Gut bacteria and memory loss: how microbes and body responses may relate to dementia

Dementia is a progressive neurodegenerative disorder which is marked with a deterioration of cognitive functions such as memory, orientation, language, and executive functions as well as behavioral changes of agitation, apathy, and aggression which diminishes quality of life (Cerejeira et al., 2012). The most common type of dementia, Alzheimer’s disease (AD), is an increasingly important health issue, particularly with aging populations (Javaid et al., 2021). Neuropathologically, AD is marked by the presence of beta-amyloid (Aβ) plaques and intracellular neurofibrillary tangles of hyperphosphorylated tau, which causes neuroinflammation, synaptic loss, and neurodegeneration (Takahashi et al., 2017). Current evidence positions the gut-brain axis not merely as an associative factor, but as a critical modulator of neurodegenerative progression. This axis functions as a bidirectional communication system linking the gastrointestinal tract and the central nervous system, where neurological, immune, and metabolic pathways interact to maintain brain homeostasis (Loh et al., 2024). Alterations of the gut microbiota, known as dysbiosis, have been associated with the onset and progression of neurodegenerative disorders such as dementia (Zhao et al., 2017; Cattaneo et al., 2017; Kim and Jazwinski, 2019). Experimental studies indicate that the microbiota modulates the brain’s neurochemical response and that an imbalance in the gut flora can impair emotional and cognitive balance, contributing to the development of cognitive and behavioral symptoms typical of dementia (Cryan and O'Mahony, 2011). Intestinal dysbiosis has been observed in patients with dementia, including AD, and is believed to contribute to systemic and brain inflammation, promoting neuronal damage. Gut bacteria can modulate the production of pro-inflammatory cytokines, which trigger neuroinflammation—a key process in neurodegenerative diseases (Zhao et al., 2017; Vogt et al., 2017; Liang et al., 2018). Zhao et al. (2017) highlighted that dysbiosis in Alzheimer’s patients correlates with elevated levels of inflammatory markers in blood and brain. Moreover, the gut microbiota can influence the production of beta-amyloid, the protein that accumulates in the brains of people with AD. Many studies suggest that the gut microbiota can also influence behavior (Cryan and Dinan, 2012; Sgambato and O’Mahony, 2020), especially in patients with dementia, where symptoms such as depression, anxiety, aggressiveness, and apathy are common. Dysbiosis disrupts the enzymatic machinery required for the synthesis of key neuroactive precursors (e.g., tryptophan availability for serotonin) (Cryan and O'Mahony, 2011; Hoban et al., 2016; Maier et al., 2017), thereby directly impairing synaptic plasticity and emotional regulation, while its alteration can cause behavioral disturbances in patients with dementia. Neuroinflammation is a key factor in the development of dementia and other neurodegenerative diseases. The gut microbiome can influence this process through the production of pro-inflammatory molecules like cytokines, which cross the blood–brain barrier and activate glial cells, causing neuronal damage (Foster and McVey Neufeld, 2013). Studies indicate that the microbiota regulates the brain’s inflammatory response by modulating central and peripheral immune cells. Dysbiosis, characterized by an increase in pathogenic bacteria or a reduction in beneficial bacteria, may promote chronic inflammation that worsens cognitive and behavioral symptoms in dementia. Recent studies show that specific gut microbiota compositions, such as the abundance of Firmicutes and Verrucomicrobia, are associated with better cognitive functions, while increased levels of Bacteroidetes and Proteobacteria correlate with memory and cognitive deficits (Cryan and Dinan, 2012). These data support the role of the gut-brain axis, through which the microbiota regulates neuroinflammation, neurotransmitters, and blood–brain barrier integrity, directly influencing cognitive health (Aburto and Cryan, 2023; Loh et al., 2024). Various therapeutic strategies based on the gut microbiota are under development for the treatment or prevention of dementia. These include the use of probiotics, prebiotics, and fecal microbiota transplantation (FMT), which have shown promising results in animal models and preliminary human studies. Hoban et al. (2016) demonstrated that probiotics can positively influence behavior and cognition in animal models of dementia, suggesting a potential therapeutic approach to modulate the microbiota and improve neuropsychiatric symptoms.

This review identifies three convergent pathogenic pathways linking dysbiosis to cognitive decline: (1) metabolic starvation via reduced short-chain fatty acids (SCFA) and lactate producers (e.g., Faecalibacterium, Lactococcus phages); (2) systemic inflammation driven by P-glycoprotein dysfunction and LPS translocation; and (3) direct neurotoxicity via secondary bile acids and p-cresol.

Given the breadth and complexity of the topic under investigation, this scoping review was conducted following the methodological framework proposed by Arksey and O’Malley (2005), later refined by Levac et al. (2010), and in accordance with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines. This type of review was chosen because it allows for a systematic mapping of the existing literature, the identification of key thematic areas, the synthesis of available findings, and the detection of research gaps. The scoping review proved particularly suitable for exploring the role of the microbiota i cognitive decline, a multidisciplinary and rapidly evolving field characterized by significant methodological and conceptual heterogeneity among studies that led us to a final selection of 35 studies, deemed sufficient for a structured and thematically organized narrative synthesis.

A comprehensive literature search was conducted across six electronic databases, initially retrieving 6,359 records. After removing 2,763 duplicates and 283 non-English articles, 3,313 records remained for title and abstract screening. During this phase, 2,946 articles were excluded based on titles, and 167 more after abstract review, leaving 200 articles for full-text retrieval. Despite efforts such as contacting authors and consulting libraries, 21 articles could not be accessed. The remaining 179 articles underwent detailed eligibility assessment, resulting in 35 studies meeting the inclusion criteria and included in this review (See PRISMA in Figure 1). The selected articles were categorized into five thematic macroareas, each representing a specific aspect of the relationship between gut microbiota and cognitive decline in dementia (Table 1). This classification allowed a systematic and coherent synthesis of the heterogeneous literature, facilitating comparison of findings and identification of common themes. This approach provided a clearer understanding of how microbiota alterations may influence cognitive impairment in dementia, improving the overall quality and focus of the review.

This scoping review analyzed numerous studies on the role of the gut and oral microbiota in cognitive decline, highlighting a growing integration between clinical, molecular, and environmental approaches. The findings outline a complex framework in which the microbiota emerges as a key element in the pathophysiology of neurodegenerative diseases, although the direction and robustness of these associations vary across studies due to methodological and demographic heterogeneity. Several studies have shown that modulation of the microbiota through probiotics and synbiotics can improve cognitive function, particularly in individuals with AD or MCI (Akbari et al., 2016; Azuma et al., 2023). Innovative approaches such as acupuncture and personalized diets (Yin et al., 2025; Zhang et al., 2021), as well as dietary interventions like intermittent fasting and ketogenic diets (Mela et al., 2025), demonstrate neuroprotective effects through microbiota regulation and inflammation control. As evidenced in the literature, the growing interest in integrative strategies—including fecal transplantation, physical exercise, and mindfulness suggests a shift toward multimodal approaches, although differences in intervention duration, strain specificity, and outcome measures make cross-study comparisons difficult (Bhadoriya et al., 2025). Nutritional and probiotic interventions appear to improve cognitive function in individuals with MCI or Alzheimer’s by acting on inflammation, SCFA production, BDNF modulation, and epigenetic regulation via miRNAs (Fu et al., 2022). However, their effectiveness is highly dependent on baseline cognitive status and individual microbiota composition, limiting generalizability across populations (Zhao T. et al., 2023; Zhao Y. et al., 2023). Microbiota profiling has revealed distinctive patterns between cognitively healthy individuals and patients with Alzheimer’s or MCI, with recurrent alterations in microbial diversity and bacterial taxa. Specifically, a reduction in butyrate-producing bacteria (e.g., Faecalibacterium, Roseburia) and an increase in pro-inflammatory species (e.g., Proteobacteria, Escherichia–Shigella) have been associated with neurodegenerative biomarkers such as amyloid burden and hippocampal volume reduction. Moreover, the presence of peripheral microbial signatures (Marizzoni et al., 2023) and plasma panels derived from the microbiota (Liu et al., 2023) opens new avenues for the development of non-invasive diagnostic tools (Zhang et al., 2025). Several studies suggest that cognitive decline is associated with the loss of beneficial bacteria (SCFA producers) and the increase of pro-inflammatory taxa, correlating with clinical and neurobiological biomarkers of AD (Zhu et al., 2022). Thus, the gut and oral microbiota emerge as potential early biomarkers and therapeutic targets (Chen et al., 2022). At the molecular level, its metabolites—such as SCFAs, bile acids, and neurotoxic compounds—influence inflammation, neuroplasticity, and brain metabolism (Wu et al., 2021; MahmoudianDehkordi et al., 2019; Garg et al., 2024). These data, consistent with the literature, indicate that SCFAs in particular play a key role in epigenetic and synaptic regulation (Zhang et al., 2025), opening new perspectives for early diagnosis and monitoring of cognitive decline. Nevertheless, metabolite measurements differ substantially across platforms (serum vs. plasma vs. feces), limiting comparability. Another interesting finding from this review is that the gut virome is emerging as a relevant component in neurocognitive health (Ghorbani et al., 2023). Alterations in bacteriophages can modify microbial composition and the production of neuroprotective metabolites, contributing to neuroinflammation and cognitive decline (James et al., 2024). The virome thus emerges as a potential biomarker, although current evidence is preliminary and limited by small sample sizes and heterogeneous methodologies. As highlighted in the literature, even the mycobiome—although still poorly studied—may influence brain function, enriching the complexity of the gut ecosystem (Bhadoriya et al., 2025). Individual and environmental factors, such as exposure to phthalates (Deng et al., 2024), genetic predisposition (Hou et al., 2021), education level (Klee et al., 2024), and intestinal motility (Ma et al., 2023), significantly influence microbiota composition. These associations, however, may be partially confounded by unmeasured variables such as diet, medications, lifestyle, and socioeconomic context, reinforcing the concept of AD as a systemic condition. Indices such as the Gut Dysbiosis Index (Li et al., 2020) and specific taxa associated with MCI progression (Yang et al., 2023) suggest the potential use of the microbiota as both a biomarker and therapeutic target. The preclinical stages of cognitive decline represent a critical window for intervention, as several studies indicate that subtle but detectable changes in microbiota composition and its metabolites occur during this phase, preceding the onset of measurable cognitive deficits (Wu et al., 2021; Yang et al., 2023; Yin et al., 2025). These modifications constitute potential early biomarkers and targets for timely preventive strategies. However, the limited number of longitudinal cohorts restricts the ability to determine causality or temporal sequencing of microbiota alterations. Numerous environmental (e.g., diet, pollutants), genetic (e.g., APOE ε4), socio-educational (e.g., education level), and physiological (e.g., intestinal motility) factors influence microbiota composition and function, contributing to cognitive decline through inflammatory, metabolic, and neurovascular mechanisms (Zhang et al., 2021; Deng et al., 2024; Hou et al., 2021). Yet the relative contribution of each factor remains difficult to quantify due to variability in study design and reporting practices. Several studies reviewed (Akbari et al., 2016; Azuma et al., 2023; Hsu et al., 2023) have shown that probiotic supplementation, alone or combined with prebiotic fibers, can improve cognitive function in individuals with MCI or AD. These results have also been confirmed in healthy elderly populations, with benefits in memory, attention, and executive flexibility (Fekete et al., 2024). The proposed mechanisms include modulation of systemic inflammation, increased SCFA production, and BDNF regulation. In parallel, microbiota analysis has revealed significant alterations in individuals with cognitive impairment, including a reduction in beneficial taxa such as Faecalibacterium and Ruminococcaceae, and an increase in pro-inflammatory bacteria such as Proteobacteria and Escherichia–Shigella (Zhu et al., 2022; Wanapaisan et al., 2022; Grabrucker et al., 2023). Another emerging area concerns the oral microbiota, increasingly recognized as a potential early biomarker. Reviewed studies (Chen et al., 2022; Sritana and Phungpinij, 2024) have highlighted the role of pathogenic oral bacteria, such as Fusobacterium nucleatum and Porphyromonas gingivalis, in neuroinflammation and amyloid aggregation, as also confirmed by other literature data (Lim et al., 2017). Still, oral-microbiome methodologies vary substantially across studies. At the molecular level, microbial metabolites, particularly SCFAs like butyrate, play a neuroprotective role, while other compounds such as Trimethylamine N-oxide (TMAO) and toxic bile acids are associated with AD biomarkers (MahmoudianDehkordi et al., 2019; Nho et al., 2019; Wu et al., 2021; Kim et al., 2024).

Finally, integrated and multi-omic approaches are revolutionizing the understanding of cognitive decline. Studies such as those by Yang et al. (2023) and Yin et al. (2025) have developed predictive models based on microbial, metabolic, and neuroimaging signatures capable of early identification of individuals at risk of progression from MCI to dementia. Nevertheless, differences in computational pipelines and lack of standardized analytical frameworks limit reproducibility. The reviewed evidence supports a model where cognitive decline is driven by a “double hit”: (1) the loss of neuroprotective metabolites (ergothioneine, SCFAs) and (2) the active production of neurotoxins (secondary bile acids, p-cresol). This shifts the therapeutic focus from simply “adding good bacteria” to restoring specific metabolic pathways. Taken together, while the existing literature highlights promising microbiota–brain interactions, these methodological limitations caution against overinterpretation and underscore the need for more rigorous, standardized, and longitudinal research frameworks. The complex interplay between gut dysbiosis, metabolite translocation, and neuroinflammatory pathways is mechanistically summarized in Figure 2.

This scoping review highlights the central role of the gut and oral microbiota in the pathophysiology of cognitive decline, reinforcing the hypothesis of the gut–brain axis in the progression of MCI and AD. Compositional alterations, including the loss of beneficial taxa and the increase of pro-inflammatory species, are associated with cognitive and neuroinflammatory dysfunctions. In this framework, the microbiota thus emerges as a non-invasive diagnostic biomarker and a modifiable therapeutic target. Interventions such as probiotics, synbiotics, dietary modifications, and multimodal approaches show promising effects, especially in the early stages of cognitive impairment. Furthermore, the integration of multi-omic strategies and predictive models opens new perspectives for precision medicine. However, the interpretation of these findings is limited by several factors. High methodological heterogeneity, geographical variability, and the lack of longitudinal studies substantially reduce the generalizability of current findings. Additional challenges include the variability of sequencing techniques, inconsistent metabolite-profiling methods, and limited control of confounding lifestyle and environmental factors. Ultimately, this review highlights that the microbiota–gut–brain axis is not just a biomarker of disease, but an active, modifiable driver of AD pathology via specific metabolic and immune checkpoints (e.g., P-gp modulation, bile acid metabolism), offering tangible targets for precision medicine.

Future research should focus on longitudinal and multicenter studies conducted on large and diverse cohorts to validate microbial and metabolic signatures as predictive biomarkers of progression from MCI to AD. Randomized controlled trials will be necessary to evaluate personalized microbiota-targeted interventions, including probiotics, dietary strategies, and fecal microbiota transplantation. The integration of multi-omic approaches (metagenomics, metabolomics, epigenomics, neuroimaging) is increasingly recognized as crucial to better understand microbiota–brain interactions. Additionally, further exploration of the oral microbiome, virome, and mycobiome is remains essential, along with the promotion of interdisciplinary approaches to develop precision medicine strategies. Importantly, standardized protocols for sample collection, sequencing, metabolite quantification, and data reporting will be necessary to improve comparability across studies and strengthen the translational potential of microbiota-based biomarkers.

The main strength of this scoping review lies in its systematic and multidimensional approach, which enabled an integrated analysis of the microbiota’s role in cognitive decline. The inclusion of clinical, observational, and experimental studies, organized into five thematic macro-areas, provided a broad and up-to-date overview, with particular attention to emerging fields such as the gut virome and oral microbiome. However, the substantial methodological heterogeneity across studies—ranging from sequencing platforms to metabolite-profiling techniques-, the predominance of observational and cross-sectional designs, and the limited integration of multi-omic data restrict significantly limit differences in dietary and environmental exposures, and inconsistent control of confounding factors further constrain generalizability. The absence of longitudinal trajectories prevents the identification of temporal dynamics and early inflection points in microbiota-related changes. Despite these limitations, the review offers a solid foundation to guide future research and inform the development of more rigorous, personalized therapeutic strategies grounded in microbiota-driven mechanisms rather than purely taxonomic changes.