Furthermore, we provide a novel mechanism for the regulation of the actin cytoskeleton during migration: LPXN-mediated phosphorylation of CaD by the extracellular-signal regulated kinase 1/2 (ERK). RESULTS Reduced adhesion and cell size of PCa cells after LPXN knockdown To investigate the influence of LPXN expression on the adhesive characteristics of PCa cells, we performed a cell adhesion assay. decreased expression of TGF-beta-activated kinase 1 (TAK1) after LPXN knockdown in PC-3 PCa cells. Subsequent analyses of the downstream kinases revealed the extracellular signal-regulated kinase (ERK) as an interaction partner of LPXN that facilitates CaD phosphorylation during LPXN-mediated PCa cell migration. In conclusion, we demonstrate that LPXN directly influences cytoskeletal dynamics via interaction with the actin-binding protein CaD and Rabbit Polyclonal to ZAR1 regulates CaD phosphorylation by recruiting ERK to highly dynamic structures within PCa cells. gene encodes five different CALD1 transcripts, resulting in two major isoforms: a high-molecular-mass isoform (h-CaD) that is expressed in smooth muscle cells and a low-molecular-mass isoform (l-CaD) expressed in non-muscle cells. The regulation of CaD is important for proper cell function because decreased expression of l-CaD has been found in many cancer cell types [12-15]. In the present study, we identify the actin-binding protein CaD as a new interaction partner of LPXN, thereby linking LPXN directly to the actin cytoskeleton for the first time. Furthermore, we provide a novel mechanism for the regulation of the actin cytoskeleton during migration: LPXN-mediated phosphorylation of CaD by the extracellular-signal regulated kinase 1/2 (ERK). RESULTS Reduced adhesion and cell size of PCa cells after LPXN knockdown To investigate the influence of LPXN expression on the adhesive characteristics of PCa cells, we performed a cell adhesion assay. After downregulation of LPXN expression in PC-3 and DU 145 cells using a specific siRNA, cells were plated on glass slides coated with fibronectin (FN), rat tail collagen (Col), bovine serum albumin (BSA) or gelatin (Gel). Adhered cells were fixed after 2 hours of incubation, and the cytoskeleton was visualized using FITC-conjugated phalloidin. Cell numbers and cell size were analyzed using confocal fluorescence microscopy. We observed that cells with LPXN knockdown showed reduced adhesion on all substrates in comparison to control cells (Figure ?(Figure1A).1A). The strongest effect of LPXN knockdown was observed for adhesion on FN-coated slides. In addition, the highest difference in cell size between LPXN knockdown and control transfected (siLuc) cells was observed on FN-coated and BSA-coated slides (Figure ?(Figure1B).1B). Thus, loss of LPXN expression seems to BI-409306 reduce the capability to adhere to the ECM in PCa cells. Open in a separate window Figure 1 LPXN knockdown decreases adhesion and cell sizeTo analyze adhesion, PC-3 and DU 145 cells transfected with siRNA against LPXN (siLPXN) or luciferase (siLuc = control) were plated on glass culture slides that were either uncoated (?) or coated with bovine serum albumin (BSA), collagen (Col), fibronectin (FN) or gelatin (Gel). Cells were fixed 2 hours after plating; the cytoskeleton was visualized using FITC-conjugated phalloidin (green), and nuclei were stained with DAPI (blue). Cell number (A) and cell size (B) were determined by confocal microscopy. After 2 hours (C) of adhesion, siLPXN-transfected cells showed a reduced surface area compared to control-transfected cells, whereas 24 hours (D) later, they were not distinguishable from each other. As summarized in Figure ?Figure1C,1C, PC-3 cells showed a significantly reduced surface area after LPXN knockdown compared with control transfected cells. After 2 hours, control cells were already spread on the substratum and had a strong contact to the fibronectin matrix, whereas cells with LPXN knockdown remained rounded BI-409306 and showed no cell protrusions. As a control and to study the effect of LPXN knockdown on long-term adhesion, cells transfected with siLPXN or siLuc (control) were allowed to adhere for 24 hours. During this time course, both cell populations could completely adhere to the substratum and showed no difference in their morphology (Figure ?(Figure1D),1D), pointing to a function of LPXN in early adhesion dynamics. LPXN interacts with the actin-binding protein CaD To identify proteins that could facilitate the cytoskeletal changes mediated by LPXN, we performed a yeast two-hybrid screen using a human prostate cDNA library with full-length LPXN as bait. This resulted in two different clones encoding the human actin-binding protein caldesmon BI-409306 (CaD, proximity ligation assay (PLA) on PC-3 cells using specific LPXN and CaD antibodies, respectively. Interaction of the two proteins is indicated by the red dots (Figure ?(Figure3D).3D). Confocal fluorescence microscopic analysis of the PLA revealed that LPXN-CaD interaction was mainly localized to the sub-membranous compartments, whereas no interaction was detected at the protrusion zone of migrating cells or at stabilized actin structures and podosomes (Figure ?(Figure3D).3D). We observed little interaction of LPXN and CaD in non-migrating or quiescent PCa.