In recent times, researchers have begun to ask questions about the presence of a microbiome in the brain, especially following reports demonstrating the presence of microbes in the brain specimens of HIV-infected subjects (83). While the healthy human brain continues to be considered a sterile environment, the role of the microbiome in other regions of the body in modulating brain development, neurochemistry and behaviors (Table 2) have been examined (84–88).

Human brain development begins in-utero and extends through adolescence into early adulthood. It is an intricate process that involves the migration of cells over long distances (in-utero), with extensions of their cell processes over even greater distances (4,89). The protracted course of pre- and post-natal brain development and the complexity of the process increase its vulnerability, providing a window of opportunity for environmental factors to exert influences that modulate brain structure, function and neurotransmitter/receptor expression (4,76, 87). In the last decade or more, there have been suggestions that the microbiota (especially the gut) influences brain development and maturation (86,90).

While examining the impact of normal gut microbiota on brain development in mice, Diaz Heijtz et al. (86) concluded that early-life but not late-life exposure of germ-free mice to normal gut microbiota was associated with the modulation of motor activity and anxiety-like behaviors (86). These modulatory effects on behaviors occurred via their ability to influence canonical signal pathways, and release of brain neurotransmitter (including serotonin, dopamine and noradrenaline) and synaptic-related proteins like synaptophysin (86). Sudo et al (90) reported that post-natal reconstitution of germ-free mice with normal gut microbiota flora containing Bifidobacterium infantis improved the response of germ-free mice to stress by influencing adrenocorticotrophic hormone levels in mice (90). Studies in rodents have also demonstrated the antidepressant effects of Bifidobacterium infantis via its ability to normalize the concentration of pro-inflammatory cytokines and tryptophan (91-92). Gut microbiota have also been reported to influence the expression and synthesis of receptors and neurotransmitters (including serotonin) which are important in brain development (93). Desbonnet et al. (91) also reported that in rat pups of dams colonized with Bifidobacterium infantis a decrease in frontal cortex concentration of serotonin, and an increase in plasma levels of kynurenic acid and tryptophan were observed, when compared to pups of non-colonized controls (91).

During the first three years of life, microbial colonization of the different niches on and within the human body occurs rapidly. There is an initial domination of the microbiota by members of the Actinobacteria and Proteobacteria phyla, and then a gradual shift towards Firmicutes and Bacteroidetes domination as adulthood is reached (94-95). This process also coincides with the period of brain development (spanning from early-life to adolescence) during which synaptogenesis and pruning goes on in the brain (94-97). In the growing infant, there is increasing advocacy for the consideration of the gut microbial communities as modifiable influencers of brain development and host behavior (98–101) The involvement of the gut microbiota in the programming of the brain circuitries that modulate a number of behavioural responses including stress, anxiety, cognition, motor coordination and social interaction during early brain maturation (90, 99,102–105) have also been reported.

Gut microbiota depletion (due to use of antibiotics from early adolescence) was associated with cognitive deficits, alteration in tryptophan metabolism, and a significant decrease in the expression of oxytocin, brain derived neurotropic factor and vasopressin in the adult brain (106). There have been suggestions that gut microbial depletion or dysbiosis may play a role in the pathogenesis of attention-deficit-hyperactivity and autism spectrum disorders. A number of studies have also implicated the central nervous system in the pathogenesis and pathophysiology of some functional gastrointestinal tract diseases including inflammatory bowel disease and irritable bowel syndrome (107-108). Studies have also demonstrated the beneficial effects of prenatal prebiotics and probiotic interventions in disease prevention (50-51,109) suggesting that these could be effective novel methods in preventing and/or managing neurodevelopmental disorders like autism and attention deficit hyperactivity disorder. Studies demonstrating the concurrence of postnatal neurodevelopment and the establishment of the gut microbiota provide evidence of a possible bi-directional modulation or regulation of the maturation of one another (110).

The constitution of the skin microbiome (like that of the gut) is a reflection of the synergy of the functional metabolic activity of the host tissues and resident microbes, which are greatly influenced by host behavior and the surrounding environment. In the last decade, there have been suggestions that the brain interacts with the skin, like in a “skin-brain” axis. Also, the extensive innervation of the skin has led researchers to consider it a sort of “diffuse brain” (111). The occurrence of skin disorders like psoriasis and atopic dermatitis is now being associated with the development of behavioral disorders like posttraumatic stress disorder, anxiety, and depression (112–114). Some studies have also shown that the colonization, virulence, and adhesion of skin microbes could be modulated by peptide neurohormone, neurotransmitters and steroid hormones (14). These neuropeptides such as vasoactive intestinal peptide, catecholamines, calcitonin gene-related peptide, nerve growth factor and substance P, can be secreted by some skin cells. However, while the idea of a skin-brain or brain–skin axis is plausible, the full extent of this relationship (unidirectional or bidirectional) remains largely speculative.

Gut microbiota influences on the central nervous system were first reported following observations that laxatives and/or oral antibiotics could mitigate the progression of hepatic encephalopathy (87). In the symbiotic relationship between man and microbes, the human body provides lodgings and nutrition, while microbes repay this favor by ensuring the maintenance of health/well-being, and prevention of outsider invasion (115). The roles of microbes in ensuring optimal brain development and functioning have been reported (86, 87, 90,103,1116). Also, current research is beginning to report the importance of the microbiota in modulating host behaviors (Table 3).

Studies in germ-free or antibiotic-treated fruit-fly (Drosophila melanogaster) showed hyperlocomotion and increased walking speed, which was reversed by mono-colonization with the fruit-fly commensal Lactobacillus brevis (117). In another study by Akami et al (118), increase in foraging behaviors of Oriental flies (Bactrocera dorsalis) was determined by the intestinal microbiota consisting mainly of members of the Enterobacteriaceae family (118). In mammals, the importance of the microbiota in modulating dietary behaviors and metabolic profile been reported. In a study in which cultured and uncultured human fecal microbiota were transferred from weight-discordant twins (obese and lean) to germ-free mice, it was observed that the germ-free mice colonized with obese fecal microbiota showed significantly-higher weight increases compared to mice colonized with microbiota from the lean twin (66). Also, co-habitation of these two mouse groups was associated with a transformation of the metabolic profile of the obese microbiota mouse groups to a lean-like state (66). There have been reports associating obesity and related dysmetabolism with an increase in the incidence of mood disorders, depression and cognitive decline (119-120). The possibility of replicating these features in murine models of obesity (121) has allowed researchers to examine the role of diet and dysmetabolism on neurobehavior (122-123). Bruce-Keller and colleagues (122) reported that mice colonized with high-fat diet microbiota (with a preponderance of bacteria from the phylum Firmicutes and a depletion of Bacteroidetes) had alteration in exploratory activity, cognition and stereotypic behavior compared to controls (122).

Diaz Heijtz et al. (86) reported increased locomotor activity and rearing behaviours in germ-free mice, while Desbonnet et al (104) reported evidence of increased self-grooming and alteration in social interaction evidenced by avoidance of conspecifics, decreased preferences for novel conspecifics, and decreased time spent in social investigation (104). In another study using germ-free mice Gareau et al (124) reported impairment of short-term recognition memory and working memory (124), while Zheng et al (125) reported antidepressant-like response (125). Gareau et al (124) also observed that colonization of mice with pathogenic bacteria Citrobacter rodentum was associated with accentuation of stress–induced memory loss (124).

In humans, there have been reports of a loss of microbial diversity with aging (with a predominance of Bacteroidetes and a decrease in Firmicutes) and the presence of an age-associated cognitive decline in the same age group (126). In a study examining gut microbial diversity of cirrhotics with or without hepatic encephalopathy (HE), it was reported that (compared to healthy controls), cirrhotics had a preponderance of commensals from the phyla Enterobacteriaceae, Alcaligeneceae, and Fusobacteriaceae, with low concentrations of Lachnospiraceae and Ruminococcaceae (127). However, in cirrhotics (with HE) a higher concentration of Alcaligeneceae and Porphyromonadaceae Veillonellaceae was observed, which correlated positively with cognitive-deficits, high levels of pro-inflammatory markers (TNF-α, IL-2, IL=6 and IL-13) and endotoxemia (127).

Oral antibiotics–induced microbiota depletion in mice have also been reported to transiently alter the microbial composition of the gut, resulting in an increase in exploratory behavior and an increase in the hippocampal expression of brain- derived neurotropic factor (84). These changes did not occur with intraperitoneal administration of the antibiotics (84). A few studies have also reported that in germ-free mice, the basal behavior was anxiolysis (86, 102) compared to the expected aversion for open spaces (a protective feature) in mice with normal gut microbiota. In the germ-free mice, this alteration in anxiety-related behaviors was accompanied by an increase in the expression of brain derived neurotropic factor (BDNF) in the hippocampus (dentate gyrus), a decrease in the expression N-methyl-D-aspartate receptor subunit in the central amygdala, and a decrease in serotonin receptor expression by the dentate gyrus region of the hippocampus (102). In another study, Bravo e al (128) reported that long-term treatment of healthy mice with Lactobacillus rhamnosus probiotic was associated with anxiolysis in the elevated plus-maze and decreased immobility time in the forced swim test; there was also a significant decrease in the levels of stress-induced corticosterone levels and region-specific increase in the expression of γ-amino butyric acid receptor mRNA. These features were however abolished with vagotomy (128), suggesting that these effects were vagus nerve dependent.

Currently, the role of the gut microbiota in social interaction and the development of social behavior has being studied extensively (123,129–132), using either germ free mice, or following colonisation with probiotic (132). Peri-conceptional exposure to antibiotics has also been shown to alter social behavior and induce anxiety (133). The use of antibiotics to deplete mouse gut microbiota during adolescence was associated with impairment of social memory in adulthood (106), features which corresponded with a reduction in mRNA levels of vasopressin and oxytocin in the mouse hypothalamus (106). In general, alterations in social memory and neurobehavior could also be attributed to the ability of gut microbiota to modulate myelination of neurons in the prefrontal cortex (134). Hoban et al. (134) also reported evidence of abnormal hypermyelination of axons in male germ-free mice (134). Studies in the mouse model of autism have also revealed that oral treatments with Bacteroides fragilis were associated with improvements in social interactions, stereotypic behaviors, anxiety-related behaviors and sensorimotor deficits (3).

The role of the microbiota in the modulation of neurobehavior and/or brain neurochemistry has also been demonstrated by studies using prebiotics or probiotic interventions. In rodents, probiotic intervention with Bacteroides infantis was associated with decreased immobility time in the forced swim paradigm (92). Buffington et al. (132) demonstrated that a maternal diet that is high in fat caused microbiota dysbiosis, deficits in social interaction, preference for social novelty and a decrease in the neuronal oxytocin immunoreactivity in the paraventricular nucleus of the hypothalamus in the maternal high-fat diet offspring, compared to normal diet fed controls. However, there was restoration of behavioral deficits, microbiota dysbiosis and oxytocin immunoreactivity following reconstitution with Lactobacillus reuteri probiotic (132), demonstrating that these deficits were possibly due to low-concentrations of Lactobacillus reuteri in mouse pups (132). Colonization with specific members of the Bifidobacterium breve species have been associated with alteration of the fatty-acid composition of the brain in favor of fatty-acids that improve of cognitive abilities (135). Also, in rats, probiotic formulations of Lactobacillus helveticus and Bifidobacterium longus were associated with a decrease in anxiety-related behaviors measured in the conditioned defensive burying test (136). The use of prebiotic intervention to promote selective proliferation of commensal bacteria has also been reported to modulate neurobehavior. Tarr et al (137) reported that a diet rich in human milk oligosaccharides 3′Sialyllactose or 6′Sialyllactose reduced stress-induced anxiety-like behavior in the open field and light/dark preference paradigms (137).

In humans, the benefits of probiotic intervention on the modulation of behavior have also been demonstrated. In healthy adults with baseline depressive symptoms or patients with chronic fatigue syndrome, the probiotic administration of Lactobacillus casei was associated with improvement in symptoms of depression and anxiety respectively (138-139). Probiotic formulations of Bifidobacterium longum and Lactobacillus helveticus administered to otherwise healthy subjects was also associated with a decrease in psychological distress including somatization, anger, anxiety and depression, using relevant clinical indices (136). Administration of prebiotic trans-galactooligosaccharide to irritable bowel syndrome sufferers was associated with significant reduction in anxiety and also an increase in the concentration of bifidobacteria in fecal samples (140).

A number of other studies have also reported the possible influences behavioral modification can exert on microbial density or diversity (141-142). Exposure to social disruption stress in rodents was associated with a decrease in the relative abundance of bacteria of the genus Bacteroides, and an increase in those belonging to the genus Clostridium. There was also an associated increase in the circulating levels of interleukin 6 (IL-6) and monocyte chemo-attractant protein (MCP)-1 (142) Increased levels of IL-6 and MCP-1 have also been observed to correlate positively with increased concentration of bacteria from 3 genera including Coprococcus spp, Pseudobutyrivibrio sp, and Dorea spp (142). In mice exposed to chronic restraint stress, a significant increase in Citrobacter rodentium colonization with an associated increase in the gene expression for tumor necrosis factor alpha in colonic tissue has been demonstrated (141). Accumulating evidence of the ability of gut microbiota to influence neurobehaviour and the alteration of microbiota diversity and density due to influences from neurobehavioural modifications further confirm the possible existence of a bidirectional communication that allows brain signals to influence sensory, motor and secretory functions of the gut, while visceral messages from the gut can in turn influence neurobehavior, social behavior and brain chemistry (123).

Also the relationships that link the gut microbiome, skin microbiome, and the brain could be crucial in understanding the contributions of bacteria to the development of odor cues, which is known to contain important information about the host’s social relationships and social behaviors. Studies have shown that the intensity of the human odor and/or its chemical composition is closely-linked to the skin’s microbial diversity and composition (143–145). The interaction of the skin microbiota, the sweat glands, and the brain is important in the creation of human odors that allows kith recognition of kin, the detection of fear in animal prey, and sexual behaviors in animals (146). There have been reports suggesting that Staphylococcus epidermidis which is part of the normal skin microbial flora has the ability to degrade leucine content of sweat causing foot odor. Also, colonization of the plantar skin by Bacillus subtilis has been associated with strong foot odor (143). Absence of axillary odor has also been linked to paucity of gram positive bacteria (145).

3.2.1. Signaling pathways/mediators and molecular alterations involved in behavior/microbiota interactions

The communication between the gut microbiota and the brain affects behavior. This communications occurs via neural, immune and endocrine pathways that connect the microbiota to the central nervous system (147). The interactions (Table 4) between microbiota and brain alter neurobehavior, signal pathways, neurotransmitters and hormones (Figure 1); it also modulates the expression of receptor mRNA in different regions of the brain (86, 90, 102, 124, 147).

Figure 1

Showing how the gut microbiota via neural, homonal or endocrine mechanisms modulatesbrain function and chemistry.

Table 4 Microbiota-mediated neurochemical and molecular alterations

Model	Microbial Composition/Mediators	Brain region/tissue	Neurochemical/Molecular alteration	Reference
Mice	Lactobacillus rhamnosus	Hippocampus, amygdala, LC Prefrontal cortex, cingulated gyrus	↑ GABAAα2 mRNA ↓ GABAB1b mRNA ↓ GABAAα2 mRNA ↑GABAB1b mRNA	(128)
Mice	GF	Hippocampus	↑ serotonin and 5-hydroxyindoleacetic	(103)
Mice	GF	Dentate gyrus	↓ serotonin receptor mRNA	(102)
Mice	GF	Dentate gyrus Cornus ammonis 1 region of the hippocampus, amygdala and cingulate cortex Hippocampus	↓BDNF ↑BDNF ↑neurogenesis	(102) (86, 90, 124) (149)
Mice	GF	Cortex Cortex and hippocampus Central amygdala	↓ N-methyl-D-aspartate receptor; the NR1 ↓NR2a ↓NR2b	(90) (90) (102)
Mice	GF	Striatum Hippocampus	↓ nerve growth factor-inducible clone A ↓Gopamine receptor ↑dopamine receptor	(86) (86)
Mice	GF	Cerebral cortex	↑Dopamine	(150)
Mice	Antibiotic-induced depletion of microbiota	Hippocampus	↑BDNF	(84)
Mice	Trichuris muri	Hippocampus	↓ BDNF mRNA	(151)
Mice	Citrobacter roduntum	CA1 region of the hippocampus	↓BDNF expression	(124)
Mice	Bifidobacterium infantis	Frontal cortex Amygdala Paraventricular nucleus	↓5–hydroxyl indole acetic acid ↓3,4-Dihydroxyphenylacetic acid ↓cFos expression	(92) (90)
Mice	GF	Hypothalamus	↑corticotropin-releasing factor gene expression and protein levels ↑Plasma ACTH	(90)
Rats	Bifidobacterium pseudocatenulatum Bifidobacterium animalis subsp lactis and Propionibacterium jensenii	Hypothalamus	↓dopamine and adrenaline ↑ACTH (reversal of maternal-separation stress induced derangements) reversal of maternal-separation stress induced biochemical derangements and microbial density	(157) (155)
Rats	IBS microbiota	Colonic	colonic genes involved in glucocorticoid receptor signaling	(156).
Mice	GF or mice deficient of SCFAs receptor FFAR2a (GPR43)	Microglia	↓ Fcgr2β, Mapk8, IL-1α, and Cd86., Janus kinase 3 and the signal transducer and activator of transcription 1 (Stat1 ↑ Nfkbiα) and the central microglia transcription and survival factors (Sfpi1 and Csf1r)	(166)
Mice	GF	Hippocampus,hypothalamus	↑Apoptosis,	(153)
Mice	GF	Prefrontal cortex	Hypermyelination, overexpression of myelin-related genes	(134)

Neural pathways involve efferents from the central nervous system to the gut, and afferents from the gut to the central nervous system. Efferent signals that are conducted from the central nervous system to the gut modulate gut motility, secretion, and permeability, which in turn affect microbiota composition. The efferent pathway relies on neurotransmitters such as acetylcholine. The afferent signals are transmitted from the gut to the central nervous system via the vagus nerve which is capable of recognizing microbial products and cellular components. The changes that had been observed in the brain include region-dependent alterations in levels of gamma-aminobutyric acid (GABA) mRNA gene expression (128). Bravo et al (128) reported that colonization with Lactobacillus rhamnosus was associated with a decrease in the expression of GABA_B1b mRNA in the amygdala, hippocampus and locus coeruleus; with an increase in the prefrontal cortex and cingulate cortex. The expression of GABA_Aα2 mRNA was decreased in the cortical region, and increased in the hippocampus/amygdala (128). The role of GABA in maintaining central excitation/inhibition balance has been reported (148). Studies examining the impact of gut microbiota on neurotransmitters concentration in germ-free mice have reported that hippocampal concentrations of serotonin and 5-hydroxyindoleacetic acid (metabolite of serotonin) were elevated compared to control (103); while a decrease in the expression of the serotonin receptor mRNA was observed in the dentate gyrus (102). The expression of brain derived neurotropic factor (BDNF) increased in the dentate gyrus (102), and decreased in the cornus ammonis (CA) 1 region of the hippocampus, amygdale and cingulated cortex (86, 90, 124). While the content and expression of BDNF is modulated by the gut microbiota, brain derived neurotropic factor itself regulates neurogenesis, supporting the result of a recent study that demonstrated increased hippocampal neurogenesis in adult germ-free mice (149). There was also reduced expression of different subunits of the N-methyl-D-aspartate receptor; the NR1 in the cortex (90), NR2a in the cortex and hippocampus (90) and NR2b in the central amygdale (102). Diaz Heijtz et al (86) reported a decrease in the expression of nerve growth factor-inducible clone A and dopamine receptor in the striatum, and an increase expression of dopamine receptor in the hippocampus of germ–free mice (86).

Antibiotic-induced depletion of gut microbiota was associated with increased expression of BDNF in the hippocampus (84). Matsumoto et al (150) also reported increase in dopamine concentration in the cerebral cortex of germ-free mice compared to mice colonized by normal microbial flora (150), which supports the increased exploratory activity reported by previous studies (86, 102). Colonization of mice with non-invasive parasite Trichuris muri was associated with colonic inflammation and anxiety-like behavior, while in-situ hybridization revealed a decrease in hippocampal expression of BDNF mRNA (151). Gareau et al (124) also reported decreased BDNF expression in mice colonized with citrobacter roduntum following induction of stress (124). Intervention with probiotic Bifidobacterium infantis was associated with a decrease in the frontal cortex concentration of 5– hydroxyl indole acetic acid and amygdaloid cortex concentration of 3, 4-Dihydroxyphenylacetic acid, metabolites of serotonin and dopamine respectively (92); and a decrease in the expression of cFos in the paraventricular nucleus (92), neurotransmitter modulations that could be related to the antidepressant effects observed in the forced swim test. A number of these studies confirm previous reports associating hippocampal depletion of neurotransmitters with the development of depression and depressive-like symptoms (152). Neuronal apoptosis has also been linked to the gut microbiota. In the hypothalamus and some regions of the hippocampus, increased apoptosis was observed in neonatal germ-free mice when compared with control mice, this was also accompanied by increase microbial count and density (153). Neuron myelination and myelin plasticity have also been reported to be impacted by the gut microbiota; in germ-free mice, evidence of hypermyelination of the axons in the prefrontal cortex as well as over-expression of myelin-related genes have been reported (134).

The endocrine pathway involves the hypothalamo-pituitary-adrenal (HPA) axis, and the secretion of hormonally-active peptides such as peptide YY. Changes in the HPA axis response can be triggered by impulses originating from the brain (e.g. due to stress), or from the gut microbes who can positively or negatively regulate the HPA axis (90). In germ-free mice, restraint stress has been observed to cause increased HPA response, leading to an elevation of hypothalamic corticotropin-releasing factor gene expression/protein levels, a decrease in cortical/hippocampal brain-derived neurotrophic factor, and an elevation of plasma adrenocorticotropic hormone (ACTH)/corticosteron (90). However, colonization by specific microbiota at certain periods in their life can either normalize their HPA response to stress, or make them to continue to show exaggerated responses (90). This influence has been suggested to profoundly impact behaviors like anxiety, attention, and certain aspects of memory.

Probiotic intervention (from postnatal day 2 to 21) with a strain of Bifidobacterium pseudocatenulatum reversed neuroendocrine and neurobehavioural changes induced by maternal deprivation induced stress (154). The neuroendocrine response included an attenuation of stress-induced increase in corticosterone levels, and catecholamine levels in the hypothalamus. A down-regulation of maternal-deprivation stress–induced increase in intestinal levels of interferon gamma and catecholamine (dopamine and adrenaline) was also demonstrated (154).

Maternal probiotic supplementation with Bifidobacterium animalis subsp lactis and Propionibacterium jensenii has also been shown to reverse the changes induced by maternal-separation stress or the adult-stress protocol on biochemical parameters and microbial density (155). In rats pups exposed to maternal-separation stress, an increase in plasma levels of ACTH and fecal concentration of enterococci, Escherichia coli, and clostridia has been shown. In adult animals exposed to stress, an increase in both ACTH and corticosterone levels was observed; there was also a decrease in the density of Bifidobacteria, and an increase in Escherichia coli and Bacteroides (155). Transplantation of fecal microbiota from patients with irritable bowel syndrome into germ free mice has also associated with anxiety-related behavior, increased gastrointestinal transit time, alteration in the intestinal barrier (156). There was also increased expression of colonic genes involved in glucocorticoid receptor signaling (156).

In line with the bidirectional communication that occurs in the gut-brain axis, the neurons of the autonomic nervous system and neuroendocrine mediators carry outputs from the brain to the gut. Biologically active peptides including peptide YY (PYY), neuropeptide Y and pancreatic polypeptide are expressed at distinct levels of the gut–brain axis (157). Peptide YY and pancreatic polypeptide are expressed exclusively by the endocrine cells of the digestive system, while NPY is expressed at all levels of the gut–brain and brain–gut axis (157). PYY is synthesized in the gut by enteroendocrine cells with the G protein-coupled receptor FFAR3 (Gpr41) which sense dietary proteins/fats, and microbial-derived short-chain fatty acids (158). The PYY that is transported to the brain affects feeding behavior, while in the GIT, it by triggers satiety, reduces food intake, and slows the GI motility (158).

The immune pathway involves the influence of microbial antigens and metabolites on the immune system, which in turn affects the central nervous system. It is known that specific microbes and their products can use different mechanisms to influence the development and functions of particular subsets of immune cells (159). Microbial antigens might trigger the immune system and contribute to the pathogenesis of neurodegenerative disorders such as multiple sclerosis which has profound behavioral components (160-161). On the other hand, the microbial metabolites that affect immune function include chemical mediators such as small chain fatty-acids (acetic acid, propionic acid and butyric acid) and tryptophan metabolites. The short chain fatty acids (SCFAs) are mainly produced by large-intestinal bacterial fermentation of the complex carbohydrates. They modulate gut motility, enhance intestinal epithelium integrity, increase mucus production, promote regulatory T-cells, and inactivate nuclear factor kappa B (162-164). In the gastrointestinal system, SCFAs induce the release of neuropeptides and hormones including glucagon-like peptide 1 and PYY from intestinal enteroendocrine cells (165), via the activation of the G protein-coupled receptors (GPRs; GPR41, GPR43, and GPR109A), and also by acting as epigenetic regulators. SCFAs modulate gene expression through the inhibition of histone deacetylases.

In the central nervous system, SCFAs are important in the modulation of microglial maturation and homeostasis (166). The microglia in germ-free mice exhibited global defects including altered cell size, immature phenotypes and impaired innate immune response. These features were also evident in mice deficient of SCFAs receptor FFAR2a (GPR43) (166). Genome-wide mRNA studies revealed significant differences in mRNA profiles of microglia genes of germ-free compared to specific-pathogen (SPF) mice (166). A total of 198 genes were observed to be downregulated while about 173 genes were unregulated in microglia from germ-free mice, compared controls (166).A number of the downregulated genes were genes that had been linked to cell activation including Fcgr2β, Mapk8, IL-1α, and Cd86. Also diminished were genes such as Janus kinase 3 and the signal transducer and activator of transcription 1 (Stat1), which are linked to signaling of type I IFN receptors (166). Upregulated genes in the germ-free mice microglia include a number of genes responsible for the inhibition of transcription (Nfkbiα) and the central microglia transcription and survival factors (Sfpi1 and Csf1r) (166), which are usually downregulated in microglia (167-168).

Bacterial populations in the gut have also been reported to be capable of producing hormones and/or neurotransmitters similar to that produced by humans (169). Members of the Bifidobacterium species produce gamma-amino butyrate (GABA), Lactobacillus species secrete acetylcholine and GABA, while Escherichia produce norepinephrine, dopamine and serotonin. Serotonin is also produced by the Streptococcus and Enterococcus spp, while dopamine and norepinephrine is produced by members of the Bacillus species. These neurotransmitters that are produced by microbes could also impact neurobehaviour (169-170).

The microbiota contributes to the maintenance of the central nervous system structure and function through direct interactions between the enteric nervous system and the gut microbiota (171-172), or indirectly through the modulation of endocrine, immunological, and neural pathways (173). Gut microbiome dysbiosis has been associated with the development of neuropsychiatric disorders, including bipolar disorder, anxiety disorder, schizophrenia, autism, chronic fatigue syndrome, major depressive disorder and stress (174–176). Neurodegenerative diseases like Alzheimer’s disease, Parkinson’s disease, and stroke have also been linked to microbiome dysbiosis (177).

A number of studies (Table 5) have demonstrated the presence of alterations in specific microbes (or their metabolites) with the occurrence of central nervous system disorders (55,178–182). Following the examination of fecal samples of children with autism, Finegold and colleagues (183-184) reported low levels of microbes within the phyla Actinobacteria and Firmicutes, and an increase in the concentration of Proteobacteria and Bacteroidetes, when compared to healthy controls (183-184). The presence of high loads of Bacteroidetes, Bacteroides vulgatus, and Desulfovibrio species were observed in the feces of severely- autistic children (183). Song et al. (178) reported an abundance of Clostridia in children with autism who had gastrointestinal disorders (178), while Kang et al (185) reported evidence of a decrease in microbial diversity and the presence of lower levels of Coprococcus, Prevotella, and Veillonellaceae in stool samples of autistic children (185). In the valproic acid model of autism in rodents, alterations in the concentrations of Firmicutes and Bacteroidetes occurred alongside autism-like social behaviors (186). In children with attention deficit hyperactivity disorder, an increase in fecal Bifidobacterium has also been reported (187). In Rete syndrome, alteration of the gut microbiota has also been described; and examination of stool samples from patients revealed a decrease in microbial diversity and an increase in levels of Bifidobacterium, Bacteroidaceae, Actinomyces, Clostridium Enterococcus, Lactobacillus, Escherichia, and Shigella species (180). In patients with major depressive disorder (MDD), there have been reports suggesting a preponderance of Bacteroidetes phyla and a decrease in Lachnospiraceae (188). Jiang et al. (189) reported an increase in the concentration of Proteobacteria, Bacteroidetes and Actinobacteria in depressed patients; while Zheng et al (125) reported a relative abundance of Actinobacteria and a decrease in Bacteroidetes in patients with MDD compared to healthy controls (125). A cause-effect relationship has also been demonstrated between the microbiota dysbiosis and mood disorders, using fecal transplantation studies (125, 190). Transplantation of fecal microbiota from depressed patients to rodents was associated with the induction of depression-like behavior (125). Anxiety and anhedonia were also reported in rats colonized with fecal microbiota from depressed patients (190).

A number of neurodegenerative disorders have been associated with microbiota dysbiosis. In Parkinson’s disease, it had been suggested that the pathogenesis may have actually arisen from the abnormalities in the gut before spreading via the gut-brain axis to the brain (191). However, what is generally obvious is increasing evidence in support of the existence of a brain-gut interaction (107,110), and possibly the involvement of the brain-skin cross talk (192) in the pathophysiology of a number of neurodegenerative disorders. Examination of fecal samples from Parkinson’s disease patients revealed a decrease in the population of Enterobacteria and Prevotella species (179). The transfer of microbiota from new-onset, treatment-naïve PD patients to groups of germ-free mice has also been shown to lead to worsening motor deficits (193). In Alzheimer’s disease, an association has been established between the presence of beta-amyloid plaques and microbial dysbiosis (182). Alzheimer’s disease has also been associated with alterations in microbial diversity and population. Studies using transgenic mice revealed Alzheimer’s disease-induced alteration of microbial diversity, with alterations in the populations of Bacteroidetes and Firmicutes; specifically lowering the population of Akkermansia and Allobaculum, and increasing the population of Rikenellaceae (182).

In recent times, the brain-skin crosstalk has been proposed, as reported by studies that have demonstrated over-activation of neuropeptides in skin disorders like psoriasis and atopic dermatitis; and revealed skin conditions related to neuroimmunological stress (194–197). Mijouin et al. (113) also reported that exposure of the skin microbe Bacillus cereus to substance P was associated with increased cytotoxicity, increased caspase-1 activity, and morphological changes in the actin cytoskeleton (113). In neurodegenerative disorders like Parkinson’s disease and Alzheimer’s disease, direct alteration of the skin microbial density or diversity is rarely reported. Instead, the evidence of skin involvement would include an increase in melanoma and non-melanoma skin cancers in Parkinson’s disease (198), or altered skin physiology in Alzheimer’s disease (199). However, the precise role of the skin microbiome in all these is still a subject of research.

Based on available evidence, it is easy to deduce that manipulation of the microbiota may be a therapeutic tool in the management or prevention of behavioural disorders. Along this line, the study of how this knowledge can be used specifically for the management of behavioral disorders is an advancing field that holds a lot of promise. Human behavioral disorders that have been linked to microbiota dysbiosis include ADHD, ASD, schizophrenia, bipolar disorders, major depression, anxiety disorders, obsessive-compulsive disorder, eating disorders; and specific neurological disorders such as AD, PD, and MS (200). However, beyond the recognition or identification of a link, the questions is to what extent can manipulation of the microbiome be used in the therapy of these disorders, and what the future holds if therapy is steered in this direction. Also, while the global market for agents that may be of benefit in the prevention or management of dysbiosis continues to expand, the extent to which they may be used in the clinical management of behavioral disorders remains largely unknown. Again, the multifactorial origin of a number of disorders implies that targeting the microbiome alone is not likely to be enough for therapy. However, some scientists are in support of the idea of targeting the microbiota as novel therapeutic options in the management of neuropsychiatric and neurodegenerative diseases. Their support for the exploration of this novel approach to therapy has been backed by a number of theories, including: a) the “old friend” hypothesis, b) the gut microbiota hypothesis, and c) the leaky-gut theory.

The old friend theory suggests that a symbiotic relationship has existed between man and microbes from time immemorial (201), this relationship is crucial to the maintenance of the balance of nature as well as the health and well-being of humans. The transformations that have occurred as a result of modernization (leading to changes in lifestyle, diet and healthcare) have greatly reduced man’s exposure to microbes, which has in turn altered man’s immune system and brain development (202). The gut microbiota theory of brain disorders initially arose from the results of a 2002 study that demonstrated resistance to disease and improvement in health and brain function following the feeding of Lactobacillus–fermented fodder to pigs; and subsequently from accumulating evidence linking the gut microbiota to neurobehavioural dysfunction and the development of mood disorders and mental illness (202-203). The leaky gut theory is based on the hypothesis that the human body has two main barriers, the blood brain barrier (BBB) and the gut barrier. The gut barrier regulates the movement of nutrient and signal molecules into the body, and prevents the entry of microorganisms. However, the BBB with the aid of tight junctions regulates the entry and exit of molecules in central nervous system. The maintenance of the integrity of these barriers is critical to health and wellbeing (204-205). The importance of the gut microbiota in the development of the BBB was demonstrated by Braniste et al. (206), when they reported that in germ-free mice an increase in the permeability of the BBB was observed when compared to mice with normal gut flora. This was observed to persist to adulthood and was also associated with a decrease in the expression of tight junction proteins (occludin and claudin-5) (206).

These theories portray the important role the microbiota–gut-brain axis could potentially play in the prevention of and development of drugs for the management of brain disorders; also, accumulating preclinical evidence continues to support these views. A number of methods directed at improving the microbiota such as the transplantation of fecal microbiota, and the use of probiotics or prebiotics have been shown to have significant benefits in health and disease. In children with ASD, seven-to-eight weeks of microbiota transfer therapy altered the gut microbiome and improved both gastrointestinal and behavioral symptoms (207). These improvements were associated with increased overall bacterial diversity and an increase in the abundance of Bifidobacterium, Prevotella, and Desulfovibrio species (207). Also, several systematic reviews have indicated that probiotics can effectively improve mood in humans (208-209), offering promise in the management of depression. In a clinical trial involving patients with mania, 24 weeks of adjunctive probiotics (using Lactobacillus rhamnosus strain GG and Bifidobacterium animalis subsp. lactis strain Bb12) were shown to prevent psychiatric re-hospitalizations after discharge, especially in individuals with elevated levels of baseline systemic inflammation (210).

From the foregoing, while it is becoming increasingly-evident that manipulations of the microbiome can be clinically-beneficial for a limited number of behavioral disorders, for the vast majority, research is yet to substantiate their benefit in humans. Therefore, more effort is needed towards conducting research in this area, especially, considering the fact that orthodox medications have not proven to be fully-satisfactory in the management of a number of such disorders.