Proteomic identification of select protein variants of the SNARE interactome associated with cognitive reserve in a large community sample

Alfredo Ramos-Miguel,Andrea A Jones,Vladislav A Petyuk,Vilte E Barakauskas,Alasdair M Barr,Sue E Leurgans,Philip L De Jager,Kaitlin B Casaletto,Julie A Schneider,David A Bennett,William G Honer

DOI: https://doi.org/10.1007/s00401-021-02282-7

Abstract:Age-related neuropathologies progressively impair cognitive abilities by damaging synaptic function. We aimed to identify key components within the presynaptic SNARE (soluble N-ethylmaleimide-sensitive factor attachment protein receptor) machinery associated with cognitive performance and estimate their potential contribution to brain reserve in old age. We used targeted SRM proteomics to quantify amounts of 60 peptides, encoded in 30 different genes, in postmortem specimens of the prefrontal cortex from 1209 participants of two aging studies, with available antemortem cognitive evaluations and postmortem neuropathologic assessments. We found that select (but not all) proteoforms are strongly associated with cognitive function and the burden of Alzheimer's disease (AD) pathology. Specifically, greater abundance of STX1A (but not other syntaxins), SYT12, full-length SNAP25, and the GABAergic STXBP1 variant were robustly associated with better cognitive performance. By contrast, greater abundance of other presynaptic proteins (e.g., STXBP5 or tomosyn, STX7, or SYN2) showed a negative influence on cognition. Regression models adjusting for demographic and pathologic variables showed that altered levels of these protein species explained 7.7% additional between-subject variance in cognition (more than any individual age-related neuropathology in the model), suggesting that these molecules constitute key elements of brain reserve. Network analyses indicated that those peptides associated with brain reserve, and closest to the SNARE fusogenic activity, showed greater centrality measures and were better connected in the network. Validation assays confirmed the selective loss of the STX1A (but not STX1B) isoform in cognitively impaired cases. In rodent and human brains, STX1A was selectively located at glutamatergic terminals. However, in AD brains, STX1A was redistributed adjacent to neuritic pathology, and markedly expressed in astrocytes. Our study provides strong evidence, indicating that select presynaptic proteins are key in maintaining brain reserve. Compromised ability to sustain expression levels of these proteins may trigger synaptic dysfunction and concomitant cognitive impairment.