If previous studies ¹ are to be believed, there may be just one step from gut microbiota to skeletal health: certain gut microorganisms are thought to increase the production of T cells, which in turn stimulate the production of immune mediators and inflammatory cytokines, promoting osteoclastogenesis and bone loss in mice. Other studies point to mechanistic links involving the production of microbial short-chain fatty acids (SCFAs) and the metabolism of dietary components involved in bone metabolism (vitamins K, D, and complex polysaccharides). This research field even has a specific name: osteomicrobiology. The field has seen very few clinical studies, but an American team recently published one.

Two separate cohorts, one suspect in common

This research was based on the cohorts from two observational studies: 831 elderly men (mean age 84.2) from the Osteoporosis in Men Study (MrOS) ² and 1,227 younger men and women (mean age 55.2) from the Framingham Heart Study (FHS) ³. An analysis of the data identified 37 microbial genera that appear to be involved in the FHS study (DTU089, Marvinbryantia, Blautia, and Akkermansia were negatively associated with bone density, while Turicibacterand Victivallis were positively associated) and 4 genera in the MrOS study (negative associations with Methanobrevibacter and DTU089 and positive associations with Lachnospiraceae NK4A136).

Thus, despite the difference between the two cohorts in terms of height, sex, and age, a common bacterium was associated with lower bone density in both cohorts: DTU089. DTU089 is known to be more abundant in those with low levels of physical activity and very limited protein intake, two factors unfavorable to bone health.

Meta-analysis

The researchers pooled the two cohorts to perform a meta-analysis. The results: a higher abundance of Akkermansia and DTU089 was associated with a less dense radius and tibia; conversely, a higher abundance of the Lachnospiraceae NK4A136 group and Faecalibacterium was associated with higher bone density.

The researchers also identified eight metabolic pathways associated with bone measurements, the most important of which involved the histidine, purine, and pyrimidine biosynthesis pathway. Previous experiments in mice had suggested a disturbance in purine metabolism in osteoporosis.

These results remain preliminary and require further studies to better understand the mechanisms by which certain bacteria can modify skeletal integrity, but they do support initial preclinical findings. Above all, they give greater hope that we may one day be able to modulate the gut microbiota and better protect bone health.

Do you think that analyses of stool samples may be soon added to tests designed to identify individuals with early Alzheimer’s disease in order to orientate them to appropriate treatments more rapidly?

After reading this article the first question which logically comes to mind is to ask oneself if the analysis of the microbiota could be proposed to identify those individuals with early stage or preclinical AD. From a neurologist’s point of view, the response is rather negative. This is because current data, both for symptomatic AD and for preclinical AD, have failed to identify a specific “standard” microbiota, which may distinguish these cases from a control population using routine stool analysis. Moreover, there are now markers of AD, reliable even at a preclinical stage, easily utilisable in a clinical context. Leaving aside PET imaging which is not available in all centres and analysis of the cerebrospinal fluid which involves a lumbar puncture which can be considered invasive, it is now possible to detect modifications in the expression and/or phosphorylation of certain proteins implicated in the neurodegenerative process in the plasma, i.e. in a simple blood sample, for cases of symptomatic AD as well as at the preclinical stage [3].

Would you consider sharing this publication with your patients to explain the relationship between the gut microbiota and the brain, in order to reinforce the key role played by the gut microbiota in human health?

To end on a more positive note, there is, however, no doubt that this article, by showing for the first time a modification to the composition of the microbiota in preclinical AD, provides new evidence that the microbiota may be involved in the development of AD, and moreover, at an early stage. In this context, its summary and a simplified version may be proposed to the general public or to some patients to emphasise the important role of the microbiota in health. Nevertheless, independent confirmation of these results by other teams will be necessary in the future.

Recurrent miscarriage: a case study on vaginal microbiota transplantation (VMT) 

^{Wrønding T, Vomstein K, Bosma EF et al. Antibiotic-free vaginal microbiota transplant with donor engraftment, dysbiosis resolution and live birth after recurrent pregnancy loss: a proof of concept case study. EClinicalMedicine 2023; 61.}

This case has clear scientific interest. A woman with a history of late pregnancy losses and severe vaginal dysbiosis undergoes a vaginal microbiota transplant (VMT). Five months later, she becomes pregnant, with a healthy vaginal flora, and subsequently gives birth to a fullterm baby. However, the limitations of the study should be noted: it involved a single patient diagnosed with antiphospholipid antibody syndrome (APLS, a thrombophilia associated with miscarriages).

Furthermore, therapy received for this APLS during her last pregnancy may explain the results in part or in full. Prior to the transplant, the 30-year-old mother of one had suffered a series of pregnancy losses, some of them late (gestational week 27 in 2019, and weeks 17 and 23 in 2020). For the previous nine years, she had complained of itching and vaginal discharge, which worsened during her pregnancy attempts, despite treatment. With good reason: in July 2021, her vaginal microbiota showed a very strong dysbiosis, with a 91.3% dominance of Gardnerella spp. According to a compassionate use protocol, a VMT from a healthy donor was performed in September 2021, on day 10 of her menstrual cycle, without antibiotic pretreatment. While orally or vaginally administered antibiotics (metronidazole or clindamycin) can have a cure rate for vaginal dysbiosis of 80%-90% one month after treatment, the recurrence rate can be as high as 60% after one year, with the added risk of resistance.The VMT rapidly corrected the dysbiosis and its symptoms and for several months Lactobacillus strains similar to those of the donor became dominant. In February 2022, the patient became pregnant naturally. She received therapy for her APLS during this pregnancy. Regular monitoring of her vaginal microbiota revealed the return of Gardnerella spp. (41.8%) at gestational week six. A second VMT was initially planned for two weeks later, but by the day in question, L. crispatus had once again come to dominate the patient’s microbiota.

At the end of the pregnancy, a perfectly healthy baby boy was born via planned cesarean section. Although they need to be confirmed by further clinical studies, these results suggest that VMT may serve as a treatment for patients with severe vaginal dysbiosis, including those at risk of complications following in vitro fertilization.

For the authors, this case study serves as a proof of concept, but it also offers hope for therapies based on the modulation of the vaginal microbiota.

Microbiome-based personalized diet to improve pre-diabetes

^{Ben-Yacov O, Godneva A, Rein M, et al. Gut microbiome modulates the effects of a personalised postprandial-targeting (PPT) diet on cardiometabolic markers: a diet intervention in pre-diabetes Gut 2023;72:1486-1496.}

Researchers at Weizmann institute have been pioneers in developing personalized dietary approaches based on gut microbiome (GM). In this paper, Ben-Yacov et al. studied the effects of personalized postprandial-targeting (PPT) versus Mediterranean diet (MED) on cardiometabolic risk factors. Overall, diet is known to affect cardiometabolic health but whether the GM modulates these effects has been scarcely studied in longitudinal settings. In this 6-months trial, 225 pre-diabetic adults were randomly assigned to PPT and MED arms. PPT was based on an algorithm and MED on dietitian judgement. Overall, there was a low-carbohydrate and high-fat pattern of the PPT intervention since dietary carbohydrate is an important component in post-prandial glucose response. Compared to MED, the PPT intervention increased GM diversity and richness more. In the PPT arm, the consumption of some catechin-rich foods including dark chocolate and cashews increased. This was further associated with the enrichment of Flavonifractor plautii that is reported to participate in flavonoid catechin metabolism. According to a statistical model, changes in specific GM species partially mediated the effects of diet on clinical outcomes. For instance, the change in UBA11471 sp000434215 (from Bacteroidales order) partially mediated the effect of change in ‘Med Oil and Fats’ consumption on HbA1c outcome. HbA1c is a glycated haemoglobin that is used to assess diabetes. Three bacterial species (from Bacteroidales, Lachnospiraceae and Oscillospirales orders) were found to mediate the effect of the PPT with clinical outcomes of HbA1c, HDL-cholesterol and triglycerides.

In conclusion, the study supports the role of GM in modifying the effects of diet changes on cardiometabolic outcomes and advance the concept of precision nutrition for reducing comorbidities in pre-diabetes.

Fecal metabolome in inflammatory bowel disease

^{Vich Vila A, Hu S, Andreu-Sánchez S, et al. Faecal metabolome and its determinants in inflammatory bowel disease. Gut 2023; 72: 1472-85.}

Ulcerative colitis (UC) and Crohn’s disease (CD) are the subtypes of inflammatory bowel disease (IBD). Several microbial metabolites are known to affect inflammatory reactions that are important players in IBD. However, untargeted fecal metabolomic studies in IBD patients are scarce. In this study, the potential of fecal metabolites as biomarkers for IBD were assessed. The study comprised of 255 healthy controls and 424 IBD patients. The untargeted metabolomic analyses were accompanied with gut microbiota composition, exome sequencing, and genomic array data analyses in both cohorts. The metabolomes of the IBD groups were characterized by depletion of vitamins and fatty acid-related molecules. In addition, IBD patients had higher levels of the phenolic compound p-cresol sulphate, which originates when gut bacteria ferment proteins. The patients with UC had lowest levels of fecal anti-inflammatory short chain fatty acids. To identify potential biomarkers, a machine learning approach was used to predict disease phenotypes. The ratio of sphingolipid and L-urobilin discriminated between IBD and non-IBD samples. In the patients with IBD, the increase of pathobionts co-occured with increased levels of sphingolipids, ethanolamine and primary bile acids. In CD patients, resection of the ileocecal valve was associated with changes in the levels of 212 metabolites, such as cholic acid. Further, the resection associated in a reduced abundance of Faecalibacterium prausnitzii, which also negatively affected the levels of anti-inflammatory metabolites.

A mediation analysis showed that the observed associations between lifestyle, clinical factors and fecal metabolites were driven by alterations in the gut microbiota. Altogether, the study shows the potential of fecal metabolites as biomarkers for IBD and that, despite the influence of lifestyle, genetics and disease, gut microbes are strong predictors of the levels of fecal metabolites.

A microbiotamodulated checkpoint directs immunosuppressive intestinal T cells into cancers

^{Fidelle M, Rauber C, Alves Costa Silva C, Tian AL, et al. A microbiota-modulated checkpoint directs immunosuppressive intestinal T cells into cancers. Science 2023; 380: eabo2296.}

Resistance of cancers to immune checkpoint inhibitors (ICIs) as a treatment can result from antibiotic (ABX) treatment, which may involve the gut microbiota. However, this relationship has not been extensively studied. Therefore, Fidelle and co-workers addressed the gaps in the knowledge using a rodent model and human patients. Based on the literature, the gut bacteria can induce the differentiation of lymphocytes primed in the mesenteric lymph nodes or homing to the intestinal lamina propria express the α4β7 integrin interacting with its counter-receptor, mucosal addressin cell adhesion molecule-1 (MAdCAM-1), which is expressed in high endothelial venules (HEVs). This importantly prevents the immigration of Treg17 cells to the gut tumors, which can further compromise the anticancer effects of ICIs. Th17 are a subset of pro-inflammatory T helper cells defined by their production of interleukin 17 (IL-17). These cells are related to T regulatory cells and the signals that cause Th17s to inhibit Treg differentiation. In rodents, ABX-treatment reduced MAdCAM-1 expression. This could be explained by the intestinal recolonization by the genus Enterocloster. Further, the oral administration of Enterocloster was sufficient to down-regulate MAdCAM-1 expression. Restoration of MAdCAM-1 expression on the ileal HEV by fecal microbial transplantation or blockade of IL-17A reversed the inhibitory effects of ABX. Ectopic expression of MAdCAM-1 in the liver caused a local maintenance of enterotropic α4β7+ Treg17 cells.

This maintenance further reduced their accumulation in the tumors as well as improved the efficacy of the immunotherapy in mice. In the cohorts of lung, kidney, and bladder cancer patients, low serum levels of MAdCAM-1 had a negative prognostic impact. In conclusion, the MAdCAM-1– α4β7 axis should be preferably considered in cancer immunosurveillance.

What do we already know about this subject?

Attention deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) are neurodevelopment disorders. These children with ADHD and ASD often have gastrointestinal disorders such as abdominal pain and constipation. Genetic abnormalities are implicated in the onset of these disorders, coupled with interaction with environmental risk factors, dietary intake in particular. For this reason, in parallel with drug therapies, dietary management is proposed, the composition of the gut microbiome being essential in the regulation of the gut-brain axis. Moreover we know that children with ASD are often selective eaters which may explain a modification of the gut microbiota. In addition to the dysbiosis, an increase in intestinal permeability has been described, in addition to low grade systemic inflammation, in both ADHD and ASD. The aim of this study was to analyse modifications to the gut microbiota in the groups ADHD, ASD and those with both ADHD/ASD, compared with healthy siblings and non-related controls. The secondary objectives were to assess intestinal permeability and the immune system.

What are the main insights from this study?

A total of 95 children aged 5-17 years, including 32 ADHD, 12 ASD, 11 ADHD/ ASD, 14, 5, 4 siblings, and 17 non-related controls. Gastrointestinal disorders were constipation: ADHD 15.6% (siblings 7.1%), ASD 8.3% (siblings 0%), ADHD/ ASD 18.2% (siblings 0%), controls 5.9%; abdominal pain: ADHD 3.1% (siblings 0%), ASD 16.7% (siblings 0%), ADHD/ ASD 18.2% (siblings 0%), controls 0%; and less frequently gastroœsophageal reflux. An atypical diet was found to be most common among the ASD children (50%), essentially a lack of variation in food intake. The analysis of the gut microbiota failed to find any variations in alpha-diversity between ADHD, ASD, ADHD/ASD and related or non-related controls; in contrast, it was significantly lower in siblings with ASD (figure 1). Gut microbiota composition was very similar between ADHD and ASD, as shown by beta-diversity, but this differed significantly between ADHD and ASD compared with non-related controls (figure 2). Analysis of the gut microbiota composition showed that some ADHD, ASD, or ADHD/ASD children had a relatively lower abundance of the Bacteroidetes phylum and greater abundance of Actinetobacteria. All groups were dominated by Bacteroides, Faecalibacterium, Blautia and Bifidobacterium genera; some children exhibited a high level of Prevotella. Differences in the abundance of bacterial genera were found between ADHD, ASD and controls (figure 3) but not between ADHD and ASD.

No difference was observed between the various groups, nor between related or non-related controls, for levels of faecal calprotectin nor for lipopolysaccharide-binding protein (LBP). However, there was no correlation either between faecal calprotectin and LBP with bacterial alpha- and beta-diversities. Measurement of various cytokines and chemokines failed to show significant differences between the various groups; however, several ADHD and ASD individuals had higher levels of IL1-RA compared with non-related controls and five ADHD children and one ASD child had concentrations of IFN- g higher than non-related controls. Lastly, weakly positive correlations were found between LBP and IL-8 (p=0.023), IL-12 (p=0.018), IL-13 (p=0.035) and PlGF (p=0.045), suggesting that disruption of the gut barrier function may result in immune dysregulation.

What are the consequences in practice?

This study was conducted on a small number of individuals, in particular concerning the small number of related controls. The gut microbiota, as well as intestinal permeability, may be relevant targets for the treatment of children and adolescents with ADHD and ASD.

What do we already know about this subject?

Crohn’s disease (CD) is an inflammatory bowel disease (IBD) characterised by recurrent chronic inflammation of the intestine. The cause of CD is unknown, but the current hypothesis is that microbial or environmental factors induce inflammation of the gut in genetically predisposed individuals, leading to inflammation and chronic lesions. Case-control studies of patients with CD have catalogued alterations in the gut microbiome composition [1]. However, these studies fail to distinguish whether the altered gut microbiome composition is associated with the onset of Crohn’s disease or is the result of inflammation or a drug treatment. To answer these questions, the Canadian project GEM (Genetic Environmental Microbial), a prospective cohort study in healthy first-degree relatives of individuals with Crohn’s disease, was designed to identify the parameters associated with the development of Crohn’s disease. Among parameters, the authors studied the gut microbiome profile preceding the onset of CD and to what extent this composition predicts the risk of developing CD. The authors applied a machine learning approach to the analysis of the composition of the gut microbiome in a large cohort of healthy first-degree relatives of individuals with Crohn’s disease (N = 3,483) with the aim of defining a microbial signature associated with the risk of developing CD.

What are the main insights from this study?

On the basis of data from the GEM cohort, the authors developed and validated a microbiome risk score (MRS) used to classify individuals who will develop CD in the future. The taxa contributing most to the MRS were an increased abundance of Ruminoccus torques and Blautia which was positively correlated to the MRS (suggesting a harmful effect of these taxa), whereas the abundance of the genus Roseburia was negatively correlated with the MRS (suggesting a protective effect of this genus). Lastly the authors observed than an increase in the abundance of the genus Faecalibacterium was inversely related to an increase in the MRS. This study is the first to demonstrate that a decrease in the abundance of Faecalibacterium may be a preclinical signature of CD which may be observed many years before the onset of the disease, suggesting a causal relationship with the decrease in this anti-inflammatory bacteria [2] . Significantly, the changes in the microbiome preceding the onset of CD were observed independently of the existence of intestinal inflammation (indicated by faecal calprotectin levels).

The authors also performed a metabolomic analysis of stool samples from a sub-group of the cohort. Cytosine and its derivative, cytidine showed the strongest negative correlation with the MRS.

Moreover, the pre-CD signature of the MRS was associated with a reduction in metabolites having anti-inflammatory or antioxidant activity such as gentisate and nicotinate. These protective metabolites were also positively correlated with the abundance of Faecalibacterium and Lachnospira, which indicates a potential biological interaction between the abundance of these metabolites and the microbial composition.

What are the consequences in practice?

This study suggests that the analysis of the microbiome of healthy individuals at risk of developing Crohn’s disease may identify the most high risk subjects and thus permit their close supervision and the possibility of initiating interventions designed to modify the microbial imbalance and therefore reduce the risk of developing the disease.

The vast majority of the presentations gave the information about the composition and functions of the microbiota and about the gastrointestinal microbiota as a therapeutic target in the treatment of various diseases.

A special symposium (“Fungi in your gut: friends or foes”) was devoted to the mycobiome, as one of the components of the microbiome, mycobiome formation, environmental factors influencing mycobiome composition, the interaction of the mycobiome with bacteria (Selena Porcati, Italy);
role in the pathogenesis of IBD (Dragos Ciocan, France), and its potential involvement in carcinogenesis (Alexander Link, Germany).

Microbiome in IBS/IBD

The information about the microbiome role in the pathogenesis of irritable bowel syndrome (IBS) and irritable bowel disease (IBD) is still growing and expanding (symposium “Disease primer: The role of gut microenvironment in IBD and IBS”). Harry Sokol (France) and Rinse K. Weersma (Netherlands) reported that changes in microbiota composition may be considered as a biomarker for IBD and could be targeted by therapeutic intervention through probiotics, postbiotics, bacteriophages, and fecal transplantation. As for changes in the microbial composition of the gastrointestinal tract in patients with IBS, it has an undeniable contribution to all the pathogenic mechanisms of the disease (inflammation of the intestinal wall, impaired motility, hypersensitivity), therefore the prescription of non-absorbable antibiotics and probiotics can be considered an essential part of the IBS treatment (Magnus Simren (Sweden) and Premysl Bercik (Canada).

What is a healthy microbiota?

Furthermore, there were some questions raised without definite answers yet. For example, we still lack knowledge of what the term “healthy microbiota” actually means. It is assumed that the more appropriate term would be “unhealthy microbiota” (the so-called B2 enterotype), reflecting the inflammatory changes in the intestine and accelerated transit –when microbiota is represented mainly by Bacteroides, is low in Firmicutes and has poor microbial diversity. Targeting the microbiota composition in order to make it shift it from B2 enterotype can be considered a new therapeutic strategy (Jeroen Raes, Belgium).

In addition, due to the unquestionable importance of intestinal microbiota composition both in maintaining human health and promoting the pathogenesis of some chronic non-infectious diseases, clinicians nowadays, often unreasonably, expect to use microbial composition assays as a diagnostic, prognostic or therapeutic tool. An increasing number of commercial organizations offer microbiota diagnostic tests which are available with neither clear indications for use nor reliable interpretation of the results. An International Consensus development has begun, bringing together more than 50 international experts with the ultimate goal of streamlining diagnostic tests, treatment approaches and advancement of knowledge in the field of the microbiome (Gianluca Janiro, Italy).

In addition to this discussion of the microbiome as a direct pathogenic factor and a target for therapeutic intervention, other aspects of the pathogenesis and treatment of diseases associated with micobiom gut dysbiosis were also presented. Among them are IBD and oncological diseases.

For more than 20 years we have indirect serological and genetic evidence of fungal role in intestinal inflammation in IBD patients, such as antisaccharomyces antibodies in Crohn disease patients and genetic polymorphism of сaspase recruitment domain-containing protein 9 (CARD9) and dectin-1. These polymorphisms are mediate signals from pattern recognition receptors to activate pro-inflammatory cytokines. A lot of studies in the last ten years prove that abundance of fungi species in the gut IBD patients decreases compared to healthy people. Alterations of the mycobiota composition associate with poor injury repair of mucomucosa (in animal model). Sacharomyces boulardii given as probiotic can reduce gut inflammation due to intestinal barrier restore (in animal model). But the use of fungal community modification to treat IBD needs further research (Dragos Ciocan).

In recent years, there has been increasing interest in the potential role of intestinal fungi and their recognition receptors, (for example, C-type lectin receptors) in the development of human cancers, such as esophageal, gastric, pancreatic, colorectal cancer, hepatocellular carcinoma and
non-gastrointestinal cancer also – melanoma, breast cancer. Some studies demonstrate that fungal pathogens may induce inflammatory responses, contributing to tumorigenesis.

Functional gastrointestinal disorders and the microbiota

The intestinal microbiota has a bidirectional relationship with motility, visceral sensitivity, GI secretory function, permeability, and the immune system. Probiotics show promise in managing functional disorders. For infantile colic, ESPGHAN recommends some strains of Lactobacillus and Bifidobacterium in exclusively breastfed infants. HCPs may suggest different strains of Lactobacillus for functional abdominal pain, or to reduce symptoms in IBS [1].

Modulating microbiota to impact infant’s life

Factors such as delivery mode, breastfeeding [2], environment and no antibiotic use positively impact neonatal colonization, fostering a healthy intestinal environment, metabolic balance, homeostasis, and immune tolerance. Conversely, preterm birth, C-section, lack of breastfeeding, NICU admission, and antibiotic use may lead to dysbiosis, contributing to immune-related diseases. Strategies for modulation and prevention of dysbiosis involve a balanced diet and the use of probiotics, prebiotics, synbiotics, and postbiotics.

Human milk microbiota

Human milk encompasses probiotics, prebiotics, synbiotics, and postbiotics, collectively contributing to the microbiota balance in breastfed children. Maternal skin and the infant’s oral cavity are identified as the primary contributors to the milk microbiota. Factors that modulate the human milk microbiota include gestational age, infant gender, mode of delivery, lactation stage, feeding mode, geographic location, social network density, maternal health status, and maternal diet [3].

Upper respiratory tract infection (URTI) and microbiota

URTIs leads to indiscriminate antibiotic prescription worldwide, and the WHO estimates that antibiotic resistance-related deaths could reach 10 million by 2050. A systematic review showed an overall 35% reduction in the number of URTIs when probiotics were used, 2-day decrease in
the severity of symptoms, and 45% reduction in antibiotic use. Certain probiotic strains appear promising in reducing the incidence of viral URTIs, severity of infections, and antibiotic use [4].

There is increasing evidence that the gut microbiome plays a significant role in influencing human health. Diet is central in this relationship, and western-style dietary patterns have played a major role in the recent exacerbation of chronic non-communicable diseases (NCDs) in socio-economically developed societies. Here, we discuss what constitutes healthy eating from a microbiome science perspective and apply this evidence to inform ongoing controversies in the nutrition field and development of microbiome-targeted nutritional strategies. We focus this article on
dietary recommendations in the general population with an aim of disease prevention, not patients with medical conditions who often have specific dietary requirements and should consult with a registered dietitian.

Whole plant foods versus processed foods

According to all dietary guidelines we reviewed [1], whole plant foods (vegetables, fruits, whole grains, legumes, and nuts) that have undergone limited processing should dominate a daily diet (figure 1). This recommendation is well-supported by a microbiome perspective (figure 2 and figure 3). Whole plant foods are the sole naturally occurring source of dietary fibers, some of which are fermentable and provide growth substrates for microbes. Plant variety may maintain microbiome diversity, and fiber fermentation results in metabolites such as short-chain fatty acids (SCFAs) that evoke a wide variety of metabolic (satiety-related hormones and improved insulin sensitivity), physiological (increased mucus production and tight junction expression), and ecological (pathogen inhibition) effects [2]. In addition, provision of substrates for bacteria prevents mucus degradation and downstream inflammation and infections in mice [3]. Phytochemicals present in whole plant foods, most of which are not absorbed in the small intestine, also undergo biotransformation by gut microbiota that increases their bioavailability, absorption, antioxidative, and immunomodulatory effects [4], but the significance of these interactions for health are less well-understood. Finally, the functional characteristics and nutritional quality (e.g., nutrient composition and accessibility) of most whole plant foods are vastly superior to processed foods, which often further contain food additives that impair microbiome composition and gut barrier function (figure 2).

Whole grains

The potential role of gut microbiome in the well-established metabolic and immunological benefits of whole grains is increasingly investigated. The bran layer of whole grains contains dietary fibers such as arabinoxylans and β-glucans that are fermented by gut microbiota into beneficial metabolites. The anti-inflammatory effects of whole grains have been linked to enrichment of SCFA-producers [5]. In ex-germ-free mice colonized with fecal microbiota from humans that responded to whole grain barley kernel bread and contained Prevotella, it was observed improvement in glucose tolerance that mirrored effect in humans [6]. In addition, individuals with excess body weight that harbor Prevotella at baseline show elevated weight loss on a whole grain-rich diet [7]. These studies suggest that at least some of the metabolic benefits of whole grains are mediated by the gut microbiome (figure 3).

Sources of protein

Most dietary guidelines recommend consuming plant-based protein foods (e.g., legumes, nuts), and fish (e.g., fatty fish) and poultry at the expense of other animal sources of protein, specifically red meat (figure 1). Legumes and nuts are rich in fibre and contain phytochemicals and omega-3 fatty acids that influence host-microbe interactions (figure 3). Daily supplementation of walnuts and almonds increased relative abundances butyrate- producers, specifically Roseburia [8]. Mung bean supplementation reduced weight gain and fat accumulation in mice fed high-fat diets but not in germ-free mice fed the same diets, establishing a causal role of the microbiome [9]. Among all animal- based protein foods, fatty fish is likely to display the largest microbiome-driven immunological and metabolic benefits [1].

Dietary patterns

The realisation that health is not primarily influenced by individual foods or nutrients but their interconnectedness and synergistic effects led to an emphasis on dietary patterns in several recently updated dietary guidelines, such as the 2020-2025 Dietary Guidelines for Americans and Canada’s food guide. Mediterranean diet combines many of the food groups that have favorable effects on host-microbe interactions. Several randomized-controlled trials were conducted to investigate these interactions, and showed that metabolic, immunological, and cognitive benefits of
a Mediterranean diet were linked to increases in Faecalibacterium prausnitzii and Roseburia abundances [10].

Red and processed meats

Most dietary guidelines and several medical societies recommend the reduction of red meat and the avoidance of processed meats, although a series of systematic reviews from 2019 concluded that there is only weak evidence of their links to adverse health outcomes [11]. The gut microbiome provides a helpful perspective in this controversy. Proteolytic fermentation of meat protein by gut microbes increases toxic metabolites such as ammonia, p-cresol, and hydrogen sulphide [12]. Being high in saturated fat, processed meats further stimulate secretion of bile acids into the small intestine that are then transformed into secondary bile acids by microbes. In addition, the curing agents used in processed meats, nitrate and nitrite, are substrates for microbial biotransformation to N-nitroso compounds. Toxicological considerations therefore support current dietary recommendations (figure 3). 

The metabolites resulting from protein fermentation (e.g., hydrogen sulphide, ammonia) are of lower toxicity and not currently classified as human carcinogens, supporting the conclusion that moderate consumption of lean red meat is likely of limited risk. In contrast, the N-nitroso compounds and secondary bile acids that result from consumption of processed meats are carcinogenic, supporting recommendations to avoid or minimize the consumption of processed meat.

Dairy

Most dietary guidelines recommend skimmed and low-fat (0-2%) dairy products and suggest avoiding high-fat (> 25%) dairy products (e.g., certain cheeses, cream-based products, butter). However, there is no consensus on full-fat dairy (~ 3.5%), which is discouraged in some dietary guidelines, although its detrimental effects have been questioned. Interactions between dairy fat and the gut microbiome are relevant for this discussion. Milk-derived saturated fats induce Bilophila wadsworthia, which is pro-inflammatory and caused diseases such as colitis [13] in mouse models. These mechanistic findings support dietary recommendations to limit dairy to low-fat varieties (figure 3).

Low-carbohydrate diets

Low-carbohydrate diets are popular as they can achieve remarkable short-term weight loss and metabolic benefits, although results may not be sustainable in the long-term. These diets are high in fat and/or protein and often low in fiber.
Consequently, they result in a detrimental metabolic profile with increased concentrations of N-nitroso compounds and decreased levels of butyrate and anti-inflammatory phenolic compounds [14]. Due to their effects on gut microbiota, low-carbohydrate diets might therefore be detrimental to health when consumed over longer periods.

Although international dietary guidelines are highly consistent and provide excellent guidance on what constitutes healthy eating, there is scope for improvements and innovations through a more systematic consideration of the microbiome.

Evolutionary considerations and microbiome restoration

The human-microbiome symbiosis has evolved over millions of years in a very different environmental and nutritional context. Industrialization, which led to a substantial increase in NCDs, depleted microbiome diversity, diminished enzymatic capacity of the microbiome for carbohydrate utilization, enriched for mucus- degrading organisms and enzymes, and led to a loss of microbial symbionts. There is therefore an evolutionary-based argument to boost fiber intake beyond the 25-38 grams per day that are currently recommended in dietary guidelines, and this is supported by both observational and intervention studies [15]. In addition to promoting increased fibre intake from whole foods, there is strong scientific rationale to redress the impact of industrialization on the gut microbiome through prebiotic, probiotic, and synbiotic strategies. Products that attempt to restore and diversify the microbiome are already on the market, but research on their clinical benefits is in its infancy (see below).

Live microbes

Another hallmark of industrialization is reduced microbial exposure. The biodiversity hypothesis states that contact with natural environments is required to enrich the human microbiome, promote immune balance, and protect from allergy and inflammatory disorders. Probiotics provide
live microbes and have been studied and marketed in this context for decades (see box “Probiotics and prebiotics”). In addition, fermented foods, such as kefir, yoghurt, kombucha, and Sauerkraut can, if consumed raw, contain high numbers of live microbes (bacteria and fungi). Although
microbes present in fermented foods do not colonize the human gut due to their non-native nature in the human gut ecosystem, they are still detectable within human fecal microbiota and might interact directly with the host.

Precision nutrition

Humans differ in their response to dietary interventions, which questions the onesize- fits-all approach currently applied in dietary guidelines. Precision or personalized nutrition aims to tailor nutritional recommendations to an individual’s biology (genes, metabolism, etc.). Microbiome measurements might become a key component of precision nutrition strategies. Although several companies already offer personalized dietary advice based on the fecal microbiome, these services are not validated by any regulatory authority, leaving uncertainty regarding the accuracy of the recommendations. National dietary guidelines currently do not consider precision or personalized approaches, and their implementation will be challenging on a population scale. Although there is a scientific rationale to personalize nutrition, it is important to emphasize that most individuals will benefit from the dietary recommendations discussed above.