Bacterial competition occurs as interference competition (killing of target cells with antibacterial weapons) or exploitation competition (consuming nutrients from the milieu) (Chassaing and Cascales, 2018). As an anti-bacterial weapon, T6SSs were primarily considered to mediate interference competition in the intestinal microbiota. The antibacterial function of T6SSs relies primarily on injection of “anti-bacterial” effectors that target conserved, essential features of the bacterial cell, such as peptidoglycan (Russell et al., 2011), membrane phospholipids (Russell et al., 2013), nucleic acids (Ma L.S. et al., 2014), NAD⁺ (Whitney et al., 2015), and the critical cell division protein FtsZ (Ting et al., 2018).

Commensal bacteria form a protective barrier that protects the host from bacterial pathogens (Belkaid and Hand, 2014). In healthy adult individuals, Bacteroidales are the most abundant order of bacteria in the intestinal microbiota (Faith et al., 2013). Recently, by performing extensive bioinformatics analyses and creating hidden Markov models for Bacteroidales Tss proteins, Coyne and colleagues identified 130 T6SS loci within 205 human gut Bacteroidales genomes. As Bacteroidales comprise approximately 50% of all colonic bacteria in many people, this suggested that T6SSs are distributed in about 25% of bacteria in the human colon (Coyne et al., 2014; Coyne et al., 2016). Moreover, more than 10⁹ T6SS-firing events were determined per minute per gram of colonic contents, further supporting the importance of this weapon in shaping gut microbiota composition (Wexler et al., 2016). T6SS loci of the human gut Bacteroidales species segregate into three distinct genetic architectures (GA), termed GA1-GA3. GA1 and GA2 loci are present on conserved integrative conjugative elements (ICE) and are transferred between diverse human gut Bacteroidales strains via ICE. GA3 loci are not contained on conserved ICE and are confined to Bacteroides fragilis (Coyne et al., 2014, 2016). The GA3 T6SSs of B. fragilis antagonizes human gut Bacteroidales species in vivo using previously undescribed effectors, likely to create a locally protected niche in the human gut (Wexler et al., 2016). The interbacterial interactions among symbiotic Bacteroidales species could be predicted according to the presence or absence of strain-specific effector/immunity (E/I) repertoires. Further, some of these strains may avoid contact-dependent killing by accumulating immunity genes to neutralize antibacterial effectors that they do not encode to persist in the gut (Wexler et al., 2016). Importantly, symbiotic non-toxigenic B. fragilis strains could restrict enteric colonization by an enterotoxigenic B. fragilis strain, dependent on a functional T6S, to protect the host against intestinal inflammatory disease, suggesting a novel role of T6SS in colonization resistance (Hecht et al., 2016; Casterline et al., 2017).

Type VI secretion systems are widely distributed and have been proposed to be present in about 25% of all sequenced Gram-negative bacteria (Boyer et al., 2009; Salomon and Orth, 2015), including enteric pathogenic microorganisms, e.g., Vibrio cholerae, Salmonella enterica, Shigella sonnei, and Yersinia pseudotuberculosis. Thus, T6SS mediated antagonism in the mammalian gut both benefits microbiota-mediated colonization resistance by preventing pathogen invasion and facilitates some pathogens to battle resident microbiota to invade the ecosystem and cause disease. Some enteric pathogens could utilize T6SS-mediated antibacterial weapons to kill symbionts and establish within the host gut. For example, S. enterica Serovar Typhimurium uses T6SS to kill commensal Klebsiella oxytoca and enhance colonization of the mouse gut (Sana et al., 2016). With a functional T6SS, S. sonnei showed an advantage in competing against E. coli and Shigella flexneri both in vitro and in vivo, which may explain the dominance and increasing global prevalence of S. sonnei in developed countries worldwide (Anderson et al., 2017). Constitutive T6SS expression provides V. cholerae with an advantage in intra-specific and inter-specific competition, such that T6SS-dependent toxicity toward other bacteria could enhance V. cholerae survival in the environment and/or during colonization of a host (MacIntyre et al., 2010; Unterweger et al., 2012; Fu et al., 2013). Through transcriptome sequencing (RNA-Seq), T6SSs and their associated toxins in the gut symbiont of 28 Snodgrassella alvi strains from diverse Apis and Bombus species were analyzed. T6SS-associated Rhs toxins with antibacterial activity could mediate both intraspecific and interspecific competition among S. alvi strains and other bee gut microbes. Furthermore, extensive recombination and horizontal transfer of toxin/immunity genes between strains in the gut microbiota have resulted in tremendous diversity in their toxin repertoires, which suggest that T6SS-mediated competition may be an important driver of coevolution (Steele et al., 2017). These studies showed that enteric pathogens could use their T6SSs for interbacterial competition in vivo and for niche occupancy.

Besides bacterial competitors, T6SS was also found to target some single cell eukaryotic competitors, including amoebae and fungi. In fact, T6SS was first identified by screening V. cholerae mutants that failed to kill the social amoeba Dictyostelium discoideum, and the lipid-binding effector VasX was found to be required for T6SS-mediated amoeba killing, possibly through plasma membrane perturbations (Pukatzki et al., 2006; Miyata et al., 2011; Zheng et al., 2011). T6SSs in Pseudomonas syringae and Serratia marcescens were reported to be required for competition against other single cell eukaryotic organisms, i.e., yeast and fungi (Haapalainen et al., 2012; Trunk et al., 2018). The first anti-fungal T6SS effectors, Tfe1 and Tfe2, were identified in S. marcescens. Tfe1 causes plasma membrane depolarization without formation of a specific pore, while Tfe2 disrupts nutrient uptake and amino acid metabolism, and induces autophagy (Trunk et al., 2018). Since single cell eukaryotes are important components of the intestinal microbiota (Coyne and Comstock, 2019), it is not surprising that intestinal bacteria might deploy anti-eukaryotic T6SSs to compete for nutrients and space against co-habiting microbial eukaryotes.

As essential micronutrients are involved in a wide range of cellular processes (e.g., DNA replication, respiration, and energy generation), transit of these metal ions is a crucial component of host-microbe interactions, including those in the gastrointestinal tract (Sorbara and Pamer, 2019). For example, E. coli mutants defective in catecholate siderophore production show impaired murine gut colonization (Pi et al., 2012). Without a high-affinity zinc transporter, Campylobacter jejuni is unable to replicate or colonize the gastrointestinal tract, as quantities of zinc in the gastrointestinal tract are reduced by the host (Gielda and DiRita, 2012). Thus, competition for micronutrients in the gut is intense.

Type VI secretion system has been reported to be involved in acquisition of essential micronutrients, such as zinc, manganese, and iron, indicating its novel function in increasing bacterial fitness through competition for essential nutrients. In Pseudomonas aeruginosa, the H3-T6SS secreted effector TseF facilitates iron acquisition by interacting with the iron-chelating molecule PQS (Pseudomonas quinolone signal) (Lin et al., 2017). In Burkholderia thailandensis, T6SS4 secretes zinc- and manganese-scavenging proteins to fulfill the increased cellular demand for these metal ions under oxidative stress (Si et al., 2017a,b). Similarly, in the enteric pathogen Y. pseudotuberculosis, T6SS4 functions to combat host nutritional immunity and multiple adverse stresses by translocating a zinc binding effector YezP (Wang et al., 2015). Distinct from the extensively studied contact-dependent “anti-bacterial” and “anti-eukaryotic” T6SSs, the “metal ion transporting” T6SS secretes metal-binding proteins into the extracellular milieu, independent of cell-cell contact (Si et al., 2017a). Thus, T6SS confers survival advantages to bacterium in niches with multiple bacterial species not only by delivering “anti-bacterial” toxins to kill competing cells for interference competition, but also by enhancing its ability to acquire essential micronutrients for exploitation competition. It will be interesting to determine the physiologic consequence of these metal acquisition T6SSs in the intestinal ecosystem in the future.

Interestingly, expression of Y. pseudotuberculosis T6SS4 was found to be nutrient status-dependently regulated by the stringent response factor RelA and the nutrient-dependent regulator RovM (Song et al., 2015; Yang et al., 2019). Further research is needed to reveal whether T6SS is involved in competition for other nutrients besides metal ions.

Although it is well-known that enteric pathogens could utilize the antibacterial T6SS weapon to eradicate competing microbes in vivo, little is known about whether the T6SS-dependent interactions with commensal bacteria have additional effects on virulence. Recently, Zhao et al. (2018) discovered that the V. cholerae T6SS could compete against commensal E. coli strains in the mouse gut to facilitate colonization. Importantly, they found that this microbial antagonistic interaction could improve fitness of V. cholera by activating host innate immune responses and improving expression of bacterial virulence genes (i.e., cholera toxin) (Zhao et al., 2018). The authors suggested that the innate immune response was induced by bacterial debris derived from lysed commensals, which together with up-regulated cholera toxin (CT), resulted in elevated disease symptoms and increased fitness of V. cholerae as a pathogen. Fast et al. (2018) used the Drosophila melanogaster model of cholera to define the contribution of T6SS to V. cholerae pathogenesis. They found that interactions between T6SS and the gut commensal Acetobacter pasteurianus intensified disease symptoms and accelerated host death. The disease severity was attenuated by inactivation of T6SS, or removal of A. pasteurianus. Interestingly, mutation of the Immune Deficiency (IMD) pathway relieved T6SS dependent lethality, implicating innate defenses in T6SS-mediated host death (Fast et al., 2018). These studies established that interactions between T6SS and commensal bacteria activate innate immune responses and promote V. cholera pathogenesis.

In addition, a few T6SS toxic effectors result in metabolic disorders or diseases in eukaryotic host cells. VgrG-1 in V. cholerae was responsible for T6SS-dependent cytotoxicity of macrophages through covalent cross-linking of host cell actin (Pukatzki et al., 2007; Ma and Mekalanos, 2010). In the diarrheal isolate Aeromonas hydrophila SSU, T6SS-secreted VgrG1 exhibited actin ADP-ribosylating activity, which can induce a rounded phenotype in HeLa cells, resulting in apoptosis (Suarez et al., 2010). A non-VgrG T6SS effector in Edwardsiella tarda, EvpP, inhibited NLRP3 inflammasome activation through repression of the Ca²⁺-dependent MAPK-Jnk pathway (Chen et al., 2017). T6SS effectors with phospholipase activity could cause damage to the cell membrane and accelerate the infection process (Dong et al., 2013; Russell et al., 2013). Notably, a few bacterial pathogens with the T6SS apparatus may spread through the digestive tract but cause diseases outside the gastrointestinal system. For example, in the newborn meningitis E. coli (NMEC) K1 strain, the Hcps of T6SS interacts with human brain microvascular endothelial cells (HBMEC) in a coordinated manner, first binding, then invading, and finally causing apoptosis of HBMEC (Zhou et al., 2012). The T6SS-secreted proteins VgrG1 and VgrG2 in Helicobacter hepaticus increases cellular innate pro-inflammatory responses, resulting in liver disease and intestinal inflammation (Bartonickova et al., 2013).

Recently, Logan and colleagues used a combination of microbial genetics, in vitro experiments, and quantitative in vivo imaging in zebrafish to determine the role of V. cholerae T6SS in gut colonization. They found that V. cholerae can expel resident microbiota of the genus Aeromonas in a T6SS-dependent manner. Unexpectedly, T6SS acted primarily to increase the strength of gut contractions, rather than killing the bacterial competitor. Coupling of T6SS activity to host contractions depended on an actin cross-linking domain (ACD) from the T6SS apparatus. When the ACD domain was deleted, V. cholerae could no longer induce enhanced host contractions, and dense Aeromonas communities remained in the gut. Deleting the ACD domain did not affect the ability of V. cholerae to kill A. veronii in vitro or enter and occupy the host intestine. These findings reveal a novel strategy by which enteric pathogens can modulate host intestinal mechanics to redefine gut communities and highlights the role of the host fluid-mechanical environment in shaping gut population dynamics (Logan et al., 2018). Whether this is a common tactic for gut-colonizing bacteria to invade the intestinal ecosystem especially in human needs to be verified in the future.

Within the intestinal lumen, enteric pathogens encounter severe stresses, such as bile salts, antimicrobial peptides, free fatty acids, enhanced osmolarity, and oxidative stress (Fang et al., 2016). To survive these adverse environments, bacteria have developed various delicate mechanisms. As a versatile molecular machine, T6SS has been found to play crucial roles in the bacterial stress response and contribute to cell survival under multiple environmental challenges. Vibrio anguillarum T6SS is regulated by the general stress response regulator RpoS and is involved in its resistance to hydrogen peroxide, ethanol, and low pH (Weber et al., 2009). Upon activation by RpoS and the global oxidative stress regulator OxyR, enterohaemorrhagic Escherichia coli (EHEC) uses its T6SS to secrete a Mn-containing catalase, KatN, thus providing a higher resistance to reactive oxygen species (ROS) produced by the host and enhancing survival levels in the host (Wan et al., 2017). Y. pseudotuberculosis T6SS4 was required for bacterial survival under osmotic, oxidative, and acidic stress and for resistance to bile salts, and its expression was regulated by various stress response regulators, such as the stationary growth phase stress σ factor RpoS, the global oxidative stress regulator OxyR, and the acid/osmotic regulator OmpR (Gueguen et al., 2013; Zhang W. et al., 2013; Guan et al., 2015; Wang et al., 2015). In V. cholerae O1, activation of T6SS has been shown to be dependent on the osmolarity of the milieu (Ishikawa et al., 2012). In addition, in the intestine, V. cholerae T6SS genes are overexpressed under mucin (intestinal host factors) and microbiota-modified bile salt conditions (Bachmann et al., 2015). Bile salts can also activate expression of T6SS genes in Salmonella typhimurium (Sana et al., 2016) and C. jejuni (Lertpiriyapong et al., 2012). These studies on gene expression regulation suggest that T6SS is expressed in the intestine and could provide resistance to multiple stresses, increasing bacterial adaptability or survival in the intestinal ecosystem.

Recently, it was suggested that T6SS in V. cholerae serves as a predatory weapon and fosters horizontal gene transfer. For incorporation of DNA released by lysed cells into a competent predator cell, natural transformation and evolution can occur (Borgeaud et al., 2015). Thus, T6SS may promote co-evolution of bacterial microflora within their environments, including intestinal ecosystems.

Compelling data showed that H. hepaticus interacts with intestinal epithelial cells (IECs) and employs T6SS to limit within-host growth and virulence (Chow and Mazmanian, 2010). This feature was regarded as T6SS function of anti-virulence (Jani and Cotter, 2010). Nevertheless, this phenomenon is rarely seen.

To summarize, we constructed a schematic diagram showing the functions of T6SS in the intestinal ecosystem (Figure 1).

Bacteroidetes, a phylum of Gram-negative bacteria, is highly abundant in the gastrointestinal tracts of humans and other mammals (Lozupone et al., 2012; Russell et al., 2014; Wexler and Goodman, 2017). Despite Bacteroidales being predominant in the mammalian gut, T6SSs in Bacteroidetes species were not identified until 2014, perhaps because the T6SS in Bacteroidetes does not have sufficient sequence similarity with core T6SS proteins of Proteobacteria for protein-protein comparisons (e.g., BLASTP) or protein-profile comparisons (e.g., Pfam, COG) (Russell et al., 2014; Coyne et al., 2016). Studies have revealed that the T6SS loci are segregated into three distinct genetic architectures (GA1-GA3), among which the T6SS loci GA1 and GA2 are contained on highly similar integrative conjugative element (ICE) (Coyne et al., 2016). Among the numerous mechanisms that compete in the extremely dense ecosystem of the gut, T6SSs are probably very prevalent antagonistic systems in gut Bacteroidales (Wilson et al., 2015; Chatzidaki-Livanis et al., 2016; Coyne et al., 2016; Coyne and Comstock, 2019), as the Bacteroides T6SS genes are widespread in human gut metagenomes (Verster et al., 2017). Given that T6SSs in Bacteroidetes mediate inter-bacterial antagonism against pathogens and in view of the observed stability of Bacteroidetes in the healthy human intestine, it predicts that T6SS in the stable Bacteroidetes community contributes to intestinal homeostasis. We believe that further research on T6SSs in Bacteroidales species will add to our understanding of microbial stability in the gut and to our ability to diagnose and treat gut microbial infections.

Vibrio cholerae is a Gram-negative pathogen consisting of over 200 serogroups that usually cause diarrhoeal diseases ranging from cholera to mild gastroenteritis after ingestion of contaminated food or water containing them (Kitaoka et al., 2011; Unterweger et al., 2014). All V. cholerae strains examined to date contain T6SS gene clusters (Unterweger et al., 2012). This microorganism provided several important and initial discoveries about T6SS, including the definition of the contact-dependent secretion model (Pukatzki et al., 2006; Basler et al., 2012), identification of T6SS-dependent effector-immunity pairs (Dong et al., 2013; Russell et al., 2013; Altindis et al., 2015), and induction of intestinal inflammation through actin cross-linking in host cells (Ma and Mekalanos, 2010). Recently, in vivo analysis presented new insight on the pathogenic mechanism of V. cholerae with T6SS. T6SS in V. cholerae acts on commensal bacteria A. pasteurianus to accelerate death of the Drosophila host (Fast et al., 2018). Another study revealed the roles of V. cholerae T6SS in antagonism against host commensal microbiota in vivo over their niche, which facilitated bacterial colonization of the mice gut (Zhao et al., 2018). Besides killing host gut bacterial symbionts, V. cholerae T6SS could modulate host intestinal mechanics, enhancing intestinal movements that led to expulsion of resident microbiota by the host (Logan et al., 2018).

Research indicated that vpT6SS2 contributed to adhesion of V. parahaemolyticus to host cells (Yu et al., 2012). In addition, the vpT6SS2 effector could induce autophagy in macrophage cells without causing apparent cytotoxicity (Yu et al., 2015).

Experiments suggested that T6SS contributed to pathogenicity, at least in the S. enterica serotypes Gallinarum (Blondel et al., 2013), Enteritidis (Blondel et al., 2010; Troxell, 2018), Typhi (Wang et al., 2011), Dublin (Pezoa et al., 2014), and Typhimurium (Mulder et al., 2012). A recent study provided in vivo evidence that S. typhimurium used T6SS as a weapon to kill commensal bacteria (K. oxytoca) to successfully colonize the mouse gut (Sana et al., 2016).

Studies of Y. pseudotuberculosis revealed roles for T6SS in responses to various stressors and identified corresponding regulators, including OmpR (Gueguen et al., 2013; Zhang W. et al., 2013), OxyR (Wang et al., 2015), ZntR (Wang et al., 2017), RpoS (Guan et al., 2015), RovM (Song et al., 2015), and YpsI/YtbI (Zhang et al., 2011). Notably, T6SS4 in Y. pseudotuberculosis was found to be involved in zinc transportation by secreting a zinc-binding protein YezP to mitigate the impact of detrimental hydroxyl radicals under oxidative stress (Wang et al., 2015). The contribution of T6SS to ion transport was also confirmed in B. thailandensis and P. aeruginosa (Lin et al., 2017; Si et al., 2017a,b).

Escherichia coli are model bacteria that have been studied extensively. Although less abundant than Bacteroidales in the gastrointestinal tract, symbiotic gut E. coli play important roles in colonization resistance against enteric pathogens of the proteobacterial phylum (Coyne and Comstock, 2019). To date, more than 150 E. coli strains have been identified. Most are harmless, but several serotypes have been proposed to be pathogenic and may cause significant diarrheal and extraintestinal diseases (Croxen et al., 2013). The current knowledge regarding the prevalence, the assembly, the regulation, and the roles of the T6SS in E. coli has been reviewed (Journet and Cascales, 2016). It has been suggested that T6SS contributes to pathogenesis in E. coli strains. For example, T6SS-dependent Hcp1 in E. coli K1 induced actin cytoskeleton rearrangement, apoptosis, and release of interleukin-6 (IL-6) and IL-8 in human brain microvascular endothelial cells (HBMEC) (Zhou et al., 2012). Further, T6SS in avian pathogenic E. coli (APEC) strains contributes to APEC pathogenesis (Ma et al., 2018). Several structural proteins of T6SS in E. coli have been studied to clarify the specific functions of these T6SS components (Felisberto-Rodrigues et al., 2011; Aschtgen et al., 2012; Durand et al., 2012). However, whether E. coli gut symbionts have T6SSs that function in colonization resistance against enteric pathogens, and how enteric pathogenic E. coli T6SSs function in the intestinal ecosystem still remain enigmatic.

Notably, the T6SS apparatus has a common evolutionary origin with phage tail-associated protein complexes. Yet, the involvement of phages in the evolution of bacterial T6SS remains unclear.