Tau deposition is associated with functional isolation of the hippocampus in aging

Abstract

The tau protein aggregates in aging and Alzheimer disease and may lead to memory loss through disruption of medial temporal lobe (MTL)-dependent memory systems. Here, we investigated tau-mediated mechanisms of hippocampal dysfunction that underlie the expression of episodic memory decline using fMRI measures of hippocampal local coherence (regional homogeneity; ReHo), distant functional connectivity and tau-PET. We show that age and tau pathology are related to higher hippocampal ReHo. Functional disconnection between the hippocampus and other components of the MTL memory system, particularly an anterior-temporal network specialized for object memory, is also associated with higher hippocampal ReHo and greater tau burden in anterior-temporal regions. These associations are not observed in the posteromedial network, specialized for context/spatial information. Higher hippocampal ReHo predicts worse memory performance. These findings suggest that tau pathology plays a role in disconnecting the hippocampus from specific MTL memory systems leading to increased local coherence and memory decline.

Fig. 1

Local and distant functional connectivity can be measured using different, related approaches. a Regional homogeneity (ReHo) measures the nonparametric concordance of adjacent voxel timeseries (TS). For each voxel in the gray matter (center voxel in the figure), a Kendall’s correlation coefficient value is computed between that voxel’s TS and its 26 neighbors who share at least a face, edge or corner with the center voxel. ReHo is a measure of local connectivity. b Conventional functional connectivity measures statistical dependency, often Pearson correlation, between the time-varying fMRI BOLD signal in two voxels or regions. In this example, mean timeseries from two distant regions in a parcellation are correlated to estimate functional connectivity

Fig. 2

Patterns of ReHo in gray matter change with age. a Group mean ReHo images of YA (n = 50) and OA (n = 89) demonstrated distinct ReHo patterns. Areas of high ReHo in YA were lateral parietotemporal junction, precuneus, and primary visual areas. In OA, high ReHo was observed in a majority of the temporal lobe. b ANOVA was used to compare ReHo in a priori cortical ROIs between YA and OA. Differences in ReHo signal were observed in all illustrated ROIs except RSC where the difference was trending (p = 0.081). ReHo in AT ROIs, PHC, and ERC was significantly greater in OA than YA (all p < 0.001), while ReHo in PCC and Prec was significantly lower in OA than YA (all p < 0.001). c There was no relationship between age and hipp-ReHo in YA (r = 0.05, p = 0.75; all r and p-value pairs reflect Pearson correlation). d In contrast to YA, there was a significant positive relationship between age and hipp-ReHo in OA (r = 0.35, p < .001). Source data are provided as a Source Data file. YA young adults, OA older adults, Prec precuneus, PCC posterior cingulate cortex, RSC retrosplenial cortex, Fus fusiform, PHC parahippocampal cortex, ERC entorhinal cortex, TP temporal pole

Fig. 3

Hipp-ReHo is associated with AD pathology in older adults. a Global PiB DVR (PiB + in red; PiB- in gray; threshold DVR > 1.065) positively predicted hipp-ReHo (r = 0.29, p = 0.01; all r and p-value pairs reflect Pearson correlation). Global PiB DVR remained a significant predictor of hipp-ReHo after adjusting for age, sex, and hippocampal volume. The band within the boxplots represents the median while the upper and lower edges of the box represent the first and third quartiles, respectively. The whiskers extend up to 1.5 times the interquartile range. Data points outside this range are plotted as outliers. b ERC FTP SUVR (FTP + in red; FTP- in gray; threshold SUVR > 1.26 in Braak III/IV) was a positive predictor of hipp-ReHo (r = 0.35, p < 0.001). ERC FTP SUVR remained a significant predictor of hipp-ReHo after adjusting for age, sex and hippocampal volume. Illustrative boxplots are used as in panel A. c, d Voxelwise regressions of ReHo images spanning all gray matter voxels with global PiB DVR or ERC FTP SUVR as the independent variable of interest illustrate significant clusters (cluster corrected p < 0.05) predominantly within the hippocampus. e A mediation analysis testing if ERC FTP SUVR mediates the observed relationship between PiB DVR and hipp-ReHo revealed a significant mediation effect (indirect path p = 0.025; direct path p = 0.239; total p = 0.005). Source data are provided as a Source Data file. ReHo regional homogeneity, PiB Pittsburg compound B, DVR distribution volume ratio, ERC entorhinal cortex, FTP flortaucipir, SUVR standardized uptake value ratio

Fig. 4

Higher hipp-ReHo is associated with functional isolation of the hippocampus. a Hipp-ReHo and hipp-FCI, a measure of hippocampal connectivity averaged over the entire cortex, were negatively correlated (r = −0.30, p = 0.004; all r and p-value pairs reflect Pearson correlation). b Hipp-ReHo was negatively correlated to hipp-FC-AT, a measure of hippocampal connectivity to AT regions (r = −0.28, p = 0.007). c There was no relationship between hipp-ReHo and hipp-FC-PM (r = 0.07, p = 0.49). d Hipp-ReHo negatively predicted mean ReHo across all cortical ROIs (r = −0.41, p < 0.001). e Hipp-ReHo was positively correlated to mean ReHo in AT regions (r = 0.68, p < 0.001). f Hipp-ReHo negatively predicted mean ReHo in PM regions (r = −0.22, p = 0.042). g Across OA participants, the correlation between hipp-ReHo and hippocampal FC to each cortical ROI indicated that increasing hipp-ReHo was related to decreasing FC between hippocampus and temporal cortex. h Across OA participants, hipp-ReHo covariance to ReHo in each cortical ROI showed that higher hipp-ReHo was associated with increased ReHo in temporal cortex. Source data are provided as a Source Data file. hipp-FCI hippocampal functional connectivity index, hipp-ReHo hippocampal ReHo, Hipp-FC hippocampal functional connectivity, AT anterior temporal, PM posterior medial, hipp-FC-AT hippocampal FC to tau-vulnerable AT regions, hipp-FC-PM hippocampal FC to Aβ-vulnerable PM regions

Fig. 5

Hippocampal FC is related to tau outside the MTL. a Hipp-FC-AT, measure of hippocampal connectivity to AT regions, negatively predicted IT FTP SUVR (r = −0.29, p = 0.007; all r and p-value pairs reflect Pearson correlation;). Hipp-FC-AT remained a significant predictor of FTP SUVR in IT (p = 0.04; multiple regression) after adjusting for age, sex, hippocampal volume and hipp-ReHo. b Hipp-FC-AT was a negative predictor of AT Tau SUVR (r = −0.27, p = 0.01). After adjusting for covariates, hipp-FC-AT predicted AT FTP SUVR at statistical trend level (p = 0.06; multiple regression). c, d There was no relationship between Hipp-FC-PM and IT FTP SUVR (r = −0.03, p = 0.82) or AT FTP SUVR (r = −0.01, p = 0.93). Source data are provided as a Source Data file. IT inferior temporal cortex, FTP flortaucipir, SUVR standardized uptake value ratio, AT anterior temporal, PM posterior medial, hipp-FC-AT hippocampal functional connectivity to tau-vulnerable AT regions, hipp-FC-PM hippocampal functional connectivity to Aβ-vulnerable PM regions

Fig. 6

Hipp-ReHo predicts episodic memory but not working memory or executive function. a Higher hipp-ReHo predicted worse episodic memory performance (r = −0.29, p = 0.01; all r and p-value pairs reflect Pearson correlation;). In a multiple regression, hipp-ReHo remained a significant predictor of memory (p = 0.046) after adjusting for age, sex, hippocampal volume and global PiB DVR, but was trending (p = 0.094) in a model that included ERC FTP SUVR, which was also trending (p = 0.070). b, c There was no significant relationship between hipp-ReHo and working memory (r = 0.05, p = 0.63) or executive function (r = −0.13, p = 026). Source data are provided as a Source Data file. ReHo regional homogeneity, hipp-ReHo hippocampal ReHo, PiB Pittsburgh compound B, DVR distribution volume ratio