APOE4 copy number‐dependent Complement Pathway Dysregulation

Background Complement has been implicated in the pathophysiology of AD based on human genetic studies, human post‐mortem pathology, and mouse genetic studies. However, the relationship between complement activation and other AD risk factors (such as the APOE4 allele) remains unclear. Method We used pathway analysis of the ADNI targeted proteomics dataset to identify protein clusters highly correlated with each other, including a complement pathway cluster. We then developed a hybrid proteomics platform, which allows for both discovery proteomics (via LC‐MS/MS) and precise quantitation of complement protein‐derived peptides via Stable Isotope Labelled (SIL) internal control peptides for the complement proteins. We then tested this hybrid proteomics platform on over 130 CSF samples from cognitively normal older adults (ages 60‐80), together with assays for CSF amyloid beta, tau and p‐tau‐181p. Spearman correlation tests were used to evaluate correlations between protein pathway expression and demographic variables, such as APOE4 allele number. Result Pathway analysis of the ADNI targeted proteomics dataset identified a cluster of complement proteins whose levels were highly correlated with each other in the CSF; this cluster included peptides from C1QB, C2, C3, C4A, C5, C6, C8B and CRP and several additional proteins. Across ∼300 individuals (including cognitively normal controls, MCI and AD patients), expression of this complement protein cluster was inversely correlated with APOE4 allele count and gender (with lower expression in females), and a positive correlation with age (P