Establishing cellular-validated standards for parcellating hippocampal subfields in aging and Alzheimer's disease

Aging and Alzheimer’s Disease (AD) differentially affect the hippocampal subfields. A robust,
reliable imaging tool is needed to evaluate the hippocampal subfields and distinguish early AD
from normal aging. Though recent neuroimaging studies have harmonized the labeling of the
whole hippocampus with success, multiple groups have parcellated the hippocampus further
into subfields and highly variable results emerged. Conflicting reports exist about atrophy
changes along the aging-Alzheimer’s continuum. The hippocampus contains selective
vulnerable cell populations. Identifying pyramidal cell profiles for all subicular cortices, CA1,
CA2, and CA3 is essential to develop rigorous, cell-validated definitions for the hippocampal
subregions and to corroborate these as reliable biomarkers is a critical step to determine AD
vulnerability. Histology is required to differentiate subfields accurately. Our overall goal is to
characterize the hippocampus subregions based on cells, not layers or microanatomical
features. The objective of this proposal is to create twelve histologically-validated hippocampal
subfields based on novel, cellular criteria (size, pattern, directionality measures) and apply this
neuroimaging map to in vivo populations. This regionally integrated map will establish new
cellular readouts for the hippocampal formation that represent the subpopulations most
vulnerable in AD. The project will combine high-resolution ex vivo MRI, a preclinical brain tissue
collection, and our neuroanatomical expertise. Aim 1 will establish a novel neuroimaging tool
that segments twelve hippocampal subfields in FreeSurfer. Aim 2 will validate hippocampal
subregions histologically with Nissl staining corresponding to the same cases and establish
novel anatomical and cellular frameworks for each subregion: pyramidal cell diameter
measures, pyramidal neuronal pattern profiles from cell directionality measures and our triple “c”
protocol to compute architectonic boundaries. Aim 3 will apply the hippocampus subfield
segmentation tool to 3 T in vivo MRI data including the Harvard Aging Brain Study, the Human
Connectome Project, Alzheimer’s disease Neuroimaging Initiative and validate against
established biomarkers. Our regionally integrated subfield map is needed to disambiguate aging
from AD in terms of specificity, sensitivity and reliability in early prediction of AD. The work – at
the microscale – will generate well-validated standards that will help to disentangle cellular and
network vulnerabilities in healthy aging from AD in future in vivo functional, longitudinal, or
therapeutic studies.

Grant Number: 5R01AG072056-03
NIH Institute/Center: NIH
Principal Investigator: Jean Augustinack