Alzheimer's risk patients may benefit from Mediterranean keto diet, study shows gut microbiome changes

By Neha MathurReviewed by Danielle Ellis, B.Sc.Apr 9 2023

In a recent study published in the Journal of the Alzheimer's Association (AA), researchers examined the effects of a low-carbohydrate-modified Mediterranean ketogenic diet (MMKD) on the gut microbiome and metabolome of patients at risk for Alzheimer's disease (AD).

Study: Effects of a ketogenic and low-fat diet on the human metabolome, microbiome, and foodome in adults at risk for Alzheimer's disease. Image Credit: Yulia Furman / Shutterstock

Background

Although AD pathology involves several systemic changes, metabolic dysregulation is a primary change observed throughout AD. However, it is not forthright to identify metabolites associated with AD because its metabolic signatures vary with gender and apolipoprotein E (APOE) genotype.

Studies have shown a correlation between AD incidence and metabolites, such as ketone body β-hydroxybutyrate, valine, and acetone. Additionally, triglycerides are implicated with mild cognitive impairment (MCI) and the onset of AD. Strikingly, the discovery of the gut-brain axis (GBA) has led to the exploration of an altered gut microbiome in AD.

Bile acids are responsible for cholesterol elimination in the brain, accumulation of which could lead to hepatic encephalopathy, a liver disease that increases the risk of developing AD. Since gut microbiota transforms bile acids, it is another fascinating area in AD research. The unifying thread between the gut microbiome, neurocognitive status, and metabolic disorders is their connection with diet.

About the study

In a previous trial, researchers observed that the MMKD augmented levels of amyloid β 42 (Aβ42), and decreased tau levels in the cerebrospinal fluid (CSF). Also, it increased hemoglobin A1c (Hb1Ac), insulin, and triglyceride levels in MCI patients. Additionally, these patients showed increased cerebral perfusion and ketone uptake, both metrics of brain function.

Moreover, the MMKD significantly affected gut microbial populations, including mycobiome and stool metabolites. For instance, MMKD increased the relative abundance of bacterial taxa Enterobacteriaceae, Akkermansia, etc., and Agaricus and Mrakia fungal species in MCI patients. Metabolically, this diet was implicated with augmented butyrate and propionate in the stool.

In the present study, researchers used shotgun metagenomic sequencing and metabolomics to study the same patient samples used in the previous trial. It helped them further characterize the complex relationship between diet, cognitive status, and the gut microbiome and metabolome of MCI patients and cognitively normal (CN) individuals. They extracted deoxyribonucleic acid (DNA) for metagenomic sequencing and metabolites for untargeted tandem mass spectrometry (MS/MS) analysis.

The team used the National Institute on Aging (NIA)-AA guidelines for the clinical detection of MCI patients and AD Neuroimaging Initiative (ADNI) criteria to identify CN patients. Further, they employed a randomized crossover design and let all participants consume either an MMKD or the low-fat American Heart Association Diet (AHAD) for six weeks, the control diet. After a six-week washout, these participants consumed the second diet for six weeks.

Next, the researchers broadly assessed the microbiome, foodome, and metabolome's associations with diet and cognitive status via dimensionality reduction. They updated this method and utilized Bayesian inferential regression to read out particular microbes, foods, and metabolites associated with diet, cognitive status, or both. Finally, the team collected stool samples from participants at the initiation and cessation of each dietary intervention and six weeks after the washout of the diet taken later.

Results

The study cohort comprised 23 adult participants, of which 20 completed the full intervention. The authors found a higher relative abundance of Akkermansia sp. on the MMKD than the AHAD and a similar surge in relative abundance of Dialister and Bacteriodes sp. in CN vs. MCI groups. Additionally, they detected that many subspecies of Akkermansia muciniphila were enriched on the MMKD but not AHAD. Among CN individuals on the MMKD, Dialister invisus, and many Bacteriodes fragilis strains were enriched.

Since this diet improves metabolism and insulin sensitivity, the enrichment of Akkermansia muciniphila was not specifically related to diet composition. It might have become enriched to improve metabolic regulation related to insulin sensitivity.

The Alistipes sp. CAG:514 to Bifidobacterium adolescentis ratio was markedly different between MCI and CN groups, but post-beginning the AHAD, not MMKD, highlighting that dietary intervention can alter the microbiome and downstream metabolites/signaling molecules. This observation was specific to cognitively impaired participants, suggesting that individuals throughout the cognitive spectrum may respond variably to interventions impacting the microbiome. Further, the authors noted that APOE4-positive individuals had more GABA-synthesizing gut microbes and correspondingly more GABA in their CSF.

The direct implications of these gut changes on the central nervous system (CNS) remain unclear. Future studies assessing the microbiome in patients with cognitive impairment and modification by intervention are needed. Clearly, GABA dysfunction in the gut and brain likely occurs differently, especially during the earlier AD stages examined in this study.

Note Alistipes synthesize gamma-aminobutyric acid (GABA); AHAD likely increased GABA production in MCI patients through an increase in GABA-producing Alistipes sp. Conversely, the MMKD regulates GABA production in both MCI and CN individuals via increasing the relative abundance of GABA-regulating Akkermansia muciniphila.

Bile acids regulate cholesterol metabolism. Accordingly, the authors found MMKD-specific associations between bile salt hydrolase (BSH)-containing microbes and some bile acids. However, they did not observe diet-specific associations between broader bile acid classifications. Thus, an individual's diet's fat content or cognitive status most likely had no effect on the bile acid categories, e.g., unconjugated and conjugated bile acids.

Conclusions

Despite several technical challenges, this longitudinal trial used statistical methods that accounted for data sparsity and compositionality and identified diet- and cognitive function-specific microbial and metabolite features. Future studies should more extensively evaluate lipid metabolism, ketogenesis, and other similar processes known to be perturbed by diet during AD.