Blood metabolites reveal lifestyle links to brain health before dementia

by · News-Medical

A large analysis suggests that blood metabolites may capture how smoking, BMI, medication use, gut microbes, and other exposures intersect with cognition and brain structure years before dementia emerges.

Review: Beyond estrobolome 1.0: unraveling endocrine-microbiome axis as a driver and therapeutic target in hormone-driven cancers. Image Credit: Lightspring / Shutterstock

Alterations in peripheral metabolism have been increasingly implicated in brain diseases, including Alzheimer’s disease (AD). Studies have linked blood metabolites, such as lipids, lipoproteins, and amino acids, to AD-related phenotypes. Functional and structural changes in the brain, as well as changes in neuropsychological markers, appear years before AD manifestation. Specifically, brain atrophy and white matter hyperintensities are neurodegenerative and vascular markers associated with AD.

Studying early metabolic changes related to these features may help identify metabolites critical to disease etiology. Besides genetics, the exposome (lifestyle factors, medication use, and other environmental exposures) and the gut microbiome can influence metabolite levels, and their management offers opportunities to stabilize metabolism and counter metabolic changes associated with disease.

The study and findings

In the present study, researchers evaluated the association of blood metabolites with general cognition and MRI markers in dementia-free middle-aged adults. First, plasma levels of 1,387 metabolites were measured in 1,082 participants of the Rotterdam study. Next, the team assessed the cross-sectional relationships of 991 frequent metabolites with general cognition and MRI markers in 1,068 participants after data preprocessing. This revealed significant associations between 14 metabolites and cognition.

Increased levels of uridine, 2-deoxyuridine, ergothioneine, and two uncharacterized metabolites, and reduced levels of seven sulfated xenobiotics and two uncharacterized metabolites were associated with improved cognition. After adjustment for education, 12 of these 14 associations remained significant; after further adjustment for smoking, diabetes, and hypertension, all 14 remained nominally associated, but no longer met FDR-adjusted significance.

Regarding MRI markers, S-adenosylhomocysteine was associated with white matter lesion (WML) volume, while 21 metabolites were associated with total brain volume. No individual metabolite was significantly associated with hippocampal volume after multiple-testing correction.

Higher levels of 6-bromotryptophan, glycerophosphorylcholine, argininate, X-11787, and three sphingomyelins were associated with increased brain volume. Meanwhile, lower levels of N-lactoyltyrosine, six metabolites related to caffeine degradation, and seven intercorrelated metabolites, including hydroxylated acylcarnitines and a hydroxylated dicarboxylic fatty acid, were associated with elevated brain volume. The team also conducted a replication analysis in an independent sample of 847 older adults from the Rotterdam study. In this cohort, nine of 14 metabolites were significantly associated with cognition.

A second replication analysis was conducted in the Alzheimer Gut Microbiome Project (AGMP) cohort of 512 participants; associations with three cognitive scores were investigated: the Uniform Data Set (UDS) Benson Figure Total Delayed score, the National Alzheimer’s Coordinating Center (NACC) Montreal Cognitive Assessment (MoCA) score, and the Craft Story Delayed Recall (CRAFTDRE) score. Seven metabolites were significantly associated with at least one cognitive score. Across the RSI-4 and AGMP replication analyses, all 14 cognition-associated metabolites were significant in at least one replication test.

The researchers also compared the metabolite signature of cognition with incident AD in the older RSI-4 cohort, finding strong concordance between metabolites linked to cognition and those linked to later AD diagnosis.

Sex-stratified analyses revealed suggestive evidence of interactions with sex for multiple metabolites associated with MRI markers and general cognition. In particular, reduced levels of five metabolites, including glycocholate and taurocholate, were significantly associated with improved cognition in females only. Conversely, higher N-acetyl-aspartyl-glutamate was associated with improved cognition in males only. However, these sex-stratified findings were exploratory and did not replicate in the older RSI-4 cohort.

Next, the team investigated how much variance of metabolites was explained by genetic, lifestyle, microbial, clinical, and medication features. Among cognition-related metabolites, lifestyle features explained a substantial part of the variance of nine metabolites. Likewise, clinical, medication, and lifestyle features were the major factors affecting the variance of metabolites associated with WML or total brain volume.

The researchers also assessed associations of cognition- and MRI marker-related metabolites with clinical features, lifestyle factors, gut microbiota, and medication use using regression analyses. This revealed a significant association of 13 (out of 14) cognition-related metabolites with distinct lifestyle factors. Specifically, smoking was associated with increased levels of sulfated metabolites and lower levels of 2′-deoxyuridine and uridine.

In addition, observationally increased ergothioneine levels, linked to improved cognition, were associated with lower body mass index (BMI), higher alcohol intake, and higher education. Meanwhile, lower ergothioneine levels were associated with antacid use. Among medications and clinical factors, antidiabetic medication and diabetes showed the most associations. BMI and alcohol intake showed the most associations with metabolites linked to MRI markers.

In particular, 20 metabolites linked to MRI markers were associated with alcohol intake or BMI. Among metabolites associated with MRI markers and cognition, 22 were significantly associated with specific gut microbiota. For instance, higher ergothioneine was associated with increased abundance of 12 microbial genera, including six genera that showed positive associations, such as Fusicatenibacter and Romboutsia. Mediation analyses revealed that ergothioneine mediated 31.5% of the negative association between antacid use and cognition in an exploratory cross-sectional mediation analysis. The authors cautioned that this does not establish causality and that confirmation is required in longitudinal and mechanistic studies before any clinical prescribing implications can be drawn.

Conclusions

Taken together, the findings suggest that lifestyle factors play a substantial role in shaping blood metabolites linked to cognition and MRI markers in middle-aged people without dementia. Smoking was identified as a crucial lifestyle factor associated with cognition-associated metabolites. Meanwhile, BMI, antidiabetic medication, diabetes, and alcohol intake were associated with metabolites linked to MRI markers. Overall, the results could inform future prevention and intervention efforts, but they do not prove that modifying any single factor will improve brain outcomes.

Want to read later? Download your PDF copy by clicking here.

Journal reference: