Medicinal Plants and Their Impact on the Gut Microbiome in Mental Health: A Systematic Review

Background: Various neurocognitive and mental health-related conditions have been associated with the gut microbiome, implicating a microbiome–gut–brain axis (MGBA). The aim of this systematic review was to identify, categorize, and review clinical evidence supporting medicinal plants for the treatment of mental disorders and studies on their interactions with the gut microbiota. Methods: This review included medicinal plants for which clinical studies on depression, sleeping disorders, anxiety, or cognitive dysfunction as well as scientific evidence of interaction with the gut microbiome were available. The studies were reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. Results: Eighty-five studies met the inclusion criteria and covered thirty mental health-related medicinal plants with data on interaction with the gut microbiome. Conclusion: Only a few studies have been specifically designed to assess how herbal preparations affect MGBA-related targets or pathways. However, many studies provide hints of a possible interaction with the MGBA, such as an increased abundance of health-beneficial microorganisms, anti-inflammatory effects, or MGBA-related pathway effects by gut microbial metabolites. Data for Panax ginseng, Schisandra chinensis, and Salvia rosmarinus indicate that the interaction of their constituents with the gut microbiota could mediate mental health benefits. Studies specifically assessing the effects on MGBA-related pathways are still required for most medicinal plants.


Introduction
Stress, anxiety, mood disorders, sleep problems, and cognitive dysfunction are the most common mental health problems for which herbal products constitute a reasonable treatment option with minor side effects and low toxicity [1,2]. The pathogenesis of mental disorders is complex and generally thought to be linked to genetic, immune-related, humoral, neural, and environmental factors. However, various neurocognitive and mental health conditions have been strongly associated with imbalances in the gut microbiome composition, referred to as dysbiosis [3].

The Microbiome-Gut-Brain Axis (MGBA)
It is important to consider the symbiotic relationship between humans and their resident microbes when discussing the role of the gut-brain axis in behavior, health, and disease [4]. The sharp increase in various disease states in recent decades [5,6] could be explained, at least in part, by the changes in modern diets and lifestyles that have negatively impacted the composition and diversity of the human gut microbiome [7]. The gut microbiome could be the missing link in the conceptualization and treatment of psychological disorders [4]. The microbiome-gut-brain axis (MGBA) provides a network for signals from the brain to influence the motor, sensory, and secretory functions of the gut while simultaneously allowing signals and metabolites from the gut microbiome to influence brain development, biochemistry, function, and behavior [8][9][10].
The human intestinal microbiome predominantly consists of anaerobic bacteria, with the Firmicutes, Bacteroidetes, Proteobacteria, and Actinobacteria phyla constituting more than 90% of the total microbiota [11]. The gut microbiome is regarded as an important factor in bidirectional communication between the gut and the brain (gut-brain axis) [12][13][14]. This communication is based on several complex pathways that typically transmit sensory information from the gastrointestinal (GI) tract and subsequently convert it into hormonal, neural, and immunological signals. These signals further transmit information to the central nervous system (CNS) either individually or cooperatively [15]. Figure 1a shows how the gut microbiome can influence brain function via the gut-brain axis, thereby regulating behavior and psychological processes [12,[16][17][18][19]. Microbiota-gutbrain interactions are thought to occur via three major pathways: (i) direct and indirect signaling via chemical transmitters such as microbial metabolites (e.g., short-chain fatty acids, or SCFAs), hormones, or neurotransmitters that can be either directly synthesized or modulated in their levels by gut microbiota; (ii) neural pathways, e.g., modulation of vagus nerve activity; and (iii) signaling within the immune system, e.g., microglia-mediated effects or effects of circulating cytokines that can modulate the activity of the hypothalamic-pituitary-adrenal (HPA) axis [11,12,16,[20][21][22].

Essential Oils and Herbs Rich in Essential Oils
Thirteen in vitro and in vivo studies assessing the effect on the gut microbiota of mental health-related essential oils (orange blossom oil, lavender oil) or herbal drugs rich in essential oils (lemon balm leaf, rosemary, lemon verbena leaf, black cumin, and turmeric root) met the inclusion criteria. For lavender (Lavandula angustifolia Mill.) oil, traditional use is stated for the indication of sleep disorders, temporary insomnia, mental activation/upregulation, Nutrients 2022, 14, x. https://doi.org/10.3390/xxxxx www.mdpi.com/journal/nutrients Figure 3. Potential gut-brain communication pathways modulated by triterpenes in mental disorders. Triterpenes (such as ginsenosides) can alter gut-brain microbiome communication in impaired brain function and promote a healthy mental state. These beneficial effects are related to rebalancing the gut microbiome and influencing neural (blue letters), immune (pink letters), and humoral/metabolic (green letters) pathways. Triterpene glycosides are metabolized by the gut microbiome into active components (e.g., ginsenosides into compound K). These active metabolites are more bioavailable than the native compounds. Ginsenosides and their metabolites promote neurotrophic factors and reduce pro-inflammatory mediators and stress levels [207,[209][210][211]. The major gut-brain mechanisms by which ginsenosides have a beneficial effect are marked with red arrows ( activation/upregulation, inhibition/downregulation). TJPs: tight junction proteins; BDNF: brain-derived neurotrophic factor; PI3K: phosphoinositol 3 phosphate; AKT: protein kinase B; IL-1β: interleukin-1β; NF-κB: nuclear factor-κB; PYY: peptide YY; GLP1: glucagon-like peptide 1; : stimulates/promotes.
Regarding chemical signaling, microbiota-derived metabolites, and in particular SC-FAs, are important signaling molecules. SCFAs are produced from carbohydrates by certain GI tract microorganisms and regularly absorbed by the colonocytes through H + -dependent or sodium-dependent monocarboxylate transporters. SCFAs are responsible for several local effects, including maintenance of intestinal barrier integrity, mucus production, and anti-inflammatory effects (lowering the risk for colorectal cancer). These beneficial effects of SCFAs, in turn, improve overall gut health [23]. Moreover, SCFAs exert substantial systemic hormone-like actions and show immunomodulatory and neuroactive properties [12,16,24]. SCFAs also control the production of gut peptides by enteroendocrine cells (EECs). These peptides modulate the gut-brain axis and stimulate the synthesis of gut-derived serotonin from enterochromaffin cells (ECs), subsequently influencing gut-brain hormonal communication [16]. Moreover, SCFAs can cross the blood-brain barrier (BBB) and control microglia homeostasis in the brain. This process is thought to be involved in proper brain development and in modulating behavior [14,16,25]. Butyrate, in particular, is of major interest given its ability to regulate gene transcription and has been shown to have an antidepressant effect in mice [26].
Apart from SCFAs, the gut microbiota can produce neurotransmitters in the epithelial lining and convert their precursors to active metabolites in the gut lumen [27]. Various GI bacteria such as Lactobacillus spp., Bifidobacterium spp., Bacillus spp., Escherichia spp., and Saccharomyces spp. are involved in the production of neurotransmitters such as gammaaminobutyric acid (GABA), acetylcholine, noradrenaline, dopamine, and serotonin, and in the production of the serotonin precursor tryptophan. These neurotransmitters can, in turn, control neural signaling within the enteric nervous system (ENS) and eventually modulate brain function and behavior [12,16,28]. While neurotransmitters produced in the gut may not directly influence the brain as they do not pass through the BBB, they are able to influence the CNS through mechanisms including direct stimulation of the vagus nerve, as well as indirect circulatory and immune pathways [29]. Serotonin, the most well-studied neurotransmitter in relation to depressive illness, appears to be particularly susceptible to being influenced by the gut microbiome. A key study in 2009 revealed that the plasma serotonin levels of germ-free mice were almost three times less than those of conventional mice [30]. It was subsequently demonstrated that this differential serotonin level was secondary to the remarkable ability of gut microbes to directly promote the synthesis of serotonin from its amino acid precursor, tryptophan, in intestinal enterochromaffin cells [31]. Furthermore, the gut microbiome was also shown to influence serotoninergic levels in the hippocampus, an area of the brain which plays an important role in stress, anxiety, and depression [32]. Lyte [33] stated that probiotics function mechanistically as delivery vehicles for neuroactive compounds and that these probiotics have the potential to act as psychotropic agents.
The aim of this review was to assess the available scientific literature for potential links between the efficacy of medicinal plants used for mental health conditions and their interaction with gut microbiota. For this purpose, we scrutinized published data from clinical studies of medicinal plants for mental disorders and from studies assessing the interaction of these plants with gut microbiota.

Eligibility Criteria
Inclusion and exclusion criteria of studies were as follows: Medicinal plants were included in the systematic survey if there was clinical evidence for effects on depression, sleep, anxiety, mood, or cognitive dysfunction and there were studies available (in vitro studies, in vivo studies involving humans and animals except for ruminants and birds) that evaluated an interaction of these medicinal plants and gut microbiota. Only studies performed with the listed plant parts or extracts were considered relevant. No studies on combinations of herbal extracts were included. In the literature survey in Tables 1-3 we also excluded published data on pure compounds occurring as main constituents in these extracts but mention them in the discussion of the results when relevant. Studies concerning neurodegenerative diseases such as Alzheimer's and Parkinson's were excluded because the neurodegenerative nature of these diseases places them in a separate category.      [193]

Search Strategy
Data were successively gathered from the PubMed/Medline and Embase databases (https://www.ncbi.nlm.nih.gov/pmc; https://www.embase.com; last accessed: 5 January 2021). The reference lists of all retrieved review articles were also checked for additional related articles. For the first aim of retrieving all studies dealing with the effects of medicinal plants on mental health, the following search strategy, steps, and general keywords were used in PubMed: ((medicinal plant *) AND ((antidepressant) OR (mental stress) OR (mood disorder *) OR (insomnia) OR (sleep) OR (anxiety) OR (cognitive impairment *) OR (circadian clock) OR (circadian rhythm) OR (dementia) OR (memory) OR (adaptogen *) OR (focus and attention) OR (fatigue)) NOT ((Alzheimer's disease *) NOT (Parkinson's disease *)). In the second step, the focus was on clinical effects of the mental health-related medicinal plants identified from the studies retrieved in the first step. Their botanical plant names were specifically searched in PubMed using the following search string: (plant name OR plant name OR . . . . . . ) AND (clinical study) AND (anxiety) OR (insomnia) OR (antidepressant) OR (cognitive impairment *) OR (fatigue) OR (memory)).
The third goal was the identification of published data on the interaction of the identified medicinal plants and the gut microbiome. The relevant literature was searched in PubMed and in Embase. For PubMed, search terms were (plant name OR plant name OR . . . . . . ) AND ((gut microbiome) OR (gut microbiota) OR (gut bacteria)); for Embase, search terms were "plant name" AND "gut microbiome" OR "gut microbiota" OR "gut bacteria" OR "intestine flora".
In the last step, the medicinal plants with reported clinical mental health effects and that were also evaluated in studies of the gut microbiome were selected.. The search strategy is shown in the PRISMA flowchart in Figure 2. The searched data were transferred to the Citavi literature management program.

Study Selection
The titles of all retrieved papers were examined, and studies inconsistent with the objectives of this systematic review were excluded. In the next step, the abstracts of the remaining studies were examined, and again, incompatible studies not meeting the inclusion criteria (see Section 2.1) were excluded. Then, data were extracted from the full texts of the compatible studies and tabulated using standardized information, such as botanical names, medicinal plant parts used, common or local name(s), main constituents, and the field in which the clinical studies had been conducted.

Results and Discussion
A total of 6844 records were identified from the database searches concerning medicinal plants used for mental health, with 1503 articles related specifically to clinical studies. The second search was for studies of gut microbiota that included the use of these plants with mental health effects, yielding 34 medicinal plants with 887 records. Of these articles, after screening of the title and abstract, 677 were excluded based on the criteria described above (Section 2.1). The remaining 210 full-text articles were further reviewed and screened based on the inclusion criteria, yielding 85 articles on gut microbiome interactions with 30 mental health-related medicinal plants for inclusion in this systematic review. The flowchart of the included studies is depicted in Figure 2. Table 1 displays the list of the 30 medicinal plants with a clinically proven impact on mental health and for which studies on gut microbiome interactions were available. The included studies on gut microbiota were performed with the same plant parts or extracts as used in the clinical studies. In vitro and in vivo data on gut microbiome interactions are detailed in Tables 2 and 3.

In vitro studies
Of the 16 in vitro studies that met the inclusion criteria, 12 were performed with colon microorganisms from human fecal samples. Nine of these twelve studies used single fecal samples from either one or several donors, and the remaining three used pooled fecal samples. In the four nonhuman studies, three used fecal samples from different experimental animals (rat, mouse, dog), and one study applied a set of single microbial strains representing major intestinal genera [88]. A total of 14 of the 16 studies used simple static batch fermentations, preceded in 4 cases by static simulation of upper GI tract digestion [66,168,191,201]. Another two studies applied more sophisticated dynamic digestion models with sequential upper intestinal tract digestion and colonic fermentation [182,193]. Nine of the sixteen in vitro studies assessed both the microbial composition and metabolite changes during incubation with a herbal material. Of the remaining seven, three assessed only microbiome changes, and four investigated only metabolite profile changes during incubation. The metabolites most often studied in vitro were the SCFAs formed by gut microbial metabolization of plant polysaccharides, followed by metabolites derived from polyphenols and triterpenes. Microbial community composition changes were most frequently monitored by 16S rRNA gene sequencing (six studies), fluorescence in situ hybridization (FISH) (four studies), or qPCR (three studies). The study with single strains used cultivation-based agar dilution.

In vivo studies
Of the 69 in vivo studies that met the inclusion criteria, 11 were clinical, and 58 involved various experimental animal species (34 in mice, 15 in rats, 5 in pigs, and 1 each in rabbits, dogs, C. elegans, and Drosophila).
The human studies enrolled comparatively small participant numbers, with intervention group sizes ranging from 6 to 38. Different intervention groups (i.e., placebo vs. treatment or different treatments) were compared in only three of these studies, whereas the remaining eight assessed different treatments in a crossover design or compared the effect of a certain treatment on gut microbiota or metabolite profiles in samples taken before and after the intervention. In all studies, fecal samples were collected for assessment of fecal microbiota changes (seven studies), metabolite changes (two), or both (two). Ten of the studies enrolled healthy (in some cases overweight) patients, and one study enrolled participants with type 2 diabetes mellitus. This latter study assessed the effect of a herbal intervention on depression scores and on the GI tract microbiome composition [68], and thus is the only human study that directly investigated a correlation between a mental health condition and the gut microbial community composition. Most of the in vivo studies in experimental animals involved mice and rats. In general, the same bacterial phyla occur in rodents and humans, predominantly Bacteroidetes and Firmicutes. The Clostridium superfamily is also widespread in rats and humans, but there are marked differences in the abundance of important genera such as Lactobacillus and Bifidobacterium between humans and rodents [202,203]. Of these 58 studies, 27 used healthy animals, and 31 relied on different disease models, most commonly obese animals and colitis induced by dextran sodium sulfate (DSS), along with models of diabetes mellitus type 2, hypercholesterolemia, nonalcoholic fatty liver disease, menopause, and colorectal cancer. In five of the studies, the effects of medicinal plants on the gut microbiota in animal models were assessed related to mental health disorders, such as depression-like behavior, anxiety-and depression-like behavior, and memory impairment [42,106,172,173,204]. Changes in the gut microbial community composition were investigated in 33 of these studies, metabolite changes in 4, and both metabolite and microbial community changes in 21, all with fecal samples from the living animals or fecal content or mucosa from different intestinal regions collected after sacrifice. The technique most widely used to assess microbiota changes in human and animal studies was 16S rRNA gene sequencing (applied in 43 studies). Other commonly used techniques were qPCR with primers targeting specific bacterial groups or genera, and cultivation-based methods (bacterial plate counting, agar dilution). The microbial metabolites most commonly studied were SCFAs, the microbial fermentation products of polysaccharides (determined in 23 in vivo studies). In some of the studies, microbial metabolites of secondary plant metabolites such as ginsenosides [148,150] or phenolic compounds [200] were investigated. In the following sections, we group the data on MGBA interactions of herbal drugs into the major secondary metabolites present in these plants.

Herbal Drugs Containing Saponins
Many herbal drugs from medicinal plants with clinical effects in mild depression, anxiety, cognitive impairment, insomnia, and fatigue contain triterpenoid saponins (Radix Astragali, Herba Centellae, Radix Ginseng, Radix Polygalae, Folium Gynostemmae) or steroid saponins (Radix Polygonatae, Semen Foenugraeci). Saponins have long been of interest for their potential therapeutic benefits in many diseases, but their poor pharmacokinetic properties, with an extremely low bioavailability (frequently < 1.0%), have hampered the translation of these compounds into drugs. Mechanisms of action of saponin-rich plants on the CNS are largely unknown, and their metabolizaton by and modification of gut microbiota have therefore emerged as potential targets.
Trigonella foenum-graecum L. and Polygonatum sibiricum Redoutè are medicinal plants with effects on mental health that contain substantial amounts of steroidal saponins. T. foenum-graecum substantially corrected the dysbiotic effect of a high-fat diet (HFD) in mice, especially regarding the Firmicutes phylum [177]. The addition of T. foenum-graecum to feed positively influenced the gut microbiome composition and immune parameters in weaning piglets [178], and in cultivation-based plate count assays, a saponin-rich P. sibiricum extract increased the abundance of probiotic bacteria and decreased the abundance of potentially harmful species [164].
Polygala tenuifolia Willdenow is mainly used as a standardized ethanolic root extract (BT-11) that is rich in triterpene saponins and has neuroprotective and antidepressant effects [158,159]. Upon in vitro incubation with intestinal bacteria, 25 triterpene metabolites formed by deglycosylation and deacetylation reactions could be detected [162]. In rats, 29 triterpene metabolites were identified in feces after the administration of an ethanolic P. tenuifolia root extract, indicating that these metabolites are not absorbed in vivo but can have local effects on the intestinal microbiome. The altered microbiome may, in turn, indirectly affect brain function through the MGBA [160].
Astragalus membranaceus root contains triterpene saponins with the marker compound astragaloside IV, in addition to various compound classes such as flavonoids, polysaccharides, and amino acids.
The authors of one animal study found a significant increase in gut microbiota richness and diversity in a mouse model of type 2 diabetes and a significantly altered relative abundance of several bacterial taxa, inducing an increased abundance of Lactobacillus and Bifidobacterium [71]. Increases in both genera have been associated with mental health.
Leaf extracts of Gynostemma pentaphyllum (Thunb.) Makino, another mental healthrelated, triterpene saponin-rich medicinal plant, also significantly increased Bifidobacterium and Lactobacillus abundance and displayed prebiotic-like effects with a significant growth stimulation of SCFA-producing bacteria [120]. Furthermore, in a murine colon cancer model, treatment with G. pentaphyllum saponins led to an increase in potentially healthbeneficial bacteria, and significantly reduced sulfate-reducing bacteria [116,119]. In addition, treatment with G. pentaphyllum saponins increased the Bacteroidetes/Firmicutes ratio in normal [120] and HFD-fed animals [115,118]. Similar to G. pentaphyllum, treatment with P. tenuifolia root aqueous extract increased the Bacteroidetes/Firmicutes ratio in HFD-fed mice [161]. The aerial parts of Centella asiatica (L.) Urban, a herbal brain tonic for mental disorders [80], significantly reduced stress-related depression and anxiety [81]. C. asiatica is rich in triterpenoids, specifically asiaticoside, and has shown gut microbiota-modulating properties in a murine colitis model [82].
The best examined medicinal drug influencing the brain and nervous system is ginseng root from Asian ginseng (Panax ginseng C.A. Mey.) or American ginseng (Panax quinquefolius L.). Numerous randomized, double-blind, placebo-controlled trials have evaluated the efficacy of ginseng for cognitive performance, neurotransmission modulation, memory and learning enhancement, and neuroprotection. Effects have been attributed to a group of ginseng-specific triterpenoid saponins known as ginsenosides. Based on their structures, they are classified into three groups: panaxadiols, panaxatriols, and oleanolic acids [141,146,147].
Ginseng root extracts exert prebiotic-like effects by increasing the abundance of Lactobacillus and Bifidobacterium in rats [120,142,144] and support the restoration of the intestinal microbiome in antibiotic-treated mice [149]. Recent studies have demonstrated a link between the community structure of the gut microbiome and the gut microbial metabolism of ginsenosides. The three most abundant gut microbial metabolites are ginsenoside Rg 3 , ginsenoside F2, and compound K, formed from the protopanaxadiol group through stepwise cleavage of the sugar moieties [153]. Very high levels of compound K and low levels of the progenitor compound ginsenoside Rb1 were found in human feces after oral administration of American ginseng in healthy volunteers [150].
Host-related factors such as stress or diet lead to changes in the gut microbiome composition and function, which affect the efficiency of ginsenoside metabolism and absorption. Different dietary habits may result in differing gut microbiota populations, in turn affecting gut microbial metabolism and absorption of herbal constituents. For example, distinct fecal levels of ginsenoside Rb1 and compound K have been observed in healthy volunteers with dissimilar dietary habits [151]. After oral administration of an ethanolic extract of American ginseng, compound K was undetectable in antibiotic-treated mice but could be detected in stool samples from vehicle-treated mice [148]. Rats with different degrees of gut microbial metabolism of ginsenosides to compound K have shown different gut microbiome compositions. Isolated colonic Bifidobacterium spp. exhibited converting activity of ginsenosides Rb1, Rb2, and Rc to compound K [145]. According to a recent literature review, the main gut microbial genera involved in ginsenoside biotransformation are Bacteroides, Bifidobacterium, and Eubacterium [205].
Ginseng saponins such as ginsenosides Rb1 and Rg1, as well as their partially deglycosylated counterparts ginsenoside Rg 3 and compound K, have shown antidepressant and anxiolytic effects in various animal models via regulation of neurotransmitters (serotonin, norepinephrine, dopamine, GABA), the HPA axis, the glutamatergic system, BDNF, and intracellular signaling pathways in the CNS. They also reduce the secretion of proinflammatory factors (IL-1β, IL-6, TNF-α) and increase the production of anti-inflammatory cytokines (IL-4 and IL-10) [206,207]. The question, which of these mental health-beneficial effects are exerted via direct effects and which are due to indirect mechanisms occurring via the MGBA cannot be answered on the basis of currently existing data. One aspect that deserves particular consideration is that compound K, the major gut microbial metabolite of ginseng saponins, has better bioavailability than its progenitor compounds [207,208].
Taken together, as shown in Figure 3, the available data suggest that triterpenes may modulate an imbalanced microbiome-gut-brain communication during impaired brain functions and promote mental health [207,[209][210][211][212][213]. G. pentaphyllum, C. asiatica, and P. ginseng exerted prebiotic-like effects and led to a recovered intestinal flora diversity or mitigated gut dysbiosis compared with control groups in rodent models [82,115,116,120,142,144,149].
According to many preclinical studies, certain types of triterpenes possess antiinflammatory, antioxidant, and antiapoptotic properties and thus may contribute to neuronal protection [206,213]. Moreover, triterpene glycosides can be metabolized by gut microbiota into better absorbable active metabolites that become systemically available [206][207][208]. According to many preclinical studies, certain types of triterpenes possess anti-inflammatory, antioxidant, and antiapoptotic properties and thus may contribute to neuronal protection [206,213]. Moreover, triterpene glycosides can be metabolized by gut microbiota into better absorbable active metabolites that become systemically available [206][207][208]. Figure 3. Potential gut-brain communication pathways modulated by triterpenes in mental disorders. Triterpenes (such as ginsenosides) can alter gut-brain microbiome communication in impaired brain function and promote a healthy mental state. These beneficial effects are related to rebalancing the gut microbiome and influencing neural (blue letters), immune (pink letters), and humoral/metabolic (green letters) pathways. Triterpene glycosides are metabolized by the gut microbiome into active components (e.g., ginsenosides into compound K). These active metabolites are more bioavailable than the native compounds. Ginsenosides and their metabolites promote neurotrophic factors and reduce pro-inflammatory mediators and stress levels [207,[209][210][211]. The major gut-brain mechanisms by which ginsenosides have a beneficial effect are marked with red arrows ( activation/upregulation, inhibition/downregulation). TJPs: tight junction proteins; BDNF: brain-derived neurotrophic factor; PI3K: phosphoinositol 3 phosphate; AKT: protein kinase B; IL-1β: interleukin-1β; NF-κB: nuclear factor-κB; PYY: peptide YY; GLP1: glucagon-like peptide 1; : stimulates/promotes.

Essential Oils and Herbs Rich in Essential Oils
Thirteen in vitro and in vivo studies assessing the effect on the gut microbiota of mental health-related essential oils (orange blossom oil, lavender oil) or herbal drugs rich in essential oils (lemon balm leaf, rosemary, lemon verbena leaf, black cumin, and turmeric root) met the inclusion criteria. For lavender (Lavandula angustifolia Mill.) oil, traditional use is stated for the indication of sleep disorders, temporary insomnia, mental Triterpenes (such as ginsenosides) can alter gut-brain microbiome communication in impaired brain function and promote a healthy mental state. These beneficial effects are related to rebalancing the gut microbiome and influencing neural (blue letters), immune (pink letters), and humoral/metabolic (green letters) pathways. Triterpene glycosides are metabolized by the gut microbiome into active components (e.g., ginsenosides into compound K). These active metabolites are more bioavailable than the native compounds. Ginsenosides and their metabolites promote neurotrophic factors and reduce pro-inflammatory mediators and stress levels [207,[209][210][211]. The major gut-brain mechanisms by which ginsenosides have a beneficial effect are marked with red arrows ( 2022, 14, x. https://doi.org/10.3390/xxxxx www.mdpi.com/journal/nutrients According to many preclinical studies, certain types of triterpenes possess anti-inflammatory, antioxidant, and antiapoptotic properties and thus may contribute to neuronal protection [206,213]. Moreover, triterpene glycosides can be metabolized by gut microbiota into better absorbable active metabolites that become systemically available [206][207][208]. Figure 3. Potential gut-brain communication pathways modulated by triterpenes in mental disorders. Triterpenes (such as ginsenosides) can alter gut-brain microbiome communication in impaired brain function and promote a healthy mental state. These beneficial effects are related to rebalancing the gut microbiome and influencing neural (blue letters), immune (pink letters), and humoral/metabolic (green letters) pathways. Triterpene glycosides are metabolized by the gut microbiome into active components (e.g., ginsenosides into compound K). These active metabolites are more bioavailable than the native compounds. Ginsenosides and their metabolites promote neurotrophic factors and reduce pro-inflammatory mediators and stress levels [207,[209][210][211]. The major gut-brain mechanisms by which ginsenosides have a beneficial effect are marked with red arrows ( activation/upregulation, inhibition/downregulation). TJPs: tight junction proteins; BDNF: brain-derived neurotrophic factor; PI3K: phosphoinositol 3 phosphate; AKT: protein kinase B; IL-1β: interleukin-1β; NF-κB: nuclear factor-κB; PYY: peptide YY; GLP1: glucagon-like peptide 1; : stimulates/promotes.

Essential Oils and Herbs Rich in Essential Oils
Thirteen in vitro and in vivo studies assessing the effect on the gut microbiota of mental health-related essential oils (orange blossom oil, lavender oil) or herbal drugs rich in essential oils (lemon balm leaf, rosemary, lemon verbena leaf, black cumin, and turmeric root) met the inclusion criteria. For lavender (Lavandula angustifolia Mill.) oil, traditional use is stated for the indication of sleep disorders, temporary insomnia, mental According to many preclinical studies, certain types of triterpenes possess anti-inflammatory, antioxidant, and antiapoptotic properties and thus may contribute to neuronal protection [206,213]. Moreover, triterpene glycosides can be metabolized by gut microbiota into better absorbable active metabolites that become systemically available [206][207][208]. Figure 3. Potential gut-brain communication pathways modulated by triterpenes in mental disorders. Triterpenes (such as ginsenosides) can alter gut-brain microbiome communication in impaired brain function and promote a healthy mental state. These beneficial effects are related to rebalancing the gut microbiome and influencing neural (blue letters), immune (pink letters), and humoral/metabolic (green letters) pathways. Triterpene glycosides are metabolized by the gut microbiome into active components (e.g., ginsenosides into compound K). These active metabolites are more bioavailable than the native compounds. Ginsenosides and their metabolites promote neurotrophic factors and reduce pro-inflammatory mediators and stress levels [207,[209][210][211]. The major gut-brain mechanisms by which ginsenosides have a beneficial effect are marked with red arrows ( activation/upregulation, inhibition/downregulation). TJPs: tight junction proteins; BDNF: brain-derived neurotrophic factor; PI3K: phosphoinositol 3 phosphate; AKT: protein kinase B; IL-1β: interleukin-1β; NF-κB: nuclear factor-κB; PYY: peptide YY; GLP1: glucagon-like peptide 1; : stimulates/promotes.

Essential Oils and Herbs Rich in Essential Oils
Thirteen in vitro and in vivo studies assessing the effect on the gut microbiota of mental health-related essential oils (orange blossom oil, lavender oil) or herbal drugs rich in essential oils (lemon balm leaf, rosemary, lemon verbena leaf, black cumin, and turmeric root) met the inclusion criteria. For lavender (Lavandula angustifolia Mill.) oil, traditional use is stated for the indication of sleep disorders, temporary insomnia, mental inhibition/downregulation). TJPs: tight junction proteins; BDNF: brain-derived neurotrophic factor; PI3K: phosphoinositol 3 phosphate; AKT: protein kinase B; IL-1β: interleukin-1β; NF-κB: nuclear factor-κB; PYY: peptide YY; GLP1: glucagon-like peptide 1; ⊕: stimulates/promotes.

Essential Oils and Herbs Rich in Essential Oils
Thirteen in vitro and in vivo studies assessing the effect on the gut microbiota of mental health-related essential oils (orange blossom oil, lavender oil) or herbal drugs rich in essential oils (lemon balm leaf, rosemary, lemon verbena leaf, black cumin, and turmeric root) met the inclusion criteria. For lavender (Lavandula angustifolia Mill.) oil, traditional use is stated for the indication of sleep disorders, temporary insomnia, mental stress, and mood disorders according to the current European Medicines Agency monograph [214]. Orange, lavender, lemon verbena, and rosemary are used to treat anxiety and insomnia, suggesting anxiolytic and sleep-promoting effects. The most abundant essential oil constituents in bitter orange (Citrus aurantium L. ssp. aurantium) flowers are limonene and linalool. D-Limonene shows antidepressant-like effects by influencing the neuroendocrine, neurotrophic, and monoaminergic systems [215]. In a cultivation-based in vitro study with 12 gut bacterial species, the essential oils of lavender and orange blossom showed preferential inhibitory activity against potentially pathogenic gut microorganisms while having a reduced impact on gut microbes regarded as beneficial [88].
In addition to essential oils, ethyl acetate extracts from C. aurantium blossoms contain flavanone glycosides, such as hesperidin, naringin, and neohesperidin. An in vivo study in HFD-fed mice performed with flavonoid-rich extracts indicated a reversal of the HFDinduced gut microbiota imbalance. In particular, the relative abundance of Bifidobacterium was increased, and the Firmicutes/Bacteroidetes ratio was significantly decreased [87].
Lemon verbena (Aloysia citriodora Paláu) ethanolic extracts contain polyphenols, iridoids, and flavonoids that contribute to their biological effects. In a study with HFD-fed mice, a lemon verbena ethanolic extract reduced intestinal dysbiosis, decreased the Firmicutes/Bacteroidetes ratio, and increased Akkermansia abundance in comparison with untreated HFD-fed mice [62]. The biological activities of rosemary (Salvia rosmarinus Spenn.) are, on the one hand, related to its volatile constituents and, on the other hand, to phenolic compounds such as the phenolic diterpenes carnosol and carnosic acid, and the phenylpropane derivative rosmarinic acid. Guo et al. (2018) found that supplementation with a rosemary extract containing 60% carnosic acid reduced depression-like behaviors alongside gut microbiota dysbiosis and inflammatory reactions in the hippocampus and serum of chronic restraint stress mice. The microbiome was rebalanced by significantly increasing the abundance of Firmicutes and Lactobacillus spp., and by significantly decreasing the abundance of Bacteroidetes and Proteobacteria. The extract exerted an antidepressive effect by suppressing the hippocampal expression of IL-1β, TNF-α, and NF-κB, thus inactivating inflammatory reactions in the hippocampus and microglia. The extract also promoted BDNF and p-AKT/AKT expression in the hippocampus [42].
Two weeks of treatment with an aqueous extract of powdered Melissa officinalis yielded an increased microbial Chao-1 diversity index in obese mice. These modifications were associated with higher cecal levels of butyrate, propionate, and ethanol [140].
The rhizome of turmeric (Curcuma longa L.) contains volatile oil rich in sesquiterpenes, polysaccharides, and yellow compounds called curcuminoids that have a dicinnamoylmethane skeleton. Petersen et al. studied turmeric powder in an in vitro anaerobic incubation with human fecal microbiota and observed potential prebiotic effects mainly based on the use of the polysaccharides in the herbal material [96]. In an animal study from 1986, colony counts of total aerobes were decreased in rats fed with turmeric, and counts of total anaerobes were increased after 3 months of application [97].
Curcumin has been shown to be metabolized by human fecal bacteria by demethylation, reduction, and hydroxylation reactions [216]. One of these metabolites, di-Odemethylcurcumin, has shown potential neuroprotective effects by attenuating LPS-induced inflammation in rat microglial cells. The metabolite was twofold more active than its parent compound curcumin [217], indicating that curcumin metabolites may have beneficial effects in mental health provided that they are able to pass the BBB.
Many of these findings indicate that it is not the essential oil but rather more polar constituents that are responsible for the interaction with gut microbiota, such as the phenolic diterpene carnosic acid in the case of rosemary, or polysaccharides and curcuminoids in the case of turmeric. This may be because essential oil constituents have a low molecular weight and are rather lipophilic, making them more likely to be absorbed in the upper intestine [218]. Therefore, they are less likely to come into contact with the gut microbiota. Hence, the pronounced mental health-promoting effects of volatile oils [219] may arise via routes other than the MGBA.

Herbal Drugs Containing Other Terpenoids
Extracts of Ginkgo biloba L. leaves are used worldwide in a standardized form, containing diterpene lactones (ginkgolides A, B, C, J), the sesquiterpene lactone bilobalide, flavonoids (mainly as glycosides), and polysaccharides. They are applied to neurological disorders connected to impaired cognitive functions and have been considered for anxiety and depression. In an in vitro study with rat intestinal bacteria, the time course of biotrans-formation of those constituents notably differed among diabetic rats, diabetic nephropathy rats, and healthy rats [107]. The composition and function of gut microbiota can change in response to diseases. If plant constituents are biotransformed by gut microbiota in vivo, their metabolism and absorption in the digestive tract may change with disease-induced changes in the microbial community composition and function. These alterations may, in turn, modulate the systemic effects of these compounds.
To study possible antidepressant mechanisms of G. biloba, the efficacy of a polysaccharide fraction from a leaf extract on the gut microbiome composition and depressive symptoms in mice was investigated. Compared with the untreated control group, the extract reduced stress-induced depression and mitigated gut dysbiosis, leading to an enhanced richness of Lactobacillus. Oral administration of L. reuteri or FMT by oral gavage from ginkgo-treated mice into depressive mice also significantly decreased the immobility time in the forced swimming test. These findings indicate that gut microbiome modulation by G. biloba polysaccharides can lead to reduced depressive symptoms, possibly via the MGBA [106].
Saffron (Crocus sativus L.) is also used in anxiety, mood disorders, and mild depression, with a considerable number of randomized controlled human clinical trials supporting its application [89][90][91][92][93]. Saffron contains four main bioactive carotenoids: crocin, crocetin, picrocrocin, and safranal, with a lipophilic character that makes them readily absorbable in the upper intestine. Crocin is rapidly hydrolyzed by enzymes in the intestinal epithelium and, to a lesser extent, by gut microbiota, resulting in deglycosylated trans-crocetin that is absorbed via the gut mucosa. trans-Crocetin is the only saffron metabolite that can cross the BBB and reach the CNS. A pilot study evaluating the effects of saffron on the gut microbiome composition in rats found a strong decrease in the Cyanobacteria and Proteobacteria phyla, and a less dramatic reduction in the Bacteroidetes and Firmicutes phyla [94].
Overall, extracts from G. biloba and C. sativus mitigated gut dysbiosis and enhanced Lactobacillus species compared with untreated control groups in animal studies. The study on C. sativus was rather preliminary and performed with healthy rats; the investigation of G. biloba was performed with rats that had stress-induced depression behaviors. Because that study was performed with a gingko fraction containing mainly polysaccharides, obviously the polysaccharides and not the terpenes or flavonoids were responsible for the apparent diminution of depressive signs.

Herbal Drugs Rich in Phenolic Constituents
Polyphenols are a broad group of phytochemicals made up of hydroxylated phenyl moieties and present in medicinal plants, tea, fruits, and cereals [27]. The polyphenolic compounds reviewed here belong to three groups: lignans (phenylpropane derivatives), flavonoids (flavan-3-ols, flavanones, flavones, flavone-3-ols, anthocyanidins, and isoflavones), and tannins (derivatives of catechin or gallic acid). Polyphenol esters, glycosides, or polymers are not usually absorbed in the small intestine, and interaction between gut microbiota and dietary polyphenols has often been reported. The gut microbiota can metabolize polyphenols, resulting in the production of potentially active metabolites that can reach the systemic circulation and, in some cases, cross the BBB and exert biological activities [220]. Moreover, polyphenols can alter the gut microbiome composition and function by increasing the population of healthy gut bacteria and decreasing the growth of pathogens, producing a prebiotic-like effect [220].

Herbal Drugs Containing Lignans
Recent clinical trials provide evidence for the use of Schisandra chinensis (Turcz.) Baill. and Eleutherococcus senticosus (Rupr. & Maxim.) Maxim. in mental (anxiety, depression) and behavioral disorders, including cognitive function, memory, and attention [170]. Most schisandra lignans have a dibenzocyclooctadien skeleton, whereas E. senticosus roots contain a mixture of the lignans eleutherosides B4, D, and E, together with phenylpropanoids.
In vivo studies indicate that some isolated constituents such as the lignans schisandrin B and eleutheroside E and the phenylpropanoid eleutheroside B contribute to the activity of the total extracts [103].
Studies in rat models have revealed that schisandrin B, the most abundant S. chinensis fruit lignan, can cross the BBB thanks to its lipophilic properties and low molecular weight [221]. Apart from lignans, schisandra fruits contain essential oil and polysaccharides. As reported by Yan et al. [173], the total extract and lignans alleviated depressive and anxiety symptoms, whereas the essential oil and polysaccharides ameliorated cognitive decline in lipopolysaccharide (LPS)-induced C57BL/6 mice. These authors also assessed the influence of schisandra total extract, lignans, polysaccharides, and essential oils on the microbiota-gut-brain axis. The total extract (95% ethanol) and the lignan fraction ameliorated depressive-like behaviors by restoring the altered intestinal microbiota composition, enhancing propionate and butyrate concentrations, and exerting anti-inflammatory effects via inhibition of the Toll-like receptor 4/NF-κB/IκB kinase α signaling pathway [173].
Lignans are also the main substances in raw S. chinensis fruits and a fruit wine prepared from these fruits, which exerted anxiolytic and antidepressive activities and modulated gut bacterial phylotypes in rats subjected to the chronic unpredictable stress procedure (CUSP). Long-term administration (35 days) restored gut microbial ecosystem dysbiosis occurring in CUSP rats. Of interest, the study authors observed improved cerebral ischemia, enhanced cerebral blood flow, and attenuated hippocampal neuritis after treatment with raw S. chinensis fruits and S. chinensis fruit wine. Hippocampal neurogenesis is involved in memory and learning, and disrupted neurogenesis is implicated in cognitive impairment and mood disorders, including anxiety and depression [172].
Su et al. also investigated the effect of a S. chinensis polysaccharide extract on the composition and diversity of the gut microbiome in mice. The polysaccharides had beneficial effects in mice with ulcerative colitis by recovering the gut microbial profile and increasing SCFA production [175].
In a randomized, double-blind clinical trial with 28 obese women, fecal microbiota community changes after the administration of an aqueous S. chinensis fruit extract were found to be different for each participant. This result indicated that S. chinensis affected the gut microbiome, but in different ways, depending on the pretreatment gut microbiome composition [174]. Overall, the data suggest that lignans are the most effective fraction of S. chinensis in the relief of depressive and anxiety disorders. Their activity may, at least in part, be related to the bidirectional connection between the gut microbiome and the brain. Furthermore, polysaccharide-rich S. chinensis extracts were able to reduce the abundance of potentially harmful bacteria through the production of SCFAs and regulate intestinal homeostasis.

Herbal Drugs Containing Flavonoids
Clinical studies indicate that flavonoid consumption may ameliorate mental disorders such as depressive symptoms [222], but the mechanisms involved in these effects have not been fully elucidated. Some flavonoids are orally bioavailable and pass the BBB, and certain flavonoid groups show binding affinity for the benzodiazepine site on the GABA A receptor and inhibit monoaminoxidases A and B [223]. Moreover, flavonoids act as antioxidant agents because of their hydrogen-donating ability, which may ultimately result in neuroprotection [224]. However, a high proportion of flavonoids are not absorbed in the upper intestine and therefore potentially interact with the gut microbiome. These compounds may possess prebiotic effects, since gut bacteria have been reported to be capable of utilizing them [225].
Glycine max L. (soy), a medicinal and food plant rich in isoflavones, has shown beneficial effects on mental health in menopausal women [108,109]. An in vivo study in mice showed that feeding with an HFD alone decreased SCFA levels, but this effect was compensated by soy addition. This was accompanied by enhanced relative abundances of Bacteroidetes, which mainly produce acetate and propionate [112]. A study in dogs re-vealed that soybean husk significantly increased levels of microbial fermentation products such as the SCFAs acetate and butyrate, as well as lactate. In addition, increased abundances of health-beneficial bacteria have been observed in vitro and in vivo [110]. In a rat model of menopause, soy supplementation reduced the Firmicutes/Bacteroidetes ratio and improved cardiometabolic health [113].
Isoflavone glycosides undergo hydrolysis in the upper GI tract and are only partially absorbed. In the colon, unabsorbed isoflavones are decomposed to smaller metabolites, i.e., aglycones and their decomposition products that are formed by reactions such as hydroxylation, hydrogenation, dehydroxylation, and C-ring cleavage [226]. Individual differences in the gut microbiome composition may influence the metabolism of isoflavone aglycones; for example, depending on the gut microbiome composition, daidzein can be further biotransformed either to O-desmethylangolensin or to S-equol, two metabolites with distinct pharmacological activities [227]. Gut microbial isoflavone metabolites may have an impact on mental health. In a placebo-controlled clinical trial in perimenopausal/postmenopausal Japanese women evaluating the effect of pure S-equol supplementation on mood-related menopausal symptoms, the pretreatment anxiety scores of equol producers were lower than those of non-producers, and S-equol supplementation improved mood-related symptoms in equol non-producers [228]. In mice, the microbial daidzein metabolite 6,7,4trihydroxyisoflavone improved scopolamine-induced cognitive impairment and enhanced learning memory, possibly by enhancing the expression of BDNF and the phosphorylation of cAMP response element binding, and by reducing acetylcholinesterase and malondialdehyde in the hippocampus [229]. These findings indicate that gut microbial isoflavone metabolites can exert beneficial effects on mental health.
In addition to isoflavones, soybean contains saponins such as soyasaponin I, which has been shown to ameliorate scopolamine-induced memory impairment in mice with intact gut microbiota, although it did not show significant effects in antibiotic-treated animals. Pre-fermentation with the bacterial strain Lactobacillus pentosus var. plantarum C29 further increased the effect, most likely because the strain can effectively biotransform glycosidic isoflavones and saponins into their more absorbable aglycones [204].
The female inflorescences of Humulus lupulus L. (hop) are used as herbal medicinal products for anxiety, mood disorders, and sleep disturbances. Hop contains a mixture of the flavonoids xanthohumol, isoxanthohumol, and 8-prenylnaringenin. These compounds have the potential to modulate and to be metabolized by the gut microbiota [124]. Furthermore, hop extracts comprise primary antimicrobial prenylated phloroglucinol derivatives such as humulones and lupulones. In an in vitro fermentation experiment with a human fecal suspension, a hop extract rich in humulone and lupulone altered the microbial community structure by favoring the growth of Enterobacteriaceae and inhibiting probiotic Bifidobacteria and butyrate-producing Eubacterium, and reduced butyrate levels. These effects were observed at high hop extract concentrations (final concentration 100-5000 µg/mL), which may be considered nonphysiological [123].
A Morus alba L. (mulberry) leaf extract significantly improved working memory and cognitive function in a clinical trial [136]. In an in vivo animal study, changes in the gut microbiome were observed in HFD-induced obese mice. Mulberry leaves partially reversed the microbiome shifts caused by the HFD, significantly increasing the Bacteroidetes/Firmicutes ratio. Additionally, a relative increase in Akkermansia and a relative decrease in Proteobacteria were observed [137].
Much of the literature on the interaction between flavonoid-containing plants used for mental health and the gut microbiome focuses on grapes (fruits of Vitis vinifera L.). Grape peels and fruit pulp are rich in flavonoids and anthocyanins. Grapes or grape-derived products (e.g., raisins, pomace, extracts) are associated with improved cognitive performance, including attention, language, and memory, as well as calmness and mood [179][180][181]. Several in vitro and in vivo studies showed an influence of grape preparations on the intestinal microbiome, but with different and partly contradictory results.
Mandalari et al. studied in vitro the influence of raisins (dried fruits of Vitis vinifera) on the human gut microbiome. Bacterial plate counting showed an increase in Bifidobacterium and Lactobacillus, and 16S rRNA gene sequencing revealed a relative decrease in Bacteroidetes and Faecalibacterium prausnitzii, indicating the potential to promote the proliferation of beneficial bacteria [191]. In contrast, in a human study assessing the effect on the intestinal microbiome of daily raisin consumption for 2 weeks, a significant increase in the relative F. prausnitzii abundance was observed, with no consistent relative increase in Bifidobacterium. In addition, no significant changes were detected for the Bacteroidetes and Firmicutes phyla in this human study. More pronounced changes were detected after 1 week of raisin consumption rather than after 2 weeks, possibly because raisin ingestion has only short-term effects on the gut microbiome composition [192].
Chacar et al. evaluated the impact of long-term feeding with polyphenol-rich grape pomace extracts on rat intestinal microbiota and observed a potentially more health-beneficial gut microbiome composition in aged rats after 14 months of treatment compared to a control group and young rats [194]. Another study that examined changes in the rat gut microbiome after consumption of polyphenol-rich grape antioxidant dietary fiber (GADF) showed a significant increase in the abundance of Lactobacillus spp. [195]. Feeding pigs a diet containing grape seed and grape marc meal extract, a polyphenol-rich byproduct of wine or juice processing, resulted in a reduction in Streptococcus abundance and total SCFA levels [196].
Three studies examining the effects of grapes on HFD-induced obesity and gut microbiota in mice showed that ingestion of grape fruit extracts could partially restore the disruption of the intestinal microbiome composition and mitigate many of the adverse health consequences caused by the HFD, such as reduced microbial alpha diversity. Grape administration also influenced the levels of several bacterial families and genera including Akkermansia, Bifidobacterium, Lachnospiraceae, Ruminococcus, and Bacteroidetes [197][198][199]. On the other hand, one study of grape pomace supplementation in healthy women found no changes in the gut microbiome composition. However, a significant increase in SCFAs was observed, likely because of the degradation of fibers or phenolic compounds in the extract. No significant changes were detected in the concentrations of phenolic metabolites, and large inter-individual variations were observed. 3-(4 -Hydroxyphenyl)-propionic acid was the only phenolic compound that clearly increased in the feces of two volunteers after grape pomace supplementation. In the urine, no differences were observed, and plasma samples were not analyzed [200].
The gut microbial metabolites of flavonoids may contribute to the mental health-related activities of medicinal plants. For example, the flavonol metabolites 4-hydroxyphenylacetic acid and 3,4-dihydroxyphenylacetic acid have shown anxiolytic activity in rats after oral and intraperitoneal application, while their progenitor flavonoids kaempferol, myricetin, and quercetin only displayed anxiolytic effects when administered orally, indicating that their gut microbial metabolization is required for activity [230]. The mechanism of anxiolytic action of these metabolites is still unclear, since 3,4-dihydroxyphenylacetic acid has been shown to be unable to cross the intestinal and blood-brain barriers in vitro, and to be rapidly eliminated from plasma in rats [231,232].
In summary, data on the influence of flavonoid-containing, mental health-related medicinal plants on the gut microbiome composition are heterogeneous. Generally, flavonoids are naturally produced by plants to deter bacterial infection and thus likely possess a certain antimicrobial potential towards gut microorganisms. Prenylated hop phloroglucinol derivatives reduced the relative abundances of certain beneficial bacterial genera at high concentrations, whereas isoflavones increased their levels. It is also reported that flavonoids beneficially impact the gut microbial community by increasing the relative abundance of known equol-producing bacteria such as lactobacilli [113]. The highest number of microbiome studies was retrieved for grape extracts and grape products. A large number of intestinal bacterial species were found to be influenced by grape preparations, but the results concerning gut microbiome changes are highly divergent. This may be because of the wide variety of different grape preparations used in the studies and the different experimental platforms for studying the interactions between grapes and the gut microbiome. In summary, studies of the interaction between the gut microbiome and flavonoid-rich grape preparations showed either no significant influence or prebiotic-like effects with no adverse impact on the gut microbiome.
Overall, most studies retrieved on flavonoid-rich, mental health-related medicinal plants were focused on their effects on gut microbiota, while the potential impact of microbial flavonoid metabolites on targets related to the MGBA remained widely unconsidered and deserves a more systematic assessment in the future.

Herbal Drugs Containing Tannins
As already mentioned in Section 3.2.2, grape preparations have positive effects on mental health. While flavonoids and anthocyanins are more abundant in grape peels, grape seeds contain large amounts of condensed tannins.
In an in vitro study with human fecal inoculum, incubation with grape seed polyphenols resulted in a significant increase in potentially beneficial bacteria such as Bifidobacterium spp. and Lactobacillus-Enterococcus groups, while the abundances of Bacteroides-Prevotella and Clostridium histolyticum groups decreased [183]. In contrast, fermentation of grape seed polyphenols in the colonic phase of the GI simulator SHIME, harboring a reproducible human microbial community, led to a general inhibition of the growth of all tested bacterial groups. This inhibition was ascribed to substrate limitation during batch incubation and to a certain antimicrobial capacity that had been previously shown for the applied grape extract [182]. In an in vitro fermentation study, a large proportion of grape seed constituents were found to be indigestible. During in vitro bacterial fermentation with rat cecal inoculum, dietary fibers and proteins were partially degraded, while 97% of the extractable polyphenols were metabolized, leading to the production of SCFAs. Metabolites of the extractable polyphenols were not analyzed in this study [233].
In rats, intake of polymeric grape seed tannins significantly increased the production of SCFAs, whereas the cecal pH and activity of various bacterial enzymes were decreased [184].
Yamakoshi et al. evaluated the effects of a procyanidin-rich grape seed extract on healthy adults after a 2-week administration (0.5 g/day). Culture-based plate counting indicated a significant increase in Bifidobacterium and a tendency to decrease for Enterobacteriaceae compared with pretreatment levels [190].
Feeding two doses of grape seed extracts to mice in combination with an HFD showed that grape seed administration could reduce HFD-induced changes in gut microbiota and improve glucose tolerance. Of interest, the lower applied dose seemed to be more effective than the higher one [189]. In ovariectomized mice, administration of a grape seed extract led to an increase in Bacteroidetes and a decrease in Firmicutes, normalizing the Firmicutes/Bacteroidetes ratio [188].
Two studies in pigs investigated the effects of ingesting grape seed meal, the residual from grape seeds after screw pressing the oil. Grosu et al. found that in healthy pigs, the additive increased the relative abundances of Bacteroidetes, Proteobacteria, and Prevotella and decreased the relative abundances of Firmicutes, Lachnospiraceae, and Lactobacillus [186]. In pigs with DSS-induced colitis, grape seed meal intake attenuated a DSS-induced Roseburia increase while stimulating the growth of Anaerovibrio and Megasphaera and butyric acid production [187].
Choy et al. examined the effects of grape seed extract ingestion on tannin metabolite production and gut microbiota in healthy pigs. The phenolic metabolites detected in feces included hydroxyphenylacetic acids, hydroxyphenylpropionic acids, hydroxyphenylvaleric acids, hydroxybenzoic acids, and caffeic acid. 4-Hydroxyphenylvaleric acid and 3-hydroxybenzoic acid were detected as major phenolic metabolites that increased during grape seed intake compared with baseline [185]. This finding is in line with the results from a study by Sánchez-Pátan et al. with a reproducible human gut microbial community in an in vitro simulator of the human GI tract [182].
Apart from their high levels of lipids, proteins, and dietary fiber, almonds (the seeds of Amygdalus communis L.) contain considerable amounts of polyphenols. The most abundant classes are condensed and hydrolyzable tannins (gallotannins, ellagitannins) and flavonoids that are readily metabolized by the human gut microbiota [201]. A randomized controlled trial showed that almonds could ameliorate post-lunch memory decline [63].
An almond-based low-carbohydrate diet significantly improved depression in patients with type 2 diabetes mellitus and induced a significant increase in the growth of SCFAproducing bacterial genera [68]. Psichas et al. reported that SCFAs in combination with free fatty acid receptor 2 can promote the secretion of glucagon-like peptide 1 [234], which is thought to influence depression and anxiety associated with metabolic dysfunction [15]. This finding suggests that the antidepressant effect of almonds may be associated with an increased abundance of SCFA-producing bacteria in the GI tract.
Three other human studies investigated the effect of almond consumption on the gut microbiome, yielding divergent results. Almond snacking for 8 weeks decreased the relative abundance of the opportunistic pathogen Bacteroides fragilis in young adults [67]. In another study, the intake of almonds for 18 days led to a decrease in lactic acid bacteria in adults, with no change in the abundance of Bifidobacteria [69]. Holscher et al. reported that the degree of almond processing, such as chopping, roasting, and grinding into butter, differently affected the gut microbiome composition [64].
In vitro fermentation of blanched finely ground almonds and blanched defatted finely ground almonds with human feces led to the conclusion that defatted almonds did not alter the composition of gut microbiota, whereas finely ground almonds stimulated the growth of Bifidobacteria and Eubacterium rectale [65]. Similar changes in the gut microbiota with natural and blanched almond skins were found in an in vitro GI digestion and fermentation model with human feces. Almond skins contain polyphenols and high amounts of dietary fiber, with higher polyphenol concentrations in natural than in blanched skins. Therefore, the authors concluded that the dietary fiber present in almond skin rather than polyphenols is responsible for their prebiotic effects [66].
Green tea, prepared from unfermented leaves of Camellia sinensis (L.) Kuntze, has a long history of use and is consumed all over the world. Thus, numerous studies have explored the beneficial effects of green tea and green tea extracts, including the modulation of cognitive function and mood in humans, reduced anxiety, improved attention, and cognitive impairment prevention [72,73]. Compounds active in mental health that are found in unfermented green tea leaves are mainly methylxanthines (caffeine), amino acids (Ltheanine), and flavan-3-ols (main compound: epigallocatechin-3-O-gallate, EGCG). EGCG possesses calming effects and relieves stress, whereas L-theanine, especially in combination with caffeine, improves attention and reduces fatigue [72]. High amounts of EGCG and other tea polyphenols are absorbed in the small intestine and undergo metabolism in different organs. The unabsorbed proportion is metabolized by colon microbiota and affects the community composition, inducing potential health-promoting effects due to gut microbiome shifts regarded as beneficial [235].
In four animal studies, changes in the gut microbiome of mice after the administration of green tea leaves or green tea extracts were detected. An aqueous green tea extract partly reversed the HFD-induced changes in the microbial community in mice at the genus and family levels. In addition, it increased total fecal SCFAs, in particular propionic acid and valeric acid [74]. Powdered leaves of purple-leaf tea, a new cultivar of C. sinensis with purple leaves, also mitigated the negative effects of an HFD on the murine gut microbiome [75].
In a murine model of chemical-induced colitis, feeding the animals green tea extracts resulted in positive effects on colitis-related signs such as tissue damage and colonic inflammation, and on gut microbiome dysbiosis [76,77]. In addition, the levels of fecal acetic, propionic, and butyric acids were significantly enhanced in one of the studies [76]. In the other study, FMT from green tea-treated to untreated mice also reduced colitis-induced inflammation and tissue damage and mitigated dysbiosis [77].
Additionally, the seeds from Paullinia cupana Kunth (guarana) contain tannins and methylxanthines as active compounds. Two double-blind and placebo-controlled studies confirmed the positive effects of standardized guarana seed extracts on mental health due to an improvement in cognitive performance in healthy participants, and on fatigue in breast cancer patients [154,155]. In two animal studies, guarana seed administration was associated with changes in the rat gut microbiome. The findings in one of these studies suggested that ingestion of guarana seed powder for 3 weeks affected the rat gut microbiome in a negative way, increasing the relative abundance of Cyanobacteria and decreasing the relative abundance of Lactobacillus and Bacteroidetes, with no impact on microbial diversity. This outcome was attributed to the possible antimicrobial effects of caffeine and other constituents [156]. The authors of the second animal study concluded that guarana administration together with an HFD did not induce considerable changes in the rat gut microbiome [157].
Although this aspect has not been thoroughly investigated in the studies reviewed herein, gut microorganisms are generally known to metabolize flavan-3-ols and condensed tannins from different herbal sources. Therefore, it can be assumed that the flavan-3-ols occurring at high levels in grapes, almonds, and green tea are also degraded by gut microorganisms. Oligo-and polymeric procyanidins are first decomposed to flavan-3-ol monomers, which are degraded by C-ring fission and dehydroxylation steps to dihydroxyphenyl-and hydroxyphenyl-γ-valerolactones and hydroxyphenylvaleric acids. These can be further metabolized to smaller phenolic acids that are also formed during gut microbial metabolism of flavonoids (Section 3.2.2) [236].
To date, there is only a low number of studies assessing the pharmacological effects of phenyl-γ-valerolactones and phenylvaleric acids available in the literature [237]. The study by Unno et al. indicated good BBB permeability for 5-(3,5-dihydroxyphenyl)-γvalerolactone, the major gut microbial EGCG metabolite in rats. Moreover, this metabolite increased the number of neurites and neurite length in SH-SY5Y neuroblastoma cells, indicating that the compound may promote neurogenesis in the brain [238].
Additionally, hydrolyzable tannins that occur at higher levels in almond skins are known to be metabolized by gut microbiota from studies performed in other tannincontaining plants. Meanwhile, it is well known that ellagitannins are decomposed to ellagic acid and further to urolithins by gut microbiota, with different metabotypes that are capable of producing differing urolithin patterns [239]. Urolithins have been predicted in silico to pass the BBB [240], and they have shown the potential to exert neuroprotective effects mainly in cellular models, but their possible beneficial effects related to mental health still need to be studied systematically [241].
In summary, condensed tannins present in grape seeds can induce changes in the gut microbiome and mitigate gut microbial dysbiosis. However, several studies have shown diverging results regarding changes in the gut microbiome composition, including increased as well as decreased abundances of Lactobacillus. Changes in the gut microbiome upon almond intake include an increase in beneficial bacteria such as Bifidobacteria and a decrease in Bacteroidetes, while an antidepressant effect may be related to an increased abundance of SCFA-producing bacteria, since SCFAs stimulate the secretion of the antidepressant glucagone-like-peptide-1. Several animal studies suggest an improvement in microbial dysbiosis and growth promotion of beneficial bacteria by green tea leaves. It remains unclear whether these changes are caused by the methylxanthines or the catechins. Guarana seed intake, on the other hand, did not lead to beneficial effects on the gut microbiome in two animal studies. This may be attributed to the antimicrobial effects of caffeine, but also to the tannins, which possess widely described antimicrobial effects [242]. The role of gut microbial tannin metabolites in mental health-related disorders has not been systematically studied to date.

Herbal Drugs Containing Other Phenolic Compounds
A medicinal herb commonly used to treat depression is Hypericum perforatum L. (St. John's wort). Numerous studies support the role of this plant in the treatment of mild to moderate depression because it has shown comparable efficacy, fewer side effects, and a lower risk of discontinuation when compared with selective serotonin reuptake inhibitors [125]. The plant contains a number of compound classes potentially involved in its antidepressant effects such as hyperforins, polyphenols (including flavonoids such as hyperoside), naphthodianthrones (hypericin), and procyanidins [243]. In a recent animal study, the effects of H. perforatum on the gut microbial community composition were investigated in ovariectomized rats. Ingestion of a H. perforatum extract could reverse gut microbiome changes at the phylum level caused by ovariectomy-induced estrogen deficiency, and extract application mitigated the increase in the Firmicutes/Bacteroidetes ratio [126].
The roots of Rhodiola rosea L. are used as a traditional medicine for their positive mental health effects on anxiety, stress, fatigue, and depression, as shown by in vivo animal and human studies [165,166]. The main phenolic compounds in the roots of R. rosea are catechins, procyanidins, and phenylpropanoids (mainly derivatives of cinnamyl alcohols and salidroside) [168]. Labachyan et al. showed that treatment with R. rosea root extract could alter the gut microbiome composition in Drosophila melanogaster as the order Lactobacillales was significantly decreased and the genus Acetobacter was increased [167]. In an in vitro incubation study with human fecal slurry, cinnamylalcohol, tyrosol, and hydroquinone were identified as the main phenolic metabolites [168]. Tyrosol is able to penetrate the BBB and has shown potent neuroprotective and neuroregenerative activities in vitro and in animal studies [244], and hydroquinone has shown protective effects against transient focal cerebral ischemia in rats [245], indicating the neuroprotective potential of these gut microbial R. rosea metabolites.
In addition to the well-known administration of Cannabis sativa L. for chronic pain and chemotherapy-induced nausea and vomiting, multiple studies have shown an effect on secondary sleep disturbance, although with only moderate evidence [78]. The main active compounds in C. sativa are cannabinoids (tetrahydrocannabinol and cannabidiol). Activation of cannabinoid receptors, which are part of the endocannabinoid system, causes multiple changes in GI function including gut motility, gastric secretions, gut-brain signaling, and interactions with the intestinal microbiome, such as increased LPS release [246]. In an animal study, the effects of three cannabis extracts with different cannabinoid concentrations on the gut microbiota composition of mice fed a high-fat/cholesterol diet (HFCD) were examined. The HFCD group receiving a cannabidiol-rich cannabis extract was the only group in which the Bacteroidetes/Firmicutes ratio decreased compared with the control group receiving the HFCD only. The two other extracts, which were either rich in tetrahydrocannabinol or contained similar concentrations of cannabidiol and tetrahydrocannabinol, had no significant impact on the gut microbiome composition [79].
Overall, although numerous studies assessed the interaction of plants with phenolic compounds used for mental health and the gut microbiota, most of them were not designed to assess MGBA-related effects. For only a limited number of plants, such as Schisandra chinensis and Amygdalus communis, studies are available that indicate potential mediation of mental health-related effects via the MGBA. The impact of the reviewed polyphenolcontaining plants on the MGBA is not yet evident from the existing data. However, for many of these reviewed plants, general beneficial and prebiotic-like effects on the gut microbiome have been shown, including mitigation of microbial community imbalances in different animal models of HFD-induced obesity, colitis, and menopause, and the enrichment of potentially health-beneficial bacteria such as SCFA producers, leading to increased intestinal SCFA production. These effects could also be relevant for mental health.
As shown in other studies, the anti-inflammatory activity of polyphenols and the metabolites produced by the gut microbiome can reduce neuroinflammation [27]. Polyphenols and their metabolites can control multiple risk factors for depression (e.g., inflam-mation, neurotransmitter levels and their precursors, neuronal innervation) and could be beneficial in the prevention and management of different mental health disorders [27,220]. Moreover, in a limited number of studies, gut microbial polyphenol metabolites such as S-Equol, 6,7,4 -trihydroxyisoflavone, 3,4 dihydroxyphenylacetic acid, 4-hydroxyphenylacetic acid, 5-(3,5-dihydroxyphenyl)-γ-valerolactone, hydroquinone, and tyrosol have shown pharmacological effects related to mental health conditions. Figure 4 shows a schematic representation of the key mechanisms of the MGBA through which polyphenols and their microbial-derived metabolites could exert a favorable effect on mental health conditions. Polyphenols exert a prebiotic-like influence on the gut microbiota that may contribute to positive MGBA effects. Moreover, inactive polyphenols are metabolized by gut microbiota to bioavailable and bioactive metabolites [247][248][249]. These active metabolites can reach the systemic circulation by crossing the intestinal epithelium and enhance brain function by regulating pro-inflammatory mediators, the HPA axis, vagus nerve communication, neurotrophic factors, and serotonin levels. Some of them may also permeate the BBB. Moreover, polyphenols may exert antioxidant effects and lower enhanced reactive oxygen species levels in the brain [27,220,250]. In addition, they can stimulate SCFA production by the gut microbiota [251].  Polyphenols and their metabolites support the rebalancing of the altered gut microbiome during dysbiosis, and the metabolites can cross the intestinal epithelium and reach the systemic circulation and brain. These molecules may modulate gut-brain communication via neural (blue letters), immune (pink letters), and humoral/metabolic (green letters) pathways. Polyphenols and their metabolites can modulate vagus nerve communication, the HPA axis, pro-inflammatory mediators, neurotrophic factors, and serotonin levels, positively influencing brain functions. Polyphenols have antioxidant effects and can reduce ROS levels in brain disorders [27,220,250], and they can also stimulate gut microbiome production of SCFAs [251]. Furthermore, polyphenols and their metabolites may have local brain effects such as improved cerebrovascular blood flow and a reduction in neuroinflammation [252]. The major gut-brain mechanisms by which polyphenols may exert beneficial effects are indicated with red arrows ( activation/upregulation, inhibition/downregulation). BBB: blood-brain barrier; IDO: indolamine 2,3 dioxygenase; TDO: tryptophan 2,3-dioxygenase; QA: quinolinic acid; PPARγ: peroxisome proliferator-activated receptor gamma; AMPK: 5′AMP-acti- . Potential microbiome-gut-brain communication pathways modulated by polyphenols in mental disorders. Gut microorganisms metabolize polyphenols to potentially active metabolites. Polyphenols and their metabolites support the rebalancing of the altered gut microbiome during dysbiosis, and the metabolites can cross the intestinal epithelium and reach the systemic circulation and brain. These molecules may modulate gut-brain communication via neural (blue letters), immune (pink letters), and humoral/metabolic (green letters) pathways. Polyphenols and their metabolites can modulate vagus nerve communication, the HPA axis, pro-inflammatory mediators, neurotrophic factors, and serotonin levels, positively influencing brain functions. Polyphenols have antioxidant effects and can reduce ROS levels in brain disorders [27,220,250], and they can also stimulate gut microbiome production of SCFAs [251]. Furthermore, polyphenols and their metabolites may have local brain effects such as improved cerebrovascular blood flow and a reduction in neuroinflammation [252]. The major gut-brain mechanisms by which polyphenols may exert beneficial effects are indicated with red arrows ( www.mdpi.com/journal/nutrients unication pathways modulated by triterpenes in mental disorides) can alter gut-brain microbiome communication in impaired hy mental state. These beneficial effects are related to rebalancing g neural (blue letters), immune (pink letters), and humoral/metiterpene glycosides are metabolized by the gut microbiome into des into compound K). These active metabolites are more bioas. Ginsenosides and their metabolites promote neurotrophic facmediators and stress levels [207,[209][210][211]. The major gut-brain s have a beneficial effect are marked with red arrows ( activaownregulation  Figure 3. Potential gut-brain communication pathways modulated by triterpenes in m ders. Triterpenes (such as ginsenosides) can alter gut-brain microbiome communication brain function and promote a healthy mental state. These beneficial effects are related to the gut microbiome and influencing neural (blue letters), immune (pink letters), and hu abolic (green letters) pathways. Triterpene glycosides are metabolized by the gut micr active components (e.g., ginsenosides into compound K). These active metabolites are vailable than the native compounds. Ginsenosides and their metabolites promote neuro tors and reduce pro-inflammatory mediators and stress levels [207,[209][210][211]. The majo mechanisms by which ginsenosides have a beneficial effect are marked with red arrows tion/upregulation, inhibition/downregulation). TJPs: tight junction proteins; BDN rived neurotrophic factor; PI3K: phosphoinositol 3 phosphate; AKT: protein kinase B; leukin-1β; NF-κB: nuclear factor-κB; PYY: peptide YY; GLP1: glucagon-like peptide lates/promotes.

Essential Oils and Herbs Rich in Essential Oils
Thirteen in vitro and in vivo studies assessing the effect on the gut mi mental health-related essential oils (orange blossom oil, lavender oil) or herbal in essential oils (lemon balm leaf, rosemary, lemon verbena leaf, black cumi meric root) met the inclusion criteria. For lavender (Lavandula angustifolia Mill tional use is stated for the indication of sleep disorders, temporary insomn inhibition/downregulation). BBB: blood-brain barrier; IDO: indolamine 2,3 dioxygenase; TDO: tryptophan 2,3-dioxygenase; QA: quinolinic acid; PPARγ: peroxisome proliferator-activated receptor gamma; AMPK: 5 AMP-activated protein kinase; ROS: reactive oxygen species; TOR: target of rapamycin; ⊕: stimulates/promotes.

Herbal Drugs Rich in Polysaccharides
Dietary fibers are plant polysaccharides that are indigestible in the upper intestinal tract but that can be metabolized by intestinal microorganisms. These fibers and their microbiota-mediated metabolic end products, i.e., SCFAs, can modulate the gut microbiome composition [37]. Traditional medicines rich in polysaccharides that are used to promote mental health include the rhizomes of Dioscorea opposita (= D. oppositifolia L.; Chinese yam) and the fruits of Lycium barbarum L. (goji). A water-ethanol extract from Chinese yam significantly improved conditions such as fatigue, stress, depression, sleep, and calmness [98], while a standardized juice of L. barbarum fruits was associated with improved cognitive function, especially semantic fluency [134].
In two animal studies, the administration of Chinese yam significantly restored the disturbance in gut microbiota during or after antibiotic treatment. Zhang et al. assessed the effects of different concentrations of dried Chinese yam powder on antibiotic-treated mice. Ampicillin-induced dysbiosis was restored by ingestion of Chinese yam powder. A significant increase was observed in Bifidobacteria and Lactobacilli, as was a decrease in Enterococcus in the group receiving the highest concentration of Chinese yam [99]. Supplying rats with a Chinese yam water extract together with imipenem/cilastatin sodium increased the abundance of Lachnospiraceae, Ruminococcaceae, Clostridiales, and Firmicutes and decreased the abundance of Blautia, Prevotella, and Eisenbergiella compared with rats receiving only antibiotics [100]. These data indicate the good prebiotic effects of Chinese yam.
Kang et al. showed that goji berry ingestion was associated with considerable changes in the gut microbiota of IL-10-deficient mice, increasing the abundance of butyrate-producing bacteria. Furthermore, the growth of Bifidobacterium and the Firmicutes/Bacteroidetes ratio increased. Thus, goji berry demonstrated strong prebiotic effects [135].
As known from other herbal materials, plant-derived polysaccharides that are indigestible in the upper intestinal tract are metabolized by gut microbiota into SCFAs that can influence the gut-brain axis via three major pathways [36,253]. Via the neural pathway, SCFAs can reduce cortisol levels; via the immune pathway, they decrease the levels of inflammatory mediators and microglial activation; and via the humoral/metabolic pathway, they can exert beneficial effects on serotonin synthesis, neurotrophic factors, and various gut neuropeptides. Moreover, SCFAs may restore tight junctions in the leaky intestinal epithelium by increasing the expression of TJPs, and they can exert local beneficial actions on gut health, such as maintaining mucus protection [36,37,[253][254][255][256]. A detailed schematic representation of the various pathways describing the possible action of plant polysaccharides (dietary fibers) on the MGBA is presented in Figure 5.
Nutrients 2022, 14, x. https://doi.org/10.3390/xxxxx www.mdpi.com/journal/nu of inflammatory mediators and microglial activation; and via the humoral/meta pathway, they can exert beneficial effects on serotonin synthesis, neurotrophic factors various gut neuropeptides. Moreover, SCFAs may restore tight junctions in the leak testinal epithelium by increasing the expression of TJPs, and they can exert local bene actions on gut health, such as maintaining mucus protection [36,37,[253][254][255][256]. A det schematic representation of the various pathways describing the possible action of polysaccharides (dietary fibers) on the MGBA is presented in Figure 5. Figure 5. Potential microbiome-gut-brain communication pathways modulated by plant-derived polysaccharides in mental disorders. Gut microorganisms metabolize polysaccharides that resist digestion in the upper gastrointestinal tract into SCFAs. SCFAs modulate gut-brain communication via neural (blue letters), immune (pink letters), and humoral/metabolic (green letters) pathways. SC-FAs may reduce cortisol levels, inflammatory mediators, and microglial activation, have a beneficial effect on serotonin synthesis, neurotrophic factors, and various gut neuropeptides, and restore tight junctions in the leaky intestinal epithelium by increasing the expression of tight junction proteins (TJPs). In addition, SCFAs exert local beneficial actions that improve gut health (e.g., maintaining mucus production, anti-inflammatory effects) [37,[253][254][255][256]. The major gut-brain mechanisms by which SCFA/active polysaccharide metabolites offer benefit are marked with red arrows ( According to many preclinical studies, certain types of triterpenes possess anti-inflammatory, antioxidant, and antiapoptotic properties and thus may contribute to neuronal protection [206,213]. Moreover, triterpene glycosides can be metabolized by gut microbiota into better absorbable active metabolites that become systemically available [206][207][208]. Figure 3. Potential gut-brain communication pathways modulated by triterpenes in mental disorders. Triterpenes (such as ginsenosides) can alter gut-brain microbiome communication in impaired brain function and promote a healthy mental state. These beneficial effects are related to rebalancing the gut microbiome and influencing neural (blue letters), immune (pink letters), and humoral/metabolic (green letters) pathways. Triterpene glycosides are metabolized by the gut microbiome into active components (e.g., ginsenosides into compound K). These active metabolites are more bioavailable than the native compounds. Ginsenosides and their metabolites promote neurotrophic factors and reduce pro-inflammatory mediators and stress levels [207,[209][210][211]. The major gut-brain mechanisms by which ginsenosides have a beneficial effect are marked with red arrows ( activation/upregulation, inhibition/downregulation). TJPs: tight junction proteins; BDNF: brain-derived neurotrophic factor; PI3K: phosphoinositol 3 phosphate; AKT: protein kinase B; IL-1β: interleukin-1β; NF-κB: nuclear factor-κB; PYY: peptide YY; GLP1: glucagon-like peptide 1; : stimulates/promotes.

Essential Oils and Herbs Rich in Essential Oils
Thirteen in vitro and in vivo studies assessing the effect on the gut microbiota of mental health-related essential oils (orange blossom oil, lavender oil) or herbal drugs rich in essential oils (lemon balm leaf, rosemary, lemon verbena leaf, black cumin, and turmeric root) met the inclusion criteria. For lavender (Lavandula angustifolia Mill.) oil, traditional use is stated for the indication of sleep disorders, temporary insomnia, mental According to many preclinical studies, certain types of triterpenes possess anti-inflammatory, antioxidant, and antiapoptotic properties and thus may contribute to neuronal protection [206,213]. Moreover, triterpene glycosides can be metabolized by gut microbiota into better absorbable active metabolites that become systemically available [206][207][208]. Figure 3. Potential gut-brain communication pathways modulated by triterpenes in mental disorders. Triterpenes (such as ginsenosides) can alter gut-brain microbiome communication in impaired brain function and promote a healthy mental state. These beneficial effects are related to rebalancing the gut microbiome and influencing neural (blue letters), immune (pink letters), and humoral/metabolic (green letters) pathways. Triterpene glycosides are metabolized by the gut microbiome into active components (e.g., ginsenosides into compound K). These active metabolites are more bioavailable than the native compounds. Ginsenosides and their metabolites promote neurotrophic factors and reduce pro-inflammatory mediators and stress levels [207,[209][210][211]. The major gut-brain mechanisms by which ginsenosides have a beneficial effect are marked with red arrows ( activation/upregulation, inhibition/downregulation). TJPs: tight junction proteins; BDNF: brain-derived neurotrophic factor; PI3K: phosphoinositol 3 phosphate; AKT: protein kinase B; IL-1β: interleukin-1β; NF-κB: nuclear factor-κB; PYY: peptide YY; GLP1: glucagon-like peptide 1; : stimulates/promotes.

Essential Oils and Herbs Rich in Essential Oils
Thirteen in vitro and in vivo studies assessing the effect on the gut microbiota of mental health-related essential oils (orange blossom oil, lavender oil) or herbal drugs rich in essential oils (lemon balm leaf, rosemary, lemon verbena leaf, black cumin, and turmeric root) met the inclusion criteria. For lavender (Lavandula angustifolia Mill.) oil, traditional use is stated for the indication of sleep disorders, temporary insomnia, mental inhibition/downregulation). HDAC: histone deacetylases; GDNF: glial cell-derived neurotrophic factor; NGF: nerve growth factor. ⊕: stimulates/promotes.

Conclusions and Outlook
The MGBA is considered a significant therapeutic target for several mental disorders. Medicinal plants contain various classes of secondary plant metabolites, and many of them are poorly absorbed in the upper GI tract due their high polarity and molecular weight. Therefore, most likely, they interact with the gut microbiome and thereby potentially modulate the MGBA. In the present review, 30 medicinal plants showing effects on mental health-related disorders in clinical and animal studies were identified in reports that also showed their potential interaction with the gut microbiota. Overall, 85 in vitro and in vivo studies on this interaction were retrieved.
With a few exceptions, the studies were not designed to directly assess the impact of the respective herbal preparations on targets or pathways related to the MGBA. Nevertheless, they provide indications of a possible interaction with the MGBA, such as positively influencing dysbiotic microbiome conditions, increasing the abundance of health-beneficial or SCFA-producing bacterial species, or exerting anti-inflammatory effects, as in the case of Salvia rosmarinus, or because they are metabolized by gut microbiota into active metabolites that affect various MGBA-related pathways, as in the case of ginsenosides.
In some studies, the results indicate that the marker compounds commonly used for their standardization are not responsible for the interaction with the gut microbiome and that other compound classes are involved. For example, in the case of Ginkgo biloba, a polysaccharide but not the terpenes or flavonoids obviously exerted positive effects on depressive symptoms in a mouse model of unpredictable chronic mild stress, possibly via modulation of the gut microbiome.
The results of this review indicate that the two-way interaction between the gut microbiome and medicinal herbs could play a role in mediating their mental health effects. We propose that the plant constituents present in these herbs exert their neuroprotective effects through a multitarget effect on the host and the microbiome and can therefore be referred to as phyto-psychobiotics. Certain compound classes such as polyphenols and polysaccharides have been shown to have prebiotic effects. Terms such as flavobiotics and phytobiotics have been used to refer to phytochemical constituents conferring health benefits on the host by positively influencing the gut microbiome [257,258]. Furthermore, recently, it has been proposed that polyphenols act as duplibiotics, meaning that these phytoconstituents have a dual effect on the microbiome by exerting antimicrobial properties, similar to antibiotics, on one hand, and by acting as prebiotics, positively stimulating the growth of beneficial bacteria, on the other hand [259]. Moreover, some of the plant constituents can be metabolized by gut microbiota into pharmacologically active compounds and other postbiotics such as SCFAs, lactate, and phenolic metabolites that can either have a local effect in the gut or be absorbed by the epithelial cells and provide other health benefits to the host via different pathways including the MGBA. Many single plant constituents have been tested for their neuroprotective effects in in vitro and in vivo studies (reviewed elsewhere) [260,261]; however, studies directly assessing the synergistic effects of multiple phytochemical constituents in medicinal plants on MGBA-related targets or pathways are scarce or even non-existent for many candidate plants with clinically proven effects on mental health. Such studies are urgently needed to generate a better understanding of the possible effects of these plants on the MGBA. We recommend that future clinical studies assessing the effect of medicinal plants on mental health should include the analysis of the gut microbiome composition and function to explore the possible action of these medicinal plants on the MBGA. This would facilitate a better understanding of why some individuals respond to interventions while others might be non-responders as they may lack the microorganisms needed to help them metabolize specific plant constituents into active metabolites. Furthermore, combining in vitro GI models, which include both upper and lower GI tract simulation, with multi-omics approaches (e.g., metagenomics, metabolomics, metatranscriptomics, and metaproteomics) can be used as a first step to explore the complex bidirectional interaction between plant constituents and the gut microbiome. These approaches will provide insight into the mode of action and health benefits of herbal medicines, and they will support the identification of new active plant constituents and how they might act via the MGBA or confer additional health benefits on the host.