Genetic Modifiers of Neurodegeneration in a Drosophila Model of Parkinson's Disease

Abstract

Disease phenotypes can be highly variable among individuals with the same pathogenic mutation. There is increasing evidence that background genetic variation is a strong driver of disease variability in addition to the influence of environment. To understand the genotype-phenotype relationship that determines the expressivity of a pathogenic mutation, a large number of backgrounds must be studied. This can be efficiently achieved using model organism collections such as the Drosophila Genetic Reference Panel (DGRP). Here, we used the DGRP to assess the variability of locomotor dysfunction in a LRRK2 G2019S Drosophila melanogaster model of Parkinson's disease (PD). We find substantial variability in the LRRK2 G2019S locomotor phenotype in different DGRP backgrounds. A genome-wide association study for candidate genetic modifiers reveals 177 genes that drive wide phenotypic variation, including 19 top association genes. Genes involved in the outgrowth and regulation of neuronal projections are enriched in these candidate modifiers. RNAi functional testing of the top association and neuronal projection-related genes reveals that pros, pbl, ct, and CG33506 significantly modify age-related dopamine neuron loss and associated locomotor dysfunction in the Drosophila LRRK2 G2019S model. These results demonstrate how natural genetic variation can be used as a powerful tool to identify genes that modify disease-related phenotypes. We report novel candidate modifier genes for LRRK2 G2019S that may be used to interrogate the link between LRRK2, neurite regulation and neuronal degeneration in PD.

Keywords: D. melanogaster; LRRK2; Parkinson’s disease; dopamine neurons.

Figure 1

Locomotor function of LRRK2 G2019S flies depends on genetic background. Negative geotaxis of Ddc-GAL4; UAS-LRRK2-G2019S flies in 148 DGRP backgrounds at 2 and 6 weeks of age. DGRP background has a strong effect on locomotor performance at both ages (2 weeks, P < 1.08 × 10⁻⁶⁶; 6 weeks, P < 4.04 × 10⁻⁴⁴).

Figure 2

Effect of candidate modifier gene KD on LRRK2 G2019S locomotor dysfunction. Top association candidates were tested at 2 weeks (A) and 6 weeks (B) of age. Neuron projection candidates were tested at 2 weeks (C) and 6 weeks (D) of age. In each case, there was a significant effect of genotype on geotaxis (individual ANOVAs, P < 0.0006). In 6-week-old flies, Bonferroni post-tests revealed a significant effect of G2019S LRRK2 expression and an enhancement or suppression effect of RNAi for some candidate genes on the performance of G2019S LRRK2-expressing flies (ns not significant, * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001). Ctrl is +/+; UAS-LRRK2-G2019S/+ and LRRK2 G2019S is Ddc-GAL4/+; UAS-LRRK2-G2019S/+.

Figure 3

Effect of candidate modifier gene KD on LRRK2 G2019S dopamine neuron loss. (A) Confocal projection images through the brain of control +/+; UAS-LRRK2-G2019S/+ (A′), and Ddc-GAL4/+; UAS-LRRK2-G2019S/+ (B′) flies and G2019S LRRK2 expressing flies with RNAi-mediated knockdown of mAChR-C (C′), tie (D′), CG14355 (E′), CG14881 (F′), CG17565 (G′), CG33506 (H′), pros (I′), pbl (J′), ct (K′) and tup (L′). Bar, 60 μM. Quantitation of total dopamine neurons in five clusters (PPM1/2, PPM3, PPL1, PPL2, and PAL) for top association (B) and neuron projection (C) candidate genes. There was a significant effect of genotype for both (individual ANOVAs, P < 0.0001) and Bonferroni post-tests revealed a significant effect of G2019S LRRK2 expression and KD of CG33506, pros, pbl and ct candidate modifiers (* P < 0.05, ** P < 0.01).