Aging is associated with significant alterations in brain structure, blood flow, and metabolism, leading to declines in cognitive function and increased risk for neurodegenerative diseases. The most notable structural change is brain atrophy, particularly in the hippocampus and other regions associated with memory and cognitive processes (Raz and Rodrigue, 2006; Fjell and Walhovd, 2010). Aging also leads to a decline in white matter (WM) integrity through various processes, including myelin degradation, and the development of white matter lesions (Molloy et al., 2021). Cerebral blood flow (CBF) is another critical aspect of brain function that changes with age. Older adults typically exhibit reduced CBF, which can negatively impact cognition (Tarumi and Zhang, 2018). Aging also affects brain metabolism, with older individuals displaying reduced glucose and oxygen metabolic rates, and the production of essential brain metabolites (Lin et al., 2014; Brinton et al., 2015; Mattson and Arumugam, 2018; Turner et al., 2022).

Accumulating evidence has implicated the gut microbiome in brain health (Cox and Weiner, 2018). Gut dysbiosis is associated with many neurological disorders, including Alzheimer’s disease, stroke, and traumatic brain injury (Vogt et al., 2017; Borsom et al., 2020; Yanckello et al., 2022a; Hammond et al., 2022b). Aging is associated with significant alterations in the gut microbiome (Hoffman et al., 2017). One prominent age-related change is the decline of beneficial bacteria, such as Bifidobacteria, and an increase in opportunistic pathogens like bacteria from the phylum Proteobacteria (Biagi et al., 2010). This shift may contribute to chronic low-grade inflammation, known as “inflammaging,” which has been linked to various age-related conditions, including cardiovascular diseases, cognitive decline, and metabolic disorders (Franceschi and Campisi, 2014). Additionally, older adults often experience reduced production of short-chain fatty acids (SCFAs) due to changes in the gut microbiota (Badal et al., 2020). SCFAs play essential roles in maintaining gut health, immune function, and metabolism. The decline in SCFAs production may contribute to age-related immune dysregulation and increases susceptibility to infections and chronic diseases (Buford, 2017).

There are other factors that may impact gut-brain interactions in aging, such as apolipoprotein E (APOE) variance, diet, body mass index (BMI), obesity, and diabetes. Individuals with APOE ε4 alleles (APOE4) display altered brain function, such as reduced amyloid-beta clearance (Castellano et al., 2011) and increased inflammation (Shi et al., 2017). Recent studies show connections between the APOE4 genotype and gut microbiome composition, with APOE4 carriers exhibiting increased pro-inflammatory microbes (Hoffman et al., 2019; Tran et al., 2019; Parikh et al., 2020; Yanckello et al., 2022b). Diet, BMI, and diabetes all have significant impacts on brain function and the gut microbiome (Asadi et al., 2022; Wachsmuth et al., 2022). A diet high in sugar and saturated fat can contribute to inflammation in the brain, impairing cognitive function and increasing the risk of dementia (Gómez-Pinilla, 2008). Additionally, obesity and diabetes are associated with decreased cognitive performance, and an altered gut microbiome, which can contribute to inflammation and neurodegenerative diseases (Arnoldussen et al., 2014; Gomes et al., 2014).

While there is growing evidence to support a link between the gut microbiome and brain health in aging, the specific associations between microbiome composition, brain imaging features, APOE genotype, BMI, diabetes, and dietary intake remain unclear. To shed light on these relationships, we performed an analysis of 30 community-swelling older adults. Our study aims to establish a foundation for understanding how gut microbes are associated with brain health. These findings could have important implications for developing precision nutrition interventions to optimize healthy brain aging older adults.

We conducted a study on the gut microbiome of 30 older adults who reside in the community. Figure 6 presents a summary of our findings, which indicate a strong association between microbiome composition in this group and several factors including APOE genotype, BMI, obesity, and diabetes, as well as brain imaging markers obtained from brain volume, CBF, metabolites, and WMI. Our analysis reveals that APOE2 and APOE4 carriers had a very distinctive microbiota composition are consistent with a number of studies examining gut microbiome in murine models (Weng et al., 2019; Parikh et al., 2020; Zajac et al., 2022). Specifically, the APOE4 genotype had relatively lower abundance of Roseburia, but higher in Eubacterium eligens and Faecalibacterium prausnitzii. On the other hand, the APOE2 genotype had higher levels of Lachnoclostridium species and lower levels of Eubacterium eligens, Oscillibacter, and Faecalibacterium prausnitzii (Chung et al., 2017). Roseburia, a well-known butyrate producer that is associated with a decreased risk of cognitive decline, was found to be decreased in the APOE4 genotype (Jin et al., 2021). This is an interesting finding as APOE2 is protective against Alzheimer’s disease (AD) while APOE4 increase the risk (Yamazaki et al., 2019).

While the functions of some species identified between APOE2 and APOE4 carriers in this study still need to be explored, our results align with existing literature, suggesting that APOE variants may shape the structure of the gut microbiome (Weng et al., 2019; Parikh et al., 2020; Zajac et al., 2022). In earlier work, we demonstrated that asymptomatic APOE4 carriers experience more pronounced gut dysbiosis than non-carriers, including those with APOE e3 alleles (APOE3), and prebiotic diet may mitigate the risk for AD by reducing the gut dysbiosis in the APOE4 mice (Hoffman et al., 2019; Yanckello et al., 2021). Other research has shown associations between APOE genotype differences and significant variations in the relative abundance of families like Prevotellaceae and Ruminococcaceae, as well as several butyrate-producing genera (Tran et al., 2019). A recent investigation even revealed that disruptions in gut microbiome composition can modify immune pathways, leading to the tauopathy underlying Alzheimer’s disease in an APOE genotype-dependent manner (Seo et al., 2023). These insights suggest that microbiome composition and diversity may play a substantial role in AD development, influenced by variations in APOE genetics. Moving forward, future studies should delve deeper into the differential effects of APOE-genotype dependent variations in the microbiome.

We found that obesity was inversely associated with Akkermansia muciniphila, which has previously been found to be decreased in obesity and is believed to be protective against metabolic disorders through its excretion of endocannabinoids that control inflammation, gut barrier function, and gut peptide secretion (Everard et al., 2013). Further, proinflammatory state was associated with higher BMI and diabetes. Diabetes was associated with increased levels of E. coli and Parabacteroides goldsteinii. While E. coli is a normal commensal gut bacteria, it has many pathologic variants that can cause disease (Leimbach et al., 2013), and increases in infectious E. coli have previously been reported in diabetes (Suri et al., 2009) and diabetes-related neurological disorders, such as stroke (Hammond et al., 2022a,b).

Our findings suggest that a healthy diet is strongly associated with a microbiota that produces SCFAs, leading to an anti-inflammatory state. We used the Dietary Screener Questionnaire from the National Health and Nutrition Examination Survey to measure effects of the diet on the microbiome. Interestingly, we observed an inverse relationship between calcium and Clostridium asparagiforme, a species that has been found to be reduced in smokers with hypertension (Wang et al., 2021). Conversely, we found that calcium was positively associated with Acidaminococcaceae, Phascolarctobacterium and Eubacterium eligens. The former two taxa are known to produce SCFAs (Wu et al., 2017), while Eubacterium eligens has anti-inflammatory effects on the host (Chung et al., 2017). Although the precise mechanisms underlying the effects of calcium on these microbial taxa remain unknown, previous research has shown that a high calcium diet increases Firmicutes species and levels of lactic, acetic, and butyric acid (Fuhren et al., 2021). We also observed a positive relationship between vegetable intake and Agathobaculum butyriciproducens, a butyrate producer, which has been shown to improve performance in behavioral tests, decrease Aβ plaque deposition and microglial activation APP/PS1 transgenic mice (Go et al., 2021). A negative effect with Eggerthella lenta, an infectious species, was found in our data (Gardiner et al., 2015; Go et al., 2021). It is interesting that fiber, whole grains, added sugars, dairy, and fruit intake did not have any significant associations with the microbial taxa in our data. Many studies have highlighted the beneficial effects of a vegetarian diet on the microbiome by limiting the abundance of inflammatory species and promoting SCFA-producing species (Tang et al., 2019; Sakkas et al., 2020). Other studies have investigated the effects of Western, Mediterranean, vegetarian, high carbohydrate, high fat, whole food, ultra-processed, and low-calorie sweetener diets on the microbiome (Dahl et al., 2020). Specific Dietary effects on the microbiome should be explored in future studies.

The results also imply that strengthening the immune barriers of the gut were associated with the volume of brain structures related to processing high levels of information. Bacteroides ovatus is related to the thalamus volume. The thalamus is an important relay station in the brain that communicates information from the spinal cord and cerebellum to the cerebral cortex (Sherman, 2007). Bacteroides ovatus is a commensal microbe that uses various carbohydrates and proteins as its fuel source (Fultz et al., 2021), is a prominent inducer of IgA (Yang et al., 2020), and promotes the reparative cytokine IL-22 (Ihekweazu et al., 2021), thereby stimulating epithelial recovery (Ihekweazu et al., 2019). The family Acidaminococcaceae is related to the volume of the mid posterior portion of the corpus callosum. The region is responsible for connecting portions of the parietal and temporal cortices on each side of the brain. This is consistent with a previous finding that Acidaminococcaceae is associated with memory performance (Khine et al., 2020).

Our results also indicated that microbiota associated with SCFAs production are highly associated with less vascular disease in the brain. Bacteroides uniformis correlates inversely with white matter hypointensities (WMH), which are a marker of vascular disease in the brain (Graff-Radford et al., 2019) and are often associated with cognitive impairment (Hu et al., 2021). Bacteroides uniformis prefers wheat bran extract as a fuel source and is a butyrate producer; it strengthens the first line of immune defense against unhealthy diets (Lopez-Almela et al., 2021) and improves glucose tolerance (Fabersani et al., 2021). This microbe has also been associated with increased dopamine transporters (Hartstra et al., 2020) and normalizing the brain reward response to reduce anxiety in rats (Agusti et al., 2021). Interestingly, the microbiota-derived phenylacetylglutamine has previously been associated with the amount of WMH in the brain (Yu et al., 2021).

Our data also suggests that gut microbiota that produce GABA, the main inhibitory neurotransmitter in the brain (Petroff, 2002), are highly associated with the amount of GABA present in the corpus callosum. Collinsella aerofaciens is negatively associated with GABA production, and Parasutterella excrementihominis is positively associated with GABA production. Collinsella aerofaciens is an obligate anaerobe (Bag et al., 2017; Qin et al., 2019) and a proinflammatory species associated with Crohn’s disease (Joossens et al., 2011; Kalinkovich and Livshits, 2019). Parasutterella excrementihominis is strictly anaerobic (Nagai et al., 2009) and has been associated with impaired GI health (Fart et al., 2020). While Collinsella aerofaciens and Parasutterella excrementihominis have not previously been identified as GABA producers or consumers, Bacteroides has been identified as a dominant GABA producer and Pseudomonas as a prominent GABA consumer (Strandwitz et al., 2019).

It was observed that microbial species Ruminococcus lactaris, a producer of SCFAs, is associated with the brain metabolites related to mitochondrial function and cognition. NAA correlates with neuronal mitochondrial function and survival (Paslakis et al., 2014), and NAAG has procognitive properties (Neale and Olszewski, 2019). Ruminococcus lactaris is a butyrate producer (Sato et al., 2021) and is negatively correlated with the inflammatory cytokine IL-8 (Shintouo et al., 2020). A previous group found a link between Ruminococcus and NAA thought to be mediated by Ruminococcus decreasing cortisol, which impacts brain NAA (Mudd et al., 2017).

Our analysis indicates a positive correlation between Collinsella stercoris, a producer for conjugated bile acid, with CBF. Previous research has shown that the conjugated bile acid Tauroursodeoxycholic acid (TUDCA) can alleviate early brain injury by reversing cerebrovascular dysfunction and ER-stress-mediated apoptosis (Chen et al., 2020). TUDCA was found to enhance neurological function and treated SAH-related cerebrovascular dysfunction by increasing cerebral cortical perfusion and decreasing blood–brain barrier permeability. These reports are in line with our findings that presence of Collinsella stercoris may be beneficial for maintaining CBF. We further showed that microbiota related to equol producers are highly associated with WMI that connecting brain regions associated with language and memory. The superior temporal sulcus is commonly referred to as Wernicke’s area and is a central area for speech recognition and processing (Nourski et al., 2021). The left pars triangularis is commonly referred to as Broca’s area and is a central area for speech production (Foundas et al., 1996) Members of the family Eggerthellaceae are able to convert the isoflavone daidzein (a soy product) into equol, an estrogen (Soukup et al., 2021). Specific members of this family, including Adlercreutzia equolifaciens, Asaccharobacter celatus, and Gordonibacter pamelaeae all have negative associations with these same areas responsible for language circuits, in addition to areas of the brain responsible for memory, learning, and reward processing. Equol supplementation has been shown to improve long and short-term memory in rats by increasing brain antioxidant activity and improving blood pressure (Liu and Tsai, 2016; Caliskan et al., 2021).

Collectively, the findings in the present study indicated that microbiome composition in aging may play an important role in determining brain metabolism and integrity, and APOE status could also be a key player. This is in line with studies showing that metabolism may play a more critical role in driving neurodegeneration than amyloid beta plaques and tau tangles in the progression of Alzheimer’s disease (AD) (Hammond et al., 2020; Hammond and Lin, 2022), and that APOE4 carriers and non-carriers develop AD through different metabolic pathways (Hammond et al., 2021). Therefore, it will be critical for to identify interventions that can promote healthy microbiome in aging, taking into consideration of APOE genotype, to protect brain function in aging and potentially mitigate AD risk for APOE4 carriers. This is in good agreement with our recent studies showing that supplementing a diet with the prebiotic inulin, a fermentable prebiotic fiber, can positively impact gut microbiome composition, boost the production of SCFAs, enhance mitochondrial function, and decrease neuroinflammation in young, asymptomatic APOE4 mice (Hoffman et al., 2019). Our results also demonstrate that these effects of dietary inulin supplementation revealed an APOE genotype-dependent response (Yanckello et al., 2021). These findings are also consistent with what we observed in the current study that higher fiber (vegetable) intake can improve microbiome in aging. Our study further provides other potential interventions in the future, such as equol or butyrate supplementation, as well as probiotic interventions with butyrate and equol producers like Roseburia, which have shown promise as easily implemented, inexpensive tools to optimize healthy brain aging.

While this study provides further understanding between gut microbiome and brain health in the aging population, it has many limitations. The observational nature of this study helps us to make many associations between gut bacteria and markers of brain health on imaging, but we are unable to determine causation. Additionally, due to the pandemic, we were only able to recruit 30 participants for the study. The relatively small sample size for a human study with a heterogeneous population potentially leaves the study underpowered to detect smaller associations and changes in the microbiome. Future work should utilize a larger sample size to allow adjusting for potential confounders and to identify which demographic factors (e.g., BMI, gender, age, APOE genotype) most strongly associate with the composition of the microbiome. Future work should investigate the long-term nature of the gut microbiome in this population. Future studies can also further include other more advanced MRI methods to measure brain oxygen metabolism or perivascular space, which will provide further insight for brain metabolism and inflammation, respectively (Lin et al., 2008, 2010; Feldman et al., 2018; Langan et al., 2022).

In summary, we demonstrated the significant correlations of microbiome in older adults with their brain imaging markers, APOE genotypes, calcium intake, vegetable intakes, diabetes, and obesity. The findings could have important implications in the future for promoting brain health through microbiome modulation and developing precision nutrition interventions to optimize healthy brain aging older adults.

The datasets presented in this article are not publicly available due to national legislation that restricts the deposition of human data. Requests to access the datasets should be directed to A-LL ai-ling.lin@health.missouri.edu.

The studies involving humans were approved by Institutional Review Board at the University of Kentucky. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

TH, AS, and A-LL designed the research. TH, SG, GC, AC, and A-LL conducted the research. TH, AC, YJ, XX, and AS performed the statistical analysis. TH, SG, GC, YJ, SH, AC, CA, XX, AF, AS, PB, and A-LL wrote the manuscript. TH, SG, AS, and A-LL had primary responsibility for final content. All authors contributed to the article and approved the submitted version.

This research was supported by grants from NIH/NIA, NIH/ODS, and American Federation for Aging Research to A-LL (R01AG054459 and RF1AG062480) and TH (T32AG057461).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.