Analysis of the function of the Nf2 tumor suppressor protein, Merlin

Abstract

The Neurofibromatosis type 2 tumor suppressor gene (NF2) is mutated in inherited and sporadically occurring central nervous system tumors. The NF2 encoded protein, merlin, shares close sequence similarity in its amino-terminal domain to members of the band 4.1 family of membrane-cytoskeletal linkers. Similarities between merlin and this family suggest a role for merlin in regulating cytoskeletal function. Thus, NF2 may be a novel type of tumor suppressor gene that mediates its tumor suppressor function through interactions with the actin cytoskeleton. However, the molecular and cellular functions of this tumor suppressor gene were largely unknown when the work described here began. Mutational analysis of Nf2 in flies has lead to the identification of a dominant-negative allele, which harbors mutations in the amino-terminal domain of the protein. The work presented here demonstrates that expression of a murine analog of this amino-terminal mutant of Nf2 (termed, Nf2BBA) leads to complete transformation of NIH3T3 fibroblasts in culture. Cells that express Nf2BBA display disruptions of the actin cytoskeleton, lack of contact inhibition of growth, and anchorage-independent growth. In addition, Nf2-deficient mouse embryo fibroblasts (MEFs) exhibited similar contact inhibition and cell-matrix adhesion defects to Nf2BBA expressing cells. Nf2BBA cells continue to cycle under normal growth inhibitory conditions, such as serum withdrawal, and exhibit high levels of the cell cycle regulator, cyclin D1. Elevated levels of cyclin D1 are necessary for cellular transformation following Nf2BBA expression. Nevertheless, the exact mechanism by which Nf2BBA results in cellular transformation remains elusive. Recently published studies have revealed that merlin may regulate members of the RhoGTPase

(cont.) family, as absence of Nf2 expression in fibroblasts leads to many phenotypes reminiscent of overactive Rac, such as increased membrane ruffling and increased activity of the c-jun N- terminal kinase (JNK). Our work has extended to the analysis of the role of merlin in the regulation of the Rac pathway. Using rat schwannoma cells and N2-deficient MEFs, we have demonstrated that merlin exerts its inhibitory effects downstream of Rac, through a direct interaction with the p21 activated kinase, Pak. We demonstrate that in the absence of merlin, Pak is active and hyperphosphorylated, and, conversely, when merlin is overexpressed, Pak activity is diminished. The N-terminal half of merlin binds to the functionally conserved Rac/Cdc42 interaction binding (CRIB) domain of Pak. Several models for merlin regulation of Pak activity will be discussed. Finally, the identification of Pak as a kinase that is misregulated in the absence of NF2 may lead to possible avenues for therapeutic intervention.