Gut microbiota and autism spectrum disorders

Overview

Autism spectrum disorders (ASD) are neurodevelopmental and neuropsychiatric disorders with clinical manifestation in childhood, characterized by difficulties in social interaction and communication, limited interests, and repetitive behavior. In recent years, there has been a significant increase in the prevalence of this disorder, currently affecting 1–2% of children. In pathophysiology, genetic factors are mainly used, which are also modified in their phenotypic expression by environmental factors. A large percentage of patients with ASD also have a variety of gastrointestinal problems; children experience constipation or diarrhea, flatulence, or nonspecific abdominal pain. In last years, the connection of intestinal symptomatology with the manifestations of ASD has come to the forefront of scientific interest with the development of new molecular-bio­logical methods of studying the community of intestinal microorganisms, the gut microbio­ta. Communication between the gut and the CNS (the gut-brain axis) is made possible by a very complex two-way neurohumoral communication system. This system ensures the effect of the intestinal microbio­ta on brain functions and allows the brain signals to affect the intestinal microbio­ta activity and the gastrointestinal tract functions. This article focuses on various pathophysiological mechanisms involved in these processes and discusses in detail the possible therapeutic modalities that affect the gut microbio­ta composition and function, such as the use of vancomycin, oxytocin, prebio­tics and probio­tics, as well as fecal transplantation, which has very promising results in relation to both gastrointestinal and ASD symptoms.

Keywords:

Autism – autism spectrum disorders – gut microbiota – probiotics – fecal transplantation

Introduction

Autism spectrum disorders (ASD) are neurodevelopmental and neuropsychiatric disorders with clinical manifestation in childhood, characterized by difficulties in social interaction and communication, limited interests, and repetitive behavior. Symptoms persist throughout the life; children with ASD switch to adult care after the age of 18. ASD are more common in men at a 4 : 1 ratio. The estimated prevalence of ASD in the population increased from 1 out of 476 to 1 out of 323 in the 1990s [1,2] to 1 or 2 out of 100 in 2014 [3,4]. In about half of the patients, the higher incidence can be explained by bet­ter dia­gnosis and awareness of ASD, older age of parents, and geographical groupings of families; in the remaining patients, the cause of the increased prevalence of ASD is unclear [5].

The pathophysiological mechanisms involved in the development of ASD remain a subject of intensive research. Genetics play an important role; a number of candidate genes involved in the pathophysiology of ASD have been identified [6]. At the current level of knowledge, the most widespread theory is the genetic predisposition of an individual, where various environmental factors can influence or trigger psychiatric symptoms. Such a “non-genetic but hereditary” factor may be the composition and diversity of the gut microbio­ta, a highly structured community of all the microorganisms that inhabit the intestine of each individual.

The link between the composition of the gut microbio­ta and the development of ASD symptoms seems very unlikely at first glance. However, it is clear from clinical practice that children with ASD have very often digestive problems and various gastrointestinal tract symptoms, such as abdominal pain, persistent constipation or, conversely, diarrhea, flatulence, etc. [7].

Recent studies show that the gut microbio­ta is involved, through a number of mechanisms, in a very important communication channel, the gut-brain axis, which will be described below. Normal colonization patterns of the gut microbio­ta are very important for the proper development and function of the CNS; they provide key homeostatic mechanisms in the development of inflammatory processes. Intestinal microbial dysbio­sis may be associated with the development or worsening of ASD symptoms.

Normal gut microbio­ta

The human gut contains up to 80–100×10¹² microorganisms [8], which in total have about 100times more genes than the human genome [9]. The gut microbio­ta is not only made up of bacteria (bacterioma), but also of a whole range of viruses, fungi, and other microorganisms. The gut is thought to colonize 40,000 different species of bacteria, the analysis of which is currently based on the detection of their specific 16S (18S) ribosomal ribonucleic acid sequences [10]. It is possible to cultivate less than 3,000 bacterial species from the human body by culture [11]. It is therefore indisputable that molecular-bio­logical methods leading to the discovery of a considerable diversity of gut microbio­ta have provided new knowledge and opened a number of questions about the composition, function, and significance of this community.

The microbial population pattern is unique for each individual and is based on genetic predisposition [10] and influenced by a number of environmental factors. The management of childbirth can have a significant effect on the composition of the microbio­ta. Cesarean section delivery does not allow the baby to come into contact with the mother’s vaginal microbio­ta and to acquire non-physiological microbial stimuli, especially from the skin. Nutrition and a type of diet also have an undeniable influence on microbio­ta composition [10]. Breastfeed­ing significantly positively affects its composition, as breast milk is a source of probio­tic oligosaccharides; plants are a source of soluble prebio­tic fiber; and important microorga­nisms such as “probio­tics” are found in fermented dairy products [10]. Animal products do not have such benefits for the microbio­ta. Industrially processed foods treated by methods leading to the removal of pathogenic microorganisms and using a wide range of chemicals have an undeniably negative impact.

Human physiological microbio­ta undergoes relatively dynamic advancement during ontogenetic development. The dominant gut microbial phyla are Firmicutes, Bacteroidetes, Actinobacteria, Proteobacteria, Fusobacteria, and Verrucomicrobia, with the two phyla Firmicutes and Bacteroidetes representing 90% of the gut microbio­ta. The Firmicutes phylum is composed of more than 200 different genera, such as Lactobacillus, Bacillus, Clostridium, Enterococcus, and Ruminicoccus. Bacteroidetes consists of predominant genera, such as Bacteroides and Prevotella. The Actinobacteria phylum is proportionally less abundant and mainly represented by the Bifidobacterium genus [12]. Significant differences in the composition of the gut microbio­ta can be observed between premature and full-term neonates. In premature infants, two essential bacterial genera are missing: Bifidobacterium and Lactobacillus with a compensatory predominance of Proteobacteria phylum [13,14]. The infant digestive tract is populated by microorganisms from the Actinobacteria and Proteobacteria phyla [10]. In adults, the highest proportion of bacteria in the gut microbio­ta is formed by the Bacteroidetes, which includes the genera Bacteroides and Prevotella. Adult intestinal bacteria include the phylum Firmicutes, which includes the genera Clostridium and Lactobacillus. Bacteria of the Proteobacteria and Actinobacteria phyla are also found in adult intestinal bacteria [10,15].

The microbial population pattern is partly genetic but is influenced by a number of external factors and is thus highly individual. For example, geographical differences in microbio­tic composition are so significant that small samples of subpopulations from a particular geographical area are highly heterogeneous and therefore healthy controls cannot be easily defined, even in a relatively homogeneous population in terms of genetics, nutritional habits, and ethnicity. In their research, He et al studied gut microbio­ta in China’s Guangdong Province with a population of 108 million. They randomly selected 14 districts in the province and analyzed samples from 7,009 individuals over the age of 18. They found that the geographical differences in the composition of the intestinal microbio­ta exceeded those differences that were evident when comparing samples from “healthy” individuals with individuals experiencing various metabolic diseases (type 2 diabetes mellitus, metabolic syndrome, and obesity). The authors believe that this explains the origin of inconsistent dysbio­tic patterns observed in studies on small population samples [16].

In terms of “healthy” microbio­ta, not only its specificity but also the diversity and number of individual bacterial species are certainly important. Studies comparing the microbio­ta populations of Western civilizations with those of indigenous tribes show a significant reduction in bacterial diversity in modern conditions, and this trend is sup­ported by globalization [17–19].

Mechanisms of communication on the gut-brain axis

Communication between the gut and the CNS (the so-called gut-brain axis) is made possible by a very complex two-way neurohumoral communication system (Fig. 1). This system ensures the action of the gut microbio­ta on brain functions and allows the brain signals to affect the microbio­ta activity and gastrointestinal tract functions. The autonomic nervous system plays an important role here. During stress reactions, the sympathetic system activity is increased and intestinal functions, specifically intestinal motility and secretion, are inhibited. With long-term stress, sympathetic hyperactivity occurs to such an extent that intestinal integrity is damaged, intestinal epithelium is destroyed, and intestinal motility and secretion are impaired. These changes lead to a change in the gut microbio­ta composition and to the development of pathological inflammatory processes [20,21]. Stressed mice have been shown to have a hyperactive hypothalamic-pituitary-adrenal axis, leading to increased serum cortisol and proinflammatory cytokines and to the development of intestinal inflammation [22]. The effects of stress leading to dysbio­sis of the gut microbio­ta are well documented, for example, in irritable bowel syndrome [23].

The results of recent research show that the effect of gut microbio­ta far exceeds the intestinal boundary. The bacterial gut microbio­ta is able to produce various metabolites that are applied at several levels. They are able to maintain the integrity of intestinal epithelia, affect nutrient resorption and fat metabolism, and affect intestinal motility. For example, short-chain fatty acids (SCFAs) play a crucial role. Bacteria obtain energy and produce metabolites, particularly from SCFAs (especially acetic, propionic, and butyric acid), by fermenting indigestible carbohydrates (e. g., cellulose, xylan, inulin) [24]. These metabolites are able to reduce in vitro production of proinflammatory cytokines interleukins (IL-1â, IL-6), and tumor necrosis factor-alpha (TNF-á). In contrast, they increase the production of anti-inflammatory cytokines, such as IL-10 [25]. SCFAs are also able to penetrate the blood-brain barrier (BBB) and exert their functions in the brain, either directly or indirectly through the G-protein coupled receptors. Butyric acid, for example, exerts its function in the brain as an epigenetic modulator through histone deacetylases [25–27].

The gut microbio­ta is able to regulate key central neurotransmitters at the level of their precursors. SCFAs are able to increase the production of enteral serotonin and ensure its homeostasis. Monoamines such as norepinephrine, dopamine, and serotonin can also be produced by the gut microbio­ta. Certain types of bacteria are able to modulate serotonin secretion from colonic enterochromaffin cells and modulate its concentration in both the intestine and the blood [25]. Bifidobacterium infantis increases plasma tryptophan levels and thus affects central serotonergic transmission [28]. Bacteria can produce neurotransmitters themselves, as has been shown in Lactobacillus and Bifidobacterium genera, which produce gamma aminobutyric acid (GABA). Animal studies evidenced that Lactobacillus can modulate the levels of serotonin and dopamine in the brain, influencing early brain development [29]. Escherichia, Bacillus, and Saccharomyces in turn produce norepinephrine. Escherichia can also produce serotonin and Bacillus dopamine [28]. These bacterially synthesized neurotransmitters are able to pass through the intestinal mucosa and exert their physiological functions in the 
brain.

Pathogenic gut microbio­ta leads to the stimulation of the secretion of proinflammatory cytokines (IL-1â, IL-6, IL-18) through intestinal epithelial and dendritic cells. These lead to the development of gut inflammation and increased intestinal permeability and are associated with various neuropsychiatric diseases, such as depression, anxiety, and schizophrenia [25].

Gut microbio­ta and gut-brain communication in ASD

There are certainly several reasons that the connection between the intestine and the symptoms of autism has been studied. The symptoms of gastrointestinal disease have long been observed in children with ASD, as mentioned in the introduction. Gastrointestinal problems are observed 6–8× more often in children with autism than in the general population [7]; interestingly, the severity of symptoms correlates with the severity of autism [7]. In rare cases, this connection can also be explained genetically. For example, chromodomain helicase DNA-binding protein 8 (CHD8) is identified as a candidate gene in the pathogenesis of ASD. Children with this mutation had a much higher rate of constipation than other patients with ASD (up to 60% of cases). Studies have shown that mutations in CHD8 gene damage the intestinal motility by interfering with neuronal colonization of the gastrointestinal tract [30].

The presence of gastrointestinal tract symptoms in children with ASD in itself encourages the study of pathological processes in the gut in these patients in relation to the severity of the clinical expression of the disorder, also because it opens various therapeutic options; due to the absence of any causal treatment, such options are very tempting in children with ASD. However, a deeper study of this methodology presents numerous obstacles. One very fundamental problem is the definition of “healthy control” in relation to the composition of the intestinal microbio­ta. The gut microbio­ta has an individual genetic imprint for each individual and is influenced by a number of factors, from the type of birth control, through breastfeeding to pharmacological influences, nutritional factors, stress, and other environmental factors, including geographical variability. Creating a “normal gut microbio­ta map” is a difficult task. Although it is possible to create a relatively sophisticated map of its composition with the help of molecular genetic methods, its interpretation is not easy [16].

Children with ASD have a lower level of Akkermansia, Bacteroides, Bifidobacterium, Escherichia, and Enterococcus, a higher level of Faecalibacterium, and slightly increased level of Ruminococcus and Clostridium. Reduced levels of beneficial bacteria combined with the increased levels of harmful bacteria contribute together to ASD symptoms [31]. Several studies showed that there was a higher level of Clostridium in the individuals with ASD compared to controls and hypothesized that Clostridium can produce neurotoxins and contributed to ASD [31]. A higher level of Bifidobacterium negatively correlates with ASD symptoms. Bifidobacterium is among first colonizers of the human gut and one of the dominant groups in the gut microbio­ta of breast-fed infants. It can ferment complex polysaccharides to regulate host functions and promote health, encouraging the interest in its use as a probio­tic [32].

Strati et al observed a significantly higher Firmicutes/ Bacteroidetes ratio in children with ASD, due to a significant reduction in Bacteroidetes phylum in these children [33]. Bacteriodetes and its genus Bacteroides are the main producers of propionate in the gut, and the abundance of propionate in feces correlates strongly with the abundance of Bacteroides. Propionate produced by the microbio­ta is used for gluconeogenesis in the liver and represents a source of glucose in the host. The study indicated that neurodevelopmental abnormality in ASD patients is accompanied with impaired propionic acid metabolism, which may relate to the change of propionate-producing bacteria [34]. The results in mouse models show that the enhancement of Bacteroides fragilis sp. can lead to altered gut microbio­ta and blood metabolite profile, modify intestinal permeability, and improve ASD behavior [35]. The elevated Proteobacteria phylum was found in ASD patients. It is noted that Proteobacteria is a major gram-negative bacteria and includes a variety of opportunistic pathogens. Meanwhile, as microbial signature of dysbio­sis in the gut microbio­ta, Proteobacteria is associated with host inflammation. These gram-negative bacteria produce a potent toxic factor lipopolysaccharide (LPS). An animal study indicated that prenatal LPS exposure reduced the level of glutathione in the brain. Glutathione is a significant antioxidant, closely related to detoxification in the brain [32]. It was found that in patients with ASD, Candida yeast is twice as present in the gut microbio­ta as in the control group. The interpretation of these results is difficult, and the authors hypothesize that alterations in the intestinal fungal population due to Candida overgrowth in the gut microbio­ta in children with ASD may adversely affect the gastrointestinal function due to cytokine network dysregulation [33].

Thus, the representation of individual bacterial genera and their taxonomic description will play an important role, but so will the mechanisms by which individual bacteria are involved in various neurohumoral, neurometabolic, and neuroimmunological processes. Immune dysregulations in patients with ASD were observed, including alterations in circulating and brain cytokines, chemokines, and other inflammatory factors, as well as abnormal distributions and/ or reactivity of various leukocyte subtypes. Certain inflammatory cytokines (i.e., interferon- ã, IL-1ß, IL-6, IL-12, TNF-á) may be associated with ASD symptoms and their severity. Increased levels of some chemokines (i.e., CCL-2, CCL-5) are associated with higher abberant behavioral scores and more impairments in the development [36]. 

A number of studies have dealt with the gut microbio­ta in relation to many major diseases in any organ system; the list is beyond the scope of this text, which is focused only on patients with ASD. The crucial question is whether the differences in the gut microbio­ta are causal in relation to ASD, or whether they have a mere association between the two diseases [10]. It has been suggested that the changes in the gut microbio­ta composition may be caused by limited nutritional patterns in these patients, leading to the reduction in bacterial diversity and overgrowth of other strains. The gut microbio­ta of patients with ASD has an as yet unspecified but characteristic pattern, as shown by the results of a recently published study [37] in which the authors transplanted the gut microbio­ta from human donors with ASD to germ-free mice and found that these mice expressed autistic behavior traits. The authors performed a metabolomic analysis of the intestinal contents of mice with ASD and a control group. They found that the dif­ferences in GABA and glycine receptor inhibitory agonists and antagonists could be detected in autistic mice. In particular, they found 5-aminovaleric acid (5-AV), a fermentation product of proline, which is considered a weak agonist of GABAA receptors with anticonvulsant effects in mice and is significantly lower in patients with ASD [37]. They also showed lower levels of taurine, a metabolic product of taurocholic acid that is essential for brain development [38], is produced by neurons and astrocytes, and is transported across the BBB by the taurine transporter TauT. It is a weak GABAA agonist and also a possible glycine receptor agonist that is reduced in patients with ASD [7].

The results support the theory that gut bacteria may affect inhibitory GABA signaling. Some forms of ASD are thought to be caused by excitatory-inhibitory imbalances in synaptic transmission and involvement in corticosubcortical circuits [6]. The possibilities of therapeutic intervention in this imbalance at the level of influencing synaptic transmission are the subject of studies and have been shown to be effective in mouse models [39–41]. The authors [37] tested whether 5-AV and taurine affect the cortical circuits and interfere with the excitatory-inhibitory balance by registering the activity from fifth-layer pyramidal neurons in the medial prefrontal cortex. Specifically, they examined whether the treatment with 5-AV and taurine from conception to adulthood affected the membrane and synaptic properties of pyramidal neurons (miniature inhibitory or excitatory postsynaptic potentials). Pyramidal neurons from the fifth cortical layer were shown to have significantly lower excitability in 5-AV-treated mice. GABA excitation is thought to shift physiologically to GABA inhibition during the development, and this signaling change does not occur in children with ASD, which is thought to be one of the possible mechanisms in the pathophysiology of autism [6]. Taurine is an inhibitor of the potassium chloride cotransporter KCC2 [42] and can thus interfere with the excitatory-inhibitory switching in GABAergic transmission, thereby modulating neuronal development and the response of neurons to inhibitory neurotransmitters. In summary, the gut microbio­ta can influence complex behavior in mice through the formation of neuroactive microbial metabolites using the mechanisms that interfere with the regulation of excitatory-inhibitory balance in the brain [37].

Therapeutic options

Influencing the symptoms of autism by interfering with the gut microbio­ta of patients has been a subject of research for 20 years (Tab. 1). The first studies were performed using the non-absorbable antibio­tic vancomycin, which may have a short-term effect on the symptoms of ASD. This has been demonstrated in small groups of autism patients treated with vancomycin for 8 weeks [43,44]. Increased attention is currently being paid to oxytocin, a hypothalamic peptide that increases sociability. Mice with blocked oxytocin receptors show a significant deficit in social behavior [45]. Interestingly, the probio­tic bacterium Lactobacillus reutri is able to influence the activity of the posterior hypothalamus and increase the level of oxytocin, which makes it possible to intervene in social behavior through the influence of gut microbio­ta [46]. Preclinical studies show that intranasally administered oxytocin can positively influence social behavior patterns [46].

A number of studies have addressed the use of prebio­tics and probio­tics in the treatment of ASD symptoms. These are mostly prospective studies, open-label or double-blind, with 8–30 participants in the ASD and control groups. Prebio­tics (colostrum, carrot powder, guar gum, vitamin A) combined with probio­tics (various genera of Lactobacillus and Bifidobacterium) were given for 2–12 months. The results show that there may be a reduction in negative gastrointestinal symptoms 
[47–49], and in some there may be a statistically significant effect on behavioral disorders within ASD [47,48,50]. However, prebio­tics and probio­tics often have no effect on gastrointestinal tract symptoms or ASD symptoms [51,52]. In summary, the results of the studies are generally not very encouraging and the effects of prebio­tics and probio­tics on the reduction of ASD symptoms are small. Inoue et al demonstrated a statistically significant reduction in microbial diversity and levels of some cytokines (IL-1â, IL-6 and TNF-á) after prebio­tic administration [50]. Another study showed a statistically significant increase in the Bacteroidetes/ Firmicutes ratio after vitamin A supplementation, but without the evidence of the improvement in ASD symptoms [51].

Compared to probio­tics, which contain only a few bacterial strains from dairy cultures, the method of fecal transplantation (microbio­ta transfer therapy; MTT) is very interesting from a therapeutic point of view, as thousands of natural gut bacterial species can be administered at once. This therapy is effective in the treatment of recurrent Clostridium difficile infections and also yields promising results in the treatment of inflammatory bowel disease [52–54]. Kang et al performed MTT in 18 children with ASD aged 7–16 years in a prospective open-label study [55]. Children received oral vancomycin for 14 days before transplantation, followed by 12–24 h of fasting to clean the bowels. Thereafter, children received a high-dose initial dose of standardized human gut microbio­ta (SHGM) either rectally or oral­­ly followed by daily oral maintenance low-dose SHGM therapy in combination with proton pump inhibitors for 7–8 weeks. There was an improvement in the gastrointestinal symptoms scale of approximately 80% (P < 0.001) with a significant improvement in symptoms such as constipation, diarrhea, and abdominal pain. This improvement persisted for 8 weeks after stopping the treatment. Likewise, the manifestations of ASD improved in several psychological tests. Significant changes were observed in the results of the Childhood Autism Rating Scale (CARS), a dia­gnostic tool used in children with ASD. The CARS score decreased by 22% at the end of the treatment and by a further two percent in 8 weeks after the end of the treatment. There was an improvement in the Social Responsiveness Scale (SRS) score, which assesses deficits in the social sphere. The Vineland Adaptive Behavior Scale (VABS) score, which is used to assess adaptive behaviors such as communication, normal daily skills, and socialization, showed an increase of 1.4 years (P < 0.001) in all spheres. The results in these tests tended to further improve when evaluated in 8 weeks after the end of the treatment. The study of gut microbio­ta showed an overall improvement in microbial diversity with an increase in Bifidobacterium, Prevotella, and Desulfovibrium, again with an effect lasting even 8 weeks after the end of the treatment. Importantly, the authors did not show a significant difference in results when comparing the group where SHGM was administered rectally with the group with oral initial high-dose administration. A follow-up study by the same scientific team, which indicated the effect of MTT two years after its implementation, also showed very positive results [56]. Further favorable results were obtained in the study of 24 patients with ASD, where fecal transplantation led to an increase in the proportion of Bacteroides fragilis and Bacteroides vulgatus in feces and improved the composition of gut microbio­ta and ASD symptoms in relation to Coprococcus comes [57]. No adverse reactions or complications of fecal transplantation were reported in any of these studies and the transplantations were generally well tolerated. Although the results of the studies with MTT appear promising, the effect is only partial and proves that the pathophysiology of ASD is complex and further research concerning the involvement of the gut microbio­ta in the development of ASD is needed.

Conclusion

Given the growing prevalence of autism and the severity of this disorder, the question of therapeutic options is increasingly relevant. It is clear that the pathophysiology of ASD is complex, with both genetic and environmental influences. If the symptoms can be influenced by interfering with the composition of the gut microbio­ta, then this possibility is certainly a great challenge.

Acknowledgments

The publication was supported by funding through the Faculty of Medicine of Masaryk University for junior researcher Pavlína Danhofer, MD, PhD (ROZV /  28 /  LF /  2020).

Conflict of interest

The authors declare they have no potential conflicts of interest concerning drugs, products, or services used in the study.

The Editorial Board declares that the manu­script met the ICMJE “uniform requirements” for biomedical papers.

Redakční rada potvrzuje, že rukopis práce splnil ICMJE kritéria pro publikace zasílané do biomedicínských časopisů.

Pavlína Danhofer, MD, PhD

Department of Pediatric Neurology

Masaryk University

University Hospital

Černopolní 9

613 00 Brno

Czech Republic

e-mail: pavlina.cahova@seznam.cz

Accepted for review: 4. 1. 2021

Accepted for print: 15. 4. 2021