Antidepressants are used not only in the treatment of depression but also play an essential role in treating anxiety and obsessive–compulsive disorders. In addition, an exciting discovery is a fact that antidepressants also have antibacterial properties (Kalaycı et al., 2014). Consistently, therefore, recently published work shows that increased use of duloxetine, sertraline, fluoxetine and bupropion at appropriate doses - in the transformation mechanism- promotes the increased prevalence of antibiotic resistance genes (Lu et al., 2022).

Tricyclic antidepressants (TCA) have so far been tested in animal models. Clomipramine and imipramine are cytotoxic to human parasitic protozoa, i.e., Leishmania donovani (Mukherjee et al., 2020) and Leishmania major (Zilberstein and Dwyer, 1984). In combination with benznidazole, clomipramine is active against Trypanosoma cruzi (García et al., 2016). A solution of amitriptyline hydrochloride with bactericidal activity against 253 bacterial strains (72 gram-positive and 181 gram-negative) and five fungal strains was also developed. Moreover, when used in rodents infected with the pathogenic strain Salmonella typhimurium NCTC 74, the drug showed antimicrobial activity against the pathogens Staphylococcus spp., Bacillus spp., Vibrio cholera, Cryptococcus spp. and Candida albicans (Mandal et al., 2010). It has also been demonstrated that amitriptyline and clomipramine are active against methicillin-resistant Staphylococcus pseudintermedius (Brochmann et al., 2016). Promethazine and imipramine, on the other hand, showed an inhibitory effect against E. coli and Yersinia enterocolitica by interfering with plasmid replication (Csiszar and Molnar, 1992), and imipramine itself was active against Giardia lamblia (Weinbach et al., 1992). On the other hand, desipramine is effective against Plasmodium falciparum (Basco and le Bras, 1990) but also Lactobacillus casei, Akkermansia muciniphila and Bacteroides fragilis (Ait Chait et al., 2020).

Selective serotonin reuptake inhibitors (SSRIs) work against Brucellae and are synergistic in combination with antibiotics against Corynebacterium urealyticum (Muñoz-Bellido et al., 1996). SSRIs also inhibit mucus production in coagulase-negative staphylococci (Cussotto et al., 2019a). After analyzing the antimicrobial activity of the four SSRIs against E. coli, sertraline was the most potent antibacterial agent, but this was not clinically significant against Gram-negative bacteria (Bohnert et al., 2011). It was additionally confirmed that the growth of S. aureus ATCC 6538, E. coli ATCC 8739 and P. aeruginosa ATCC 9027 after plating with sertraline, especially in combination with antibiotics, is inhibited. In fungi, growth inhibition affected Aspergillus niger and A. fumigatus (Ayaz et al., 2015) Fluoxetine has a strong dose-dependent antimicrobial effect against L. rhamnosus and E. coli, including K12 strain (Jin et al., 2018) while escitalopram has only a minor antimicrobial effect on E. coli. In an in vivo experiment, it was also confirmed that fluoxetine and escitalopram significantly increase intestinal permeability in the ileum (Cussotto et al., 2019b). Fluoxetine is active against S. aureus, standard and resistant strains of P. aeruginosa and E. coli, especially in combination with gentamicin and erythromycin (Karine de Sousa et al., 2018; Batista de Andrade Neto et al., 2019; Dos Santos et al., 2020). Recent studies indicate that administration of fluoxetine in an in vitro model of sepsis inhibits the synthesis of interleukin 1 beta (IL-1β), interleukin 6 (IL-6), tumor necrosis factor alpha (TNF-α), myeloperoxidase (MPO) activity, monocyte chemoattractant protein – 1 (MCP-1), high sensitivity C-reactive prhs-CRP, procalcitonin (PCT), lactate, and the oxidative stress index (OSI), and di-sulfide levels (Cakir et al., 2022). In a 2019 study, it was confirmed that escitalopram oxalate inhibits the growth of E. coli at a concentration of 45, 15, and 5 mM, while Bacillus subtilis is not inhibited (Valipour et al., 2019). Escitalopram oxalate and clonazepam inhibited multidrug-resistant clinical isolates and standard bacterial strains from the American Type Culture Collection in an in vitro study, especially in combination with ciprofloxacin and sulfamethoxazole-trimethoprim. The mechanism of bacterial inhibition by clonazepam is the digestion of plasmid DNA (da Rosa et al., 2021). Exposure of bacteria to fluoxetine and sertraline solutions for 30 days has also been shown to promote the development of antibiotic resistance (Gurpinar et al., 2022). This observation is supported by other studies showing that fluoxetine and amitriptyline may increase the prevalence of aminoglycoside (aph3iiiA), multidrug (mdtK, mdtP, mdtH, mdtG, acrA), and tetracycline (tetM) resistance genes in the rats model (da Rosa et al., 2021).

Ketamine, whose mechanism of antidepressant action is not fully understood, also exhibits in vitro antibacterial activity against six different bacterial strains: S. aureus, S. epidermidis, E. faecalis, S. pyogenes, P. aeruginosa and E. coli, with S. aureus and S. pyogenes being the most sensitive. It is worth emphasizing that the culture of these organisms in this study was carried out using propofol - a compound that stimulates the growth of bacteria (Begec et al., 2013). Ketamine at a concentration of 2.49–3.73 mM inhibits the growth of methicillin-resistant S. aureus (MRSA; Coutinho et al., 2021), and in vitro model, it has activity against Stachybotrys chartarum, Staphylococcus epidermidis and Borrelia burgdorferi (Torres et al., 2018). In combination with fluconazole and itraconazole, it is active against C. albicans resistant to class azoles (de Andrade Neto et al., 2020).

In vivo studies using the supply of antidepressants are not numerous. In a recent study, Lukić et al. (2019) administered fluoxetine, escitalopram, venlafaxine, duloxetine or desipramine to lab animals and analyzed the drug-induced changes at the microbiota level using next-generation sequencing (NGS). After analyzing the results, it was found that alpha diversity and richness decreased in the test animals, and at the species level, the number of Ruminococcus, Adlercreutzia, and an unclassified Alphaproteobacteria decreased. When Ruminococcus flavefaciens or Adlercreutzia equolifaciens were transplanted into animals treated with fluoxetine, it was noticed that the first bug abolished the antidepressant effect of the drug. RNA analysis showed that R. flavefaciens stimulated changes in gene expression in the cerebral cortex, including downregulation of neuroplasticity genes and upregulation of oxidative phosphorylation genes in mitochondria. In turn, in mice exposed to mild stress (Zhang et al., 2021), administration of fluoxetine and amitriptyline resulted in a decrease in the Firmicutes/Bacteroidetes ratio, with fluoxetine having a tremendous potential to cause these changes. In turn, the increase in the Bacteroidaceae was associated primarily with amitriptyline. Both drugs significantly affected the counts of Parabacteroides, Butyricimonas, and Alistipes. At the metagenomic level, these drugs have been shown to alter the metabolic pathways involved in carbohydrate catabolism, membrane transport and signal transduction. This new direction of research gives hope for more effective treatment of psychiatric patients, but also for helping those patients who until now belonged to the group of people resistant to psychopharmacological treatment. Thanks to the better effectiveness of the administered drug, it is possible to lower its dose, reducing the risk of side effects.

Human studies are scarce. Liśkiewicz et al. (2019) observed that a six-week inpatient treatment with escitalopram in an acute depressive episode resulted in an increase in alpha diversity in the fecal microbiota, but a causal relationship with patients’ mental health was not proven. A reanalysis of these results using other bioinformatics and statistical methods did not confirm these observations (Liśkiewicz et al., 2021). Similarly, in the study by Lin et al., no effect of escitalopram (29 days, 10 mg/d) on the composition of the microbiota in patients with major depressive disorder (MDD) was found (Lin et al., 2017). These observations indicate the need to conduct further research in this area and, above all, standardize the microbiome analysis methodology and the statistical methods used to evaluate the results obtained. Bharwani et al. (2020), in a study conducted in a small group (n = 15) of patients with MDD, observed that although the administration of citalopram or escitalopram did not affect the composition of the microbiota during the 6-month treatment, but the microbiota differed between those who responded to the therapy and those who did not improve clinically. This may suggest a link between the microbiota and treatment response.

Microbiota modification with probiotics is a very promising method of supporting antidepressant treatment. Nikolova et al. (2021), in their updated review and meta-analysis of 404 depressed patients, found that psychobiotics were effective in reducing depressive symptoms when administered as an adjunct to antidepressant therapy (SMD = 0.83, 95%CI 0.49–1.17) but did not show benefit in monotherapy (SMD = 0.02, 95%CI 0.34–0.30). Potential mechanisms may be associated with an increase in the concentration of brain-derived neurotrophic factor (BDNF) and a decrease in the concentration of C-reactive protein (CRP) under their usage. Results of another meta-analysis, including 603 patients (Misera et al., 2021), showed the influence of psychobiotics on psychometric tests in patients with depression. However, the effect of psychobiotics was not of great clinical importance and was effective in combination with other antidepressants, which was confirmed in the meta-analysis of Nikolova et al. (2021). It should be emphasized that taking probiotics is well tolerated, safe and does not pose a risk in patients with MDD.

Additionally, this intervention is not associated with significant side effects (Misera et al., 2021). In mechanistic studies, Rudzki et al. (2019) and Kazemi et al. (2019a,b) observed an improvement in tryptophan metabolism due to probiotic therapy. Akkasheh et al. (2016) and Reiter et al. (2020) observed the anti-inflammatory effects of probiotics, which, however, have not been confirmed in other studies (Romijn et al., 2017; Rudzki et al., 2019). Heidarzadeh-Rad et al. (2020) reported an increase in the level of BDNF, which correlated with the response to antidepressant treatment. However, the meta-analysis did not confirm the effectiveness of the use of probiotics in the regulation of parameters related to the HPA axis (cortisol level), inflammation (interleukins, tumor necrosis factor—TNF) and the tryptophan degradation pathway (kynurenine). In another meta-analysis, Amirani et al. (2020) reported that the intake of probiotics by patients with various mental disorders (not only MDD) had a beneficial effect on CRP, IL-10 and malondialdehyde (MDA) levels but did not affect other markers of inflammation (TNF-alpha, IL-1B) and oxidative stress. However, due to the ambiguous results of the studies and the high heterogeneity of the studied populations, it can be said that the clinical mechanism of action of probiotics in improving the symptoms of MDD remains the subject of speculation.

It is known that the effect of probiotics/psychobiotics is strain-dependent. However, the meta-analyses conducted so far have not confirmed the effectiveness of a particular strain in patients with MDD. Interestingly, the administration of probiotics is associated with a better clinical response (Misera et al., 2021) and improves patients’ cognition. However, it has not been shown that the effect of probiotics is always associated with the impact on the intestinal microbiota composition, although it may determine its functions (Liśkiewicz et al., 2019). Similar observations apply to metabolic diseases, in which the microbiota does not seem to change its composition but the functions of the microbiome (Kaczmarczyk et al., 2022). Finally, it is worth noting that one study observed that gut barrier integrity and markers of inflammation (gut microbiota can affect both of these parameters) might be related to the response to treatment in MDD patients and the severity of symptoms (da Rosa et al., 2021). The observed changes, however, were of a correlation rather than a cause-and-effect relationship (Łoniewski et al., 2021).

Antipsychotics are, next to antidepressants and sedatives, the most commonly used drugs in psychiatry. Taking into account the antimicrobial properties of antipsychotic drugs, it seems that the metabolic disorders observed in the course of therapy with these drugs may be caused by unfavorable changes in the composition of microorganisms responsible for metabolic activity (Skonieczna-Żydecka et al., 2019). Indeed, clinical studies and animal studies have shown that treatment with Second Generation Antipsychotics (SGAs) induces significant changes in the abundance of the main types of intestinal microorganisms, resulting in weight gain, hypertriglyceridemia, hypercholesterolemia, hypertension and glucose metabolism, which significantly increases the risk of developing metabolic syndrome and cardiovascular events (Davey et al., 2012, 2013; Bahr et al., 2015a,b; Flowers et al., 2017; Riedl et al., 2017; Kao et al., 2018). Furthermore, the results of the latest scoping review, in which 46 studies were analyzed, showed that antipsychotic drugs were causing disturbances within the intestinal barrier, including intensification of gut-associated lymphoid tissue (GALT) activity and reduction of short-chain fatty acids (SCFA) synthesis, are associated with the induction of metabolic changes observed with such therapy (Singh et al., 2022; Figure 4).

Microbiota analysis of 76 hospitalized elderly patients by Ticinesi et al. (2017) showed that the use of antipsychotic drugs was strongly associated with the composition of the gut microbiome. An increased abundances of Prevotella, Victivallis and Desulfovibrionaceae species were observed during treatment. In an in vitro study by Cussotto et al. (2019b) it was found that administration of aripiprazole (ARI) significantly increased the richness and diversity of microbial species. At the genus level, several species belonging to Clostridium, Peptoclostridium, Intestinibacter and Christensenellaceae were increased after ARI treatment. Moreover, increased permeability in the ileum was recorded. Ma et al. (2020) observed the difference in the composition of the intestinal microbiota between a group of patients with schizophrenia who did not take medication and those treated with antipsychotics. Yuan et al. (2021) observed an increase in α-diversity and decreased Lachnoclostridium, and increased Romboutsia abundance in schizophrenia (SCH) patients after 24 weeks of risperidone (RIS) treatment. In addition, treatment response was significantly associated with basal levels of these bacteria.

The influence of the type of antipsychotic drugs on the microbiota has not been confirmed by Nguyen et al. (2021) in a well-designed and analyzed study that took into account the compositional nature of the microbiota and used machine learning. However, the results were affected by the data’s multidimensional nature and the study’s cross-sectional nature.

Long-term use of antipsychotics is associated with side effects, including significant weight gain, particularly with olanzapine (OLZ) and clozapine, moderate with quetiapine and risperidone (RIS), and lowest with ARI (Dayabandara et al., 2017). Such adverse metabolic effects also apply to children and adolescents in whom antipsychotics are used (Bretler et al., 2019).

Morgan et al. (2014), Davey et al. (2012, 2013), and Bahr et al. (2015a,b), in their experimental work, observed that weight gain during treatment with OLZ and RIS is closely related to the composition of the intestinal microbiota. Davey et al. reported that the co-administration of olanzapine and an antibiotic in rats attenuates the metabolic effects associated with olanzapine use (Davey et al., 2013). Kao et al. (2018) proved that the prebiotic galacto-oligosaccharides (B-GOS) significantly reduced body weight gain in rats treated with OLZ, while not affecting the drug’s effect. A recent study by Luo et al. (2021) showed that olanzapine given to rats for 35 days increased the Firmicutes:Bacteroidetes ratio. At the species level, OLZ treatment reduced the relative abundance of Ruminococcus bromii compared to controls. Phenotypically, the drug’s effect resulted in an increase in body weight, glucose concentration and unfavorable changes in lipid metabolism, as well as an increase in adipocyte volume and fatty liver. These symptoms subsided after the introduction of metformin - a drug inhibiting weight gain (Hu et al., 2014), which probably also causes microbiota changes (Silamiķele et al., 2021). Consequently, metformin restored microbiota homeostasis, although at the level of the intestinal ecosystem, there was a further reduction in the number Lactobacillus reuteri and Bacteroides pectinophilus, and also the growth of the abundance of Bacteroides uniformis, Bacteroides acidifaciens, Ruminococcus bromii, Desulfovibrio simplex, and Arcobacter butzleri (Luo et al., 2021).

Other side effects include adverse effects on glucose tolerance, increased cholesterol and triglyceride levels, and hypertension. These symptoms very often lead to the development of metabolic syndrome in patients (De Hert et al., 2011), and its prevalence is close to 30% (Sanchez-Martinez et al., 2018). Flowers et al. (2017) investigated the interactions between treatment and the gut microbiota in a cross-sectional study of 117 patients with bipolar disorder. Forty-nine people were subjected to SGA therapy, while 68 did not take any medications, constituting the control group. It has been noted that taking medications is associated with changes in the composition of the intestinal microbiota. The number of Lachnospiraceae, Akkermansia and Sutterella genera were changed. The number of Lachnospiraceae bacteria increased in the SGA group, while the number of Akkermansia decreased. Such a pattern was also noted in obese patients who were not using antipsychotics.

A cross-sectional study of 18 boys aged 8–15 years taking RIS was conducted by Bahr et al. (2015a). The control group included 10 patients aged 10–14 with psychiatric disorders who were not taking SGA. Chronic use of RIS led to an increase in the subjects’ body mass index (BMI) and a significantly lower ratio of Bacteroidetes to Firmicutes compared to the control group. Li et al. (2021) observed that after 24 weeks of RIS treatment, BMI, glucose, Homeostatic Model Assessment (HOMA-IR), Total-cholesterol, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C) and triglyceride levels were significantly changed compared to pre-treatment values. It was also found that the baseline abundance of Christensenellaceae and Enterobacteriaceae was significantly associated with changes in triglycerides, BMI and HOMA-IR after 24 weeks of RIS treatment. Similarly, Yuan et al. (2018) found that 24-week administration of RIS causes significant metabolic disorders and changes in the composition of the microbiota, among which changes in the abundance of Bifidobacterium spp. were correlated with changes in body weight.

Significant changes in the composition of the intestinal microbiome during 6 weeks of treatment with OLZ in patients with acute schizophrenia were not observed by Pełka-Wysiecka et al. (2019). The study involved 20 hospitalized patients taking daily doses of 5 to 20 mg of OLZ. Analysis of the intestinal microbiota of patients with SCH has shown that it is highly individual in taxonomy and functionality. In addition, it does not change during the 6-week therapy. Two types of enterotypes of the intestinal microbiome in patients with SCH were distinguished: type 1 was characterized by a predominance of Prevotella bacteria, while type 2 was characterized by a high amount of Bacteroides, Blautia and Clostridium. Pharmacotherapy was associated with clinical improvement in all patients and significant weight gain in women. The severity of symptoms at the beginning of therapy varied depending on the expected metabolic activity of the microbiota (Pełka-Wysiecka et al., 2019).

First-generation antipsychotics (FGA) also have antimicrobial properties. Thioridazine, a phenothiazine drug, has in vitro antimicrobial activity against methicillin-sensitive strains of S. aureus (Ordway et al., 2002; Hahn and Sohnle, 2014; Tozar et al., 2020), resistant to vancomycin: Enterococcus (Wainwright et al., 1999), Mycobacterium tuberculosis (Ordway et al., 2003; Amaral and Viveiros, 2017), Pseudomonas aeruginosa and Mycobacterium avium (Viveiros et al., 2005; Ruth et al., 2020). In addition, it has been shown to inhibit the growth of Plasmodium falciparum and Trypanosoma spp. (Hrynchuk and Vrynchanu, 2019), and also Salmonella enterica serovar Typhimurium 74 (Dasgupta et al., 2010). Fluphenazine has a pronounced activity against Gram-positive and Gram-negative bacteria, including Salmonella typhimurium (Dastidar et al., 1995). Recent results indicate that this drug protects against Candida glabrata infection due to inhibiting calmodulin [128]. Trifluoperazine strongly inhibited the growth of Shigella spp., Vibrio cholerae and Vibrio parahaemolyticus at concentrations of 10–100 μg/ml (Mazumder et al., 2001) but has also been shown to be active against replicating, non-replicating and refractory Mycobacterium tuberculosis (REDDY et al., 1996; Advani et al., 2012), and also Cryptococcus neoformans (Wang and Casadevall, 1996). Chlorpromazine, in turn, has anti-mycobacterial properties in vitro (Crowle et al., 1992) and restricts the growth of S. aureus and E. coli.(Coutinho et al., 2008). A recent study gave the possibility to use chlorpromazine-enriched catheters to prevent biofilm formation by Escherichia coli, Proteus mirabilis and Klebsiella pneumoniae in urinary tract infections (Sidrim et al., 2019). In one of the few and at the same time the most recent studies, Manchia et al. (2021) assessed the effect of both SGA and FGA on the microbiome of patients with SCH. At the type level, Bacteroidetes and Fusobacteria increased in FGA users, and at the family level, Fusobacteriaceae, Streptomycetaceae, Helicobacteraceae, and Bacillaceae. Clostridiales Family XII Incertae Sedis were not observed in this group of patients in contrast to the group treated with SGA. In the group taking typical antipsychotics, bacteria Fusobacterium, Butyricimonas, Blautia, Paraprevotella, Klebsiella were more numerous. At the species level, this group has increased counts of Parabacteroides merdae, Ruminococcus sp. AT10, Clostridium piliforme, Enorma massiliensis, Megamonas rupellensis, Bacteroides ovatus, Butyricimonas sp. GD2 and Ruminococcus sp. DJF VR70k1. At the same time, they were generally absent in patients taking SGA.

For other drugs used in psychiatry, such as anticonvulsants or opioid analgesics, the literature is scarce. Lithium and valproate do not inhibit E. coli and L. rhamnosus in vitro. However, in rodents, they can significantly change the microbiota inhabiting the caecum, causing changes in the number of Ruminococcaceae and Bacteroides (Cussotto et al., 2019b). Valproate used in patients with epilepsy for 3 months did not cause statistically significant changes in the microbiota, including diversity indices, but at the end of the observation, an increase in the Firmicutes/Bacteroidetes ratio was observed (Gong et al., 2022). One recent study also showed that valproate stimulates the production of trans-9-elaidic acid in S. ludwigii TBRC2149, a biomolecule previously found to inhibit beta-oxidation in macrophages (Poolchanuan et al., 2020). In turn, the latest research by Huang S, et al. (2022) showed that lithium carbonate used in a model of intestinal inflammation induced beneficial changes in the microbiota by normalizing diversity, an increase in the abundance of A. muciniphila and an increase in the synthesis of SCFA. In addition, in lamina propria, it stimulated the activation of anti-flooded regulatory T cells.

TREG cells (Huang S, et al., 2022). Lamotrigine has antibacterial activity against Gram-positive bacteria: B. subtilis, S. aureus and S. faecalis (Journal of Chemical Sciences | Indian Academy of Sciences, n.d.). Together with carbamazepine, it caused toxic effects in intestinal epithelial cells in vitro and inhibited the growth of several species of bacteria representing a healthy digestive system in children up to 9 years of age (such as Ruminococcus gnavus, Clostridium ramosum, and Roseburia intestinalis) and patients with drug-resistant epilepsy (including “anti-epileptic” Parabacteroides species), while the growth of Staphylococcus caprae, Dorea longicatena, E. coli, and Klebsiella aerogenes growth was enhanced (Ilhan et al., 2022). Tramadol has strong in vitro bactericidal activity against E. coli and S. epidermidis and weak antimicrobial activity against S. aureus and P. aeruginosa (Tamanai-Shacoori et al., 2007), and methadone against S. aureus, P. aeruginosa and S. marcescens (Sheagren et al., 1977). Tramadol administered at concentrations of 12.5 and 25 mg/ml to BALB mice infected with Staphylococcus aureus and Pseudomonas aeruginosa induced a reduction in wound diameter (by inflammation induction and phagocytic activity) caused by S. aureus (Farzam et al., 2018). After the administration of morphine, changes in the composition of the gut microbes in rodents were associated with a significant increase in the number of pathogenic bacteria and a decrease in stress tolerance communities (Wang et al., 2018). In a cohort of patients with cirrhosis, chronic opioid use was associated with a reduction in the relative abundance of Bacteroidaceae and Ruminococcaceae (Acharya et al., 2017). Cruz-Lebron et al. examined the microbiota in an opioid-dependent mouse model (Cruz-Lebrón et al., 2021). Repeated intraperitoneal treatment with methadone resulted in a decrease in A. muciniphila numbers and, consequently, in the synthesis of SCFA. In another study patients received opioid agonists (heroin, prescription opioids), antagonists (naltrexone), a combination agonist–antagonist (buprenorphine-naloxone), or no medication. Those using opioid agonists (no antagonists) had lower microflora diversity, Bacteroides enterotype, and lower relative abundance of Roseburia, a butyrate-producing genus, and Bilophila, a bile acid metabolizing genus (Gicquelais et al., 2020).