Actin-rich protrusions and nonlocalized GTPase activation in Merlin-deficient schwannomas.

Schwannomas lack both alleles for the tumor suppressor Merlin, a cytoskeleton-membrane linker. Previous results showed increased cell spreading of schwannoma cells, but little is known about the underlying mechanisms. Electron microscopy reveals that schwannoma cells not only show more lamellipodia/ruffles but also multiple filopodia. We show that Cdc42, important in filopodia formation, is activated. Both Rac1 and Cdc42 are found all around the cell periphery and in colocalization with their effector phospho-p21 activated kinase in human schwannoma cells. We therefore claim that Rac1 and Cdc42 are activated in a nonlocalized manner, which explains the disperse distribution of lamellipodia/ruffles and filopodia. Using live cell imaging, we further demonstrate continuous remodeling of the many actin-rich protrusions in schwannoma cells. The underlying cytoskeleton of these structures is thin and extensively branched. The actin-related protein 2/3 complex, a major regulator of actin branching, is enriched in the many lamellipodia and ruffles of human primary schwannoma cells. We suggest that the Merlin deficiency in human primary schwannoma cells leads to a random, nonlocalized activation of Rac1 and Cdc42, inducing many actin-rich protrusion zones, not only at the leading edge but also all around the cell periphery. Their nondirectional occurrence together with the continuous and highly dynamic actin remodeling results in the dedifferentiation of these tumor cells.

Reduced apoptosis rates in human schwannomas.

Schwannomas, tumors originating from Schwann cells, represent a frequent neurological tumor and can occur both in a genetic disorder called neurofibromatosis type 2 (NF2) and sporadically. In both cases the genetic background is identical as all schwannomas are caused by biallelic mutations in the tumor suppressor gene NF2 coding for merlin. Mutations in this gene have also been found to be responsible for 50% to 60% of spontaneous and 100% of the NF2 associated meningiomas. The NF2 gene product, merlin, links transmembrane proteins to the cytoskeleton and is involved in intracellular signaling processes. It has previously been shown that reexpression of wild-type merlin in primary human schwannoma cells leads to an increase in the number of apoptotic cells. Here, we report in vivo and in vitro evidence that the basal apoptosis rate of primary human schwannoma cells is reduced in comparison to that of normal Schwann cells, supporting the idea that in this benign tumor type, apoptosis has a role in tumorigenesis.

Pathological adhesion of primary human schwannoma cells is dependent on altered expression of integrins.

Mutations in the tumor suppressor gene coding for merlin cause Neurofibromatosis type 2 (NF2), all spontaneous schwannomas, and a majority of meningiomas. Merlin links transmembrane proteins to the cytoskeleton. Accordingly, primary human schwannoma cells lacking merlin show an increased number of lamellipodia and filopodia as well as increased cell spreading. We show enhanced adhesion in primary human schwannoma cells and present evidence that this is dependent on the integrin chains alpha6beta1 and alpha6beta4. We further demonstrate that the integrin chains beta1 and beta4 are upregulated in schwannomas using different complementary methods, and report higher expression of these integrins per schwannoma cell by fluorescence assisted cell sorting (FACS). Finally we report clustering of the integrin chains alpha6, beta1, and beta4 on schwannoma cells. Our findings fit well into recent data on the role of merlin in signaling cascades connected to integrins and help explain pathological ensheathment of extracellular matrix or pseudomesaxon formation which is a hallmark of schwannoma histopathology.

Upregulation of the Rac1/JNK signaling pathway in primary human schwannoma cells.

Schwann cells lacking the tumor-suppressor-protein merlin tend in man to build benign tumors (schwannoma). We observed that characteristic features of these cells which are relevant to tumorigenicity resemble those described in cells with high Rac activity. Moreover this small GTPase also phosphorylates merlin via PAK activation. We hypothesized that merlin deficiency might cause an activation of Rac and its dependent signaling pathways, in particular the pro-tumorigenic JNK pathway. We show an enhanced activation of Rac1 in primary human schwannoma cells, find both Rac and its effector PAK at the membrane where they colocalize, and describe increased levels of phosphorylated JNK in the nucleus of these cells. Further we describe regulation at post-transcriptional level with upregulated protein, but not mRNA levels for Rac1, and JNK1/2. We conclude that merlin regulates Rac activation, and suggest that this is important for human schwannoma cell dedifferentiation.