF165 (50ng/mL)
for 60 min. Samples were fixed, permeabilized and immuno-stained for uPAR (red), the integrin
subunits β1 (green) and β3 (blue). The effects seen in polarized cells were reminiscent of the effects in
resting endothelial cells. Unstimulated polarized migrating cells showed little colocalization of
integrin β1-subunit and uPAR, while focal adhesions stained positive for integrin-β3 and uPAR. Upon
VEGF stimulation, the β1-integrin subunit colocalized with uPAR in vesicles (white arrows) while β3
and uPAR were colocalized in focal adhesions. The addition of RAP prevented colocalization of
uPAR and integrin β1 (as in non-polarized cells). scale bars 10 and 2,5µm
Supplementary Figure 5. Dose dependent inhibitory effect of Pep 243-251 on the VEGF-induced
and uPAR mediated internalization of integrin α5β1. HUVECs were pretreated with either Pep 243251 or Pep Scrambled in dose dependent manner for 15 min and stimulated with VEGF 165 (50ng/ml)
for 60 min. Samples were analyzed for integrin subunit β1 surface expression by flow cytometry. The
inhibitory effect of Pep 243-251 on VEGF-induced internalization was statistically significant
beginning with a peptide concentration of 0.25 µM. Bar diagramme summarizes 5 experiments
(means±S.D.) with immunostaining in control cells representing 100%.
Supplementary Figure 6. Urokinase receptor knock down by siRNA impairs the VEGF-induced
internalisation of integrin subunit 1. HUVECs were transiently transfected with uPAR siRNA
(100nM) for 36 h. The samples were stimulated with VEGF165 (50ng/ml) for 60 min and analyzed for
1 integrin subunit surface expression by flow cytometry. Bar diagramme summarize data from 3
experiments (means ± S.D.) (**p < 0.01).
Supplementary References.
1. Schwarz HP, Lenting PJ, Binder B, Mihaly J, Denis C, Dorner F, et al. Involvement of lowdensity lipoprotein receptor-related protein (lrp) in the clearance of factor viii in von
willebrand factor-deficient mice. Blood 2000;95:1703-1708.
2. Ronne E, Behrendt N, Ellis V, Ploug M, Dano K, Hoyer-Hansen G. Cell-induced potentiation
of the plasminogen activation system is abolished by a monoclonal antibody that recognizes
the nh2-terminal domain of the urokinase receptor. FEBS Lett 1991;288:233-236.
3. Hohenegger M, Mitterauer T, Voss T, Nanoff C, Freissmuth M. Thiophosphorylation of the G
protein beta subunit in human platelet membranes: evidence against a direct phosphate
transfer reaction to G alpha subunits. Mol Pharmacol 1996;49:73-80.
4. Chavakis T, Willuweit AK, Lupu F, Preissner KT, Kanse SM. Release of soluble urokinase
receptor from vascular cells. Thromb Haemost 2001;86:686-693.