The “efficacy paradox” of phytochemicals, low bioavailability yet significant health benefits, is associated with gut microbiota, which biotransforms dietary precursors into bioactive metabolites, enabling systemic effects. Thus, health outcomes of diet depend not just on intake “the rainbow”, but on gut microbial metabolism, redefining “you are what you eat” to “you are what your microbes make of what you eat”. The health-promoting properties of dietary phytochemicals have been extensively documented in preclinical data from in vitro and animal studies, where they exhibit potent bioactivities 1. However, when translated into human interventions, the benefits of these compounds are often inconsistent, highly variable across individuals, and sometimes disappointingly ineffective 2-4. This disconnect is often termed the “efficacy paradox”, representing a significant bottleneck in the field of nutrition, since it challenges the conventional dietary recommendations and highlights our incomplete understanding of how phytochemicals are metabolized within the human body (Figure 1). Multiple factors contribute to this paradox, including the food matrix, processing methods, dosage, host genetics, hepatic metabolism, and lifestyle. However, a rapidly growing number of evidence indicates that the gut microbiota is the most influential and personalized determinant of phytochemical's efficacy 5. In fact, the vast majority of dietary phytochemicals (up to 90%–95%) cannot be absorbed in the upper gastrointestinal tract and enter the colon, where the microbial “black box”—trillions of anaerobic microbes, act as an essential, highly personalized, and biological processing unit for phytochemical biotransformation (Figure 2A). These microbes possess a vast and diverse arsenal of metabolic enzymes (e.g., β-glycosidase, carbohydrate-active enzymes (CAZymes), esterase, reductase, decarboxylase, dehydroxylase, demethylases, and dehydrogenases) that are most absent in the human host, enabling them to deconstruct the complex dietary precursors into smaller, more lipophilic, and often more bioactive metabolites that can be readily absorbed through the colonic epithelium into circulation and exert systemic health effect 6. Consequently, the composition of individual's gut microbiota determines whether a specific phytochemical is activated into its potent form, giving rise to the concept of “metabotypes” (metabolic phenotypes), the distinct population subgroups defined by their ability to produce certain key metabolites 7. Although previous studies have comprehensively identified the biotransformation pathways of specific phytochemicals (e.g., ellagitannins to urolithins, daidzein to S-equol, and flavan-3-ols to phenyl-γ-valerolactones), the field now faces a more profound challenge—how can we translate preclinical mechanisms into personalized nutritional interventions? Despite decades of research, several critical questions remain unanswered, like (1) which steps within these pathways are supported by the microbe–enzyme–gene evidence in the human body, rather than inferences from preclinical studies; (2) what limitations exist in the current human research; (3) what is the optimal way to combine multi-omics and traditional biochemical methods to understand how correlation and causation are connected; (4) and most importantly, how can this mechanistic understanding be translated into personalized nutrition practice? This work, therefore, aims to discuss the aforementioned questions through a comprehensive framework that closely links mechanistic discoveries with practical applications. Specifically, we rigorously evaluate evidence regarding the biotransformation pathways of key phytochemicals, clearly distinguishing among definitive validation, in vitro inferences, and human trial evidence, while highlighting their respective limitations. Furthermore, we integrate current methodologies into a systematic discovery-to-validation strategy encompassing multi-omics, bioinformatics, and classical biochemistry. Finally, a concrete roadmap for developing next-generation probiotics (NGPs), precision synbiotics, and postbiotic nutraceuticals is proposed. By doing so, we can unlock the true potential of dietary phytochemicals and pioneer a new generation of truly effective functional ingredients from their gut metabolites. The journey to uncover these microbial secrets has yielded fascinating insights, transforming our understanding of how diet influences nutrition and health. In recent years, increasing work has been carried out to answer two fundamental questions involved in gut microbiota-mediated metabolism of phytochemicals. First, what are the key gut bioactive metabolites of phytochemicals, and how do the key metabolites contribute to the health benefits of phytochemical precursors? Second, how is gut microbiota-mediated metabolism and biotransformation process implemented, especially specific gut bacteria species/strains, their metabolic enzymes and encoding genes? Several well-characterized examples are elaborated and discussed below. Ellagitannins, abundant in many plant-based foods like pomegranates, berries, and walnuts, are large polyphenols with poor bioavailability. Upon reaching the colon, they are hydrolyzed to ellagic acid (EA) by gastric acid, which is then sequentially converted by gut bacteria into a series of smaller, more absorbable compounds dibenzopyran-6-one derivatives known as urolithins (Figure 2B) 8. The biological significance of this transformation is well-established. Urolithin A and Urolithin B are the major end-products of EA, and Urolithin A in particular exhibits significant health functions including antioxidant, anti-inflammatory, anti-cancer, anti-aging, cardioprotective properties, and muscle function-enhancing effects that exceed those of EA itself 8, 9. However, this microbial biotransformation is not universal. Based on the gut microbial capacity, individuals can be classified into distinct “urolithin metabotypes”, including metabotype A (produces only Urolithin A), metabotype B (produces Urolithin A, Isourolithin A, and/or Urolithin B), and metabotype 0 (produces no above urolithins at all) 10. These urolithin metabotypes are directly linked to the presence of a functional microbial “assembly line” with key dehydroxylation capabilities. Foundational work identified Gordonibacter urolithinfaciens DSM 27213T, and G. pamelaeae DSM 19378T capable of converting EA into intermediate urolithins such as Urolithin C 11. Subsequently, Ellagibacter isourolithinifaciens DSM 104140T was isolated and shown to produce isourolithin A from EA 12. However, these species alone cannot complete the pathway to Urolithin A/B, as they lack the capacity for the final dehydroxylation steps. A subsequent study revealed that certain Enterocloster species (e.g., E. asparagiformis, E. bolteae, and E. citroniae) are responsible for this final dehydroxylation step that produce the end-stage Urolithins A/B, defining each metabotype 13. The dehydroxylation mechanism, especially the challenging sequential dihydroxylation of urolithins, has been clarified recently at the enzymatic and genetic levels. Pidgeon et al. uncovered an inducible Urolithin C dehydroxylase (ucd) operon in Enterocloster species, which includes ucdC (FAD-binding subunit), ucdF (2Fe2S-binding subunit), and ucdO (molybdopterin cytosine dinucleotide (MCD)-binding subunit), and this operon encodes a novel molybdopterin-dependent enzyme complex that specifically dehydroxylates urolithins at its 9-hydroxy position 14. To move forward, Bae et al. systematically identified four urolithin dehydroxylases from Gordonibacter strains and E. isourolithinifaciens, including three catechol dehydroxylases (Eadh1, Eadh2, and Eadh3) belonging to the dimethyl sulfoxide reductase family, and a xanthine oxidase family enzyme (Ucdh) from E. bolteae 15. These enzymes target individual hydroxyl groups on the urolithins and exhibit substrate specificity. For Eadh2, and are specifically responsible for the groups at and of urolithins, it the previous that Gordonibacter can only produce Urolithin while Ellagibacter Isourolithin A the specifically on the of Urolithin which Enterocloster to produce Urolithin A 14. several gut strains have been to ellagitannins EA, like and However, the metabolism of and strains is and their biotransformation Thus, while they that capacity be more than the evidence for these strains is to in vitro therefore, whether they contribute to urolithin in the complex gut of is The that Urolithin A and Urolithin B in the but not in the highlights the critical of and the of strains from The concept of urolithin metabotypes is supported by studies in human with into metabotype A, 0 of dietary and health 10. studies with have shown that metabotype is in most individuals, metabotype 0 individuals can be converted to a 8, metabotypes have been associated with in gut microbial with Urolithin A of Gordonibacter species and distinct [8. Urolithin A has the trial with trial its ability to health and muscle in the critical remain that most human studies metabotypes to health outcomes are and no trial has by metabotypes to whether the metabolic capacity the to Urolithin A the of the operon and dehydroxylase by dietary factors host has not been and the potential of microbes to urolithin are health benefits, such as antioxidant, and have been extensively but remain of the The of S-equol, a key gut microbial of a for this in the efficacy of In fact, exhibits biological than its as by its to and it the determinant of health effects However, the capacity of is not since only of in and in many are and the are a on the specific gut bacteria studies have identified the specific in responsible for biotransformation of most of which to the family, such as and The genetic for this has been comprehensively across metabolic enzymes and their encoding for including daidzein reductase reductase and reductase with a have been identified in and bacteria like is responsible for the step to then into its which can be to by Finally, the pathway by converting into the (Figure These definitive microbe–enzyme–gene evidence, as they have been through in (e.g., and biochemical of from the strains of the family, gut microbes in the including the strains from (e.g., and (e.g., and (e.g., have been to in However, not in these bacteria on a in of species The of in these indicates that the of daidzein to is by distinct enzyme and the for and work to understand in these This microbial and is of the examples of a gut studies and have shown that the health benefits of for are more in However, the human evidence is studies on a to but the of this metabotype within individuals their is not In most not to it to regarding the of this the for to efficacy and the presence of this is variable among known species it as that species not functional This the of functional including such as and their derivatives like and as as are the most abundant in the human with and as the major dietary Although a of can be absorbed in the the with of their the 13. they are by microbes into a of intermediate metabolites known as and their (Figure Specifically, the microbial of flavan-3-ols a sequential including the of (e.g., and by microbial to acid, which is to the of like and to the and to produce and to These metabolites, as the in the gut and are to the health benefits of dietary flavan-3-ols (e.g., and to their potent bioactivities like antioxidant, anti-inflammatory, anti-cancer, and effects 13. studies identified diverse gut bacteria capable of flavan-3-ols in For and have been shown to to and acid acid bacteria such as and possess capable of For several species belonging to have been including and and these bacteria can the of and to produce such as For the critical step of strains (e.g., and belonging to have been to the of and to produce to be the only to this and it on from and of the a major study revealed DSM This can the in the of and a novel reductase Subsequently, dehydroxylation of the is by two distinct including dehydroxylase for and dehydroxylase for from Despite this the responsible for remain clearly this but the and enzymes involved have not been E. alone can complete the pathway to produce in the that a of microbes, be the gut microbial capacity to produce individuals, giving rise to distinct that in to foods Several studies have documented in the et al. the microbial metabolism of in from and identified three distinct metabotypes on However, are limitations that to be For most studies studies and a number of of metabotype and no study by metabotypes to whether capacity the to In the in vitro capacity and in systemic and it is the of to exert biological within the target the above many phytochemicals have been to be metabolized and activated by gut microbiota to exert their health such as to by to by However, gut microbiota-mediated metabolism and biotransformation mechanisms of many of them remain and are yet to be clarified in the the gut microbiota certain phytochemicals since their ability to functional groups cannot be Bae et al. that Gordonibacter pamelaeae can acid, an acid, into an by its catechol dehydroxylase This is with two previous classical metabolism can be by gut bacteria and E. and and the can be and by gut bacteria E. it is complex for the health of gut microbiota-mediated To systematically these metabolic pathways and move from correlation to nutritional has a that multi-omics and with biochemical (Figure The of a specific microbial for phytochemical biotransformation is the step to understand on traditional methods, highly effective for functional are and to the vast majority of gut microbial The of multi-omics has our capacity for systematically the metabolic potential of gut microbiota, enabling to from discoveries to of microbial functions 6. This current multi-omics and into a discovery-to-validation clearly how these can be to from correlation to The aims to phytochemical with microbial and host and can in the gut microbiota and its metabolic dietary This the of potential and their associated capable of phytochemicals. a genetic of the microbial is and is their potential of can be to microbial and that for particular metabolic A study by et al. data from to systematically the of This revealed their presence within the family and identified novel species capable of is a functional to in and with the of microbial metabolites that the intake of phytochemicals, they are can novel and metabolites, while of specific compounds of can be through For three urolithin metabotypes and have been defined by through the of the of Urolithin A, Isourolithin A, and Urolithin B in human 10. a critical of by which microbial are in to a dietary they are This is in the of a novel For can be to in to specific dietary The key of this is its ability to that are and those that are activated to metabolism specific a with for the processing of data from The of (e.g., and and like is in the generation of These specific microbes, their and and the of metabolites. has the capacity to to microbial metabolic capabilities. of can to the generation of on microbial The aims to these into specific is critical to the from the to functional genetic is an that not this by into an host, such as E. and for specific This is effective in involved in the of is to that this to the of new enzymes from human gut The key is that functional directly links a to a genetic of the The field of has a in the of which has to significant in The of the of metabolic strains (e.g., with those that do not possess this the of that are the and are For the a of across species of as by et al. through the of and evidence was for its to and to for with specific of in For et al. have a that the of the human gut to biotransformation of foods and These are of particular in the of enzymes from only by have been shown to make significant to the of microbial associated with phytochemical The of the operon was not through but rather through and of from individuals with defined urolithin by to in Enterocloster species The which encodes a molybdopterin-dependent enzyme be to in but it as a through the of and This the that systematic of multi-omics and can the of novel microbial functions and mechanisms for subsequent biochemical In many the process through multi-omics and is a To we to the of the for This target and in vitro biotransformation validation, and validation, and functional First, the strains be isolated from human and their metabolic can be Although can the functional it cannot that a specific is responsible for the biotransformation in is of by can be to target strains from This with the phytochemical as a the of specific to for particular For the of G. urolithinfaciens and E. isourolithinifaciens was through of from individuals with defined urolithin ellagic acid as a key substrate a is its metabolic be in anaerobic The be with the phytochemical with like and to the target by the This step the evidence that the alone can the the is specific microbial enzymes and encoding be and genetic of the in a host is to enzyme in it is challenging for the genetic of anaerobic bacteria and anaerobic specific enzymes responsible for biotransformation a of including and a is it be into a host (e.g., E. where the is and with the enzyme and its substrate are then to the and to its and substrate specificity. These have been to enzymes involved in the For et al. these methods to and from E. the into and dihydroxylation In the host, are genetic the the metabolic and the a this in evidence for the This strategy was in the of the operon in Enterocloster Pidgeon et al. that of the operon the of Urolithin A from Urolithin and definitive that this operon is critical to the final dehydroxylation step 14. Finally, the is in animal to and its (e.g., and its of its in health. to to animal it is to that the within complex human This can be by phytochemical and with to for effects. Furthermore, the for in is the of with defined a particular microbial can a specific phytochemical and the is within this the evidence that the are capable of biotransformation within the Finally, the of the identified mechanisms can be by to human a particular enzyme is as the of a metabolic (e.g., urolithin metabotype A), then its presence in the population be associated with that these fundamental is not an the way for a in how we nutrition, and food The in our from the dietary to the bioactive microbial which is by four key as discussed below. The concept of probiotics has the two from and often to highly applications. to as are defined as well-characterized strains from the human gut microbiota, and these strains are for the of specific traditional probiotics (e.g., and are on the of their metabolic their with the host, and their ability to in the gut them capable of specific metabolic within an individual's gut probiotics a of health benefits, but their of is often and it to outcomes in specific By the of has a clearly defined For E. has been as an to its ability to produce the potent from of the defined biochemical the efficacy of can be specific metabolic (e.g., Urolithin A enabling a precision The for such the of Although not a phytochemical this has been associated with metabolic health and has preclinical and has been with and efficacy in and This particular that can be as on the of Despite the potential of these novel challenges remain in them from the to The key is the which the ability of an to in the which is highly variable and on many factors such as microbiota, dietary and the For a be not but it be to produce the as by the Urolithin A studies 14. to such as with a specific are the of In the complex in human including the of and bacteria such as and that depend on the microbial is many to that and (e.g., and Thus, an be a comprehensive including to out factors and studies in animal For the of through that such preclinical can the way for for strains not as and The efficacy of an individuals to genetic and gut individuals exhibit an to a while it which the for studies and the of diverse in The of defined microbial rather than as in the of with a the of by the of the Although can metabolic their faces practical including host and the for metabolic as of specific and their and precision compounds to specific with metabolic to the identified can be to an individual's microbiota and the truly personalized The of is on two key of a to their to and metabolic the of for the biotransformation metabolic For a for Urolithin A combine strains of Gordonibacter and Enterocloster strains with in evidence the of Enterocloster and of Urolithin A in capacity that such a effective an combine and enabling the substrate to be at the and where the is For individuals microbial functions are at a precision a and more to bacteria The is to dietary compounds that specifically the and of the target the microbial For certain polyphenols (e.g., and have been shown to the of in and which is by a gut as an polyphenols more substrate for like the of phytochemical can be to Gordonibacter Enterocloster in individuals possess these microbes but exhibit functional levels. This the and of the target To translate into of be studies the nutritional of key strains and their For by which polyphenols specifically the of Gordonibacter Enterocloster of in individuals the effects of substrate and the target in and while and the of effect Furthermore, studies are to the of metabolic and their with and enabling more of which from specific Finally, with microbiota to human metabolic can the effects of specific on microbial and metabolic The most way to is to microbial which can be through the of the bioactive metabolites of phytochemical By through and compounds like Urolithin A S-equol, we can a This a for of their gut microbiota The of to the gut microbial metabolic For by and and in vitro and in validation, it has been that it is not the ellagitannins but Urolithin A that the health benefits a is identified as the key bioactive its be This can be through specific microbial strains their by to The of Urolithin A this its as the of a process enabling its preclinical and A postbiotic pathway preclinical studies are for and the for human in the and For a trial that of of Urolithin A for was and in and in and This including mechanistic preclinical and a for postbiotic such as Although is a for as a in its evidence is The of Urolithin A has in developing gut metabolites as including and For the be by clearly the bioactive preclinical to evaluate efficacy and and gut microbiota However, are such as the for data and the of that bioavailability. Furthermore, this field into whether postbiotic such as Urolithin A and S-equol, can benefits, the of microbial metabolites in complex The discussed synbiotics, and to specific microbial functions their metabolic to the human these microbial metabolic in vitro to phytochemicals within a food This often termed precision traditional food with and process the of probiotics microbes to the precision as to produce target compounds within for in functional dietary This several that it the to for in gut microbial metabolism, for and to The of for precision directly the mechanistic of and than a more practical is to them in with traditional strains to functional foods with health benefits, as not the For daidzein and an such as to a and a with This of concept several practical challenges that be First, like Gordonibacter are to the and of food the of the traditional a anaerobic enabling the to and complete their Second, the of these gut bacteria into food be rigorously including to out factors as discussed of metabolites (e.g., and Urolithin that the remain strategy is to these challenges by the key such as the pathway enzymes urolithin into like E. These strains have been for and can be to of target This has been for in and for Urolithin A enzymes in E. the way for of these bioactive metabolites. the of with traditional and genetic is in probiotics to for with and specific such as to the target low the food This no genetic and can strains that are to the For acid bacteria in have been to and acid, and these in a to the of traditional and within complex food their and metabolic genetic is not to nutritional but rather to as a to the of functional For a Urolithin A as a while strains a of The of on factors such as and the at which outcomes can be including and pathway to the of precision as our understanding of gut microbial metabolic mechanisms the of compounds that can be to enabling the of a new generation of foods to health The field of nutrition is at a The of is giving way to a more and where our gut microbiota as the critical diet and health. The of phytochemical efficacy in is not a but a to a understanding of our profound with our gut By multi-omics, bioinformatics, and we are to the microbial metabolic that the health benefits of our The is that we in this fundamental to move from a examples to a comprehensive of phytochemical metabolism in our In doing so, we not only the efficacy but the for a new of personalized be to and functional foods to our microbial the profound and health potential within our The of functional ingredients not in the but more in the potent by our gut and have the final and it for This work was supported by the and the of for not to work to limitations. the position of for and the no of involved in this the of this work, we and to and the as as to and these we and the as translated and be in the
Yang et al. (Sat,) studied this question.