Protein Coaggregation in Caribbean Atypical Parkinsonism: The Contribution of Annonacin

Abstract

Aims: There is an unexpectedly high proportion of atypical forms of degenerative parkinsonism in the French Caribbean islands. Residents of these islands are thought to be susceptible to Caribbean atypical parkinsonism (CAP) owing to their consumption of Annonaceae plant products containing the mitochondrial toxin annonacin. Here, we aimed to better correlate the clinical diagnosis of CAP with the misfolded protein pathology observed in affected individuals and to further investigate how annonacin could contribute to CAP pathogenesis.

Methods: We conducted postmortem histopathological analysis of brain samples from eight patients; more specifically, we assessed the distribution and burden of α-synuclein (αS) and tau pathologies. Additionally, we studied how annonacin influences αS and tau aggregation using biophysical assays, with the corresponding recombinant human proteins serving as substrates.

Results: CAP was associated with heterogeneous clinical and histopathological features. Tau/αS copathology with a predominance of either αS or tau aggregates was observed in the majority (5/8) of patients. Tau and αS aggregates were sometimes colocalised in the same brain regions or cells. In biophysical assays, we showed that annonacin leads to an increase in αS aggregation and the formation of αS fibrils that could cross-seed tau aggregation.

Conclusions: Annonacin may contribute to degenerative CAP by modulating the production of tau and αS pathogenic protein assemblies.

Keywords: annonacin; atypical parkinsonism; copathology; tau; α‐synuclein.

FIGURE 1

The majority of Caribbean atypical parkinsonism patients have a mixed p‐tau/αS pathology. Tau and αS pathology was observed in five out of eight patients in different regions of the brain through the use of anti‐tau (clone AT8) (p‐tau) and anti‐αS (clone 5G4) antibodies that detect hyperphosphorylated (Ser202, Thr205) tau and aggregated αS, respectively. In some of these regions, concomitant αS and p‐tau pathological aggregates were observed. P1 showed marked pathological αS and p‐tau aggregates in the CA2 region of the hippocampus (a, b). αS and p‐tau aggregation was moderate in P2 in the same region as in P1, and a few intraneuronal neurofibrillary tangles were seen (inset) (hippocampus; CA2) (c, d). In P3, there was a moderate amount of pathological αS and tau aggregates in the reticular formation of the medulla oblongata (e, f). In contrast, in P4, there were rare Lewy neurites (g: arrow and arrowhead [inset]) in the dorsal nucleus of the vagus nerve (dmX) in the medulla oblongata associated with a low amount of p‐tau pathological aggregates (h). P5 (i, j) was primarily a tauopathy in which we observed a moderate number of p‐tau deposits in the dmX within the medulla oblongata, with only rare αS deposits (i, arrow and arrowhead [inset]). In P6, only p‐tau aggregates were observed in the reticular formation of the medulla oblongata, whereas αS immunostaining was negative (k, l). Finally, in P8, a marked amount of pathological p‐tau was detected in the locus coeruleus, with no αS aggregates (m, n). Scale bars: 50 μm.

FIGURE 2

Codetection of p‐tau and p‐αS aggregates in the brains of CAP patients. Coexistence of p‐αS and p‐tau aggregates in a large pyramidal cell soma from the CA2 hippocampal region in Patient 1 (P1). p‐αS and p‐tau aggregates are present concomitantly in neuropil (a, first column). Coexistence of p‐αS and p‐tau aggregates in cell somas but less often in neuritic processes in the CA2 hippocampal region of Patient 2 (P2) (a, second column). p‐tau aggregates are comparatively more abundant in patient P2 than in P1. p‐αS and p‐tau aggregates coexist in cell somas but not in neurites in the nucleus basalis of Meynert from Patient 3 (P3) (a, third column). Note the presence of small p‐tau puncta in the cell body with a large αS deposit. A higher magnification of selected areas is presented above (b). Large and thin arrows point to cell bodies and neuropil containing both p‐αS and p‐tau aggregates, respectively in (a, b). Arrowheads point to cells comprising a single type of aggregate. Scale bar: 20 µm (a) and 10 µm (b).

FIGURE 3

The impact of the mitochondrial Complex I inhibitor annonacin on αS and tau aggregation. αS aggregation measured by ThT fluorescence after 72 h of incubation of αS_m (70 μM) with (1–10 μM) or without annonacin (−). Data in % of maximal ThT fluorescence levels are presented as mean values ± SEM (n = 3). ***p < 0.001 versus αS. One‐way ANOVA followed by Dunnett's test (a). Fibrillization kinetics (0–72 h) of αS measured by ThT fluorescence in the presence or absence of αS_m (70 μM) mixed (red line and squares) or not mixed (black line and circles) with annonacin (a; 10 μM). Variations in ThT fluorescence levels are shown for samples containing annonacin only (green line and triangles). Data expressed in % of maximal ThT fluorescence at the endpoint stage are presented as mean values ± SEM (n = 3) (b). TEM images of αS samples incubated at 37°C for 72 h under continuous orbital agitation in the absence (upper panel) or presence (lower panel) of annonacin (10 μM). The right panel shows enlarged images depicted by the dotted line squares. Scale bars: 500 nm (left) and 50 nm (right) (c). Measurements of the length (in μm) and diameter (in nm) of fibrils formed in the presence (αS:A_f) or absence (αS_f) of annonacin. Data are presented as mean values ± SEM (n = 16). ***p < 0.001 versus αS. Student's t‐test (d, e). Tau fibrillization kinetics monitored by ThT fluorescence in conditions where tau_m (22 μM) is mixed with 7 μM (monomer equivalent) fibrillar species of αS formed in the presence (αS:A_f; red line and circles) or absence (αS_f; black line and squares) of annonacin (10 μM). Variations in ThT fluorescence levels are also shown when annonacin (10 μM) was added in the presence (inverted green triangle and line) or absence (blue triangle and line) of tau_m (f). Tau fibrillization kinetics monitored by ThT fluorescence in conditions where tau_m (22 μM) was mixed (red square and line) or not (green circle and line) with 0.2 mg/mL heparin to promote aggregation. ThT fluorescence levels are estimated comparatively under conditions where test samples contain tau_m with heparin and annonacin (black triangle and line) or heparin alone (inverted yellow triangle and line) (g). In (f) and (g), data expressed in % of maximal ThT fluorescence signal at the endpoint stage are presented as mean values ± SEM (n = 3).