2025 年 72 巻 4 号 論文ID: 7204106
Alginate, a heteropolysaccharide composed of α-L-guluronic acid (G) and β-D-mannuronic acid (M), comprises poly-G, poly-M, and mixed poly-MG regions. Alginate lyases, classified within the polysaccharide lyase (PL) family, degrade alginate into unsaturated saccharides via β-elimination. Due to the abundance of alginate in brown algae, various marine bacteria produce alginate lyases for its assimilation. Recently, alginate lyases have also been identified in gut bacteria such as those of the genus Bacteroides. In this study, we purified an alginate lyase from enrichment culture supernatants containing alginate, using a human fecal sample, and isolated B. xylanisolvens strain MK6803, which can grow on alginate as a sole carbon source-unlike the type strain B. xylanisolvens XB1A. Draft genome sequencing of strain MK6803 revealed an alginate-metabolizing gene cluster encoding three alginate lyases belonging to PL6_1, PL17_2, and PL38, along with a putative oxidoreductase. This gene cluster was shared with B. ovatus CP926 and B. xylanisolvens CL11T00C41, but not with the type strain XB1A. Bacteroides species lacking this gene cluster exhibited no alginate assimilation, even if they possessed genes encoding one or more of the three alginate lyases. This suggests that the presence of the putative oxidoreductase, alongside the lyases, is essential for alginate assimilation in Bacteroides species. Phylogenetic analysis indicated horizontal gene transfer within the genus Bacteroides. These findings highlight the role of alginate metabolism in the adaptation of human gut microbiota.
ABm, alginate-containing culture medium; Abs280, absorbance at 280 nm; ΔAbs235, increase of absorbance at 235 nm; DTT, dithiothreitol; G, α-L-guluronic acid; GBm, glucose-containing culture medium; M, β-D-mannuronic acid; OD600, optical density at 600 nm; PL, polysaccharide lyase; YT, yeast extract- and tryptone-containing.
Alginate, a heteropolysaccharide composed of linear polymerized units of α-L-guluronic acid (G) and its C5 epimer β-D-mannuronic acid (M), is an abundant dietary fiber derived from brown algae such as kelp and seaweed [1, 2]. It is widely used in the food industry as a gelling agent and thickener [3, 4, 5]. The proportions of G and M (G/M ratios) vary among alginate samples, ranging from approximately 0.5 to 2.5 [6, 7]. Alginate consists of poly-G, poly-M, and mixed poly-MG regions.
Alginate lyase, a member of the polysaccharide lyase (PL) family (EC 4.2.2.-), is produced by various organisms, including seaweeds, mollusks, and microbes [8]. This enzyme degrades the 1,4-glycosidic linkages in alginate via a β-elimination reaction, releasing unsaturated saccharides with a double bond between C4 and C5 at the nonreducing end of the polymer [9]. Alginate lyases are classified by substrate specificity into poly-G-specific (EC 4.2.2.11), poly-M-specific (EC 4.2.2.3), and bifunctional types. PLs are categorized into PL families 1 through 44 based on amino acid sequences and are cataloged in the Carbohydrate-Active enZYmes (CAZy) database [10]. As of April 2025, alginate lyases-including both endo- and exo-type enzymes-have been identified in PL families 5, 6, 7, 14, 15, 17, 18, 31, 32, 34, 36, 38, 39, 41, and 44.
Alginate is degraded by intestinal bacteria and metabolized into low-molecular-weight compounds, including short-chain fatty acids [11]. Mathieu et al. reported horizontal gene transfer from marine to intestinal bacteria and suggested that the acquisition of alginate-metabolizing genes facilitates survival and adaptation in the human gut, where alginate and other dietary fibers are abundant [12]. Recently, Rønne et al. identified three alginate lyases in a newly isolated Bacteroides ovatus strain capable of growing on alginate [13]. This research group also showed in another report that the activity of endo-acting alginate lyase A1-I from Sphingomonas sp. A1 supported the growth of B. eggerthii strain DSM 20697 that possesses exo-acting PL6 and PL17 alginate lyase but is incapable of alginate assimilation [14]. In the genomes of Bacteroides species, genes involved in polysaccharide degradation, uptake, and metabolism are often organized into genomic clusters known as polysaccharide utilization loci (PULs) [15].
In this study, we performed enrichment culture using a human fecal sample and successfully isolated B. xylanisolvens strain MK6803, which was capable of growing on alginate as the sole carbon source. Biochemical tests and draft genome sequencing revealed that this isolate exhibited characteristics distinct from those of the B. xylanisolvens type strain XB1A.
Preparation of alginate blocks. Sodium alginate extracted from algae (Eisenia bicyclis) was obtained from Nacalai Tesque Inc. (Kyoto, Japan). G-abundant and M-abundant blocks were prepared following a previously described method [16]. Proton nuclear magnetic resonance (1H NMR) analysis revealed that approximately 90 % of the G-blocks were composed of G residues, whereas the M-abundant block contained around 30 % G residues.
Purification of alginate lyase. This study was approved by the Ethical Committee of Graduate School of Agriculture, Kyoto University (Approval number H29-4) and performed in compliance with the Helsinki Declaration. Feces collected from a healthy volunteer were suspended in 0.15 M NaCl. A 50 µL aliquot of this suspension was inoculated into 100 mL of culture medium consisting of alginate-rich culture medium [17] containing 0.2 % sodium alginate, 0.45 % yeast extract, 0.3 % tryptone, 0.45 % sodium chloride, 0.25 % potassium chloride, 0.045 % magnesium sulfate heptahydrate, 0.04 % potassium dihydrogen phosphate, 0.02 % calcium chloride dihydrate, and 0.04 % L-cysteine. After overnight incubation at 37 °C, the full preculture was transferred to 24 L of the alginate-rich culture medium and cultured at 37 °C for three days. To reduce oxygen exposure, the flask was sealed with a silicone plug and securely wrapped in plastic film. Bacterial cells were removed by centrifugation (4,383 × G, 4 °C, 10 min), performed twice. Ammonium sulfate was added to the supernatant to achieve 80 % saturation. The resulting precipitate was dissolved in 750 mL of 20 mM Tris-HCl (pH 7.5) and dialyzed against the same buffer. Solubilized proteins were fractionated using TOYOPEARL DEAE-650M (Tosoh Corp., Tokyo, Japan), TOYOPEARL Butyl-650M (Tosoh Corp.), Q-Sepharose HP (Danaher Corp., Washington, DC, USA), Superdex 200 pg (Danaher Corp.), and Mono Q 10/100 GL (Danaher Corp.). TOYOPEARL DEAE-650M was equilibrated with 20 mM Tris-HCl (pH 7.5) containing 1 mM dithiothreitol (DTT), followed by application of the sample, washing with the same buffer, and elution of proteins using a linear gradient of 0-1 M NaCl. Ammonium sulfate was then added to the active fractions to 30 % saturation before applying them to TOYOPEARL Butyl-650M, which had been equilibrated with 20 mM Tris-HCl (pH 7.5) and 30 % saturated ammonium sulfate containing 1 mM DTT. This was followed by washing with the buffer and protein elution using a linear gradient of 30-0 % saturated ammonium sulfate. After dialysis against 20 mM Tris-HCl (pH 7.5) containing 1 mM DTT, the active fractions were applied to Q-Sepharose HP equilibrated with the same buffer. The column was washed, and proteins were eluted using a 0-1 M NaCl linear gradient. Active fractions were then dialyzed against 20 mM Tris-HCl (pH 7.5) containing 1 mM DTT and 0.15 M NaCl. After concentration using Centrifugal filters-10K Regenerated Cellulose 10000NMWL (Merck KGaA, Darmstadt, Germany), the active fractions were applied to Superdex 200 pg, pre-equilibrated with 20 mM Tris-HCl (pH 7.5), 1 mM DTT and 0.15 M NaCl. Active fractions were further dialyzed against 20 mM Tris-HCl (pH 7.5) containing 1 mM DTT and subsequently applied to Mono Q 10/100 GL, also equilibrated with the same buffer. After washing, proteins were eluted using a 0-1 M NaCl linear gradient. Protein concentration was estimated by absorbance at 280 nm (Abs280) or determined using Bradford Reagent (Merck KGaA). An Abs280 of 1.0 was assumed to correspond to a protein concentration of approximately 1 mg/mL.
Enzymatic activity assay of fractions. For the fecal-derived fractions, alginate lyase activity was measured as an increase in absorbance at 235 nm (ΔAbs235) in a mixture of each fraction and sodium alginate as follows. Twenty microliter of each fraction was mixed with 980 µL of 50 mM Tris-HCl buffer (pH 7.5) containing 0.8 % sodium alginate and incubated at 37 °C. The reaction was terminated by adding 350 µL of 0.2 M sodium hydroxide at two time points: 30 s and 30 min and 30 s after initiation. The increase in absorbance over the 30-min interval (ΔAbs235) was used as a measure of alginate lyase activity in each fraction.
Isolation of B. xylanisolvens from human feces. Feces obtained from a healthy volunteer were suspended in 0.15 M NaCl. A 10 µL aliquot of the suspension was inoculated into a culture medium containing 0.2 % sodium alginate, 0.1 % ammonium sulfate, 0.22 % dipotassium hydrogen phosphate, 0.09 % potassium dihydrogen phosphate, 0.09 % sodium chloride, 0.003 % calcium chloride dihydrate, 0.002 % magnesium chloride hexahydrate, 0.001 % manganese chloride tetrahydrate, 0.001 % cobalt chloride hexahydrate, 0.0004 % iron (II) sulfate heptahydrate, and 0.04 % L-cysteine (pH 7.0) (ABm medium), supplemented with 0.03 % yeast extract and 0.03 % tryptone (YT-ABm medium). After 48 h of anaerobic culture at 37 °C, the supernatants were removed by centrifugation, and the pellets were streaked onto YT-ABm plates containing 1.5 % agar. Following anaerobic incubation, single colonies were isolated. The colonies were inoculated into YT-ABm medium. After 51 h of anaerobic culture at 37 °C, the supernatants were spotted onto thin-layer chromatography (TLC) plate, followed by development in a solvent of 1-butanol:acetic acid:water = 3:2:2 ratio. The saccharides were detected by spraying 10 % sulfuric acid in ethanol and heating. The colonies which exhibited degradation of alginate were identified by 16S rRNA gene sequencing.
Bacterial strains. Bacteroides caccae JCM 9498, B. cellulosilyticus JCM 15632 (CRE21), B. clarus JCM 16067, B. dorei JCM 13471, B. eggerthii JCM 12986 (DSM20697), B. faecis JCM 16478, B. finegoldii JCM 13346, B. nordii JCM 19687, B. oleiciplenus JCM 16102, B. ovatus JCM 5824, B. stercoris JCM 5826, B. thetaiotaomicron JCM 5827, and B. uniformis JCM 5828 were provided by the Japan Collection of Microorganisms, RIKEN BRC, a participant in the National BioResource Project of the Ministry of Education, Culture, Sports, Science and Technology, Japan. Bacteroides vulgatus NBRC 14291 was obtained from the National Institute of Technology and Evaluation (Japan).
Biochemical test. The B. xylanisolvens isolate was subjected to biochemical testing using API 20A and Rapid ID32A (bioMérieux SA, Marcy-l'Étoile, France) by TechnoSuruga Laboratory Co., Ltd. (Shizuoka, Japan), both designed for the identification of anaerobic bacteria. The biochemical profile was compared with that of B. xylanisolvens strain XB1A, as described previously [18] (Table 1 and Table S1; see J. Appl. Glycosci. Web site).
Table 1. Purification of alginate lyase from an alginate enrichment culture derived from fecal samples.
| Step | Total protein (mg) | Total activity (U) | Specific activity (U/mg) | Yield (%) | Purification (fold) |
| Ammonium precipitation | 1,160 | 411,000 | 353 | 100 | 1 |
| TOYOPEARL DEAE-650M | 108 | 57,000 | 528 | 13.9 | 1.49 |
| TOYOPEARL Butyl-650M | 8.67 | 50,100 | 5,780 | 12.2 | 16.4 |
| Q-Sepharose HP | 0.390 | 19,300 | 48,900 | 4.69 | 139 |
| Superdex 200 pg | 0.0310 | 8,200 | 263,000 | 2.00 | 745 |
| Mono Q 10/100 GL | 0.0160 | 3,870 | 233,000 | 0.940 | 660 |
Growth curves. The strains were precultured in GAM medium (AccuDia GAM Broth, Shimadzu Diagnostics Corp., Tokyo, Japan) for 48 h at 37 °C. Following centrifugation, bacterial cells were washed with 0.15 M NaCl and suspended in ABm medium or a glucose-containing medium in which alginate was replaced with glucose (GBm), with an initial cell density of 2.0 × 106 cells/mL. Bacterial growth was monitored by measuring the increase in optical density at 600 nm (OD600).
Genome sequencing. Genomic DNA from the isolate was extracted using the DNeasy Blood & Tissue Kit (QIAGEN N.V., Venlo, The Netherlands). A library was prepared with the TruSeq DNA PCR-free Kit (Illumina Inc., San Diego, CA, USA), followed by sequencing on a NovaSeq 6000 system (Illumina Inc.) in Macrogen Japan Corp. (Tokyo, Japan). Adaptor sequences and low-quality reads were removed using Trimmomatic [19], and the cleaned reads were mapped to a reference genome (RefSeq ID: GCF_000210075.1) using BWA [20]. Unmapped reads (24.2 % of total) were de novo assembled with SPAdes [21]. Raw read data of B. xylanisolvens isolate (strain MK6803) were deposited to DNA Data Bank Japan (BioProject ID, PRJDB11427; BioSample ID, SAMD00290071).
Recombinant BxPL38 protein. A gene encoding a putative PL38 alginate lyase from the B. xylanisolvens isolate (BxPL38 protein), whose signal peptide as predicted by SignalP-5.0 [22] was trimmed and cloned into pET-21b(+). Escherichia coli BL21(DE3) harboring the plasmid was cultured in Luria-Bertani medium containing 100 µg/mL sodium ampicillin at 37 °C and 120 strokes per min, followed by induction with 0.4 mM isopropyl-β-D(-)-thiogalactopyranoside and incubation overnight at 16 °C under the same shaking conditions. Cells were lysed by ultrasonication, and supernatants were collected by centrifugation. The recombinant BxPL38 protein was purified using TALON Metal Affinity Resin (Takara Bio Inc., Shiga, Japan). The protein concentration of the purified BxPL38 protein was measured using Bradford Reagent (Merck KGaA).
Enzymatic activity of BxPL38 protein. The purified BxPL38 protein (0.2 µM) was incubated with 0.3 % sodium alginate, G-abundant block, or M-abundant block in 50 mM Tris-HCl buffer (pH 7.5). Enzymatic activity was measured by monitoring the increase in Abs235 using UV-2600 spectrophotometer (Shimadzu Corporation, Kyoto, Japan). One unit (1 U) of activity was defined as a 0.1 increase in absorbance at 235 nm per min.
Phylogenetic analysis. The complete genomes of 18 B. ovatus and 11 B. xylanisolvens strains were retrieved from the NCBI database, as listed in Table S2 (see J. Appl. Glycosci. Web site). Phylogenetic trees were constructed using multilocus sequence analysis with UBCG2 [23].
Purification of alginate lyase from fecal samples. Alginate metabolism in the gut microbiome remains largely unclear. We first investigated whether alginate lyase could be extracted from fecal samples. Feces from a healthy volunteer were incubated in alginate-containing media, followed by collection of the supernatant to isolate alginate lyases secreted by alginate-metabolizing gut bacteria. After protein precipitation using 80 %-saturated ammonium sulfate, the proteins were solubilized and applied to five chromatography columns in individual consecutive purification steps. Following a linear NaCl gradient elution on TOYOPEARL DEAE-650M, an anion exchange resin, alginate lyase activity was detected in fractions #17-20 and #23-31 (Fig. 1A), suggesting that presence of more than two lyases in the applied solution. Based on specific activity (ΔAbs235/Abs280), fractions #17-20 were selected for further analysis. Subsequently, alginate lyase activity was observed in the flow-through and wash fractions during hydrophobic interaction chromatography using TOYOPEARL Butyl-650M resin (Fig. 1B). These fractions were then dialyzed and subjected to anion exchange chromatography using Q-Sepharose HP and size-exclusion chromatography with Superdex 200 pg (Figs. 1C and D). A final anion exchange step using Mono Q 10/100 GL revealed alginate lyase activity as a single peak (Fig. 1E). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by silver staining indicated the purified fraction contained a ~65 kDa protein (Fig. 1F). Overall, we achieved a 660-fold purification of alginate lyase with a 0.94 % yield (Table 1). The purified enzyme acted on alginate and M-abundant block but not on G-abundant block or the following glycosaminoglycans: hyaluronan, chondroitin sulfate C, and heparin, indicating poly-M-specific lyase activity (Figs. 1G and H). Although a large portion of the substrate remained at the point of origin, the products were exclusively disaccharides, indicating that the purified enzyme functions in an exo-acting manner, selectively releasing disaccharide units. To identify the lyase, the Coomassie brilliant blue (CBB)-stained band was excised and analyzed via liquid chromatography-tandem mass spectrometry (LC-MS/MS). While some detected peptides showed low identity to PL6, PL7, and PL17 alginate lyases, most matched unrelated proteins, such as DNA-directed RNA polymerase subunits from Dichelobacter nodosus and Halorhodospira halophila. Nonetheless, these experiments demonstrated the presence of alginate lyase(s) derived from gut microbes.
The samples were applied to (A) TOYOPEARL DEAE-650M, (B) TOYOPEARL Butyl-650M, (C) Q-Sepharose high performance (HP), (D) Superdex 200 pg, and (E) Mono Q 10/100 GL. Protein concentration (Abs280) and enzymatic activity (ΔAbs235) are shown as closed circles and open rhombuses, respectively. Fractions indicated by arrows were collected and used in subsequent experiments and/or analyses. (F) Silver-stained SDS-PAGE image of collected fractions. (G) Relative activity of the purified enzyme for alginate, M-abundant (M) block, G-abundant (G) block, hyaluronate, chondroitin C, and heparin. Activity on sodium alginate was set at 100 %. (H) TLC analysis of supernatants. Digestion products were visualized using 10 % sulfuric acid in ethanol. Arrows indicate oligosaccharide produced by the purified alginate lyase.
Isolation of B. xylanisolvens strain MK6803. Using a fecal sample from a healthy volunteer, we performed enrichment culture in alginate-containing media. Eleven colonies were isolated and assessed for their ability to digest alginate. Among them, four isolates showed significant alginate degradation after incubation (Fig. 2). These four isolates, but not the remaining seven, were able to grow on alginate after 24-48 h of incubation. All four were identified as B. xylanisolvens based on their 16S rRNA gene sequences. Isolate #1, designated MK6803, was selected for further analysis.
Spots near the baseline indicate residual alginate in culture supernatants.
Biochemical characteristics of strain MK6803. We compared the biochemical properties of strain MK6803 with those of the B. xylanisolvens type strain XB1A. While MK6803 lacked β-glycosidase and α-arabinosidase activity, as well as certain fermentation abilities, it produced indole-a feature not observed in XB1A (Table 2 and Table S1; see J. Appl. Glycosci. Web site). In total, eight biochemical characteristics differed between the two strains.
Table 2. Biochemical characteristics of the Bacteroides xylanisolvens isolate (strain MK6803).
| MK6803 | XB1A＊ | |
| β-Glycosidase | − | + |
| α-Arabinosidase | − | + |
| Indole production | + | − |
| D-Mannitol fermentation | − | + |
| Salicin fermentation | − | + |
| Glycerol fermentation | − | + |
| L-Rhamnose fermentation | − | + |
| D-Trehalose fermentation | − | + |
＊The characteristics of B. xylanisolvens strain XB1A were described in a previous report [18]. Additional characteristics are provided in Table S1 (see J. Appl. Glycosci. Web site).
Alginate-metabolizing gene cluster. Draft genome sequencing of B. xylanisolvens strain MK6803 revealed the presence of a gene cluster conserved in B. xylanisolvens strain CL11T00C41 but absent in the type strain XB1A (Fig. 3). This gene cluster was also found in B. ovatus strain CP926 but not in the type strain NCTC 11153, indicating that only a subset of B. xylanisolvens and B. ovatus strains harbor this cluster. The cluster encodes a putative PL6_1 guluronate-specific lyase, PL17_2 alginate lyase, a major facilitator superfamily (MFS) transporter, glucose 1-dehydrogenase (also annotated as an oxidoreductase), and PL38 alginate lyase. A nucleotide BLAST search indicated that, aside from B. xylanisolvens and B. ovatus, no other species possess this cluster.
The figure was generated using Easyfig [38]. Alignment was performed using the nucleotide Basic Local Alignment Search Tool (BLAST).
Growth of Bacteroides strains in alginate culture. To examine the ability to grow using alginate as the sole carbon source, we cultured strain MK6803 and various Bacteroides strains in ABm medium (Fig. 4). Bacteroides xylanisolvens strain MK6803 exhibited growth in ABm medium, although its growth rate was slower compared to that in GBm medium (Fig. 4A). In contrast, the type strains of 15 Bacteroides species grew well in GBm medium but failed to grow in ABm medium, indicating an inability to assimilate alginate (Fig. 4). The B. xylanisolvens type strain XB1A, which lacks alginate lyase genes, showed no ability to utilize alginate. Although B. eggerthii strain 10128 harbors a gene cluster encoding putative PL6_1 and PL17_2 alginate lyases, as well as SusC/D and MFS transporters, this strain also failed to grow in the presence of alginate. Similarly, B. ovatus strain NCTC 11153 did not grow despite possessing a gene cluster encoding putative PL38 alginate lyases. These findings suggest that alginate assimilation in Bacteroides species requires the presence of a gene cluster encoding PL6_1, PL17_2, PL38, and a putative oxidoreductase, as found in B. ovatus strain CP926, B. xylanisolvens strain CL11T00C41, and B. xylanisolvens strain MK6803 (Fig. 4A).
Strains were cultured anaerobically in alginate-based medium (ABm; open squares with solid lines) or glucose-based medium (GBm; closed circles with dotted lines). Arrows indicate the presence of the gene clusters possessed.
Alginate lyase activity of BxPL38 protein. Alginate lyases of the PL6 and PL17 families have been well characterized, whereas the PL38 family remains less understood. To investigate whether the BxPL38 protein possesses alginate lyase activity, we purified the recombinant protein expressed in E. coli (Fig. 5A). After 24 h of incubation, slight degradation of alginate was observed in the BxPL38/alginate mixture on the TLC plate; however, M-abundant block was completely degraded into low-molecular-weight products by BxPL38 (Fig. 5B). The specific activity toward M-abundant block (7.98 ± 0.25 U/nmol) was higher than those toward alginate (1.22 ± 0.09 U/nmol) and G-abundant block (0.87 ± 0.03 U/nmol) (Fig. 5C).
(A) Purified recombinant BxPL38 protein. SDS-PAGE followed by staining with Coomassie brilliant blue. (B) TLC analysis of reaction mixtures containing alginate, G-abundant (G) block, or M-abundant (M) block in the presence of BxPL38 protein. (C) Enzymatic activity on alginate, G-abundant (G) block, or M-abundant (M) block.
Horizontal gene transfer. To investigate the horizontal gene transfer of the gene cluster, we constructed a multilocus sequence analysis (MLSA)-based phylogenetic tree (Fig. 6). The phylogenetic trees revealed that B. ovatus (Fig. 6A) and B. xylanisolvens (Fig. 6B) strains harboring the gene cluster were positioned separately, suggesting that the gene cluster was horizontally transferred between B. ovatus and B. xylanisolvens.
(A) B. ovatus. (B) B. xylanisolvens. Asterisks indicate strains harboring the gene cluster encoding PL6_1, PL17_2, the putative oxidoreductase, and PL38.
Bacteroides xylanisolvens was first isolated from human feces and described as a novel Bacteroides species capable of degrading xylan, a type of hemicellulose [18]. Pudlo et al. later reported that some B. xylanisolvens isolates could assimilate mucin O-glycan [24]. In this study, we isolated B. xylanisolvens strain MK6803 through alginate enrichment culture from a fecal sample. This strain lacked several fermentation capabilities but produced indole (Table 2). Indole production by B. ovatus has been shown to exhibit anti-inflammatory effects via activation of the aryl hydrocarbon receptor [25]. Notably, the amount of indole produced varies considerably among Bacteroides species [26]. Since we did not quantify indole production in strain MK6803, further studies are needed to assess the potential health benefits of this strain in the human gut.
One of the key characteristics of strain MK6803 is its ability to assimilate alginate (Figs. 2 and 4). With respect to alginate degradation and fermentation, Fu et al. reported that B. xylanisolvens AY11-1 produced oligosaccharides and short-chain fatty acids from alginate [27]. Harada et al. isolated a B. xylanisolvens strain that assimilated low-molecular weight-alginate and enhanced NO secretion from RAW264.7 cells when incubated with alginate [28]. Li et al. demonstrated that the predominance of B. xylanisolvens led to acetate accumulation in human fecal cultures supplemented with alginate [17]. Additionally, Rønne et al. recently reported that metabolites produced by alginate-assimilating B. xylanisolvens supported the growth of other Bacteroides species that are unable to utilize alginate [14]. These findings suggest that colonization by alginate-assimilating B. xylanisolvens in the gut may benefit host health and promote the establishment of other beneficial gut bacteria.
In this study, we focused on a gene cluster encoding PL6_1, PL17_2, and PL38 alginate lyases, along with a putative oxidoreductase. The PL6_1 family in the CAZy database includes alginate lyases with varied substrate specificities [10, 29, 30]. Rønne et al. reported that the PL6_1 and PL17_2 alginate lyases encoded in the cluster exhibited poly-G and poly-M specificity, respectively [13]. The same group also demonstrated that PL38 alginate lyases preferred partially hydrolyzed low-molecular-weight alginate over high-molecular-weight alginate and showed the highest activity in the order of poly-MG, poly-G, and poly-M [13]. The PL38 alginate lyase encoded by B. xylanisolvens strain MK6803 seems to exhibit M-abundant block-preferred activity (Fig. 5). The amino acid sequences of the PL38 lyases were conserved between B. ovatus and B. xylanisolvens strains with 100 % identity, suggesting that the two enzymes are unlikely to differ in substrate specificity. The PL38 lyases showed 30 % (query coverage 72 %), 34 % (86 %), 37 % (83 %) identity with the three PL38 lyase from B. ovatus NCTC 11153 (GenBank IDs, ALJ44837.1, ALJ44842.1, and ALJ44843.1). As mentioned in Materials and Methods, the M-abundant block utilized in this study contained 30 % G residues [16]. On the other hand, Rønne et al. [13] utilized G residue-absent poly-M block obtained from an epimerase AlgG-negative mutant of Pseudomonas fluorescens [31]. This research group thoroughly analyzed characteristics of BoPL38 protein and reported its structure and mechanism [32]. The observed differences in substrate specificity between the two enzymes may be influenced by the M/G composition and the block length of the substrates used. Another possibility is that the observed specific activity of the BxPL38 protein is influenced by the presence of G included in the M-abundant block. Thus, similar to B. ovatus PL38 lyase, BxPL38 might exhibit specificity toward multiple bond types. In addition, because we and Rønne et al. calculated enzymatic activities in different methods, comparison of the activity was incapable. Further analysis, such as calculation of kcat/Km and enzymatic activity assay using alginate blocks derived from other source, is required for biological function of the PL38 lyase in Bacteroides species.
As shown in Fig. 4, B. eggerthii strain 10128 and B. ovatus strain NCTC 11153 exhibited no growth when alginate was used as the sole carbon source, despite possessing alginate lyase genes. This inability to assimilate alginate suggests that the putative oxidoreductase is essential for alginate utilization. BLAST search indicated that, in the genomes of B. ovatus and B. xylanisolvens, the putative oxidoreductase genes appeared exclusively within the alginate-utilization gene clusters. We previously reported that Sphingomonas sp. A1 produces an NADPH-dependent reductase (A1-R), a member of the short-chain dehydrogenase/reductase superfamily, which catalyzes the conversion of 4-deoxy-L-erythro-5-hexoseulose uronic acid to 2-keto-3-deoxy-D-gluconic acid [33]. The putative oxidoreductase identified in this study shares 35 % sequence identity with A1-R and exhibits 98 % coverage across the full length of A1-R. AlphaFold3 model of the putative oxidoreductase was similar to the overall structure of A1-R (PDB ID, 3AFM) (Fig. 7A) [34]. Structural alignment indicated that the three catalytic amino acid residues Ser, Tyr, and Lys are topologically conserved in the putative oxidoreductase (Fig. 7B). The structural prediction supported our hypothesis that the putative oxidoreductase is involved in alginate assimilation through conversion of 4-deoxy-L-erythro-5-hexoseulose uronic acid to 2-keto-3-deoxy-D-gluconic acid.
(A) Crystal structure of A1-R (PDB ID, 3AFM) (left) and AlphaFold3 model of the putative oxidoreductase (right). (B) Structural alignment of the two protein. The three catalytic amino acid residues (Ser, Tyr, Lys) are shown in sticks. Alignment was conducted in PyMOL software (Schrödinger Inc., New York, NY, USA).
As shown in Fig. 6, gene cluster-positive B. ovatus and B. xylanisolvens strains are located in distinct clades in the MLSA-based phylogenetic tree, supporting the hypothesis of horizontal gene transfer among Bacteroides species. In relation to gene transfer events in Bacteroides, the tetracycline resistance gene tetQ has previously been reported to transfer between Bacteroides and Prevotella species, which typically colonize different hosts [35, 36]. Coyne et al. proposed that extensive horizontal gene transfer occurs among Bacteroidales species coresiding in the human gut [37]. Although acquisition of the alginate-utilization gene cluster may confer a colonization advantage in the gut where alginate is frequently present due to the consumption of brown algae-containing foods, the cluster has so far been detected only in Bacteroides species. Whether this gene cluster can be transferred to other gut-commensal bacteria remains unknown.
In conclusion, we isolated an indole-producing and alginate-assimilating B. xylanisolvens strain and demonstrated the involvement of a putative oxidoreductase as a key factor for growth on alginate as the sole carbon source. Given that Bacteroides species are known to produce various beneficial compounds for the host, the horizontal transfer of this gene cluster among Bacteroides species may contribute to enhanced gut colonization and potentially improve human health.
The authors no conflict of interests.
We would like to thank Enago (www.enago.jp) for the English language review. This work was partially supported by JSPS KAKENHI Grant Numbers 25K01979, 21H02156, and 18H02166.
