Calcineurin-mediated regulation of growth-associated protein 43 is essential for neurite and synapse formation and protects against α-synuclein-induced degeneration

Abstract

Introduction: Elevated calcium (Ca²⁺) levels and hyperactivation of the Ca²⁺-dependent phosphatase calcineurin are key factors in α-synuclein (α-syn) pathobiology in Dementia with Lewy Bodies and Parkinson's Disease (PD). Calcineurin activity can be inhibited by FK506, an FDA-approved compound. Our previous work demonstrated that sub-saturating doses of FK506 provide neuroprotection against α-syn pathology in a rat model of α-syn neurodegeneration, an effect associated with the phosphorylation of growth-associated protein 43 (GAP-43).

Methods: To investigate the role of GAP-43 phosphorylation, we generated phosphomutants at the calcineurin-sensitive sites and expressed them in PC12 cells and primary rat cortical neuronal cultures to assess their effects on neurite morphology and synapse formation. Additionally, we performed immunoprecipitation mass spectrometry in HeLa cells to identify binding partners of these phosphorylation sites. Finally, we evaluated the ability of these phosphomutants to modulate α-syn toxicity.

Results: In this study, we demonstrate that calcineurin-regulated phosphorylation at S86 and T172 of GAP-43 is a crucial determinant of neurite branching and synapse formation. A phosphomimetic GAP-43 mutant at these sites enhances both processes and provides protection against α-syn-induced neurodegeneration. Conversely, the phosphoablative mutant prevents neurite branching and synapse formation while exhibiting increased interactions with ribosomal proteins.

Discussion: Our findings reveal a novel mechanism by which GAP-43 activity is regulated through phosphorylation at calcineurin-sensitive sites. These findings suggest that FK506's neuroprotective effects may be partially mediated through GAP-43 phosphorylation, providing a potential target for therapeutic intervention in synucleinopathies.

Keywords: FK506; GAP-43; calcineurin; neurite branching; neuroprotection; synapses; α-synuclein.

FIGURE 1

Calcineurin-dependent phosphosites S86 and T172 of GAP-43 contribute to neurite branching in PC12 cells. (A) A heat map representing phosphoproteomic hits identified in the rat striatum after correction for protein abundance, showing statistically significant differences between control and α-synuclein-expressing rats (Caraveo et al., 2017). Each row represents a protein from which the corresponding phosphorylated peptide was detected. Peptides highlighted in orange indicate GAP-43 phosphorylation sites that were restored to control levels in α-synuclein-expressing rats treated with sub-saturating doses of FK506. (B) Phosphopeptides from GAP-43 rescued with <5 ng/mL FK506 treatment in α-syn rat striatum. The identified phosphorylation sites are highlighted in red. Data represent n = 3 rats. *p < 0.05 (two-tailed t-test). (C,D) PC12 cells were co-transfected with GFP and empty vector or GAP-43 in Wild type (WT) or the indicated phosphomutant form. The analysis includes measurements of the longest neurite (C) and the number of branches (D) from >300 cells across 4 independent experiments. All experiments are normalized relative to GFP control (red dashed lines; each line represents ± SEM); WT GAP-43 serves as reference (black lines; each line represents ± SEM). *p < 0.05; **p < 0.01; ***p < 0.001 comparison between phosphomimetic and phosphoablative mutants and ^∙p < 0.05; ^∙⁣∙∙p < 0.001 comparison between phosphomutants and WT; One-way Anova with post hoc Tukey’s test.

FIGURE 2

Calcineurin-dependent phosphosites S86 and T172 of GAP-43 contribute to neurite branching and spine formation in primary cortical neurons. (A) Representative confocal images of rat primary cortical neurons transduced with GFP or GFP fusions of GAP-43 WT, phosphoablative mutant (S86A,T172A), or phosphomimetic mutant (S86E,T172E) at the CaN-dependent phosphorylation sites. Cultures were transfected with a fluorescent protein mScarlet. Scale bar is 50 μm. (B) Sholl analysis of the cultures in conditions presented in panel (A). N = 30–50 cells per condition. (C) Total number of spines/10 μm from conditions in panel (A). (D) Distribution of spine morphology from conditions in panel (A). One-way Anova with post hoc Tukey’s test was used to analyze spine morphology distribution amongst the treatment groups. The distribution was not significantly different. (E) Representative images of dendritic spines from cultures in panel (A). Scale bar is 10 μm. In all conditions, SEM and *p < 0.05. One-way Anova with post hoc Tukey’s test.

FIGURE 3

Calcineurin-dependent phosphosites S86 and T172 of GAP-43 bind structural components of ribosomes. (A) Venn diagram representation of all MS hits that passed the selection criteria described in main text and materials and methods. N = 3 per condition. One-Way Anova with a multi-comparison test using a two-stage linear step-up procedure of Benjamini, Krieger, and Yekutieli, q = 0.05. (B) Gene ontology analysis for molecular processes from the MS hits in panel (A).

FIGURE 4

Calcineurin-dependent phosphosites S86 and T172 of GAP-43 regulate α-synuclein-mediated pathobiology. (A) Representative western blot from rat primary embryonic cortical neurons transduced with either GFP, or GFP fusions of GAP-43 Wild type (WT), phosphoablative (S86A,T172A) or phosphomimetic (S86E,T172E) mutants of the CaN-dependent phosphorylation sites, or the previously established PKC-dependent site (S41) in phosphoablative, S41A, or phosphomimetic, S41D, mutant form, in combination with Empty vector (EV) or αSyn A53T. Actin serves as a loading control. (B) Quantitation of GFP fluorescence intensity over actin from panel (A). (C) Quantification of the neuronal marker MAP2 of the neuronal cultures immunostained for MAP2 (D). (E) Quantification of ATP levels from conditions in panel (A). Experiments in panels (A–E), N = 3 per condition. (F) Representative western blot for PSD95, GFP and actin from cultures in panel (A) and quantitation of PSD95 signal/actin, the loading control (G). One-way Anova with post hoc Tukey’s test; *<0.05; **<0.01.