FERMENTED FOODS: HARNESSING THEIR POTENTIAL TO MODULATE THE MICROBIOTA-GUT-BRAIN FOR MENTAL HEALTH

Fermented foods: Harnessing their potential to modulate the microbiota-gut-brain axis for mental health

Highlights

• Fermented foodsoffer an affordable dietary intervention strategy.

• Fermented foods comprises of microbes and molecules with neuroactive potential.

• Food substrate and environmental conditions shape fermented food microbiome.

• Most preclinical and human studies lack accounting for appropriate controls.

• Long term intervention with accurate measurement of fermented food intake is critical.

Abstract

Over the past two decades, whole food supplementation strategies have been leveraged to target mental health. In addition, there has been increasing attention on the ability of gut microbes, so called psychobiotics, to positively impact behaviour though the microbiota-gut-brain axis. Fermented foods offer themselves as a combined whole food microbiota modulating intervention. Indeed, they contain potentially beneficial microbes, microbial metabolites and other bioactives, which are being harnessed to target the microbiota-gut-brain axis for positive benefits. This review highlights the diverse nature of fermented foods in terms of the raw materials used and type of fermentation employed, and summarises their potential to shape composition of the gut microbiota, the gut to brain communication pathways including the immune system and, ultimately, modulate the microbiota-gut-brain axis. Throughout, we identify knowledge gaps and challenges faced in designing human studies for investigating the mental health-promoting potential of individual fermented foods or components thereof. Importantly, we also suggest solutions that can advance understanding of the therapeutic merit of fermented foods to modulate the microbiota-gut-brain axis.

1. Introduction

“Long is the calm brain active in creation; Time, only, strengthens the fine fermentation”.

Johann Wolfgang von Goethe

Keeping the brain active in creation as referenced by Goethe relies on a variety of intrinsic and extrinsic signals. Over the past two decades, there has been a growing appreciation that the gut microbiota is a key mediator with respect to responding to external signals and triggering intrinsic functions within the body. The microbiota-gut-brain axis consists of a diverse microbial community contained in the intestinal environment that is in constant communication with the central nervous system, and vice versa. The gut microbial community is influenced by a variety of factors ranging from diet (Latorre-Pérez et al., 2021, Qin et al., 2022, Asnicar et al., 2021), age (Yatsunenko et al., 2012, Ghosh et al., 2022), medication use (Ghosh et al., 2022), sex (Cuesta-Zuluaga et al., 2019), ethnicity (Ang et al., 2021, Dwiyanto et al., 2021) and geographical location (Bai et al., 2022, Kabwe et al., 2020) and other factors, with diet playing a particularly important role in influencing the microbiota and the metabolites they produce. These microbes and their metabolites influence the hosts intestinal (Del Bo et al., 2021, Peron et al., 2021), immunological (Longhi et al., 2021, Wastyk et al., 2021, Bisgaard et al., 2016, Li et al., 2016) and neural components (Shi et al., 2021, Tessier et al., 2021, Ribeiro et al., 2022) of the microbiota-gut-brain axis. In parallel, meta-analyses examining the relationship between mental health and dietary patterns have hinted at the promise of dietary intervention strategies to influence the cognitive wellbeing of an individual (Abbott et al., 2019, Ocklenburg and Borawski, 2021, Taylor et al., 2021). Thus, as a consequence of a combination of these advances, there is a growing interest in targeting the microbiota via diet so as to confer mental health benefits to the host (Berding et al., 2021b). The majority of microbiota-targeted interventions involve probiotics (Long-Smith et al., 2020), prebiotics (Mörkl et al., 2020), fibres (Berding et al., 2021a), polyphenols (Rodríguez-Daza et al., 2021), fatty acids (Silva et al., 2020), or, more broadly via changes in habitual dietary consumption (Ghosh et al., 2020, Millman et al., 2021, Valls-Pedret et al., 2015). Whole food based dietary interventions are increasingly applied to study mental health (Berding et al., 2021a). These interventions are usually lifelong and are integrated into the habitual diet of the patient. Recently studies have investigated fermented foods as a potential avenue of microbiota-targeted intervention strategy (Marx et al., 2020). Although ancient in origin, fermented foods are now seen as conduits for introducing beneficial microbes and molecules. Moreover, fermented foods are applicable therapeutics across various socioeconomic sectors given their potential affordability and cross-cultural accessibility.

Food fermentation was traditionally employed to enable longer storage/shelf-life of food substrates that would otherwise spoil quickly, which was crucial in times of scarcity (Amato et al., 2021, Borremans et al., 2020, Ross et al., 2002), whilst concurrently enhancing flavour profile (Liu et al., 2019b, Cai et al., 2019, Mandha et al., 2022, Peraza and Perron, 2022), reducing the toxicity of raw materials/controlling pathogenic microorganisms and simultaneously allowing for their digestion (Reddy and Pierson, 1994, Maixner et al., 2021). Fermented foods are mainly classified into categories based on the substrate used, e.g., cereal, dairy, meat, fish, vegetable and legume (Tamang et al., 2020), which differ with respect to their primary food substrate and type of fermentation (e.g., defined starter culture, spontaneous or back-slopped culture). Table 1 in this review highlights the diverse nature of fermented foods that is referred to in this review along with the substrate category and the nature of fermentation employed in its production. The microbial community present in a fermented food is associated with a number of factors, including the type of substrate (Achi and Asamudo, 2019, Leech et al., 2020), geographical location (Van Reckem et al., 2019, Jung et al., 2018, Zhong et al., 2016, Li et al., 2017), pH (Yang et al., 2020) and method of preparation (Lee et al., 2021, Van Reckem et al., 2019) (See Fig. 1). Fermented foods are a rich source of beneficial microorganisms (potential probiotics) (Okada et al., 2018, Wang et al., 2022) as well as bioactive peptides (Chaudhary et al., 2021), phytochemicals and vitamins (Septembre-Malaterre et al., 2018, Shahbazi et al., 2021b). As researchers increasingly investigate the impact of different dietary intervention strategies and habitual dietary practices on sculpting the gut microbiota (Losasso et al., 2018, Tanes et al., 2021, Stege et al., 2022) and, consequently, their metabolites (Wu et al., 2016, Chen et al., 2022), it is not surprising that fermented foods receive particular attention. This is in no small way related to their capacity to modulate the composition and/or diversity of the gut microbiota (Bellikci-Koyu et al., 2019, Le Roy et al., 2022, Wastyk et al., 2021) and/or the production of microbial metabolites, such as short chain fatty acids (SCFA), polyphenolic (Johnson et al., 2019, Zorraquín-Peña et al., 2021), tryptophan and bile metabolites (Scott et al., 2020) and, as a result, can modulate the pathways that relay information from gut to the brain. Table 2 provides a primer on the nomenclature of various microbial components and bioactive components along with consensus statement that aptly captures their function. The frequent, yet incorrect, description of fermented food-associated microorganisms as probiotics in the literature and by industrial stakeholders has led the International Scientific Association for Probiotics and Prebiotics (ISAPP) to establish a consensus statement regarding terminology that is frequently used in microbiota based therapeutic strategies.

Table 1. Compilation of fermented foods that are frequently referred to in this review and the type of fermentation process most commonly employed in their production. The methods of fermentation might be subjected to variation in their production, geographical location, cultural practices and are not strictly confined the method specified in the table.

Fig. 1. Fermented foods are diverse in their preparation, substrate category and type of fermentation. This can influence the microbiota and the molecular composition of fermented foods. Subsequently, these components can influence the communication between the gut and the brain resulting in modulation of the intestinal barrier and blood brain barrier, peripheral and central immune system and nervous system and thereby modulating the intestinal mileu, general gut and brain health. Created with BioRender.com.

Table 2. Nuances of Nomenclature: This table serves as a guide to aid the understanding and usage of appropriate terminology.

Here, we review the components of fermented foods that can exert beneficial effects to the individual with a particular focus on mental health. We also summarise existing literature from preclinical and clinical research relating to the impact of fermented foods on individual components of the microbiota-gut-brain axis. Lastly, we discuss the current challenges associated with preclinical and human studies aimed at understanding the beneficial effects of fermented foods as a potential intervention strategy targeting mental health.

2. Fermented food and the microbiota-gut-brain axis

The microbiota-gut-brain axis facilitates a constant bidirectional relay of information from the intestine via the enteric nervous system (ENS) and from the intestinal milieu consisting of microbial communities, microbial metabolites, gut associated lymphoid tissue (GALT), peripheral immune cells and cytokines to the brain and vice versa via the sympathetic/parasympathetic nervous system, neurotransmitters and the circulatory immune system (Cryan et al., 2019). Fermented foods are rich in microbes and their metabolites. In addition to the phytochemicals, these metabolites can take the form of neurotransmitters, neuroactives and neuromodulators (Yu et al., 2020), and thereby stimulate the connecting pathways of the microbiota-gut-brain axis: immune, neuroendocrine, enteric nervous and circulatory system. Upon digestion, they can result in the production of microbial metabolites that can modulate the permeability of the intestinal barrier (Scott et al., 2020) and the blood brain barrier (BBB) (Stachulski et al., 2022, Angelino et al., 2019). Before we discuss the effects of fermented food supplementation on mental health, we outline the current evidence of the ability of fermented foods to modulate the individual components of the microbiota-gut-brain axis so as to exploit them for future microbiota-targeted

2.1. Fermented food and the immune system

Microbial metabolites are capable of regulating the immune system (Lavelle and Sokol, 2020) via systemic circulation (Colombo et al., 2021) or by the vagus nerve (Bluthe et al., 1996, Namgung et al., 2022, Huffman et al., 2019). Growing evidence on the presence of receptors for bacterial cell wall components and immunostimulants such as peptidoglycan and lipopolysaccharides in the brain depict the far reaching ability of the gut microbiota on the brain and behaviour of the host (Wheeler et al., 2023, Arentsen et al., 2017, Tillinger and Mravec, 2021). This microbiota-gut-immune-brain connection is influenced by different dietary and lifestyle choices across lifespan (Bostick et al., 2022, Ratsika et al., 2023). Fermented foods, a subset of dietary intervention strategies can therefore be leveraged to boost this bidirectional communication.

Several components of fermented foods are being individually studied for their ability to modulate the immune system. The bacterial cellular components of fermented foods have shown the ability to instigate the release of IL-10 from dendritic cells and CD4 + T cells (Kim et al., 2019). Other components, such as fermented food exopolysaccharides (EPS) are also being explored for their modulatory effects on the immune system, as seen with kefiran, an EPS product produced by Lactobacillus kefirofaciens (Bourrie et al., 2016). Microbial metabolites produced in fermented food also have immunomodulatory abilities. For example, oleamide, a microbial metabolite produced by Penicillium candidum, is present in products such as camembert cheese. Oleamide has been shown to suppress TNF-α release from microglia by acting as an agonist for the P2Y and cannabinoid receptors (Kita et al., 2019).

A potential mechanism by which fermented foods are able to exert immunomodulatory effects is through activation of the hydrocarboxylic acid receptor (HCA3R), as a consequence of consuming lactic acid bacteria (LAB) fermented food. Sauerkraut, also rich in LAB upon consumption is shown to elicit a chemotactic response from monocytes via D-Phenylacetic acid, a potent HCA3R receptor agonist (Peters et al., 2019). Large scale genome wide association analysis have revealed the LAB found in human gut likely to be from foods (Pasolli et al., 2020). Interestingly, a human observational study on the TwinsUK cohort showed elevation in B. animalis subsp. lactis with yogurt consumers that positively correlated with 3-hydroxyoctanoate levels, which is an agonist for HCA3 and could be potentially implicated in host immunoregulation (Le Roy et al., 2022). Further studies on the hydrocarboxilic receptor revealed only humans and great apes to possess three hydrocarboxylic receptors, as opposed to the two (HCA1R, HCA2R) seen in other organisms, indicating the development of evolutionary adaptations to accommodate the consumption of fermented foods (Peters et al., 2019).

2.2. Fermented food on inflamed barrier integrity

Microbial surface molecules and microbial metabolites are implicated in the integrity of the intestinal epithelial barrier (Spiljar et al., 2017, Liu et al., 2020b) and the blood brain barrier (Knox et al., 2022a). SCFA (Knox et al., 2022b), methylamines (Hoyles et al., 2021), bile acid metabolites (Lajczak-Mcginley et al., 2020) and amino acid metabolites (Scott et al., 2020, Stachulski et al., 2022) are some of the metabolites produced by the host microbiome that can influence the intestinal and BBB integrity of the host. Fermented foods are a rich source of microorganisms and on consumption can result in production of microbial metabolites that can elicit an immunological response, thereby influencing the integrity of these physiological barriers.

Several studies have examined the impact of fermented food products on intestinal integrity in the context of Inflammatory Bowel Disease (IBD). Murine models of IBD when supplemented with Bacillus subtilis fermented milk showed rescue in intestinal morphology as well as elevation in tight junction protein level as opposed to the disease control (Zhang et al., 2021b). This restoration of intestinal tight junction proteins has also been observed using in-vitro and murine models of IBD supplemented with fermented barley and soybean (Woo et al., 2016c). Studies have also measured rescue of intestinal integrity via measurement of bacterial translocation. As the microbial community resides within the gut, translocation can indicate a compromise to intestinal integrity. The translocation of microbes and their cellular components can activate the peripheral immune system resulting in the release of cytokines that can ultimately trigger altered BBB integrity and neuroinflammation (Maes et al., 2013, Vujkovic-Cvijin et al., 2022). Indeed, pre-treatment of mice with milk fermented using Lacticaseibacillus paracasei subsp. paracasei lowered intestinal permeability as visualised by the reduced translocation of Salmonella typhimurium (Acurcio et al., 2020). Although less frequently studied, downstream reinforcing effects of fermented food on BBB integrity was observed following kimchi and kefir supplementation, which increased expression of BBB tight junction proteins along with a sex-specific reduction of mRNA levels of IL-1β and TNFα in the pre-frontal cortex and hippocampus in murine models (Kim et al., 2022b, Murray et al., 2019).

2.3. Fermented food and central/peripheral inflammation

Beyond their influence on barriers, fermented foods are also studied with respect to their role on central/peripheral immunomodulation. Preclinical studies assessing the impact of fermented foods on the immune system have either studied effects on the baseline unperturbed immunological state of the host or when the host is subjected to an immune challenge that mimics a pathological condition. For instance with respect to the latter, Lactobacillus delbrueckii and Streptococcus thermophilus fermented yogurt supplementation in a recurrent model of IBD in murine models, decreased levels of TLR4 + and IL-17 along with elevation of IL-10 and TLR9 + cells in the colon (Chaves et al., 2011). Similar reductions in the levels of inflammatory cytokines, especially IFN-γ and TNF-α, were observed in the colon of mice receiving milk kefir post-infection with Giardia intestinalis (Franco et al., 2013). These effects could be due to the ability of some kefirs to upregulate mRNA levels of IL-17 and downregulate IL-6, TNF-α, and IFN-γ when subjected to immune challenge in mice (Acurcio et al., 2020). Moreover, repeated treatment with other milk kefirs in mice has been shown to increase levels of Treg and IL-10 cells coupled with an attenuation of the levated neutrophils and CXCL1 levels induced by the stress of repeated oral gavage (Van De Wouw et al., 2020). Further studies have highlighted that this potential peripheral immunomodulation observed in kefir could be attributed to IgA-stimulating Lactobacillus kefiri that reduces expression of pro-inflammatory cytokines in the mesenteric lymph nodes and increases anti-inflammatory cytokines, CXCL-1 and expression of mucin-6 genes. Other strains of the same species were also associated with a lowered expression of IL-6 in the ileum and colon post-LPS exposure (Carasi et al., 2015). It should be noted that studies focusing on a single strain isolated from fermented food aid in the discovery of potential probiotic strains and can give insights into the mechanisms via which a fermented food may confer a health benefit. However, these studies do not capture the broader bioactive potential of fermented foods that include more complex mixtures of bacterial strains, bioactive peptides, microbial metabolites, fibres and other components that can impart beneficial biological effects on the individual (Vieira et al., 2021).

Studies are increasingly shifting towards understanding the far-reaching effects of fermented foods on neuroinflammation. A spatial observation of the immunomodulatory effects of fermented foods on the brain revealed that hippocampal and cortical regions of the brain exhibit lower levels of mRNA expression pertaining to IL-6, TNF-α, TLR4 receptor and MCP1 protein, which was significantly upregulated in mice fed a high-fat diet (Kim et al., 2022b). This was also observed in the hypothalamic regions of the brain and serum in mice subjected to chronic unpredictable mild stress when supplemented with fermented rice germ (Batsukh et al., 2022). Future work is needed to explore the impact of fermented foods at each compartment of the microbiota-gut-brain-immune axis, particularly the effects of microbial metabolites on the intestinal barrier, peripheral immune system, BBB and central immune system. This information would provide valuable insight into understanding the molecular underpinnings of the microbiota-gut-brain-immune axis.

Importantly, some of the aforementioned findings relating to the immunomodulatory effects of fermented foods have translated to human studies, albeit only a few studies have also concurrently looked at the gut microbiome profile. For instance, in a study conducted with healthy volunteers, a 10-week fermented food-based intervention resulted in decreased circulating cytokine levels, especially IL-6, IL-10 and IL-12b, along with a concurrent reduction of activation proteins from 4 major immune cell types: CD4⁺, CD8⁺, T and B cells (Wastyk et al., 2021). The study also revealed a correlation between faecal butyrate and lower circulating B cells (Wastyk et al., 2021), mirroring the preclinical findings of an interplay between microbial metabolite and host immune status. Furthermore, a reduction in circulating cytokine levels such as IL-4, IL-10, IL-12 and MCP-1 was observed in IBS patients supplemented with kimchi for a period of 12 weeks (Kim et al., 2022a). This is however not replicated in other studies that have looked at the impact of fermented, unfermented and pickled vegetables for 6 weeks (Galena et al., 2022). The study showed no difference in their serum C – reactive protein (CRP) and TNF-α profile, when measured at the end of the intervention. Similarly, a fermented food enriched diet, when administered for a duration of 4 weeks, did not show any alteration in serum cytokines (Berding et al., 2023). Epidemiological studies have also revealed no associations between the consumption of fermented and unfermented dairy with circulating CRP levels (Voutilainen et al., 2022). It is therefore surprising that a recent meta-analysis on fermented dairy products such as yogurt, fermented milk and kefir revealed lowered CRP and elevated IFN-γ with no significant effect on IL-12, IL-10 and IL-6 (Zhang et al., 2023). It is clear that the effects of fermented foods on circulatory inflammatory cytokines cannot be generalised given the varied responses observed with intervention strategies. A potential reason for this could be due to the heterogeneity of food substrates and microorganisms present in the fermented food and duration of the dietary intervention. The immunological response observed is highly tailored to the fermented food that is being studied and could be influenced by the microbial community it hosts (Spindler et al., 2022) and is addressed in greater detail in the final section of the review. It is also to be acknowledged that sex plays a critical role in modulating microbiota-gut-brain-immune axis communication (Jaggar et al., 2020, Klein and Flanagan, 2016). Some studies have only looked at fermented food supplementation in male (Schoen et al., 2009, Bourrie et al., 2023) and female volunteers (Han et al., 2015, Galena et al., 2022). One study, however, examined sex differences on moderate beer consumption and observed that women had higher CD3⁺ counts than male volunteers, with both groups showing an overall reduction in the IFN-γ/IL-10 ratio when compared to their baseline abstinent phase (Romeo et al., 2007). However, there is a general dearth of human studies that explore sex-selective immunomodulatory responses associated with the consumption of fermented foods and its effects on the overall health of the individual.

2.4. Fermented food and the hypothalamic-pituitary adrenal (HPA) axis

The HPA axis is a key regulator of mood and behaviour and forms the neurohormonal component of the microbiota-gut-brain axis. Dysregulation of the HPA axis and its interplay with the immunological system is implicated in multiple neuropsychiatric disorders (Wingenfeld and Wolf, 2011, Rinne et al., 2002, Cruz-Pereira et al., 2020). Cortisol, the main stress hormone of the HPA axis, has been shown to modulate the immune system (Bellavance and Rivest, 2014), BBB permeability (Varanoske et al., 2022) and intestinal barrier integrity (Zhao et al., 2019, Karl et al., 2017). The latter is responsible for sustaining homoeostasis in the gut micro-environment (Amini-Khoei et al., 2019, Uren Webster et al., 2020) and can be influenced by the neurological state of the host.

In murine models, the impact of the HPA axis on various behavioural parameters such as appetite, aversion and cognition are studied by using a multitude of tests such as forced suspension, and immobilisation (despair related behaviour), sucrose water preference (depression-like behaviour), elevated plus maze (anxiety-like behaviour) and startle reflex (responsiveness) and many others (Packard et al., 2016). Consumption of fermented foods including red bean tempeh, cheese and fermented plant (Laminaria japonica) as assessed by murine models, was shown to reduce anxiety-like behaviours and corticosterone levels when subjected to stress (Chen et al., 2020a, Chen et al., 2021b, Fourman et al., 2021a, Jung et al., 2020). Similar attenuations in depression and anxiety-like phenotype accompanied by lowered plasma corticosterone and inflammatory markers such as NF- Κb, TNF-α and IL-6 were observed in the colon of mice supplemented with fermented ginseng extract for a period of 5 days (Han et al., 2020). In a comparable manner, fermented porcine placenta supplementation for a 21-day period, which showed lowered cortisol along with creatinine kinase and lactate. The study also showed concurrent lowering of circulating cytokines- IL-6, TNF-α, IL-1β, IL-4 and IFN-γ (Kim et al., 2016). All of the mentioned studies report a strong interplay between the immune system and the HPA axis. A potential mechanism by which fermented foods are able to modulate cortisol release could be via blunting the response peripheral immune challenge by causing a reduction in circulating cytokines and other inflammatory markers which are responsible for activating neuronal projectionsinto the PVN (Bellavance and Rivest, 2014). Interestingly, a study on mice subjected to restraint stress and receiving milk kefir treatment, reported the probiotic containing fermented food was able to block HPA axis dysregulation by attenuating the altered expression of glucocorticoid receptors in the PVN, when compared to mice receiving unfermented milk. The study also hinted at the HPA axis dysregulation to be highly sex specific and a potential connection between IL-6 and glucocorticoid receptor expression (Smith et al., 2021). Such interesting exploratory efforts could shape our understanding of the multifaceted role played by fermented foods in modulating the microbiota-gut-immune-brain axis.

Fermented food interventions in human cohorts have been limited and conflicting. An 8-week intervention of fermented porcine placenta resulted in lower serum cortisol levels along with concurrent reduction in the mRNA expression of IL-1β following treadmill stress testing (Yoon et al., 2020). Similarly, reductions in salivary cortisol was observed in students receiving Lacticaseibacillus casei strain Shirota and subjected to examination stress. The same strain was also shown to blunt the release of corticosterone in rats subjected to water avoidance stress (Takada et al., 2016). Conversely, some studies reported no effects of fermented foods on post-stress cortisol levels (Jaatinen et al., 2014, Marcos et al., 2004), potentially due to the small quantity, diverse nature of fermented food being consumed, and short duration of intervention.

2.5. Fermented foods influencing serotonin secretion and receptor expression

The enterochromaffin cells in the intestine are major producers of serotonin (Lund et al., 2018). Gut-derived serotonin is shaped by host diet (Bruta et al., 2021, Horn et al., 2022), baseline microbiota composition (Reigstad et al., 2015a, Yano et al., 2015, Hata et al., 2017) and microbial metabolites produced as a consequence (Reigstad et al., 2015a, Yano et al., 2015). Although serotonin cannot pass the BBB, it can influence fundamental aspects of the gastric system, such as regulating secretion, motility and tonicity. There are a growing number of studies focusing on the influence of fermented foods on serotonin secretion with special emphasis on tryptophan metabolites. Tryptophan metabolites are critical precursors for serotonin and melatonin biosynthesis and also lead to the production of indole and kynurenine-related metabolites, all of which are increasingly being appreciated for their neuroactive role in addition to maintaining gut health.

Preclinical studies report lowered serotonin turnover in the colon of mice supplemented with milk kefir compared to unfermented controls (Van De Wouw et al., 2020), which was also reflected in mice subjected to immobilisation stress and supplemented with fermented Chinese medication such as fermented Mentha arvensis and fermented Cornus officinalis (Tian et al., 2018, Tian et al., 2020). Other preclinical models have demonstrated the ability of fermented soy-based food such as tempeh to modulate the level of serotonergic genes. Tempeh-fed zebrafish showed upregulation in genes involved in transportation, synthesis and signalling of serotonin, namely tph1b, tph2 and slc6a4a genes in the brain (Chen et al., 2021b). The ability of fermented foods to modulate peripheral serotonin levels could be attributed to the presence of endogenous bacteria within the food, which are capable of producing these metabolites as a result of the fermentation process (Jeong et al., 2021, Gallardo-Fernández et al., 2022, Kumar et al., 2022). The ability of fermented foods to modulate gut health could also be credited to synbiotic components present in the food. A study of a murine model of constipation employed a synbiotic yogurt containing konjac mannan oligosaccharide (prebiotic) and Bifidobacterium animalis spp lactis (probiotic). The study reported an increase in levels of acetylcholine, substance P and upregulation of motilin, vasoactive intestinal peptide receptor (VIPR-4) and 5-HT4 receptors in colonic tissue upon intervention with the synbiotic containing yogurt (Li et al., 2021b).

Despite these developments, there remains a shortage of human studies that have extensively explored the involvement of fermented foods in modulating the serotonergic landscape. In one instance, daily consumption of a fermented milk product containing Lactobacillus casei for 8 weeks showed increased faecal serotonin and decreased abdominal discomfort without changes in serum tryptophan, kynurenine, salivary cortisol and IgA levels compared to participants receiving unfermented controls (Kato-Kataoka et al., 2016). Other studies on white wine consumption have shown lowered myogastric electrical activity, variation in plasma serotonin and dopamine profiles (Boyer et al., 2004, Levanon et al., 2002). Peripherally produced serotonin cannot cross the BBB but downstream metabolic products of tryptophan metabolism can cross the BBB and influence neurological state of the host. In recent years, research has been dedicated towards understanding the effects of tryptophan metabolism, especially the kynurenine pathway, on host behaviour and health. A meta-analysis of prebiotic and probiotic supplementation on tryptophan metabolism revealed significant decrease in kynurenine and kynurenine:tryptophan ratio with probiotic supplementation (Purton et al., 2021). Indeed, metabolomic analyses of fermented foods have revealed that they are reservoirs for tryptophan metabolites (Yılmaz and Gökmen, 2020), revealing that it could be interesting to see how peripheral supplementation of tryptophan metabolites can affect mood and behaviour modulated by the gut microbiota.

2.6. Fermented food modulating enteroendocrine signalling

The enteroendocrine system (EES) is capable of secreting molecules that can influence the afferent vagus nerve, and receptors of these molecules are expressed higher up in the nucleus tractus solitarius (NTS) and hypothalamic regions of the brain. These regions undermine energy expenditure, food preferences and satiety of the host thereby mediating feeding behaviour (Latorre et al., 2016, Holst, 2013). Functioning of the EES is also influenced by microbial metabolites that are introduced through dietary intake. The EES primarily hosts a network of gut hormones such as serotonin, neuropeptide-Y, GLP-1, ghrelin, peptide YY (PYY), motilin and somatostatin that continuously relay information to the enteric nervous system (ENS). The impact of each of these gut hormones and peptides on the microbiota-gut-brain axis has been described in detail across multiple extensive reviews (Richards et al., 2021, Wachsmuth et al., 2022, Sun et al., 2020). In brief, the hormones of the EES regulate motility, appetite, release of insulin and bile acids (Sun et al., 2020). The submucosal and the myentric plexsus of the ENS communicate with each other via a small collection of neurotransmitters and neuropeptides, which can be secreted and modulated by the fermented food microbiome, microbial metabolites and co-occurring active principles.

2.7. Fermented food and gut hormone secretion

The enteroendocrine cells (EEC) receive multiple stimuli ranging from nutrients, toxins and microorganisms present in fermented and unfermented food and respond by releasing peptides and hormones. Modulating the levels of GLP-1 is being used to attain improved glucose homoeostasis (Yadav et al., 2013) as well as to target obesity (Aldawsari et al., 2023) resulting in the development of wide array of GLP-1 analogues aimed at managing obesity (Dailey and Moran, 2013) and type 2 diabetes mellitus (Maselli and Camilleri, 2021). Dietary intervention strategies are also being employed to target circulating levels of GLP-1 peptides in patients diagnosed with type 2 diabetes (Di Mauro et al., 2021) and abdominal obesity (Fuglsang-Nielsen et al., 2021). This recent exploration in the use of dietary strategies to combat type 2 diabetes mellitus has resulted in the identification of microbial metabolites, predominantly SCFA that can improve production of GLP-1 and peptide YY, both of which are implicated in regulation of insulin release. (Zhao et al., 2018). Fermented foods as a potential source of prebiotics and probiotics to shape the gut microbiota and its metabolites are also being explored to manipulate the levels of gastric peptides that are implicated in satiation and insulin release (Manaer et al., 2015, Fallah et al., 2018).

Both unfermented and fermented foods contain bioactive components that can influence the secretion of incretins such as GLP − 1, a gut hormone implicated in satiety (Johnson and De Mejia, 2016). It is therefore necessary for studies focusing on the impact of fermented foods on the enteric environment to concurrently account for the effect of unfermented controls on EEC secretion profiles. The patterns of endogenous and exogenous release of GLP-1 have been well reported in murine models and human studies (Flint et al., 1998, Terrill et al., 2019, Chen et al., 2021a) and therefore can be adopted to study the effects of fermented foods on satiation. Interestingly, in-vitro studies on kippuku-cha, a fermented Japanese beverage, was found to activate GPR-120 and stimulate the release of GLP-1 via phosphorylation of the Erk1/2 (p42/44 MAP kinase) pathway (Nagasawa et al., 2020). Similarly, phenol and polysaccharide extracts from quinoa yogurt fermented with starter cultures of S. thermophilus and Lactobacillus bulgaricus stimulated the release of GLP-1 and influenced expression of proglucagon mRNA, CCK and c-FOS along with DDP-1 V inhibitory potential (Obaroakpo et al., 2020). The study revealed that the fermented product elevated concentrations of GLP-1 and proglucagon than the unfermented controls. Such studies assessing the incretin release profile using in-vitro studies must be interpreted with caution, as their effects can be inflated when compared to in-vivo conditions. For example, fermented dairy-based products such as whey possess the ability to stimulate the release of GLP-1 and CCK (Chaudhari et al., 2017, Sánchez-Moya et al., 2020), however these effects were not translated to in-vivo studies (Kondrashina et al., 2018). The study stated that this lack of translatability between in-vitro to in-vivo models could be due to the action of gastrointestinal enzymes, which may reduce the GLP secretogogue capacity of fermented dairy products unless enterically protected (Kondrashina et al., 2018).

Human studies often adopt self-reported questionnaires to gauge the effect of intervention strategy on satiety and hunger. In humans, fermented dairy, fermented dairy alternates and fermented bread are predominantly studied for their ability to modulate satiety in the individual, but only a few studies investigated circulating GLP profiles. For example, consumption of sourdough bread by Swedish adults resulted in pronounced satiety compared to consumption of unfermented whole grain and yeast fermented controls. However, the effects of sourdough on insulin release and satiety has not been replicated in two other clinical trials (Shah et al., 2020, Iversen et al., 2018). The GLP-1 release profiles across all the cohorts were unaffected, suggesting fermented foods enhance satiety via mechanisms independent of GLP-1 (Zamaratskaia et al., 2017). Apart from sourdough, high levels of vinegar supplementation has been associated with increased satiety and lower postprandial blood glucose (Ostman et al., 2005). Researchers have hypothesised that the satiating effect of vinegar and some sourdough could be due to the acid content of the fermented product. The inconsistent results observed on satiety profiles across studies could be attributed to factors such as sex, sample size, duration of intervention, method of sourdough preparation (including species and strain level differences) and quantity of sourdough consumed (Ribet et al., 2023).

2.8. Impact of fermented food consumption on ghrelin and leptin levels

Desacyl ghrelin and acyl ghrelin are orexigenic hormones that are the predominant forms of circulating ghrelin and are implicated in the regulation of appetite and its neuroprotective effects (Lach et al., 2018, Rhea et al., 2018, Huang et al., 2019). Unlike serotonin, ghrelin is released in the periphery and can cross the BBB, thereby binding to several targets associated with food reward including corticolimbic, amygdala, insula and orbitofrontal cortex (Zanchi et al., 2017, Farokhnia et al., 2019). Leptin, an anorexigenic hormone produced by adipocytes and enterocytes, is shown to activate several regions of the brain such as the brain stem, parahippocampal gyrus, middle frontal gyri, middle temporal gyrus, right hypothalamus and lingual gyrus (Zanchi et al., 2017). Several detailed reviews have highlighted the role of gut hormones in regulating food reward-motivated behaviour in the host and also in contributing towards the conversation between the gut and the brain (Lach et al., 2018, Zanchi et al., 2017, Decarie-Spain and Kanoski, 2021). It is therefore essential to also understand if fermented foods can alter levels of ghrelin/leptin in circulation and consequently affect the food reward-motivated behaviour of the host.

Preclinical murine models have addressed this topic by showing a reduction in plasma ghrelin levels post-intervention with fermented goat and cow milk and mixed results with respect to impacts on plasma leptin and adiponectin levels (Diaz-Castro et al., 2017, Muñoz Alférez et al., 2020). Moreover, rats receiving a fermented whey beverage alongside a high-fat diet showed no difference in leptin and ghrelin profiles, despite a decrease in food intake, compared to high-fat diet controls (Hong et al., 2015). Some studies have included unfermented controls and have reported a greater reduction in serum leptin (Lu et al., 2021) and ghrelin levels (Liu et al., 2019a) upon consuming the fermented counterparts. Such reports serve as an important reminder to interpret other studies that lack critical controls with caution, so as to prevent inflating the potential health benefits associated with fermented food intake. A number of human studies have tried to ascertain if the satiating effects of fermented food are due, at least partially, to protein content. In one randomised crossover study, the diets of participants were supplemented with fermented and unfermented tempeh (protein control) for a period of two weeks. The group receiving the fermented product showed a reduction in circulating acyl-ghrelin levels by 30% with concurrent reduction in circulating insulin compared to unfermented controls (Noer et al., 2021, Diaz-Castro et al., 2017). Other clinical studies have reported varied gastric hormone responses depending on the fermented product consumed. The effects have ranged from an increase in ghrelin after moderate beer consumption without changes in leptin levels (Beulens et al., 2008) to almost no differences in the level of leptin and ghrelin release amongst different fermented dairy products (Hansson et al., 2020) and no effect on sensation of satiety post-intervention with red wine despite lowering of plasma ghrelin levels (Ismail et al., 2022). These results provide evidence of the highly individualistic enteric hormone release profiles towards the consumption of fermented foods along with potential involvement of other pathways/gut hormones, in driving satiation.

2.9. Other pathways and mechanisms through which fermented food can influence appetite

Another gut hormone associated with satiety is somatostatin, which has long been implicated in satiety and hunger. Activation of somatostatin positive neurons in the brain shows increased preference towards high-calorie foods and may be implicated in the development of metabolic conditions such as obesity (Kumar and Singh, 2020); but it is important to note that somatostatin has alsobeen shown to regulate the release of gastric enzymes, increased gastric emptying and mediating inflammatory response. The stomatostatin release profile after consumption of fermented soy bean (natto) was demonstrated in mice subjected to gastric mucosal injury. The mice receiving the fermented soy bean showed higher levels of somatostatin, vasoactive intestinal polypeptide, and motilin (Suo et al., 2015) and concurrently showed lowered levels of serum cytokines. Microbial metabolites are also being studied for their participation in modulating satiety of the host. Several studies exploring the effects of SCFAs have shown that feeding behaviour can also be influenced by these bi-products of bacterial fermentation (Brooks et al., 2017). For example, yogurt supplementation to rats reduced weight and elevated faecal SCFAs (Qu et al., 2018). Similarly, a type 2 diabetes model of mice receiving kombucha lowered fasting blood glucose and increased faecal SCFAs compared to unfermented tea (Xu et al., 2022). However, both interventions maintained intestinal permeability and restructured the gut microbiome by increasing the abundance of Lactobacillus, Butyricicococcus and Lachnospiraceae with a concurrent increase in the levels of GLP-1 and PYY secretion. A potential mechanistic explanation for this could be the stimulating action of SCFA on the mRNA for POMC, AgRP, CART in the hypothalamus (regions of the brain implicated in feeding and energy expenditure) (Pichiah et al., 2016), a pattern which is in agreement with targeted peripheral administration of acetate (Frost et al., 2014). Together, these studies hint at a potential crosstalk between gut microbiota, microbial metabolite profiles and the ENS/CNS within the individual. Most of the discussed pre-clinical studies have included only a single time point post-intervention with the fermented food. The analysed parameters include only a subset of the phenotypical traits such as body weight, feeding pattern and adiposity. In addition to this, molecular markers implicated in understanding satiety and feeding behaviour is also being incorporated. This method of experimental design might be insensitive in capturing transient changes in the gut microbiota and microbial metabolite composition, and gut hormone profile, and can therefore be resolved by sampling at multiple time points.

2.10. Fermented food and gut microbiota

The gut microbial community produces metabolites that can modulate host health and reinforce/reduce the integrity of the intestine and BBB thereby modulating the levels of pro-inflammatory triggers reaching systemic circulation and eventually the brain. It is therefore important to understand the impact of fermented food on the intestinal milieu and intestinal health, as they continuously relay information to the peripheral immune system, enteric and central nervous system.

In addition to the modulation of the gut microbiota, preclinical studies have also looked at the impact of fermented dairy products on intestinal permeability (Putt et al., 2017), microbial metabolites (Gao et al., 2021) and peripheral cytokine profile (Du et al., 2022). Frequently studied microbial metabolites, SCFA’s were elevated in interventions containing fermented plant-based products such as fermented carrot juice (Liu et al., 2021, Yu et al., 2022, Ma et al., 2021), fermented beverages containing fruits and vegetables (Wang et al., 2019), and fermented raspberry pomace enriched with lactic acid bacteria (Wu et al., 2021). SCFA’s are implicated in the maintenance of intestinal barrier integrity (O’riordan et al., 2022) and might be the driving force in increasing the expression of tight junction proteins thereby reinforcing the permeability of the intestine (Zorraquín-Peña et al., 2021, Taladrid et al., 2021).

It is to be noted that sex is a critical factor that can influence the composition of the gut microbiome (Bridgewater et al., 2017). Studies exploring the sex-selective effects on gut microbiome composition have been conflicting, and this may be due to small sizes of study population, body mass index, xenobiotic use, diverse dietary patterns, pathological condition, inconsistencies in tools and pipelines employed to study microbiota composition, diverse ethnicities and strains of models employed in preclinical studies (Kim et al., 2020). The aforementioned factors influence the gut microbiome to a greater effect than sex in humans (Kolde et al., 2018, Jaggar et al., 2020, Lloyd-Price et al., 2017). Murine models, however, seem to show a much more pronounced difference in gut microbiota profile than human studies (Mcgee and Huttenhower, 2021), which has led to differential responses to dietary intervention strategies (Bridgewater et al., 2017). Considering this sexual dimorphic nature of the gut microbiota, a recent study of dietary fibre supplementation for 20 weeks to mice revealed sex-specific responses in the gut microbiome and faecal metabolome. Female mice on a high inulin-based diet showed increased abundance of Bacteroidota and a decline in Faecalibaculum and Lactococcus, the latter of which is exaggerated in diets rich in dietary fibre (Lloyd-Price et al., 2017). The sexually dimorphic nature of the microbiota in mice along with the differences in the gonadal hormone profile results in a distinct humoral response. As a result, there is a marked increase in genes pertaining to inflammation in the hippocampus and hypothalamus profile in a sex-specific manner when subjected to probiotic intervention (Yahfoufi et al., 2023). Fermented foods being a possible source of prebiotics and probiotics, as previously described, have resulted in a sex-selective response (Murray et al., 2019). Therefore, preclinical murine models should be controlled for sex-selective variations in the gut microbiome and downstream immunological profiles.

2.11. Microbiota modulatory components of fermented foods

Several literature works have documented the capacity of fermented foods to restructure the gut environment and modulate host health (Stiemsma et al., 2020, Leeuwendaal et al., 2022, Selhub et al., 2014, Melini et al., 2019, Aslam et al., 2020, Costa et al., 2021, Moreno-Arribas et al., 2020), hinting at the potential capacity of fermented foods to influence microbiota gut-brain-axis communication. Some of the modulatory components are produced as a result of the fermentation process. For instance, fermented beverages can contain a percentage of alcohol and other volatile components that can influence the composition of the gut microbiota. The impact of the alcoholic fraction of fermented food on the gut microbiome has been reported in murine (Lee et al., 2020) and human studies (Marques et al., 2022). The alcoholic portion of the fermented food has to be accounted for as it has been recently implicated for its ability to confer a neurological benefit to the food consumer. A retrospective cohort study analysing the impact of alcohol consumption and changes in their intake frequency with risk of dementia revealed its potential neuroprotective activity. The study reported mild and moderate drinkers had a lower risk of all-cause dementia compared to non-drinkers whereas heavy drinkers were posed to have a higher risk of all cause dementia (Jeon et al., 2023). In fact, recent systematic reviews have revealed that low to moderate percentage intake of alcohol in several of the fermented foods and beverages have been linked with lowered risk of Alzheimer’s disease and dementia (Porras-García et al., 2023). For this reason, it is also important to control for the effect of alcohol on the gut microbiota and gut brain communicatory pathways. Fermented foods can also alter metabolite production in the gut even without altering the microbial composition or diversity in some circumstances. This can be due to the presence of dietary fibres and bioactive components that are intrinsically present in the food (Zhou et al., 2019, Wang et al., 2021). Fermented foods can also be rich in phytochemicals and other bioactive agents, though some of these originate from the food substrate and hence are also present at similar concentrations in the unfermented counterparts. In light of this, it is important to include unfermented controls in preclinical and human study design, so as to accurately understand the components of fermented foods driving the change on the intestinal milieu and on cognitive health. Supplementary table 1 provides a non-exhaustive compilation of important phytochemicals present in both fermented and unfermented foods that can influence microbiota-gut-brain communication. Fermented foods also contain a variety of potentially health-promoting microorganisms that can confer better gut health in addition to contributing towards the restructuring of the gut microbiota. Indeed, strains isolated from fermented camels’ milk, pickled Chinese cabbage (LAB fermented), fermented yogurt when formulated into a probiotic cocktail (Lactobacillus species) all showed the ability to restore the colonic microbial community in antibiotic-treated mice and shift the gut microbiota at a phyla level from a Pseudomonadota-dominated environment to a profile similar to the control group (Shi et al., 2018).

In other instances, it would seem that the beneficial impacts are not attributable to a single strain but rather a combination of factors, including the introduction of multiple microbes, metabolites and substrates that confer benefit to the host. This was evident in mice receiving fermented barley juice, which resulted in an altered faecal metabolomic profile in a manner that was distinct from the group receiving Lactobacillus plantarum alone (Zhu et al., 2021). It should be noted that the ability of fermented foods to shape the gut microbiome in preclinical models does not necessarily translate to human studies (Nguyen et al., 2015, Hugenholtz and De Vos, 2018). To address this, many studies have explored the impact of fermented foods on the gut microbiome in humans. For instance, observational studies reveal high alpha gut microbiome diversity in Korean participants who habitually consumed high amounts of fermented legumes, fermented vegetables, tea, seaweed and nuts (Noh et al., 2021). Similarly, self-reported fermented food consumption patterns in 130 participants from northern Spain also revealed that fermented dairy consumption correlated with higher levels of Akkermansia and low levels of Bacteroides with concurrent high levels of SCFA such as propionate and butyrate, a pattern that was also observed among cheese consumers when compared to non-consumers (González et al., 2019). Greater beta diversity has been seen in regular fermented food consumers than occasional consumers with the former displaying higher proportions of Faecalibacterium prausnitzii, Prevotella spp, Pseudomonas spp, Clostridiales, Enterobacteriaceae, Lachnospiraceae and Bacteroides spp (Taylor et al., 2020). On the other hand, the effects of fermented foods on gut microbiome diversity are inconsistent between intervention and observational study designs with studies reporting little to no change in the diversity metric after intervention with fermented food (Alvarez et al., 2020, Berding et al., 2023, Le Roy et al., 2022, González et al., 2019). The conflicting nature of these studies could be attributed to the gut microbiome being predominantly resistant to change over time (Faith et al., 2013). It could also be due to the study duration being too short to capture subtle changes, diverse nature of fermented foods being incorporated into the study design, high degree of interindividual-variability of baseline gut microbiome amongst participants warranting need for crossover studies with sufficient washout period. Other factors include studies employing 16 S rRNA gene sequencing as opposed to shotgun metagenomics sequencing, thereby lacking species and further strain level resolution that can significantly contribute towards understanding the subtle changes in microbiota composition along with variation in functional capabilities of microbiome.

3. Preclinical and clinical landscape: application of fermented foods to study microbiota-gut-brain axis communication

The ability of fermented foods to influence the major communication pathways involved in gut to brain signalling has been foundational in its adoption to target mental health. To identify fermented foods that modulate the microbiota-gut-brain axis, it is important to know the current ways by which fermented foods are able to alter communication pathways that are involved in relaying information from the gut to the brain. The priming of the immune system and the restructuring of the gut microbiota and microbial metabolites via fermented food supplementation has been extensively studied. In addition to the immune system, the ENS, along with its gastric peptides and hormones, is also being shaped by fermented food supplementation. Table 3 provides a summary of the current understanding of fermented foods on communication pathways between the gut and brain derived from Section 2.

Table 3. Summary of the potential mechanisms of how fermented foods modulate microbiota-gut-brain signalling. See Section 2 for details.

Microbiota-gut-brain axis modulation mechanism	Model employed	Fermented food investigated	Potential mechanism of modulation
Immune system:	Preclinical	• Milk kefir • Sauerkraut • Camembert cheese • Yogurt	1. Stimulate the release of IL-10, TNF-α 2. Release of microbial metabolites that can bind to HCA3R receptor 3. Lower levels of IL-6, TNF-α and IFN-γ upon challenge with pathogen 4. Stimulation of IgA release by probiotic components present in milk kefir
	Clinical	• Yogurt • Sauerkraut • Kimchi • Fermented food based diet	1. Lowered circulating levels of IL-6, IL-10 and IL-12b and immune cells such as CD4+, CD8+, T and B cells 2. Reduction in circulating cytokines such as IL-4, IL-10, IL-12 and MCP-1
HPA axis	Preclinical	• Red bean tempeh • Cheese • Fermented Laminaria japonica • Fermented Ginseng • Fermented porcine placenta • Milk kefir	1. Lowered cortisol circulating cytokines- IL-6, TNF-α, IL-1β, IL-4, NF- κb, and IFN-γ 2. Altered expression of glucocorticoid receptors in the paraventricular nuclei in the brain
	Human studies	• Fermented porcine placenta • Fermented milk • Fermented food diet	1. Lowered serum cortisol and mRNA expression of IL-1β
Serotonin and Serotonin receptor Expression	Preclinical	• Milk kefir • Fermented Mentha arvensis • Fermented Cornus officinalis • Tempeh • Synbiotic containing yogurt	1. Lowered serotonin levels and increased serotonin turnover 2. Upregulation of genes pertaining to synthesis and transportation of serotonin in the brain 3. Increase in colonic levels of 5HT-4 receptors
	Human studies	• Fermented milk product • Wine	1. Increased levels of serotonin when measured in faeces and plasma
Enteroendocrine system: Via peptides and hormones	Preclinical studies	• Kippuku-cha • Fermented quinoa yogurt • Fermented dairy • Fermented whey • Kombucha • Natto	1. Release of GLP-1 via activation of GPR-120 2. Release of GLP-1 by phosphorylation of Erk1/2 (p42/44 MAP kinase) 3. Activation of GLP-1 release via increased expression of CCK and c-FOS along with DDP-1 V 4. Reduction in serum leptin, insulin and ghrelin 5. Higher levels of somatostatin, vasoactive polypeptide and motilin with consumption of fermented food
	Human studies	• Tempeh • Sourdough • Vinegar • Beer • Fermented dairy • Red wine	1. Pronounced effects of satiety reported post consumption of sourdough and vinegar 2. Increased/absence of difference in leptin levels with consumption of fermented foods
Intestinal milieu	Preclinical	● Fermented fruit and vegetable beverages ● Fermented raspberry pomace	1. Production of SCFA, which is implicated in the maintenance of BBB and intestinal barrier 2. Restructuring gut microbiota and microbial metabolite composition
	Human studies	● Fermented food based diet	1. Increased diversity metrics of the gut microbiota in self-reported consumers of fermented food

READ THE ORIGINAL ARTICLE FOR DATA ANALYSIS AND REFERENCES