Investigation of the Interaction between Cdc42 and Its Effector TOCA1: HANDOVER OF Cdc42 TO THE ACTIN REGULATOR N-WASP IS FACILITATED BY DIFFERENTIAL BINDING AFFINITIES

Abstract

Transducer of Cdc42-dependent actin assembly protein 1 (TOCA1) is an effector of the Rho family small G protein Cdc42. It contains a membrane-deforming F-BAR domain as well as a Src homology 3 (SH3) domain and a G protein-binding homology region 1 (HR1) domain. TOCA1 binding to Cdc42 leads to actin rearrangements, which are thought to be involved in processes such as endocytosis, filopodia formation, and cell migration. We have solved the structure of the HR1 domain of TOCA1, providing the first structural data for this protein. We have found that the TOCA1 HR1, like the closely related CIP4 HR1, has interesting structural features that are not observed in other HR1 domains. We have also investigated the binding of the TOCA HR1 domain to Cdc42 and the potential ternary complex between Cdc42 and the G protein-binding regions of TOCA1 and a member of the Wiskott-Aldrich syndrome protein family, N-WASP. TOCA1 binds Cdc42 with micromolar affinity, in contrast to the nanomolar affinity of the N-WASP G protein-binding region for Cdc42. NMR experiments show that the Cdc42-binding domain from N-WASP is able to displace TOCA1 HR1 from Cdc42, whereas the N-WASP domain but not the TOCA1 HR1 domain inhibits actin polymerization. This suggests that TOCA1 binding to Cdc42 is an early step in the Cdc42-dependent pathways that govern actin dynamics, and the differential binding affinities of the effectors facilitate a handover from TOCA1 to N-WASP, which can then drive recruitment of the actin-modifying machinery.

Keywords: BAR domain; CDC42; CIP4; FBP17; TOCA1; WASP; actin; endocytosis; nuclear magnetic resonance (NMR); protein-protein interaction.

FIGURE 1.

The TOCA1 HR1-Cdc42 interaction is low affinity. A, curves derived from direct binding assays in which the indicated concentrations of Cdc42Δ7Q61L·[³H]GTP were incubated with 30 nm GST-PAK or HR1-His₆ in SPAs. The SPA signal was corrected by subtraction of control data with no GST-PAK or HR1-His₆. The data were fitted to a binding isotherm to give an apparent K_d and are expressed as a percentage of the maximum signal; B and C, competition SPA experiments were carried out with the indicated concentrations of ACK GBD (B) or HR1 domain (C) titrated into 30 nm GST-ACK and either 30 nm Cdc42Δ7Q61L·[³H]GTP or full-length Cdc42Q61L·[³H]GTP. The K_d values derived for the ACK GBD with Cdc42Δ7 and full-length Cdc42 were 0.032 ± 0.01 and 0.011 ± 0.01 μm, respectively. The K_d values derived for the TOCA1 HR1 with Cdc42Δ7 and full-length Cdc42 were 6.05 ± 1.96 and 5.39 ± 1.69 μm, respectively.

FIGURE 2.

The Cdc42-HR1 interaction is of low affinity in the context of full-length protein and in TOCA1 paralogues. A, diagram illustrating the TOCA1 constructs assayed for Cdc42 binding. Domain boundaries are derived from secondary structure predictions; B, binding curves derived from direct binding assays, in which the indicated concentrations of Cdc42Δ7Q61L·[³H]GTP were incubated with 30 nm GST-ACK or His-tagged TOCA1 constructs, as indicated, in SPAs. The SPA signal was corrected by subtraction of control data with no fusion protein. The data were fitted to a binding isotherm to give an apparent K_d and are expressed as a percentage of the maximum signal. C–E, representative examples of competition SPA experiments carried out with the indicated concentrations of the TOCA1 HR1-SH3 construct titrated into 30 nm GST-ACK and 30 nm Cdc42Δ7Q61L·[³H]GTP (C) or HR1^CIP4 (D) or HR1^FBP17 (E) titrated into 30 nm GST-ACK and 30 nm Cdc42FLQ61L·[³H]GTP.

FIGURE 3.

The structure of the TOCA1 HR1 domain. A, the backbone trace of the 35 lowest energy structures of the HR1 domain overlaid with the structure closest to the mean is shown alongside a schematic representation of the structure closest to the mean. Flexible regions at the N and C termini (residues 330–333 and 421–426) are omitted for clarity. B, a sequence alignment of the HR1 domains from TOCA1, CIP4, and PRK1. The secondary structure was deduced using Stride (64), based on the Ramachandran angles, and is indicated as follows: gray, turn; yellow, α-helix; blue, 3₁₀ helix; white, coil. C, a close-up of the N-terminal region of TOCA1 HR1, indicating some of the NOEs defining its position with respect to the two α-helices. Dotted lines, NOE restraints. D, a close-up of the interhelix loop region showing some of the contacts between the loop and helix 1. NOEs are indicated with dotted lines. All structural figures were generated using PyMOL.

FIGURE 4.

Mapping the binding surface of Cdc42 onto the TOCA1 HR1 domain. A, the ¹⁵N HSQC of 200 μm TOCA1 HR1 domain is shown in the free form (black) and in the presence of a 4-fold molar excess of Cdc42Δ7Q61L·GMPPNP (red). Expansions of two regions are shown with peak assignments, showing backbone amides in fast or intermediate exchange. B, CSPs were calculated as described under “Experimental Procedures” and are shown for backbone and side chain NH groups. The mean CSP is marked with a red line. Residues that disappeared in the presence of Cdc42 were assigned a CSP of 0.2 but were excluded when calculating the mean CSP and are indicated with open bars. Those that were not traceable due to spectral overlap were assigned a CSP of zero and are marked with an asterisk below the bar. Residues with affected side chain CSPs derived from ¹³C HSQCs are marked with green asterisks above the bars. Secondary structure elements are shown below the graph. C, a schematic representation of the HR1 domain. Residues with significantly affected backbone or side chain chemical shifts when Cdc42 bound and that are buried are colored dark blue, whereas those that are solvent-accessible are colored yellow. Residues with significantly affected backbone and side chain groups that are solvent-accessible are colored red. A close-up of the binding region is shown, with affected side chain heavy atoms shown as sticks. D, the G protein-binding region is marked in red onto structures of the HR1 domains as indicated.

FIGURE 5.

Mapping the binding surface of the HR1 domain onto Cdc42. A, the ¹⁵N HSQC of Cdc42Δ7Q61L·GMPPNP is shown in its free form (black) and in the presence of excess TOCA1 HR1 domain (1:2.2, red). Expansions of two regions are shown, with most peaks in fast or intermediate exchange. B, CSPs are shown for backbone NH groups. The red line indicates the mean CSP, plus one S.D. Residues that disappeared in the presence of Cdc42 were assigned a CSP of 0.1 and are indicated with open bars. Those that were not traceable due to overlap are marked with an asterisk. Residues with disappeared peaks in ¹³C HSQC experiments are marked on the chart with green asterisks. Secondary structure elements are indicated below the graph. C, the residues with significantly affected backbone and side chain groups are highlighted on an NMR structure of free Cdc42Δ7Q61L·GMPPNP; those that are buried are colored dark blue, whereas those that are solvent-accessible are colored red. Residues with either side chain or backbone groups affected are colored blue if buried and yellow if solvent-accessible. Residues without information from shift mapping are colored gray. The flexible switch regions are circled.

FIGURE 6.

Model of Cdc42·HR1 complex. A, a representative model of the Cdc42·HR1 complex from the cluster closest to the lowest energy model produced using HADDOCK (67). Residues of Cdc42 that are affected in the presence of the HR1 domain but are not in close proximity to it are colored in red and labeled. B, structure of Rac1 in complex with the HR1b domain of PRK1 (46) (PDB code 2RMK). C, sequence alignment of RhoA, Cdc42 and Rac1. Contact residues of RhoA and Rac1 to PRK1 HR1a and HR1b, respectively, are colored cyan. Residues of Cdc42 that disappear or show chemical shift changes in the presence of TOCA1 are colored cyan if also identified as contacts in RhoA and Rac1 and yellow if they are not. Residues equivalent to Rac1 and RhoA contact sites but that are invisible in free Cdc42 are gray. D, regions of interest of the Cdc42·HR1 domain model. The four lowest energy structures in the chosen HADDOCK cluster are shown overlaid, with the residues of interest shown as sticks and labeled. Cdc42 is shown in cyan, and TOCA1 is shown in purple.

FIGURE 7.

The N-WASP GBD displaces the TOCA1 HR1 domain. A, the model of the Cdc42·TOCA1 HR1 domain complex overlaid with the Cdc42-WASP structure. Cdc42 is shown in green, and TOCA1 is shown in purple. The core CRIB region of WASP is shown in red, whereas its basic region is shown in orange and the C-terminal region required for maximal affinity is shown in cyan. A semitransparent surface representation of Cdc42 and WASP is shown overlaid with the schematic. B, competition SPA experiments carried out with indicated concentrations of the N-WASP GBD construct titrated into 30 nm GST-ACK or GST-WASP GBD and 30 nm Cdc42Δ7Q61L·[³H]GTP. C, Selected regions of the ¹⁵N HSQC of 145 μm Cdc42Δ7Q61L·GMPPNP with the indicated ratios of the TOCA1 HR1 domain, the N-WASP GBD, or both, showing that the TOCA HR1 domain does not displace the N-WASP GBD. D, selected regions of the ¹⁵N HSQC of 600 μm TOCA1 HR1 domain in complex with Cdc42 in the absence and presence of the N-WASP GBD, showing displacement of Cdc42 from the HR1 domain by N-WASP.

FIGURE 8.

Actin polymerization downstream of Cdc42·N-WASP·TOCA1 is inhibited by excess N-WASP GBD but not by the TOCA1 HR1 domain. Fluorescence curves show actin polymerization in the presence of increasing concentrations of N-WASP GBD or TOCA1 HR1 domain as indicated. Maximal rates of actin polymerization derived from the linear region of the curves are represented in bar charts below. Error bars, S.E.

FIGURE 9.

A simplified model of the early stages of Cdc42·N-WASP·TOCA1-dependent actin polymerization. Step 1, TOCA1 is recruited to the membrane via its F-BAR domain and/or Cdc42 interactions. F-BAR oligomerization is expected to occur following membrane binding, but a single monomer is shown for clarity. Step 2, N-WASP exists in an inactive, folded conformation. The TOCA1 SH3 domain interacts with N-WASP, causing an activatory allosteric effect. The HR1^TOCA1-Cdc42 and SH3^TOCA1-N-WASP interactions position Cdc42 and N-WASP for binding. Step 3, electrostatic interactions between Cdc42 and the basic region upstream of the CRIB initiate Cdc42·N-WASP binding. Step 4, the core CRIB binds with high affinity while the region C-terminal to the CRIB displaces the TOCA1 HR1 domain and increases the affinity of the N-WASP-Cdc42 interaction further. The VCA domain is released for downstream interactions, and actin polymerization proceeds. WH1, WASP homology 1 domain; PP, proline-rich region; VCA, verprolin homology, cofilin homology, acidic region.