Epigenetic Regulation at the Interplay Between Gut Microbiota and Host Metabolism

1The human body co-habit with a diverse community of symbiotic microorganisms and their set of genes, collectively known as the microbiome. The acquisition of the initial microbiome is a dynamic rather than a static process during early life. A recent estimation of the number of bacterial cells over human cells in our body has reduced the ratio from 10:1 to a 1.3:1. This implies a similar number of bacterial and human cells in and on the human body. However, the human microbiome encodes for at least 100 times more genes than our genome. Therefore, the corresponding higher functionality of bacterial genes is a key aspect to understand existing metabolic interactions between the host and its microbiota.

2The microbiota helps their hosts to digest dietary fiber; produces some important neurotransmitters, hormones, and vitamins; helps in training the host immune system; and protects against pathogens, among many other functions. However, unbalanced microbiota can also cause disease. Some common diseases in western societies such as obesity and type 2 diabetes are associated with shifts in the relative abundance of gut bacteria composition and functionality, compared to the ones observed in healthy individuals. The cause of the microbiota imbalance (dysbiosis) of unhealthy individuals across age and geography has been mainly correlated with dietary habits. Therefore, the so-called chemical crosstalk between the microbiota and its host has tangible consequences for the physiological state of the host. However, the molecular mechanisms by which microbiota chemically interacts with host cells and regulate gene expression remain largely unknown. In this regard, the role that certain host-microbiota derivate metabolites may exert on epigenetic events at the DNA, RNA, and histone level needs to be further investigated.

3Differences in the microbiota or epigenome in two genetically identical organisms, such as same-sex inbred mice or monozygous twins, can create differences in susceptibilities to diseases including obesity and type 2 diabetes. As with gut microbiota, new studies have demonstrated that epigenetic events are highly dynamic, changing in response to nutrient availability or physical exercise. DNA methyltransferases, DNA hydroxylases, histone acetyltransferases, histone deacetylases, histone methyltransferases, and histone demethylases are enzymes responsible for adding or removing these dynamic epigenetic modifications. In this regard, endogenous metabolites can regulate gene expression through epigenetic events in host cells. For instance, histone deacetylation regulated by sirtuin family deacetylases is regulated by the NAD+/NADH ratio, acetyl-CoA, O-acetyl-ADP-ribose, and nicotinamide. Whether gut microbiota metabolism is regulating the concentration and/or activity of endogenously produced metabolites by the host remains largely unexplored, and it is only recently that an increasing number of researchers have started to investigate this possibility. have demonstrated that microbial colonization regulates global histone acetylation and methylation in multiple host tissues in a diet-dependent manner.

4Short-chain fatty acids (SCFAs) are exclusively produced by the microbial fermentation of dietary carbohydrates, and their abundances are regulated by the composition of the microbiota. Importantly, SCFAs, particularly butyrate and acetate, produced by the microbiota, inhibit histone deacetylases ( Figure 1 ). Increased levels of histone acetylation promote decondensation and relaxation of chromatin, supporting a more transcriptionally active state of chromatin.

5Alteration of chromatin state is a possible mechanism by which gut microbiota induces host immune maturation. Recognition of a “self” antigen should not only be limited to mammalian host antigens, but also symbiotic microbiota antigens forming part of the whole human ecosystem in a healthy state. The human leukocyte antigen (HLA) or major histocompatibility complex (MHC) gene system encodes many antigen-presenting proteins, which are essential to recognize and distinguish “self” from “non-self” antigens. Colonization of germ-free mice has demonstrated the capacity of microbiota-specific species to activate MHC class II genes. However, little is known about the immunomodulatory effect of microbial metabolites. A poorly explored possibility is that epigenetically relevant metabolites such as SCFA, highly influenced by microbiota composition, would regulate MHC gene expression by coordinating activity of enzymes that acetylate and methylate histones and DNA allowing chromatin accessibility.

6Regulation of DNA and histone methylation may be driven by complex microbiota–host metabolism interactions involving S-adenosyl methionine (SAM), derived from the essential amino acid methionine through diet. Folate plays an essential role by re-methylating homocysteine to methionine ( Figure 1 ), thereby ensuring the provision of SAM. In this regard, enzymes that are depleted in obese microbiomes are frequently involved in cofactor and vitamin metabolism, including the production of cobalamin (vitamin B12), pyridoxal phosphate (the active form of vitamin B6), tetrahydrofolate, and folate. Taken together, dysbiosis of microbiota can influence SAM levels and, as a result, alter the methylation status of DNA and histones. Whether dysbiosis of microbiota can alter α-ketoglutarate and succinate levels in specific peripheral host tissues, and regulate the rate of DNA demethylation, is a plausible but little explored possibility. Ten-eleven translocation (TET) enzymes are a key family of DNA and histone demethylases that use α-ketoglutarate as co-substrate. However, due to the structural similarity with α-ketoglutarate, Tets are susceptible to competitive inhibition by fumarate and succinate, causing an increase in histone and DNA methylation levels.

7Another modification that could be regulated by microbial metabolism is histone phosphorylation. In response to a low ATP/AMP ratio indicative of energy status, the AMP-activated protein kinase (AMPK) can translocate to chromatin and phosphorylate histone H2B. Changes in AMPK activity have been reported in obesity, type 2 diabetes, metabolic syndrome, and cardiovascular disease. Interestingly, germ-free mice were resistant to obesity and insulin resistance that develop after consuming a Western-style, high-fat, and sugar-rich diet. The persistently lean phenotype of germ-free animals was associated with increased skeletal muscle and liver levels of phosphorylated AMPK. It is also tempting to speculate that phosphotransferase systems (PTS) overrepresented in the Western diet microbiomes could have an impact on this histone modification.

8In short, dysbiosis and reduction of the microbiota diversity can potentially alter the levels of nutrients and metabolites that can potentially act as regulators of DNA methylation and histone modifications either by directly inhibiting enzymes that catalyze the processes, or by altering the availability of substrates necessary for the enzymatic reactions.

9Hosts and their microbiota have a very intimate relationship and should be considered as a single biological and evolutionary unit, termed holobiont. In this regard, we could arguably talk about holo–genome, –transcriptome, –proteome, or –metabolome, referring to the combination of both, host and host microbiota molecular layers or modules of information at the DNA, RNA, protein, or metabolite level, respectively ( Figure 2 ). To investigate the dynamics of the holobiont ecosystem network, multi-omic approaches bring unprecedented advantages. Diseases such as obesity and diabetes are known to be multifactorial, and the collection of several -omic data ( Table 1 ) from the same holobiont specimen, may provide a detailed molecular description and new mechanistic insights of how dietary nutrients and gut microbiota metabolism can regulate the host phenotype through gene expression and epigenetic and metabolic regulation.

10To study a complex metabolic disorder such as familial type 1 diabetes mellitus (T1D), used a combination of host genomics and proteomics together with metagenomics, metatranscriptomics, and metaproteomics to demonstrate a pronounced family membership effect in the structuration and functionality of the microbiomes. They observed a correlation between certain human pancreatic enzymes and the expression of specific microbial genes involved in key T1D metabolic transformations. used a combination of metabolomic, proteomics of histones, transcriptomic, and metagenomic techniques in conventional and germ-free mice, to demonstrate how SCFAs produced by the microbiota, or supplemented exogenously to germ-free mice, regulate histone post-translational modifications (PTMs). Comparing histone PTMs and transcriptional profiles between conventional, germ-free mice and germ-free mice supplemented with SCFAs, Krautkramer et al. concluded that SCFAs alone are partially causative for histone PTMs. In the same direction, used a combination of metagenomics, transcriptomics, epigenomics, and metabolomics together with imaging electron microscopy, to identify a diurnal rhythmicity in the microbial biogeography, metabolic profile, and metagenomic functionality as critical orchestrators of host epigenetic marks and gene expression. Thaiss et al. have demonstrated that host epigenetic and transcriptional circadian oscillations are partially dependent on environmental signals such as microbiome metabolite dynamics in the intestines and that peripheral organs “sense and adapt” to this circadian metabolite rhythms in a similar manner.

11Overall, multi-omic approaches will facilitate the structuration of future research, improve patient stratification toward a more personalized care, and open new avenues to evaluate the effectiveness of functional probiotics, functional foods, or nutritional interventions aimed at regulating host gene expression in health and disease.

12The development and improvement of new technologies and bioinformatic tools are advancing biomedical research at a fast pace. Multi-omic experiments are allowing researchers to obtain mechanistic insights on the human holobiont homeostasis, improving decision-making for next experiments to be performed. In a research context, extensive collection of -omic data will allow integration of information into modulable and controlled models of microbial communities. However, it is important to identify confounding factors in longitudinal -omic studies. Standardization of methods and techniques to reduce noise and bias in microbiome research will improve how we translate lab findings into the clinic.

13Using germ-free or gnotobiotic mouse models provide a framework to manipulate the gut microbial composition in a controlled manner. These models can be used to study the chemical crosstalk between host and microbiota by uncovering specific epigenetic changes in host cells induced by colonization of specific bacterial strains. Colonizing gnotobiotic mice with single strains or small bacterial consortia, instead of whole fecal transplants, might bring more accurate information to understand specific molecular mechanisms and molecular pathways involved in the host epigenetic regulations. Microbiota-induced dysbiosis with antibiotics might provide a useful approach to validate findings in gnotobiotic models by partially mimicking the effects of absence of microbiota or a reduction in the microbial diversity. Interestingly, organoids might also provide a challenging but very useful in vitro culturing system to study host–microbiota interactions.

14In a clinical context, gaining knowledge on the “microbiota–nutrient metabolism–host epigenetics axis” using multiple -omic approaches in combination with microbiota modulation therapies, has the potential to prevent and treat more efficiently metabolic diseases:

15Fecal microbiota transplantation (FMT): The use of microbiota modulation therapies such as FMT to treat recurrent Clostridium difficile infections has been proved significantly more efficient than a vancomycin treatment. Recently, the potential of FMT as microbiome modulation technique for treating metabolic, neurological, and immunological disorders has been tested, improving these conditions by partially restoring microbiota diversity and functionality. Assessing long-term host epigenetic effects of FMT has to be further investigated.

16Microbiota as drug target: The unique microbiome composition of each person is probably responsible for different susceptibilities to the same nutrient, pollutant, or drug treatment. Microbiota-derived metabolites can enter the bloodstream and interact with drug treatments, impacting the efficacy, toxicity, and clearance of the drug. Diagnosing or treating diseases using microbiota-targeted drugs, probiotics, and use of bacteriophages or engineered bacteria has recently re-emerged.

17Nutritional intervention, probiotics, and prebiotics: The acute consumption of a cocktail of probiotics containing a selection of five strains of Lactobacillus and five strains of Enterococcus modulates the microbiome and enhances SCFAs production in human and mice. The consumption of some probiotics has proven to be beneficial and ameliorates stress felt in healthy women, improves insulin sensitivity, protects against infections, and helps to restore microbiota after distortion by antibiotics, among other benefits. Promoting SCFAs producing bacteria in the host gut by a nutritional intervention that increases fiber consumption may have an epigenetic effect in the host.

18The study of the “microbiota–nutrient metabolism–host epigenetic” axis has great potential to reveal the molecular mechanisms by which gut microbiota composition affects the expression of genes in their hosts. In a human holobiont context, this axis is relevant to understand, prevent, diagnose, and treat the existing epidemic of metabolic disorders such as type 2 diabetes and obesity. Microbiota is a key player in health outcomes due to the potential myriad of metabolites that can produce and interact with any cell of our body through systemic circulation. Those metabolites are coming from direct transformations of nutrients available in the gut microbiota or from secondary transformed host products. The link between epigenetic marks and gut microbes appears to be mediated by host-microbial metabolites that act as substrates and cofactors for key epigenetic enzymes in the host. A disruption in the composition of the gut microbiota may lead to unbalanced key metabolites that sequentially may impact epigenetic pathways and alter gene expression. The implementation of multi-omic approaches to study the human holobiont will facilitate the stratification of patients toward a personalized-oriented care, improving disease prevention, diagnostics, drug election, and treatment efficiency.

19All datasets analyzed for this study are cited in the manuscript and/or the supplementary files.

20JM and OY designed, wrote, revised, and approved the submitted version of the manuscript.

21This work was supported by the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No 675610. OY thanks the following bodies for funding: Ministerio de Economia y Competitividad (MINECO) (BFU2017-87958-P) and the Spanish Biomedical Research Centre in Diabetes and Associated Metabolic Disorders (CIBERDEM), an initiative of Instituto de Investigacion Carlos III (ISCIII).

22The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.