The gut microbiota is a complex ecosystem consisting of bacteria, viruses, fungi and archaea, though the bacterial kingdom is the most studied of these populations (Thursby and Juge, 2017).

Several studies have linked the role of the gut microbiota to overall health status (Cresci and Bawden, 2015). In this way, some bacterial taxa have emerged as important treatments of dysbiosis/unbalanced microbiota (Cristofori et al., 2021). In addition, the inclusion of probiotics and prebiotics in the diet and fecal microbiota transplantation are validated strategies in the treatment of specific infections, especially those caused by resistant strains of Clostridioides difficile (Mills et al., 2018; Sandhu and Chopra, 2021). Moreover, strategies based on modulation of the gut microbiota appear as promising options to treat a wide variety of other pathologies, such as intestinal bowel disease, inflammatory diseases, metabolic diseases, cancer, and other pathologies (Marchesi et al., 2016; Yao et al., 2021; Fernández et al., 2022). However, further research is needed to verify the suitability of their clinical use in these diseases.

Emotional states are processes that regulate brain and body, and represent a reciprocal brain–body dialogue (Colombetti and Zavala, 2019). In recent years, the interaction between the gut microbiota and the central nervous system has acquired a special significance referred to as the ‘gut-brain axis’ (Du et al., 2020), which is a two-way communication between gut bacteria and the brain that takes place via the nervous, endocrine and immune systems. This bidirectional communication involves neuronal modulation, immune response and hormone release (Neufeld et al., 2011; Selkrig et al., 2014). As it is a bidirectional response, it involves regulation of the permeability of the intestinal epithelium and the blood–brain barrier. The response of intestinal microbiota is mediated by various metabolism products, including short-chain fatty acids (SCFAs); bacterial neurotransmitters such as gamma-aminobutyric acid (GABA) or serotonin; modulators such as quinolinic acid, which modifies the immune system; and hormones such as cortisol (Fernández et al., 2016; Puricelli et al., 2021; Chen et al., 2022). The microbiome is also thought to influence brain function, behavior (Vuong and Hsiao, 2017) and mental health (Valles-Colomer et al., 2019).

Alterations in the human gut microbiota composition have been linked with mood disorders and neuropsychiatric disorders (Huang et al., 2019) and, in turn, with neurotransmitters imbalances (Frankiensztajn et al., 2020). Increasing evidence has linked the gut microbiota with symptoms of autism spectrum disorder (ASD) that are regularly affected by gastrointestinal problems and the gut microbiota dysbiosis (Yang et al., 2018). Furthermore, many patients who experience gastrointestinal discomfort are more likely to present mental disorder comorbidities (Huang et al., 2019). In this sense, an alteration of the gut microbiota brain axis has been associated with autistic behaviors (Li and Zhou, 2016). In terms of emotional states, this axis is also involved in the regulation of stress-related responses (Foster et al., 2017), specifically the hypothalamus-pituitary-adrenocortical (HPA) axis. The gut microbiota is closely connected to the development and function of the HPA axis (Frankiensztajn et al., 2020). In this regard, alterations in the metabolic, immune and endocrine systems have been found in patients with depression, pointing to an association between the pathophysiology of depression and the gut microbiota (Caspani et al., 2019). In addition, the gut microbiota of depressed patients are significantly different from those of healthy controls (Liang et al., 2018). Thus, modulation of the gut microbiota and neuroinflammation could alter brain function and have an influence on depressive and anxiety-like behaviors (Koopman and El Aidy, 2017). With respect to severe mental illness, dysbiosis has also been identified as a comorbidity of schizophrenia (Castro-Nallar et al., 2015) and has been associated with the severity of psychotic symptoms and global functioning in patients during their first episode of psychosis (Nguyen et al., 2018). The interaction with the gut-brain axis has also been studied in the context of eating disorders (Seitz et al., 2020). In this sense, some research suggests that the gut microbiota is altered in patients with anorexia nervosa (Hanachi et al., 2019), since it seems to play a role in different metabolic functions (regulation of weight gain, insulin secretion and energy production from food) (Tremaroli and Bäckhed, 2012). The gut microbiota has also been linked to neurodegenerative diseases and may play a key role in the aging process, the loss of quality and strength of muscle, the loss of skeletal mass and in the decline in cognitive function (Ni et al., 2019). The aim of the present review is to provide an overview of the results of human clinical studies published over the last 11 years that highlight the relationship between gut microbiota and psychiatric disorders.

Based on the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guide (Hutton et al., 2016), we employed an evidence-based model to frame a PICO question model (PICO: Participants, Intervention, Control, and Outcomes).

The following question was posed: “Is there an association between psychiatric disorders and gut microbiota?” Participants: individuals diagnosed with psychiatric disorders. Interventions: untreated or treated (dietary intervention, psychological programs, antipsychotic drugs, probiotics or microbiota transfer therapy…) individuals. Controls: healthy individuals, of similar age and weight, with no psychiatric disorders. Outcome measures: microbiota content and diversity in individuals with psychiatric disorders.

Three thousand five hundred twenty-four articles were identified with the keywords (microbiota, psychological disorders and human) and the following filters were applied to both data bases (yielding 92 articles in PubMed and 167 articles in Scopus): last 11 years (2012–2022); full text; clinical trial; randomized controlled trial. We removed 20 articles about animal studies, 13 systematic reviews, 2 cohort study articles, 61 no case–control articles, 117 articles dealing with a topic other than ours, 5 articles without full text, 2 duplicated articles and 2 articles not written in English, thus rendering a new total of 18 articles. Of these, 15 papers were case–control studies and 3 were randomized clinical trials (Figure 1). Table 1 shows the general characteristics and results of each of the 18 articles. The CASP quality assessment of the reviewed case–control studies and randomized clinical trials is shown in Tables 2, 3, respectively. The total quality score of case–control studies reached a maximum of 9/11 and that of the randomized clinical trial studies reached a maximum of 8–10/11.

The present systematic review provides an update of studies characterizing the gut microbiota in psychiatric disorders and captures the large number of studies published in the recent years. We confirm an association between psychiatric disorders and the gut microbiota composition. The studies we reviewed demonstrate differences between subjects with psychiatric disorders and controls. In light of the reported results, the microbiota may be a regulator of mood disorders and behavior through the brain-gut-microbiota axis, thus positioning itself as a promising target in disease diagnosis and therapeutic interventions.

The studies included in the present review suggest that there is an alteration in microbial diversity in patients with depression compared to controls. Uemura et al. (2019) found that individuals with improved gut microbial diversity had a lower CES-D score, indicating better mental status, as shown by previous studies (Aizawa et al., 2016; Liang et al., 2018). Moreover, the results highlighted a correlation between gut microbiota and stress. In the case of endometriosis, a clear correlation has been demonstrated between the proinflammatory cytokines NF-ĸB p65 and COX2 and Prevotella genus (Hemmings et al., 2017; Xu et al., 2017; Márquez-Morales et al., 2021), and similar results have been reported for bacterial vaginosis caused by pathogenic and cytokine-producing Prevotella species such as P. bivi (Randis and Ratner, 2019). Therefore, exposure to stress would appear to change the composition of the intestinal microbiota in the body, and the microbiota in turn modulates stress levels (Foster et al., 2017; Ilchmann-Diounou and Menard, 2020). Nutritional intervention studies with a fermented drink containing L. plantarum, L. paracasei and L. brevis showed a beneficial effect on stress reduction, pointing to probiotics as a potential therapy in such cases (Márquez-Morales et al., 2021).

Khine et al. (2020) found significant differences in the gut microbiota between aging and healthy subjects. SCFA-producing and anti-inflammatory taxa have been correlated with improved brain function. In contrast, pro-inflammatory taxa, such as Fusobacterium or some members of the Enterobacteriaceae family, have been correlated with decreased brain function (Khine et al., 2020). Similarly, other studies have shown a relationship between microbiota and cognitive behavior. Over the aging period, the organism produces an inflammatory response that can affect the balance of the intestinal microbiota (Hu et al., 2016). Therefore, the intestinal microbiota in the elderly differs from that in healthy adults (Ni et al., 2019). The microbiota can influence complex behaviors, such as learning, stress, depression and anxiety (Cryan et al., 2019), leading us to conclude that the status of the brain may be influenced by inflammatory processes, which are strongly related with the brain-gut microbiota communication. With respect to eating disorders, few differences in bacterial genera have been observed between BED and non-BED groups, but the changes reported are of interest, namely, a reduction of bacteria that produce SCFAs. These bacteria play an important role in the regulation of inflammation, immunity and secretion of peptides related to ingestion behavior (Leyrolle et al., 2021). Moreover, Akkermansia depletion may affect the regulation of the intestinal barrier, increasing permeability and the risk of infection. In addition, the use of this bacterium in clinical trials, directly or as a postbiotic, could potentially be of use in weight loss and protection against diabetes (Depommier et al., 2019). Similar results have been observed in patients with PSTD, in whom a reduction of A. muciniphila was associated with higher host-mediated inflammation and increased intestinal permeability, suggesting a key role for this bacterium in the treatment of this pathology (Hemmings et al., 2017). In Anorexia Nervosa, an altered composition of the intestinal microbiota was observed, with a reduced proportion of carbohydrate-fermenting genera correlating with a lower butyrate concentration. Furthermore, a higher proportion of taxa (such as the Enterobacteriacee family) was associated with intestinal inflammation, which promotes bacterial translocation and systemic inflammation (Borgo et al., 2017), a tendency opposite to that seen in obese subjects (Rieder et al., 2017). These studies showed that proinflammatory cytokines are elevated in patients with eating disorders, so increasing SCFA in patients with anorexia nervosa, due to the production of taxa such as Lachnospiraceae, may help to improve host histone epigenetic states and decrease levels of inflammatory markers (Morita et al., 2015; Yuan et al., 2022). Therefore, gut microbiota composition may affect the total amount of energy extracted from food intake, which is relevant for weight regulation (Seitz et al., 2020). The effect of the gut microbiota on patients with first-episode psychosis (FEP) is also relevant. Schwarz et al., 2018 identified differences in the gut microbial composition between FEP-patients and controls in relation to the severity of psychotic symptoms and global functioning assessment. A decrease in butyrate-producing taxa such as Faecalibacterium, Blautia, Ruminococcus or Roseburia could be a factor influencing the severity of symptoms (Schwarz et al., 2018). Similarly, in different studies on neurodevelopmental diseases (Kang et al., 2013; Vuong and Hsiao, 2017), the gut microbiota dysbiosis was identified in ASD subjects, who exhibited a higher abundance of pathogenic bacteria and a lower number of beneficial taxa, reduced levels of SCFAs and several metabolic disorders, in this sense a strong correlation was observed in Desulfovibrio with the severity of autism manifestations. Following an intervention consisting of a prebiotic compound and several probiotics, such as B. longum and Lactobacillus (L. paracasei and L. rhamnosus), a change in the gut microbiota was observed, as it became more similar to that of control children (Tomova et al., 2015), in addition to an increase in fermenting taxa that contributed to a significant increase in the concentration of SCFAs and an improvement in the metabolism of dopamine and tryptophan. In fact, microbiota modulation was shown to lead to an improvement in the health of ASD patients (Wang et al., 2020). Similar results showed that the gut microbiota is related to gastrointestinal symptoms and ASD (Kang et al., 2017; Kong et al., 2019). In the light of all this research, understanding the role of the microbiota in brain development should be a therapeutic target of future research into mental illness (Gárate et al., 2011; Sandhu et al., 2017; Valles-Colomer et al., 2019), since multiple studies have demonstrated this relationship (Hu et al., 2016; Borgo et al., 2017; Colica et al., 2017) and its influence on the psychoneuroimmunology network (El Aidy et al., 2015). More robust studies should be performed to demonstrate causality.

There are several limitations to the studies reviewed herein. It is very difficult to prove causality between bacterial taxa and these diseases due to the lack of both sufficient statistical power and consensus concerning interpretations of microbiota data. In many of the studies analyzed the sample size was too small to draw conclusions. In addition, there were many differences found in specific microbial members, although this could be linked to issues with the methodology and not with real inconsistency between groups. The absence of a metagenomic study of the patient’s intestinal populations over time preclude from performing a causal correlation. There are differences at the metagenomic sequencing level with some studies showing low sequences per sample that limit the results and should be taken into account when making comparisons between studies. Exogenous sources of inter-study heterogeneity (antidepressants, diet, sex) were also observed, such as Morita et al. (2015), that did not take into account the psychiatric medication of the subjects, if they were prescribed, or Guo et al. (2022), that took into account psychoactive substances, but not psychiatric medication. In addition, one of the most prominent limitations that can be observed is the inconsistency found in the different methods used for recruitment, screening and categorization of cases and controls, as many diagnoses are carried out by a self-reported methodology. In this sense, Wang et al. (2020) sampled medical subjects that represent a possible bias. In relation to age, Yuan et al. (2022) recruited both children and adults for the sample, which is quite heterogeneous, whereas Tomova et al. (2015) included two-year-old children with autism in the study, with the diagnosis being premature at such age (Ghosn et al., 2022). Another limitation is that all psychiatric symptoms in the reviewed studies should match the DSM-5 diagnostic criteria. Furthermore, in most studies the recruitment is at a local level so the diversity and composition of the gut microbiota of that population may not be representative of that from other countries or continents, making comparison more problematic. Further studies are needed to determine whether gut microbiota dysbiosis is a cause or an effect of the disease. In addition, given the heterogeneity of the studies, meta-analysis cannot be performed.

The articles evaluated for this review show a clear dysbiosis of the gut microbiota in all the psychiatric disorders studied: depression, stress, brain function, eating disorders, psychotic episodes and ASD. In general, a reduction of fermenting taxa has been observed, parallel to a decrease in SCFAs concentrations. This occurs in addition to an increase in certain opportunistic or pathogenic bacteria that can maintain a proinflammatory status. Interventional studies with prebiotic fibers, probiotics (especially Lactobacillus and Bifidobacterium species) or postbiotics (for example, pasteurized compounds of A. muciniphila) have shown the possibility of mitigating the symptoms associated with these diseases. Thus, it is feasible that the status of the brain is influenced by inflammatory processes, which reinforces the theory of a gut-brain axis communication.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

CG-DV and JF: conceptualization, formal analysis, investigation, methodology, validation, visualization, writing—original draft, and writing—review and editing. ES, IM-C, and CM: supervision, validation, visualization, and writing—review and editing. CB: conceptualization, formal analysis, funding acquisition, supervision, validation, visualization, and writing—review and editing. All authors contributed to the article and approved the submitted version.

This study was funded by a grant (PI18/00932 and PI21/01160) from the Institute of Health Carlos III and the European Regional Development Fund (ERDF, “A Way of Doing Europe”), CIPROM/2022/32 from Generalitat Valenciana and UGP-20-132 from FISABIO. CG-DV was a beneficiary of a PFIS contract (FI19/00076), JF was a beneficiary of a Plan de Ciencia, Tecnología e Innovación 2018-2022 del Principado de Asturias contract from Ficyt (Principado de Asturias, AYUD/2021/58584), and CB was a beneficiary of a Miguel Servet type I contract from the Institute of Health Carlos III (CP19/00077).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.