From bugs to bedside: functional annotation of human genetic variation for neurological disorders using invertebrate models

Abstract

The exponential accumulation of DNA sequencing data has opened new avenues for discovering the causative roles of single-nucleotide polymorphisms (SNPs) in neurological diseases. The opportunities emerging from this are staggering, yet only as good as our abilities to glean insights from this surplus of information. Whereas computational biology continues to improve with respect to predictions and molecular modeling, the differences between in silico and in vivo analysis remain substantial. Invertebrate in vivo model systems represent technically advanced, experimentally mature, high-throughput, efficient and cost-effective resources for investigating a disease. With a decades-long track record of enabling investigators to discern function from DNA, fly (Drosophila) and worm (Caenorhabditis elegans) models have never been better poised to serve as living engines of discovery. Both of these animals have already proven useful in the classification of genetic variants as either pathogenic or benign across a range of neurodevelopmental and neurodegenerative disorders-including autism spectrum disorders, ciliopathies, amyotrophic lateral sclerosis, Alzheimer's and Parkinson's disease. Pathogenic SNPs typically display distinctive phenotypes in functional assays when compared with null alleles and frequently lead to protein products with gain-of-function or partial loss-of-function properties that contribute to neurological disease pathogenesis. The utility of invertebrates is logically limited by overt differences in anatomical and physiological characteristics, and also the evolutionary distance in genome structure. Nevertheless, functional annotation of disease-SNPs using invertebrate models can expedite the process of assigning cellular and organismal consequences to mutations, ascertain insights into mechanisms of action, and accelerate therapeutic target discovery and drug development for neurological conditions.

Figure 1

Scheme for the functional assessment of human disease SNPs within invertebrate models. GWAS use population-based metrics to discover disease-associated SNPs. Moreover, genome or exome sequencing data from individual patients can also be analyzed to identify putative disease-causing mutations. Following identification of SNP candidates from humans, in silico algorithms can optionally be used to screen for pathogenic mutations (with accuracy rates of ~70%); see Table 1 for example programs. SNPs of interest are then modelled in flies or worms using transgenic expression of human genes (which permits the investigation of any human mutation). Alternately, an SNP can be introduced using CRISPR editing at the conserved position of the orthologous gene (which facilitates the investigation of human mutations occurring at evolutionarily conserved sites). In this manner, human mutations can be functionally annotated via cellular, molecular and behavioral assays using gene-edited invertebrate strains. Examples displayed here represent ocular degeneration in Drosophila and dopaminergic neurodegeneration in C. elegans.

Figure 2

Neurodegeneration caused by transgenic expression of human SNPs in C. elegans. SNPs in the human VPS41 gene, encoding a conserved endolysosomal fusion and trafficking protein (hVPS41), cause a newly defined class of movement disorder that includes symptoms of neurodegeneration, a phenotype revealed initially through functional analysis using transgenic C. elegans strains (49). The genetic lesions in patients occur in a compound heterozygous orientation and include the two exemplar mutations depicted in the line diagram (A). To examine the impact of VPS41 patient-derived SNPs on neurodegeneration, transgenic worms expressing human VPS41 variants exclusively in the C. elegans dopamine neurons were crossed to animals overexpressing human alpha-synuclein in the same cells. The resulting strains, which also express GFP to illuminate the dopamine neurons, were scored for the evidence of neurodegeneration (B, C). Approximately 35% of transgenic worms overexpressing alpha-synuclein alone display the normal complement of six anterior dopamine neurons (the remaining 65% exhibit a loss of neuronal processes and cell bodies). At the same point in lifespan, co-expression of WT hVPS41 did not significantly alter the amount of alpha-synuclein-dependent neurodegeneration (not shown). Similarly, if either one of the two VPS41 SNPs was co-expressed with WT VPS41 in the dopamine neurons (i.e. WT + S285P or WT + R662^*—stop codon), no significant change in alpha-synuclein-induced neurodegeneration was observed. Only co-expression of both human VPS41 SNPs (S285P and R662^*) resulted in more significant dopaminergic neurodegeneration. One-way analysis of variance with Tukey post-hoc test; day 10, post-hatching analysis. Arrowheads indicate intact dopamine neurons; arrows signify a degenerating neuron. Scale bar = 20 μm.