Microbiomes in physiology: insights into 21st‐century global medical challenges

New Findings What is the topic of this review? The role of the gut microbiome in physiology and how it can be targeted as an effective strategy against two of the most important global medical challenges of our time, namely, metabolic diseases and antibacterial resistance. What advances does it highlight? The critical roles of the microbiome in regulating host physiology and how microbiome analysis is useful for disease stratification to enable informed clinical decisions and develop interventions such as faecal microbiota transplantation, prebiotics and probiotics. Also, the limitations of microbiome modulation, including the potential for probiotics to enhance antimicrobial resistance gene reservoirs, and that currently a ‘healthy microbiome’ that can be used as a biobank for transplantation is yet to be defined. Abstract The human gut microbiome is a key factor in the development of metabolic diseases and antimicrobial resistance, which are among the greatest global medical challenges of the 21st century. A recent symposium aimed to highlight state‐of‐the‐art evidence for the role of the gut microbiome in physiology, from childhood to adulthood, and the impact this has on global disease outcomes, ageing and antimicrobial resistance. Although the gut microbiome is established early in life, over time the microbiome and its components including metabolites can become perturbed due to changes such as dietary habits, use of antibiotics and age. As gut microbial metabolites, including short‐chain fatty acids, secondary bile acids and trimethylamine‐N‐oxide, can interact with host receptors including G protein‐coupled receptors and can alter host metabolic fluxes, they can significantly affect physiological homoeostasis leading to metabolic diseases. These metabolites can be used to stratify disease phenotypes such as irritable bowel syndrome and adverse events after heart failure and allow informed decisions on clinical management and treatment. While strategies such as use of probiotics, prebiotics and faecal microbiota transplantation have been proposed as interventions to treat and prevent metabolic diseases and antimicrobial resistance, caution must be exercised, first due to the potential of probiotics to enhance antimicrobial resistance gene reservoirs, and second, a ‘healthy gut microbiome’ that can be used as a biobank for transplantation is yet to be defined. We highlight that sampling other parts of the gastrointestinal tract may produce more representative data than the faecal microbiome alone.


INTRODUCTION
The microbiome is critical to physiological homoeostasis, influencing health and disease status in the host. The human body contains trillions of microbes encompassing bacteria, archaea, viruses and microeukaryotes (Zhang et al., 2019). In the current report, the gut microbiome consists of symbiotic or pathobionts that are resident in the host and opportunistic pathogens that are acquired from the environment or other parts of the body (Casadevall & Pirofski, 1999;Chow et al., 2011). While opportunistic pathogens can cause acute effects, pathobionts are only able to cause deleterious effects to host health in certain circumstances, such as when the immune system has become compromised. Commensals/symbionts contribute to the maintenance of physiological homoeostasis as well as providing colonisation resistance to opportunistic pathogens (Hornef, 2015).
Acute deleterious effects of pathogens in physiology can lead to infections and in extreme cases cause bacteraemia leading to sepsis and death. This has led to heavy overuse of antibiotics to combat infections, which has provided the selective pressure that is driving increases in antibiotic resistance (Ahmed, 2005;Casadevall & Pirofski, 1999;Lau et al., 2004).
As well as harbouring antimicrobial resistance genes, the gut microbiome may also influence systemic physiological functions by competing for essential nutrients or digesting complex molecules to produce substrates for host energy metabolism and cell signalling (Martin et al., 2019). The gut microbiota can therefore also cause subtle but chronic physiological effects, which contribute to the epidemic of metabolic/inflammatory diseases such as diabetes, cardiovascular disease and neurodegenerative diseases Wang et al., 2020). Coupled to antibiotic resistance, metabolic diseases are among the leading global medical challenges of our time, posing a socio-economic burden worldwide. Nevertheless, mechanisms through which the microbiome influences physiology remain relatively poorly understood. It is envisaged that advances in sampling techniques, multi-omic approaches (genomic, transcriptomic, methylomic, proteomic, metabolomic), and bioinformatic tools will increase the resolution at which these pan-kingdom interactions can be studied, thus expanding our understanding of the influence of the microbiome on host physiology in health and disease. Such advances will likely revolutionise future clinical practices in disease prevention, treatment and management.

New Findings
• What is the topic of this review?
The role of the gut microbiome in physiology and how it can be targeted as an effective strategy against two of the most important global medical challenges of our time, namely, metabolic diseases and antibacterial resistance.
• What advances does it highlight?
The critical roles of the microbiome in regulating host physiology and how microbiome analysis is useful for disease stratification to enable informed clinical decisions and develop interventions such as faecal microbiota transplantation, prebiotics and probiotics. Also, the limitations of microbiome modulation, including the potential for probiotics to enhance antimicrobial resistance gene reservoirs, and that currently a 'healthy microbiome' that can be used as a biobank for transplantation is yet to be defined.
biomes in Physiology" symposium, which took place virtually on the 14th July 2021 at the main Physiological Society meeting and the current report is on the role of the gut microbiome in regulating physiological functions locally in the gut and remotely in various gut-organ axes, including the heart, the liver and the brain.

MICROBES ACROSS THE LIFESPAN
The overall composition of the microbiome is determined by early life events such as mode of delivery, breastfeeding and frequency of antibiotic use. However, the abundance of each microbe may fluctuate across the lifespan due to factors such as age, diet, lifestyle, cultural practices and geographical location (Arboleya et al., 2012;Johnson & Versalovic, 2012). In general, greater microbial diversity and functional redundancy provide resilience to perturbation by the aforementioned factors, and therefore are associated with beneficial impacts on the 2. While to date a 'healthy microbiome' has not been defined owing to intra-and inter-individual variations in the core microbiome, various microbial products can be used, due to their common functionality, to stratify host disease phenotypes such as irritable bowel syndrome and cardiovascular disease and can predict clinical outcomes after hospitalisation from diseases such as heart failure.
3. Plasma trimethylamine N-oxide (TMAO) is an example of a microbially derived metabolite shown to be an important prognostic marker of adverse events after heart failure and all-cause mortality and is comparable to traditional markers such as B-type natriuretic peptide (BNP) and N-terminal (NT)-pro-hormone BNP (NT-proBNP). Currently, there are no specific drugs that effectively and specifically alter the microbiome to reduce TMAO burden.
4. While there is a debate as to whether alterations in microbial structure and their metabolites such as TMAO may be causative, a mere correlation or indeed a symptom of disease, recent studies using faecal microbiota transplantation and conventionalised animal models indicate that the microbiome has direct/causative effects on host physiology, including effects on inflammaging. The supplementation of TMAO has direct atherogenic effects. In general, high microbial trimethylamine (TMA) production is indicative of altered microbiome composition and structure.
5. The microbiome can also affect the host's health by supporting pathogenic bacteria. This may be mediated by metabolic cross-feeding, whereby commensal species produce metabolites, which support the growth of pathogens and pathobionts. In addition, the microbiome can serve as a reservoir for antimicrobial resistance genes, which can transfer horizontally to pathogenic bacteria. While probiotics have been widely used to prevent or treat diseases (potentially through the modulation of the microbiome), and have been postulated to reduce resistance genes, it has been shown that probiotics may increase and exacerbate the number of antimicrobial resistance genes enhanced by antibiotics.
6. Most data on the diversity and function of the microbiome have been inferred from sampling the faecal microbiome but emerging evidence suggests that the faecal microbiome may misrepresent effects of probiotics on the intestinal microbiome community and the abundance of resistance genes in the gastrointestinal (GI) tract. Therefore, the use of direct sampling from the GI tract is paramount in future research.
At the symposium, Swann discussed age-dependent variability in the neurobiochemical profiles of mice across the lifecourse, with fluctuations in several microbially derived metabolites. This included metabolites such as 3-indoxyl-sulphate, γ-aminobutyric acid, TMAO, hippurate and phenylacetylglutamine . While certain metabolites were abundant during the neonatal period and declined into adulthood, others gradually increased with age, and some peaked in abundance at puberty before returning to neonatal levels in adulthood . As many of these compounds are involved in brain function and development, it is important to At the symposium, a direct impact of the microbiome on ageassociated inflammation in the brain was described by Parker who reported that faecal microbiota transplantation (FMT) treatment was effective in switching from an age-associated chronic lowgrade inflammatory phenotype (inflammaging) and a younger less inflammatory phenotype (Parker et al., 2021). Young mice receiving microbiota from aged mice exhibited an elevated inflammatory phenotype, whereas age-associated serum and brain inflammatory changes in mice could be reduced or reversed by transplantation with microbiota from young donor mice (Parker et al., 2021 (Brunt et al., 2020;Geng et al., 2018;Tan et al., 2019).

MICROBIAL COMPONENTS, CELL SIGNALLING AND DISEASE STRATIFICATION
In their talks, Swann and Parker highlighted various signalling pathways that are affected by microbially derived compounds (such as metabolites, cell wall components and extracellular vesicles), which can regulate immune function, metabolic homoeostasis and brain function (Figure 1). Among the metabolites, SCFAs are perhaps the most-studied gut microbially derived metabolites. SCFAs interact with a range of receptors such as G protein-coupled receptors (GPCRs) on host cells, both locally in the gut and in remote organs such as the brain, heart and the liver. Through their interaction with GPCRs, SCFAs have been shown to modulate the secretion of hormones including glucagon-like peptide-1 and peptide YY, which impact on the brain functions such as mood, appetite, food intake and energy expenditure (Frost et al., 2014;Modasia et al., 2020). In the gut, SCFAs are also utilised by intestinal epithelial and colonic cells as energy sources, positively promoting gut barrier integrity, as well as maintaining low intestinal pH that is unfavourable to opportunistic pathogens and pathobionts (Pérez-Reytor et al., 2021). Nevertheless, SCFAs may also be used by pathogens such as Salmonella, Clostridium and Citrobacter species as a cue for expressing virulence genes . This is a particularly good example of how the interaction of specific microbial species with the complex multi-organism gut microbiome may influence disease risk. In irritable bowel syndrome (IBS) patients with constipation (IBS-C), reduced levels of SCFAs in faecal samples are coupled with reduced levels of acetate in mucosal biopsies (Mars et al., 2020). Although SCFAs derive from dietary fibre, these observations were independent of dietary fibre intake suggesting that other factors may influence the availability of SCFAs.
Recent data suggest that SCFAs may derive from the metabolism of L-carnitine to trimethylamine, a pathway highlighted by Suzuki at the symposium (Suzuki et al., 2021) and discussed below in relation to the atherogenic phenotype (Rajakovich et al., 2021). In their talk, Swann further demonstrated that in contrast to IBS-C patients, IBS patients with diarrhoea (IBS-D) present higher levels of tryptophan and its indoleamine microbial metabolite, tryptamine, which again acts locally hypoxanthine, provides an excellent example of competitive nutritional partitioning between the microbiota and the host .
Hypoxanthine is an important energy source for intestinal epithelial cells, promoting epithelial cell development and recovery from injury, however, Swann demonstrated increased hypoxanthine use by the gut microbiome with decreased levels being observed in IBS-C patients (Mars et al., 2020). As such, alterations in microbiota composition and metabolites during critical developmental periods may prove detrimental to health.

STRATIFYING DISEASE OUTCOMES AND CLINICAL DECISIONS BASED ON MICROBIAL METABOLITES: THE CASE OF TRIMETHYLAMINE OXIDE AND HEART FAILURE
The possibility of stratifying patients and guiding clinical decisions based on metabolic profiles was highlighted by Suzuki. Plasma TMAO levels were shown to be a strong predictor of adverse secondary events after heart failure compared to traditional markers such as Nterminal pro-B-type natriuretic peptide (NT-proBNP) (Senthong et al., 2016;Suzuki et al., 2016). TMAO is produced in the liver by flavin- is further metabolised in a multistep process involving several gene clusters to produce TMA and other metabolites including SCFAs, such as acetate and butyrate, which have been shown to be the endproducts in this process (Day-Walsh et al., 2021;Rajakovich et al., 2021). The factors that regulate the formation of TMA from carnitine are yet to be understood although it seems that this pathway may be more important in the production of the atherogenic TMA than that involving choline. In their talk, Suzuki further highlighted that the associations of TMAO with adverse events after heart failure are influenced by geographical location, being higher in individuals from Norway, the Netherlands, Germany, Sweden and the United Kingdom than in those from Italy and Greece (Suzuki et al., 2019). Of note, this geographical variation was shown to be independent of polymorphisms in the FMO3 gene along with diet, indicating that there is a yet unknown factor influencing the predictive capacity of TMAO on all-cause mortality and death after heart failure.
To date, it has been difficult to clarify the compositions or indeed microbial species that can be used to predict disease phenotypes in the host. However, the abundant microbial metabolites provide an opportunity to profile and characterise individuals who may be at risk of not only heart failure but other metabolic diseases as well as hospitalisation and adverse events after hospitalisation including death. In their talk, Suzuki demonstrated that heart failure patients presenting high levels of TMAO when they are admitted to the hospital might still have high levels after treatment; that is to say, current treatments of heart failure patients and management of their future risk do not target their TMAO status. This presents an opportunity to stratify patients who may be at risk and to make clinically relevant informed decisions based on microbial metabolites. Nevertheless, there is an urgent need for therapies that can target the microbiome to reduce the burden of TMAO in those at risk, as current medications such as β-blockers do not target the microbiome.

MANIPULATING THE MICROBIOME FOR THE BENEFIT OF THE HOST: PROBIOTICS, PREBIOTICS AND FAECAL MICROBIOTA TRANSPLANTATION
Supplementation with live probiotic microorganisms has been proposed as a means for beneficially altering the microbiome, for example by reducing the production of disease-associated metabolites (including TMAO) or reducing the burden of pathobionts and commensals carrying antibiotic resistance genes. As highlighted by Swann probiotics, including those commonly used as dietary supplements, could support health by preventing the colonisation of pathobionts in preterm babies (Alcon-Giner et al., 2020). However, in their talk, Suez highlighted the complexity and limitations of using probiotics in adults, as the colonisation success of supplemented probiotics shows high inter-individual variations, which may underlie heterogeneity in probiotics' efficacy . For example, members of the gut microbiome encode for antibiotic resistance genes, creating a reservoir (resistome) that can transfer horizontally to pathogens and pathobionts, facilitating the emergence of antibioticresistant strains. In their work, Suez demonstrated that probiotics can reduce the reservoir of antibiotic resistance genes in the human gut, but only in individuals permissive (receptive) to probiotic colonisation (Montassier et al., 2021).
In addition to being a major contributor to the expansion of the gut resistome, the use of antibiotics perturbs the gut microbiome, leading to dysbiosis associated with an elevated risk for non-communicable diseases. Probiotics are often consumed in conjunction with antibiotic therapy to prevent detrimental effects of antibiotics on the microbiome. In their talk, Suez reported that, surprisingly, probiotics delay, rather than facilitate recovery of microbiome diversity from a course of antibiotics. Furthermore, probiotics contributed to an expansion in the number of antibiotic resistance genes in the gut, and in particular increased the abundance of the clinically relevant vancomycin resistance gene (VanG) .
In addition to probiotics, nutraceutical compounds such as complex carbohydrates and polyphenols have been investigated for their capacity to alter the microbiome for the benefit of the host (prebiotics). While the increase in certain microbes in response to these nutraceuticals may suggest beneficial effects, caution has to be exercised as metabolic cross-feeding may promote the growth of a beneficial species, which will consequently produce metabolites that facilitate the survival of pathobionts (Eloe-Fadrosh & Rasko, 2013;Mohajeri et al., 2018). FMT has also been proposed as a mode to increase microbial diversity or rebalance a dysbiotic microbial composition resulting from infection or ageing for example. As demonstrated by Parker et al., FMT from young donors proved effective in preventing age-associated symptoms and inflammation (Parker et al., 2021). In the context of the resistome, Suez demonstrated that FMT was more effective than probiotic supplementation at restoring the resistome back to pre-antibiotic status. However, the complexity of the microbiome and variations between individuals make it difficult to pinpoint a 'healthy' or optimal microbial composition that can be used as a biobank to treat all those with gut dysbiosis.

CURRENT LIMITATIONS AND THE FUTURE OF THE MICROBIOME IN MEDICAL PHYSIOLOGY
The microbiome offers a unique albeit challenging opportunity to improve host metabolic physiology and revolutionise future clinical practices in disease prevention, treatment and management. However, there was a consensus among the speakers at the symposium on the requirement for the standardisation of sampling and experimental approaches, which will greatly improve our ability to understand the role of the microbiome in physiology. In particular, Swann highlighted the need for averaging longitudinal data from an individual collected from multiple sampling points while Suez demonstrated the disparity between the microbiome and resistome within the stool sample and that from different sites within the gastrointestinal tract. To date, most research has focused on the bacterial component (bacteriome) of the microbiome, but there is an increasing appreciation of the importance of the viral (virome) and fungal (mycobiome) fractions of the microbiome, which is likely to become more apparent as our ability to study these elements evolves, in particular as the bacteriome along with its metabolome is also substantially influenced by phage predation (Hsu et al., 2019).
Thus, it is evident that many of the non-communicable diseases

COMPETING INTERESTS
Authors declare no conflict of interest.