IQGAP1 Interaction with RHO Family Proteins Revisited: KINETIC AND EQUILIBRIUM EVIDENCE FOR MULTIPLE DISTINCT BINDING SITES

Abstract

IQ motif-containing GTPase activating protein 1 (IQGAP1) plays a central role in the physical assembly of relevant signaling networks that are responsible for various cellular processes, including cell adhesion, polarity, and transmigration. The RHO family proteins CDC42 and RAC1 have been shown to mainly interact with the GAP-related domain (GRD) of IQGAP1. However, the role of its RASGAP C-terminal (RGCT) and C-terminal domains in the interactions with RHO proteins has remained obscure. Here, we demonstrate that IQGAP1 interactions with RHO proteins underlie a multiple-step binding mechanism: (i) a high affinity, GTP-dependent binding of RGCT to the switch regions of CDC42 or RAC1 and (ii) a very low affinity binding of GRD and a C terminus adjacent to the switch regions. These data were confirmed by phosphomimetic mutation of serine 1443 to glutamate within RGCT, which led to a significant reduction of IQGAP1 affinity for CDC42 and RAC1, clearly disclosing the critical role of RGCT for these interactions. Unlike CDC42, an extremely low affinity was determined for the RAC1-GRD interaction, suggesting that the molecular nature of IQGAP1 interaction with CDC42 partially differs from that of RAC1. Our study provides new insights into the interaction characteristics of IQGAP1 with RHO family proteins and highlights the complementary importance of kinetic and equilibrium analyses. We propose that the ability of IQGAP1 to interact with RHO proteins is based on a multiple-step binding process, which is a prerequisite for the dynamic functions of IQGAP1 as a scaffolding protein and a critical mechanism in temporal regulation and integration of IQGAP1-mediated cellular responses.

Keywords: CDC42; GTPase activating protein (GAP); IQGAP1; Ras homolog gene family, member A (RHOA); Ras-related C3 botulinum toxin substrate 1 (Rac1); Rho (Rho GTPase); binding affinity; fluorescence anisotropy; protein-protein interaction; signal transduction.

FIGURE 1.

Schematic representation of domain organization, various constructs, and proteins of IQGAP1. Shown is IQGAP1 domain organization along with the PKCϵ phosphorylation sites Ser-1441 and Ser-1443, constructs, and proteins relevant to this study. Coomassie Brilliant Blue (CBB)-stained SDS-PAGE (12.5%) of purified IQGAP1 proteins used in this study is shown. CHD, calponin homology domain; CC, coiled-coil repeat region; WW, tryptophan-containing proline-rich motif-binding region; IQ, four isoleucine/glutamine-containing motifs.

FIGURE 2.

GRD1-CT but not GRD selectively associates only with mGppNHp-bound, active RAC1 and CDC42. A, chemical structure of mGppNHp, a fluorescently labeled, non-hydrolyzable GTP analog, used in this study. B, the stopped-flow device. The stopped-flow device consists of two motorized, thermostated syringes, a mixing chamber, and a fluorescence detector. Two different protein solutions indicated in brackets were rapidly mixed and transferred to a fluorescence detection cell within <4 ms. mGppNHp- or mGDP-bound RHO proteins were used in this study as the fluorescent reporter groups. C and D, association of GRD1-CT with active, mGppNHp-bound CDC42 and RAC1. Kinetics of association were followed by rapidly mixing 2 μm GRD1-CT, GRD1, or GRD2 with 0.2 μm mGppNHp- or mGDP-bound RAC1 (C) or CDC42 (D). The obtained data are the average of four to six independent measurements. The k_obs values obtained for the association of GRD1-CT with mGppNHp-bound CDC42 and RAC1 were 1.68 and 1.80 s⁻¹, respectively. No change in fluorescence was observed for GRD1-CT- and mGDP-bound RAC1 or CDC42 (black), GRD1 or GRD2 with mGppNHp-bound RAC1 or CDC42 (red), and GRD1-CT with RHOA-mGppNHp (blue). E and F, overlapping binding sites of CDC42 and RAC1 for GRD1-CT. Association of RAC1-mGppNHp (0.2 μm) with GRD1-CT (2 μm) was blocked in the presence of excess amount of non-fluorescent CDC42-GppNHp (10 μm) (E). Association of CDC42-mGppNHp (0.2 μm) with GRD1-CT (2 μm) was blocked in the presence of excess amount of non-fluorescent RAC1-GppNHp (10 μm) (F).

FIGURE 3.

IQGAP1 GRD binds CDC42 but not RAC1 in a nucleotide-independent manner. A, fluorescence polarization assay. Fluorescence polarization signal of a fast tumbling fluorescent molecule, e.g. RAC1-mGppNHp in its unbound state, increased if a larger protein, e.g. IQGAP1, bound to it and formed a slow tumbling complex. B, fluorescence polarization experiments were conducted by titrating mGppNHp-bound, active forms of CDC42, RAC1, and RHOA (1 μm, respectively) with increasing concentrations of GRD1-CT (0–20 μm) or GRD2 (0–120 μm), respectively. C, evaluated dissociation constant (K_d) shown as bars illustrates a significant difference in the binding properties of these two IQGAP1 proteins measured in B. D, fluorescence polarization experiments were conducted under the same conditions as in B using mGDP-bound, inactive forms of CDC42, RAC1, and RHOA. E, calculated K_d values shown as bars clearly indicated interaction of GRD1-CT and GRD2 with CDC42-mGDP but not with RAC1 and RHOA. Data are expressed as the mean ± S.D. All experiments were performed in duplicate.

FIGURE 4.

IQGAP1 variants significantly differ in their interaction properties with CDC42 and RAC1. A and B, association of different GRD1-RGCT variants and CT (2 μm, respectively) with mGppNHp-RAC1/CDC42 (0.2 μm) was measured. A, association of GRD1-RGCT^WT (black), GRD1-RGCT^S1441E (green), and GRD1-RGCT^S1443E (red), but not with CT (blue), with RAC1-mGppNHp. B, the k_obs values, shown as bars, comparatively illustrate association rates of GRD1-CT, GRD1-RGCT^WT, and GRD1-RGCT^S1441E with mGppNHp-bound forms of CDC42 and RAC1, which is significantly reduced in the case of GRD1-RGCT^S1443E and completely absent in the case of GRD1, GRD2, and CT under these experimental conditions. C–E, fluorescence polarization experiments were conducted to measure the interaction of mGppNHp-bound forms of RAC1 (C) and CDC42 (D) with increasing concentrations of GRD1-RGCT variants (WT, S1441E, and S1443E; 0–36 μm, respectively). E, calculated K_d values, shown as bars, reveal a significant decrease in the binding affinities of GRD1-RGCT^S1443E as compared with GRD1-RGCT^WT and GRD1-RGCT^S1441E. Data are expressed as the mean ± S.D. All k_obs measurements were performed in triplicate, and fluorescence polarization experiments were conducted in duplicate.

FIGURE 5.

Phosphomimetic mutation of RGCT and deletion of CT affect the IQGAP1 association with CDC42 and RAC1. A and B, individual rate constants for the GRD1-CT interaction with RAC1 and CDC42 are represented in A and B. Left panel, association of mGppNHp-bound RAC1 or CDC42 (0.2 μm, respectively) with increasing concentrations of GRD1-CT (2 to 12 μm). Middle panel, evaluated association rate constant (k_on) from the plot of the k_obs values, obtained from the exponential fits to the association data in the left panel against the corresponding concentrations of the GRD1-CT. Right panel, evaluated dissociation rate constant (k_off) measured by the displacement of the GRD1-CT (2 μm) from its complex with mGppNHp-bound RAC1 or CDC42 (0.2 μm, respectively) in the presence of excess amounts of non-fluorescent RAC1-GppNHp (10 μm). Other kinetics are given in supplemental Figs. S1 and S2. C, calculated individual rate constants for the interaction of the IQGAP1 variants with RAC1 and CDC42, respectively, plotted as bar charts. Dissociation constants (K_d) were obtained from the ratio k_off/k_on. Data are expressed as the mean ± S.D. All k_obs measurements experiments were accomplished in triplicate.

FIGURE 6.

Q61L mutation greatly increased CDC42 affinity for IQGAP1. A, individual rate constants for the GRD1-CT interaction with CDC42^Q61L. Association of tGppNHp-bound CDC42^Q61L (0.2 μm) with increasing concentrations of GRD1-CT (2 to 10 μm) is shown in the left panel. The middle panel shows evaluated association rate constant (k_on), and the right panel shows the evaluated dissociation rate constant (k_off) measured by the displacement of the GRD1-CT (2 μm) from its complex with tGppNHp-bound CDC42 (0.2 μm) in the presence of excess amounts of non-fluorescent CDC42^Q61L-GppNHp (10 μm). B, calculated individual rate constants for the interaction of the GRD1-CT with CDC42^Q61L, plotted as bar charts. A dissociation constant (K_d) of 0.05 μm was obtained from the ratio k_off/k_on. C–F, fluorescence polarization were performed by titrating CDC42^Q61L-tGppNHp (1 μm) with increasing concentrations of GRD1-CT (0–2.5 μm) (C) and GRD2 (0–7.5 μm) (D) or CDC42^Q61L-mGppNHp with increasing concentrations of GRD1-CT (0–3 μm) (E) and GRD2 (0–7.5 μm) (F).