The Biology of Akkermansia

Abstract

Members of the genus Akkermansia are the only known representatives of the Verrucomicrobiota within the gastrointestinal (GI) tract. Akkermansia muciniphila, the best-characterized representative of the genus, is a mucin-degrading specialist that has emerged as a microbe of significant interest due to its influence on the health of its hosts. We describe emerging themes in the biology of Akkermansia species, including species diversity, their cellular structures, physiology, interactions with other intestinal microbes, responses to diet, and effects on mammalian hosts, particularly their role in modulating immune responses. We also summarize some tools available to explore the molecular biology of Akkermansia and discuss its increasingly complex interactions with other members of the microbiota and their implications for gastrointestinal health.

Introduction

Members of the genus Akkermansia, the only known representatives of the Verrucomicrobiota phylum within gastrointestinal (GI) tracts, have garnered significant interest due to their ability to modulate host health and disease. Akkermansia muciniphila, the most studied representative of the genus, is a Gram-negative, obligate anaerobic, rod-shaped bacterium with a unique capacity to degrade mucin. This review will outline what is known about the biology of Akkermansia sp, including species diversity, cell structures, physiology, ecology, and impact on their hosts (Figure 1).

Figure 1. Akkermansia biology.

This graphic summarizes the key aspects of Akkermansia cellular structures, its colonization of mucus layers in the gastrointestinal tract, interactions with the microbiota and the immune system, and its impact on human health. Image created with BioRender.

Akkermansia and the PVC superphylum

The Verrucomicrobiota phylum (formerly Verrucomicrobia) is part of the PVC superphylum based on 16S rRNA phylogenetics and the level of conservation among core proteins (reviewed in (121)). The phylogenetic grouping of the Planctomycetota, Verrucomicrobiota, Chlamydiota, Lentisphaerota, Kirimatiellota, and Candidatus Omnitrophota into the PVC superphylum highlights their potential marine ancestry (28). The biology of members of the PVC group is relatively underexplored, and critical findings in these microbes will provide insights into prokaryotic evolution, pathogenesis, and commensalism.

Members of the PVC group display unique external and internal membrane structures. For example, Verrucomicrobium spinosum, the first identified member of the Verrucomicrobiales order, has many fimbriae extending out of multiple prosthecae that resemble warts, resulting in its naming (127). Early observations among PVC members led to the proposal of intracellular membrane-bound compartments (83). However, three-dimensional reconstructions of the planctomycete Gemmata obscuriglobus indicate that these structures consisted of invaginations contiguous with the cytoplasmic membrane (125). Thus, except for anammox bacteria, which have membrane-defined compartments for the highly toxic anammox reaction (100), PVC membranes are more consistent with those of Gram-negative bacteria (27). However, abnormally distended periplasms, termed paryphoplasms in planctomycetes (83), represent intracellular structures for an energy-dependent uptake of glycans and proteins for intracellular degradation (87, 11). A. muciniphila also selectively accumulates mucin glycoproteins into a distended periplasmic space termed “mucinosome” (20). Indeed, degradation of complex polysaccharides is shared among many marine and fresh water Verrucomicrobiota that degrade algal polysaccharides such as fucoidan (134). Similarly, metagenomic analyses of Verrucomicrobiota in ruminal animals suggest they can degrade lignocellulose (reviewed in (41)), and Verrucomicrobiota from extreme environments contribute to polysaccharide degradation and carbon cycling (101).

A. muciniphila was first isolated from human feces in media containing type III pig gastric mucin (PGM) as the sole carbon and nitrogen source (25). Earlier generations of prokarya-specific 16S rRNA primers failed to detect Verrucomicrobiota sequences (10), but it has since been established that Akkermansia is abundant in fecal samples (9), suggesting an adaptation to life in the GI tract. Akkermansia species have been cultured from various vertebrates (106). With the expansion of Akkermansia isolates and additional phylogenetic and phenotypic analyses, the current proposed list of species within the Akkermansia genus includes the species A. biwaensis (66), A. durhamii (98), A. glycaniphila (106), A. ignis (60), A. massiliensis (99), and A. muciniphila, which is further subdivided into subspecies communis and muciniphila (25, 98). Pairwise analysis of the genomic average nucleotide identities of the proposed species and subspecies fits the current standards for species (95% ANI threshold) and subspecies (98% ANI) definitions (98). Additional potential species are likely present within clade AmIII (45), and candidatus species have been proposed from fecal metagenomic reconstructions. Throughout the review, we highlight some of the phenotypic diversity observed across Akkermansia species and strains, including differences in metabolic pathways such as assimilatory sulfate reduction and cobalamin biosynthesis. Just as understanding how phyla and superphyla grouping provides insight into shared biology, distinct species and strain-level features offer insights into Akkermansia’s unique biology and its impacts on microbiome ecology and host physiology (98).

The Akkermansia bacterial cell surface

Few studies have focused on characterizing the Akkermansia bacterial cell. Genomic information has identified predicted ORFs that encode proteins relevant to cell surface structures, although a significant portion of Akkermansia genomes encode proteins of unknown function. Figure 2 depicts our current understanding of A. muciniphila cell surface structures, with most work performed in the Muc^T (ATCC BAA-835) isolate. Akkermansia sp. are Gram-negative bacteria ranging from 0.5-1.0 μm (25, 66, 106) and display different morphologies depending on the growth media used (25, 82, 85). Cells often grow as single bacilli but can also grow as short chains or aggregates (25), with doubling times in porcine gastric mucin ranging from 0.3h to over 20h, depending on species and strain (7). The major metabolic properties of A. muciniphila have been reviewed elsewhere and will not be covered here (55). A. muciniphila cell envelope-associated structures include:

Figure 2. A. muciniphila cell.

(a) Schematic of A. muciniphila subsp. muciniciphila cell, highlighting its pili, exopolysaccharide (EPS)/capsule, and paryphoplasm. Inset: TEM image of A. muciniphila Muc^T. Note the EPS/capsule on the bacterial surface. (b) A. muciniphila proteins involved in pili biogenesis and mucin acquisition (Gene ID (Amuc_) (Davey et al. (2023). Inset: Image of A. muciniphila Muc^T (white) with or without pretreatment of CCCP (carbonyl cyanide 3-chlorophenylhydrazone) before the addition of fluorescein-labelled mucin (Fl-mucin, green). The arrow points to a pilus with a mucin focus at the terminus. CCCP dissipates the cell’s proton motive force, resulting in diminished uptake of mucin. Image obtained from Davey et al. (2023); copyright 2023.

Pili

Early electron microscopy images of A. muciniphila indicated the presence of pili (25). Based on sequence and structural homology as well as predicted N-methylphenylalanine domains, the Muc^T genome predicts multiple proteins involved in type IV pili assembly (20, 104). The Amuc_1098 to Amuc_1102 locus encodes the most well-characterized Muc^T type IV-like pili proteins, although other proteins predicted to be involved in pilus formation are encoded elsewhere in the genome. PilQ (Amuc_1098) is likely the secretin for type IV pili in A. muciniphila (104). Proteomic analysis of bacterial fractions separated by sucrose-density gradient centrifugation indicates that Amuc_1099 and proteins in the putative pili gene cluster are highly enriched in membrane fractions and cell envelope proteins (105). Amuc_1100 shares structural homology with PilN and PilO (97, 146), proteins in the alignment subcomplex in Pseudomonas aeruginosa involved in pili formation. Amuc_1100 is also an immunostimulant that has garnered significant attention due to its anti-obesity effects (113). Mul2b (Amuc_1101) belongs to the PilM family of proteins, and Mul2A (Amuc_1102) (20) shares structural similarity with archaeal type IV pilin (153). Significantly, this locus is conserved across Akkermansia species, and genetic evidence indicates that it is required for mucin uptake by A. muciniphila, leading to its designation as MUL2 (20). Transposon mutagenesis identified additional potential mucin utilization loci, including other putative pili proteins, suggesting a broad role for Akkermansia pili in proteoglycan capture (20).

Capsule and exopolysaccharides

A. muciniphila produces exopolysaccharides (EPS) (40) and India ink staining, along with TEM imaging of cells stained with ruthenium red, appears to confirm capsular structures (25, 106, 122). A. muciniphila EPS is mainly composed of fucose, galactose, glucose, and N-acetylglucosamine (GlcNAc) (1:1:2:1) (40). Akkermansia genomes have multiple gene clusters encoding proteins involved in EPS and/or capsule biosynthesis, with A. massiliensis and A. biwaensis encoding two additional capsular biosynthesis loci (7). Hence, capsules and EPS likely vary between different Akkermansia species. Genes predicted to be involved in EPS biosynthesis in Muc^T include Amuc_0636, Amuc_0939-45, Amuc_1139-42, Amuc_1411-14, and Amuc_2077-98, with the Wzx/Wzy-dependent pathway proposed for EPS production (40). Genes in this cluster are differentially expressed under microaerobic conditions (108), in the presence of polyphenols (122), and in media containing glucose (103) or high GlcNAc levels (40). Notably, genes predicted to be involved in EPS/capsule production are required for A. muciniphila’s fitness in the GI tract but not for growth in vitro in mucin (20).

Lipooligosaccharide and peptidoglycan

The A. muciniphila strain Muc^T lacks an O-polysaccharide repeating unit in the lipid A portion of its outer membrane (38). A. muciniphila’s lipooligosaccharide (LOS) contains a 26-sugar unit core oligosaccharide of non-repeating sugar motifs distributed along a hexadecasaccharide and an undecasaccharide chain with distinct fucosylation. In addition, the lipid A portions consist of a heterogeneous mix of differently phosphorylated and acylated lipid A species, including tetra-, penta-, and hexa-acylation (38). These modifications likely impact their ability to be recognized by Toll-like receptors (TLRs), which will be discussed in greater detail below. Because genes in Muc^T are predicted to be involved in LOS biosynthesis and overlap with genes predicted to be required for EPS/capsule biosynthesis (38), it is still unclear whether there are shared oligosaccharide components between both structures.

Like most bacteria, the A. muciniphila cell wall contains peptidoglycan and ββ-lactamases to counter the activity of penicillins (39, 107). Peptidoglycan fragments from A. muciniphila Muc^T display some unusual features, including non-acetylated glucosamine residues and O-acetylated N-acetylmuramic acid subunits (39). Despite these differences, A. muciniphila peptidoglycan fragments can stimulate human NOD1 and NOD2 (39).

Paryphoplasm and mucinosomes

Some bacteria in the PVC superphylum exhibit enlarged periplasmic spaces and membrane invaginations known as paryphoplasms (36), which have also been observed by TEM in A. muciniphila (104). In G. obscuriglobus, high molecular weight dextran and proteins are scavenged from the environment and accumulate in these structures (87, 11). The binding of high molecular weight dextrans to stalk or pili-like fibers is associated with the formation of crateriform structures in G. obscuriglobus and P. limnophila (11). An analogous process occurs in A. muciniphila, where polarized intracellular structures within bacterial cells accumulate fluorescently labeled mucins in an energy-dependent manner. These structures have been termed mucinosomes based on the prediction that this is where mucin degradation products are stored and degraded (20). The intracellular processing of mucin is consistent with earlier observations that mucin degradation is associated with intact bacteria (104), and there is genetic evidence that mutants in mucin hydrolases are not complemented in trans by neighboring cells (20). Furthermore, mutants in the mul loci fail to form mucinosomes (20). Live imaging of fluorescent mucin interaction with A. muciniphila suggests it first binds to the bacterial cell surface as discrete puncta, collapsing into a single focus (20).

Akkermansia interactions with the microbiota

Akkermansia interacts with the rest of the gut microbiota (Figure 3), which places a selective pressure on using mucins as a nutrient source. This is exemplified by the observation that the genetic requirements for A. muciniphila to colonize mice increase drastically as the complexity of the microbiota increases (20).

Figure 3. A. muciniphila ecology.

Interactions between A. muciniphila and other bacteria, phages, and the host in the GI tract. Arrows between bacteria indicate agonisms (green) or antagonisms (red) between A. muciniphila and GI bacteria. Arrows without directionality indicate that interactions may be positive, negative, or unknown. Predicted interactions are based on in vitro studies, mouse models, and human metagenomic data, as indicated. Gray symbols denote that positive and negative interactions have been observed depending on the model system. The mechanisms mediating these interactions remain unclear and are likely context-dependent. GlcNAc, N-acetylglucosamine. The following references were used to construct the figure: 7, 8, 17, 26, 30, 34, 35, 37, 42, 46, 53, 59, 69, 77, 84, 108, 112, 114, 120, 128, 133, 137, 142, 151, 152, 158, 163. ASR, assimilatory sulfate reduction. Akk, Akkermansia.

Intragenus competition:

Analysis of 16S rRNA profiles and human stool metagenomic datasets suggests that a single Akkermansia species typically dominates the microbiota (7, 53, 59). Based on colonization experiments in mice, endogenous Akkermansia species inhibit engraftment by other Akkermansia, and species like A. biwaensis outcompete A. muciniphila in antibiotic-treated animals (7, 128). Additionally, colonization by different strains of subspecies muciniphila is controlled by priority effects, with competition outcomes determined by colonization order (128). While competition for the same nutrients (such as mucin) has been proposed as the mechanism governing competitive exclusion between Akkermansia strains (128), other factors like bacteriophages or expanded nutritional capabilities may provide a selective advantage to new Akkermansia entering a pre-established microbiota. For example, A. massiliensis outcompetes A. muciniphila in gnotobiotic mice, via A. massiliensis-derived extracellular vesicles (53).

Competition between Akkermansia and other gastrointestinal microbes

Akkermansia competes predominantly with mucin-degraders, such as Ruminococcus, Bacteroides, and Allobaculum (126, 142). Indeed, Akkermansia mutants that cannot utilize mucins colonize germ-free mice but fail to engraft in those with intact microbiota (20). A. muciniphila growth is also limited when co-cultured with R. gnavus or R. torques in mucin medium (114). Although co-culture with B. thetaiotaomicron does not inhibit A. muciniphila growth in vitro, A. muciniphila levels decrease in the distal colon during co-colonization of germ-free mice, suggesting that competition occurs in the GI tract (69). A. muciniphila increases the expression of ABC transporters and capsule biosynthesis genes during co-culture with B. thetaiotaomicron, presumably for defense-related functions (69). In human microbiome data, Akkermansia and Allobaculum exhibit an inverse relationship (120) likely linked to A. mucolyticum’s ability to utilize mucin (142). Allobaculum secretes a sialidase that can arrest Akkermansia’s growth in vitro by limiting nutrient access and distorting cell division (141). Significantly, co-colonization alters the immune response to either bacterium individually, with A. muciniphila ameliorating Allobaculum-induced colitis and Allobaculum blunting A. muciniphila-induced antibody responses (120).

Analysis of the interactions between Akkermansia and other mucin-degrading bacteria highlights Akkermansia’s unique approach to mucin degradation. The mucin generalist B. thetaiotaomicron is capable of mucin degradation but will preferentially degrade dietary fiber (26, 69). During mucin degradation, R. torques liberates monosaccharides and O-glycan fragments that can enhance the growth of B. thetaiotaomicron but not A. muciniphila, underscoring the enzymatic differences in how and where mucins are catabolized (126).

The gut microbiota also includes many bacteriophages that can directly (59) and indirectly (54) control Akkermansia abundance. Akkermansia genomes encode prophages (77) and phage defense systems, such as CRISPR/Cas (59) and type 3 BREX defense systems (30), which may limit viral infections. Analysis of the prophages in A. muciniphila (strain YL44) shows that they can be activated in vitro and the mouse GI tract (77, 163). Thus, phages actively infect Akkermansia in animals and modulate Akkermansia levels.

Cooperation of Akkermansia with other gut microbes

A. muciniphila contributes to the gut microbiome’s metabolic web (8, 34). For example, the non-mucin degrading butyrate producers Eubacterium hallii, Anaerostipes caccae, Faecalibacterium praustnizii, Roseburia inulinivorans, and Roseburia intestinalis will grow on mucins as their sole carbon and nitrogen sources when in co-cultured with A. muciniphila (8, 17, 112). A. muciniphila also uses pseudovitamin B12 produced by E. hallii to facilitate the conversion of succinate to propionate, thus resulting in bi-directional metabolic cross-feeding between the two organisms (8). However, not all Lachnospiraceae benefit from co-culture with A. muciniphila. For example, Roseburia faecis and Agathobacter rectalis, which cannot use sialic acid, are undetectable after 24 hours of co-culture with A. muciniphila (133). By metabolizing mucin within intact cells, A. muciniphila likely limits the release of glycans that other members of the microbiota can scavenge. A. muciniphila can also benefit from the presence of other mucin-degrading microbes, such as Phocaeicola vulgatus (formerly known as Bacteroides vulgatus) (158).

The effects of A. muciniphila mucin breakdown on pathogenic microbes are context-dependent. For instance, A. muciniphila metabolites can support the growth of Clostridium difficile when co-cultured in media containing mucin, and A. muciniphila supernatants enhance chemotaxis towards mucin (34). However, daily gavage with A. muciniphila protects mice challenged with C. difficile (152). The impact on other infections is more complex. For example, A. muciniphila mucolysis can exacerbate Salmonella infection in mice (37) , but it can also be protective when live or pasteurized A. muciniphila is administered before infection (84). Other examples of the context-dependent effects of A. muciniphila on enteric pathogens or pathobionts include Citrobacter rodentium and Adherent and Invasive E. coli (AIEC) infections (26, 137, 151).

The physiology of Akkermansia

A. muciniphila can grow in several defined media formulations without mucin. For instance, glucose, GlcNAc, and soy peptone supplemented with threonine support the robust growth of A. muciniphila (20, 25, 79). The primary source of energy generation for A. muciniphila growth is the Embden-Meyerhof-Parnas (EMP) pathway, which converts glucose into pyruvate (20, 79, 85). Despite its ability to grow on other carbon and nitrogen sources, A. muciniphila displays a higher growth rate when grown on mucin (103). Its adaptation to mucin is reflected in physiological changes in vitro and the extensive mucin-degrading machinery encoded by the A. muciniphila genome, including glycoside hydrolases (GHs) and multiple predicted outer membrane proteins (104). Transcriptional profiling studies indicate that ~25% of all coding sequences in A. muciniphila are differentially regulated by mucin (20, 103).

Mucin is a heavily glycosylated protein containing a domain consisting of numerous PTS (proline, threonine, serine) repeats. The serine and threonine residues act as targets for O-linked glycosylation (reviewed in (50)). These residues are first modified by the addition of GalNAc to form core 1-4 glycan structures that are then further processed by adding galactose, GalNAc, or GlcNAc into increasingly complex glycan chains. These chains are capped by sialic acid and fucose, which can be sulfated. The degree of mucin sulfation and fucosylation varies along the gastrointestinal tract (50). Despite this molecular heterogeneity, A. muciniphila can consume different mucin types (6, 32, 103, 133).

Glycoside hydrolases

A. muciniphila encodes an array of GHs that target the heterogeneous mixture of glycan linkages found in mucin (6, 19, 130, 133, 149, 156). A. muciniphila expresses several members of the GH families (Figure 4, Table S1) (6, 69, 133). These enzymes liberate glycans from the mucin backbone and cleave glycan chains, even when sulfation or mucin composition is altered (6, 31, 68, 133). Biochemical characterizations of GH against their substrates, coupled with their predicted locations within bacterial cells, have provided a granular understanding of events that occur during mucin hydrolysis by A. muciniphila (6, 133). Proteins predicted to be exported across the cytoplasmic membrane facilitate the cleavage of β1,3-linked galactose residues, sialic acid, fucose residues, and sulfate groups (6, 133). Additional breakdown of mucin is facilitated by periplasmic GHs that require the prior removal of fucose and sialic acid, suggesting that mucin degradation begins outside of the cell but is mainly contained within the periplasmic space (6), which is consistent with findings showing the formation of mucinosomes (20) and that A. muciniphila mucolytic activity is predominantly associated with intracellular fractions (104). The diversity of GHs in A. muciniphila likely enables the bacteria to target a large spectrum of mucin glycan modifications. Indeed, a genetic screen in A. muciniphila indicated that most individual GHs were not required for growth in mucin media or colonization of mice (20) (Table S1).

Figure 4: Mechanism of mucin degradation by A. muciniphila.

Major A. muciniphila enzymes involved in breaking down host glycans are represented by their locus tag. These can be searched as “Amuc_” followed by the displayed number. Those that cannot act on whole mucin are shown in separate sections, as are those that act on host glycans and human milk oligosaccharides (HMOs). The following references were used to construct the figure: 6, 19, 20, 68, 79, 89, 92, 93, 103, 104, 130, 131, 133, 138, 139, 149, 156. A compendium of mucin-degrading enzymes and their exact specificities can be found in Supplementary Table 1.

Proteases

A. muciniphila can also cleave the protein backbone of mucin. Metallopeptidases (Amuc_0627 (138), Amuc_0908 (130), Amuc_1438 (92), Amuc_1514 (131), OgpA (Amuc_1119) (139)) and an aspartic protease (Amuc_1434) (93) have been identified and functionally characterized. Though glycan chains limit access to the protein backbone (50), these glycoproteases can function in the presence of glycosylated residues, and in some cases even require them (130), indicating that proteolysis of mucin does not require prior deglycosylation.

Degradation of other glycans

A. muciniphila can utilize or modify glycans beyond mucin. For instance, Akkermansia can grow on human milk oligosaccharides (HMOs) similar to mucin glycans (89, 109). HMOs may enable A. muciniphila to establish itself as a resident of the GI tract early in life (18), as HMO degradation capability is relatively rare among GI microbes (89). Interestingly, some Akkermansia species are better at colonizing the infant gut (89), which may be linked to the expanded repertoire of GHs observed in Akkermansia species (6, 20, 109). Colonization by HMO degraders in neonates enables the engraftment of other GI microbes, as metabolites produced by Akkermansia grown in HMOs can be utilized by non-mucin degrading bacteria (33).

A. muciniphila-host metabolite interactions

The major byproducts of Akkermansia mucin catabolism are acetate and propionate, with minor levels of 1,2-propanediol and succinate produced (20, 79, 116). These molecules, especially acetate, subsequently influence host physiology. Circulating acetate levels are associated with increased insulin sensitivity and lower visceral fat levels (52). Propionate and A. muciniphila have been linked to intestinal epithelial cell development and stem cell maintenance (29, 63).

A. muciniphila can convert proline to glutamate, a precursor in the γ-aminobutyric acid (GABA) synthesis pathway (116). GABA is a potent neurotransmitter that modulates the central nervous system and stress responses (4, 67). GABA levels are depleted in germ-free mice, suggesting that the microbiota modulates GABA levels in vivo (67, 75). Although several A. muciniphila strains can synthesize GABA (67) and express enzymes necessary to make both GABA and succinate in vitro; their levels in culture media are low, indicating that most glutamate may ultimately be used to generate propionate (116).

Vitamin B12

Vitamin B12 (cobalamin) is a critical cofactor for methionine synthase and methylmalonyl CoA synthase, which are central to many cellular and organismal functions (reviewed in (44)). The host acquires Vitamin B12 primarily through diet and potentially via the microbiota, although it is unclear whether the host ultimately absorbs microbiota-derived cobalamin (44). The A. muciniphila methylmalonyl-CoA synthase (Amuc_1983 and Amuc_1984) (8) enables the maximum energy generation from mucin by producing propionate from succinate. Indeed, when cobalamin is limited, A. muciniphila strain Muc^T generates succinate during growth in mucin (8, 95). Other Akkermansia species, like A. massiliensis, produce propionate independently of exogenous cobalamin because they synthesize their own (72) and can support the growth of cobalamin auxotrophs (65). INSeq characterization of methylmalonyl-CoA synthase (Amuc_1984) transposon mutants indicates that this enzyme is required for colonization in mice (20), suggesting that maximal energy harvesting from mucin is needed for competition with other members of the microbiota.

Sulfate metabolism

Bacterial assimilatory sulfate reduction (ASR) produces reduced sulfur from inorganic sulfate for biosynthetic pathways (73). Sources of sulfate in the GI include diet (73), sulfated mucins (50), and other microbes (13). A. muciniphila sulfatases can remove sulfate from mucins (6, 133) and A. muciniphila strain Muc^T encodes all the genes required for ASR (Amuc_1294 – Amuc_1301) (7). In contrast, Akkermansia biwaensis and Akkermansia massiliensis species lack ASR components and thus are highly dependent on reduced sulfur for growth in the mucin medium (7). ASR-deficient A. muciniphila mutants colonize germ-free mice and display a growth advantage in conventionally raised mice (7, 20). This suggests that Akkermansia can harvest reduced sulfur from the host and other microbes. Interestingly, an analysis of strains of A. muciniphila subspecies muciniphila indicates that loss of ASR, an energetically intensive process (161), is common (98).

Bile acids

Host bile acid production and profile are influenced by host genetics and gut microbiome composition (136). As primary bile acids are converted into secondary bile acids by GI microbes, changes to the overall microbiome mediated by A. muciniphila can lead to downstream shifts in bile acid production (145). Though sensitive to the primary bile acids CA, GCA, GCDCA and the secondary bile acids GDCA and TDCA, A. muciniphila growth is not inhibited by the secondary bile acid DCA or low amounts of mixed bile (46). A. muciniphila exposed to bile acids increases expression of genes involved in stress response and the production of hopanoids. These steroid-like lipids enhance membrane permeability and confer bile acid tolerance (46).

The influence of diet on Akkermansia abundance

Diet is a key determinant of the abundance and diversity of gut microbes (21). Specific diets, prebiotics, and caloric content are all factors reported to modulate Akkermansia abundance. Because Akkermansia can degrade HMOs it can colonize the infant GI during early development (18). Furthermore, its detection in human colostrum correlates with the content of fucosylated HMOs (1). Different Akkermansia species exhibit varying glycoside hydrolase profiles, with A. biwaensis demonstrating increased growth in the presence of HMOs (109).

Extreme ketogenic diets can increase A. muciniphila abundance and protect against seizures in a mouse model of refractory epilepsy (102). However, the impact of other ketogenic diets on A. muciniphila abundance in a human cohort has been reported to be minor (2), possibly reflecting different diet formulations or the combined influence of the host.

In mouse and rat models, polyphenols from various sources (i.e. cranberry, grapes, cloudberry, peach, curcumin) have been reported to enhance Akkermansia abundance and function as prebiotics (reviewed in (123)). Studies in healthy volunteers revealed an increase in Akkermansia following the intake of polyphenols from European black elderberry (117) or pomegranate extract (81). Pomegranate is a rich source of ellagitannins, a type of tannin that is metabolized by the gut microbiota into ellagic acid and urolithins. A. muciniphila converts these ellagitannins into ellagic acid, which is necessary for their prebiotic effect (51).

Lastly, reducing caloric intake or dietary fiber content can favor the outgrowth of A. muciniphila. Severe nutrient limitation enables Akkermansia to thrive on host glycans with limited microbial competition (143, 151). In the context of fasting (118), low-calorie diets (143), low-fiber diets (111), extreme loss of appetite during chemical and radiation therapy (132), viral infections (74), or anorexia nervosa (90), the increase in mucin degraders is expected to adversely affect gut barrier function. Indeed, blooms in Akkermansia have been associated with increased risks of developing food allergies (111), neutropenic fevers, and graft-versus-host disease (GVHD) following hematopoietic stem cell transplantation (132), as well as increased vulnerability to intestinal pathogens (26, 151) or delayed viral clearance (74). The magnitude of Akkermansia’s expansion is particularly notable in animal models with small, well-defined microbiotas, where it can represent 40% or more of the microbial community under fiber-free diets (26, 111, 151). However, this effect is not as pronounced in specific pathogen-free (SPF) mice (111). It is important to note that an increase in Akkermansia abundance under specific fiber-free diets does not necessarily lead to the induction of inflammatory markers; the interactions of dietary components with both the host and other microbiota members also play a significant role (111).

Akkermansia and its impact on host immunity

Given that most lymphocytes travel through the GI lymphatics, it is unsurprising that the microbiota can significantly affect immune homeostasis. Akkermansia immunomodulation at the GI can impact local and systemic inflammation. Multiple mechanisms have been implicated, including strengthening gut barrier function, directly activating immune receptors, generating specific serum and mucosal antibody responses, and skewing immune cell populations such as regulatory T cells.

Intestinal barrier function

Intestinal epithelial cells maintain a barrier function through tight junction (TJ) proteins like occludin, claudins, and junctional adhesion molecules (JAM) (78). Disruption of TJs is associated with GI infections and intestinal inflammation (78). A. muciniphila’s proximity to the intestinal epithelium (24, 108) enables it to play a prominent role in protecting the GI barrier from damage caused by mucin layer depletion (49) or TJ dysfunction (96).

Paradoxically, while A. muciniphila degrades mucins, its colonization leads to a thicker mucin layer and an increase in mucin-producing goblet cells (94, 111, 151). In addition to promoting mucin production, A. muciniphila stimulates the expression of TJ proteins under both normal (111, 151) and high-fat diets (80, 150). Furthermore, A. muciniphila administration also increases intestinal endocannabinoid levels, which bolster the gut barrier in the context of high-fat diet-induced damage (35). Some of these effects are likely driven more by host responses to A. muciniphila than by its metabolic activity, as similar protection is observed with pasteurized A. muciniphila, its outer membrane vesicles (OMVs), or the administration of the A. muciniphila protein Amuc_1100 (14, 80, 84, 113, 115, 144).

A. muciniphila’s barrier-promoting effects can provide protection in contexts of infection-induced mucin layer and epithelial layer damage. Clostridium difficile-infected mice treated with A. muciniphila showed less intestinal damage, increased MUC2 expression, and higher TJ expression than their untreated counterparts (152). Similarly, in S. Typhimurium infection in mice, administration of A. muciniphila restored mucosal, epithelial, and TJ stability (84).

Activation of innate immune sensors

TLRs are transmembrane proteins that recognize microbe-derived ligands and activate innate immune responses through the Toll/IL-1R signaling pathways. TLR2 and TLR4 regulate the balance between Th1 and Th2 lymphocytes, which is essential for activating adaptive immunity. Akkermansia signaling through these receptors has been proposed as central to their immunomodulatory functions. Amuc_1100 (105, 146) and lipooligosaccharide (LOS) (38) can activate TLR2 and TLR4 in vitro. Additionally, other Akkermansia-derived molecules, such as Lipid A (38), a15:0-i15:0 phosphatidylethanolamine (present in the cell membrane) (5), and the secreted threonyl-tRNA synthetase AmTARS (64) , act as TLR2 agonists. Conversely, Akkermansia-produced ornithine lipids upregulate the expression of ATF3 (Gene Activating transcription factor 3), a negative regulator of TLR4 signaling (160). Notably, the activation of TLR2 and TLR4 can vary among distinct Akkermansia species (7).

TLR activation is regarded as a key mechanism through which Akkermansia can influence health by ameliorating metabolic disorders (113), regulating serotonin via TLR2 to improve depression (16, 147), enhancing the efficacy of immunotherapies (129), and mediating the deworming effect through TLR2 (57). In contrast, Akkermansia is linked to the downregulation of TLR4 and TLR2 in non-alcoholic fatty liver disease (47) and in oral bacterial infections (135). Discrepancies between preclinical and human-based studies (110) make it challenging to define TLR regulation as a generalized mechanism.

Akkermansia can also modulate host immunity systemically by skewing immune cell populations such as T follicular helper cells (3) and regulatory T cells (48, 71, 86, 159). This has important consequences for the hosťs health. For example, A. muciniphila is one of several gut commensals that can predict (23, 98) and improve response to immune checkpoint blockade cancer therapy (124). Oral supplementation with Akkermansia has been shown to improve response to immune checkpoint blockade, likely via Akkermansia-induced secretion of IL-12 by dendritic cells, increasing the percentage of small intestine Th1 cells in mesenteric lymph nodes and MCA-205 sarcoma tumors (124). Alternative mechanisms, such as A. muciniphila’s production of STING-activating cyclic dinucleotides, have been suggested as providing anti-tumor activity (76). Similarly, Akkermansia derived inosine has also been proposed to mediate Akkermansia’s ability to improve the response to immune checkpoint blockade (91). Thus, Akkermansia may modulate systemic immune responses through multiple microbial molecules and immune cell pathways.

Secretion and post-translational modification of antibodies

Immunoglobulin A (IgA) is central to mucosal immunity and is important in shaping the microbiota composition. Akkermansia and other mucophilic bacteria are preferential targets of IgA antibodies, likely due to their proximity to antigen sampling sites in the GI (70). Notably, A. muciniphila deglycosylates IgA subclass 1 (IgA1), specifically desialylating the hinge region of IgA1. This reaction appears to be enhanced in the presence of mucins. While the consequences of this deglycosylation under homeostasis remain unclear, deglycosylated IgA translocates more efficiently across the intestinal epithelium. It is also the target of autoantibody generation, which can lead to IgA nephropathies. In mouse models of these kidney diseases, Akkermansia contributes to forming immune aggregates in the glomerular mesangium and the development of autoimmunity (42).

Finally, Akkermansia is unique among microbiota members as it induces IgG1 and antigen-specific T-cell responses (3, 71). Yet, the consequences and significance of these responses in modulating the immune system are not well understood.

Current models for Akkermansia biology and future perspectives

The increased gene content (~15-20%) in Akkermansia species like A. massiliensis and A. biwaensis, compared to A. muciniphila, indicates that there are additional Akkermansia metabolic properties that contribute to the complexity of host-microbe interactions beyond what is known for A. muciniphila. These include the ability of A. massiliensis to synthesize cobalamin and clade III (65), as well as the expanded use of HMOs like 2’-fucosyllactose by A. biwaensis (109).

The bulk of our understanding of Akkermansia biology has been through the application of various genomic approaches. A. muciniphila transcriptional responses to mucin presence in the media have been characterized at the RNA (20, 26, 62, 85, 103), protein (79, 103, 104) , and metabolite (85) level. When comparing transcriptional responses of Akkermansia to mucin O-glycans under fiber-deficient or fiber-rich diets, no significant gene signature of a shift in substrate utilization was observed (26) , suggesting that mucin is the preferred nutrient source. However, the expression of A. muciniphila enzymes involved in mucin consumption can be regulated by the presence of other members of the microbiota, both non-mucolytic bacteria like Anaerostipes caccae (17) or mucus-degraders such as B. thetaiotaomicron (69). In vivo, Akkermansia prefers colonic to gastric mucins (116). However, most studies in vitro have been conducted using PGM. It remains to be determined whether the structural differences of colonic mucin lead to differential responses in Akkermansia metabolism.

Metabolomic studies of A. muciniphila culture supernatant identified compounds such as the Arg-Lys-His (RKH) peptide, a TLR4 antagonist that protects against sepsis (155); harmaline, an alkaloid that induces BAAT (bile acid-coenzyme A: amino acid N-acyltransferase) expression and offers protection against the severe fever with thrombocytopenia syndrome virus (SFTSV) (154); and palmitoleic acid, which inhibits tumor necrosis factor (TNF) and modulates the immune response to Mycobacterium tuberculosis (15). Similarly, proteomic analysis of A. muciniphila culture supernatants (strain SNUG-61027) led to the discovery of the glucagon-like peptide-1-inducing protein P9, which improves metabolic disease (157), and Amuc_1409, a beneficial effector for intestinal regeneration (58). Future comparative studies will aid in understanding the impact of Akkermansia genetic diversity on these and other important secreted proteins and metabolites on host functions. For instance, Akkermansia supernatants can induce transcription factors and genes involved in lipid metabolism and cellular proliferation in intestinal organoids (88). A. muciniphila can also impact the methylation of mRNA associated with diseases and disorders altered in the cecum of mono-associated mice (56). The impact of Akkermansia on post-transcriptional regulation is poorly understood.

The interactions between Akkermansia and mammalian hosts have been studied in vitro and in vivo. In vitro studies include interactions of Akkermansia or secreted molecules and vesicles with immune cells (5), epithelial monolayers (119), and intestinal organoids (58). Experiments in animals primarily involved mice of different genetic backgrounds or disease predispositions fed different diets (26, 113) or infected with pathogens such as helminths (57), viruses (154), and bacteria (15). These studies have some limitations, as the experimental setup where mice were either mono-colonized, colonized with defined communities, or subjected to repeated gavages such that Akkermansia constituted a significant portion of their microbiota, does not accurately reflect the complex human microbiota. Repetitive gavages, in particular, introduce the variable of repeated exposure to non-physiological levels of the bacteria and non-specific immunostimulation that can act as confounders when interpreting the role of Akkermansia in the intestinal ecosystem.

Host genetics also influence the microbiota composition and how the host responds to the microbiota (43, 140). One study leveraging the mouse Diversity Outbred cohort (12) identified a Quantitative Trait Locus (QTL) associated with A. muciniphila-produced ornithine lipids’ immunomodulatory effects (160). Similarly, QTL analysis identified loci that influence host bile acid profiles, which modulate the abundance of Akkermansia (61).

In humans, Akkermansia abundance has been associated with both positive (i.e., protection from obesity, type 2 diabetes, sepsis, asthma, tuberculosis, chronic kidney disease, alcoholic liver disease) and negative (i.e., multiple sclerosis, Parkinson’s disease, neutropenic fevers, autoimmune nephropathy) health outcomes (reviewed in (110)). Positive clinical trials of A. muciniphila oral administration have been reported in overweight insulin-resistant volunteers (22). Notably, the relationship between A. muciniphila and health status can vary depending on the geographical location of the studied population, underscoring the potential influence of host genetics. For example, studies on autism spectrum disorder (ASD) found increased Akkermansia levels in children with ASD in Ecuador (164), while the opposite has been observed in studies from Australia (148) and China (162).

Genetic manipulation of microbes remains a limiting factor in understanding the function of members of the intestinal microbiota. Until recently, Akkermansia was considered a genetically intractable bacterium. No site-directed mutagenesis method has yet been reported, although random mutagenesis and exogenous expression of genes have been accomplished in A. muciniphila via conjugation of a modified transposon (20, 109).

The growing interest in Akkermansia biology is primarily driven by its potential as a modulator of various diseases, including metabolic disorders, as well as conditions affecting the CNS, liver, and kidneys. However, a comprehensive understanding of its biology requires improvement, particularly regarding its impact on immunity. Such knowledge will enhance our ability to determine how and when this microbe can be utilized therapeutically and ultimately optimize its beneficial potential while minimizing any adverse effects.

Summary Points list.

• Akkermansia is a member of the PVC superphylum and possesses cellular characteristics that may facilitate the intracellular catabolism of complex glycans.

• The genus Akkermansia is diverse, with different species and strains colonizing various vertebrate animals.

• A. muciniphila produces pili transport systems required for mucin transport into intracellular degradation sites.

• Akkermansia encodes numerous glycan hydrolases and proteases to catabolize mucin.

• Akkermansia sp. competes and participates in metabolic networks with other microbes in the GI tract.

• Akkermansia in the gastrointestinal tract can affect local and systemic immunity.

Terms and Definitions

GalNAc

N-Acetylgalactosamine

GlcNAc

N-Acetylglucosamine

EPS

Exopolysaccharides

HMO

Human Milk Oligosaccharide

LOS

Lipooligosaccharide

MUL

Mucin Utilization Locus

PGM

Porcine Gastric Mucin

TJ

Tight Junction

TLR

Toll-Like Receptor