Colonization resistance: Metabolic warfare as a strategy against pathogenic Enterobacteriaceae

Abstract

The intestine is home to a large and complex bacterial ecosystem (microbiota), which performs multiple beneficial functions for the host, including immune education, nutrition, and protection against invasion by enteric pathogens (colonization resistance). The host and microbiome symbiotic interactions occur in part through metabolic crosstalk. Thus, microbiota members have evolved highly diverse metabolic pathways to inhibit pathogen colonization via activation of protective immune responses and nutrient acquisition and utilization. Conversely, pathogenic Enterobacteriaceae actively induce an inflammation-dependent disruption of the gut microbial ecosystem (dysbiosis) to gain a competitive metabolic advantage against the resident microbiota. This review discusses the recent findings on the crucial role of microbiota metabolites in colonization resistance regulation. Additionally, we summarize metabolic mechanisms used by pathogenic Enterobacteriaceae to outcompete commensal microbes and cause disease.

Introduction

The intestines of the human host are home to a complex ecosystem of microbes, the gut microbiota[1,2]. The concept that the gut microbiota may play a role in human health originates back to antiquity, with Hippocrates stating that “all disease begins in the gut” [3]. Empiric evidence of gut microbes affecting human physiology can be traced to discoveries made by 19^th-century microbiologists such as Theodor Escherich[4]. However, the gut microbiota’s crucial impact on human health and disease could only be truly appreciated in the 21^st century through the development of sequencing-based methods to identify complex microbial communities. Together with germ-free animal models, this approach has revolutionized microbiota research and our understanding of host-microbe interactions beyond infectious diseases.

The gut microbiota is a collection of majority obligate anaerobic bacteria that functions as a microbial-organ[5]. This microbial-organ educates the immune system[6], increases caloric uptake[7], and provides resistance against enteric bacterial pathogens, termed colonization resistance[5]. As a concept, colonization resistance was first introduced in the 1950s by Bohnhoff et al.[8]. This study pioneered the use of the broad-spectrum aminoglycoside antibiotic streptomycin as a method of disrupting the resident populations of gut flora in mice. Disturbance of the gut microbiota by streptomycin significantly decreased the infectious dose of the bacterial pathogen Salmonella enterica serovar Enteritidis leading to higher disease incidence. Since then, multiple groups have sought to understand how this complex consortium of bacteria inhabiting the intestines works alongside the host to protect against intestinal colonization by enteric pathogens.

As a defense mechanism against the invaders, the gut microbiota occupies major nutrient niches and produces inhibitory fermentation products that directly and indirectly limit pathogen expansion in the gastrointestinal tract[5,9]. Enteric pathogens must develop “counter-attack” strategies, allowing these invading microbes to alter the gut environment directly or indirectly to overcome colonization resistance. For instance, Cholera’s causative agent, Vibrio cholerae, has been reported to kill commensal bacteria via its type VI secretion system (T6SS)[10,11]. This secretion system is decorated with VgrG proteins such as VgrG-3, found to break down the cell wall and induce lysis of prey commensal bacteria[12]. Pathogenic Enterobacteriaceae such as Salmonella spp. and Citrobacter rodentium use their type III secretion system (T3SS) to cause intestinal inflammation that provides a metabolic advantage against the resident microbiota[13]. Indeed, the production of electron acceptors by the host inflammatory response enables pathogens to perform aerobic and anaerobic respiration to access carbon sources made available upon gut microbiota disruption (dysbiosis) [14–16]. Research continues to implicate dysbiosis in varying disease states from cancer to autoimmunity, and digestive disorders[17,18]. Thus, it is essential to understand the homeostatic processes that govern the mutualistic relationship between host and resident gut microbes to provide novel insights leading to the development of microbiota-targeted therapeutics for human disease. This review aims to discuss the metabolic communication that confers colonization resistance against enteric bacterial pathogens, and the known strategies used by pathogenic Enterobacteriaceae to overcome colonization resistance.

Short-chain fatty acids: Friends or foes in colonization resistance?

Short-chain fatty acid (SCFAs) production by gut microbes is the most well-characterized beneficial metabolic interaction between the gut microbiota and the human host. SCFAs are fermentation by-products resulting from dietary fiber digestion such as inulin or cellulose [19], by the dominating bacterial classes in the gut microbiota, Bacteroidia, and Clostridia[20,21]. Microbial digestion of dietary fiber results primarily in butyrate, acetate, and propionate in the cecum at around 120-130mmol/kg, with production decreasing distally towards the rectum [20,22]. Intestinal microbiota-derived SCFAs are not just found in the large intestine but can also circulate in the bloodstream, and modulation of dietary fiber intake can affect intestinal and systemic SCFA concentrations [23], Thus, the SCFA’s beneficial effects in colonization resistance are not restricted to their function in the intestinal lumen but can also be explained by their effects on the intestinal mucosa and systemic organs[24] (Fig. 1 and Table 1).

Figure 1. Mechanisms of short-chain fatty acids (SCFA) mediated colonization resistance against pathogenic Enterobacteriaceae.

Fermentation of the dietary fiber by the gut microbiota members promotes mitochondrial β – oxidation by the gut epithelium, inhibits HDACs and stimulates GPCRs 41,43, and 109A promoting resident T-regulatory cell populations in the gut. SCFAs also act directly inhibiting pathogenic Enterobacteriaceae. HDAC – Histone deacetylase; GPCR – G-protein coupled receptor; PPAR-γ – Peroxisome proliferator-activated receptor – gamma.

Table 1.

The effect of gut microbiota-derived metabolites on colonization resistance.

Class	Compound	Producer(s)	Function	Key References
Short-chain fatty acids	Acetate	Bacteroides spp. Bifidobacterium spp. Lactobacillus spp. Prevotella spp. Etc.	Impedes pathogenic E.coli expansion via intracellular methionine pools.	[20,21,35,37,41,43]
	Butyrate	Ruminococcaceae Lachnospiraceae	Impedes S. Typhimurium Maintains anaerobiosis via host epithelium β-oxidation Inhibits HDACs Activates PPAR-γ
	Propionate	Bacteroides spp. Clostridium spp.	Limits S. Typhimuirum expansion Inhibits HDACs
Amino acids	Acidic (Aspartate-D & Glutamate-E)	Dietary sources via proteolysis by the gut microbiota.	Fortifies tight junctions and enterocyte metabolism	[51,52,55,65]
	Nitrogenous side-chain (Arginine-R, Glutamine-Q)		Modulate inflammatory response Expand Firmicutes	[58,69–71]
	Aromatic (Phenylalanine-F, T ryptophan-W, Tyrosine-Y)		Activation of TAARs and AhRs Impedes EHEC & C. rodentium LEE pathogenicity island expression. Fortify tight junctions of gut epithelium	[49, 59–64]
	Sulfur-containing (Methionine-M & Cysteine-C)		Limits respiration of pathogenic facultative anaerobes	[74–76]

SCFA protective roles through the host response

Primarily produced by Firmicutes from acetyl-CoA, butyrate can interact with many host facets. Butyrate has been found to fuel β-oxidation by the mitochondria of the colonic epithelium (colonocytes) via activation of the nuclear receptor peroxisome proliferator receptor γ (PPAR-γ)[25]. Mitochondrial β-oxidation increases oxygen consumption by colonocytes and ultimately maintains anaerobiosis of the distal gastrointestinal tract[25,26], creating the ideal environment for the resident microbiota.

Butyrate and other SCFAs can, directly and indirectly, influence the gastrointestinal tract’s inflammatory status. Research demonstrates that both butyrate and propionate inhibit histone deacetylases (HDACs) in the colonic epithelium, limiting apoptosis and epigenetic chromatin modification[27]. Additionally, many G-protein coupled receptors (GPCRs), present in colonocytes and immune cells, are capable of sensing all three SCFAs. Specifically, GPCR 41, 43, and 109A are responsible for this interaction[28,29]. Signal transduction and antagonism of HDACs results in an enrichment in resident T-regulatory cells, known to be responsible for curtailing immune responses and enhancing the production of anti-inflammatory cytokines such as IL-10[30]. Therefore, the production of SCFAs allows the microbiota to develop a mutually beneficial relationship with the host by decreasing the inflammatory tone of the intestinal mucosa while creating an anaerobic environment necessary for strict anaerobic commensal microbes to thrive.

Many enteric bacterial pathogens, especially those belonging to the family Enterobacteriaceae, are facultative anaerobes[31]. Unlike the dominant members of a healthy microbiota, Enterobacteriaceae can conduct both aerobic and anaerobic respiration and thus can leverage these metabolic strategies to outcompete resident microbes upon induction of inflammation[32,33]. Pathogenic Enterobacteriaceae have been shown to rely on inflammation-generated alternative electron acceptors such as oxygen [34], nitrate [32,33], and tetrathionate [35] for gut expansion. Thus, maintenance of the anaerobic environment and limiting the gut’s inflammatory status acts as a first gate that must be surpassed for enteric pathogens to succeed[36].

SCFA direct effects on pathogenic Enterobacteriaceae

By comparison, acetate is the most abundant SCFA in the intestinal lumen, with concentrations up to three times higher than other SCFAs[24]. Acetate production occurs primarily from either the hydrolysis of acetyl-CoA or the Wood-Ljungdahl pathway[37]. Many enteric commensals produce acetate, including (not limited to) Akkermansia muciniphila, Bacteroides spp., Bifidobacterium spp., Lactobacillus spp., and Prevotella spp.[38]. Acetate has a role in inhibiting the expansion of pathogenic Escherichia coli by depleting intracellular methionine pools[39,40]. Roe et al. suggest that exposure of E. coli to weak acids results in the inhibition of homocysteine catabolism. High ratios of homocysteine to methionine leads to competitive inhibition of methionine tRNA synthetases resulting in a growth defect. After acetate treatment, bacteria supplemented with methionine recover to wild-type growth[39,40]. Salmonellae, on the other hand, have been reported to use acetate as a signaling molecule. Lawhon et al. describe that Salmonella can respond to intestinal acetate via the BarA/SirA two-component system. This system regulates the Salmonella pathogenicity island-1 (SPI-1) via increasing expression of HilA, a OmpR/ToxR family transcriptional regulator. SPI-1 contains genes that encode the type-III secretion system (T3SS-1) used by S. Tm to invade the gut epithelium and cause inflammation. Importantly, BarA/SirA-mediated increase in HilA expression induces expression of T3SS-1 invasion genes and promotes S. Tm invasion of the intestinal mucosa. The resulting intestinal inflammation provides electron acceptors which enable S. Tm to perform respiratory metabolism to outcompete the resident gut microbiota [41].

To understand the effects of butyrate on the luminal expansion and invasion of the intestinal epithelium by pathogenic Salmonellae, Gantois et al. developed a transcriptomic analysis approach. Butyrate exposure was found to downregulate 19 genes in total, with 17 of which being located on the SPI-1. Genes included hilD, encoding the master virulence regulator [42], invF, a transcriptional activator of the molecular effector sopB[43], and the effector sopE2[44]. Interestingly, Salmonella enterica serovar Typhimurium (S. Tm) can overcome the detrimental effects of butyrate in the expression of invasion genes in vivo. Using a mouse model of Salmonella-induced gastroenteritis, Bronner et al. showed that S. Tm uses the YdiQRSTD operon to perform β-oxidation of butyrate upon nitrate exposure and induce intestinal inflammation via intestinal epithelium invasion [45].

Propionate, the last of the three described SCFAs, is found in the highest concentrations in the cecum and proximal colon. Propionate results from both Bacteroides spp. via the succinate pathway and the lactate pathway via Firmicutes such as Clostridium spp.[46,47]. In a 2018 publication, Jacobson et al. investigate Bacteroides spp.’s ability to impede the expansion of S. Tm via SCFAs. The group found that propionate exposure decreased intracellular pH, resulting in reduced division time and maximal growth of the bacterium in vitro. Furthermore, mice inoculated with live Bacteroides spp. had significantly less S. Tm burden two days post-infection (p.i.) compared to those who received heat killed. Thus, suggesting that propionate production by Bacteroides spp. limits S. Tm expansion [47].

Further research is required to understand if this dynamic is all there is to the story. The pathogen in question, S. Tm, is known to harbor necessary enzymes to break down both propionate and an associated molecule 1,2-propanediol[44,48], Jacobson et al. demonstrate that S. Tm is sensitive to propionate exposure, inducing acidification of the cytoplasm, an increase in generation time, and reduced ability to colonize the murine gut. However, S. Tm expresses a microcompartment to shield its DNA from the genotoxic intermediate aldehyde, ultimately producing propionyl-CoA [49]. Propionate utilization converges on this same intermediate, as outlined by Yoo et al., suggesting that the inhibition seen by Jacobson could be only contextually relevant[49]. It is possible that pathogen-induced changes in intestinal environment may allow S. Tm to utilize this SCFA as a carbon source. What parameters would enable utilization of propionate instead of inhibition are yet to be identified. However, one likely avenue is the induction of inflammation necessary for S. Tm to overcome propionate-mediated colonization resistance in the gut.

Amino acids: New allies in the microbiota war against pathogenic Enterobacteriaceae

Aside from SCFA production, the microbiota plays a vital role in dietary peptide proteolysis, liberating essential and non-essential di-peptides and amino acids[50–52]. Recent studies have described a new role for amino acids in conferring colonization resistance indirectly via host immune responses like those of SCFAs [53–55], and directly via interspecies competition and bacterial growth inhibition [40] (Fig. 2 and Table 1). Interestingly, pathogenic Enterobacteriaceae have been found to switch from carbohydrate to amino acid utilization in the inflamed gut[56]. Thus, the healthy microbiota must sequester these potential carbon and nitrogen sources from invaders[36,56]. Additionally, some amino acids are used directly for the production of SCFAs via fermentation by the gut microbes reifying the interweaving metabolic networking that occurs in the gut microbiota to resist colonization by enteric pathogens[57].

Figure 2. Mechanisms of amino acids mediated colonization resistance against pathogenic Enterobacteriaceae.

Proteolysis of dietary peptides empowers the microbiota to impede colonization by pathogenic Enterobacteriaceae directly and indirectly via host immune responses and epithelium barrier functions. Arg – Arginine; Asp – Aspartate; Cys – Cysteine; Gln – Glutamine; Glu – Glutamate; Met – Methionine; Phe – Phenylalanine; Tyr – Tyrosine; Trp – Tryptophan; Ahr – Aryl-hydrocarbon receptor; TAAR – Trace-amine associated receptor.

Amino acids protective role through the host response

Glutamate, a precursor to glutathione (GSH), is a closely monitored amino acid in the gut lumen. Like aspartate, this amino acid can be fed directly into the TCA cycle and is a source of energy for the gut epithelium [55]. Studies have demonstrated that glutamate supplementation reduces epithelial permeability, measured by transepithelial electrical resistance (TEER), and upregulates expression of tight junction proteins such as Claudin-3, Claudin-4, and ZO-3 in vitro[54] (Fig. 2 and Table 1). Decreased epithelial permeability prevents enteric pathogens and pathobionts from reaching systemic sites and causing disease [58]. The respiratory burst associated with an influx of innate immune cells during inflammation produces reactive oxygen species (ROS) [59,60]. GSH is essential in the host response to reactive oxygen species (ROS), by allowing the conversion of cytotoxic hydrogen peroxide to inert water. Glutamate via glutathione production modulates ROS production via the nrf2 pathway[61], and thus may help ameliorate low-grade inflammation in the gastrointestinal tract by cutting off sources of alternative electron acceptors that are necessary for expansion of facultative anaerobes[62].

Aromatic amino acids (AAAs), specifically tryptophan, tyrosine, and phenylalanine, have been implicated in immune education[63] and gut homeostasis [63]. These amino acids are converted into signaling molecules by the gut microbiota through several different enzymatic processes. For instance, indole and indole-like molecules generated via microbiota-dependent tryptophan metabolism regulate epithelium homeostasis, specifically the tight junctions that prevent microbial migration[64]. Expression of AAA converting enzymes is highly dispersed amongst the gut microbiota, and functional redundancy does exist[63]. AAA by-products are sensed via the expression of trace amine-associated receptors (TAARs) and aryl-hydrocarbon receptors (AhRs) by the colonic epithelium and immune system[53,64–67]. TAARs have been implicated in energy metabolism of intestinal epithelial cells [53] (Fig. 2 and Table 1). Indole derivatives have been found to educate and active protective immune responses, via AhR-dependent upregulation of IL-22 secretion by lamina propria lymphocytes [68]. Additionally, indole propionic acid, a ligand for pregnane X receptor (PXR), promotes gut barrier homeostasis via toll-like receptor (TLR4) and IL-10 receptor signaling [69,70]. Kiss et al. found that Innate lymphoid cells (ILCs) which express Ahr, require stimulation by indole-derivatives for the formation of innate lymphoid follicles (ILF) postnatally in mice. Formation of ILF and secretion of IL-22 were required for protection against the enteric pathogen C. rodentium [71]. Further research is required to extrapolate the role of AAAs in colonization resistance against pathogenic Enterobacteriaceae.

Amino acids protective role against enteric pathogens

Aspartate concentrations in the gut lumen are elevated in digestive disorders such as Ulcerative colitis and Crohn’s Disease[72]. This non-essential amino acid is also associated with the initial expansion of S. Tm in the mammalian gut through aspartate-dependent fumarate respiration[73]. As previously stated, one of the gut microbiota’s strategies against enteric infections is intrafamily competition[74]. Like E.coli Nissle 1917, other commensal E.coli species play a key role in limiting the expansion of S. Tm and are known to utilize aspartate for expansion62,63, 64.

Glutamine, closely related in molecular composition to glutamate differing by a terminal amide group, is also tightly regulated in the gut lumen. Glutamine and glutamate are highly abundant in the body, serving as neurotransmitters, fuel, and ligands for many signal transduction cascades [55,62]. Glutamine and glutamate, in concert with arginine, serve as nitrogen sources for the gut epithelium and supplement urea hydrolysis by the gut microbiota [76,77]. Without proper nitrogen, DNA replication and mRNA transcription are altered in Enterobacteriaceae due to environmentally-triggered stress responses [78]. Menezes-Garcia et al. have demonstrated that enterohemorrhagic E.coli (EHEC) and C. rodentium regulate virulence based on the presence of L-arginine[79]. The arginine sensor, ArgR, modulates the expression of the T3SS used by EHEC and C. rodentium to induce inflammation. Interestingly, mutants lacking the arginine transporter ArtP, demonstrated decreased virulence in vivo [79].

Aromatic amino acids, such as tryptophan, play an important role in beneficial host responses [52]. However, they are also capable of working directly against enteric bacterial pathogens. Kumar et al. demonstrate that tryptophan metabolism via Bacteroides thetaiotaomicron leads to the production of indoles (Fig. 2 and Table 1). Indoles have been shown to modulate the expression of virulence genes in EHEC and C. rodentium[80], including the major virulence factor locus of enterocyte effacement (LEE) pathogenicity island[81] encoded by these bacteria. Kumar et al. constructed B. thetaiotaomicron mutants lacking tnaA, which encodes a tryptophanase used for indole production, to demonstrate the relationship between microbiota-derived indoles and the ability of pathogens EHEC and C. rodentium to attach and efface enterocytes. Indeed, mice colonized with B. thetaiotaomicron ΔtnaA harbored more EHEC, an effect that was abrogated upon supplementation with exogenous indoles [80].

Sulfur-containing amino acids, specifically methionine and cysteine, have been shown to fuel hydrogen sulfide production by the gut microbiota[82,83]. Production of hydrogen sulfide via δ-proteobacteria has been implicated in inhibiting pathogenic Enterobacteriaceae expansion[84]. Hydrogen-sulfide specifically inhibits aerobic respiration and promotes the gut microbiota’s colonization resistance after previous insults [82,83]. In a recent publication, Stacy et al. investigate the mechanism of increased colonization resistance to enteric invaders due to history of previous infection[84]. Using a combination of germ-free and conventional mouse models with 16S rRNA microbiota sequencing, the group could identify previous exposure of the microbiota to pathogenic invaders that led to a concerted expansion of δ-proteobacteria. These commensal bacteria can produce H₂S, which conferred resistance towards secondary infection by the pathogenic facultative anaerobe Klebsiella pneumoniae by blocking anaerobic respiration (Fig. 2). Bismuth subsalicylate, the active compound of Pepto-Bismol, was used to verify that gut microbiota derived H₂S inhibits expansion of K. pneumoniae. Interestingly, decreased H₂S availability after bismuth subsalicylate treatment caused a significant expansion of K. pneumoniae in mice with a history of infection. Although the study by Stacy et al. found that the commensal bacteria use taurine for H₂S production, it is possible that δ-proteobacteria use other sulfur-containing amino acids to produce H₂S or obtain taurine via host production of this amino acid from sulfur-containing amino acids [85].

Conclusions

The human gastrointestinal tract is a battleground always at war; whether trying to tamp down the immune response or prevent potential microbial invaders from taking a foothold. The gut microbiota assists in this fight, acting both as front-line soldiers and support for the host. SCFA production, namely butyrate, acetate and propionate activate beneficial host responses, limiting the inflammatory status of the gut. These molecules have also been implicated in directly limiting the expansion of pathogenic Enterobacteriaceae by depleting intracellular methionine pools or promoting intracellular acidification. However, whether pathogens like S. Typhimurium can overcome SCFA-mediated colonization resistance remains unclear, especially considering they harbor necessary enzymes to use them as fuel sources. Secondly, further studies are necessary to determine the impact of SCFA utilization on virulence or global gene expression in vivo.

Similar to SCFAs, amino acids fuel gut epithelial metabolism helping to maintain the anaerobic environment favoring the populations of obligate anaerobes. Amino acid sequestration and interspecies competition prevent expansion of the bacterial invaders by limiting both carbon and nitrogen sources. In turn, pathogens respond to amino acid availability by regulating expression of virulence genes. Dynamics of pathogenic and commensal Enterobacteriaceae in response to microbiota derived-amino acids have yet to be fully explored. These dynamics could help identify novel metabolic targets for the development of therapeutics against enteric bacterial pathogens. This paradigm is supported by data suggesting that pathogenic Enterobacteriaceae alter their metabolic strategy in response to inflammation to take advantage of gut-derived metabolites [86].

Highlights:

• The gut microbiota and host have a symbiotic relationship to maintain homeostasis.

• Short-chain fatty acids work directly and indirectly to inhibit pathogen expansion.

• Amino acid metabolism confers resistance against pathogenic Enterobacteriaceae.

• Metabolic flexibility allows pathogens to overcome colonization resistance.