Hyperphosphorylated forms of the tau protein are pathological agents in Alzheimer's disease (AD) and also
found in cognitively normal older adults. Tau first accumulates in the entorhinal cortex (ERC) and then spreads
through the brain in a stereotypical spatiotemporal pattern driven by trans-synaptic and activity related
mechanisms. These findings suggest that patterns of ERC functional connectivity (FC) may be critical to
understanding the initial progression of tau spread in humans, though this association has not yet been
investigated. Additionally, previous literature suggests a bidirectional relationship between FC and tau spread,
such that while strong FC may initially drive tau spread, high levels of tau could eventually disrupt FC networks.
The current project will investigate for the first time in humans whether tau spread is associated with FC
networks of the ERC using an innovative multimodal neuroimaging approach in cognitively normal older adults
(OA). Tau will be quantified in vivo using the positron emission tomography (PET) tracer [18F] Flortaucipir.
ERC FC networks will be derived from resting-state functional MRI using seed-to-voxel FC analyses.
Additionally, Aß will be measured with [11C] PiB-PET, neurodegeneration with structural MRI, and episodic
memory with a composite neuropsychological test score. In Aim 1, we will investigate the proposed
bidirectional model between tau spread and ERC FC networks in a cross-sectional design in OA subjects. In
Aim 2, we will assess within OA subjects whether ERC FC strength at baseline is predictive of longitudinal tau
deposition at follow-up 1.5-2 years later, and whether these factors are also predictive of longitudinal episodic
memory decline. Findings from this study will help explain patterns of tau spread through the human brain,
providing a key target for interventions to reduce or prevent the spread of tau in aging and AD. Additionally, the
impact of tau on the medial temporal lobe memory system will be further characterized, allowing predictors of
memory decline to be identified. Completion of the proposed research will achieve the applicant's training
goals, including (1) development of neuroimaging skills in PET, fMRI, and sMRI, (2) conceptual training in the
clinical pathophysiology of AD, (3) improvement of scientific communication skills, and (4) growth of teaching
and mentoring skills. The Helen Wills Neuroscience Institute at UC Berkeley provides a network of innovative
neuroscience researchers and world-class facilities for neuroimaging. Dr. William Jagust, the sponsor, is a
leader in applying multimodal neuroimaging methods to the study of aging and AD, and has a proven track
record of successfully mentoring students. The combination of the proposed research and training plan will
provide the applicant with a comprehensive foundation on which to build a successful post-doctoral fellowship
and research career using multimodal neuroimaging to study memory impairment in aging and disease.