SUPPLEMENTARY INFORMATION

Transcription

1 DOI: /ncb2271 Supplementary Figure a! WM266.4 mock WM266.4 #7 sirna WM266.4 #10 sirna SKMEL28 mock SKMEL28 #7 sirna SKMEL28 #10 sirna WM1361 mock WM1361 #7 sirna WM1361 #10 sirna 9 WM266. WM SKMEL28! % elongated cells! * 95! 9! 8! 75! b! 5! WM266. * WM136 SKMEL28! * p-mlc * 1.5! 0.5! α pmlc α tubulin! #sirna α Figure S1 Ras2 depletion affects the morphology of melanoma cells. (a) Micrographs showing the morphologies of WM266.4, WM1361 and SKMEL28 melanoma cell lines transfected with scrambled control sirnas (mock) or with sirna #7 against Ras2, cultured in a thick layer of collagen. Scale bar: 100 μm. Bar charts show the percentage of elongated cells after transfection of 2 sirnas against Ras2 (#7, 10) (0/cells per experiment n=3. Error bars indicate SE. p values: * p< 0.05 by Student s t test). (b) Down-regulation of Ras2 increases phosphorylation of MLC2. Total lysates of WM266.4, WM1361 and SKMEL28 cells transfected as in (a) were monitored for phosphorylated MLC2 and the indicated proteins by immunoblotting. Bar charts show the fold increase in p-mlc2 relative to mock transfected cells (n=3). Error bars indicate SE. p values: * p< 0.05; ** p<0.01 by Student s t test Macmillan Publishers Limited. All rights reserved.

2 Supplementary Figure a! PH1 IQ DH * PH2 REM Cdc Ras ΔPH1 ΔIQ ΔDH ΔPH2 ΔCdc 1/2- ΔDH- DH- b! rounded elongated CONTROL RI: Y27632 Ras2 WT Ras 2ΔCdc Ras 2ΔDH Figure S2 (a) Diagram depicting Ras constructs used in the study. Ras2 constructs present an N-terminal Flag epitope that permits their detection by immunobloting. -tagged constructs of Ras1/2 have their Cdc domain replaced by. The DH domain of Ras1 (DH1) and Ras2 (DH2) was also epitoped with. Notice that a common region (*) corresponding to aminoacids in Ras1, is deleted in both ΔDH and ΔPH2 mutants, whereas such region is present in the DH- constructs. (b) Ras2 affects the morphology of A375M2 melanoma cells. Pictures show the morphologies of A375M2 cells: control, control treated with the ROCK inhibitor RI: Y27632, stably trasfected with Ras2: wild type, Cdc or ΔDH cultured in a thick layer of collagen. Red arrows show cells with typical elongated morphology. Green arrows show rounded cells. Scale bar: 100 μm Macmillan Publishers Limited. All rights reserved.

3 Supplementary Figure a! Rho Rac fold activation! C si 2 C si 2 C si 2 C si 2 starved growing starved growing A375 M2 b! -GTP! A375M2 Jurkat c! p-pak! 2.5! 1.5! ** 0.5! C si 2 C si 2 C si 2 C si 2 d! starved growing starved growing -GTP! sirna #7 p MLC A375M2 -GTP! total! α p-mlc! α! α Flag! C C wt ΔDH Δ! wt! ΔDH, Δ! C C wt ΔDH Δ! sirna #7 Figure S3 (a) Statistics for Rac1 and Rho activation levels in A375M2 and cells after depletion (si) or over-expression of Ras2, in cells exponentially growing or serum-starved for 24 h. n= 3, error bars indicate SE. Student s t test reveals no significance relative to the corresponding controls. (b) Statistics for activation levels in A375M2, Jurkat and cells after depletion (si) or over-expression of Ras2, in exponentially growing or 24 h-starved cells. n = 5, error bars indicate SE. p values, * p< 0.05 by Student s t test. (c) Statistics for Pak1 phosphorylation levels in A365M2 cells under the same conditions. n = 5, error bars indicate SE p values, *** p< by Student s t test relative to the corresponding controls. (d) Ectopic expression of Flag-tagged murine Ras2 wild-type and ΔCdc but not the ΔDH mutant reverts activation and phospho-mlc2 levels resulting from the down-regulation of endogenous Ras2 by sirna #7, in Hela and A375M2 cells, blots correspond to A375M2 cells. n = 5, error bars indicate SE. p values, * p< 0.05 by Student s t test relative to the values in control cells transfected with sirna#7. Right panels: representative western blots for the indicated proteins Macmillan Publishers Limited. All rights reserved.

4 a IP Sos Jurkat TL PI TL IP TL PI TL IP H-Ras Sos b HA-Ras1 N17 N17 wt QL Flag- HA-Ras1 N17 - N17 wt QL Flag- IP Rho A Rho A PI IP Flag TL PI IP HA IP Rac 1 Flag-Ras2 Flag-Ras2 c Rac 1 N17 N17 wt QL AU5- N17 - N17 wt QL AU5- H-Ras-GTP st FS TNF LPA iono PI IP AU5 PI IP Flag Total H-Ras TL e C 1 DH(1) 2 DH(2) st st TNF LPA iono -GTP Total Fold HA PI IP TL d GST-RhoA GST-Rac1 GST- f C - 1 Cdc DH input NF GDP GTP S NF GDP GTP S NF GDP GTP S -GTP Total HA 1- Fold DH1- p-pak1 DH2- Dock10- Vav- fold increase p-pak * *** * ns Pak1 0 Figure S4 GDP-bound preferential associates with Ras GEFs. (a) Sos1 does not bind to, neither Rho or Rac1 bind to Ras2. Lysates from and Jurkat cells were immunoprecipitated for Sos1, RhoA and Rac1 and immunoprecipitates were probed for the indicated proteins. (b) Rass bind to GDP-bound. COS-7 cells were transfected with plasmids encoding HA-Ras1 or Flag-Ras2, and, respectively, Flagor AU5-epitoped constructs: N17, wt or QL. Rass or s were specifically immunoprecipitated with anti-ha, anti-flag or anti-au5 antibodies, as indicated and associated proteins were revealed by immunoblotting. (c) Top panel: Agonist-induced Ras activation in cells, starved (st) or after 2 min stimulation with the indicated stimuli. FS=foetal serum. Lower panel: agonist-induced binding of Ras2 to. The association of the endogenous proteins was analyzed in agonist-stimulated cells. (d) Rass bind to GDP-bound in vitro. The indicated, bacterially-purified, GST-Rho GTPases nucleotide free (NF), GDP- or GTPγS- loaded, were incubated with lysates of cells expressing the indicated -tagged GEFs and Ras 1 and 2 DH domains (DH1 and DH2 respectively) and association was determined by immunoblotting. (e) Rass DH domain is sufficient to inhibit activation. activation was assayed in growing COS-7 cells transfected with HA- together with vectors encoding for Ras1 and 2 or their respective DH domains (DH1 and DH2). (f) Ras DH but not Cdc domain is required to inhibit activation and Pak1 phosphorylation. activation and Pak1 phosporylation were analyzed in cells transfected with HA- and onco, plus Ras1, and its ΔCdc and ΔDH mutants. Bar charts show the fold increase in p-mlc2 relative to control cells (n=3). Error bars indicate SE. * P< 0.05; *** P<0.001 by Student s t-test relative to the levels of -transfected cells Macmillan Publishers Limited. All rights reserved.

5 a! wt wt N17 QL! α! α! Dock1 wt wt N17 QL! α! α! PI! IP α FLAG! PI! IP α FLAG! TL! α! α! TL! α! α! b Ost + AU5- N17! c + AU5- N17! α Ost! - - "1 ΔCdc ΔDH! α! α! α! TL! PI! IP α AU5! α Ost! α! TL! PI! IP α AU5! α! α! α! d c α HA-DH-PH! α! α HA! α GST! Figure S5 (a) Bona fide GEFs associate preferentially with dominant negative. COS-7 cells were transfected with onco or Dock10DHR2 as shown. Where indicated, Flag-epitoped wt, N17 or QL was included. GEF- interaction were analyzed in anti-flag immunoprecipitates. (b) Rass outcompete Ost for binding to. COS-7 cells were transfected with AU5- N17 plus Ost in addition to Ras s. was anti-au5 immunoprecipitated and associated proteins were revealed by immunoblotting. (c) Ras1 DH but not Cdc domain is required to inhibit / association. Cells were transfected with AU5- N17 and onco, plus Ras1 or its Cdc and DH deletion mutants. was anti-au5 immunoprecipitated and associated proteins were revealed by immunoblotting. (d) Purified Ras1 DH-PH domain displaces from binding to in vitro. was bound in vitro to purified GST- N17 and incubated by increasing concentrations of a purified peptide spanning Ras1 DH-PH domains. Proteins associated to were determined by immunoblotting Macmillan Publishers Limited. All rights reserved.

6 Figure S6 Ras GEFs do not affect Rac1-dependent cellular events. (a) Ras GEFs prevent NIH3T3 soft agar colony formation induced by onco-. Bars represent the number of colonies, mean +/- s.d of 5 independent experiments, expressed as a percentage relative to those induced by onco alone (ns p > 0.05; ** p < 0.01; (*** p < by Student s t-test). (b) Ras GEFs do not affect NIH3T3 transformation by onco-vav. NIH3T3 cells were transfected with oncovav plus the indicated cdnas. Bars represent the number of transformed foci, mean +/- s.d of 5 independent experiments, expressed as a percentage relative to those induced by oncovav alone (ns p > 0.05, Student s t-test). (c) Ras GEFs do not affect NIH3T3 lamellipodia formation induced by Vav. Representative confocal micrographs of NIH3T3 cells transfected with oncovav, plus the indicated constructs. Cells were immunostained with the appropriate primary antibodies to monitor the expression of Ras1/2 forms (blue). Cells were then stained with Phalloidin (red) to mark filamentous actin. Symbol on the lower, right hand corner shows the confocal plane displayed. Bar: 10 μm Macmillan Publishers Limited. All rights reserved.

7 a b filopodia per cell! filopodia per cell! c A375 WM266.4 WM1361 SKMEL28! C #7 #1 C #7 #1 C #7 #1 C #7 #1 d fold increase in p-mlc 2.5! 1.5! 0.5! α pmlc! α tubulin! α! ** ** C C! Dock10DHR si! α α tubulin! fold increase in p-mlc2 e A375P WM266.4 WM136 si2 #7! si2 #7! si2 #7! C C si DN! C C si DN! C C si DN! α pmlc 7! 6! 5! A375P WM266.4 WM1361 SKMEL28 ** ** * * C C si DN- si2 #7 pe ** ** * * α tubulin! α α! f si A375P WM226.4 g fold increase in p-pak- * * * * Invasion index C C si DN C C si DN si2 #7 si2 #7 Figure S7 Ras2 regulates invasion and lung colonization by melanoma cells. (a) Ras GEFs impair Dock10-induced filopodia formation in NIH3T3 cells. Graph shows the number of filopodia per cell upon transfection with the indicated constructs. Bars represent the average +/- SE of at least cells. (b) Ras2 levels modulate filopodia-formation in cells. Graph shows the number of filopodia per cell upon transfection with the indicated constructs after 24 h starvation. Cells transfected with either Ras2 sirna or shrna were co-transfected with pe (5% of the total DNA transfected) to detect transfected cells. Bars represent the media +/- SE of at least cells. (a and b) Only similar sized, nonconfluent cells with the same levels of phalloidin staining were analyzed. Filopodia counting registers the number of filopodia per cell in multiple confocal planes. Significance in (a) was calculated relative to Dock10 DHR2-transfected cells, except when indicated otherwise; in (b) it was calculated relative to pe-transfected cells (*** p <0.001; Student s t-test). (c) Down-regulation of endogenous Ras2 levels in the indicated melanoma cell lines by two Ras2 sirnas (#7, 10). (d) Depletion of Ras2 results in increased phosphorylation of MLC2 in a - dependent fashion. WM266.4 cells were transfected as indicated with sirna #7 against Ras2 plus vector (c) or increasing concentrations of a sirna against. The levels of the indicated proteins were revealed by immunoblotting. Bar charts show the fold increase in p-mlc2 relative to control cells (n=3). Error bars indicate SE. ** p<0.01 by Student s t test. (e) Depletion of Ras2 results in increased phosphorylation of MLC2 in a -dependent fashion. Top panel: as in (b) but down-regulation was achieved both by sirna interference and by the use of N17 dominant negative mutant (DN). Chart: pmlc2 levels in the indicated melanoma cell lines. Error bars indicate SE. p values: * p< 0.05, ** p < 0.01 by Student s t test relative to the respective levels in control cells transfected with sirna #7 against Ras2. (f) Pak-1 phosphorylation in response to Ras2 knock-down is dependent on. Analysed in the indicated cell lines transfected with sirna #7 against Ras2. down-regulation was achieved both by sirna interference (si) and by the use of N17 dominant negative mutant (DN). Bar charts show the fold increase in p-mlc2 relative to control cells (n=3). Error bars indicate SE. p values: * p< 0.05, by Student s t test relative to control cells transfected with sirna #7. (g) Inter-cellular comparison of invasive indexes among the melanoma cell lines used. Absolute invasion index into collagen-i matrix of A375M2, SKMEL28, WM1361 and WM266.4 cells (n=3, error bars +SE). Student s t-test was used to generate p values: *p< Macmillan Publishers Limited. All rights reserved.

8 Panel 1 of Fig 1a Panel 3 of Fig 1a Blots from Figure 1 Panel 1 of Fig 1b Panel 3 of Fig 1b Panel 1 of Fig 1c 2 Flag Panel 1 of Fig 1d pmlc2 1 Panel 2 of Fig 1d 2 Panel 3 of Fig 1d pmlc2 RhoA Rac1 Blots from Figure 2 Rac1 Panel 4 of Fig 1d RhoA Panel 1 of Fig 2a Panel 2 of Fig 2a Panel 1 of Fig 2b 10 1 ppak1 Blots from Figure 3 Panel 1 of Fig 1a (Jurkat) 1 Panel 4 of Fig 1a (Jurkat) 2 IgG 1 Panel 5 of Fig 1a (Jurkat) 2 IgG Panel 8 of Fig 1a (Jurkat) Panel 1 of Fig 1a (Mm Brain hom.) 1 Panel 4 of Fig 1a (Mm Brain hom.) 1/2 1 Panel 5 of Fig 1a (Mm Brain hom.) 1/2 Panel 8 of Fig 1a (Mm Brain hom.) Panel 1 of Fig 3h Panel 2 of Fig 3h Panel 3 of Fig 3h Panel 4 of Fig 3h 1 Panel 1 of Fig 5d Blots from Figure 6 1 Panel 3 of Fig 5d Figure S8 Uncropped, full scans of the films showing the key experiments displayed in the main figures Macmillan Publishers Limited. All rights reserved.

9 Supplementary Movies Movie S1 Time-lapse video-microscopy using a 10x magnification objective of A375P cells plated on top of collagen 48 h post Mock transfection. Movie S2 Time-lapse video-microscopy using a 10x magnification objective of A375P cells seeded on top of a collagen matrix 48h post-transfection with Ras2 OT7 sirna. Movies S3 Time-lapse video-microscopy using a 10x magnification objective of empty vector A375M2 stable cell line plated on top of collagen. Movies S4, S5, S6 Time-lapse video-microscopy using a 10x magnification objective of Ras2WT A375M2 stable cell line (Movie 4), RasΔCdc (Movie 5) and RasΔDH (Movie 6) plated on top of collagen. Compare to Movies S Macmillan Publishers Limited. All rights reserved.