Diet and APOE as moderators of the relationship between trimethylamine N‐oxide and biomarkers of Alzheimer's disease and glial activation
John C. Thorstenson,Margo B. Heston,Katie R Zarbock,Cynthia M Carlsson,Corinne D. Engelman,Yuetiva Deming,Sterling C. Johnson,Tyler K. Ulland,Sanjay Asthana,Gwendlyn Kollmorgen,Ivonne Suridjan,Kaj Blennow,Henrik Zetterberg,Nathaniel A Chin,Federico E. Rey,Barbara B Bendlin
DOI: https://doi.org/10.1002/alz.051827
2021-12-01
Abstract:Dietary patterns appear to impact cognitive trajectories in aging, and gut microbiota have been implicated in Alzheimer’s disease (AD) pathogenesis, potentially as modulators of neuroinflammation early in the disease. Diets featuring low meat and dairy consumption have been linked to reduced AD risk, and recently, the gut microbial metabolite trimethylamine N‐oxide (TMAO) was found in cerebrospinal fluid (CSF) and linked to CSF biomarkers of AD. Because TMAO is largely derived from dietary sources of choline, carnitine, and betaine, we examined whether these precursors drive the association between TMAO and sTREM2, a marker for glial activation. Additionally, TMAO has been found to inhibit cholesterol metabolism, a strong risk factor for AD, which is further dysregulated by the APOE4 allele. Therefore, we aimed to determine whether TMAO‐glial activation relationships are moderated by APOE4 carrier status. Participants from the Wisconsin Registry for Alzheimer’s Prevention and the Wisconsin Alzheimer’s Disease Research Center provided CSF samples (n=570, Table 1). sTREM2 and YKL‐40 biomarkers were measured with the exploratory Roche NeuroToolKit assays, a panel of robust prototype immunoassays (Roche Diagnostics International Ltd). TMAO, carnitine, choline, and betaine relative abundance were obtained using Metabolon’s UHPLC‐MS/MS metabolomics platform. A subset of participants (n=159) completed the MIND diet questionnaire. Metabolite and biomarker levels were log‐transformed for analysis; models were adjusted for age, sex, and APOE4 carrier status. Linear regression tested associations between intake of TMAO precursor‐containing foods (red meat, butter, cheese, fish) and CSF levels of each TMAO precursor. Path analysis with Satorra‐Bentler adjustments tested whether TMAO mediated precursor‐biomarker relationships. Linear regression tested whether APOE4 carrier status moderated TMAO‐biomarker relationships. Red meat and cheese consumption predicted levels of CSF carnitine (βs=0.009, ‐0.017; ps=0.0609, 0.0007; Figure 1 A, B; respectively). TMAO mediated the relationship between carnitine and sTREM2, although effects were marginal; several individual relationships throughout both path models showed strong associations (Tables 2, 3; Figures 2, 3). APOE4 carrier status did not significantly moderate TMAO‐glial activation relationships. This study suggests that CSF carnitine reflects dietary intake, and may drive the TMAO‐sTREM2 association previously identified. Future studies in animal models are required to confirm these results mechanistically.