Date Published: March 21, 2017
Publisher: Public Library of Science
Author(s): Stuart G. Snowden, Amera A. Ebshiana, Abdul Hye, Yang An, Olga Pletnikova, Richard O’Brien, John Troncoso, Cristina Legido-Quigley, Madhav Thambisetty, Carol Brayne
Abstract: BackgroundThe metabolic basis of Alzheimer disease (AD) pathology and expression of AD symptoms is poorly understood. Omega-3 and -6 fatty acids have previously been linked to both protective and pathogenic effects in AD. However, to date little is known about how the abundance of these species is affected by differing levels of disease pathology in the brain.Methods and findingsWe performed metabolic profiling on brain tissue samples from 43 individuals ranging in age from 57 to 95 y old who were stratified into three groups: AD (N = 14), controls (N = 14) and “asymptomatic Alzheimer’s disease” (ASYMAD), i.e., individuals with significant AD neuropathology at death but without evidence for cognitive impairment during life (N = 15) from the autopsy sample of the Baltimore Longitudinal Study of Aging (BLSA). We measured 4,897 metabolite features in regions both vulnerable in the middle frontal and inferior temporal gyri (MFG and ITG) and resistant (cerebellum) to classical AD pathology. The levels of six unsaturated fatty acids (UFAs) in whole brain were compared in controls versus AD, and the differences were as follows: linoleic acid (p = 8.8 x 10−8, FC = 0.52, q = 1.03 x 10−6), linolenic acid (p = 2.5 x 10−4, FC = 0.84, q = 4.03 x 10−4), docosahexaenoic acid (p = 1.7 x 10−7, FC = 1.45, q = 1.24 x 10−6), eicosapentaenoic acid (p = 4.4 x 10−4, FC = 0.16, q = 6.48 x 10−4), oleic acid (p = 3.3 x 10−7, FC = 0.34, q = 1.46 x 10−6), and arachidonic acid (p = 2.98 x 10−5, FC = 0.75, q = 7.95 x 10−5). These fatty acids were strongly associated with AD when comparing the groups in the MFG and ITG, respectively: linoleic acid (p < 0.0001, p = 0.0006), linolenic acid (p < 0.0001, p = 0.002), docosahexaenoic acid (p < 0.0001, p = 0.0024), eicosapentaenoic acid (p = 0.0002, p = 0.0008), oleic acid (p < 0.0001, p = 0.0003), and arachidonic acid (p = 0.0001, p = 0.001). Significant associations were also observed between the abundance of these UFAs with neuritic plaque and neurofibrillary tangle burden as well as domain-specific cognitive performance assessed during life. Based on the regional pattern of differences in brain tissue levels of these metabolites, we propose that alterations in UFA metabolism represent both global metabolic perturbations in AD as well as those related to specific features of AD pathology. Within the middle frontal gyrus, decrements in linoleic acid, linolenic acid, and arachidonic acid (control>ASYMAD>AD) and increases in docosahexanoic acid (AD>ASYMAD>control) may represent regionally specific threshold levels of these metabolites beyond which the accumulation of AD pathology triggers the expression of clinical symptoms. The main limitation of this study is the relatively small sample size. There are few cohorts with extensive longitudinal cognitive assessments during life and detailed neuropathological assessments at death, such as the BLSAConclusionsThe findings of this study suggest that unsaturated fatty acid metabolism is significantly dysregulated in the brains of patients with varying degrees of Alzheimer pathology.
Partial Text: Alzheimer disease (AD) is a neurodegenerative disorder characterised by progressive cognitive decline, with impairment in multiple cognitive domains including memory, executive function, and language . AD accounts for between 60% and 80% of total dementia cases worldwide  and represents a major cause of global morbidity and mortality. It is currently estimated that there are over 46 million people suffering from the disease worldwide, with the number of patients estimated to rise to 131.5 million by 2050 . As well as a major human cost, dementia also imposes a significant economic impact costing US$818 billion in 2015, and estimated to rise to US$1 trillion by 2018 .
The clinical characteristics of the three diagnostic groups analysed are summarised in Table 1. The three groups did not differ significantly in age at death, postmortem interval, sex, or APOE ε4 status. In this study, a total of 4,897 metabolite features were measured, 3,482 by LC-MS and 1,415 by GC-MS. Of these measured metabolite features, 126 were successfully annotated, representing 100 structurally distinct metabolites (S1 Table, S2 Table).
The metabolic basis of vulnerability to AD pathology and the subsequent expression of symptoms of AD are poorly understood. In this study, we applied mass spectrometry-based metabolomics to human brain tissue samples from a well-characterized longitudinal cohort, i.e., the BLSA, to address this issue. Our results suggest that perturbations in brain UFA metabolism are closely related to AD pathogenesis. To the best of our knowledge, this report is the first to measure brain tissue levels of UFAs to demonstrate a relationship with both severity of AD pathology and the expression of AD symptoms. Including brain tissue samples from “ASYMAD” individuals who represent an intermediate group in the gradation of neuropathology from controls to AD patients in the absence of cognitive impairment during life allowed us to relate measures of brain UFAs to incremental levels of AD pathology and symptom expression. Importantly, by measuring UFA levels in brain regions both vulnerable to distinct pathological features of AD, i.e., MFG (amyloid deposition) and ITG (tau accumulation) as well as in a region relatively resistant to AD pathology, i.e., CB, we were able to ask whether the observed alterations in these metabolites were related to AD-defining pathological processes. We can categorize our results broadly as follows: