Rac1 selectively binds a specific lamellipodin isoform via a noncanonical helical interface

Abstract

Lamellipodin (Lpd) is a multifunctional adapter protein that regulates cell migration and adhesion by recruiting Ena/VASP proteins to the leading edge and modulating actin polymerization. The interaction of Lpd and Rho family or Ras family GTPases is crucial for regulating actin dynamics. Contrary to previous assumptions that the main Lpd isoform interacts with Rac1, here we show that strong and specific binding to Rac1 is instead mediated by the short isoform Lpds. This interaction is dependent on Rac1's GTPase activity and a short insertion (cs2) within the coiled-coil region unique to the Lpds isoform. Structural modeling and mutagenesis analyses further reveal that Lpds engages Rac1 through a noncanonical, single-helix binding mode distinct from the canonical helical-pair configuration. Our results uncover a novel isoform-dependent GTPase:effector binding mode for Rac1-driven actin dynamics, with implications for therapeutic targeting in Rac1-associated cancer progression.

Keywords: GTPase effector; RAPH; Rac1; cell migration; isoform; lamellipodin; splicing variant.

Figure 1

The Lpds isoform of Lpd has strong binding affinity with activated Rac1 specifically.A, schematic representation of Lpd isoforms. A talin-binding site (green) is located at the N-terminal end of Lpd and is conserved across all isoforms. The CS is depicted in orange, the CC is in red, the RA (Ras-associating) domain is in yellow, and the PH (pleckstrin homology) domain is in cyan. B, the co-IP assay of GFP-tagged full-length Lpd and Lpds and HA-tagged constitutively active Rac1 mutant (Rac1 Q61L). Cell lysate and IP samples are detected by Western blot using anti-HA antibody and anti-GFP antibody. CC, coiled-coil region; CS, insertion of cc; IN, autoinhibitory segment; Lpd, lamellipodin; PH, pleckstrin homology domain; RA, Ras-association domain; TBS, talin-binding site.

Figure 2

Rac1:Lpds interaction requires the GTPase activity of Rac1 and the CS insertion of Lpds.A, co-IP assay of GFP-tagged Lpds1-PH with HA-tagged Rac1 Q61L, Rap1 G12V, M-Ras Q71L, M-Ras WT, H-Ras G12V, and R-Ras G12V. B, co-IP assay of HA-tagged WT Rac1 and Rac1 Q61L with GFP-tagged Lpds FL, 1-PH, csRAPH, cs-cc, and Lpd ccRAPH. The schematic diagrams of the constructs are shown above, with the same domain color codes as in Figure 1A. C, co-IP assay of HA-tagged WT Rac1, constitutively active Rac1 mutant (Rac1 Q61L), and inactivated Rac1 mutant (Rac1 Q61L/T17N) with GFP-tagged Lpds csRAPH. D, pulldown assay of GST-tagged WT Rac1, Rac1 Q61L, and Rac1 Q61L/G12V with His-tagged Lpds csRAPH. CS, coiled segment; Lpd, lamellipodin; PH, pleckstrin homology.

Figure 3

Lpds binds Rac1 through a non–β-sheet interface distinct from the RIAM:Rap1 interaction.A, structural alignment of the AlphaFold-predicted, Lpds alone model with the crystal structure of the RIAM:Rap1 complex. Lpds domains are colored as follows: CS insertion (orange), CC (red), RA domain (yellow), and PH domain (cyan). The RIAM:Rap1 complex is shown in gray. The zoomed-in panel (bottom right) highlights key salt bridges in the RIAM:Rap1 interface (Asp33:Lys213 and Tyr40:Lys193). The corresponding residues in Lpds (Lys335 and Lys355), selected for mutagenesis, are labeled. B, sequence alignment of the switch I region from Rac1, Rap1, H-Ras, and M-Ras. Conserved residues are highlighted in green. Rac1 residues Ile33 and Tyr40, corresponding to salt bridge–forming residues in Rap1, are marked in red. C, sequence alignment of Lpd, Lpds, and RIAM, with lysine residues involved in Rap1:RIAM interaction highlighted in the boxes. Lpds numbering is based on the sequence including the CS insertion. D, co-IP assay of HA-tagged Rac1 Q61L with GFP-tagged Lpds csRAPH, csRAPH K335A, and Lpd ccRAPH. E, co-IP assay of HA-tagged Rac1 Q61L with various truncations of GFP-tagged Lpds csRAPH and csRAPH K355A. CC, coiled coil; CS, coiled segment; Lpd, lamellipodin; PH, pleckstrin homology; RA, Ras-associating; RIAM, Rap1-GTP–interacting adaptor molecule.

Figure 4

Structure prediction and co-IP assay suggest that the cs2 insertion stabilize the CC region via hydrophobic interaction.A, amino acid sequence of the two Lpd splicing insertions, cs1 and cs2, are shown. The cs2 insertion residues are numbered 1′ to 25’. Two hydrophobic leucine clusters, LLL (3L) and LL (2L), in the cs2 insertion are highlighted in cyan. Leu4′ and Ile15′ of cs2 are colored in pink. B, predicted interface between the cs2 insertion (in pink) and CC region (in red). Hydrophobic residues contributing to the cs2:CC interaction are shown in surface representation. Residue numbers of CC are based on the Lpd main isoform to avoid confusion. C, co-IP assay of HA-tagged Rac1 Q61L and GFP-tagged Lpds csRAPH, Lpds csRAPH 3L3A, Lpds csRAPH 2L2A, and Lpds csRAPH 3L3A/2L2A. D, structure of Rac1 (in gray) in complex with the helical-pair for Prk1 (in pink). E, structure Lpds:Rac1 complex predicted by AlphaFold3. Upper: Lpds is colored as indicated in Figure 1A. Rac1 is colored in gray. Lower: Interface between Lpds-cs2-CC and Rac1. Switch I and switch II regions of Rac1 are colored in dark gray. Residues forming the hydrophobic interaction interface are shown in surface representation. CC, coiled coil; CS, coiled segment; Lpd, lamellipodin.