Mechanotransduction refers to the transformation of physical forces into chemical signals. It generally involves stretch-sensitive channels or conformational change of cytoskeleton-associated proteins. Mechanotransduction is crucial for the physiology of several organs and for cell migration. The extent to which mechanical inputs contribute to development, and how they do this, remains poorly defined. Here we show that a mechanotransduction pathway operates between the body-wall muscles of Caenorhabditis elegans and the epidermis. This pathway involves, in addition to a Rac GTPase, three signalling proteins found at the hemidesmosome: p21-activated kinase (PAK-1), the adaptor GIT-1 and its partner PIX-1. The phosphorylation of intermediate filaments is one output of this pathway. Tension exerted by adjacent muscles or externally exerted mechanical pressure maintains GIT-1 at hemidesmosomes and stimulates PAK-1 activity through PIX-1 and Rac. This pathway promotes the maturation of a hemidesmosome into a junction that can resist mechanical stress and contributes to coordinating the morphogenesis of epidermal and muscle tissues. Our findings suggest that the C. elegans hemidesmosome is not only an attachment structure, but also a mechanosensor that responds to tension by triggering signalling processes. We suggest that similar pathways could promote epithelial morphogenesis or wound healing in other organisms in which epithelial cells adhere to tension-generating contractile cells.

Myotubularin is a ubiquitously expressed phosphatase that acts on phosphatidylinositol 3-monophosphate [PI(3)P], a lipid implicated in intracellular vesicle trafficking and autophagy. It is encoded by the MTM1 gene, which is mutated in X-linked myotubular myopathy (XLMTM), a muscular disorder characterized by generalized hypotonia and muscle weakness at birth leading to early death of most affected males. The disease was proposed to result from an arrest in myogenesis, as the skeletal muscle from patients contains hypotrophic fibers with centrally located nuclei that resemble fetal myotubes. To understand the physiopathological mechanism of XLMTM, we have generated mice lacking myotubularin by homologous recombination. These mice are viable, but their lifespan is severely reduced. They develop a generalized and progressive myopathy starting at around 4 weeks of age, with amyotrophy and accumulation of central nuclei in skeletal muscle fibers leading to death at 6 -14 weeks. Contrary to expectations, we show that muscle differentiation in knockout mice occurs normally. We provide evidence that fibers with centralized myonuclei originate mainly from a structural maintenance defect affecting myotubularin-deficient muscle rather than a regenerative process. In addition, we demonstrate, through a conditional gene-targeting approach, that skeletal muscle is the primary target of murine XLMTM pathology. These mutant mice represent animal models for the human disease and will be a valuable tool for understanding the physiological role of myotubularin.