Testing roles of Hereditary Spastic Paraplegia (HSP) proteins in organization of axonal endoplasmic reticulum (ER) and ER-mitochondria contacts

Abstract

The Hereditary Spastic Paraplegias (HSPs) are a group of genetically heterogeneous, neurodegenerative and neurodevelopmental diseases characterised by spasticity and lower limb weaknesses. Some known causative genes imply the importance of endoplasmic reticulum (ER) function and morphogenesis in HSPs. These genes encode ER-shaping proteins: spastin (SPG4), atlastin (SPG3A), Receptor Expression Enhancing Protein 1 (REEP1/SPG31) and reticulon (SPG12). These proteins share a common feature of one or two intramembrane hairpin domains that can recognise or drive curvature of the ER membrane. In Drosophila, removing widely expressed members of the reticulon and REEP families leads to fewer ER tubules in axons of wider diameter, although there is no widespread absence of tubules. Therefore, other proteins must be involved in shaping the tubular ER network in Drosophila axons, and a major goal of mine was therefore to identify and assess additional candidate proteins with roles in this process. One potential source of ER-shaping proteins is other HSP genes. C19orf12 (SPG43) is an HSP gene; C19orf12 reportedly localise to ER and has a predicted intramembrane domain. It is therefore a candidate for helping to shape the axonal ER network. To investigate possible roles of C19orf12 in ER structure and function, I generated transgenic flies carrying a fusion of GFP to CG3740, which showed strong localisation to axons. I also generated loss-of function mutants of the most widely expressed Drosophila ortholog of C19orf12, CG3740, using P element excision and CRISPR/Cas9. These mutants are homozygous viable, as are quadruple mutants lacking CG3740 and all the widely expressed reticulon and REEP proteins, suggesting that these 4 proteins together are not sufficient for tubular ER formation. I did not see any overt effect in ER level due to CG3740 mutation. It also did not enhance the ER fragmentation phenotype was seen in Rtnl1, ReepA, and ReepB triple mutant. Another potential source of ER-shaping proteins is proteomic studies highlighting the proteins enriched in ER tubules. I therefore, performed proteomic analysis from previously conducted proteomic studies. These analyses identified most of the known proteins with roles in ER shaping, implying that it may also be a way to identify additional proteins with similar roles. Around 15 proteins were finalised as potential candidates for an ER-shaping role and two of them appeared as strong candidates. 8 Many HSP proteins localised in ER, but some localised in mitochondria. How might proteins in two different organelles give a similar disease phenotype? I, therefore, wanted to have a way to monitor ER-mitochondria contacts in HSP mutants. I adapted a published sensor that is based on a split-GFP strategy for ER-mitochondria contacts in transgenic Drosophila. Split GFP peptides were fused with ER and mitochondria localised proteins. When two organelles are nearby, GFP is reconstituted and becomes fluorescent. In Drosophila larval axons, the reporter showed a punctate distribution similar to that of mitochondria, consistent with localisation to contact sites. The reporter also showed alteration in different diet conditions showing that it responds to different physiological conditions.