Gut microbiome may lead to increased neuroinflammation and decreased synaptic plasticity, both of which are thought to contribute to the development and progression of AD (Jiang et al., 2021). Gut microbiomes can alter the composition of the immune cells in the brain, leading to an increase in neuroinflammation (Erny et al., 2015). For example, studies in animal models have shown that colonization with certain types of gut bacteria can lead to microglial activation. Deficiency of the gut microbiome first affects the microglial cell transcriptome, which primarily regulates the interconversion of microglial cell subpopulations, and transcriptome changes are primarily associated with AD (Huang et al., 2023). Studies have shown that changes in the gut microbiome can alter synaptic plasticity in the brain. For example, as recently reviewed, germ-free mice (raised without exposure to microorganisms) have impaired synaptic plasticity compared to conventionally-raised mice (Glinert et al., 2022).

Synaptic dysfunction and microglia may interact through the microbiome (Bairamian et al., 2022), and synaptic dysfunction may precede symptoms of cognitive impairment (Selkoe and Hardy, 2016), as evidenced by diminished long-term potentiation (LTP) and long-term depression, which are closely related to synaptic plasticity (Bairamian et al., 2022). In addition, animal model studies have identified gut microbiome metabolites that regulate LTP in physiological and AD-related pathologies mechanisms, including influencing the expression of pre-and postsynaptic neuronal membrane receptors and membrane genes that further influence ion channels and thus affect synaptic activity (Albuquerque et al., 2015; Tong et al., 2015). These findings suggest a complex relationship between the gut microbiome, neuroinflammation, synaptic plasticity, and AD development. While much more research is needed to fully understand these relationships, the emerging evidence suggests that targeting the gut microbiome may represent a promising avenue for developing new treatments for this devastating disease.

Inflammation in AD is not confined to neuroinflammation but includes the complex signals between microbial involvement in peripheral and central interaction. This crosstalk may comprise many neuronal, endocrine, immunological, and metabolic pathways. To be more precise, by exploiting the MGBA. This section concentrates on interacting with the brain through neurological and humoral pathways (Abdel-Haq et al., 2019). Figure 1 concisely summarizes the role that the gut microbiome plays in AD in terms of epigenetics as well as central and peripheral crosstalk.

Escherichia coli-derived neurotoxins in Proteobacteria are associated with neuropathology in AD and increase the release of pro-inflammatory cytokines that induce systemic inflammation and exacerbate AD pathology (Kitazawa et al., 2005; Cattaneo et al., 2017). Many bacteria strains are capable of producing amyloid (Megur et al., 2020; Tran and Mohajeri, 2021), and the forms show similarity with the CNS amyloid (Hill and Lukiw, 2015), which can lead to an enhanced immune response and endogenous formation of neuronal amyloid in the brain (Friedland and Chapman, 2017). LPS and gram-negative E. coli fragments coexisted with amyloid plaques in AD brain tissue (Zhan et al., 2016). Aβ loading may initially occur in the gastrointestinal tract. Aβ_1-42 migrates after being injected into the gastric wall, and Aβ deposition is present in the vagus nerve and brain and develops symptoms of cognitive impairment (CI; Sun Y. et al., 2020). Therefore, this could provide some basis for the movement of the GM to trigger the amyloid pathogenesis of AD. The microbiome is necessary for the normal development of hippocampal and microglial cell morphology (Luczynski et al., 2016). Recently, Liu et al. observed microbiome deficiency altered dendritic signaling integration in the Cornu Ammonis1 region of mice (Liu et al., 2022). The presence of activated microglia and reactive astrocytes in the vicinity of amyloid plaques is characteristic of AD neuroinflammation, mainly in the hippocampus (Crews and Masliah, 2010; Cattaneo et al., 2017). Moreover, in the elderly brain, microglia are dysfunctional and prone to chronic activation (Spittau, 2017). In the early pathological stages of AD, the overall microglial response can support neurons by phagocytosing Aβ fibers. However, as the disease progresses, the loss of the branching phenotype of microglia with excessive activation near plaques and aggregation of other immune cells triggers a neurotoxic environment, leading to neural network damage (Leng and Edison, 2021; Thu Thuy Nguyen and Endres, 2022). Other studies have also observed numerous roles for aberrantly regulated GM in triggering Aβ amyloidosis, neuroinflammation, and microglia regulation in AD mice (Dodiya et al., 2019).

The intestinal microbiome contains two major bacterial phylotypes, the Firmicutes, the Bacteroidetes, and to a lesser extent, the actinomycetes, fusobacteria, proteobacteria, and weberia (Grenham et al., 2011). Some studies have found less diverse Firmicutes and genera, but a higher prevalence of Proteobacteria in AD (Nagpal et al., 2019; Sheng et al., 2021; Hung et al., 2022). Intestinal α-diversity and β-diversity are altered in patients on the AD spectrum (Hung et al., 2022). Several studies have reported a significant decrease in α-diversity in AD patients (Vogt et al., 2017; Liu P. et al., 2019), but not obvious in mild cognitive impairment (MCI) patients, and there is a similar gradual decline trend from MCI to AD (Nagpal et al., 2019). Nevertheless, results vary regarding β-diversity studies, with one Austrian study showing that malnutrition and drug intake affect the abundance of the specific microbiome, SCFAs, and butyrate production, and statins may be one of the reasons affecting beta diversity (Stadlbauer et al., 2020). A recent systematic review and meta-analysis of 11 studies included 378 normal control (NC) and 427 AD patients to analyze the impact of different countries and clinical stages on GM abundance. GM diversity was significantly lower in AD patients than in NC but not in MCI patients. Compared to NC, the AD spectrum group had an increased abundance of Proteobacteria, Bifidobacterium, and Phascolarc to the Bacterium, and a decreased abundance of Firmicutes, Clostridiaceae, Lachnospiraceae, and Rikenellaceae (Hung et al., 2022). More importantly, the abundance distribution of the Alistipes and Bacteroides in the NC and AD differed by country (Hung et al., 2022).

Several studies have observed alterations in gut microbiology in patients with AD, but mostly with the influence of drugs and other interventions and varying disease duration. Although there is no direct evidence in clinical trials that the effects on MGBA of current medications for AD (acetylcholinesterase inhibitors or N-methyl-D-aspartate (NMDA) antagonists) have been studied (La Rosa et al., 2018), the latest study provides evidence that donepezil affects GM via amino acid pathways and sugar metabolism (Jo et al., 2022). In light of the results mentioned above, the researchers’ attention was drawn to MCI and subjective cognitive decline (SCD) patients’ gut changes. A study used 16S ribosomal ribonucleic acid (16SrRNA) sequencing to examine stool samples from 18 AD patients without treatment or intervention, 20 MCI patients, and 18 age-matched NC. The findings revealed that Prevotella levels were higher and Bacteroides, Lachnospira, and Ruminiclostridium_9 levels were lower in AD patients. Additionally, greater cognitive performance was favorably correlated with Bacteroides, Lachnospira, and Ruminiclostridium_9, although Prevotella had the opposite relationship (Guo et al., 2021). Based on reduced effects such as drugs, the above results likewise support the idea of a progressive worsening of the degree of intestinal dysbiosis from MCI to the disease stage of AD (Guo et al., 2021). Based on this, Sheng et al. further explored gut microbial composition changes in SCD, the earliest symptom of preclinical AD. The study included 38 NC, 53 patients with SCD, and 14 patients with CI and compared the gut microbial composition of the three groups and the relationship with cognition using 16SrRNA technology (Sheng et al., 2021). Findings revealed a decreasing trend in the abundance of the phylum Firmicutes, class Clostridia, order Clostridiales, family Ruminococcaceae, and genus Faecalibacterium from NC to SCD and CI (Sheng et al., 2021). In particular, the abundance of the anti-inflammatory genus Faecalibacterium was significantly lower in the SCD group than in the NC group. Notably, altered bacterial taxa were associated with cognitive performance and were validated in amyloid-positive SCD participants (Sheng et al., 2021). The results further suggest that the microbiome may influence the development of amyloid pathology.

Proteobacteria is a major phylum of gram-negative bacteria (Zhao and Lukiw, 2018). More gram-negative bacteria have also been reported in AD patients compared to NC (Vogt et al., 2017). In addition, the abundance of Proteobacteria increased with increasing memory dysfunction (Hossain et al., 2019; Jeong et al., 2019). Bacteroides have been shown to preserve the intestinal barrier and reverse leaky gut (Hsiao et al., 2013), indicating that they may be a microbiome protective factor. Bacteroides levels are reduced in AD and MCI patients (Cattaneo et al., 2017; Zhuang et al., 2018). Moreover, the relative abundance of the Actinomycete phylum was associated with diffusion tensor imaging of the thalamus, hypothalamus, and amygdala, as well as cognitive test scores (Fernandez-Real et al., 2015). Certain Lactobacillus and Bifidobacterium secrete gamma-aminobutyric acid, which may mediate cognitive function in AD (Barrett et al., 2012), and different levels of abundance of the Lactobacillus family have been observed at different stages of dementia (Stadlbauer et al., 2020). Vogt et al. compared the different bacterial classifications of stool samples from AD patients versus NC and observed levels of differentially abundant genera associated with cerebrospinal fluid (CSF) biomarkers of AD pathology (Vogt et al., 2017).

In addition to the intestinal bacteria noted by most researchers, some studies have also found specific alterations in MCI patients in the genus Fungi and a clear correlation with AD biomarkers (Nagpal et al., 2019). More importantly, the different microbiome and fungal biota composition in 3xTg-AD mice is similar to the microbial variation found in humans, and significantly different taxa may contribute to the pathogenic cues of AD being identified (D’Argenio et al., 2022). However, there are some differences in studies regarding oral microbiology. A Canadian study used 16SrRNA sequencing to analyze stool and oral samples from AD and NC comparatively. The study found that oral microbiota exhibited greater differences between patients and controls than GM but was not associated with cognitive function (Cirstea et al., 2022). Although some results support the possibility that oral bacteria or their components may themselves invade the brain (Laugisch et al., 2018; Jungbauer et al., 2022), analyses using publicly available genome-wide association studies on periodontitis and AD failed to reveal any genetically predicted association between AD and periodontitis risk (Sun Y.-Q. et al., 2020). However, preventing and treating periodontitis and its associated inflammation could reduce neuroinflammation and prevent and treat neurodegenerative diseases (Li et al., 2022).

Based on many characteristic changes in the GM detected in the pre-AD period, this finding may provide clues as a diagnostic biomarker in the pre-AD period (Guo et al., 2021; Sheng et al., 2021; Verhaar et al., 2022). Hence, numerous studies have attempted to use GM for early diagnostic identification of AD.

APP/PS1 mice combining 16SrRNA gene sequencing and extensively targeted metabolomics revealed that B. firmus, Rikenella, Clostridium sp. culture-27, and deoxyuridine might be important biomarkers of AD (Feng et al., 2022). Based on the GM differences in feces and blood established in the preliminary study, Li et al. built a random forest model based on 11 genera of differences in stools and blood from AD patients and NC. Using the random forest model’s cut-off values with all different stool input genera, 28 of 30 MCI patients could be accurately identified with a sensitivity of 93% (Li et al., 2019). In addition, 33 AD, 32 amnestic MCI (aMCI), and 32 NC were examined in cross-sectional research. According to the findings, AD patients had a reduced level of fecal microbial diversity compared to aMCI patients and NC. Further, as patients advanced from NC to aMCI and AD, Gammaproteobacteria, Enterobacteriales, and Enterobacteriaceae gradually enriched abundance. A strong association was also discovered between the clinical severity scores of AD patients and the abundance of changed microbiomes. Notably, the model based on the abundance of Enterobacteriaceae bacteria could distinguish AD from aMCI and NC (Liu P. et al., 2019). This initially validates the value of intestinal biomarkers in the early identification of patients with CI and affirms the great potential of gut biomarkers. To improve the robustness of gut biomarkers, a study in 2022 simultaneously combined the latest proposed [AT(N)] framework. The study collected 34 Aβ (−) cognitively normal (CN−), 32 Aβ (+) cognitively normal (CN+), and 22 (11 MCI, 11 AD) CI patients. According to the findings, the relative abundance of bacteria that produce SCFAs reduced from CN− to CN+ and CI. Moreover, all CN subjects had a negative correlation between total brain Aβ load, plasma Aβ₄₂/Aβ₄₀, and 3 particular types of bacteria. It was discovered that combining plasma Aβ markers, altered GM, and cognitive ability improved discrimination between CN+ and CN− (Sheng et al., 2022). The above promising findings suggest the possibility and accuracy of GM in diagnosing AD.

Patients with newly diagnosed AD or MCI have dysbiosis of the intestinal microbiome, including a decrease in the potentially protective microbiome and an increase in the pro-inflammatory microbiome (Chen et al., 2020; Bello-Medina et al., 2021). Significant changes in gut microbial composition are observed, especially in older people, and diet may be the most important driver of change (Kumar et al., 2016). The severity of human microbiome dysbiosis shows characteristic changes in different disease stages of AD (Stadlbauer et al., 2020); an assessment of disease progression and prognosis provides some basis, suggesting that these may be meaningful biomarkers for predicting AD development (Li et al., 2019). In addition, recent studies have reported that neuroimaging of the MGBA will further improve our understanding of GM by altering neural microstructure and function in time and space (Montoro et al., 2022).

Tight junctions are an important mechanism affecting the intestinal barrier and the BBB, with the interaction of intestinal bacteria with epithelial cells being a key factor in regulating epithelial permeability by modulating tight junctions (Allam-Ndoul et al., 2020). It has been found in the last decade to be associated with AD (Schoultz and Keita, 2020). Calprotectin, a marker of intestinal inflammation (Kowalski and Mulak, 2019), is significantly increased in the CSF and brain of AD patients (Wang et al., 2014). It implies that intestinal permeability could be associated with AD pathogenesis.

Human targeting studies-the sugar absorption assay (Ménard et al., 2010), identified plasma lipopolysaccharide-binding protein (LBP) as a promising biomarker of intestinal permeability in adults (Seethaler et al., 2021). However, the markers of intestinal barrier integrity LBP and intestinal fatty acid binding protein (IFABP), are not associated with independent AD dementia, MCI, cognition, and neuropathology (Voigt et al., 2021). It is worth noting that disruption of the apical junction complex was not evaluated, and the small sample size may have influenced the results. Another study indirectly illustrated the relationship between intestinal permeability and AD. The peripheral platelet alterations are involved in the pathological process of AD and are based on the fact that C-type lectin-like receptor 2 (CLEC-2) is an activated receptor on the platelet surface (Mukaetova-Ladinska et al., 2012; Akingbade et al., 2018). Wang et al. found elevated levels of CLEC-2 and zonulin in MCI and AD patients, with a progressive increase compared to NC. Furthermore, the study identified high levels of CLEC-2 and zonulin as significant factors for decreased Mini-Mental State Examination scores (Wang X. et al., 2020). An Austrian clinical study included CI and normal population controls with serum diamine oxidase (DAO) and fecal zona pellucida protein to detect intestinal permeability and found higher levels of DAO in patients with dementia. The results suggest a correlation between dementia and increased intestinal permeability biomarkers (Stadlbauer et al., 2020). In addition, GM disorders resulted in low expression of the major tight junction proteins of the BBB (zonula occludens-1, occluding, and claudin-5) in the frontal cortex, hippocampus, and perisylvian striatum of mice (Braniste et al., 2014). More importantly, these brain regions affect memory and cognitive functions.

The findings above support a link between intestinal barrier dysfunction and AD. Some bacteria, such as Lactobacillus plantarum, Escherichia coli, or Bifidobacterium infantis, can improve the intestinal wall barrier by boosting the expression of tight junction-related proteins (Hakansson and Molin, 2011; Kowalski and Mulak, 2019). Increasing butyrate-producing bacteria enhanced tight junction protein expression in mice’s frontal cortex and hippocampus (Marizzoni et al., 2020). These findings give more evidence for the role of intestinal permeability in AD and may have biomarkers potential. However, it is crucial to highlight that due to the difficulties of performing sugar absorption tests in people with dementia, blood, and stool indicators are typically employed clinically instead (Stadlbauer et al., 2020). Various biomarkers associated with epithelial damage, including citrulline, IFABP, and LBP, have been used as indirect indicators of the reduced intestinal barrier (Galipeau and Verdu, 2016). However, studies have shown that even before leakage occurs, when the Notch signaling pathway involved in regulating the BBB is altered, the BBB has been disrupted, and there is already an impact on synaptic function, triggering dementia (Calderon et al., 2022). Conceivably, this will provide insights into discovering additional early markers of barrier function. Nonetheless, the mechanistic pathways that connect the gut barrier with the BBB should be studied further.

In AD, the intestinal microbiome and permeability characteristics have led researchers to explore the potential of gut microbial metabolites as biomarkers. The Ruminococcaceae metabolite isoamylamine recognizes and binds the Recombinant S100 Calcium Binding Protein A8 promoter region and promotes neuronal cell death, leading to cognitive decline in mice (Teng et al., 2022). Moreover, memory deficit in APP/PS1 mice was correlated with metabolite status. Fasudil et al. found that glutamate, hypoxanthine, thymine, hexanoyl coenzyme A, and leukotriene metabolize nucleotides, lipids, and sugars as well as inflammatory pathways involved in AD mouse-related metabolism (Yan et al., 2021). These novel results provide a valuable reference for using gut microbial metabolites as diagnostic biomarkers and therapeutic targets in clinical studies of AD. Several bacterial metabolites have been used as fecal biomarkers to characterize patients with AD. Trimethylamine N-oxide (TMAO) is a microbial-dependent metabolite (Connell et al., 2022). The study found higher TMAO levels, enhanced oxidative stress, and intestinal barrier dysfunction in MCI and AD patients (Vogt et al., 2018; Yan et al., 2021). TMAO topped the list of 56 microbial metabolites considered biomarkers of AD in a study identifying associations between AD and microbial metabolites through a web-based algorithm, successfully predicting changes in memory and fluid cognition in aging individuals (Xu and Wang, 2016). A clinical study included AD (n = 40), MCI (n = 35), and NC (n = 335) results show TMAO was significantly and positively correlated with p-tau, p-tau/Aβ₄₂, and t-tau, neurofilament light in the CSF of AD (Vogt et al., 2018). These results suggest that TMAO may be involved in AD pathology and neurodegeneration (Vogt et al., 2018; Arrona Cardoza et al., 2022). Changes in tryptophan, methionine, tyrosine, and purine metabolism were observed in CSF from AD patients, suggesting these may be risk factors for CI (Kaddurah-Daouk et al., 2013). Some derived molecules produced by GM in the body’s circulation are also considered potential biomarkers of AD. The outer membrane of gram-negative bacteria is primarily made up of LPS (Wu L. et al., 2021), which is present in the hippocampus and superior temporal lobe neocortex of AD brains and triggers neuroinflammatory properties (Zhao et al., 2017) that have important implications for CI (Sparkman et al., 2006). In aMCI and AD patients were found to have elevated serum LPS levels compared to NC patients (Wu L. et al., 2021). In another study, the metabolomic analysis revealed significant differences between AD and NC in tryptophan metabolites, SCFAs, and staphylococcal acid, and most were associated with the altered microbiome and CI. Notably, tryptophan disorders are already present in aMCI, and SCFAs show a decreasing trend from aMCI to AD. More importantly, indole-3-pyruvic acid, a tryptophan metabolite that accurately classifies aMCI and AD with NC, is a marker for identifying and predicting AD, and five SCFAs were identified as markers for the pre-onset and progression of AD (Wu L. et al., 2021).

Several studies have found that hormones synthesized by intestinal cells and gut microbiomes can affect AD through neural and humoral pathways (Welcome, 2018). Notably, the microbiome may also regulate hormone levels (Yan et al., 2016; Engevik et al., 2020), suggesting that gut hormones may also be potential biomarkers of AD. Leptin affects the cortex and hippocampus and reduces Aβ in the brains of AD mice (Fewlass et al., 2004; Farr et al., 2015). In addition, chronic lateral ventricular injection of leptin effectively alleviated Aβ_1-42-induced spatial memory deficits and reversed Aβ_1-42-induced hippocampal late-phase LTP inhibition in rats (Tong et al., 2015). Cholecystokinin (CCK) is a satiety hormone that binds directly to CCK receptors in the hypothalamus and hindbrain to regulate appetite and is highly expressed in the hippocampus, which is essential for protecting and enhancing memory (Blevins et al., 2000; Plagman et al., 2019). A study examined the CCK levels of CSF in 287 subjects with AD and found high levels of CCK were associated with higher total CSF tau and p-tau181, associated with better cognition and more gray matter volumes in the posterior cingulate gyrus, parahippocampal gyrus, and medial prefrontal cortex. These results suggest that CCK levels may reflect compensatory protective mechanisms during AD pathology (Plagman et al., 2019). Cortisol mediates cognition through two types of receptors: Mineralocorticoid Receptors (MRs) and Glucocorticoid Receptors (GRs) (Joëls, 2006; Daskalakis et al., 2013). MRs have a higher affinity, 6–10 times higher than GRs (de Kloet et al., 1999; Joëls, 2006). In animal models, Enterococcus faecalis was found to promote social activity and reduce Corticosterone (CORT) levels in mice after social stress (Wu W.-L. et al., 2021); GM imbalance causes elevated CORT in AD mice (Hendrickx et al., 2021) and remodeling of the GM was able to reduce CORT levels (Liu et al., 2021). In addition, Ghrelin was found to modulate the function of hippocampal neurons and synapses in mice, thereby enhancing memory (Li et al., 2013). The above results indicate that elevated levels of stress hormones may be a potential biomarker for pre-symptomatic AD (Hendrickx et al., 2021). In humans, Sami Ouanes et al. measured cortisol and Dehydroepiandrosterone sulfate (DHEAS) levels in CSF and found that cortisol and cortisol/DHEAS ratios were positively correlated with tau and p-tau CSF levels and negatively correlated with the amygdala and insula volumes at baseline. More importantly, higher CSF cortisol and DHEAS levels predicted a more pronounced cognitive decline and disease progression over 36 months (Ouanes et al., 2022a). The higher CSF cortisol may reflect or contribute to more severe neuropsychiatric symptoms at baseline unrelated to AD pathology and more pronounced deterioration over 3 years (Ouanes et al., 2022b). These findings have important implications for identifying AD. Other studies have found reduced levels of Ghrelin messenger ribonucleic acid (mRNA) in the temporal lobe of AD patients (Gahete et al., 2010). Ghrelin may be involved in changes in neuroinflammation and cognitive function in AD (Jeon et al., 2019). These promising results further demonstrate the potential of gut hormones as biomarkers of AD.

Dysbiosis may result in the degeneration of the CNS immune response and the significant release of inflammatory cytokines that activate macrophages (Wu and Wu, 2012; Belkaid and Hand, 2014). Animal models and clinical studies have demonstrated that GM contributes to AD’s pathogenesis by regulating microglia function to involve peripheral and central inflammation processes (Bradt et al., 2000; Wu M.-L. et al., 2021). Numerous studies have demonstrated a connection between the presence and metabolism of Aβ or tau protein and the release of pro-inflammatory cytokines such as Interleukin (IL)-1, IL-6, and IL-8 by activated microglia (Teixeira et al., 2008; Breer et al., 2012; Uslu et al., 2012). IL-6 has been found to have potential properties for detecting the severity of CI in AD patients (Lai et al., 2017). Two meta-analyses found that inflammatory biomarkers such as high-sensitivity c reactive protein, transforming growth factor-beta 1 (TGF-β1), IL-1β, IL-2, IL-6, IL-12, IL-18, monocyte chemotactic protein-1 (MCP-1), MCP-3, IL-8, and interferon-γ-inducible protein 10 were consistently elevated in AD patients (Decourt et al., 2017; Su et al., 2019). The results in AD, MCI, and controls supported the view that AD and MCI are accompanied by both peripheral and CSF inflammatory responses. In addition, AD patients have higher levels of soluble tumor necrosis factor (TNF) receptor (sTNFR)-1 and sTNFR-2, which previous studies have shown to exacerbate the major pathological changes in AD (Aβ and tau pathology) (Decourt et al., 2017; Shen et al., 2019; Varesi et al., 2022a). The significant differences in peripheral inflammatory factor concentrations found in the study suggest that these markers may be useful in monitoring disease progression.

Cattaneo et al. were the first to describe patients with cerebral amyloidosis-related CI. IL-1β, C-X-C motif ligand (CXCL)-2, and NOD-like receptor family pyrin domain containing (NLRP)-3 were all positively connected with E. coli/Shigella abundance, Eubacterium rectale abundance was negatively correlated with IL-1β, CXCL-2, and NLRP3 and positively correlated with IL-10. It was also confirmed that an increased abundance of E. coli/Shigella and a decreased abundance of anti-inflammatory Rectal fungi might be associated with peripheral inflammatory status in patients with CI in brain amyloidosis (Cattaneo et al., 2017). The above study improves the robustness of microbiome-based inflammatory factors as biomarkers. Several recent studies have validated the important role of microbiome-based inflammatory biomarkers in diagnosing and detecting disease development (Shen et al., 2020; Park J.-C. et al., 2020; Park et al., 2021; Xin et al., 2021). However, larger samples and the establishment of longitudinal studies to further narrow the range of inflammatory markers remain to be addressed.

Increasing the body of evidence further validated the important role of characteristic GM, intestinal permeability, bacterial metabolites, gut hormones, and various peripheral inflammatory biomarkers in diagnosing and monitoring disease progression. But some issues need to be further addressed. First, the gut microbial species are diverse and numerous, and we have mainly focused on bacterial species; although the influence of fungi and viruses is beginning to be noted, numerous impacts and mechanisms are unclear (Ye et al., 2022). Based on the current findings, further expansion of the scope and sample size is needed to further validate the feasibility and accuracy of intestinal biomarkers in AD (Sorboni et al., 2022). Second, due to the insufficient number of countries included in the existing studies and the limitations of some animal models to which the test results can be referred (Rahman et al., 2020; Hung et al., 2022). It is not easy to obtain consistent results across various dietary, environmental, genetic, and other confounding factors (David et al., 2014; Mullane and Williams, 2020). Consider that a single biomarker may not be sufficient to describe the pathophysiology of AD fully. Finally, combining multiple markers representing different stages of disease development may be a better option (Gu et al., 2021; Zhang X. et al., 2021). Therefore, based on these issues, using intestinal biomarkers for preclinical and clinical diagnosis is still some time away. However, previous studies have focused on serum, CSF, and tissue metabolomics of AD. Because fecal biomarkers are non-invasive, readily available, and found to have great potential as biomarkers for AD, their utility can be developed in the future, perhaps as an alternative or complementary tool. Combining imaging, plasma, and other biomarkers may improve early identification rates (Mahaman et al., 2022; Sheng et al., 2022). Numerous studies have used compound biomarkers to enable more sensitive detection and early disease diagnosis. For example, diet, GM, and microRNAs can detect MCI patients (Zhang X. et al., 2021). The combination of inflammatory factors (IL-6 and interferon-γ), phosphatidylcholine, and single-chain fatty acid-producing bacteria enables early diagnosis of AD (Gu et al., 2021). Hence, gut biomarkers may be a non-invasive and cost-effective diagnostic tool for early AD screening (Table 1) summarizes the latest potential human intestinal biomarkers.