NG2, novel proapoptotic receptor, opposes integrin a4 to mediate anoikis

Background: Disruption of cell–matrix interactions can lead to anoikis—apoptosis due to loss of matrix contacts. Altered fibronectin (FN) matrix or FN fragments as are found in periodontal disease induce anoikis of primary human fibroblasts by a novel signaling pathway. This is characterized by reduced focal adhesion kinase (FAK) phosphorylation and transcriptional downregulation of p53. However, the receptors involved are unknown. pFAK is regulated by nerve/glial antigen 2 (NG2) receptor signaling through PKCa, a point at which signals from integrins and proteoglycans may converge.

Objective: We tested the hypothesis that NG2 acts in concert with integrin a4b1 to regulate anoikis and that PKCa and FAK signaling are control points where these two receptor-mediated signaling pathways converge.

Methods: We used immunoblotting, immunofluorescence and flow cytometry to determine protein expression levels and a flow cytometry based apoptotic assay. Fibroblasts were transfected with NG2, integrin a4, and PKCa cDNAs or siRNA to examine the regulation of pFAK mediated by NG2 and integrin a4.

Results: Altered FN induced anoikis by upregulating NG2 and downregulating integrin a4. Suppressing NG2 expression or overexpressing a4 rescued cells from anoikis. NG2 overexpression alone induced apoptosis and, by reducing pFAK, increased anoikis induced by an altered matrix. NG2 overexpression or an altered matrix also suppressed PKCa expression, but overexpressing integrin a4 enhanced pFAK independently of PKCa. Cotransfection with NG2 cDNA and integrin a4 siRNA did not lower PKCa and pFAK levels more than transfection with either alone. PKCa was upstream of pFAK, as silencing PKCa decreased pFAK. PKCa overexpression reversed this behavior and rescued cells from anoikis.

Conclusion: NG2 is a novel proapoptotic receptor, and NG2 and integrin a4 oppositely regulate anoikis. NG2 and integrin a4 regulate pFAK by PKCa-dependent and -independent pathways, respectively. These mechanisms may underlie periodontal disease pathogenesis (Study supported by NIH grant R01 DE013725 to YLK).