Scientists at Washington University in St. Louis have identified specific alterations in gut microbiota composition related to preclinical neuropathology of Alzheimer’s disease. The findings might help identify novel gut-derived biomarkers to predict the disease risk.
The study is published in the journal Science Translational Medicine.
Study: Gut microbiome composition may be an indicator of preclinical Alzheimer’s disease. Image Credit: ArtemisDiana / Shutterstock
The human gut microbiota contains a plethora of microbial communities that are compositionally and functionally district from each other. Gut microbes and metabolites derived from them play crucial roles in regulating many physiological processes, ranging from immune functions to neurological functions.
Gut microbiota dysbiosis is defined as altered or imbalanced composition and diversity of gut microbiota. Evidence suggests a link between gut microbiota dysbiosis and the pathogenesis of Alzheimer’s disease (AD) and other neurodegenerative diseases.
AD is a progressive disease characterized by memory loss and cognitive decline. Imaging studies have shown that AD pathophysiology gradually progresses from preclinical AD to symptomatic AD through a series of neurodegenerative changes. Preclinical AD is characterized by the presence of disease-specific biomarkers without any apparent symptoms. The most prominent clinical biomarkers of AD are pathogenic β-amyloid (Aβ) and tau protein.
In the current study, scientists have examined cognitively normal individuals with and without preclinical AD to determine whether changes in gut microbiota can predict AD risk.
The scientists compared gut microbiota composition and function between cognitively normal individuals with and without markers of early preclinical AD. In total, 164 individuals were examined, with 49 exhibiting preclinical AD biomarkers.
Stool samples collected from the participants were analyzed to identify specific gut microbiota characteristics associated with preclinical AD or known biomarkers of AD. Furthermore, the scientists evaluated whether gut microbiota characteristics can improve the performance of machine-learning models developed to identify individuals susceptible to developing AD.
Preclinical AD status of the participants was determined based on several clinical examinations, including magnetic resonance imaging (MRI), positron emission tomography (PET), cerebrospinal fluid (CSF) analysis for Aβ and tau protein, and cognitive testing. This led to the identification of 115 healthy individuals and 49 individuals with preclinical AD.
The comparison of gut microbiota characteristics between the groups was conducted after adjusting for dietary intake and several clinical covariates, including age, body mass index (BMI), apolipoprotein ε4 (APOE ε4) carrier status (genetic risk factor), diabetes, and hypertension.
Metagenomic sequencing of stool samples was conducted to determine the relative abundance of microbial species and microbial pathways. The findings revealed no significant difference in the Firmicutes/Bacteroidetes ratio between healthy individuals and those with preclinical AD. The gut microbiota alpha diversities were also similar between the groups.
Taxonomic profiles of the gut microbiota showed significant differences between the groups. This finding suggests that the human gut microbiota may change early in AD before the onset of specific symptoms.
Association between gut microbiota profiles and preclinical AD characteristics
The comparison of gut microbiota measures with the amount of Aβ plaques and tau proteins revealed a significant association of gut microbiota taxonomic profiles with Aβ plaques and tau proteins.
However, no such association was observed for neurodegenerative markers. This could be explained by the fact that neurodegeneration occurs later in the clinical course of AD and that Aβ plaques and tau proteins are the early biomarkers to detect preclinical AD.
Specific gut microbiota characteristics in preclinical AD
The analysis of taxonomic and microbial pathway data after adjusting for clinical covariates revealed that certain microbial species, including Dorea formicigenerans, Oscillibacter sp. 57_20, Faecalibacterium prausnitzii, Coprococcus catus, and Anaerostipes hadrus have the most specific association with preclinical AD status.
A total of 13 gut microbial species showed the most specific association with health status. Of them, seven belonged to the Bacteroides genus.
Regarding microbial pathways, arginine and ornithine degradation pathways and glutamate degradation pathways showed the most specific association with preclinical AD status and health status, respectively.
Performance of machine learning models for preclinical AD status
The performance of preclinical AD status prediction models was tested with and without gut microbiota characteristics. In addition to participants’ demographic data and clinical covariates, these models initially included all available AD markers, including Aβ, tau, neurodegeneration, and genetic risk markers.
To determine the highest predictive performance, AD biomarkers, except for genetics, were rationally omitted from the models, leaving behind only demographic characteristics and clinical covariates.
The findings revealed that in models that omitted AD biomarkers except for genetics, the inclusion of gut microbiota taxonomic characteristics caused 6.8% and 27.1% improvements in prediction accuracy and specificity, respectively. Moreover, in models that omitted AD biomarkers, including genetics, the inclusion of taxonomic characteristics caused 11.2 and 13.7% improvements in accuracy and sensitivity, respectively.
These findings indicate that the utility of microbial features as an indicator of preclinical AD increased with greater scarcity of available data for established AD biomarkers.
The study provides global and specific differences in gut microbiota composition between individuals with and without preclinical AD. In other words, the study identifies gut microbiota composition as an indicator of preclinical AD.