Despite decades of research, little is known about the pathomechanisms underlying Motor Neuron Degenerative diseases (MND), limiting the therapeutic and pre-symptomatic diagnosis options. Despite clear phenotypic and genotypic differences, there is an abundance of evidence supporting the existence of common molecular pathways operating in Spinal Muscular Atrophy (SMA) and Amyotrophic Lateral Sclerosis (ALS). Among these, the preponderance of mutations in genes linked to RNA metabolism pathways suggests it is a critical hub in Motor Neuron dysfunction, raising the intriguing question of how ubiquitously expressed genes involved in housekeeping functions can trigger tissue specific alterations.
The FlySMALS consortium aimed to identify common RNA-dependent mechanisms underlying ALS and SMA by the combination of transcriptomic and proteomic studies in fly models. Throughout the project we produced three ALS and SMA genetic fly models presenting knock-down (KD) of the disease gene orthologues Tbph, Caz (ALS) and Smn (SMA) to identify common dependent- transcripts (DE). We further generated models that allowed us to map the direct interactions of these proteins with neuronal mRNA molecules, providing a detailed view of their regulatory networks and insights into the mechanisms underlying observed gene expression changes.
Although hundreds of transcripts were commonly altered by the KDs, providing evidence for the common regulation of molecular pathways, none were found to be physically bound to the three proteins, suggesting that the underlying mechanisms are set well upstream of the observed phenotypic convergence, requiring an integrative approach to distinguish causal events from the common downstream alterations. For this purpose, we developed a computational strategy to integrate our experimental datasets over the fly protein-protein interactome. By exploiting network theory approaches we identified the most critical proteins connecting Smn, Tbph and Caz with commonly altered transcripts, confining our search space.
This approach led to the selection of ~100 candidate proteins that specifically link Smn, Tbph and Caz to their commonly induced molecular dysfunctions. This set contains a large number of ubiquitously expressed genes enriched in neuron-specific functions. Our results suggest a novel mechanism for MND that explains their phenotypic convergence, which in turn may contribute to the identification of potential drug targets or biomarkers for asymptomatic states of the disease.