Alteration of RNA binding proteins (RBPs) has been linked to Amyotrophic Lateral Sclerosis (ALS). We have recently identified HuD/ELAVL4 as a novel neural RBP involved in FUS-linked and sporadic ALS (De Santis et al., 2017; De Santis et al., 2019). New preliminary evidence from multiple FUS models revealed that: a) altered HuD levels are possibly due to competition between mutant FUS and FMRP for HuD 3'UTR binding; b) as a consequence, levels of HuD targets NRN1 and GAP43 are also increased; c) FUS mutant motoneurons display aberrant axon branching and growth upon injury. NRN1 and GAP43 are HuD-regulated genes that might underlie this phenotype. It has been recently shown that co-cultures of iPSC-derived motor neurons and myotubes from patients with FUS-ALS display endplate maturation defects due to intrinsic FUS toxicity in both motor neurons and myotubes (Picchiarelli et al., 2019).
Here we aim at establishing a link between the altered formation of neuromuscular junctions and the molecular circuitry involving mutant FUS and HuD. This project is based on a relevant in vitro model, consisting of engineered human iPSCs, raised in the laboratory of Prof. Rosa. We will, in particular, take advantage of the possibility to convert iPSCs into disease-relevant cell types to establish co-cultures of motoneurons and skeletal muscle cells in microfluidics devices. The participation of Prof. Bozzoni will be instrumental for the analysis of skeletal muscle cells from iPSCs. This multidisciplinary project will greatly benefit from the expertise of Prof. Scopigno in optical imaging techniques and in particular coherent Raman microscopy, as a new tool to real-time, label-free in-vivo imaging of neuromuscular junction. Overall, we expect from this project insights into the contribution of the motoneuron to altered formation of the neuromuscular junction, as a therapeutic target in FUS ALS.
ALS has been genetically linked to individual mutations in several RNA-binding proteins (RBPs), including FUS, TDP-43, ATXN2, HNRNPA1 and TAF-15 (Lagier-Tourenne et al., 2010). In particular, ALS mutations in the FUS nuclear localization domain resulting in strong cytoplasmic mislocalization have been reported in patients affected by severe and juvenile ALS (Dormann et al., 2010). It has been hypothesized that the altered activity of an individual RBP results in defective RNA metabolism in the MN. However, which is the effect of a mutation in a given RBP on the functions of other RBPs remains an unexplored point. Strikingly, ALS-linked RBPs are ubiquitously expressed while their mutations, delocalization and/or altered levels selectively cause motoneuron degeneration. Interestingly, recent work by our lab and others identified an extensive crosstalk between FUS and other RBPs (Nakaya et al., 2013; Zhou et al., 2013; Masuda et al., 2015; Blokhuis et al., 2016; De Santis et al., 2017; De Santis et al., 2019). Taken together, these evidences suggest that a network of RBPs exists in the nervous system, where ubiquitous and neural-specific RBPs cooperate in gene expression regulation. Cross-regulation and feedback loops are crucial in establishing proper levels of each RBP. ALS mutations in one node of the network, e.g. FUS, might impact other RBPs and produce broader, unexpected, effects on the RNA metabolism of the MN. The results of our previous work and new preliminary data revealed that FUS mutant MNs have altered levels of HuD, an important neural RBP. HuD, encoded by the ELAVL4 gene, has been very recently linked to ALS by us (De Santis et al., 2017; De Santis et al., 2019) and others (Loffreda et al., 2020). Two of the genes upregulated by HuD, NRN1 and GAP43, play crucial roles in axon outgrowth.
Despite this proposal is focused on ALS-FUS models, the mechanistic insights obtained here could be relevant for other genes whose mutations are causally linked to familial and sporadic ALS. To this regard, our preliminary results and work by others (Akiyama et al., 2019) suggest that mutant FUS MNs have an aberrant axon outgrowth phenotype. Notably, this alteration is remarkably similar to aberrant phenotypes recently reported in an ALS-SOD1 mouse model (Osking et al., 2019). Importantly, acute expression of SOD1-G93A in WT MNs for 10 days was sufficient to increase axonal regeneration. Thus, at least in the case of SOD1, increased axonal outgrowth and neurite branching unlikely represent a mere compensatory effect (Osking et al., 2019). Alteration of the NMJ, due to cell-autonomous FUS toxicity in both MNs and myotubes, has been previously reported (Picchiarelli et al., 2019). However, only the contribute of the muscle has been explored.
Our project is based on ALS-relevant model systems consisting of human iPSC-derived motoneurons developed in our lab. It is INNOVATIVE and ORIGINAL because it aims to uncover broader effects of FUS mutations on other RBS and links two aspects of ALS pathogenesis: the alteration of the RBP network and the aberrant axon phenotypes in MNs. These aspects will be studied in the context of the neuromuscular junction, by taking advantage of microfluidics devices. Interestingly, overexpression of GAP43 induced MNs death in mouse (Aigner et al., 1995; Harding et al., 1999). Whether NRN1 and/or GAP43 play a role in ALS downstream of HuD remains unknown. In the long term, they might be used for designing new therapeutic strategies and/or as biomarkers.
The direct, non-invasive CARS imaging of such systems provides a unique method to directly unveil the mechanism of ALS pathogenesis. Taking advantage of the 3D reconstruction available with CARS microscopy, we expect to find the spatial distribution and the chemical concentration of ACh during the disease progression stages in the in-vitro model.