JCS010744 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. FGF-induced cell proliferation in NIH-3T3 and L cells requires Erk1/2 activation. (A) NIH3T3 cells transfected with either a control siRNA or with siNCAM1 were cultured for the indicated duration after transfection, followed by immunoblotting of the cell lysates with an anti-NCAM monoclonal antibody, an anti-FGFR1 scFv (see Materials and Methods) or an anti-tubulin monoclonal antibody (to verify equal loading). The transfection with siNCAM1 (as well as with siNCAM3, data not shown) caused a dramatic reduction in the level of NCAM, whereas the expression of FGFR1 was not affected. (B) NIH-3T3 cells transfected with either a control siRNA or with siNCAM1 were serum-starved and then stimulated for the indicated durations with FGF2 in the presence or PD98059 or DMSO. Cell proliferation was then measured as described in the Material and Methods. The transfection with siNCAM3 showed the same effect as siNCAM1 (not shown). (C) L cells transfected with either an empty vector, NCAM or ΔFN2-NCAM were serum-starved and stimulated as above, and cell proliferation was measured as described in the Material and Methods. In both NIH3T3 and L cells, FGF2-induced cell proliferation was abolished by PD98059.

• Supplemental Figure S2 -

  Fig. S2. Full-length and ΔFN2-NCAM show similar cellular localization and expression levels in L cells. (A) Lysates from L-mock, L-NCAM and L-ΔFN2 cells were subjected to immunoblotting with an anti-NCAM antibody and then with an anti-tubulin antibody to verify the equal loading of the lanes. Comparable levels of NCAM were present in L-NCAM and L-ΔFN2 cells. (B) L cells transfected with either an empty vector, NCAM or ΔFN2-NCAM were grown on coverslips, fixed and stained with an anti-NCAM monoclonal antibody. NCAM showed a very similar pattern of cellular localization, with a prominent surface expression, in L-NCAM and L-ΔFN2 cells. At higher cell density, both full-length and ΔFN2-NCAM were enriched at cell-cell contacts.

• Supplemental Figure S3 -

  Fig. S3. NCAM represses the activation of p52^SHC induced by FGF2 in L cells. L-mock, L-NCAM and L-ΔFN2 cells were stimulated with FGF-2 for 10 minutes and then subjected to immunoblotting with an antibody against phopsho-Shc, followed by immunoblotting for total Shc. FGF2 readily stimulated the phosphorylation of p52^SHC in L-mock cells. By contrast, FGF-induced phosphorylation of p52^SHC was abolished by the expression of NCAM, but not of ΔFN2.

• Supplemental Figure S4 -

  Fig. S4. NCAM induces neither cell proliferation nor protection from apoptosis. (A) Serum-starved L cells, transfected with either empty vector or with NCAM, were incubated for the indicated durations. In a separate set of experiments, L-mock cells were also stimulated daily with FGL. At each time point, cell proliferation was measured as described in the Materials and Methods. Neither membrane-associated NCAM nor the FGL peptide exhibited any proliferative activity. (B) Mock- or NCAM-transfected L cells were incubated in serum-free medium for the indicated durations, either in the absence or presence of the apoptosis-inducing agent staurosporine. Immunoblotting of cell lysates with an antibody against activated caspase-3 (Cell Signaling) was performed to monitor apoptotic cell death. Immunoblotting for vinculin (monoclonal antibody from Sigma) was performed to verify equal loading. NCAM expression did not show any effect on the apoptotic rate of L cells.

• Supplemental Figure S5 -

  Fig. S5. NCAM inhibits FGF-induced cell-matrix adhesion by binding to FGFR. L cells stably transfected with empty vector, full-length NCAM or ΔFN2 were stimulated for 10 minutes with FGF2 and then subjected to adhesion assays on collagen-IV-coated wells coated with 4 µg/cm² of collagen IV (BD Biosciences). After 2 hours, non-adherent cells were removed by washing with PBS, and adherent cells were fixed for 30 minutes with 2% formalin and stained with crystal violet. Bound dye was solubilized with 10% acetic acid and absorbance measured at 595 nm. Cell-free wells served as blanks. For each transfectant cell line, the adhesion of FGF-stimulated cells was calculated as the percentage of the adhesion of control, untreated cells. Whereas full-length NCAM abolished FGF2-induced cell-matrix adhesion, the inhibitory effect of ΔFN2-NCAM was markedly reduced, indicating that the NCAM-FGFR association is required for the repression of FGF-2 function. *P<0.005.

• Supplemental Figure S6 -

  Fig. S6. NCAM does not affect the levels of total or surface-exposed FGFR1. (A) Lysates from mock- or NCAM-transfected L cells were immunoblotted for FGFR1, using an anti-FGFR1 scFv (see Materials and Methods). Modulating the expression of NCAM did not alter the level of FGFR1. (B) Surface proteins of mock- or NCAM-transfected L cells were biotinylated using Sulfo-NHS-Biotin (Pierce Biotechnology). Cell lysates were then immunoprecipitated with anti-FGFR1 (C15, Santa Cruz) or control IgG, followed by immunoblotting with HRP-conjugated streptavidin (upper panel) and with the anti-FGFR1 scFv (lower panel). No significant differences were observed between L-mock and L-NCAM cells in the levels of either total or surface-exposed FGFR1. (C) A lysate from NIH-3T3 cells was immunoblotted for FGFR1, using the polyclonal antibody C15, confirming the specificity of the antibody and supporting the TIRF data shown in Fig. 1.

• Supplemental Figure S7 -

  Fig. S7. NCAM does not bind to FGF2. The Fc-tagged ectodomains of either FGFR1α (isoform IIIc; R&D Systems) or NCAM were immobilized on 96-well plates (0.3 pmol/well) previously coated with 0.67 pmol/well of goat anti-human Fc (Sigma). The Fc fragment alone (Calbiochem) was also immobilized as a negative control. Biotinylated FGF2 (kindly provided by G. Taraboletti, Bergamo, Italy) was then added to the wells and incubated for 3 hours at room temperature. The binding of biotinylated FGF2 to immobilized NCAM-Fc was also assayed in the presence of 10 mg/ml heparin. After washing, HRP-conjugated streptavidin was added and bound FGF2 was revealed by measuring the absorbance at 415 nm upon incubation with the ABST reagent (Sigma). FGF2 bound to immobilized FGFR1 with high efficiency, whereas no binding was observed to immobilized NCAM-Fc, either in the absence or presence of heparin. Thus, FGF2 does not bind to NCAM, supporting the notion that NCAM does not prevent the FGF2-FGFR interaction by sequestering FGF2.