The isolation and identification of strain CPCC 101271 T , which represents a novel species of the genus Stenotrophomonas
Strain CPCC 101271T was recovered from the intestinal lumen of C. elegans using LB agar plates (see Materials and Methods). The nearly full-length sequence of the 16S rRNA gene (1543 bp) of strain CPCC 101271T was obtained and submitted to GenBank under accession number MT126327. A BLAST search of GenBank showed that the 16S rRNA gene of strain CPCC 101271T exhibited 98.1%-99.7% similarity to the 16S rRNA genes of members of the genus Stenotrophomonas, and < 98.0% similarities to those of other bacterial species in the family Lysobacteraceae. In a phylogenetic tree based on the 16S rRNA gene sequences of all members of the family Lysobacteraceae, strain CPCC 101271T formed a sublineage with S. rhizophila JCM 13333T and S. bentonitica DSM 103927T within the genus Stenotrophomonas (Fig. 1). Therefore, it is reasonable to designate strain CPCC 101271T as a member of the genus Stenotrophomonas.
Genome sequencing of strain CPCC 101271T yielded a draft genome of 4,402,751 bp, assembled from 126 qualified contigs, with 100-fold coverage and an N50 length of 738,821 bp. Genes putatively encoding glucosylglycerol-phosphate synthase (G9274_RS14805, B861_RS0201980, E5352_RS0097) and alpha-trehalose-phosphate synthase (BN96_RS08035, BIZ42_RS05395, C0R07_RS03270, GDJ08_RS08470), which might endow strain CPCC 101271T the ability to maintain homeostasis of the host, were identified in its genome. The ANI values between the draft genomes of strain CPCC 101271T and its closest phylogenetic neighbors S. rhizophila JCM 13333T and S. bentonitica DSM 103927T were 84.7 % and 85.0 %, respectively. These values were both far lower than 95%, which was proposed as the cutoff value for defining different species (Kim et al., 2014). Accordingly, strain CPCC 101271T represents a species genetically different from other validly described species of the genus Stenotrophomonas. This conclusion was supported by chemotaxonomic data (described below) and the phenotypic characteristics given in the species description and in Supplementary Table S1.
Description of Stenotrophomonas nematodicola sp. nov.
Stenotrophomonas nematodicola (ne.ma.to.di'co.la. N.L. pl. n. Nematoda, a taxonomic group of animals; L. suff. -cola, inhabitant, dweller; N.L. fem. n. nematodicola, an inhabitant of nematodes).
Cells are Gram-reaction-negative, facultatively aerobic and motile coccoid rods, 1.0-1.2 µm in width and 1.9–2.3 µm in length. Colonies on LB agar medium are smooth, pale yellow in color, and circular and entire, with a diameter of 1.0-1.1 mm after 48 h of incubation. Growth occurs at 10–37°C (optimum 32°C) and at pH 6.0–8.0 (optimum pH 7.0) with 0–5 % (w/v) NaCl (optimum 0–1 %). Catalase- and oxidase reactions are positive. Positive for hydrolysis of gelatin and nitrate reduction reaction, while negative for hydrolysis of starch and urea, peptonization of milk, and production of H2S and indol. Positive for acid phosphatase, alkaline phosphatase, α-chymotrypsin, esterase (C4), esterase lipase (C8), cystine arylamidase, β-glucosidase, leucine arylamidase and valine arylamidase in an API ZYM strip. Acetic acid, acetoacetic acid, bromo-succinic acid, citric acid, dextrin, D-fructose-6-PO4, D-lactic acid methyl ester, D-maltose, D-mannose, glycyl-L-proline, L-alanine, L-glutamic acid, L-histidine, methyl pyruvate, L-lactic acid, L-malic acid, N-acetyl-D-galactosamine, N-acetyl-D-glucosamine, pectin, propionic acid, Tween 40, α-D-Glucose and α-Keto-glutaric acid can be utilized as the sole carbon source, and amygdalin, arbutin, D-fructose, D-glucose, D-maltose, esculin ferric citrate, N-acetylglucosamine, potassium 2-ketogluconate and potassium 5-ketogluconate can be assimilized and produce acid. Resistant to ampicillin (10 µg), cefaclor (30 µg), chloramphenicol (30 µg), clindamycin (2 µg), erythromycin (15 µg), gentamycin (10 µg), kanamycin (30 µg), netilmicin (30 µg), novobiocin (5 µg), penicillin (10 IU), vancomycin (30 µg), tetracycline (30 µg), tobramycin (10 µg) and treptomycin (10 µg), while sensitive to polymyxinB (300 IU) and rifampin (5 µg). Diphosphatidylglycerol (DPG), phosphatidylglycerol (PG), phosphatidylethanolamine (PE) and an unidentified phospholipid (UPL) were detected in a polar lipids extract (Figure S1). The respiratory quinone is Q-8. The major fatty acids are iso-C15:0 (38.2%) and antesio-C15:0 (16.6%), with moderate amounts of iso-C11:0 (8.7%) and C16:0 (5.1%) and small amounts (< 5%) of cyclo-C17:0, C13:02-OH, C12:03-OH, iso-C17:0, iso-C14:0, C14:0, iso-C11:0 3-OH, iso-C16:0, iso-C13:03-OH, C16:1 ω7c/C16:1 ω6c, and iso -C17:1 ω9c/C16:0 10-methyl.
The type strain CPCC 101271T (= W5) was isolated from a surface-sterilized C. elegans worm cultured in a lab in Nanyang, a city in middle of China. The DDBJ/EMBL/GenBank accession numbers of the 16S rRNA gene sequence and draft genome sequence of strain CPCC 101271T are MT126327 and WIAY00000000, respectively. The genome of the type strain is characterized by a size of 4.4 Mbp and a G + C content of 67.3 mol%.
A filled circle indicates that the node was also recovered in trees generated with the maximum-likelihood method and maximum-parsimony method. Bootstrap values are shown as the percentage of 1,000 replicates; only percentages above 50% are shown.
Escherichia coli JCM 1649T (GenBank accession no. X80725) was used as an outgroup (not shown). Bar, 0.005 substitutions per nucleotide position.
Increase in Caenorhabditis elegans survival rate and lifespan by feeding on CPCC 101271T
We compared the longevity of worms fed either on CPCC 101271T, JCM 13333T or OP50. The results showed that worms fed on JCM 13333T or OP50 had almost the similar lifespans. However, worms fed on CPCC 101271T lived approximately 40% longer than worms fed on E. coli or JCM 13333T, indicating that the nematodes fed on CPCC 101271T lived longer than those fed on E. coli OP50 or S. rhizophila JCM 13333T and had greatly increased survival rates (Fig. 2)
Strain CPCC 101271T confers the host with resistance to B. nematocida colonization
To investigate whether strain CPCC 101271T has colonization-resistance activity against B. nematocida B16, we first performed an in vitro bacteriostatic activity test. The results showed that strain CPCC 101271T could not inhibit B. nematocida B16 but could be inhibited by B16 (Fig. 3). A transparent inhibition zone with clear edges formed around the paper containing B16 after 48 h of incubation on an LB agar plate spread with strain CPCC 101271T. The clear zone, which was about 1.6 cm in diameter, was slightly smaller than the zones surrounding the positive drug controls polymyxin B (300 IU) and rifampin (5 µg). And no clear inhibition zone formed around the papers containing E. coli or LB medium (Fig. 3).
GFP-expressing strain B16g was used to confirm the specificity of B16 colonization activity assays. The results of colonization-resistance activities indicated that strain CPCC 101271T could also inhibit the colonization of B16 in the nematode intestine. During the first 24 h of infection with B. nematocida B16, almost no nematodes pre-fed with CPCC 101271T were scored as being in the “full” colonization category (see Materials and Methods). By contrast, almost 20% of animals directly fed with B. nematocida B16 were scored as “full”. In addition, after infection for 48 h, 50% of the animals fed only B. nematocida B16 were scored as “full”. However, only 10% of worms pre-fed with CPCC 101271T were scored as “full” at the same time point. After 72 h, B. nematocida B16 showed notably strong colonization ability, with 90% of the worms not pre-fed with CPCC 101271T scored as “full”. In contrast, only 10% of the animals pre-fed with CPCC 101271T were scored as “full”. Moreover, compared with worms fed only B16, the percentage of worms fed with both B16 and CPCC 101271T that had undetectable B16 colonization was much higher at 48 h (10% vs. 80%) and at 72 h (0% vs. 70%). The worms pre-fed with E. coli showed little difference compared with those in the no pre-feeding group. The ability of B16 to colonize the nematodes pre-fed with JCM 13333T was stronger than its ability to colonize those pre-fed with CPCC 101271T, but a little weaker than its ability to colonize the negative controls pre-fed with E. coli OP50. Differences between the abilities of B16 to colonize the nematodes pre-fed with CPCC 101271T and pre-fed with OP50 were notable when we compared the changes in the severity of colonization at 72-h (Fig. 4, chi-squared test, P < 0.0001). For example, only 10% of worms that were pre-fed with CPCC 101271T could be categorized as having ‘full’ colonization. However, 90% of worms that were pre-fed with OP50 were categorized in the ‘full’ colonization category. The results indicated that colonization of B. nematocida B16 was markedly attenuated in C. elegans pre-fed with CPCC 101271T.
The differences in the mortalities of the B16-infected nematodes in the different treatment groups indicated that pre-feeding with CPCC 101271T reduced the mortality caused by infection with the pathogenic bacteria B16 (Fig. 5). The mortalities of the nematodes pre-fed with OP50 and then infected by B16 and the nematodes directly infected by B16 (without pre-feeding with any other bacteria) were 85 and 90% within 60 h, respectively. By contrast, for worms pre-fed with CPCC 101271T then infected with B16, the mortality dropped to 40%. The natural mortality rate of the negative control nematodes (no pre-feeding or B16 infection) was only 18%. At other time points, the mortalities of nematodes pre-fed with CPCC 101271T were significantly lower than those of nematodes pre-fed with E. coli or directly infected with B16.
The variation in Caenorhabditis elegans microbiota community structure during competition between CPCC 101271T and B16
In a previous study, we collected free-living terrestrial C. elegans from soil and rotten fruits, and analyzed the variation in intestinal flora following B. nematocida B16 infection by performing macrogenomic analysis. We found significant differences in the diversity and distribution of microbiota between the control worms and those infected with B16 for 24 h. The diversity of the intestinal microbiome decreased after B16 infection (Niu et al., 2016). Here we aimed to investigate the variation in the intestinal bacterial community structure of C. elegans during competition between CPCC 101271T and B16.
A total of 332314, 280966, 705161, 227126 and 597664 sequences comprising 44, 26, 49, 19 and 30 operational taxonomic units (OTUs) were obtained from the five groups CW00h, CW04h, CW08h, CW12h and CW16h, respectively. At the genus level, these OTUs represented 26, 14, 23, 10 and 14 genera, respectively. It was obvious that the bacteria diversity was greatly decreased during competition between CPCC 101271T and B16 (Fig. 6, Fig. 7). At the first stage of infection (CW00h) in nematodes pre-fed with strain CPCC 101271T, the microbiota community structure predominantly consisted of the genera Bacillus, Acetobacter, Lactobacillus, Phytobacter, Stenotrophomonas, Pichia and Sphingomonas. At the second stage (CW04h), 4 h after the worms were infected by B16, dysbiosis occurred, and in the course of re-construction of the microbiota community, the bacteria diversity was drastically reduced. Besides Lactobacillus spp., Acetobacter spp. and Pichia spp., which remained the major groups, the abundance of Bacillus spp. increased slightly and the abundance of Stenotrophomonas spp. increased. At the third stage (CW08h), which we termed “the breaking period”, a large number of CPCC 101271T and B16 bacteria were co-existing and competing; the diversity of the intestinal flora had partially recovered, but the abundance of Bacillus spp. had greatly decreased. At the fourth stage (CW12h), the abundance of Bacillus spp. was even lower and the abundance of Stenotrophomonas spp. was higher. The newly reconstructed intestinal flora was disrupted again, and the species composition was the most similar to that observed at the second stage (Fig. 7). By the fifth stage (CW16h), B16 overwhelmed CPCC 101271T and only Lactobacillus spp. and Pichia spp., together with Bacillus spp., remained the major microbiota.
In a summary, over the course of B16 infection, the abundance of CPCC 101271T and its relatives (Stenotrophomonas spp.) kept on increasing until 8 h after B16 infection and then decreased sharply. The change in abundance of Stenotrophomonas spp. was accompanied by similar changes in the abundance of alpha-trehalose-phosphate synthase-encoding genes calculated from the metagenome data, except at the final stage (Figure S2). By contrast, the abundance of Bacillus spp. first decreased, then increased rapidly, which was similar to the changes in the abundance of trehalose-6-phosphate hydrolase-encoding genes (Figure S3). The abundances of alpha-trehalose-phosphate synthase-encoding genes (present in the genome of B16) and trehalose-6-phosphate hydrolase-encoding genes (present in the genome of CPCC 101271T) together with Stenotrophomonas spp. and Bacillus spp. reached the highest level at 8 h (CW08h) after B16 infection. The results suggested that strain CPCC 101271T possibly participated directly or induced some other bacteria in the community to participate in resistance to B16 colonization. However, by 16 h after infection, B16 dominated the microbiota community, and the growth of CPCC 101271T was completely suppressed.