Palmitoylation of the Parkinson's disease-associated protein synaptotagmin-11 links its turnover to α-synuclein homeostasis

Abstract

Synaptotagmin-11 (Syt11) is a vesicle-trafficking protein that is linked genetically to Parkinson's disease (PD). Likewise, the protein α-synuclein regulates vesicle trafficking, and its abnormal aggregation in neurons is the defining cytopathology of PD. Because of their functional similarities in the same disease context, we investigated whether the two proteins were connected. We found that Syt11 was palmitoylated in mouse and human brain tissue and in cultured cortical neurons and that this modification to Syt11 disrupted α-synuclein homeostasis in neurons. Palmitoylation of two cysteines adjacent to the transmembrane domain, Cys³⁹ and Cys⁴⁰, localized Syt11 to digitonin-insoluble portions of intracellular membranes and protected it from degradation by the endolysosomal system. In neurons, palmitoylation of Syt11 increased its abundance and enhanced the binding of α-synuclein to intracellular membranes. As a result, the abundance of the physiologic tetrameric form of α-synuclein was decreased, and that of its aggregation-prone monomeric form was increased. These effects were replicated by overexpression of wild-type Syt11 but not a palmitoylation-deficient mutant. These findings suggest that palmitoylation-mediated increases in Syt11 amounts may promote pathological α-synuclein aggregation in PD.

Figure 1.. Syt11 is physiologically palmitoylated.

(A) Palmitoylation of Syt11 in mouse brain (homogenized tissue in modified RIPA buffer), analyzed by the ABE assay. Hydroxylamine (NH₂OH) dependence was a control for signal specificity; PSD-95 and actin are positive and negative control proteins, respectively, for the ABE assay. 5% of the total lysate was used for input. (B) Palmitoylation of Syt11 and PSD-95 in rat cortical neuronal cultures, analyzed as in (A). (C) Palmitoylation of Syt11 in human brain tissue, extracted in 2% SDS, otherwise analyzed as in (A). Blots in (A to C) are representative of 3 independent samples.

Figure 2.. Syt11 is palmitoylated at Cys³⁹ and Cys⁴⁰.

(A) HEK293 cells were transfected with the indicated FLAG-tagged mutant constructs of Syt11 and analyzed for palmitoylation by ABE assay. (B) Quantification of (A). Data are plotted as mean ±SD from n = 3 biological replicates, analyzed by one-way ANOVA with Sidak’s multiple comparisons: **P<0.01, ***P<0.001, and ****P<0.0001; ns, not significant. (C) Rat cortical neurons were transduced with lentivirus encoding Syt11-wt-FLAG or Syt11-C39S/C40S-FLAG (abbreviated “CS”) and analyzed by ABE.

Figure 3.. Palmitoylation of Syt11 regulates its degradation by the endosomal-lysosomal system.

(A) Rat cortical neurons were transduced with lentivirus containing a bicistronic vector encoding either Syt11-wt-FLAG or the palmitoylation-deficient (CS) mutant together with a ZsGreen1 reporter. Protein levels were analyzed by Western blot and quantified (right). Data are plotted as means ±SD from n = 4 biological replicates, analyzed by unpaired t-test: ****P<0.0001. (B) Human SNCA gene triplication induced neurons (iNs) were transduced, and protein levels were measured, as in (A) with quantification (right). Data are plotted as means ±SD from n = 5 biological replicates, analyzed by unpaired t-test: ***P<0.001. (C) Representative images of rat neurons from (A) transduced with either Syt11-wt-FLAG or Syt11-CS-FLAG. (D) Quantification of the ZsGreen1 signal in (C), expressed as % of the Syt11-wt-FLAG condition. Data are plotted as means ±SD from n = 8 fields from 4 biological replicates, analyzed by unpaired t-test: ns, not significant. (E) Rat neurons transduced with either Syt11-wt-FLAG or Syt11-CS-FLAG were treated with cycloheximide (CHX) for the indicated times and protein levels analyzed by WB, expressed as % that at 0 hours. Data are plotted as means ±SD from n = 3 biological replicates. (F) GAPDH protein levels from experiments in (E) were examined. Data are plotted as means ±SD from n = 3 biological replicates. (G) Rat neurons expressing Syt11-CS-FLAG were treated with the lysosomal inhibitor bafilomycin A (BafA, 50 nM) or proteasomal inhibitor epoxomycin (epox, 1 μM) or MG132 (5 μM), as indicated, followed by analysis of Syt11 protein levels. LC3B levels are used as a marker of lysosomal inhibition. Data are plotted as means ±SD from n = 3 biological replicates, analyzed by one-way ANOVA with Dunnett’s multiple comparisons. *P<0.05; ns, not significant.

Figure 4.. Palmitoylation of Syt11 confers resistance to digitonin extraction from membranes without broadly affecting Syt11 localization.

(A) Rat neurons were transduced with Syt11-wt-FLAG lentivirus or left untreated followed by immunofluorescence imaging for endogenous Syt11 and MAP2. Scale bar, 20 μm. (B) Rat neurons were transduced with Syt11-wt-FLAG lentivirus (MOI 5) or Syt11-CS-FLAG lentivirus (MOI 20), immunostained with FLAG antibodies, and imaged. Right-most panel contains an artificially brightened version of the image in the middle panel. Scale bar, 10 μm. (C) 293T cells were transfected with the indicated wild-type or mutant construct of Syt11 and subjected to 800 μg/mL digitonin extraction. The digitonin-extractable fraction is designated “Sol” (for digitonin soluble), and the resistant fraction is designated “Ins” (or “Insol” in the graphs; for digitonin insoluble). Membrane marker calnexin was a control for consistency of extractions for each mutant. Data are plotted as means ±SD from n = 4 biological replicates, analyzed by one-way ANOVA with Dunnett’s multiple comparisons: ****P<0.0001; ns, not significant. (D) Human SNCA triplication iNs were transduced with Syt11-wt-FLAG or Syt11-CS-FLAG and subjected to 1000 μg/mL digitonin extraction, and analyzed as in (C), with Na/K ATPase as the control for consistency of the extractions. Data are plotted as means ±SD from n = 5 biological replicates, unpaired t-test: *P<0.05.

Figure 5.. Syt11 modulates αS homeostasis in PD patient-derived neurons.

(A) SNCA triplication iNs expressing either Syt11-wt-FLAG or Syt11-CS-FLAG were analyzed by digitonin extraction with 1000 μg/mL of digitonin with quantification (right), calculated as the cytosol (C) to membrane (M) ratio, expressed as a percentage of vector control. Data are plotted as means ±SD from n = 4 biological replicates, analyzed by one-way ANOVA with Dunnett’s multiple comparisons: *P<0.05; ns, not significant. (B) SNCA triplication-corrected iNs were analyzed as in (A). Data are plotted as means ±SD from n = 5 biological replicates, analyzed by one-way ANOVA with Dunnett’s multiple comparisons: **P<0.01; ns, not significant. (C) Triplication iNs were transduced as in (A) and αS tetramers measured by intact cell DSG cross-linking. DJ1 was a control for consistency of the cross-linking method across conditions. Data are plotted as means ±SD from n = 4 biological replicates, analyzed by one-way ANOVA with Dunnett’s multiple comparisons: **P<0.01; ns, not significant. (D) Triplication-corrected iNs were analyzed as in (C). Data are plotted as means ±SD from n = 5 biological replicates, analyzed by one-way ANOVA with Dunnett’s multiple comparisons: *P<0.05; ns, not significant.

Figure 6.. Effect of Syt11 on pSer¹²⁹ αS.

(A) SNCA triplication iNs were transduced with the indicated lentiviruses and pSer¹²⁹ αS assessed by WB, normalized to total αS, and quantified as % vector control. Data are plotted as means ±SD from n = 13 biological replicates, analyzed by one-way ANOVA with Dunnett’s multiple comparisons: ns, not significant. (B and C) SNCA triplication-corrected iNs (B) and primary rat neurons (C) were transduced with the indicated lentiviruses followed by assessment of pSer¹²⁹ αS as in (A). Data are plotted as means ±SD from n = 8 biological replicates each, analyzed by one-way ANOVA with Dunnett’s multiple comparisons: ns, not significant. (D) Comparison of Syt11-wt-FLAG levels in SNCA triplication iNs transduced with Syt11-wt-FLAG and HEK cells transiently transfected with Syt11-wt-FLAG. To ensure valid comparisons across experiments, levels of Syt11-wt-FLAG on WB were normalized to GAPDH levels for each respective condition and expressed as a fold change of Syt11-wt-FLAG levels in the lentivirus transduced iNs. Data are plotted as means ±SD from n=6 biological replicates for iNs and 4 biological replicates for HEK cells, analyzed by unpaired t-test with Welch’s correction: **P<0.01. (E) HEK293 cells were co-transfected with either αS-wt or αS-E46K along with the indicated Syt11 mutants. pSer¹²⁹ αS was measured as in (A). Data are plotted as means ±SD from n = 4 biological replicates for αS-wt and n = 5 biological replicates for αS-E46K, analyzed by one-way ANOVA with Dunnett’s multiple comparisons: ***P<0.001, ****P<0.0001; ns, not significant. (F) HEK293 cells were co-transfected with αS-E46K and either Syt11-wt-FLAG, Syt1-wt-FLAG, or Syt4-wt-FLAG. pSer¹²⁹ was measured, presented, and analyzed as in (A) from n = 6 biological replicates. Statistical analysis: one-way ANOVA with Dunnett’s multiple comparisons: ****P<0.0001; ns, not significant.