Microbial communities play a significant role in maintaining ecosystems in a healthy homeostasis. These systems are ubiquitous in nature including both the plant and animal kingdoms, and are important in natural processes (eg, marine biogeochemical processes and soil nutrient cycling) [[1], [2], [3]]. The role of these microbial ecosystems in human health is vital, and more influences are being discovered daily. An estimated 500–1,000 microbial species (equaling about 1013 cells) reside in the human body at any time [4,5]. When these ecosystems are in balance, the microbial community supports healthy metabolic activity, immune function, and other essential processes [6].

The various species found in any ecosystem are of variable importance. The term “keystone taxa” was used by Paine in 1969 to describe the hypothesis that certain taxa co-occur in microbial networks and are particularly important for the integrity and structure of the communities where they are found [7]. However, this term was somewhat ambiguous, and there is currently no uniformly accepted definition, especially in microbial ecology. Nonetheless, these taxa play a crucial role in microbial communities, and their disruption can lead to a dramatic shift in the microbiome structure and function (ie, dysbiosis) [8,9]. This phenomena has been demonstrated repeatedly in the gastrointestinal (GI) tract (the subject of this review) and the oral cavity where evidence has highlighted the role of certain gram-negative species (Porphyromonas gingivalis, formerly Bacteroides gingivalis, along with other species) that can cause periodontal disease by altering innate immunity and other functional characteristics [[10], [11], [12]]. Given the importance of these microbes, an enhanced and clarified definition of the keystone taxa in the human ecosystem for microbiologists and clinicians would certainly be useful.

The gastrointestinal tract contains the largest quantity of microbes anywhere in the human body, and they outnumber human GI cells by more than 100-fold [4]. Presently, in the human GI tract, we acknowledge the importance of certain taxonomic groups while admitting that it is unlikely to be a single species that plays the keystone role [8,9]. Bacteroides are known to be major players in the maintenance of eubiosis in the human gastrointestinal tract, showing many connections and positive interactions in the GI microbiota [6,9]. With metagenomics analyses, certain Bacteroides species have been shown to dominate the interspecies interactions in individuals, strongly suggestive of the role of keystone species in maintaining the structure of the microbiome [13].

The human GI tract contains many beneficial microbes as well as microbes that can cause dysbiosis or frank infection; much of the literature focuses on bacteria, though viruses also play a role in dysbiosis or pathogenesis [14,15]. In the GI tract, the use of broad-spectrum antibiotics can significantly disturb the healthy flora of an individual, which is usually dominated by Bacteroidota and Firmicutes, leading to a predominance of Pseudomonadota (formerly known as Proteobacteria) and other deleterious species [16,17] that lack the favorable metabolic processes and immunomodulatory activities of the healthy flora [18].

Optimizing the population of microbes that will result in eubiosis in various human ecosystems, such as the GI tract and the oral cavity, is essential for good health. If we can enhance the population that will result in eubiosis, we need to carefully identify keystone microbe(s) needed to achieve this status. In this narrative review, we propose to expand the understanding that the microbial community of the human GI tract is composed of many beneficial species, with many keystone microbes belonging to the phylum Bacteroidota (formerly known as Bacteroidetes; the name change was made to align with updated rules for nomenclature) [[19], [20], [21]].

Bacteroidota is one of 41 bacterial phyla [22]. The phylum Bacteroidota includes the classes Bacteroidia, Chlorobiota and Fibrobacterobiota and are anaerobic, non-spore forming bacilli. Current classification describes the majority of the gastrointestinal Bacteroidota as belonging to Bacteroidaceae, Prevotellaceae, Rikenellaceae, and Porphyromonoadaceae families [23].

Bacteria belonging to the genus Bacteroides are anaerobic non-sporing gram-negative organisms that are also resistant to bile acids, generally thriving in the gut and generally having a beneficial relationship with the host (Table 1) [24,25]. Bacteroides are abundant in the gastrointestinal tract, reaching up to 1011 cells/g of intestinal material [24].

Bacteroides fragilis was the first member of the genus to be recognized in 1898 (Table 2) [23,26]. Bacteroides thetaiotaomicron was first described in 1912 under the name Bacillus thetaiotaomicron and moved to the genus Bacteroides in 1919 [27,28]. Bacteroides sensu stricto comprises the following species: B. caccae, B. eggerthii, B. fragilis, B. ovatus, B. stercoris, B. thetaiotaomicron, B. uniformis, and B. vulgatus [29,30]. In addition, several former Bacteroides species have been reassigned as Parabacteroides (there are now 15 species; P. distasonis is the type species.) [30] Species of note in these genera include B. fragilis, B. thetaiotaomicron, B. uniformis, B. vulgatus; P. distasonis, and P. goldsteinii (Fig. 1).

Bacteroidota are considered very adaptable organisms [24]. For example, while the genus is generally considered anaerobic, B. fragilis can grow under low O2 (nanomolar) conditions, which could benefit it in establishing an initial infection in oxygenated host tissues [31]. B. thetaiotaomicron adapts its food source to that of its host, preferentially using host-derived polysaccharides and mono- and oligosaccharides from mother's milk (in weaning mice) during the suckling period, and then expanding its metabolism to use plant polysaccharides after weaning [24,32]. B. fragilis also exhibits the ability to use a wide range of polysaccharides for carbohydrate metabolism, likely providing advantages to survive and flourish in the GI tract [24]. Bacteroides further gain a foothold in the GI tract using a system similar to the type VI secretion system to translocate substances to recipient microbes in the GI tract, exporting antimicrobial effectors and antagonizing other GI bacteria [33].

The genome of Bacteroides has been described as “incredibly fluid”, with the organism capable of acquiring, activating, and repressing a variety of phenotypic functions [26]. This plasticity is due to its ability to 1) incorporate genes from other bacteria (including pathogenicity islands) via horizontal gene transfer (HGT) and 2) turn specific genes on or off as needed. HGT is an important means of spreading antimicrobial resistance genes in Bacteroides. The Bacteroides genome contains multiple copies (similar but not identical) of many of its genes; presumably it can utilize whichever version is most beneficial. In a study of 174 clinical isolates of B. fragilis group, there were at least 10 distinct gene sequences encoding for porins and 8 distinct sequences for penicillin-binding proteins, which could affect the organism's antimicrobial susceptibility [34]. Similar evidence exists for multiple versions of genes encoding efflux pumps in B. fragilis, which can be influenced by the organism's environment [35]. The high numbers of Bacteroidota in the intestinal microbiota and the high rates of horizontal gene transfer result in significant impact on the gene transfer in the microbiome between Bacteroidota and other bacteria, leading to important genetic changes in the whole bacterial community, with one example being a widespread resistance to tetracyclines spreading in the community over 30 years [36]. Recently, CRISPR/Cas-based genome editing tools have been developed for Bacteroides, that could facilitate future mechanistic studies of the gut microbiota [37].

In the gastrointestinal microbiota, Bacteroidota have a very broad metabolic potential and are regarded as one of the most stable part of gastrointestinal microbiota [16]. These organisms can rapidly adapt to changes in nutrient availability by sensing the amount of a particular polysaccharide breakdown intermediate, and then regulating the transcription of the corresponding gene cluster, known as polysaccharide utilization loci (PUL) to make different enzymes available for metabolism as needed [38]. These organisms also employ invertible promoters, or DNA promoter regions that can be reversibly inverted, to turn gene expression on or off as needed [39,40]. Utilizing these systems, Bacteroides can metabolize a variety of polysaccharides found in the GI tract. Although many nutrients are absorbed through the gastrointestinal tract, most of the easily utilized simple sugars are absorbed or consumed before they arrive at the colon, where most of the intestinal microbiota reside [41]. With their versatile polysaccharide-utilization machinery, Bacteroides spp. can degrade complex plant polysaccharides such as starch, cellulose, xylans, and pectins, making them the most stable members of the microbiota and able to influence the microbiome as a keystone species [38,[42], [43], [44]].

Via its carbohydrate metabolic chain, gastrointestinal Bacteroidota species produce short-chain fatty acids (SCFA) such as succinate, propionate, acetate, and butyrate as the major end-products [[45], [46], [47]]. The Bacteroidota species also produce many proteases, thereby facilitating protein metabolism [48]. Butyrate is used as the primary energy source for colonic epithelial cells, while propionate and acetate are necessary for lipogenesis and gluconeogenesis in the liver [49,50]. Although metabolic end products of carbohydrate metabolic chain of Bacteroidota are mainly propionate and acetate [45,51], the presence of Bacteroidota species increases butyrate levels in the total microbiota [52], which is consistent with the role of Bacteroidota species as keystone species in the microbiota. Increased production of butyrate is associated with protection against developing colon cancer [53].

The type and proportion of SCFA produced are highly influenced by the carbohydrate substrate available, which also influences the composition of the gut microbiota that consumes the carbohydrates [53]. SCFA are absorbed across the gut epithelium, and can be released into enterohepatic circulation to be metabolized by the liver or reach extra-intestinal organs [48,53].

SCFA produced by gut microbiota appear to have wide ranging influence on the host. In cell- and rodent-based experiments, B. ovatus produced acetate, propionate, isobutyrate, and isovalerate, which were associated with increased levels of the neuro-active compounds glutamine and GABA, highlighting the connection between microbial colonization and neurotransmitter production [45]. In a study of human gut isolates of Bacteroides, almost all isolates produced GABA, over a wide concentration range (depending on precursor availability and concentration) [54]. There is strong preclinical and clinical evidence that the microbiota influences the development and function of the nervous system, potentially affecting brain function and mental health [55].

SCFA are also likely involved in regulating the host's glucose homeostasis and lipid metabolism [48]. The makeup of the gut microbiota is implicated in the pathophysiology of obesity [49]. Reduced abundance of Bacteroides and increased abundance of Firmicutes was associated with obesity in rodent models, likely due to increased efficiency of caloric extraction in certain gut microbial profiles [50]. Some clinical studies show a negative association of type 2 diabetes and presence of Bacteroides, though this is not consistent among all studies of this nature [56]. The presence of B. acidifaciens in a mouse model that exhibits increased insulin sensitivity was associated with glucose homeostasis and an increased energy expenditure, that could suggest a role in preventing metabolic diseases such as diabetes [57].

Bacteroidota play a role in in colonization-resistance of other organisms and maintenance of gut integrity. For example, several species of Bacteroides produce bile salt hydrolases, which deconjugate bile acids that escape enterohepatic recirculation, which is the first step in a two-step process changing primary bile acids into secondary bile acids, the second step being 7α-dehydroxylase (deoxycholic acid [DCA] from cholic acid, and lithocholic acid [LCA] from chenodeoxycholic acid [CDCA]) [58]. It has been shown that taurocholic acid, a conjugated primary bile acid, is a strong germinant of C. difficile spores into vegetative cells [59], while deoxycholic acid, a secondary bile acid, is an inhibitor of C. difficile growth [60]. In fact, bile salt hydrolase activity was shown to be increased in fecal microbiota transplant (FMT)-treated patients who subsequently were protected from recurrent C. difficile infections [61].

The association of Bacteroides with the ability of C. difficile to colonize the gut is not yet clear. In one study, oral gavage with B. thetaiotaomicron significantly attenuated the colonization of C. difficile, with a concomitant significant decrease in the number of infiltrating lymphocytes in the gut mucosa in a mouse model of C. difficile infection (CDI) [62]. In a pathogen-free mouse model of CDI, treatment with a single strain of B. fragilis increased the survival rate of mice with CDI, as well as increased the bacterial diversity and relative abundance of commensal bacteria in the intestine, indicating B. fragilis exerted protective effects by modulating gut microbiota and alleviating barrier destruction, thereby relieving pathogenic colitis triggered by C. difficile [63]. Additionally, in one study, B. thetaiotaomicron inhibited the production of C. difficile toxins A and B in vitro, apparently through cell wall-associated glycans of B. thetaiotaomicron inhibiting cell autolysis (and therefore release of toxin) by C. difficile [64]. Conversely, in other studies, B. thetaiotaomicron was shown to stimulate growth of C. difficile through utilization of succinate [65] and through liberation of sialic acid [66]. These conflicting findings regarding the ability of Bacteroides to interfere with colonization by C. difficile suggest that the genus may possess additional protection mechanisms against C. difficile.

Data from clinical studies also support the beneficial/protective effect of Bacteroides against C. difficile. A report from 59 individuals who underwent a stool assay for C. difficile toxin demonstrated an inverse relationship between the abundance of Bacteroidota and the presence of C. difficile [67]. In another paired study of 6 patients with CDI who received fecal microbiota transplantation (FMT) from 6 paired donors, levels of Bacteroides and other genera increased in FMT recipients, which was accompanied by an increase in secondary bile acids DCA, LCA and ursodeoxycholate [68]. Additionally, levels of the major SCFAs acetate, propionate, and butyrate increased in all patients treated with FMT, reaching similar levels as observed in the donor.

Bacteroides also appear to be influential in other GI-related pathologies. In a cohort of 37 individuals with well-characterized irritable bowel syndrome, the majority of individuals had a gut microbiota composition that was distinct from that of control participants [69]. In these individuals, there were decreased Bacteroidota-associated taxa relative to control participants, and increased Firmicutes-associated taxa.

While they are generally commensal organisms, some Bacteroides spp. can be opportunistic pathogens [[70], [71], [72]]. Various scenarios, including GI disease, trauma, cancer, and GI surgery, may allow Bacteroides to escape their niche in the GI tract, invade other anatomical locations, and cause infection. Virulence factors of B. fragilis that facilitate this invasion include its production of the Bacteroides fragilis toxin (BFT) which increases permeability and induces reactive oxygen species formation, the enzyme neuraminidase that cleaves mucin polysaccharides, and capsular polysaccharides which promotes abscess formation [73].

Bacteroides are most often associated with intra-abdominal infection, usually occurring because the intestinal wall integrity has been compromised [70,71]. If left untreated, these infections can progress to bacteremia [72]. Other possible infections caused by Bacteroides include skin and soft tissue infections, respiratory abscesses, and brain abscesses [71,72,74].

Monomicrobial infections caused by B. fragilis are usually treated with metronidazole due to its generally good activity against anaerobic organisms although this cannot be taken for granted [75]. Because infection associated with B. fragilis is usually polymicrobial, antimicrobial therapy should cover other potential pathogens such as Enterobacteriaceae as well as other anaerobes. Coverage of Pseudomonas aeruginosa and Enterococcus faecalis is also suggested for hospital-acquired intra-abdominal infections [75]. Other antimicrobials that show good activity against Bacteroides are piperacillin-tazobactam, imipenem, and meropenem [76], although there are significant pockets of resistance to these agents in certain geographic regions [74,77,78].

Bacteroides spp. are inherently resistant to penicillins and broad-spectrum cephalosporins due to their production of β-lactamase [79]. Additionally, there is evidence that Bacteroides release outer membrane vesicles containing β-lactamase, that thereby can degrade β-lactam antibiotics remotely and protect other gut microbes from destruction by β-lactam antibiotics [79]. Bacteroides possess a wide variety of antimicrobial resistance mechanisms, including efflux pumps, DNA repair, and reduced iron transport [26,70]. B. fragilis demonstrates increasing rates of antimicrobial resistance, including to metronidazole [26,35].

Bacteroides appear to have protective and beneficial anti-cancer effects, in part due to the effects of their metabolic by-products, including butyrate, on the host [53]. In one animal model, mice with enteritis-associated intestinal cancer were treated with broad-spectrum antimicrobials. Subsequent transplantation of B. fragilis helped guard against weight loss and growth of intestinal tumors that occurred in the control group receiving antimicrobials [80]. B. fragilis seems to have a beneficial effect in animals treated with ipilimumab, including a net benefit of the bacteria on ipilimumab efficacy and reduced histopathological signs of colitis induced by CTLA-4 blockade [81].

On the other hand, individuals who are infected with enterotoxigenic strains of B. fragilis and have subclinical colon inflammation were shown to have significant risk for colon carcinogenesis, as demonstrated by observational studies showing increased number of B. fragilis toxin gene in the colon mucosa of colon cancer patients [82,83] and by animal studies showing tumorigenesis by B. fragilis colonization [84,85]. Another study suggesting crosstalk between Bacteroidota and T cells in the modulation of tumor progression tested the administration of a broad-spectrum cocktail of oral antibiotics in an established mouse model of cancer characterized by injection of an established pancreatic cell line for a clinically-relevant model of pancreatic ductal adenocarcinoma. Gut microbiome depletion – defined as an expected ablation of 16S ribosomal DNA and decrease in relative abundance of Bacteroidota and Firmicutes – led to a significant decrease in subcutaneous tumor burden in pancreatic cancer. On the other hand, the tumor-suppressing effect of gut microbiome depletion was abolished when the subcutaneous experiments were carried out in Rag1 knockout mice lacking mature T (and B) lymphocytes. This suggests that the tumor-decreasing effect of antibiotics required active participation of adaptive immunity [86]. Further study is needed to elucidate the specific species of Bacteroidota producing either the most beneficial anti-carcinogenic metabolome or an inflammatory environment fostering carcinogenesis, respectively.

New evidence is also emerging for the role of diet and the gut microbiota on host immune function [87]. SCFA produced by gut microbes can provide an anti-inflammatory effect in the host, as well as stimulate a variety of functions in the host immune system including production of cytokines, induction of apoptosis, and production of reactive oxygen species [87,88]. One strain of B. vulgatus (FTJS7K1) showed a significant protective role against acute inflammation in a lipopolysaccharide-induced acute intestinal injury mouse model, including restoration of the intestinal microbiota, which was disturbed by the acute injury [89]. Graft-versus-host disease (GVHD) is a proinflammatory disorder that develops from donor T cells following allogenic hematopoietic cell transplantation. In a preclinical mouse GVHD model study, administering B. fragilis enhanced the gut diversity of the mice, reduced the development of acute GVHD, and prevented chronic GVHD [52].

Parabacteroides goldsteinii has been shown to assist in the maturation and development of the immune system [90]. Beta-hexosaminidase, a conserved enzyme in the Bacteroidota, helped protect against inflammation in a mouse colitis model. Specifically, P. goldsteinii promoted the development a subtype of CD4+ effector T cells that regulate adaptive immunity at the intestinal mucosa, thereby playing an important role in homeostasis and inflammation of the intestinal mucosa. Additionally, P. goldsteinii MTS01 was evaluated in the treatment of maternal immune activation-induced autism using an autism-spectrum–like mouse model and ameliorated relevant behaviors [91].

As most of the inflammation and oncogenesis data from Bacteroidota are derived from preclinical studies, it is unknown what conclusions we can draw for their effects on human health, and future studies would be welcome in these areas.

The gastrointestinal tract is composed of billions of micro-organisms, with bacteria being the most studied. As with any ecological system, some organisms play a greater role than others. This impact of the micro-organism is not always in relation to its abundance. Although not the most common phylum, Bacteroidota are key organisms in the human gastrointestinal microbiota. There are several species, including B. thetaiotaomicron, that stand out as essential or keystone species in maintaining a healthy GI microbiota and defending against C. difficile (toxin suppression, sphingolipids, colonization resistance, and others). It is not yet clear what triggers the conversion of the Bacteroides from its positive attributes to its potentially negative effects.

Bacteroidota as part of the gut microbiota, and their broader effects on the human body, demonstrate the keystone role of this phyla and thus the requirement of a broad gut microbial consortium that contains this phylum. The challenge is to ensure the holistic nature of a healthy microbiota whether by beneficial diet, microbiota transplantation, or selected probiotics, and the avoidance of medications known to have deleterious effects on the gut microbiota, such as broad-spectrum antimicrobials and proton-pump inhibitors. The complexity of the essential phyla such as Bacteroidota help suppress those which cause infection or other adverse outcomes in the GI tract. The challenge is to maintain the ecology of the healthy microbiota. For dysbiosis-induced intestinal disease such as CDI, live biotherapeutic products with a broad consortium of bacteria are one means of re-establishing a healthy microbiota after it has been disrupted.