Theses Doctoral

Biomarkers of Alzheimer-Associated Endosomal Dysfunction

Neufeld, Jessi

Endosomal dysfunction has been mechanistically linked to Alzheimer’s Disease (AD). To date, no in vivo biomarkers for this cellular deficit exist. Yet such biomarkers are required for determining its prevalence in AD and tracking its time course—both in disease progression and potential clinical trials. With this goal in mind, we made use of an assortment of mouse models bearing AD-related endosomal trafficking defects through selective deletion of retomer core proteins. We collected CSF and brain exosomes from these retromer-deficient models and performed a battery of molecular inquiries which included lipidomic and proteomic screens, as well as hypothesis-driven biochemistry. The results of this comprehensive investigation include the first characterization of the murine CSF lipidome and the deepest characterization to date of the murine CSF proteome.

Herein, we report that VPS26a haploinsufficiency in the brain imparts no detectable protein changes in the CSF as measured by labeled LC-MS/MS at three months of age. This deficit does, however, cause a reliable reduction of CSF sphingomyelin d18:1/18:1, which is exacerbated by age, extending to other sphingomyelins and other lipid classes including dihydrosphingomyelins and monohexosylceramides.

Complete knockout of its paralog VPS26b promotes an enrichment of BACE1-cleaved APP CTFs (Beta-CTFs) in brain-derived exosomes and may alter exosomal biogenic pathways. Similar trends were seen in a neuronal-specific knockout (via Camk2-Cre recombinase) of retromer’s linchpin, VPS35.

Most importantly, an unbiased proteomic screen of CSF collected from mice with a selective knock out of VPS35 in forebrain neurons (engineered using the Camk2 system) uncovered a total of 71 hits (52 parametric and 19 nonparametric) from the 1505 proteins detected. Pathway analysis and follow-up studies identified two distinct molecular categories with previously established relevance to AD: BACE1 substrates and MAPT (more commonly referred to as tau). We report that, both in vivo and in vitro, neuronal-selective knockout of VPS35 causes increased secretion of the N-terminal fragments (NTFs) of BACE1 substrates APLP1 and CHL1 as well as total tau, and importantly, that these events occur independent of cell death. Further, we find evidence of convergence of these pathways in both mouse and human CSF. However, as these BACE1 substrates likely accumulate in plaques, we propose CSF total tau as a biomarker of endosomal dysfunction with utility over the entire course of AD progression.

We have identified and validated a series of in vivo biomarkers that are reflective of AD-associated endosomal dysfunction. While clearly sensitive to this cellular pathology, future work is required to determine their specificity. Additionally, follow-up studies are required to show that interventions which rescue endosomal dysfunction affect this molecular profile. The identified biomarkers hold great promise for early detection of endosomal dysfunction in AD and for tracking its course, during the disease progression and for clinical trials. Furthermore, the unexpected but validated finding, showing that increased CSF tau is reflective of AD-associated endosomal dysfunction, suggests that endosomal dysfunction is a universal deficit shared among AD patients in its earliest stages of disease.


  • thumnail for Neufeld_columbia_0054D_14486.pdf Neufeld_columbia_0054D_14486.pdf application/pdf 4.39 MB Download File

More About This Work

Academic Units
Pathobiology and Molecular Medicine
Thesis Advisors
Small, Scott A.
Ph.D., Columbia University
Published Here
March 30, 2018