Fig 5. CDC42 and RAC1 interact with RIT1, and NS-associated RIT1 amino acid substitutions increase complex formation between RIT1 and RAC1/CDC42.

(A and C) HEK293T cells transfected with empty vector (EV) or expressing wild-type RIT1 (WT) or RIT1 p.G95A were either serum-deprived (0.1% serum) or kept under full serum (10% serum). Endogenous CDC42 and RAC1 were immunoprecipitated with an anti-CDC42 antibody ([CDC42 (#1)] (A) and an anti-RAC1 antibody (C), respectively (IP). Co-precipitated HA-RIT1 and HA-RIT in total cell lysates (TCL) was detected by an anti-HA antibody. Enrichment of CDC42 and RAC1 in the precipitates and amount of endogenous CDC42 and RAC1 in TCL was demonstrated with an anti-CDC42 antibody ([CDC42 (#2)] (A) and an anti-RAC1/2/3 antibody (C), respectively. As a control, immunoprecipitation with an anti-V5 antibody was carried out (IP ctrl). Data shown are representative of three independent experiments. (B and D) HEK293T cells were transfected with empty vector (EV) and RIT1 expression constructs as indicated and cultured under serum deprivation (0.1% serum). Cell lysates were processed as described in (A) and (C). Data shown for CDC42 and RAC1 are representative of seven (B) and four (D) independent experiments, respectively. Autoradiographic signals were quantified by scanning densitometry. The amount of co-precipitated HA-RIT1 was double-normalized relative to amounts of immunoprecipitated CDC42/RAC1 and HA-RIT1 in total cell lysates. To conserve the relative variance of the samples, values for RIT1 wildtype and RIT1 mutants were divided by the mean of the wildtype samples [79]. The graphs show the relative amount (in arbitrary units) of co-precipitated RIT1 protein variants. The mean of five (B) and four (D) independent experiments ± SD is given. One-way ANOVA between groups: P < 0.05 (B and D). Post hoc P values were calculated by t-tests and Bonferroni correction; *, P < 0.05; **, P < 0.01 (B and D).