Our extended microbiome: the human relevant metabolites and biology of fermented foods

Summary

One of the key modes of microbial metabolism occurring in the gut microbiome is fermentation. This energy-yielding process transforms common macromolecules like polysaccharides and amino acids into a wide variety of chemicals, many of which are relevant to microbe-microbe and microbe-host interactions. Analogous transformations occur during the production of fermented foods, resulting in an abundance of bioactive metabolites. In foods, the products of fermentation can influence food safety and preservation, nutrient availability, and palatability, and once consumed, may influence immune and metabolic status, disease expression and severity. Human signaling pathways perceive and respond to many of the currently known fermented food metabolites, though expansive chemical novelty remains to be defined. Here we discuss several aspects of fermented food-associated microbes and metabolites including: a condensed history; current understanding of their interactions with hosts and host-resident microbes; connections with commercial probiotics; and opportunities for future research on human health and disease and food sustainability.

Introduction

In food, fermentation includes “foods made through desired microbial growth and enzymatic conversions of food components”¹, occurring in either aerobic and anaerobic contexts. The food fermentation process (Box 1) depends upon the microbial community to metabolize nutrients², altering the substrate (i.e., food) and producing metabolites and other molecules that can impact taste, texture, safety, and nutritional composition, as well as signal to the host and gut microbiome.

Callout Box 1-. What are fermented foods?

With the fermented food market expected to grow by over half a billion dollars in the next 4 years²⁷, fermentation has been of growing research interest. Numerous reviews have offered an extended explanation of fermented food production^1,2. Here we offer a brief summary.

Fermented foods (also referred to as ferments or microbial foods) are foods transformed by microbial growth and enzymatic conversions¹. During fermentation, members of the microbial community, which can include bacteria, yeast, or filamentous fungi, compete for resources with strain-specific nutritional and growth-condition preferences, accompanied by secretion of metabolites such as organic acids and antimicrobial peptides^43,50 to gain advantage over competitors. Most fermented food is produced for specific sensory qualities including taste and aroma, which result from the production of specific metabolites. Key to the fermentation process is encouraging the growth of microbes through the manipulation of extrinsic factors such as temperature, pH, salinity, and humidity.

Where are the microbes coming from?

The source of microbes in a fermented food can be spontaneous, from a starter culture, or through ‘back-slopping’. In spontaneous fermentation, microbes from the environment (e.g. the surface of the raw ingredient, the hands of the producer, the room and equipment used to make the fermented food) initiate fermentation¹⁸¹. While the initial microbial community will be more diverse, the raw ingredient and environmental factors will select for specific members of the community to dominate⁵⁴. Use of a starter culture (e.g. selected strain or group of strains) in a pasteurized substrate allows for a reliable final profile of the fermented food, preventing competition between the native microbial community and the added starter community, largely seen in fermented dairy production like yogurt and cheeses²⁰. In addition, starters can be constructed of characterized and sequenced strains deliberately chosen based on the presence or absence of specific traits, for example to minimize the presence of antibiotic resistance genes that might be transferable to resident gut microbes¹⁸². When a portion of a previous fermented food is incorporated into a next batch, propagating the microbial community, the process is called back-slopping fermentation² and is analogous to subculturing or serial passage of microbial cultures in the lab. Examples of back-slopping include sourdough bread and kombucha¹⁸³, where a higher abundance of a previously successful microbial community can be used to initiate fermentation, although the community itself might be poorly characterized.

Dominant microbes involved in fermentation have been largely characterized, but there remains a need for the continued characterization of microbes involved in fermentation. In particular, characterization of microbes at lower abundance which might play a role in shaping the microbial community or contribute to the metabolite pool. In addition, while there is general consistency in microbes that will be most abundant at the final stages of fermentation, like L. plantarum in vegetable fermentation or S. thermophilus and L. bulgaricus in yogurt, their final relative abundance and the presence and abundance of other microbes in the community remains variable from batch to batch. In line with microbial heterogeneity between fermentation batches, metabolite production and abundance will also be variable¹⁰⁵. Ongoing work modeling microbiota and metabolite dynamics during fermented food community assembly will shed light on how extrinsic factors might be used to decrease product variability¹⁸¹.

What does it mean for a food to be fermented?

While there are broad categories of microbes that can be involved in the fermentation process (e.g., lactic acid bacteria, acetic acid bacteria, filamentous molds, and yeast), lactic acid fermenting microbes are one of the most common classes, seen in the fermentation of many vegetables, dairy, and grains (e.g. kimchi, yogurt, and injera). In lactic acid fermentation of vegetables, addition of salt and access to preferred carbohydrate nutrient sources promotes growth of halophilic lactic acid bacteria (LAB), leading to early community dominance and secretion of antimicrobial peptides and organic acids (predominantly lactic acid), further enhancing growth^51,68,184. LAB outcompetes potential spoilage-associated microbes and competitors, excluding undesirable microbes from the final ferment. Secretion of protein hydrolases to enhance nutrient accessibility, shown primarily in mold fermentation microbes like Aspergillus oryzae, and Rhizopus oryzae, not only promotes fungal growth but also plays an important role in enhanced palatability of the fermented food. Additional metabolites produced during this process play important roles in enhancing palatability, food safety, and mediate microbe-microbe and microbe-host interactions. It is important to note that many non-fermented foods are produced to mimic the flavor and preservation of fermented foods (e.g., vinegar/salt brined pickles)¹. Manufacturers may add viable microbes to pickled foods or pasteurized fermented foods at the end of production, attempting to recreate the living microbial content of raw fermented foods and enabling “probiotic” or “contains live microbes” label claims; in many cases the supplemented microbes are different strains than those associated with fermentation and may be spores and therefore not metabolically active. Ranging widely across products, microbial foods produced via lactic acid fermentation can reach 10⁵–10¹⁰ CFU/g¹⁸⁵, producing and consuming metabolites during microbial growth. While addition of probiotic microbes to consumer-ready food products increases CFU count, fermentation (and the accompanying metabolite production) does not occur. Therefore, the simple presence of bacteria or fungi on an ingredient label does not necessarily mean it is a fermented product. For example, an olive brined in dilute lactic acid and salt with spores of a Bacillus species (distant relative of Lactobacillus that is able to form spores) added before sealing can create a shelf stable product with highly reproducible flavor that mimics a lactic acid ferment and enables a claim of “contains live microbes”. However such a product deviates significantly from a traditionally fermented olive. Data is lacking as to whether mimicking food fermentation by adding purified metabolites or live microbes has the same benefit, particularly when consuming microbes distinct from those that are typically found in fermented foods. In addition, in the addition of probiotics to food (considered “dietary supplements”) FDA draft guidelines recommend including the total weight of the added microbial mass and CFU of viable microbes, with no recommendation as to maximum allowed supplemented probiotic CFU or weight count¹⁸⁶. While estimates from the National Health and Nutrition Examination Survey (NHANES) data shows there is a trending increase in consumption of foods containing live microbes in the American diet, estimates of CFUs or microbial type are still lacking¹⁸⁷. Characterizing the typical range of dietary microbe consumption and defining a healthy personalized range is necessary for the establishment of dietary recommendations and regulation of commercial products²⁸.

While this type of food fermentation happens ex vivo, a comparable microbial metabolism occurs in the gastrointestinal (GI) tract. Research on the gut microbiome has highlighted the importance of microbiome-derived metabolites in mediating human health and disease^3–9. By applying the same framework developed to study metabolite-host interactions in the gut, we can better understand the effects of fermented food metabolism on human health, offering an unexplored additional source of human relevant microbial metabolites.

From the perspective of the gut microbiome, fermentation can be defined as a primary metabolic process occurring in the absence of oxygen, such as the consumption of complex carbohydrates in the distal colon. Dietary compounds metabolized by the gut microbiome lead to production of energy and metabolites, small chemical compounds that can influence the structure of gut microbial communities and signal directly to the host via interaction with different cell types within the gut^10,11, or signal peripherally following absorption into circulation. Nutrients exiting the stomach that are not absorbed by the host small intestine make their way to the distal small intestine and colon providing a pool of dietary compounds that gut microbes can access and metabolize.

Microbial strains that are well-matched to incoming nutrients tend to outcompete other members of the community, resulting in changes in the overall composition of the microbial community^12,13. This can occur within days of a major dietary change¹⁴. The metabolites produced during microbial fermentation in the gut are a direct product of microbes that are being fueled by nutrient inputs. These compounds provide an energy source for other microbes, impact the ecosystem chemistry (e.g., pH), directly signal to the host, further restructuring the microbial community, and/or enter host systemic circulation, broadening metabolite reach^10,15–17. Postbiotics refer to the molecules that microbes produce¹⁸, as opposed to substrates that promote the growth of gut microbes, known as prebiotics. In this framework, diet can be considered a source of nutrients for both host and microbes, as well as a source of precursors for metabolites produced during fermentation by the gut microbiome.

There is a growing appreciation that microbially-produced metabolites have a mediating role in health and disease, as seen in type 2 diabetes (T2DM), inflammatory bowel diseases (IBD), and metabolic-associated fatty liver disease^3–9. Evidence of the influence of microbial-derived metabolites primarily come from mechanistic studies, where they may promote health and disease states, as well as in their use as diagnostics or biomarkers. Production of these metabolites is influenced by a vast array of factors including diet, medications, baseline gut microbiota composition, and genetics¹⁹. Similarly, metabolites produced during food fermentation offer a novel pool of signaling compounds that can interact with the GI tract, acting directly and indirectly on the host, with the potential to be further metabolized.

Consumption of fermented foods expands the possible interactions in the gut through increased chemical diversity. By considering food fermentation as an “extended microbiome”²⁰, a term previously used to describe fermented foods by Dunn et al. 2021, we can reframe our relationship to fermented foods and their role in human health. Important considerations include how the practice of fermentation has shaped human biology, how fermented food consumption impacts health directly and indirectly, and future applications of food fermentation in biomedicine and extending beyond health. This perspective focuses on the role of metabolites as key mediators in the diet-microbe-host landscape, presenting current and potential roles fermented food microbiota-derived metabolites play in human health. We also review the need for expanded foundational information, such as characterizing the microbial communities and molecules associated with the wide array of fermented foods consumed globally.

Fermented Foods: A Brief History

While the earliest evidence of human engagement with food fermentation dates back to at least 14,000 years BCE²¹, evolutionary genetics supports a much longer relationship between humans and microbially produced metabolites. Roughly 10 million years ago, corresponding to the movement of our great ape ancestors from trees to the forest floor, evolution of the alcohol dehydrogenase class IV (ADH4) enzyme allowed our hominid ancestors to metabolize alcohol, likely a primary metabolite produced by yeast spontaneously fermenting fallen fruit²².

Over the course of human history, food fermentation has been practiced by most culinary traditions, dependent on substrate availability²⁰. The use of fermented ingredients is not uncommon in historical medicinal practices, with documented use of wine, doenjang (Korean fermented soybean paste), and garum (fermented fish sauce)^23–25 for the treatment and prevention of ailments. The modern conceptualization of microbial foods as “health foods” stems from Élie Metchnikoff, who hypothesized Bacillus bulgarian (now Lactobacillus delbrueckii subsp. bulgaricus), which had been isolated from Bulgarian curdled milk referred to as “yahourth,” was responsible for the increased longevity observed in the Balkan population compared to other European populations. Work on the isolation and administration of this strain established probiotics as both a concept and a commercial product (Box 2). A revitalization of at-home fermentation production²⁶ accompanied by the rise in the commercial fermented food market²⁷ has focused on marketing fermented foods for ‘health’ properties, although there is a lack of consensus on how and in what context health can be defined²⁸. Here we present the current understanding and limitation of the use of microbial foods for human health, focusing on the role of metabolites as microbe-host mediators, and highlighting areas of future research.

Callout Box 2: Fermented Foods and Probiotics

L. delbrueckii subsp. bulgaricus is one of the key starter strains used to make yogurt today²⁰², and was also the first described probiotic²⁰³, isolated from fermented dairy. The term probiotic refers to a well-defined and characterized live microorganism with demonstrated health benefits¹, which might be consumed in pure form (e.g., probiotic pill) or added to a food (e.g., probiotic yogurt). According to this definition, fermented foods made with spontaneous fermentation or through back-slopping do not fit the formal definition of probiotic unless the microbiome of the fermented food has been characterized and the strains are proven to exert health benefits. In commercial fermented food preparation, probiotic strains might be added into the yogurt once fermentation is complete, as with L. plantarum and L. rhamnosus²⁰⁴, two strains also found in fermented foods⁸⁴, or many of the Bifidobacterium spp. strains that are not associated with food fermentation, but rather were isolated from other sources such as infant feces^205,206.

While probiotics have been extensively reviewed²⁰⁷, success in trials is highly variable and appears dependent on a number of factors. A randomized, double-blind, placebo-controlled study looking at the impact of stomach acid on probiotic intervention showed that suppression of stomach acid production with proton pump inhibitors (PPI) might enhance probiotic effectiveness possibly via improved viability²⁰⁸. More recently, a randomized controlled trial looking at the effects of probiotics on metabolic syndrome identified diet as a key differentiating factor between responders and non-responders, with non-responders having higher levels of serum glucose and insulin at the end of the intervention²⁰⁹. Unexpectedly, total and added sugars, lactose, and sucrose intake was higher in the probiotic responders compared to non-responders. Further research is needed to better understand basic interactions between probiotics, diet, host gut microbiome, and host.

Current research points to metabolites and other probiotic products (sometimes commercially referred to as ‘postbiotics’) as key contributors to many of the benefits seen by probiotic consumption^18,210. While probiotics offer a characterized bacteria for a specific indication, in supplement form these products lack fermentation end products. Alternatively, fermented foods offer an extensive reservoir of potential postbiotics that can directly impact the microbiome, immunity, and the enteroendocrine system.

The extended microbiome

Metabolites produced during fermentation provide a key role in the increased safety²⁹, nutrient availability^30–41, and enhanced gustatory qualities^42–44 of fermented foods. Understanding how the microbial strains present combined with other factors, including salinity, pH, and temperature, contribute and influence metabolite production remains a necessary consideration when characterizing the final fermented product, particularly when it comes to food safety.

Enhancing food safety

Preserving food through fermentation is a reliable technique to enhance safety and stability over a prolonged period of time. During fermentation, metabolites produced during microbial competition for nutrients inhibits growth of competitors, concomitantly preventing growth of potential human pathogens. Production of various organic acids, primarily lactic or acetic acid, decrease the pH and create an inhospitable environment for food-borne pathogens like C. botulinum, L. monocytogenes, E. coli O157:H7, and S. flexneri^45–47. Secretion of antimicrobial compounds like bacteriocins, antibacterial peptides with both narrow and broad-spectrum activity, have been extensively studied in fermented food-associated lactic acid bacteria (LAB), including in kimchi, cheese, and fermented cereals^48–51, and show high specific activity against known food pathogens⁵¹. To date in the US, no cases of botulism have been reported in vegetable fermentation, mold-based fermentation, or dairy fermentation⁵², with the rare cases being home-fermented meats including beaver tail and uneviscerated fish, and tofu⁵³.

Much of the insight into the mechanisms underlying the ecology of fermented foods has come from their use as a model microbial ecosystems. One example is the identification of novel antimicrobials that mediated microbe-microbe interactions^54–56, highlighting fermented foods as a potentially rich source of novel antimicrobials for medical application. More generally, the reduced microbial diversity in fermented foods compared to the gut microbiome provides improved tractability for exploring community assembly, transkingdom interactions, strain-diversity, and microbial evolution⁵⁷.

Fermentation can also enhance food detoxification. Raw cassava, a starchy tuberous root native to South America, and a staple food for over 500 million people⁵⁸ contains high levels of neurotoxic cyanogenic glycosides⁵⁹, which can be reduced by >70% through fermentation⁶⁰, with Lactiplantibacillus plantarum and Loigolactobacillus coryniformis believed to play a major role^61,62. Consumption of unprocessed cassava can lead to development of the irreversible neurological disorder konzo, which disproportionately affects rural areas in Africa⁶³. Gaining mechanistic insight into cassava detoxification would allow for improvement in food safety, with potential application to other foods.

Increasing nutrient availability and antioxidants

Fermented foods also have a documented history of use as nutrient supplements. James Cook, the British explorer, successfully prevented and treated scurvy among his crew by instructing them to consume 2 pounds of sauerkraut each a week, estimated to provide about 150mg of ascorbic acid³⁰. However, these increased concentrations in nutrients might be specific to the microbial strains found in the ferment. Early work looking at prevention of scurvy in guinea pigs fed commercial sauerkraut found that two of the sauerkrauts successfully prevented scurvy while two other sauerkrauts did not. No difference was identified between the sauerkraut manufacturing conditions, however microbial communities that might be involved in the observed differences were not investigated for potential differences³¹. More recent studies support an increase in vitamins and nutrients in fermented foods compared to unfermented ingredients, including vitamins C, B₂, B₁₂, K, and folate^32–37,64. Additionally, release of encrypted bioactive components like polyphenol and flavonoid from tea leaves over the course of kombucha production increase antioxidant content of the drink³⁸. There is some rodent evidence pointing to host benefits following consumption of fermentation-driven enhanced nutrient bioavailability^39,40, however the chemical complexity combined with depletion of precursors (e.g., glucose) provides many confounders. Additional work is needed to isolate the effects of the wide array of relevant variables such as (i) diversity and variability in nutrients across fermented food; (ii) the fermented food microbiota at a strain-specific level; and (iii) elucidating factors, including host and microbiome genetics, that can impact nutrient absorption.

Flavor and taste

Fermentation is often practiced for its culinary use, where transformation of taste, smell, texture, and other sensory stimuli of the food during this metabolic process act in synthesis to change the perceived flavor of the food. Spoilage is often associated with ‘off-flavors’, while novel flavors not typically associated with the starting ingredient are often a desired result of the fermentation process. For example, carbohydrate fermentation can result in the production of thiazoles and furfural, associated with nutty and almond-like flavors^65,66 via sugar degradation and Maillard reaction pathways. Promotion of the Ehrlich pathway, an amino acid catabolic process, during fermentation can also lead to an increase in sulfur-containing, aromatic, and branched chain volatiles like 2-phenylethanol, associated with a rose-like odor^67,68. Other metabolites like organic acids, flavonoids and polyphenols can also contribute to the change in flavor of the final ferment^42,43. In addition, filamentous fungi can secrete hydrolases during fermentation⁶⁹, increasing simple amino acids and oligosaccharides, and thus promoting umami and sweet taste sensation.

As expression of sweet, bitter and umami sensors are not limited to the oral cavity but are expressed across the human GI tract⁷⁰ metabolites typically associated with flavor may play a role beyond the host gustatory system, particularly when coupled with nutrient absorption. Sensing of sweetness by T1R2/T1R3, a GPCR heterodimer expressed in enteroendocrine cells (EECs) in the human gut, can promote secretion of the incretin hormones glucagon like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP), as well as glucagon like peptide-2 (GLP-2)^71–73. Activation of T1R2/T1R3 is not limited to sucrose, but includes a variety of sweet tasting stimuli including fructose and lactose, sweet amino acids such as glycine and d-tryptophan, sweet proteins such as monellin and thaumatin and synthetic sweeteners such as aspartame and sucralose. Whether these ligands promote the same downstream T1R2/T1R3 mediated effects remains to be explored⁷⁴. Similarly, the GPCR T1R1/T1R3 mediates umami taste perception in the gut. The heterodimer is activated by l-amino acids such as monosodium glutamate (MSG; Sodium l-glutamate) or l-cysteine, and potentiated by inosine 5′-monophosphate (IMP) and other nucleotides. Activation of T1R1/T1R3 by MSG initiates peristaltic reflexes in rodent models, along with secretion of the pain mediating neurotransmitter CGRP (calcitonin gene-related peptide) in the colon⁷⁵. CGRP receptor antagonists and anti-CGRP ligand monoclonal antibodies have more recently been approved for treatment and prevention of migraines⁷⁶.

In humans, bitter taste perception is mediated by 25 putative members of the TAS2Rs (taste 2 receptors) GPCR family. Use of the bitterant caffeine has been shown to promote gastric acid secretion in the stomach through the TAS2R43 receptor⁷⁷, as well as lead to gut-derived serotonin secretion. Matured hop bitter acids (MHBA), often added in the production of beer, have been shown to increase production of cholecystokinin (CCK) production by EECs⁷⁸, which plays a role in anorexigenic signaling and gastric emptying⁷⁹. Interestingly, olfactory receptors (ORs), a class of GPCRs typically found in the olfactory mucosa have also been identified in the human gut. In purified form, the aromatic compound citronellal, also a volatile product of yeast geraniol fermentation in beer and wine production^80–82 with an intense lemon scent, has been shown to stimulate GLP-1 secretion in both a human EEC line, and mouse tissue across the small intestine⁸³. These examples illustrate the ability of molecules that trigger taste perception to impact host physiology. With growing interest in fermentation as a source of novel flavors⁴⁴, flavor-associated metabolites could impact important biological processes such as secretion of gut-derived peptides. Greater mechanistic insight into the roles metabolites produced during fermentation play both in perception of flavor, and host biology, is an exciting area for exploration- particularly focusing on how flavor-associated metabolites can modulate EECs (discussed later). In addition, inclusion of taste-preference and taste perception surveys when conducting clinical fermented food research would aid in translatability and facilitate broader research into taste-associated compounds in food fermentation.

Food Fermentation: Metabolite-host interactions

Consumption of a fermented food includes potential benefits beyond increased nutrient availability and food safety^35,41,84. In the gut, microbiome derived metabolites such as short-chain fatty acids (SCFAs) and secondary bile acids signal to the host through a variety of receptors, including transcription factors and GPCRs playing a role in host incretin secretion, energy expenditure, and immune response^85–89. Insight into metabolite production in fermented foods will allow us to map similar metabolite-GPCR interactions, allowing deeper understanding of the role fermented food-associated microbes play on the host.

Metabolites and intestinal receptors

Gut microbiota-derived nicotinic acid⁹⁰, butyrate, and β-hydroxybutyrate produced in a keto-acidosis state⁹¹ can activate the hydroxycarboxylic acid receptor 2 (HCA₂), expressed in a number of cell types, including the gut epithelium⁹². Activation has been shown to have downstream anti-inflammatory effects^93,94, while chronic inflammation has been shown to downregulate HCAR2 expression^95,96. HCA₂ activation has also been shown to play a role in increasing bone density in weanling mice⁹⁷, and suppresses NF-κB activation in colonic cell lines and in mouse colonic tissue⁹². Additional metabolites, including hippuric acid (HA) and 3-(3-hydroxyphenyl) propionic acid (3–3-PPA)activate HCA₂⁹⁸. HA and 3–3-PPA are both products of hydroxycinnamate reduction in L. plantarum⁹⁹, commonly found in vegetable fermentation. While GPCRs are expressed across multiple cell and tissue types, their expression on immune and EECs makes them of particular interest in understanding metabolite-host interaction in metabolic and immune disorders.

One of the indications of this fermented food metabolite-host relationship is the hydroxycarboxylic acid receptor 3 (HCA₃). Described by Peters et al. 2019, while most mammals express HCA₁ and HCA₂, only great apes (including humans) and siamang have an active HCA₃, with humans exhibiting strongest activation. D-phenyllactic acid (D-PLA), is the only known HCA₃ ligand^100,101 and is produced by LAB. Measured in high concentrations in sauerkraut and up to 12.0–21.1 μg/ml in kimchi¹⁰², it is elevated in the plasma following sauerkraut consumption¹⁰³. Lack of adequate animal models and currently no known HCA₃ inhibitor has remained a challenge to the understanding of the biological relevance of HCA₃¹⁰⁴. However, HCA₃ expression on human innate immune cells and adipose tissue lend to the hypothesis that HCA₃ activation plays a role in host immuno-metabolism and energy storage¹⁰³. The extent to which consumption of other lactic acid based ferments (including yogurt) can activate this receptor remains to be explored¹⁰⁵.

Examining the effects of two primary fermented food metabolites, lactic and acetic acid, provides insight into how food fermentation can impact the host immune system. When produced by the host or gut microbiota, lactic acid activation of macrophage GPR81 reduces inflammatory responses in the colon^106,107. Oral administration of lactic acid increases microbiota-dependent regulatory T-cells in the small intestine, promoting immune tolerance¹⁰⁸. Gut microbiota-derived acetate increases colonic IgA production and alters the capacity of the IgA pool to bind to specific members of the microbial community¹⁰⁹. Effects of dietary acetate have been primarily studied in disease models (Figure 1). The role of dietary acetate in healthy models remains an open area of investigation, with open questions similar to lactic acid. For both lactic and acetic acid, there is a need to (i) define the variability in organic acid production by microbial community and fermentation time point (ii) characterize the bioavailability of organic acids based on substrate type (sauerkraut compared to yogurt), and (iii) gain further mechanistic insight into downstream impact of organic acid consumption based on dosage and location of the metabolite-receptor interaction.

Figure 1: Fermented food microbes and metabolites can influence the immune system, endocrine system, and host gut microbiome.

Consumption of fermented food introduces microbes and their metabolites, along with nutrients from the food, to the gut environment. These fermented food components can stimulate the small intestinal mucosal immune system and enteroendocrine system directly, or interact with the host microbiome. (A) Metabolite-host endocrine interaction. Short chain fatty acids (SCFAs) are products of microbial metabolism of undigested carbohydrates studied. The relevance of these molecules to gut biology has been studied largely in the context of gut microbiome production. For example, butyrate and propionate produced by gut microbes can stimulate PYY and GLP-1 secretion^188,189. In addition, butyrate and propionate have been shown to increase expression of the umami taste receptors TASR1/TASR3 in FFA2/3 co-activation dependent mechanism in enteroendocrine cells, altering their nutrient sensitivity¹⁹⁰. Following fermentation, the concentration of SCFAs known to be major fermentation products of the gut microbiome (acetate, butyrate and propionate) increases in a variety of fermented foods, including yogurt, kombucha, and carrot fermentation¹⁹¹, although more research is needed for appropriate quantification following fermentation. In mouse models, dietary SCFAs have been shown to protect against HFD-induced obesity, insulin resistance and impaired glucose tolerance in mouse models through FFA3 activity, although the role of PYY and GLP-1 remains to be understood¹¹³ (B) Metabolite-gut microbiome interaction. Dietary tryptophan, an essential amino acid found in both animal and plant proteins is catabolized by the gut microbiome, leading to production of a variety of indole-containing molecules including indolepropionic acid (IPA), which can reach plasma concentrations of about ~100 μM, falling into the range of typical drug dose^192,193. IPA has been implicated in several aspects of host biology including regeneration and functional recovery of sensory neurons, suppression of intestinal inflammation and decreases gut barrier dysfunction through PXR activation in mouse models^192,194,195. Lacticaseibacillus casei and Lactobacillus helveticus, two species associated with cheese production have been studied for their role in production of indole acetic acid (IAA) and resulting in off flavors¹⁹⁶ Tryptophan abundance increases during fermentation of protein-rich substrates like meju¹⁹⁷, a fermented soybean paste. Whether tryptophan is catabolized by fermentation-associated microbes and how relevant production is to the host and host microbiome remains to be explored. (C) Metabolite-host immune interaction. Dietary acetate affects colonic Treg homeostasis, suppresses allergic airway disease in mouse models, and mediates microbial IgA response through a variety of mechanisms^7,198,199. Additionally, acetate administered in drinking water to mice with DSS-induced colitis can signal through GPR43, regulating inflammation²⁰⁰. Acetate in these studies ranged from 150mM-300mM concentrations. Observed range of acetate in kombucha is around 50mM, while vinegar can contain around 700mM acetate²⁰¹. How dietary acetic acid consumption method, concentration, and dosage impacts the human immune system remains an open area of exploration.

Additional metabolites previously identified in fermented foods, including the host immune modulating metabolite D-PLA, have shown to reach μM plasma concentrations following ingestion of sauerkraut¹⁰³. A randomized human dietary intervention showed consumption of fermented foods lead to decreased markers of inflammation¹¹⁰, while promoting increased gut microbiome diversity when compared to a high fiber consuming cohort. In this study, specific serum metabolites, which may be derived from diet, microbiome, or host, (or combinations thereof) correlated with specific aspects of the immune response. Expanded understanding of fermented food metabolites, their absorption, and interaction with human biology is the next step in characterizing ferments-immune interplay (Figure 2).

Figure 2: Metabolic paradigms of fermented-food derived metabolites.

While the direct impact of fermented food metabolites on host health needs to be elucidated, metabolic paradigms allow for the identification of potential mechanistic mediators. For the majority of these examples, understanding whether consumption of a fermented food containing the metabolites has any downstream effect remains an important area to address. (A) Lactate can reach concentrations of about 50mM during food fermentation, produced by lactic acid bacterial metabolism^211,212. Both D-lactate and L-lactate are known GPR81 ligands, with an EC50 of 1–5mM, depending on cell type^213–215. The GPCR GPR81 is present on ghrelin secreting enteroendocrine cells in the stomach¹²⁰, on immune cells^{106,216–219} and on adipose tissues^220,221, where it can play a number of downstream roles. Secretion of ghrelin, an enteroendocrine hormone primarily characterized by its role in stimulating hunger and food intake, is inhibited following GPR81 activation by lactate¹²⁰. The direct role of fermented food-derived lactate on ghrelin secretion and hunger cues remains to be characterized. Lactate sensing by immune cell types via GPR81 has been previously reviewed²¹⁸, sometimes having opposite effects, possibly due to activation of additional lactate receptors, cellular uptake, and/or metabolic effects. Orally administered lactate increases small intestinal T-regulatory cells, promoting a tolerogenic intestinal immune environment in a mouse model¹⁰⁸. On adipose tissue, lactate-GPR81 activation can suppress lipolysis²¹⁵. GPR81 is also found in numerous neuronal cell types in the brain^222,223, tumor tissue, and other intestinal cell types²²⁴. (B) During food fermentation, acetic acid bacteria can metabolize glucose or ethanol to produce acetic acid². As a GPR41 and GPR43 ligand, acetic acid can activate enteroendocrine cells in the colon, stimulating PYY and GLP-1²⁹, two enteroendocrine hormones important for satiety and food intake regulation, as well as insulin secretion. GPR43 activation can also mediate the protective effect of acetate on IBD^29,41. In addition, colonic microbially-derived acetate has been shown to increase colonic IgA production¹⁰⁹, and play a role in microglial maturation²²⁵. (C) More recently, additional fermented food-derived metabolites have been potential players in metabolite-host signaling and health. Hippuric acid (HA) and 3-(3-hydroxyphenyl) propionic acid (3–3-PPA), two HCA₂ agonists⁹⁸, are a product of hydroxycinnamate reduction in L. plantarum⁹⁹. HCA₂, which has a number of other agonists, has been shown to play a role in regulation of homeostasis, nutrient sensing and inflammation^94,226. Whether fermented food-derived HA and 3–3-PPA are sufficient to interact with HCA₂ and lead to a clinically significant health benefit remains to be understood. (D) Related to HCA₂ is HCA₃ (hydroxycarboxylic acid receptor 3), with D-phenyllactic acid (D-PLA) being the only known ligand to date¹⁰³. With significant levels measured in vegetable ferments like sauerkraut, kimchi, and fermented carrots,^102,103,105, and identified in plasma of individuals following consumption. While HCA₃ expression on human innate immune cells and adipose tissue indicate a role for HCA₃ on immuno-metabolism¹⁰³, lack of appropriate animal models make mechanistic work a challenge. Characterizing and quantifying microbial metabolites consumed and produced during food fermentation and performing the necessary mechanistic experiments to understand clinical benefits, particularly in human subjects, remains an essential validation to any purported health claims around fermented foods. More recently, two additional aryl-lactates, 4-hydroxyphenyllactic acid (4-HPLA), and indole-3-lactic acid (ILA), have also been quantified in fermented foods, derived from aromatic amino acids in high abundance by L. plantarum in both dairy and vegetable lactic acid fermentation. Concentrations of all three aryl-lactates, D-PLA, 4-HPLA, and ILA increased over storage condition in yogurt¹⁰⁵. Aryl-acetates are ligands of the aryl hydrocarbon receptor (AhR), a toxicity mediator important to immune homeostasis, in a dose-dependent manner^227,228. In addition, AhR plays an important role in barrier function and neuronal regeneration²²⁹. In mice, enteric AhR expression is microbiome dependent²³⁰, highlighting the importance of microbially derived metabolites in AhR expression and activation. (E) In miso, a soybean product fermented with Aspergillus oryzae, succinic acid is produced following carbohydrate metabolism²³¹, contributing to depth of flavor of the final miso product^232,233. Succinic acid is the endogenous ligand of GPR91, with an EC₅₀ ranging from 20–50 μM^234,235. In miso and nuruk, succinic acid has been reported to be around 300–450μM²³³. As native succinate production is largely understood within a framework of metabolic stress, GPR91 activation has interesting downstream effects dependent on cell-type activation, including collagen type I production²³⁶, increased inflammatory signaling capacity^237,238, lipolysis inhibition and increased thermogenic activity²³⁹. However, in the small intestine succinate has been shown to decrease inflammation, highlighting the complexities of site-specific activation^240–242.

EECs, are hormone-producing cells accounting for about 1% of the gut epithelium. Acting at the interface between luminal content and host signaling, EECs produce more than 20 hormones in response to nutrient absorption. These hormones can signal systemically to the pancreas, as well as the brain via vagal afferent neurons¹¹¹. While EECs can be stimulated directly by nutrients consumed, colonic EECs are also activated by bacterial SCFAs, notably butyrate and propionate, resulting from fiber fermentation. EEC-type distribution changes across the GI tract, with hormones like ghrelin and GIP primarily found in the upper small intestine, and GLP-1 and peptide YY (PYY) primarily found in the colon¹⁶ resulting in region specific effects^112–115. Specific changes in gut colonization status, such as the absence of a microbiota (germ-free mice) or administration of a probiotic (Limosilactobacillus reuteri), have shown to increase GLP-1 levels, highlighting a role for microbial signaling in incretin regulations (reviewed in Arora et al. 2021). Recently, Akkermansia muciniphila, has been shown to secrete a GLP-1 inducing metabolite resulting in a reduction of a high-fat diet induced metabolic defects in mice¹¹⁶. In conventional mice, PYY-positive EEC cell numbers increase in the colon in the presence of SCFAs via a FFA2-dependent pathway, reflected by an increase in circulating PYY, an enteroendocrine anorectic hormone¹¹⁷. In humans, a study in individuals with diabetes consuming fermentable fibers for 84 days, with hemoglobin A1c (HbA1c) as primary endpoint, showed that those in the fiber arm resulted in significant reduction of HbA1c and fasting blood glucose at the end of the study. The authors found this was due to an increase in butyrate-producing bacteria and measured increases in butyrate levels resulting in significantly increased blood levels of GLP-1 and PYY over time¹¹⁸. Another human study found that direct delivery of the SCFA propionate bound to inulin successfully delivered the propionate to the colon, resulting in significantly increased PYY and GLP-1 240 minutes after consumption of the oral propionate. Six-months of daily supplementation resulted in attenuated weight gain and reduced intrahepatic lipids¹¹⁹. Just as metabolites produced by the gut microbiota can influence host biology, metabolites produced during food fermentation can have similar interactions. The diversity of metabolites, many of which are likely to be distinct from those produced by gut-resident microbes, and interactions with microbes and host receptors in the small intestine likely produce an array of biological effects remaining to be defined.

Ghrelin secreting EECs also express GPR81, where lactate binding inhibits secretion of this “hunger hormone”¹²⁰. Unlike GLP-1-producing cells which are primarily located in the colon, ghrelin-producing cells are primarily located in the stomach and duodenum, where they are more likely to interact with fermented food-derived lactic acid, with downstream effects on hunger signaling and metabolism. Additionally, mapping fermented food metabolites to GPCR receptors, particularly the number of orphan GPCRs¹²¹, would allow for potential identification of novel GPCR-ligand pairings, expanding the development of a mechanistic framework of microbe-host interactions² and allowing for a targeted exploration of downstream effects.

Fermented Foods in the Human Disease

The rise in health claims about fermented foods is not matched by an increase in clinical evidence. The few current studies are primarily focused on assessing the impact of fermented food on healthy individuals^50,110, with few studies looking at the role of fermented foods in patients with health conditions. Early data from irritable bowel syndrome (IBS)^122,123 and metabolic diseases^124–127 (including T2DM and obesity) offer examples of how fermented foods can improve human health through altering nutrient accessibility and impacting the human microbiome. These positive outcomes highlight the need for a greater understanding of the fermented food metabolome landscape to pursue molecular mechanistic mediators.

Meta-analyses of observational studies support a role of fermented dairy intake for the decreased risk of T2DM and cardiovascular disease^124–127. However, human clinical trials looking at the risk and prevention of these diseases in the context of fermented food consumption are lacking, with the strongest current evidence coming from animal models¹²⁸. A recent study on yogurt consumption in obese mouse models identified branched chain hydroxy acids (BCHA) produced during LAB driven yogurt fermentation can supplement host BCHA and improve metabolic parameters, including improved glucose tolerance and resistance to high-fat, high-sucrose diet induced obesity. The effects appear to be driven by production of three hyodeoxycholic acids, alpha-hydroxyisocaproate, 2-hydroxy-3-methylvalerate, and alpha-hydroxyisovalerate, produced by the gut microbiome¹²⁹.

Kimchi consumption has also been studied for its impact on health, but few clinical trials have looked at the impact of consumption on metabolic disease. A small crossover clinical trial in obese patients reported significant decreases in body fat and improved metabolic markers¹³⁰. A follow-up crossover clinical trial followed prediabetic participants over the course of a 16-week period, reporting a decrease in insulin resistance and increased insulin sensitivity during the period of 10-day fermented kimchi consumption compared to 1-day fermented (defined as ‘fresh’) kimchi¹³¹. Insight into how fermentation-derived metabolites impact secretion of incretin hormones like GLP-1 is an important step in understanding the mechanistic implications of fermented food consumption. Incorporating measurements of the foods’ differences in chemical composition, patient incretin hormones, insulin and blood glucose in future clinical trials would greatly aid the understanding and translatability of laboratory work on fermented food metabolite induced responses. There remains a clear need for randomized human clinical trials and mechanistic interrogation into how the fermented food microbiome and its metabolites interact with the host including the resident microbial landscape.

While exploring fermented foods for health benefits remains an exciting frontier, there are examples where metabolites produced during fermentation can lead to negative health consequences in sensitive health populations. Biogenic amines, amino acid derivatives produced during fermentation which can include tyramine, histamine, cadaverine, and putrescine, with tyramine and histamine being of particular interest. Even small amounts of dietary tyramine (8–10mg), can lead to hypertensive episodes in the presence of the antidepressant class monoamine oxidase inhibitors¹³² with occasional fatalities reported¹³³. Dietary tyramine is increased in ferments like fish sauce, sauerkraut, and aged cheeses, but abundance can vary widely depending on food and t fermentation time¹³⁴, thus patient dietary recommendations need to consider sensitivity, food type, serving size, and drug interactions. While tyramine production is enhanced by increased sodium content in the ferment, a variety of approaches have been applied to limit tyramine production during fermentation. For example, successful tyramine reduction was achieved with the addition of tyramine- or tyrosine-oxidizing LAB in a micro-cheese model¹³⁵.

Similarly, certain LAB associated with food fermentation have been described as histamine-producers, including Lentilactobacillus parabuchneri, Lentilactobacillus buchneri, and Oenococcus oeni^136–139, primarily with proteinaceous substrates like in dairy and meat ferments. While recommendations for a low-histamine diet show high efficacy for histamine hypersensitivity, fermented foods show a wide range of histamine abundance, with length of fermentation time, storage, and cooking method all influencing histamine abundance at ingestion¹⁴⁰. Mechanistic understanding of histamine production, characterization of extrinsic factors impacting production, and rapid biogenic amine detection methods are all necessary both from a clinical and consumer perspective. It is likely that extending such detection methods to other biomolecules will aid in quality control and inform health claims particularly as potential benefits of fermented foods on other physiological aspects are explored further.

Previous studies have shown associations of fermented food or pickled vegetables (terms used interchangeably in the studies; see Box 1) intake and esophageal and stomach cancer^141–144. These concerns are related to retrospective studies where the quality and contamination of the fermented foods was not tested and many confounders exist. For example, one widely cited study published in 1980 focused on the prevalence of esophageal cancer in Northern Chinese populations¹⁴¹, with data from 1959–1970. The study found a positive correlation between “pickled vegetable” consumption (likely fermented based on literature description) and esophageal cancer risk, but due to long fermentation storage period, the authors note that much of the food became covered with a white mold. This association was also seen in regions with higher consumption of Laozao, a fermented sweet grain dish. However, samples were found to have high fungal contamination, including Aspergillus flavus, associated with increased production of carcinogenic compounds^145–149. In addition, enzymatic activity of other molds in the sample is hypothesized to allow for synthesis of N-nitroso compounds, associated with higher cancer risk^150,151. With the exception of dairy fermented products that are primarily produced using well defined cultures, vegetable ferments and kombuchas are largely driven by spontaneous fermentation or uncharacterized microbial communities, respectively. However, fermented food found at grocery stores are required to meet HACCP compliance and go through rigorous safety testing before being commercially available, significantly decreasing the risk of contamination by toxin producing microbes. Conversely, several studies have shown fermented food intake is associated with no increased risk or a decreased risk from certain cancers and diabetes^{124,126,127,152–156}. Future prospective studies with better characterization and documentation of types of fermented foods consumed are required to understand if such concerns are valid.

Food Fermentation: Microbe-microbe interactions

Composition of an individual’s gut microbiome is influenced by host genetics and lifestyle, including diet, medical practices, and environment. Studies comparing gut microbial composition across different lifestyles note decreasing gut microbiome diversity in industrialized populations, believed to be driven by factors associatefd with an industrialized lifestyle including antibiotic overuse, high-fat and high-sugar diet, and decreases in breastfeeding¹⁵⁷. Not only does immigration to an industrialized environment from an non-industrialized environment lead to a decrease in microbial diversity¹⁵⁸, but this decrease in microbial diversity compounds over time and over the course of generations^157–160. Within industrialized populations, more diverse gut microbiomes are a conserved metric associated with resistance to Western-associate diseases, such as T2DM, heart disease, and colorectal cancer^161–163. Understanding how diet can increase an individual’s microbiome diversity is an active topic of research in the gut microbiome field. Fermented food consumption appears to be a promising tool in gut microbiome reconstruction.

Fermented Foods and the Human Gut Microbiome

Dietary nutrients and host-derived metabolites, including lactate and 3-hydroxybutyric acid can impact the human gut microbiome, shifting microbial composition with downstream implications on host health from metabolic disease to immune status^13,19,114. Similarly, fermented food consumption appears to have important effects on the gut microbiome. The American Gut Project, a citizen science project, reported subtle but significant shifts in the beta-diversity and taxa of fermented food consumers compared to non-consumers across 6811 participants. Of interest, fermented food consumers had an increase in conjugated linoleic acid (CLA) producers and fecal CLA¹⁶⁴, fatty acid metabolites which activate the nuclear receptor PPAR-γ^165,166 to exert beneficial health effects on a number of conditions including IBD and T2DM.

A randomized clinical trial comparing high-fermented food consumption to high-fiber consumption showed fermented food consumption increased gut microbiome diversity compared to baseline, with alterations in the gut microbiome composition that were durable beyond the trial period¹¹⁰. Nine amplicon sequence variants (ASVs) in the Firmicutes phylum increased across the high-fermented food compared to high-fiber, including four members of the Lachnospiraceae family, two Ruminococcaceae, and one Streptococcaceae. This increase in gut microbiome diversity was not observed in the fiber-consuming groups. This trend in increased microbiome diversity has been further supported by a more recent randomized control trial comparing gut microbial composition following six weeks of fermented vegetables, pickled vegetables, or no vegetables¹⁶⁷. Not only was there an increase in gut microbiota diversity from baseline to completion of the study in the fermented vegetable group, but microbiota diversity did not change in the pickled vegetable or non-vegetable consumer group. Remaining questions include whether this increased diversity is mediated via host immune effects, live microbes in the fermented foods, metabolites alone or other macromolecules, and how durable the response is after cessation of fermented food consumption.

Interestingly, the majority of microbes responsible for the increase in diversity were not from the ferments, and were likely either acquired from the environment or present in the hosts before the intervention but below detection limits. Whether fermented foods may act as a source of microbes to increase diversity, or lead to compositional shifts that promote incorporation of transient microbes remains to be understood. Very few studies have assessed human microbiota composition before, during, and after fermented food consumption^110,167. A study comparing mother-infant strain sharing in a rural Ethiopian cohort identified two microbes from injera (Fructilactobacillus sanfranciscensis and Lactiplantibacillus xiangfangensis), a locally produced teff flour ferment, as contributors to diversity in maternal-infant cohorts¹⁶⁸. However, due to the popularity of injera consumption, whether these microbes were transient or had become incorporated into the gut microbiome could not be determined.

Comparison of LAB genomes from fermented foods to the human microbiome reveals distinct sequences supporting the idea that fermented foods are a transient source of LAB for the gut microbiome. Differences in LAB species identified across western and non-western gut microbiome populations appear to reflect lifestyle¹⁶⁹. Understanding which LABs are transient or engrafted stable members of the gut community should be aided by more investigation and improved genomic resolution. Early work on engraftment into the gut microbial community shows successful incorporation of microbes is largely dependent on method of administration, diet, absence of similar strains (i.e., open niche space), and host genetics^12,170–172. Shifts in the gut microbiome and host environment following fermented food consumption likely play an additional role in engraftment success, offering additional complexity and opportunity for targeted incorporation of missing key members. Reports of enteric pathogens and antimicrobial resistance genes within commercial fermented foods highlight the importance of vigilance in monitoring the microbes that inhabit fermented foods¹⁷³.

While current evidence shows an exciting role for fermented foods in modulation of the gut microbiome community, future clinical work would shine a light on a number of open questions: (i) understanding the role of specific fermented foods, including duration and conditions of fermentation and storage , in modulating the gut microbiome, (ii) clinical trials across a number of cohorts, including those with clinical needs, to understand the limitations in the impact on the gut microbiome community, and (iii) defining how differences in microbial load or pasteurization (or other post-fermentation processing) or cooking status might impact the gut microbial community.

Future Directions and Conclusions

While fermented foods offer potential for modulation of human health and disease through microbes and metabolites, expanded scientific understanding is needed at multiple levels. Foundational insight into the basic biology of fermented foods, such as the strains, genomes, metabolites, ecological succession and stability of microbes over time is critical information to enable pursuing connections to human health. In addition, use of tools for secondary metabolite predictions, such as antiSMASH¹⁷⁴and MicrobeMASST¹⁷⁵, and application of both targeted and untargeted metabolomics will contribute greatly to the identification of metabolites at the interface of microbe-microbe and metabolite-host interaction. Such data will aid mechanistic studies that can employ model systems such as cell culture and animal models. The field of gut microbiome science serves as a useful template for advancing such investigation, and many of the experimental approaches are directly applicable. At the same time, human studies are needed to delineate which fermented foods at what levels of intake impact what biology in which cohort or population. These studies should include well designed and carefully executed dietary interventions in healthy populations to survey changes in biology and physiology such as immune and metabolic parameters.

Additionally, clinical trials in patient cohorts targeting specific outcomes are necessary to support the incorporation of fermented food, or its components, into current medical practice; integral to medical use is the need for advanced types of quality control for the fermented food product to ensure safety and that specific bioactive components are present at required levels and no off-target effects are introduced. Whether there is a future for the prescription of medical fermented foods with certified microbial or metabolite abundances, particularly for groups with health vulnerabilities, remains an open area of exploration, but will require a number of basic mechanistic questions to be addressed.

With unhealthy food driving many common diseases in the industrialized world, incorporation of fermented foods will also serve to replace commonly consumed foods that are incompatible with long-term health (e.g., ultra-processed, high-glycemic). Fermented foods may serve as a gateway for consumers and patients to more broadly understand the power of diet to directly impact health. Several key points (Box 3) make fermented foods a candidate in catalyzing a movement toward expanded food-based health care. The ability of individuals to easily implement fermentation in their own kitchen enables economical production, engagement with food, and ability to customize to personal preferences; importantly, variation between at home and commercial production of fermented foods remains to be explored, including heterogeneity in microbe and metabolite production between commercially available fermented foods. The wide variety of fermented foods also present a wide range to meet the needs of people with different dietary restrictions and taste preferences. Indeed the molecules produced during fermentation that dictate texture and flavor are likely to also play biological roles in many cases.

Callout Box 3:

Areas to Explore in Fermented Foods and Health

	Status of Fermentation Research
General benefits of fermentation	Established  • Enhanced food safety through organic acid and antimicrobial production, decreased abundance of potential pathogens45,48–53  • Increased nutrient availability (e.g. antioxidants and vitamins)30–41  • Substrate detoxification (e.g. cassava)59–61,63 Future Area of Research  • Fermented food microbiota strain specific roles in nutrient enhancement, bioavailability, detoxification, flavor and other final fermented food properties  • Host factors affecting fermented food derived nutrient absorption   • Factors affecting microbial community development in starter and wild fermentation  • Characterization of toxin production on novel substrates, including development of at-home food fermentation testing tools  • Identification of commercially- or medically-useful fermented food products, e.g., antimicrobials to combat the rise of fungal pathogens and antibiotic resistance
Food fermentation microbiome-host interactions	Established  • Consumption associated with trends toward decreases in markers of inflammation in healthy cohorts106–110 Future Area of Research  • Identification of metabolites produced during the food fermentation process across various substrates, including use of targeted and untargeted mass spectrometry, including LC- and GC-MS for broad compounds discovery  • Assessment of live microbe-host interaction compared to metabolite-host interaction, particularly in disease phenotypes  • Characterization of metabolite-host interaction, including mechanism on host immune and enteroendocrine systems
Food fermentation microbiome-host microbe interactions	Established  • Fermented food consumption increases gut microbiome diversity110,164,168    Future Area of Research  • Identification of source of microorganisms in spontaneously fermented foods  • Role of fermented food consumption on engraftment  • Role of fermented food microbe or metabolite on gut microbial community composition and metabolism
Beyond fermented food-gut interactions	Established  • Effective models for the study of microbe-microbe interactions54–57 Future Area of Research  • Role of fermented food metabolites in gut-brain-axis signaling, particularly role in satiety and hunger signaling  • Application of fermentation for promotion of sustainability efforts176

Still, investigating the role of microbial metabolite production in a fermented food environment offers numerous intriguing yet understudied areas of research. Fermented foods lend themselves as simpler and tractable yet highly relevant models of microbe-microbe interactions compared to the human gut⁵⁷. Characterizing the fermented food metabolome and developing tools to better understand metabolite-microbe production dynamics within the fermented foods community has interesting applications for novel flavor and texture development. Transformation of unconventional substrates into familiar flavors can aid in sustainability efforts¹⁷⁶. In addition, with the growing interest in the gut-brain-axis research, the fermented food landscape offers a pool of understudied small molecules that might influence taste preference behavior^177–180, as well as novel compounds relevant to human health. Despite the exciting potential of fermented foods, research is still needed to understand the complexities.