Gut microbiota is associated with several diseases, including neurodegenerative diseases such as Parkinson’s disease (PD) and Alzheimer’s disease (AD) (Petra et al., 2015; Jiang et al., 2017). Notably, the microbiota-gut-brain axis is a bi-directional communication system that is not fully understood; however, it is known to include neural, immune, endocrine, and metabolic pathways. Studies in germ-free animals and those exposed to pathogenic microbial infections, antibiotics, probiotics, or fecal microbiota transplantation suggest the link of gut microbiota with host cognition or AD-related pathogenesis (Pistollato et al., 2016; Harach et al., 2017; Liu et al., 2017; Rieder et al., 2017; Russo et al., 2017).

Several studies have supported the theory of the occurrence of a pathway of communication between the gut and the brain, modulated by gut microbiota (Gareau, 2014; Mayer et al., 2014). It has been speculated that targeting the microbiota can affect the behavior and modulate brain plasticity and cognitive functions while aging (Leung and Thuret, 2015). Some studies demonstrated that gut-targeted intervention by consuming lactic acid bacteria such as those in yogurt, could improve or delay the onset of cognitive decline associated with aging (Jung et al., 2012; Choi et al., 2015; Scott et al., 2017). Therefore, using probiotics for ameliorating the cognitive and behavioral disorders could be a potential treatment.

The hypothesis that the microbiota-gut-brain axis (Rhee et al., 2009) plays a critical role in health and disease, including neuropsychiatric disorders (Liu et al., 2017), is rapidly progressing. Nurturing a beneficial gut microbiome with prebiotics, such as fructooligosaccharides and inulin, is an appealing but under-investigated microbiota-induced manipulation. A previous study showed that the prebiotic treatment could modify the behavior across domains relevant to anxiety, depression, cognition, stress response, and social behavior (Burokas et al., 2017). Previous findings strengthened the evidence supporting the therapeutic targeting of gut microbiota in brain-gut axis disorders, thereby opening new prospects in the field of nutritional neuropsychopharmacology. Thus, it is imperative to develop novel and effective drugs or foods with prebiotic effects from natural resources.

Morinda officinalis How. (M. officinalis), as a Chinese traditional natural herbal medicine, contains a number of active components. Reportedly, the content of saccharides in M. officinalis radix is 49.79–58.25%, which is highly composed of oligosaccharides, such as inulin-type hexasaccharide exerting antidepressant effects in the model systems (Cai et al., 1996; Li et al., 2001). This phenomenon can be effectuated by up-regulating the expression of neurotrophic factors and/or down-regulating the [Ca²⁺]_i overloading (Li et al., 2004). Bajijiasu, another oligosaccharide, protects PC12 cells from Aβ_25-35-induced neurotoxicity (Chen et al., 2013), ameliorates the cognitive deficits induced by D-galactose in mice, and protects against ischemia-induced neuronal damage or death (Tan et al., 2000a,b). Our previous study suggested that oligosaccharide extracted from M. officinalis (OMO) might inhibit the oxidative stress and neuronal apoptosis, restore normal energy metabolism, as well as, increase the cell viability and mitochondrial membrane potential in AD animal models significantly (Chen et al., 2014a). However, the underlying mechanism is yet to be elucidated.

Alzheimer’s disease is the most common neurodegenerative disorder, affecting approximately >5% of the worldwide population aged >65 years, annually. AD is a chronic neurodegenerative disease that frequently exhibits a slow progression accompanied by a greater recession of the disease over a period. Although the cause of AD is poorly understood, among various biochemical and morphological events, the presence of neurofibrillary tangles, senile plaques, and neuronal and synaptic loss are considerably noted (Chen et al., 2017). Several pieces of evidence have confirmed that the accumulation of intracellular β-amyloid (Aβ) may be an early event in the development of AD. Aβ is a peptide comprised of 36–43 amino acids formed by a large transmembrane glycoprotein, such as amyloid precursor protein (APP), expressed on the cell. Aβ may activate the inflammatory and neurotoxic processes, including the excessive generation of free radicals and oxidative damage among intracellular proteins and other macromolecules (Abbott, 2011). D-galactose can form advanced glycation end products (AGEs). The administration of D-galactose to human populations can induce cognitive deficits and disruptions in the synaptic communication. Thus, D-galactose-treated rats with synaptic disruption and memory impairment have been extensively used as model rodents (Zhan et al., 2014; Gao et al., 2016; Huang et al., 2016; Li et al., 2016). In our previous, we have used these two models to screen the effective compounds for curing the AD (Chen et al., 2014b, 2017). Thus, in this study, the two AD-like models were induced by D-galactose and Aβ, respectively.

Herein, we investigated whether OMO is effective in alleviating AD by targeting the microbiota-gut-brain axis. OMO was administered in rats exhibiting AD-like symptoms induced by D-galactose and Aβ_1-42, respectively. Some indexes and histological deterioration, including learning and memory abilities, histological alterations, production of cytokines, microbial communities, and transcriptome in small intestine and brain, were identified in AD-like animals. And the prebiotics were directly evaluated in an overdose antibiotics-treated trinitro-benzene-sulfonic acid (TNBS)-induced mice model.

Increasing evidence suggests that the microbiota-gut-brain axis plays a key role in regulating brain functions, and prebiotics are widely considered to have potential as modulators of brain dysfunctions; however, only limited studies are yet available. Herein, we reported that fructooligosaccharides from M. officinalis could markedly modify the behavior, improve oxidative stress and inflammation disorder, regulate the synthesis and secretion of neurotransmitter, ameliorate the swelling of brain tissues, and reduce neuronal apoptosis. We also reported that OMO administration alters the diversity and stability of the microbial community, the expression of the genes of AD intracellular markers such as Tau and Aβ_1-42. In addition, OMO administration exerted an adequate prebiotic role in regulating the composition and metabolism of gut microbiota in an overdose antibiotics-treated IBD mice model.

Gut microbiota plays a major role in maintaining normal physiological functions in the host. The changes in gut microbiota can lead to changes in brain function, thereby affecting the host behavior (Vuong et al., 2017). Recent studies showed a significant correlation between the changes in gut microbiota and cognitive behavior (Dinan and Cryan, 2017). The modulation of gut microbiota by germ-free animals, probiotics or antibiotics intervention, and fecal microbiota transplantation (FMT) can influence the cognitive behavior of the host (Hu et al., 2016). Our data were in agreement with previous studies, showing that the gut microbiota in two AD-like model rats was altered as compared to the normal rats, as the abundance of Clostridia and Clostridiales in Aβ_1-42-induced groups increased significantly (p < 0.05), while the groups administered OMO can reverse those changes, especially the probiotic Lactobacillus and Akkermansia increased distinctly (p < 0.05).

The probiotic administration had a marked effect on the cognitive behavior. The prototype probiotic bacterium has been found to up-regulate the hormone oxytocin and systemic immune responses in order to achieve a broad range of health benefits involving wound healing, mental health, metabolism, and myoskeletal maintenance (Servin, 2004; Erdman and Poutahidis, 2016). Studies showed that Lactobacillus pentosus var. plantarum C29 from kimchi, a traditional food manufactured by fermenting vegetables (Park et al., 2014), was beneficial to health. It can also protect the memory deficits by inducing the expressions of BDNF and p-CREB in scopolamine-induced memory-deficient mice (Jung et al., 2012), anti-inflammatory amelioration of age-dependent memory impairment in Fischer 344 rats (Jeong et al., 2015), and ameliorate memory impairment and inflammation in D-galactose-induced accelerated aging mouse (Woo et al., 2014). Another study also showed that L. plantarum could attenuate anxiety-related behavior and protect against stress-induced dysbiosis in adult zebrafish (Davis et al., 2016). Lactobacilli and Bifidobacteria exhibited antagonistic activities against microbial pathogens (Servin, 2004). Our data were in agreement with previous studies showing that Lactobacillus can ameliorate memory deficiencies (Figure 4H); the relative abundance of Aβ_1-42-induced Lactobacillus was reduced starkly, especially in the high-dose group. As a result of OMO administration, the relative abundance of Lactobacillus was increased significantly; also, the MWM tests showed that the learning and memory abilities were improved (Figure 3B). These results suggested that OMO can promote the abundance of Lactobacillus and ameliorate the memory deficiencies.

Fructo-oligosaccharides (FOS) are commonly regarded as a type of prebiotics, favorably stimulating the growth of Bifidobacteria and Lactobacilli. FOS from Stevia rebaudiana roots enhanced the growth of specific strains of both Bifidobacteria and Lactobacilli, especially, with respect to their fermentation ability (Sanches Lopes et al., 2016). FOS are reserve carbohydrates with important positive health effects and technological applications in the food industry. Another previous study indicated that short-chain fructooligosaccharides could be used optimally in combination with Bifidobacterium animalis or B. longum strains for the development of synbiotic foods or dietary supplements (Valdés-Varela et al., 2017). Our results showed that FOS from M. officinalis also enhances the growth of some probiotics similar to that of Bifidobacteria and Lactobacilli.

The microbiota can modulate the changes in the gut barrier as well as metabolic and inflammatory responses. Gut barrier function is a key to maintaining a balanced response between the host and its microbiome. This highly complex system involves numerous microbiota-derived factors. Akkermansia muciniphila is positively correlated with a lean phenotype, reduced body weight gain, amelioration of metabolic responses, and restoration of gut barrier function is effectuated by the modulation of mucus layer thickness (Bland, 2016; Derrien et al., 2016; Greer et al., 2016; Henning et al., 2017; Van Herreweghen et al., 2017). Our data in D-galactose-induced deficient rats showed that the Akkermansia were increased (p < 0.05, Figure 2D), and the intestinal pathological tissue changes (Figure 1E) were improved after administration of OMO, thereby indicating that OMO administration can improve the gut barrier function via targeting the abundance of Akkermansia, but need more studies.

Increased gut permeability (leaky gut) and alterations in gut microbiota are now widely accepted as an important link with the etiology, course, and treatment of several neuropsychiatric disorders (Anderson et al., 2016). Gut microbiota-released LPS contributes to chronic inflammation and oxidative stress (Le Sage et al., 2017). Moreover, inflammation was first implicated in AD pathology and development, with the neuropathological findings of activated inflammatory cells (microglia and astrocytes) and inflammatory proteins (for example, cytokines and complement), surrounding the amyloid plaques and the nerve fiber tangles (Alkasir et al., 2017). Our study showed that OMO administration could reduce the levels of LPS in TNBS-induced IBD mice and some pro-inflammatory cytokines in both Aβ_1-42 induced deficit rats (Figure 3C) and TNBS-induced IBD mice (Figures 5C, 6). On the other hand, it can increase the levels of some anti-inflammatory cytokines, which suggested that the administration of OMO (prebiotics) can improve the host inflammatory immune response. The data also showed that OMO administration could enhance the oxidative stress, similar to elevated SOD, MDA, CAT, and inhibiting the MDA production in the D-galactose-induced deficit rats (Figure 1D). The gut microbiota is known to play a vital role in those responses (Vanegas et al., 2017), although not fully understood. Taken together, we can summarize that OMO administration influences the host inflammatory immune response and oxidative stress by regulating the gut microbiota.

Preliminary data suggested that FOS increases fecal Bifidobacteria, induce immunoregulatory dendritic cell (DC) responses, and reduce the disease activity in patients with Crohn’s disease (Benjamin et al., 2011). Moreover, Bifidobacteria are predominant bacterial species in the human gut microbiota and have been considered to exert a beneficial effect on human health by maintaining the equilibrium of the resident organisms. B. longum with FOS reduces TNF-α, CRP, serum AST levels, HOMA-IR, serum endotoxin, steatosis, and the non-alcoholic steatohepatitis activity index significantly (Malaguarnera et al., 2012). Human milk contains B. breve, Streptococcus thermophilus, and short-chain galactooligosaccharides/long-chain fructooligosaccharides with pectin-derived acidic oligosaccharides conferring a protective role against gastrointestinal infections: ameliorating the AD symptoms, modulating the immune response, binding the viral particles, and protecting against rotavirus infection (Rigo-Adrover et al., 2017). The combination of fermented formula with short-chain galactooligosaccharides and long-chain fructooligosaccharides was well-tolerated showing a low overall crying time, low incidence of infantile colic infection, and a stool-softening effect in healthy term infants (Vandenplas et al., 2017). After a broad spectrum antibiotics treatment, the IBD mice model showed that FOS from M. officinalis also increases fecal Bifidobacteria, ameliorates the symptoms of IBD, and modulates the immune response; the Aβ_1-42 induced deficient rats showed a similar effect.

The gut microbiota can regulate the activity in the peripheral and central nervous system by various means of communication including vagal nerve and adrenergic nerve activation as well as producing several molecular candidates such as neurotransmitters, neuropeptides, endocrine hormones, and immunomodulators. Host stress hormones, such as noradrenaline, might affect the bacterial activities or signal between bacteria may change the microbial diversity and actions of the gut microbiota. However, these bacteria are capable of synthesizing and releasing several neurotransmitters and neuromodulators or eliciting the synthesis and release of neuropeptides from enteroendocrine cells. Previous studies showed that Lactobacillus and Bifidobacterium species could produce short-chain fatty acids; Escherichia, Bacillus, and Saccharomyces spp. can produce norepinephrine; spore-forming microbes can produce 5-HT; Bacillus can produce dopamine, and Lactobacillus can produce acetylcholine (Wall et al., 2014; Potgieter et al., 2015; Yano et al., 2015; Alkasir et al., 2017). In this study, the monoamine neurotransmitter (NE, DA, 5-HT, and 5-HIAA) levels in the brain tissue were reduced in Aβ_1-42 induced deficient rats, and the OMO administration can reverse this decreasing tendency. Thus, we can conclude that OMO influences some bacteria that affect the synthesis and release of some neurotransmitters and neuromodulators. Similar effects were observed in mice subjected to chronic stress, where the observed behavioral, neurochemical, genetic, and neuroendocrine changes after prebiotic (fructooligosaccharides and galactooligosaccharides) administration could be mediated partially by short chain fatty acids (SCFAs) (Burokas et al., 2017); the increased levels of acetate, propionate, and n-butyrate correlated with behavior and gene expression.

We also observed novel changes in microbiota composition, especially the increase in Bifidobacterium, immunological enhancement (Figures 5, 6), and gut barrier impairment (Figure 5) in TNBS-induced IBD mice. Previous reports showed that Bifidobacterium combined with L. acidophilus, L. casei, and L. fermentum for 12 weeks can affect the cognitive function and some metabolic statuses in AD patients (Akbari et al., 2016), and B. longum 1714 reduced the stress and improved the memory in healthy volunteers (Allen et al., 2016). The abundance of Bacteroides was also increased with OMO administration in the two AD-like animal models (Figures 3, 5). Bacteroides are strict anaerobes critical since the initiation of life (Arboleya et al., 2015), and some strains have been used as probiotics. Previous studies have shown that Bacteroides fragilis could reverse the autism-like behavior in mice (Hsiao et al., 2013).

The gut microbiota contains highly diverse microbial communities that play a critical role in the metabolic, immunological, and protective functions in health. This phenomenon is influenced by several factors including genetics, host physiology (age of the host, disease, and stress) and environmental factors such as living conditions and use of medications. Increasingly, diet has been recognized as a key environmental factor that mediates the composition and metabolic function of the gut microbiota. Furthermore, the consumption of specific dietary ingredients, such as oil, fibers, and prebiotics, is an avenue that modulates the microbiota. Studies on pigs also suggests that the combination with fructooligosaccharides might represent a valuable symbiotic strategy to increase the probiotic levels of bacteria and survival in the gastrointestinal tracts for feed and food applications (Tanner et al., 2015). The administration of such oligosaccharides is attributable to high interindividual variation of the communities in fecal bacteria from pet cats and dogs (Garcia-Mazcorro et al., 2017). In this study, the fructooligosaccharides were extracted from M. officinalis, which was widely used in soup, wine, and sweetmeats in South China. Fructo-oligosaccharides are soluble fiber extensively used as prebiotics that is conventionally associated with the stimulation of beneficial bacteria such as Bifidobacteria and Lactobacilli, among other gut members. However, the mechanisms underlying the fructooligosaccharides stimulation of the beneficial bacteria are yet unknown. A previous study showed that the obtained nutrients process of bacteria require cell membrane protein machines called SusCD complexes (extracellular substrate binding proteins and SusC transporter) (Glenwright et al., 2017), in order to detect the binding capacities between SusCD and nutrients (as starch and other dietary polysaccharides) can evaluate the activity of prebiotics of nutritional ingredients. In this study, the molecular docking analysis was carried out for the evaluation of the binding capacities between SusCD and fructooligosaccharides from M. officinalis. The molecular docking study was conducted using the CDOCKER protocol for the four polysaccharides from M. officinalis using the Discovery Studio 2.5 (DS2.5) Provisional software. A total of four components (Figures 14A,B) from M. officinalis were assimilated by literature search, and 1048 poses were generated for all the compounds investigated. The docked poses were ranked by the CDOCKER-ENERGY, and the top 10 poses with the co-crystal ligand for SusCD were retained (Figure 14). The data revealed 4 hits namely, compounds 3 (Nystose), 4 (F-fructofuranosylnystose), 5 (fructooligosaccharide, GF5), and 6 (fructooligosaccharide, GF6) with CDOCKER-ENERGY -52.0244180, -75.5881, -101.88, and -110.387, respectively (Figure 14C). This indicated that the 4 polysaccharides might exert a potent binding activity on SusCD. The interaction between the SusCD protein and the four compounds was further analyzed using the receptor-ligand interaction module in DS. The analysis between SusCD and compound 3 revealed that 9 hydrogen bond interactions appeared in the docked pose. The analysis between SusCD and compound 5 revealed that 9 hydrogen bond interactions appeared in the docked pose. The analysis between SusCD and compound 5 revealed that 7 hydrogen bond interactions appeared in the docked pose. The analysis between SusCD and compound 6 revealed that 9 hydrogen bond interactions appeared in the docked pose. These results suggested that the fructooligosaccharides from M. officinalis could be absorbed sufficiently by bacteria with the help of SusCD, serving as optimal nutritional ingredients or prebiotics for bacteria.

The changes in behavior and gut microbiota, as a result of different concentrations of Aβ_1-42, coincided with the alterations in gene expression in critical brain and intestinal regions. The transcriptome analysis of intestine and brain tissues at the 5th week induced by Aβ_1-42 showed that the expression of some genes changed significantly (Figures 10, 12). Furthermore, to understand the genetic pathways of the gut-brain axis, the significant DEGs both in the brain and intestine were analyzed to identify the gene elements by an interactive information using the Wayne chart (Figures 15A–C), which revealed that the interactive genes were increased by the Aβ_1-42 levels (14 for Aβ-10 vs. normal, while 25 for Aβ-20 vs. normal, p < 0.05). The KEGG pathway enrichment analysis showed that the DEGs were primarily enriched in protein digestion and absorption and platelet activation for Aβ-10 induced group (Figure 15D). This phenomenon might be attributed to the colonic bacteria that might not be able to run well or are lost, the non-digestible peptides and proteins (collagen) could not be fermented, and some short-chain fatty acids such as butyrate, propionate, and acetate are deficient, which resulted in platelet activation. Thus, the expressions of Col1a1, Col3a1, and Col14a1 mRNA were changed. With the altered mRNA expression of Lamc2, Sstr3, Nts, Spp1, Il6r, Vip, Kcnk3, and Nr4a1, the PI3K-Akt signaling pathway, neuroactive ligand-receptor interaction, focal adhesion, and ECM-receptor interaction were activated in the Aβ-20-induced group (Figure 15E). With the increased levels of Aβ_1-42 levels, the mRNA expressions of Igsf8, Kcnk3, and Mef2c were changed, and the KEGG pathway enrichment analysis showed that the aldosterone synthesis and secretion and MAPK signaling pathway were influenced (Figure 15F). Together with the changes in the gut microbiota community diversity shown in Figure 8, and the transcriptome analysis data of intestine and brain, we concluded that the Aβ_1-42 levels in hippocampus interact the gut and microbiota, although further studies are essential for substantiation.

Antibiotics intervention in APPSWE/PS1ΔE9 mouse model suggests that the diversity of the gut microbiota community can regulate the host innate immunity mechanisms that impact Aβ amyloidosis (Minter et al., 2016). The fecal microbiota transplantation implemented from Aβ precursor protein (APP) transgenic mice to non-transgenic wild-type mice showed a drastically increased level of cerebral Aβ levels, thereby indicating a microbial involvement in the development of Aβ pathology, and microbiota contributes to the development of neurodegenerative diseases (Harach et al., 2017; Liu et al., 2017). We also observed that the diversity in the gut microbiota community altered with the levels of Aβ and the induced time (Figure 8). Although the complex networks of communication between the gut microbiota and the brain are not yet fully elucidates, it is clear that prebiotics strongly modulates the ecology of the microbiota. However, the role of the microbial composition and the vast quantity, diversity, and functional capabilities of all these gut microorganisms on the brain and behavior are yet to be determined.

Taken together, these data provide further evidence for a beneficial role of fructooligosaccharides (prebiotics) from M. officinalis and the effects on microbiota-brain-gut axis in AD, but need more studies. This study characterized OMO as a promising naturally occurring chemical constituent and suggested microbiota-brain-gut axis as a putative new therapeutic target for the treatment of various neurological diseases by using M. officinalis in conventional medicine.

The animal protocols used in this work were approved by the Institutional Animal Care and Use committee of the Center of Laboratory Animals of the Guangdong Institute of Microbiology (GT-IACUC20160426).

DC designed the study, carried out the computational analyses and wrote the manuscript. JY and GL collected animal physiological data and fecal samples and extracted ruminal DNA. XY and XT collected animal physiological data and brain samples. DC, TY, and OS collected data regarding the microbial metabolic networks and transcriptome analysis. YX and QW helped to design the study and to develop the metagenomic analysis tools and wrote the manuscript. GZ helped with computational tool development and statistical analyses (PCA). All authors read and approved the final manuscript.

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.