Abstract
An in vivo RNAi screen identified nonmuscle myosin IIa (MYH9) as a tumor suppressor in SCC.
Major finding: An in vivo RNAi screen identified nonmuscle myosin IIa (MYH9) as a tumor suppressor in SCC.
Concept: MYH9 loss is common in human SCC and reduces the stability and nuclear localization of p53.
Impact: In vivo RNAi screens can facilitate the validation of low-penetrance mutations as driving events.
Genomic studies have identified many low-penetrance mutations in squamous cell carcinoma (SCC) with unknown functional relevance. Schramek and colleagues sought to identify SCC driver mutations through an in vivo RNA interference (RNAi) screen in which epithelial cell–specific lentiviral short hairpin RNA delivery was performed in utero. Knockdown of the nonmuscle myosin IIa heavy chain gene, Myh9, accelerated SCC formation, and its loss triggered SCC when accompanied by TGF-β pathway inactivation or Hras mutation. Myh9 knockdown induced metastatic skin SCC or head and neck SCC (HNSCC) in Tgfbr2 conditional knockout mice with median latencies of 3 to 7 months and led to the development of skin SCC after 1 year in Tgfbr2–wild-type mice. Consistent with the role of myosin IIa in regulating cell motility, myosin IIa deficiency increased cell migration and invasion in vitro. Interestingly, Myh9 knockdown did not exacerbate disease in animals expressing an epithelial-specific oncogenic Trp53 mutation, suggesting that myosin IIa may influence p53 or act in a common pathway. Indeed, myosin IIa deficiency impaired DNA damage–induced p53 activation and target gene upregulation in keratinocytes. Further, p53 activation depended upon myosin IIa ATPase activity, and myosin IIa was required for both p53 stability and nuclear retention. The authors found that myosin IIa expression was weak or absent in 24% of human skin SCCs and 31% of HNSCCs, and low expression of MYH9 correlated with poor survival in patients with HNSCC, suggesting that myosin IIa may also be a tumor suppressor in humans. MYH9 was mutated in approximately 5% of HNSCCs sequenced as part of The Cancer Genome Atlas, with mutations clustering in the ATPase domain. Moreover, expression of several of these cancer-associated mutations blocked p53 activation after DNA damage in vitro. Collectively, these data provide support for an unexpected tumor-suppressive role for myosin IIa in SCC that is associated with regulation of p53 nuclear accumulation.