New discoveries in metagenomics and clinical research have highlighted the importance of the gut microbiota for human health through the regulation of the host immune response and energetic metabolism. The microbiota interacts with host cells in particular by intermingling with the mitochondrial activities. This mitochondria–microbiota cross-talk is intriguing because mitochondria share many common structural and functional features with the prokaryotic world. Several studies reported a correlation between microbiota quality and diversity and mitochondrial function. The mitochondrial production of reactive oxygen species (ROS) plays an important role during the innate immune response and inflammation, and is often targeted by pathogenic bacteria. Data suggest that excessive mitochondrial ROS production may affect ROS signaling induced by the microbiota to regulate the gut epithelial barrier. Finally, the microbiota releases metabolites that can directly interfere with the mitochondrial respiratory chain and ATP production. Short chain fatty acids have beneficial effects on mitochondrial activity. All these data suggest that the microbiota targets mitochondria to regulate its interaction with the host. Imbalance of this targeting may result in a pathogenic state as observed in numerous studies. The challenge to find new treatments will be to find strategies to modulate the quality and diversity of the microbiota rather than acting on microbiota metabolites and microbiota-related factors.
MITOCHONDRIA AND MICROBIOTA: AN INTRIGUING COMMUNE STORY
Recent advances in microbiology and clinical medicine have shed new light on the importance of the microbiota for human health. Composed of a thousand different species, representing 1013 cells, the microbiota plays an important role in the development of a functional intestine, and by helping the digestion of food, provides nutrients for growth and well-being. Colonization of the gut by microorganisms is also necessary for the regulation of a well-balanced immune system (Kamada and Núñez 2014). The gut microbiota interacts with the enteric nervous system and may modulate brain activities (Cryan and Dinan 2012; Foster and McVey Neufeld 2013). Interestingly, several studies have reported the important role of mitochondria during the host/microbiota cross-talk (Walker et al. 2014; Zorov et al. 2014; Lobet, Letesson and Arnould 2015). The aim of this paper is to highlight the particular role of mitochondria during this process.
Common structure and function
Intriguingly, despite this role, mitochondria and bacterial members of microbiota share many features. Not surprisingly, these common features are probably due to the probable prokaryotic origin of mitochondria. Based on the endosymbiotic theory, the ancestor of mitochondria is a member of the alphaproteobacteria phylum that developed a symbiotic relationship with the eukaryotic ancestor cell (Degli Esposti et al. 2014). For example, degraded mitochondrial proteins or mitochondrial DNA can activate formylated protein receptors (FPRs) in the way that microbial formylated proteins do to signal alien proteins in eukaryotic cells (Neish 2013). Both bacterial and mitochondrial membranes can be degraded through similar autophagic systems. Mitochondrial and bacterial ribosomes are more related to each other than either is to eukaryotic ribosomes and are both sensitive to antibiotics (Kalghatgi et al. 2013). Some bacterial proteins can be imported into the host mitochondria due to the similarity of the mitochondrial targeting sequence and bacterial cytoplasmic protein targeting sequence (Lucattini, Likic and Lithgow 2004). The membranes of both are maternally inherited because the microbiota of the newborn is derived from the mother's microbiota and male mitochondria are eliminated during ovocyte fertilization.
Mitochondrial and microbiota DNA both colonize the nuclear genome
The insertion of bacterial and mitochondrial DNA may continuously happened in the nuclear genome of the host cell. Several reports have described the presence of DNA of mitochondrial and bacterial origin in the nuclear genome (Fig. 1). Mitochondrial DNA insertion (also called nuclear DNA sequences of mitochondrial origin, or NUMTs) has been well documented (Ricchetti, Tekaia and Dujon 2004). Transfer of mitochondrial DNA into the nucleus continues to occur in human cells during repair of DNA double-strand breaks. Mitochondrial DNA integrates preferentially into coding or regulating regions, increasing mutation rates and favoring cancer or the inflammatory response (Ricchetti, Tekaia and Dujon 2004). A recent study showed that bacterial DNA sequences can be found in human somatic cells and are enriched in cancer cells (Riley et al. 2013). The mechanism associated with these particular insertions remains unknown. However, other examples of bacterial DNA insertion within the nuclear genome has been previously documented. For example, Agrobacterium tumefaciens injects DNA provoking plant tumor growth and changes in host cell metabolism (Gelvin 2003). These changes induce optimal growth conditions for bacteria. Additionally, studies have demonstrated the ability of Bartonella henselae to integrate its plasmid into human cells in vitro through its type IV secretion system and induce the formation of benign tumors in blood vessels (Schröder et al. 2011). The strong and constant promiscuity of bacteria with host cells is known to favor oxidant stress. One can hypothesize that increased oxidant stress may favor nuclear DNA alteration. Subsequent repair may favor bacterial and mitochondrial DNA insertion. Whether these mechanisms of mitochondria and/or microbiota DNA integration in the nuclear genome of eukaryotic cells are similar or not is still unclear. However, such insertions may induce mutations, at least in somatic tissue, and may provoke cancer as previously observed.
Microbiota and pathogenic bacteria target host cell through ROS regulation and DNA insertion. Commensal and pathogenic bacteria release factors that modulate cell ROS concentration by acting on mitochondrial activity. Pathogen-associated molecular patterns activate the pattern recognition receptors (PRR) and induce mitochondrial ROS production and nuclear gene expression. In parallel, commensal bacteria release formylated protein recognized by the formylated protein receptor (FPR) that activates NADPH oxidase (NOX) and increases cytoplasmic ROS that are sensed by redox sensor proteins. High ROS production is able to trigger an inflammatory response and increases cell oxidative stress. Furthermore, cell stress can trigger mitochondrial and bacterial DNA insertion in the nuclear genome leading to alteration of cellular gene expression. (1) Arrow 1: mitochondrial DNA insertion into the nucleus. (2) Arrow 2: bacterial DNA insertion into the nucleus.
A mitochondria–microbiota inter-talk may be crucial for human health
Several reports show that syndromes like obesity, diabetes mellitus, Crohn's disease or even autism and depression are associated with specific microbiota composition, in particular the differential level of Bacteroides and Firmicutes phyla (Prakash et al. 2011). The mechanisms regulating microbiota quality and diversity are clearly multifactorial, including diet, presence of pathogens, resistance to stress or general health conditions. Several observations have shed light on interactions between mitochondrial function and microbiota quality and diversity. A recent report assessed the association of single nucleotide polymorphism (SNP) of mitochondrial DNA haplogroups and their association with specific microbiotal composition (Ma et al. 2014). The study shows that among SNPs of 89 European subjects, polymorphism in ND5, CYTB genes or the D-LOOP region are strongly associated with specific microbiota composition. Moreover, some mitochondrial disorders have been associated with increased rate of infection (Walker et al. 2014). Patients with mitochondrial neurogastrointestinal encephalomyopathy or carnitine palmitoyltransferase 1A deficiency are more prone to bacterial infection than the general population (Garone, Tadesse and Hirano 2011; Gessner et al. 2013). A new example of mitochondria–microbiota functional interaction has been recently published: rats fed with human milk compared with cow's milk or donkey's milk display higher energy efficiency associated with change in quality and diversity of their microbiota (Trinchese et al. 2015). These modification of mitochondrial energy metabolism are associated with an increased production of butyrate known to be produced by microbiota and to enter the TCA cycle (see paragraph below). This suggests that diet can modulate mitochondrial function related depending on the quality and diversity of microbiota. All these data suggest that mitochondria play an important role during the interaction of the microbiota with the host cell. Moreover, mitochondrial activity may be an important factor that modulates microbiota diversity and quality, probably due to the role of mitochondria during the inflammatory and immune responses.
REACTIVE OXYGEN SPECIES (ROS) SIGNALING – A KEYSTONE OF THE IMMUNE SYSTEM AND INFLAMMATION – INVOLVES MITOCHONDRIAL FUNCTIONS
Immune system and mitochondria
ROS cell concentration is determinant for the innate immune response. Mitochondria are the main source of cellular ROS and its concentration is directly correlated to the activity of the electron transfer chain. Depending on its level in the cell, the ROS concentration can induce cell proliferation and differentiation, cytokine release, or cell death by apoptosis.
Pathogenic bacteria release lipopolysaccharides (LPS), flagelin, lipoteichoic acid, lipoprotein or other toxins, known as pathogen-associated molecular patterns (PAMPs), which the host cell recognizes by means of the pattern recognition receptor (PRR) system through different pathways (Fig. 1). Four different classes of PRR receptors sense the microbiota's factors: Tol-like receptor (TLR), Rig-1-like receptor (RLR), Nod-like receptor (NLR) and C-type lectin receptor (CLR). They generate downstream signals and induce activation of the nuclear factor (NF)-κB pathway and inflammatory response to release of pro-inflammatory cytokines and antibacterial factors (Weissig and Guzman-Villanueva 2015). NLR and TLR tend to increase reactive oxygen species (ROS) production by the mitochondrial respiratory chain. In macrophages, LPS via TLR pathways reduce the expression of uncoupling protein 2 (UCP2) and increase the activity of the electron transfer chain resulting in an increase in mitochondrial ROS production (Emre and Nübel 2010). Furthermore, activation of TLR induces the translocation of the proteinTRAF6 into mitochondria and its subsequent association with ECSIT (evolutionarily conserved signaling intermediate in Toll). This protein increases mitochondrial electron transfer chain assembly and the resultant increase of mitochondrial ROS production (West et al. 2011). The most described member of the NLR family is NLRP3. Activation of this protein, partly by mitochondrial ROS, induces its re-localization from endoplasmic reticulum to mitochondria and allows the activity of the inflammosome (Lobet, Letesson and Arnould 2015).
Activation of the adaptive immune system generally increases ATP production in both lymphocytes B and T, in order to switch from quiescent state to proliferation and differentiated states. In lymphocyte T, this higher ATP production is usually due to high glycolytic activity and mitochondrial fatty acid oxidation associated with a reduction of electron transfer chain gene expression (Walker et al. 2014). This high glycolytic activity related to the Warburg effect allows the high production of different precursors involved in the biosynthesis of NADPH, amino acids, nucleotides and fatty acids. On the other hand, stimulation of lymphocyte B leads to the upregulation of both glycolysis and oxidative phosphorylation in order to allow the production of IgG or IgA antibodies.
Pathogenic bacteria target mitochondria
Interaction of the host cell with pathogenic bacteria induces several effects depending on the cell type (colonocyte, dendritic cell, macrophage…) and the PRR system that is activated. However, mitochondria are often targeted by pathogenic bacteria. For example Listeria infection is associated with fragmentation of the mitochondrial network (Lebreton, Stavru and Cossart 2015). To overcome the mitochondrial effect on the immune response, numerous bacterial species of microbiota tend to directly reduce mitochondrial ROS production (Lobet, Letesson and Arnould 2015). Mycobacterium tuberculosis downregulates the LPS-mediated signaling pathway. Other microbial toxins can upregulate the activity of the detoxification enzyme mitochondrial superoxide dismutase (MnSOD), which results in a lower ROS content and reduces host cell apoptosis, as observed in Ehrlichia chaffeensis (Liu et al. 2012).
Microbiota ROS signaling modulates the gut epithelial barrier
The presence of commensal bacteria is crucial to reduce the effect of pathogenic bacteria. Pathogenic bacteria must compete with commensal bacteria for dietary products and commensal bacteria boost the immune system (for review see Kamada and Núñez 2014). Furthermore, commensal bacteria are an important regulator of the gut epithelial barrier function. Luminal bacteria produce and release small formylated peptides that bind to the formyl peptide receptor. These receptors form a distinct class of PRR, located in the apical surface of gut epithelia. Activation of this receptor triggers the production of superoxide anion by NADPH oxidase 1 (NOX1). This results in an increased level of cellular ROS, independently of mitochondria, and induces the activity of redox sensor regulatory proteins. These proteins modulate signaling pathways, cell motility, immune suppression and epithelial cell proliferation (for review see Neish and Jones 2014). These activities are necessary to ensure epithelial barrier function and to induce anti-inflammatory cytokines such as IL-10.
Interestingly mitochondrial ROS may also be involved in the regulation of the gut epithelial barrier. First, mitochondria also release protein or nucleotides that can activate the FPRs. Secondly, induced mitochondrial dysfunction, using dinitrophenol, affects epithelial barrier dysfunction allowing transepithelial flux of Escherichia coli (Wang et al. 2014). Moreover, addition of mitochondria-targeted antioxidant suppresses epithelial barrier dysfunction. These results suggest that dinitrophenol increases mitochondrial ROS production and directly affects epithelial barrier function leading to a state of gut inflammation. In addition, the resultant gut inflammation is associated with structurally abnormal mitochondria in patient tissue (Nazli et al. 2004).
Altogether these data suggest that mitochondria are a key element in the regulation of the immune system and inflammatory process in particular through the production of ROS and its downstream effect on cell gene expression.
METABOLITES PRODUCED BY GUT MICROBIOTA MODULATE ENERGY METABOLISM
H2S and NO released by the microbiota are inhibitors of mitochondrial host respiratory chain
The most direct evidence of a mitochondrial–microbiota interaction came from a study about metabolites produced by intestinal flora. Leschelle et al. have shown that several enteric bacteria (E. coli, Salmonella) can produce a large quantity of hydrogen sulfide (H2S) due to the degradation of sulfur amino acids in the gut. An elevated concentration of H2S is known to inhibit cytochrome oxidase, one of the major complexes of the mitochondrial respiratory chain (Fig. 2) (Leschelle et al. 2005). However, small concentrations of H2S may have a positive effect on mitochondrial respiratory chain activity. Some bacteria are able to reduce H2S into sulfide. Sulfide can be metabolized by the colonocyte mitochondrial respiratory chain via the sulfide quinone reductase (Goubern et al. 2007). These data show that the microbiota can directly regulate oxidative phosphorylation activity depending on the level of H2S present in the colon. It also suggests that the presence of bacteria in the gut microbiota able to produce sulfide from H2S may be a target to improve bacteria–mitochondria cross-talk and helpful for the treatment of metabolic disease.
Microbiota release metabolites that promote or decrease mitochondrial energy metabolism. Nitric oxide (NO) is able to inhibit the tricarboxylic acid cycle (TCA) by reducing acetyl-CoA production. In addition high production of hydrogen sulfide (H2S) by the microbiota inhibit complex IV of the electron transfer chain (ETC). In contrast, short chain fatty acids (SCFAs), in particular butyrate, are able to fuel the TCA cycle. In parallel, SCFAs can induce release of anti-inflammatory IL-10 cytokines and signaling hormone GLP-1 to reduce energy intake.
Interestingly, nitric oxide (NO) also impaired energy metabolism by reducing acetyl-CoA production. NO is produced by the host during inflammation due to L-arginine conversion or nitrite reduction. Recent data showed that NO can also be produced by microbiota. Interestingly, dietary thiol compounds may increase H2S production and in combination with nitrate contribute to NO production (Vermeiren et al. 2012). High NO production in the gut may affect host mitochondrial activity and favor bacterial infection as previously explained.
Based on these data, the release of metabolites by microbiota is dependent on the diet of the subjects and the composition of the microbiota. These results highlighted the importance of dietary compounds to regulate microbiota activity that in turn modulates mitochondrial energy metabolism.
The microbiota's release of short chain fatty acids positively regulates gut function and homeostasis
In a similar way, recent studies have shown that the microbiota produced short chain fatty acids (SCFAs) such as butyrate or propionate from the fermentation of dietary fiber (Kumar et al. 2009). Interestingly, butyrate is known to be used as a source of carbon by colonocytes. Indeed, butyrate can enter the TCA cycle to reduce NAD+ to NADH, a donor of the mitochondrial electron transfer chain (Fig. 2). Notably, butyrate can be used as the only source of carbon by colonocyte mitochondria even in the presence of glucose (Donohoe et al. 2011). Moreover, butyrate not only regulates mitochondrial activity but also promotes release of signaling hormones such as GLP-1 that favor a lower food intake (Yadav et al. 2013). In line with this observation, the addition of butyrate to a high fat diet given to mice prevents the induced obesity generally observed (Lin et al. 2012). The receptor FFAR3 has been identified as a SCFA receptor expressed in GLP-1-secreting endocrine L cells. The diversity and quality of the microbiota and the degree of methylation of the FFAR3 promoter were significantly lower in the obese and type 2 diabetic patients compared to lean individuals (Remely et al. 2014). These data demonstrate a correlation between a higher body mass index and a lower methylation of FFAR3 promoter. In addition it has been reported that butyrate and propionate promote the generation of peripheral regulatory T-cells (Furusawa et al. 2013). Butyrate and propionate are known to inhibit histone deacetylase. The presence of SCFAs enhances histone H3 acetylation in the promoter of the Foxp3 locus and reduces the development of colitis.
These previous data suggest that short chain fatty acids not only induce the release of hormones that reduce food intake, but also seem to increase metabolic rate and regulate the immune system and inflammation. This example highlights the mitochondria–microbiota direct interaction for the regulation of energy metabolism.
CONCLUSION
The role of mitochondria during the host microbiota cross-talk is essential in order to modulate the innate immune response. Microbiota species tend to control mitochondrial activity in order to favor interaction and infection. Indeed, the response of the host cell toward the presence of microbiota is dependent on the presence of factors released by the microbiota that will increase (SCFAs…) or decrease (NO; MnSOD…) mitochondrial activity and ROS production. Unknown mechanisms by a variety of metabolites originating from the microbiota may be relevant for mitochondrial homeostasis and remain to be discovered. The balance between these factors may trigger an adequate host response. Differences in microbiota quality and diversity have been associated with several diseases including bowel inflammatory disease and obesity (Turnbaugh et al. 2006; Sartor and Mazmanian 2012). Alternatively, based on current available data, bacterial species can also trigger insertion of bacterial or mitochondrial DNA within the host genome and induce mutation of the somatic cell independently of mitochondria.
Consistent with these effects, it is tempting to think that targeting the microbiota could be useful to manage intestinal ROS, oxidative stress, inflammation and metabolic anomalies due to the alteration of the microbiota as we previously reported (Edeas and Weissig 2013; Weissig and Edeas 2015a,b). The perspective will be to modulate the quality and diversity of the microbiota of each person rather than acting on the microbiota metabolites and the microbiota-related factors (NO, H2S, SCFAs). Probiotics, diet or fecal transplantation are new emerging strategies to modulate the quality and diversity of the microbiota.
FUNDING
This project is supported by the World Mitochondria Society.
Conflict of interest. None declared.
REFERENCES
