Other, more recent examples of this type of molecular lesson involve entire groups of mendelian disorders that share overlapping clinical phenotypes, even though they arise from mutations in different genes. These syndromes have highlighted the existence of complex molecular networks or pathways that include distinct but functionally converging genes. A paradigmatic example has come from a genetically heterogeneous group of inherited neurologic disorders that are characterized by progressive spasticity and weakness of the lower limbs. These disorders, which are caused by corticospinal motor neuron axonopathy, are the hereditary spastic paraplegias. They have autosomal dominant, recessive, and X-linked inheritance. To date, 20 genes have been identified, half of which are involved in membrane trafficking along the exocytic and endocytic pathways. The remainder are involved in mitochondrial functions, myelination, lipid metabolism, and DNA repair.

More than 50% of patients with hereditary spastic paraplegia carry mutations in one of three genes: spastin (SPG4), receptor-expression-enhancing protein 1 (SPG31 or REEP1), or atlastin-1 (SPG3A). Spastin encodes an ATPase with a microtubule-severing activity that has different splice variants with different subcellular localizations, including the endosomes and the endoplasmic reticulum. Notably, spastin interacts with the other hereditary spastic paraplegia protein, REEP1. REEP proteins, and the structurally related reticulon proteins, have a major morphogenetic role at the endoplasmic reticulum45 because of a conserved domain of approximately 200 amino acids with two hydrophobic segments that form a hairpin in the membrane and have membrane-bending properties. Through this domain and its ability to oligomerize, the REEP and reticulon proteins can shape membranes of the endoplasmic reticulum into tubules. Intriguingly, spastin also interacts with the third major hereditary spastic paraplegia protein, atlastin. These collective observations led to the hypothesis that atlastin itself might have a role in the morphogenesis of the endoplasmic reticulum. This disease-inspired hypothesis turned out to be correct and revealed that atlastin is involved in the generation of the tubular endoplasmic-reticulum network, since it mediates homotypic fusion of tubules in the endoplasmic reticulum. Finally, in a further tightening of the relationships among atlastin-1, spastin, and REEP1, these three proteins have recently been reported to interact with one another.

This emerging scenario supports a convergent mechanism of disease in the many forms of hereditary spastic paraplegia that involve a defect in the formation of the endoplasmic reticulum tubular network. This might be particularly detrimental for long spinal neurons, since the endoplasmic reticulum is a conduit for many important small molecules with signaling or structural roles (e.g., calcium and lipids). Thus, the pervasiveness and continuity of the endoplasmic-reticulum network might well be essential in these extremely elongated cells, whereas such a network may be at least partially dispensable in smaller cells.