Development of an All-Hydrocarbon Stapled Peptide Targeting BRK1 in Triple-Negative Breast Cancer

Abstract

Wiskott-Aldrich syndrome protein family (WASF) members are key regulators of actin cytoskeleton dynamics at the leading edge of the cell membrane. WASF3 has been demonstrated to directly promote cancer invasion and metastasis in triple-negative breast cancer. WASF3 is incorporated into a heteropentameric protein complex with BRK1, CYFIP1/2, NCKAP1/1L, and ABI1/2/3 termed the WASF Regulatory Complex (WRC) that links upstream signaling pathways to Arp2/3-mediated actin nucleation. Disruption of the complex inhibits actin remodeling and presents a novel approach to targeting cancer invasion and metastasis. Here we report the development of a first-generation all-hydrocarbon stapled BRK1 mimetic peptide, BASH-2, designed to inhibit BRK1 binding within the WRC to disrupt proper WRC assembly and function. BASH-2 was found to permeate cells, bind to WASF3 and ABI2, and inhibit cancer cell migration and invasion in a dose-dependent manner. BASH-2 may present a novel approach to targeting WASF3-promoted invasion and metastasis in triple-negative breast cancer.

Keywords: BRK1; Stapled peptide; WASF Regulatory Complex; WASF3; triple-negative breast cancer.

Figure 1.. Design of BRK1-derived stapled peptides.

(A) Representation of the WASF3 WRC showing ABI2 (purple), CYFIP1 (orange), and NCKAP1 (blue) in surface view and BRK1 (red), WASF3 (green) in ribbon view (derived from PDB ID: 3P8C). The inset shows the N-terminal portion of BRK1 from which BASH-1 and BASH-2 were derived as it interacts with an N-terminal helix of WASF3. (B) In silico alanine scanning using Bristol University Docking Engine (BUDE) AlaScan highlights predicted ‘hotspot’ residues of the BRK1 sequence (residues 19-41) that are thought to contribute to binding the WRC. (C) Sequence of BRK1 residues 19-41 and the stapled sequences incorporating synthetic olefinic residues (*) at i, i+4 positions to introduce the hydrocarbon staple within the sequence. Lysine substitutions (K, bold) were also added at positions not predicted to contribute to the binding interface to promote a net positive charge.

Figure 2.. BRK1-derived stapled peptides permeate cells.

(A) Fluorescence microscopy images of MDA-MB-231 cells treated with fluorescein-labeled peptides for 4 hours (n=3) demonstrate cytosolic peptide localization at this time point. (B) Flow cytometry analysis of MDA-MB-231 cells treated with either 1 μM fluorescein-labeled BASH peptides or 2 μM fluorescein-labeled WAHMIS-2 for 4 hours demonstrates a shift in fluorescence intensity after peptide treatments. Each plot indicates 10,000 counts/treatment.

Figure 3.. BRK1-derived stapled peptides reduce cell migration in a dose-dependent manner.

(A) Representative images of migration assays (n=3) performed with treatments of BASH-1, BASH-2, WAHMIS-2 or a DMSO control for 18 h in MDA-MB-231 cells. The wound area is indicated by red lines determined using the wound healing size tool in ImageJ. Both BRK1-derived peptides downregulated wound healing as compared to the DMSO control. (B) Quantification of cell migration as measured by % wound closure across three independent wound healing assays showing a significant reduction in cell migration for WAHMIS-2, BASH-1, and BASH-2 by approximately 45-60%. BASH-2 led to a slightly more pronounced effect relative to BASH-1. Wound area was calculated using the wound healing size tool for ImageJ. Error bars represent standard deviation. ***p<0.001. (C) Representative images of migration assays (n=3) performed with MDA-MB-231 cells treated with a dose range of 0.1 μM to 1 μM BASH-2 or a scramble control for 18 h. BASH-2 demonstrated a dose-dependent reduction in migration. (D) Quantification of cell migration as determined by % wound closure across three independent wound healing assays showing a significant reduction in migration for BASH-2 at 1 μM. The scramble control had no effect on migration relative to the DMSO control. Error bars represent standard deviation. ***p<0.001.

Figure 4.. BASH-2 inhibits cell invasion in a dose-dependent manner.

(A) Representative images indicating Matrigel invasion assays (n=3) of cells treated with a concentration range of BASH-2 (0.1-1 μM). Controls include a scramble control (1 μM), WAHMIS-2 (2 μM) or DMSO. Invading cells were fixed with paraformaldehyde and subsequently stained with crystal violet for microscopy and quantification. (B) Quantification of invading cells across 3 independent experiments indicates a significant, dose-dependent reduction in invasion by over 90% by BASH-2. Further, it was found to have more pronounced effects relative to our previously developed inhibitor, WAHMIS-2, at the highest concentration tested. Error bars represent standard error. **p <0.01, ***p<0.001.

Figure 5.. BASH-2 interacts with WASF regulatory complex proteins ABI2 and WASF3.

(A) Representative (n=3) western blots demonstrate that WAHMIS-2 and BASH-2 can interact with ABI2 as assessed via streptavidin pulldown of biotinylated peptides incubated with MDA-MB-231 cell lysates. Input represents total protein levels of ABI2 in untreated lysate fractions. (B) Densitometric quantification of three independent pulldown assays that BASH-2 and control peptide WAHMIS-2 both interact with ABI2 relative to the DMSO control. Weak, statistically insignificant interactions were observed for the BASH-2 scramble control. Error bars represent standard deviation. ***p<0.001. (C) Representative (n=3) western blot measuring WASF3 interactions after streptavidin pulldown of biotinylated peptides incubated with MDA-MB-231 cell lysates. BASH-2 was found to interact with WASF3, while its scramble control was not. In addition, WAHMIS-2 was also shown not to interact with WASF3 since it derived from the WASF3 sequence and is therefore not predicted to interact with this protein. Input represents total protein levels of ABI2 in untreated lysate fractions. (D) Densitometric quantification of three independent pulldown assays indicates interactions between BASH-2 and WASF3 relative to the DMSO control. No binding was observed for WAHMIS-2 or the BASH-2 scramble control. Error bars represent standard deviation. ***p<0.001.

Figure 6.. BASH-2 increases formation of actin stress fibers and reduces localization of actin to the leading edge of lamellipodia.

Representative fluorescence images of MDA-MB-231 cells treated with indicated biotinylated peptides for 4 hours, followed by fixation and staining with phalloidin-iFluor 594 (n=3). Cells treated with BASH-2 and WAHMIS-2 indicate increased formation of actin stress fibers (indicated with white arrows) and reduced accumulation of actin at the leading edge. Stress fibers were not observed in cells treated with the BASH-2 scramble or DMSO control.