Understanding the molecular basis of resilience to Alzheimer's disease

Abstract

The cellular and molecular distinction between brain aging and neurodegenerative disease begins to blur in the oldest old. Approximately 15-25% of observations in humans do not fit predicted clinical manifestations, likely the result of suppressed damage despite usually adequate stressors and of resilience, the suppression of neurological dysfunction despite usually adequate degeneration. Factors during life may predict the clinico-pathologic state of resilience: cardiovascular health and mental health, more so than educational attainment, are predictive of a continuous measure of resilience to Alzheimer's disease (AD) and AD-related dementias (ADRDs). In resilience to AD alone (RAD), core features include synaptic and axonal processes, especially in the hippocampus. Future focus on larger and more diverse cohorts and additional regions offer emerging opportunities to understand this counterforce to neurodegeneration. The focus of this review is the molecular basis of resilience to AD.

Keywords: aging; cognition; computational models; dementia; machine learning; neuropathologic lesion; proteomic analysis.

Figure 1

Diagram of the hypothetical relationships among environmental and social stressors, brain aging, and neurodegenerative disease(s). While time passes for all of us, two sources of cellular and molecular stress occur in everyone’s brain to varying extents: episodic environmental and social stressors (layer 1, green) and the biology of aging with attendant reduced fitness for all cell types in brain (layer 2, blue). Biology of aging progressively amplifies with advancing age, although at varying rates across individuals and across organs within an individual. On their own, mounting stressors from these two layers appear insufficient to cause neurodegenerative disease(s). Addition of a third source (layer 3, red)—the neurodegenerative etiologic factors underlying injury plus response to injury—initiates additional cellular and molecular damage, leading to degeneration as signaled by disease-specific formation of hallmark pathologic lesions, including amyloid beta plaques and tau tangles for AD. It is important to recognize that damage may contribute to degeneration through pathways that are dependent or independent of hallmark lesion formation. Sufficiently severe degeneration ultimately expresses as clinically detectable dysfunction. Because this direct model of damage to degeneration to dysfunction is insufficient to account for all observations in humans and experimental animals, two impedance terms are required: resistance and resilience. Resistance signifies suppressed damage despite usually adequate stressors (left), and resilience signifies suppressed functional impairment (right) despite usually adequate degeneration. Examples are provided for some resistance and resilience factors; see text for details.

Figure 2

A framework that quantitatively defines resilience and its relationship to other cognitive terms suggest minimal overlap between mechanisms underlying damage and resilience. (A) Hypothetical relationships between increasing damage as an individual ages and how it impacts existing cognitive reserve, resilience, compensation, and cognitive function (left graph). The right graph shows an example of resistance with minimal damage across the lifespan and therefore no reduction in cognitive function. (B) Potential scenarios of cognitive impairment depending on the different quantities of damage, compensation, and reserve. (C) Correlations between estimated damage and cognitive resilience (CR score) from Eq. (1) stratified by cognitive tests. (D) Correlations between presence of APOE ε4 and damage or CR score stratified by cognitive tests. (E) Actual CR scores vs. CR scores predicted by clinical features. (F) Correlation between CR scores from number of story recall or cognitive status with educational attainment. (G) Correlation between CR scores from number of story recall or cognitive status with geriatric depression score (GDS).

Figure 3

(A) Samples (N = 155) from up to four matched brain regions were donated by 43 research participants who were assigned to three clinico-pathologic groups: healthy control (HC), cognitive resilience to AD (RAD), or AD dementia (ADD). Samples were quantified by data independent tandem mass spectrometry and data analyzed by differential expression and co-expression network analyses. (B) Illustration of differential expression analysis and summary of the final number of RAD-associated differentially expressed proteins (RAD DEPs). (C) Consensus protein co-expression analysis identified 9 modules across four brain regions. Pearson correlation with two-sided p-values was used to evaluate the relationships between clinico-pathologic groups and eigenprotein expression. (D) Module 5 (M5) eigenprotein expressions in HC, RAD, and ADD for the study set. For the boxplots, the interior horizontal line represents the median value, the upper and lower box edges represent the 75th and 25th percentile, and the upper and lower bars represent the 90th and 10th percentiles, respectively. (E) Top enriched GO biological process terms in M5 and their enrichment analysis and the corresponding Q-values (with FDR B&H method) and number of proteins hit in query. (F) Using Aβ abundance in HIPP and PA1B3 concentration in IPL to distinguish RAD from other groups. The number of samples in the study cohort: HC = 11, RES = 12, ADD = 20. CAUD, caudate; HIPP, hippocampus; IPL, inferior parietal lobule; SMTG, superior and middle temporal gyrus; HC, healthy control; RAD, resilience to AD; ADD, AD and dementia; DLPFC, dorsolateral prefrontal cortex; PC, precuneus; GO, gene ontology. ^*p < 0.05; ^**p < 0.01; ^***p < 0.001; ^****p < 0.0001; ns, not significant.

Figure 4

Cellformer deconvolutes epigenetic bulk expression into cell type–specific expression enabling an unprecedented open chromatin profiling of RAD. (A) Cellformer was fed data from comorbidity-free bulk samples from individuals with clinico-pathologic characterization as normal control (NC), Resilient to AD (RAD), and AD dementia (ADD). Three brain regions were used per individual to gain insight into the regional and cellular epigenetic profile of RAD. Cellformer generated cell type–specific expression for 6 main cell types across the whole genome, leading to an unprecedented chromatin profiling of RAD. (B) Cell type–specific open chromatin region (OCR) between RAD and ADD/NC were mainly found in HIPP (93%) and distributed between microglia (28%), and neuron cells (55%) (adjusted p < 0.05, logFC >0.5). (C) Number of OCRs (x-axis) differentially upregulated and downregulated in RAD compared to ADD/NC across cell types (y-axis). (D) GO enrichment applied to RAD-specific OCR (FDR 5%).