The fly genome contains many highly conserved orthologues to human disease genes (Reiter et al, 2001; Bier, 2005), including neurological, cardiovascular, endocrine and metabolic disease-genes. Among these, nearly all components of the Dg–Dys complex, which is involved in muscular dystrophies, are present in flies (Deng and Ruohola-Baker, 2000; Greener and Roberts, 2000; Deng et al, 2003). We now show that Dys and Dg interact genetically and biochemically and are required in the same cell types in Drosophila. A fluorescence polarization assay revealed that the Dg–Dys binding interface is highly conserved in humans and Drosophila (Figure 2). Both proteins are required for oocyte cellular polarity and interact in this process (Figure 1). Futhermore, mutants of both Dg and Dys genes show symptoms observed in muscular dystrophy. Reduction of Dg and Dys proteins results in age-dependent mobility defects (Figure 3). Eliminating Dg and Dys specifically in mesoderm derived tissues reveals that these proteins are required for muscle maintenance in adult flies: age-dependent muscle degeneration was observed in mutant tissues (Figure 4). Dg–Dys complex is also required for neuron path-finding and has both cell autonomous and non-cell autonomous functions for this process (Figures 5 and 6). Further, we have now shown that in neuronal path-finding process Dg interacts with InR and an SH2/SH3-domain adapter molecule Nck/Dock (Figure 7).
Drosophila as a muscular dystrophy model
Animal models have been used efficiently in muscular dystrophy studies. Some of the models are naturally occurring mutations (mdx-mouse, muscular dystrophy dog, cat and hamster), others have been generated by gene targeting (Watchko et al, 2002). However, the regulation and the control of Dg–Dys complex are not understood, and no successful therapeutics exist yet for muscular dystrophies (however, systemic delivery-studies using adeno-associated viral vectors show promise (Gregorevic et al, 2004)). Studies in new model organisms with easy-to-manipulate genetics might reveal the mode of regulation of the complex by identifying key regulatory components through suppressor screens. In addition, careful functional analysis of the complex in different cell types in model organisms might result in a unifying theme that will reveal its molecular mechanism of function. Such recently developed models for muscular dystrophy exist in C. elegans and zebrafish (Gieseler et al, 2000; Parsons et al, 2002; Bassett and Currie, 2003). In C. elegans Dys mutant, the transporter snf-6 that normally participates in eliminating acetylcholine from the cholinergic synapses, is not properly localized, resulting in an increased acetylcholine concentration at the neuromuscular junction and muscle wasting (Kim et al, 2004). The function of Dys in neuromuscular junctions has also been recently addressed in Drosophila (van der Plas et al, 2006). These results bring up the possibility that muscular dystrophies in humans might also at least partly be attributed to the altered kinetics of acetylcholine transmission through neuromuscular junctions.
We have now shown that Drosophila melanogaster acts as a remarkably good model for age-dependent progression of muscular dystrophy. Dg and Dys reduction in Drosophila show age-dependent muscle degeneration and lack of climbing ability. It is tempting to speculate that the common denominator between different defects observed in Dg–Dys mutants in Drosophila and C. elegans is defective cellular polarity. The defects observed in C. elegans could be due to a defect in polarization of a cell, which will generate a neuromuscular junction that leads to miss-targeted snf-6. Similarly, we have shown that Drosophila Dg–Dys complex is required for cellular polarity in the oocyte. In addition, neural defects observed are plausibly due to polarity defects in the growing axon.
Dg–Dys complex in axon path-finding
Similar to neuronal defects observed in human muscular dystrophy patients, neuronal defects were also found in Drosophila Dg and Dys mutant brains. In vertebrate brains, Dg affects neuronal migration (Montanaro and Carbonetto, 2003; Qu and Smith, 2004) possibly through interaction of neurons with their glial guides. The neuronal migration and process outgrowth have been shown to require supportive input from glial cells and involve the formation of adhesion junctions along the length of the soma. Also, the outgrowth of the leading process involves rapid extension and contraction over the length of the glial fiber (Rivas and Hatten, 1995; Shaham, 2005). Disruption of the cytoskeletal organization within the neuron, either of actin filaments (Rivas and Hatten, 1995) or microtubule interactions (Vallee et al, 2000), has been shown to inhibit glial-mediated neuronal migration. The glial function in this process is less well studied.
Drosophila photoreceptor path-finding provides an excellent system for genetic dissection of neuronal outgrowth and target recognition (Dickson, 2002). During the formation of the nervous system, newly born neurons send out axons to find their targets. Each axon is led by a growth cone that responds to extracellular axon guidance cues and chooses between different extracellular substrates upon which to migrate. Recent work has also identified a variety of intracellular signaling pathways by which these cues induce cytoskeletal rearrangements (Guan et al, 1996; Rao, 2005), but the proteins connecting signals from cell surface receptors to actin cytoskeleton have not been clearly determined. Dg is a good candidate for linking receptor signaling to the remodeling of the actin cytoskeleton and thereby polarizing the growth cone. We have now shown that perturbation of Dg–Dys complex causes phenotypes that resemble Nck/Dock-Pak-Trio axon path-finding phenotypes (Figure 5) (Rao, 2005), suggesting that Dg may be one of the key players in Nck/Dock signaling pathway for axon guidance and target recognition in Drosophila.
Interestingly, Insulin receptor-tyrosine kinase (InR) mutants also show similar phenotypes to those of Nck/Dock signaling in photoreceptor axon path-finding and these two proteins show genetic and biochemical interactions (Song et al, 2003). These data have led to speculations of mammalian InR acting in conjunction with Nck/Dock pathway in learning, memory and eating behavior (Dickson, 2003; Song et al, 2003). Our data now add Dg–Dys complex to this pathway; similar to what is seen in the case of Dg and Dys photoreceptor mutants, InR mutants show no obvious defects in patterning of the photoreceptors. However, the guidance of photoreceptor cell axons from the retina to the brain is aberrant (Song et al, 2003; Figures 5 and 6). Furthermore, genetic and biochemical evidence suggests that InR function in axon guidance involves the Dock-Pak pathway rather than the PI3K-Akt/PKB pathway. Independently, biochemical interaction between Nck/Dock and Dg has been reported (Sotgia et al, 2001) supporting the hypothesis that InR, Dg and Nck/Dock interaction regulates Dg–Dys complex. Furthermore, we have now shown that Dg interacts genetically with InR and Dock in photoreceptor axon path-finding. Since Dys interacts with Dg but not with InR and Dock, it is tempting to speculate that Dg can selectively interact with either Dys or InR and Dock (Figure 7). One possibility is that the tyrosine kinase activity of InR could regulate the Dg–Dys interaction by tyrosine phosphorylation in the Dg–Dys binding interphase (Figure 2). This tyrosine phosphorylation could prohibit the Dg–Dys interaction and thereby result in rearrangements in the actin cytoskeleton. Alternatively, other components observed in Dg–Dys complex might be involved in this regulation (Zhan et al, 2005). However, it is also possible that potential polarity defects in the Dg mutant axons result in defective InR membrane localization. Interestingly, in another cell type, the Drosophila oocyte, InR, Dg and Dys also show similar phenotypes (Deng et al, 2003; LaFever and Drummond-Barbosa, 2005; Figure 1). In addition, insulin-like growth factors (IGF) and InR are important in maintaining muscle mass in vertebrates (Singleton and Feldman, 2001). Further connection of InR to Dg–Dys complex comes from experiments showing that muscle specific expression of IGF counters muscle decline in mdx-mice (Barton et al, 2002; Shavlakadze et al, 2004; Dobrowolny et al, 2005). The work presented in this study is the first demonstration of genetic interaction between Dg and InR. Future biochemical studies should unravel the molecular mechanism of this interaction.
Furthermore, we have now shown that Dg–Dys complex is required both in neural and in targeting glial cells for correct neuronal axon path-finding in Drosophila brain. These data reveal that Dg–Dys complex also has a non-cell autonomous effect on axon path-finding and suggest that Dg–Dys-controlled ECM both from neuron and glial cells regulate neuronal axon path-finding. Further experiments are required to reveal whether long-range Laminin fibers are involved in this process, as has been shown in epithelial planar polarity (Bateman et al, 2001; Deng et al, 2003), or whether glial processes are observed in close proximity to the neural growth cone (Georges-Labouesse et al, 1998). Interestingly, similar phenotypes are observed with Integrin mutants (Tanaka and Sabry, 1995; Campos, 2005; Curtin et al, 2005), suggesting that, as in planar polarity (Bateman et al, 2001; Deng et al, 2003), Integrin and Dg–Dys complex might act in concert to regulate the process of ECM organization that will regulate the cytoskeleton of the cells involved.
Taken together, the phenotypes caused by Drosophila Dg and Dys mutations are remarkably similar to phenotypes observed in human muscular dystrophy patients, and therefore suggest that functional dissection of Dg–Dys complex in Drosophila should provide new insights into the origin and potential treatment of these fatal neuromuscular diseases. As a proof of principle, using Drosophila as a model we have now identified InR as a signaling pathway that genetically interacts with Dg. Future studies are directed to unravel the molecular mechanism of Dg and InR–Dock interactions in invertebrates as well as vertebrates.