Rab11 modulates α-synuclein-mediated defects in synaptic transmission and behaviour

Abstract

A central pathological hallmark of Parkinson's disease (PD) is the presence of proteinaceous depositions known as Lewy bodies, which consist largely of the protein α-synuclein (aSyn). Mutations, multiplications and polymorphisms in the gene encoding aSyn are associated with familial forms of PD and susceptibility to idiopathic PD. Alterations in aSyn impair neuronal vesicle formation/transport, and likely contribute to PD pathogenesis by neuronal dysfunction and degeneration. aSyn is functionally associated with several Rab family GTPases, which perform various roles in vesicle trafficking. Here, we explore the role of the endosomal recycling factor Rab11 in the pathogenesis of PD using Drosophila models of aSyn toxicity. We find that aSyn induces synaptic potentiation at the larval neuromuscular junction by increasing synaptic vesicle (SV) size, and that these alterations are reversed by Rab11 overexpression. Furthermore, Rab11 decreases aSyn aggregation and ameliorates several aSyn-dependent phenotypes in both larvae and adult fruit flies, including locomotor activity, degeneration of dopaminergic neurons and shortened lifespan. This work emphasizes the importance of Rab11 in the modulation of SV size and consequent enhancement of synaptic function. Our results suggest that targeting Rab11 activity could have a therapeutic value in PD.

Figure 1.

Rab11 reverses aSyn-dependent increases in average mEJP and eEJP amplitudes. Representative mEJP trace and summary graphs of averaged mEJP amplitudes for both Model 1 (A) and Model 2 (B) aSyn transgenic lines and their respective controls in third instar wandering larvae. Pan-neuronal expression of aSyn via the elavGAL4 driver in Model 1 induced a strong increase in mEJPs in aSyn animals. Co-expression of Rab11 with aSyn returned the amplitudes back to control values (N = 8–19). No change was observed in Model 2 (elavGAL4 driver) animals (N = 8–13). Relative cumulative frequency histograms and cumulative frequency curves for the mEJP amplitudes for both Model 1 (C) and Model 2 (D) aSyn transgenic lines and their respective controls are shown. eEJP sample recordings, summary graphs of averaged eEJP amplitudes and QC for Model 1 (E) and Model 2 (F) aSyn transgenic lines corrected for non-linear summation, which takes into account any changes in resting membrane potential. Pan-neuronal expression of aSyn via the elavGAL4 driver in Model 2 aSyn animals induced an increase in eEJPs and QC (F; N = 6–13). Co-expression of Rab11 with aSyn led to a reduction and normalization of eEJP amplitudes in the φC31 animals. No change was observed with Model 1 larvae regarding eEJP amplitudes or QC (E; N = 5). Data are mean ± SEM. ANOVA with Newman–Keuls post hoc tests. *P < 0.05, **P < 0.01 and ***P < 0.001.

Figure 2.

aSyn localizes to CSP in larval NMJ boutons. (A) Representative third instar larval NMJs of aSyn and Rab11 + aSyn individuals (elavGAL4 driver with Model 1 line) immunostained with aSyn and CSP antibodies. Scale bar = 8 µm. (B) The spatial distribution of aSyn and CSP overlaps as indicated by Pearson's and ICQ coefficients. Mean ± SEM. ANOVA with Newman–Keuls post hoc tests. **P < 0.01.

Figure 3.

Increased SV size in aSyn larvae is rescued by Rab11 overexpression. Representative images of 1b (A) and 1s (B) boutons from WT NMJs. Average vesicle diameter in 1b (C) and 1s (D) boutons for UAS control, aSyn and Rab11 + aSyn larvae (Model 1 line). Increased vesicle size was observed upon expression of aSyn, which is fully reversed to control levels by Rab11 overexpression. Relative cumulative frequency plots and histograms and of 1b (E and G) and 1s (F and H) boutons; a right shift is observed in aSyn larvae in comparison with controls. SV size is rescued by overexpression of Rab11 in both boutons (1b boutons, N = 2737–3190 and 1s boutons, N = 804–1019). Data are mean ± SEM. ANOVA with Newman–Keuls post hoc tests. ***P < 0.001.

Figure 4.

Rab11 overexpression rescues impaired locomotor behaviour in third instar larvae expressing aSyn. Representative crawling behaviour of control and larvae expressing aSyn, and Rab11 + aSyn in the motorneurons (A). Mean distance travelled by Model 1 larvae expressing aSyn, GFP + aSyn and Rab11 + aSyn in the motorneurons (B) and in the dopaminergic neurons (C). Similar crawling behaviour was observed in Model 2 larvae expressing aSyn in motorneurons (D) or dopaminergic neurons (E). Data are mean ± SEM. ANOVA with Newman–Keuls post hoc tests. ***P < 0.001. N = 30.

Figure 5.

Dopaminergic neuron loss is reversed by Rab11 overexpression. Representative adult (Day 30) Drosophila posterior inferiorlateral protocerebrum confocal images (PPL1, circled) of control (left panel) and Model 1 flies expressing GFP + aSyn (middle panel) or Rab11 + aSyn (right panel), driven by elavGAL4 and immunostained with anti-TH antibody. Scale bar = 40 µm. (A) Numbers of PPL1 neurons in flies at Day 1 and Day 30 (elavGAL4 in B and pleGAL4 in C). The expression of aSyn and GFP + aSyn by both drivers decreases the number of PPL1 cells, which is restored to normal levels by the overexpression of Rab11 (N = 19–65 hemispheres). Luciferase intensity measurement in Model 1 flies expressing aSyn in dopaminergic neurons (D). The reduction in luciferase signal is restored by the overexpression of Rab11 (N = 4). Quantification of average rhabdomeres per ommatidum in aSyn flies with and without photoreceptor Rab11 overexpression at Day 30 (E). Expression of aSyn via gmrGAL4 leads to a reduction in the number of photoreceptors in Model 1 flies, whereas Rab11 overexpression prevents this neurodegeneration (N = 60–136 ommatidia). (F) aSyn levels in elav-driven aSyn and Rab11 + aSyn fly heads. Representative immunoblot of aSyn soluble and insoluble fractions (left panel). The graphs represent immunoblot quantification of aSyn in the soluble (middle panel) and insoluble (right panel) fractions. The overexpression of Rab11 reduces the level of aSyn in the insoluble fraction, whereas no changes are found in the soluble fraction (N = 4). Data are mean ± SEM. ANOVA with Newman–Keuls post hoc tests. *P < 0.05, ***P < 0.001 and ns = not significant.

Figure 6.

aSyn-dependent adult phenotypes are ameliorated by Rab11 overexpression. Mean climbing pass rate for Model 1 (A) and Model 2 (B) flies expressing aSyn, GFP + aSyn and Rab11 + aSyn pan-neuronally by the elavGAL4 (N = 50–60 per condition). aSyn expression causes a reduction in climbing at all post-eclosion ages, which are strongly rescued by Rab11 overexpression. Lifespan was evaluated in Model 1 and Model 2 flies expressing aSyn pan-neuronally (C and D, respectively, N = 100 per genotype). aSyn expression increases mortality, which is reversed by Rab11 overexpression (P < 0.001 for both models). Data are mean ± SEM. ANOVA with Newman–Keuls post hoc tests. **P < 0.01, ***P < 0.001 and ns = not significant.

Figure 7.

Model suggesting how aSyn may alter normal synaptic function of Rab GTPases. Left panel: normal synaptic transmission. Right panel: aSyn expression causes an enlargement of SVs by sequestering Rab5, which is normally involved in preventing homotypic fusion between SVs. Alternatively, aSyn may function as a scaffold protein attracting and promoting the fusion between multiple vesicles. Either mechanism could lead to an increase in synaptic transmission. aSyn interaction with Rab11 reduces its activity and the vesicle trafficking associated with it. The overexpression of Rab11 reverses these defects, by reducing aSyn aggregation and enhancing its secretion, therefore restoring normal vesicle trafficking.