The gut microbiome and neuropsychiatric disorders: implications for attention deficit hyperactivity disorder (ADHD)

Abstract

Neuropsychiatric disorders (NPDs) such as depression, anxiety, bipolar disorder, autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD) all relate to behavioural, cognitive and emotional disturbances that are ultimately rooted in disordered brain function. More specifically, these disorders are linked to various neuromodulators (i.e. serotonin and dopamine), as well as dysfunction in both cognitive and socio-affective brain networks. Increasing evidence suggests that the gut environment, and particularly the microbiome, plays a significant role in individual mental health. Although the presence of a gut–brain communication axis has long been established, recent studies argue that the development and regulation of this axis is dictated by the gut microbiome. Many studies involving both animals and humans have connected the gut microbiome with depression, anxiety and ASD. Microbiome-centred treatments for individuals with these same NPDs have yielded promising results. Despite its recent rise and underlying similarities to other NPDs, both biochemically and symptomatically, connections between the gut microbiome and ADHD currently lag behind those for other NPDs. We demonstrate that all evidence points to the importance of, and dire need for, a comprehensive and in-depth analysis of the role of the gut microbiome in ADHD, to deepen our understanding of a condition that affects millions of individuals worldwide.

Introduction

The presence of a bidirectional biochemical signalling channel between the gastrointestinal tract and the central nervous system, now known as the gut–brain axis, was initially -recognized back in the 1980s [1, 2]. More recent studies argue that in particular the gut microbiota plays a major role in communication along this channel (Fig. 1), with some now even referring to the system collectively as the gut–brain–microbiome axis [3–6]. Neuroactive molecules including neurotransmitters, some of which ultimately govern dysfunctional behaviours in individuals with neuropsychiatric disorders (NPDs), can be produced either directly or indirectly (through a downstream chemical reaction involving one or more of its products) by the gut microbiota. In turn, the brain can activate signalling pathways (i.e. anti-inflammatory or fight or flight) that impact on the gut microbial ecosystem, including its population and composition [7]. It therefore becomes reasonable to suspect that the gut microbiota can have a significant role in the diagnosis of NPDs, and also could theoretically be modified to treat NPDs. Much of the literature currently supports this idea, particularly for NPDs such as depression, anxiety and autism spectrum disorder (ASD).

Fig. 1.

The gut–microbiome-brain axis. Image is adapted from http://www.micronutrients.com/. The gut and brain operate as interdependent units, with gut microbiota directly or indirectly producing metabolites (e.g. serotonin) that play important roles in proper neural functionality, and the brain sending signals (e.g. dopamine) to the gut that directly or indirectly affect the gut microbial community. Note that this implies neuromodulators are produced both in the brain and the gut.

Meanwhile, the 21st century has witnessed a rise in another significant NPD, attention deficit hyperactivity disorder (ADHD). The Centres for Disease Control and Prevention now estimates that more than 1 in 10 American children have this disorder, a 43 % increase in the last 15 years [8]. ADHD can also emerge into adulthood, with an estimated one in eight men diagnosed at some point in their lifetimes [9]. Most alarming is that only an estimated one in five people will seek treatment. The current breadth and depth of analyses suggest connections and influence between the gut microbiome and other NPDs, and this naturally invites the question of whether similar relationships could exist for the ADHD. The heritability estimate for ADHD is 74 % [10]. However, the precise genetic aetiology for ADHD has not been delineated. Significant associations of ADHD with genes governing the production of neuromodulators have been observed [11]. However, overwhelmingly ADHD appears polygenic or even omnigenic, with environment playing a major role [12]. In addition, any dependence between ADHD and underlying metabolic pathways argues for the microbiome playing a significant role.

This review starts with a general discourse on the gut microbiome and its overall significance with respect to human health, followed by a discussion of the brain, including germane sections and neuromodulators that contribute to dysfunctional behaviours in NPD patients, including ADHD. Then the review will focus on relevant animal and human studies that have established connections between the gut microbiome and NPDs, particularly depression, anxiety and ASD. Some of these involve microbiome-mediated efforts for treating these NPDs. This will be followed by addressing the symptomatic overlap between the NPDs and ADHD, some of which involves gastrointestinal disorders that further suggest that the gut microbiota may play a key role, thus opening potential new routes for treatment. The review will delve into current challenges in diagnosing NPDs, particularly when distinguishing one from the other because of this symptomatic overlap, and how the gut microbiome could serve as a differentiating factor. The review will end with an examination of the current state of ADHD and gut microbiome research, and conclude with recommendations for future explorations to deepen our knowledge of an NPD that currently affects millions of individuals around the world.

Gut microbiota

Microbiota, a term coined by J. Lederberg, refers to the community of microbes inhabiting a particular niche [13]. Although ‘microbiome’ has become synonymous with ‘microbiota’ its meaning is distinct; a microbiome refers to the combined genetic material of all its member microbes. The microbiome is a highly complex system that is regulated by interactions between its ex vivo and in vivo environments. For the human gut microbiome, the host physiology (including metabolites), age, sex, race and genetics contribute to the ex vivo environment. The in vivo environment is generated by its microbiota – their collective physiology, and metabolites produced by individual microbes that can coordinately influence both the composition and function of the microbiome as a whole.

Microbiome research received an enormous boost with the funding of the Human Microbiome Project by the National Institutes of Health (NIH) [14]. A subsequent press release [15] indicated that there are approximately 100 trillion microbes in the human body, outnumbering our human cells by at least 10 to 1 and accounting for about 2–6 lb of body weight in a 200 lb adult. It should be noted that more recent studies put this ratio as actually closer to one-to-one [16], although this certainly does not eliminate its significance. The NIH study also revealed that microbiomes are highly niche-specific, for example, the microbial profiles of the human gut, urogenital and skin regions are sufficiently distinct [17]. In humans, the most abundant microflora resides in the gut microbiome, which is composed mostly of anaerobes that have evolved to have a commensal relationship with the host. In particular, the colon contains the largest number of microbiota and is also the most diverse [18]. These discoveries make it increasingly apparent that humans are not just an organism but a superorganism [19]. For a review of human microbiome research over the past 10 years, refer to the NIH Human Microbiome Portfolio [20].

Gut microbiome dysbiosis has long been correlated with many disorders, including inflammatory bowel disease [21], irritable bowel syndrome [22, 23], coeliac disease [24, 25], obesity [26], metabolic syndrome [27], immune disorder [28–31], cardiovascular diseases [32] and allergies [33]. In addition, antibiotic-associated diarrhoea due to a proliferation of Clostridium difficile and other pathogenic bacteria in the absence of a healthy microbiome has for many years been a serious problem in hospitals [34, 35]. Gut microbiota have also been linked to human cancer, and patients treated with probiotics before chemotherapy have been shown to have better outcomes [36]. The established impact of the gut microbiome on human health coupled with the presence of the gut–brain–microbiome axis certainly support the theory that the gut microbiome can have an essential role in NPDs as well, motivating many to pursue this area of research.

The brain and NPDs

The brain is the ultimate dictator of individual behaviour [37, 38]. Collectively, behaviour consists of both socio-affective and cognitive elements [39]. Examples of ‘social’ elements include processing social cues, relating well with others and communicating properly. ‘Affective’ elements relate to moods, feelings and attitudes that in turn govern emotion. ‘Cognition’ includes mental processes involving perception, memory and reasoning; and critically for this review, ‘executive function’. The executive function relates to the ability to selectively monitor behaviour, and can also be referred to as ‘cognitive control’ or advanced cognitive processes [40].

The neural substrates of emotion and cognitive control are well characterized. The human brain is composed of a complex system of tracts (Fig. 2), with those most relevant for this discussion being the limbic system (which includes the thalamus, hypothalamus, hippocampus and amygdala) and the prefrontal cortex. The limbic system of the brain is concerned with instinct, mood and basic emotions – namely, the socio-affective element of behaviour [41]. The two organs of the limbic system involved in emotional regulation are the hippocampus, which is necessary for long-term memory and consolidation, and the amygdala, which is critical for emotional processing [42]. The prefrontal cortex of the brain is involved in cognition, especially sensory perception and executive function [41, 43, 44].

Fig. 2.

Brain and element-associated behaviour.

Social and cultural upbringing profoundly influence brain development and behaviour, and neuromodulators play a major role in proper brain functionality throughout a lifetime. Three key neuromodulators are dopamine (DA), serotonin (5-hydroxytryptamine, 5-HT) and noradrenaline. Noradrenaline modulates plasticity, learning and memory [45], while the other two (dopamine and serotonin) influence behaviour more directly (Fig. 3).

Fig. 3.

The role of neuromodulators and the microbiome in neuropsychiatric disorders. Neuromodulators such as dopamine and serotonin, which are affected at least indirectly by the microbiome, impact on the functionality of areas of the brain such as the amygdala, hippocampus and prefrontal cortex. These in turn play roles in emotional regulation and executive function, which are deficient in patients with NPDs. Arrows indicate how well a connection has been established; green arrows indicate convincing connections supported by multiple large-scale studies that link them directly (straight) or indirectly (dotted), whereas the red arrows show less well-established connections.

Serotonin is associated with socio-affective processing [42, 46–48]. It is produced by both the brain and the gut, with an estimated 90 % being synthesized in the gut [49]. Serotonergic system deficiencies have already been implicated multiple times in mood-related disorders. For example, dysregulation of the amygdala, which releases serotonin, has been associated with depression [50]. A lower serotonergic activity is associated with the depressive phase of bipolar disorder [51]. Fear and anxiety have also been linked to changes in the serotonergic systems [52].

Dopamine is synthesized in both the brain and the kidneys, and is associated with processes involving executive function and reward [53]. Developmental disorders such as ASD and ADHD are associated with impaired executive function [53–55]. Dysregulation of the dopaminergic system derails executive function [56], and hypotheses as to the precise neurobiological roles of this system in ASD are beginning to emerge [57].

NPDs are characterized by behavioural disturbances that are rooted in disordered brain function (Table 1) [58]. NPDs afflict both adults and children, currently accounting for the largest percentage of disabilities in developing countries [59]. The NPD burden is particularly high in the United States — where the National Institute of Mental Health estimated in 2017 that 18.9 % of adults and more than 20 % of children suffer from debilitating mental illness [60]. NPDs are categorized as mood-related or developmental. Major depression, anxiety and bipolar disorder are all examples of mood-related afflictions [61], while ASD and ADHD are classified as developmental [55]. Since all of the NPDs are associated with behavioural elements, they are most likely linked somehow to dysfunctionality in the serotonergic and/or dopaminergic systems, which have indeed become the targets of many currently administered medications.

Table 1.

Neuropsychiatric Disorders

Neuropsychiatric disorders	Disorders*	% affected	Characteristics or symptoms†
Mood-related	Major depression	7	Prolonged periods of low mood, lack of energy, irritability, loss of concentration, increased or decreased appetite, feelings of hopelessness and/or thoughts of suicide
	Anxiety	18	Feelings of restlessness or feeling on edge, difficulty concentrating, muscle tension, sleep disturbances and excessive worry
	Bipolar disorder	3	Periods of elevated mood (mania and/or hypomania) and depression
Developmental-related	ADHD	10–15	Inattention, hyperactivity and impulsivity
	Autism (ASD)	1–5	Social problems and repetitive behaviours, typically emerging in the first 2 years of life

*ADHD, attention deficit hyperactivity disorder; ASD, autism spectrum disorder

†Retrieved from the National Institutes of Health (NIH) website

The two systems and NPD categories are not mutually exclusive, either. For example, alterations in the dopaminergic system can affect a mood-related disorder such as depression [62]. Prefrontal depletion of serotonin has been implicated in impaired cognitive functions [46, 47]. This non-mutual exclusivity makes sense, since current studies show the serotonergic and dopaminergic systems to be inter-dependent [63]. This interdependence probably explains the symptomatic overlap between different NPDs [64], which currently makes proper diagnosis challenging [65].

In addition to their essential roles in the brain, the neuromodulators also play a critical function in gut physiology (controlling blood flow, motility, nutrient absorption), and the innate gastrointestinal immune system. Dysregulation of these neuromodulators has already been linked to several digestive disorders [66].

Neuropsychiatric disorders and gut microbiome

Brain development and behaviour are governed to some degree by the gut microbiome [4]. The gut microbiome can directly release neuromodulators, including dopamine and serotonin, impacting on emotional regulation [5, 67]. With the established connections between these neuromodulators and NPDs, the importance of the gut microbiome in NPD research becomes all the more likely. The link is further supported by the observed comorbidity between NPDs and gastrointestinal problems. For example, children with ASD tend to suffer from diarrhoea, constipation and abdominal pain [68, 69]. Recently, gastrointestinal symptoms were found to account for 15–16 % of the variance in the presence of ASD-associated externalizing and internalizing symptoms in a study involving almost 2756 children and adolescents [70]. Increased comorbidity between anxiety disorder and irritable bowel syndrome and inflammatory bowel disease have also been observed [71, 72]. Elevated levels of both anxiety and depression among patients with irritable bowel syndrome were recently demonstrated in a study involving 769 subjects [73].

Several recent studies uncovered parallel improvements in both mental health and the gut microbiome resulting from changes in the diet. Gluten-free and ketogenic diets have both been shown to impact on the gut microbiome [74]. A gluten- and casein-free diet in ASD patients yielded positive results, including a reduction in ASD-associated gastrointestinal symptoms and anti-social behaviours [75]. A small-scale study presented a ketogenic diet as a potential treatment for type II bipolar disorder in humans [76]. A Mediterranean-style diet was shown to improve the gut microbiome and metabolome; specifically yielding increases in fibre-consuming bacteria and short-chain fatty acid concentrations [77], which have been shown to nourish the gut lining, protect against inflammation, help to control appetite and protect against type 2 diabetes [78]. Conversely, diets that are high in sugar have been shown to correlate with lower levels of (ironically) critical gut microbiota for metabolizing carbohydrates, including Eubacterium eligens and Streptococcus thermophilus [79]. In a randomized control trial involving 152 adult individuals, it was shown that a Mediterranean-style diet with fish oil supplements reduced symptoms of depression [80]. In this study, similar improvements were seen with diets involving vegetables, legumes and omega-3/omega-6 fatty acids [80]. Diets that are high in omega-3s have been positively correlated with healthy gut diversity in humans and mice [81]. High omega-3 diets were also shown to decrease anxiety-like behaviour in mice, while diets that were high in sucrose had the opposite effect [82]. Interestingly, negative correlations were observed between ADHD diagnosis and the Mediterranean-style diet, and positive correlations were observed with diets high in sugar [83]. These studies suggest that the diet should be streamlined based on the disorders, and argues for a critical role of the gut microbiota in NPDs.

The earliest studies explicitly connecting the gut microbiome to NPDs were performed on rodents, often by comparing mice raised in isolated germ-free environment to those raised in a typical environment. Two studies found altered anxiety-like and anti-social behaviours in the germ-free mice compared to the normal mice [84, 85]. Subsequently, this behavioural change was associated with altered neurotransmission and gene expression in the hippocampal serotonergic system [5, 84] and altered microRNA expression [86]. The latter had been previously implicated in mouse anxiety [87]. Other studies directly explored metabolites produced by members of the gut microbiome. Rats treated with the metabolite propionic acid (produced by the gut microbe Propionibacterium ) demonstrated restricted response to particular objects, impaired social behaviour and cognitive dysfunction compared with controls [88].

Recent studies have also explored more specific connections between the rodent gut microbiome and the limbic system. There appears to be a dependence between the gut microbiome and inflammation within the ventral hippocampus in rodents [89]. The presence of an inhibitor of the enzyme 5-alpha-reductase led to a long-term alteration in both the gut microbiome and hippocampal neurogenesis in mice [90]. In germ-free mice, there was an altered neuronal transcription within the amygdala with an increase in genes significant for general nervous system development and activity [91]. The gut microbiome influences dopamine levels in the frontal cortex and striatum in rodents [92], two brain regions involved in executive function. It is possible that the gut microbiome is linked to ASD in part through its influence on dopamine levels in prefrontal–striatal regions, affecting executive function. If this effect proves to be true, it would also highlight a possible role of the gut microbiome in other NPDs linked to dopaminergic dysfunction and prefrontal–striatal networks, including the less well-studied ADHD.

A natural question then arises as to whether these animal studies can be mapped to humans. In some cases, it has been possible. For example, a distinct difference in microbiome composition was shown between depressed and non-depressed rodent groups [93] that was later proved to also be true in humans [94]. Liang et al. [95] demonstrated an even stronger correlation between the human and the mouse, with many of the same microbial taxa appearing as highly and lowly abundant in the gut microbiomes of mice and humans with depression. A faecal microbiota transplant from depressed humans to mice was able to transmit depression symptoms [96], further demonstrating some overlap between human and animal studies. At other times there is no overlap [97], illustrating the importance of running both types of experiments.

More than a decade ago, experiments involving human subjects demonstrated the presence of compositional differences in the gut microbiome for mood-related NPDs – depression [67], anxiety [98] and bipolar disorder [99]. A comparative analysis of more than 1000 gut microbiomes led to the identification of a Bacteroides enterotype as overrepresented in patients with depression [100–102]. A set of gut microbiome-derived compounds have been implicated in stress response and inflammation, ultimately playing a role in anxiety and depression [103]. In individuals with bipolar disorder, compositional differences in the microbiome were observed to alter levels of tryptophan, a precursor to serotonin [104].

The gut microbiome of children with ASD has been shown to be less diverse [105] and tends to harbour fewer commensal bacteria [106] compared to healthy samples. A proliferation of unwanted organisms known to produce potent neurotoxins (e.g. Clostridium ) is observed in ASD patients, although this could be a by-product of the reduced diversity [105, 107, 108]. More significant compositional differences were discovered among ASD patients, particularly within the most abundant gut phyla [109]. Li et al. [110] even propose links between members of the gut microbiome and environmental risk factors associated with human ASD, including maternal stress, infection and exposure to pesticides.

Implications

Microbial compositional differences illustrate the potential for the gut microbiome to serve as a differentiating factor between NPDs, given the current challenges in proper diagnosis due to symptomatic overlap. Noting an abundance in a particular taxon (e.g. Bacteroides in depression; Clostridium in ASD) could offer additional feedback when distinguishing one NPD from another. Additionally, connections between NPDs and the gut microbiome offer significant implications regarding NPD treatment. Most obviously, there is potential for developing vaccines to stimulate the host’s immune system to prevent infection by these harmful microbes. One example is already being evaluated for battling the ASD-associated Clostridium bolteae that is partially responsible for the eosinophil-associated gastrointestinal disorders [111]. In addition, there is the potential for more ‘natural’ NPD treatments.

Since the gut microbiome can be manipulated through diet, exercise and other means, identifying specific bacterial biomarkers could be therapeutically significant [112, 113]. Such therapies have already been helpful in animal experiments, where probiotic modulation of behaviour has been observed in both zebrafish [114] and rodents [115–117]. Both meta- [118] and clinical analyses [119] demonstrated positive results for probiotics in battling depression and anxiety in humans. Similar treatments, including microbiome-mediated therapies [120] and gluten- and casein-free diets [75], have been prescribed to individuals with ASD. Grimaldi et al. [75] actually demonstrated that diet coupled with a prebiotic for children with ASD resulted in social and behavioural improvements along with compositional changes in the gut microbes and metabolites. A few studies have explored the effects of probiotics [121] and faecal microbial transplant (FMT) [122]) on ASD symptoms, with encouraging results. For example, Lactobacillus reuteri was recently discovered as a treatment that could reverse social deficits in rodents, regardless of whether they were germ-free or not [123]. Microbiota transfer therapy (MTT), a technique involving FMT preceded by cleansing mechanisms such as antibiotics and stomach acid suppressants, was recently patented as a safe and effective method for addressing ASD symptoms (including gastrointestinal) in children [124, 125].

ADHD, current gaps and future directions

A particular area that still demands attention is ADHD, as illustrated in Fig. 3. A great deal of overlap is observed symptomatically between ADHD and other NPDs, especially ASD (including impulsivity, social awkwardness and limited areas of focus [126]). Also similar to the other NPDs, multiple studies have demonstrated a generally higher incidence of gastrointestinal symptoms in ADHD children compared to controls [127], particularly constipation [128, 129]. A significant association between ADHD and body-mass index has been indicated, suggesting a shared genetic risk [130–132]. Although the underlying mechanisms by which ADHD is linked with obesity is not clear, ADHD may increase the risk of future development of obesity [133]. An unequivocal link has been established between obesity and the gut microbiome [134, 135], and there is a growing call to use gut microbiota for treating obesity [136]. All of these results indicate the potential for a similarly significant role of the gut microbiome in ADHD, with similar implications for treatments. Given its high prevalence among all NPDs (Table 1), such developments in ADHD research would be highly beneficial.

There is very limited research explicitly connecting the gut microbiome to ADHD. A compositional difference in the early gut microbiome in children who were later diagnosed with ADHD was observed [137]. Another study found a lower level of alpha-diversity in the gut microbiome of young ADHD patients [138]. Further, differential analyses through metagenomic profiling of patients show specific ADHD-associated taxa compared to controls [139]. However, a similar analysis did not see any significant change in alpha-diversity and obtained a different set of abundant taxa for the two groups [140]. Explicit reasons for the differences are not immediately clear; both studies were performed on children, but the first used the same size for the ADHD and control groups (both n=30), while the second was slightly unbalanced (51 vs 32). There were also other inevitable differences between the two studies that can impact on the microbiome, such as age (6–16 years in the first vs 6–10 in the second) and location (Taiwan vs PR China). These additional dependences create challenges, and it therefore becomes evident that understanding the impact of the gut microbiome in ADHD patients will demand more in-depth analysis of the underlying web of interactions along the gut–microbiota-brain axis and the direct and indirect involvement of molecules produced by specific taxa.

Several proposed models integrate signalling and chemical reaction pathways along the gut–microbiome-brain axis and argue for the participation of neuromodulators [141], which now need to be experimentally verified. Interestingly, a noticeable increase in the abundance of the genus Bifidobacterium in ADHD patients’ gut microbiomes, with a corresponding increase in the concentration of a metabolite cyclohexadienyl dehydratase, was discovered [142]. Pathway analysis revealed that cyclohexadienyl dehydratase is involved in the production of phenylalanine, which is a precursor to dopamine [143]. The same researchers also discovered a lower level of response to reward anticipation in these subjects (a process known to be partially governed by dopamine). More work needs to be done to untangle the web of interactions between phenylalanine and the reward anticipation responses, which could involve both other metabolites and microbiota. There is an urgent need for a comprehensive and in-depth analysis involving the gut microbiota and ADHD, at a level that exceeds compositional analysis and ventures into inter-taxa relationships, or even multi-omics [144].

Conclusion

There is exciting future potential for research that connects the gut microbiota/microbiome to neurocognitive elements and NPDs. Understanding the impacts of the gut microbiome on relevant neuromodulators, and compositional differences in the microbiome, has led to positive outcomes with microbiota-based therapies for several NPDs – particularly depression, anxiety and ASD. ADHD is currently on the rise, and we are thankfully beginning to explore the impact of the gut microbiome in these patients. Since there is clear biochemical and symptomatic overlap between ADHD and other NPDs, the role of the microbiome merits the pursuit of at least equally in-depth analyses. It will be particularly important to explore the gut microbiome below the compositional level (as has been done more significantly with other NPDs). This will help to understand the underlying reaction dynamics within the microbiome and pathways that affect communication along the gut–microbiome-brain axis, particularly those involving neurotransmitters. Multi-omics analyses have the potential for enormous impact in this area. A relational-level analysis [145] to thoroughly understand the ecological significance of taxa may lead to the identification of potential biomarkers. Finally, future approaches should integrate host and gut microbiome analyses. These could, for example, illuminate connections between the microbiome and host gene expression in the prefrontal cortex, which has been much less significantly studied (compared to, for example, the hippocampus) but plays a more significant role in the cognitive processes for which individuals with ADHD tend to be deficient. These same studies could also determine if there are specific genetic components that coincide with a specific gut microbiotaenterotype, further helping to complete the big picture.