Kinase Activity Of Pak2, An Effector Of Rac/CDC42 and Its Interaction With β-PIX Is Required For Murine Hematopoietic Stem Cell Shape, F-Actin Formation, Directional Migration In Vitro and For HSPC Homing To Bone Marrow In Vivo

Hematopoietic stem and progenitor cell (HSPC) migration, marrow homing and engraftment are key physiological processes regulating hematopoiesis post transplantation. These processes are the result of the orchestrated actions of multiple extracellular stimuli, which regulate actin remodeling, cell polarity, chemotaxis and cell-cell interactions. In HSPC, the Rho GTPases Rac and CDC42 act as molecular switches that integrate extracellular stimuli in a spatially regulated manner to control cell migration and mediate homing to marrow and mobilization as well as cell survival/ proliferation pathways to mediate engraftment(Gu et al., Science 2003; Cancelas et al., Nature Medicine 2005; Wang et al., Blood 2006). Using an inhibitory peptide against Group A p21 activated kinases (Pak1-3), key effectors of Rac/ CDC42 and individual Pak1 & 2 genetic knock-out mice, we recently demonstrated that Pak kinases, specifically Pak2, are important for HSPC homing and engraftment (Dorrance et al., Blood 2013). Pak2 is a multi-domain protein that contains a C-terminal kinase domain and multiple N-terminal protein-interaction domains. Among these is a non-classical SH3-binding site for the guanine-nucleotide-exchange factor β-PIX, which was shown to be critical for both activation of Rac1 and its localization to and induction of membrane ruffles (Klooster et al., Journal of Cell Biology 2006). In this study we further explored the role of these domains of Pak2 in key HSPC functions, including homing to bone marrow in vivo. We employed a multi-cistronic retrovirus vector that simultaneously deleted floxed endogenous Pak2 gene sequences and rescued with either wild type (WT), a kinase dead (KD) mutant (K278A, defective in auto/ trans phosphorylation) or a Δβ-PIX mutant, (P185/R186A, that cannot bind to β-PIX). As previously demonstrated deletion of Pak2 (Pak2 ^Δ/Δ) was associated with abnormal SDF-1 stimulated cell protrusions containing F-actin (as demonstrated by confocal and electron microscopy) and these HSPC displayed decreased directional migration (Euclidean distance in Pak2^Δ/Δ vs. Pak2^WT/WT: 39.6µm ±9.6 vs. 96.6µm ±21.6; P<0.05). This phenotype of abnormal cell protrusions and decreased directional migration was rescued by expressing Pak2-WT (Pak2^WT/WT vs. Pak2-WT: 96.6µm ±21.6 vs. 74.0µm ±18.7; P: not significant) but not by expressing Pak2-KD (Pak2^WT/WT vs. Pak2-KD: 96.6µm ±21.6 vs. 33.6µm ±6.3; P<0.05) demonstrating the requirement of Pak2 kinase activity in SDF1-induced cell polarization and directed cell migration. Interestingly, we found abnormal F-actin clustering associated with defective polarization (by confocal microscopy) and decreased velocity of cell migration in time-lapsed video microscopy when Pak2-deletion was rescued with Pak2-Δβ-PIX (velocity of migration Pak2^WT/WT vs. Pak2-Δβ-PIX, 0.32µm/minute ±0.02 vs. 0.13µm/minute ±0.02; P<0.001), indicating the requirement of β-PIX exchange factor interaction with Pak2 in directed migration. To test whether these in vitro phenotypes were associated with changes in homing efficiency to bone marrow, we performed in vivo homing assays of rescued HSPC. Transduced, GFP-sorted Lin^-Sca1⁺Kit⁺ cells of each genotype were injected into lethally-irradiated C57BL/6 recipient mice (N= 12-29 /genotype). Twelve hours post-transplantation the number of EGFP⁺ cells in the bone marrow was determined and percent homing is calculated. Compared to Pak2 ^WT/WT, Pak2^Δ/Δ HSPC displayed reduced homing (99.26%± 4.9 vs. 53.4% ± 4.2; P< 0.0001). The homing defect was rescued by Pak2-WT (Pak2^WT/WT vs. Pak2-WT rescue: 99.26%± 4.9 vs. 86% ± 8.5; P: not significant). However neither Pak2-KD nor Pak2-Δβ-PIX rescued in vivo homing: 99.26% ±4.9 vs. 38.9% ±3.7 vs. 33.0%± 6.0; P< 0.0001 each mutant vs.Pak2^WT/WT) proving the necessity of kinase activity and interaction with β-PIX for bone marrow homing. Taken together we show that both Pak2-kinase activity and its interaction with β-PIX exchange factor are required for coordinated HSPC F-actin formation and cell polarization, directed cell migration in vitro and homing to bone marrow in vivo. These data directly link the in vitro effects of Pak2 kinase with in vivo bonemarrow homing.