PMC

FAK is required for 3-D cell invasion through Matrigel. (A) FAK^−/− v-Src cells do not invade through Matrigel as do FAK-expressing cells. Random cell invasion was assessed in the presence of BSA. Values are means of triplicates ± SD from three independent experiments. (B) Crystal violet–stained cells that invaded through the Matrigel can be distinguished from the 8-μm membrane pores. (C) Equivalent wound closure motility of FAK^−/− v-Src and DA2 v-Src cells on Matrigel-coated slides in the presence of serum. Cell migration was assessed as in the legend to .

FAK activates Rac and serum stimulation promotes GFP-FAK localization to lamellipodia. (A) Rac activation was visualized in the indicated cells by pull-down assays using GST-PAK (67–150) followed by Rac blotting. Total Rac expression is shown in whole cell lysates. (B) FAK^−/− v-Src cells were infected with Ad-TA or the indicated Ad-FAK constructs, and Rac activation was evaluated by GST-PAK binding. (C) FAK^−/− v-Src cells were transfected with GFP-FAK, starved, and then stimulated with 10% FBS. Time-lapse confocal video-microscopy at 1-min intervals was obtained over 3 h (Video 1 available at http://www.jcb.org/cgi/content/full/jcb.200212114/DC1). Panels show a 9-min time series where GFP-FAK is stably localized to focal contacts and transiently recruited to lamellipodia extensions (arrows).

Chemotaxis motility and serum-stimulated redistribution of v-Src to lamellipodia in FAK-containing cells. (A) ILK indirect IF of cells plated on FN in the presence of serum shows reduced numbers of focal contacts in FAK^−/− v-Src and DA2 v-Src cells. Indirect v-Src IF shows that FAK promotes v-Src association with focal contacts and membrane ruffles. Bar, 20 μm. (B) v-Src enhances the wound closure activity of FAK^−/− cells. Cell migration was assessed by 1-mm grid comparisons of four image sets; representative images are shown. (C) Serum-stimulated chemotaxis assays show that v-Src–transformed FAK^−/−, DA2, and FAK^+/+ cells exhibit similar properties of reduced migration compared with DA2 cells. Random motility was assessed in the presence of BSA. Values are means ± SD of triplicates from two independent experiments.

FAK expression promotes elevated v-Src and serum-stimulated JNK activity. (A) Whole cell lysates from the indicated serum-starved cells were blotted with phospho-specific antibodies to tyrosine (pTyr), Tyr-705 of STAT3 (pSTAT3), Ser-473 of Akt (pAkt), Thr-202 and Tyr-204 of ERK2 (pERK), and Thr-183 and Tyr-185 of JNK (pJNK). Equal protein expression was verified by reprobing blots with antibodies to STAT3, Akt, ERK2, or JNK1. Numbers represent M _r in kD. (B) ERK2 IP/IVK assays were performed with lysates from the indicated starved or 20% FBS-stimulated cells. Values are means ± SD from two independent experiments that were normalized for protein levels in IPs and represent fold differences in ERK2 activity from starved FAK^−/− cells. (Inset) Representative image of MBP phosphorylation. (C) JNK IP/IVK assays were performed as described for ERK2 above. (Inset) Representative image of GST-ATF2 phosphorylation.

v-Src transformation rescues spreading and haptotaxis defects of FAK ^−/− cells. (A) Whole cell lysates from the indicated cells were blotted with antibodies to FAK, v-Src, or β-actin. (B) The indicated cells were plated on FN-coated (10 μg/ml) coverslips in the absence of serum for 6 h. Phalloidin staining of actin shows that v-Src transformation promotes the conversion of FAK^−/− and DA2 cells to a fusiform morphology. Bar, 20 μm. (C) FN-stimulated haptotaxis assays show that expression of FAK or v-Src can rescue FAK^−/− motility defects. Random motility was assessed in the presence of BSA. Values are means ± SD of triplicates from three independent experiments. (D) p190 RhoGAP IPs were made from the indicated cells plated onto FN, analyzed by anti-pTyr blotting, and show enhanced pTyr levels stimulated by FAK or v-Src expression. Blots were reprobed by anti-p190RhoGAP blotting.

Ad-JNK rescues FAK ^−/− v-Src invasion defects and promotes increased MMP-9 expression. (A) Polyclonal IPs were used to isolate total JNK1 from mock- and Ad-JNK1 WT-infected FAK^−/− v-Src cells or from mock- and JNK1 DN- (Thr-180 and Tyr-182 mutated to Ala and Phe) infected DA2 v-Src cells. JNK1 expression was evaluated by blotting (top), and JNK activity was determined by blotting phospho-specific antibodies to Thr-183 and Tyr-185 of JNK (pJNK, middle). Equivalent total lysate protein was verified by β-actin blotting. (B) c-Jun IPs were prepared from either mock or Ad-JNK WT-infected FAK^−/− v-Src cells after serum stimulation (20% FBS for 30 min). Activating c-Jun phosphorylation was determined by blotting with phospho-specific antibodies to Ser-63 of c-Jun. (C) Matrigel invasion assays were preformed with the indicated mock- or Ad-infected cells. Values are means ± SD of triplicates. (D) Semiquantitative RT-PCR was performed with RNA isolated from mock or Ad-JNK WT-infected FAK^−/− v-Src cells after serum stimulation (20% FBS for 3 h) using primers to MMP-9 and to β-actin. JNK overexpression promoted a twofold increase in MMP-9 mRNA. (E) Gelatin zymography of conditioned media from mock- or Ad-JNK WT-infected FAK^−/− v-Src cells.

Ad-p130Cas rescues FAK ^−/− v-Src invasion defects; multiple connections to Dock180. (A) FAK^−/− v-Src cells were infected with Ad-TA, Ad-FAK, Ad-p130Cas, or Ad-Cas ΔSH3. Total p130Cas was isolated by IP and sequentially analyzed by Cas and pTyr blotting. (B) Matrigel invasion assays were performed with FAK^−/− v-Src cells infected with Ad-TA, Ad-FAK, or the indicated Ad-Cas constructs. Values are means ± SD of triplicates from two independent experiments. (C) JNK IP/IVK assays were performed with lysates of FAK^−/− v-Src cells infected with Ad-TA, Ad-FAK, or the indicated Ad-Cas constructs and with lysates from DA2 v-Src cells. Shown is GST–ATF-2 phosphorylation and values reflect fold changes in JNK activity from two independent experiments normalized to Ad-TA. (D) Lysates (1 mg) were prepared from FAK^−/− v-Src cells, incubated with GST or the indicated GST-SH3 domain fusion proteins, and collected by binding to glutathione-agarose beads. The GST-associated Dock180 was visualized by blotting, and bead-associated GST fusion proteins were visualized by Coomassie blue staining. (E) Lysates from serum-stimulated FAK^−/− v-Src or DA2 v-Src cells were prepared, and IPs were performed with antibodies to Crk, p130Cas, Dock180, and FAK. The IP- associated proteins were resolved by SDS-PAGE and visualized by anti-pTyr blotting. The membrane was cut and the indicated regions reprobed with antibodies to Dock180, p130Cas, v-Src, or Crk. The p130Cas membrane region was sequentially reprobed by HA tag blotting to visualize HA-FAK expression in lysates from DA2 v-Src cells.

FAK Tyr-397 phosphorylation, kinase activity, and SH3-binding sites in the FAK COOH-terminal domain are required to promote JNK activation and cell invasion. (A) HA tag IPs from FAK^−/− v-Src cells infected with Ad-TA or Ad-TA in combination with the indicated Ad-FAK constructs were used to evaluate Ad-FAK activation by blotting with antibodies to HA, pTyr, and phosphorylated FAK at Tyr-397 (pY397). (B) Matrigel invasion assays were performed with FAK^−/− v-Src cells infected with Ad-TA or Ad-TA in combination with the indicated Ad-FAK constructs and with noninfected DA2 v-Src cells. Values are means ± SD of triplicates from two separate experiments. (C) Ad-FAK expressed in FAK^−/− v-Src cells is localized to the tips of invadopodia emerging from Matrigel and is phosphorylated at Tyr-397 (arrowheads) as detected by indirect IF. Only the 8 μm (scale bar) Matrigel-filled membrane pores are seen in assays with FAK Pro-null–infected cells. (D) JNK IP/IVK assays were performed in Ad-TA– or Ad-FAK–infected FAK^−/− v-Src cells. Values are means ± SD of duplicates from independent experiments. (E) Matrigel invasion assays were performed with DA2 v-Src cells in the presence of DMSO or the JNK inhibitor (SP600125) at the indicated concentration in the top chamber. Values are means ± SD from two independent experiments.

FAK and activated Q61L Rac synergize to promote JNK activation, increased MMP-9 expression, and an invasive cell phenotype. (A) Flag tag blotting was used to visualize Q61L Rac expression in the indicated Ad- infected cells. (B) Matrigel invasion assays were performed with the indicated Mock or Ad-Q61L Rac-infected cells. Values are means ± SD of triplicates from two independent experiments. (C) Gelatin zymography was performed with conditioned media from the indicated Mock or Ad-Q61L Rac-infected cells. Migration of pro and active forms of MMP-9 and MMP-2 are shown. (D) Conditioned media from FAK^−/− v-Src or DA2 v-Src cells was concentrated, and MMP-9 or MMP-2 secretion was evaluated by blotting. (E) Semiquantitative RT-PCR was performed using primers to MMP-9 and to β-actin and showed that MMP-9 mRNA levels were 6.5-fold higher in DA2 v-Src compared with FAK^−/− v-Src cells. (F) Matrigel invasion assays were performed with DA2 v-Src cells with the addition of recombinant TIMP-1 at the indicated concentration in the top chamber. Values are means ± SD of triplicates from two independent experiments. (G) JNK IP/IVK assays performed with lysates from FAK^−/− cells or Ad-Q61L Rac-infected FAK^−/−, FAK^−/− v-Src, or DA2 v-Src cells. Values are means ± SD of duplicates from two independent experiments. Flag tag blotting of whole cell lysates was used to verify equivalent Q61L Rac expression.

Models of differential FAK signaling connections at focal contacts or lamellipodia/invadopodia. (A) Integrin-associated FAK and Src activation promotes increased p190RhoGAP tyrosine phosphorylation or ERK2/MAP kinase activation. FAK–Src modulation of p21 Rho activity in FAK^−/− cells is associated with increased cell motility, focal contact turnover, and actin cytoskeletal rearrangements. (B) In either EGF-stimulated human adenocarcinoma cells () or serum-stimulated v-Src–transformed fibroblasts, FAK functions to coordinate a Src-p130Cas signaling complex localized to lamellipodia in two-dimensional cell culture or to invadopodia in 3-D invasion assays. Through direct Cas SH3-mediated and indirect Crk-mediated interactions with the Dock180 guanine-nucleotide exchange factor, the FAK–Src signaling complex promotes Rac and JNK activation with effects upon gene expression. Although PI 3-kinase is a direct target of both v-Src and Rac (; ), FAK^−/− v-Src cells exhibit defects in JNK but not Akt activation. Overexpression of JNK in FAK^−/− v-Src cells coupled with serum stimulation promoted c-Jun Ser-63 phosphorylation, MMP-9 expression, MMP-2 activation, and rescued the FAK^−/− v-Src cell invasion defects.