Introduction

The genus Hymenobacter was proposed as a novel genus by Hirsch et al. [9]. The genus Hymenobacter belongs to the family Cytophagaceae, the order Sphingbacteriales, and the phylum Bacteroidetes and is well known for having higher DNA G+C contents (55–70 mol%) compared with other genera of the family. Hymenobacter species are pink to red pigmented, rod-shaped, non-motile, and Gram-negative bacteria that contain phosphatidylethanolamine (PE) as the main phospholipid, menaquinone 7 as the major quinone, and iso-C15:0, anteiso-C15:0, and summed feature 3 (16:1 ω7c/16:1 ω6c) as the major fatty acids.

At the time of writing, the genus was comprised of 28 recognized species, which were discovered in various environments. Overall, many species of the genus have been isolated from soil samples. These include the following: Hymenobacter arizonensis, isolated from the southwestern arid lands soil [22]; H. daecheongensis, from stream sediment [37]; H. deserti and H. xinjiangensis, from the desert soil of Xiinjian, China [39, 40]; H. ginsengisoli, from the soil of a ginseng field [10]; H. psychrophilus, from the soil of an industrial site [38]; and H. ruber and H. soli, from grass soil [12, 14].

In this study, strain DG7AT was isolated from a soil sample. Based on a polyphasic approach including phenotypic, phylogenetic, genomic, and chemotaxonomic characteristics, strain DG7AT was designated as a new species in the genus Hymenobacter.

Materials and Methods

Isolation of Bacterial Strain and Culture Conditions

Strain DG7AT was isolated from a soil sample (1.0 g) collected in Seoul (N 37° 34′ 30″ E 127° 00′ 30″), South Korea. The soil sample was suspended in 10.0 ml sterile water, after which the resulting supernatant was serially diluted. 100 µl of each dilution was spread on a plate of R2A agar (Difco, USA) which had been diluted 10 times, and was incubated at 25 °C. The purified colonies were tentatively identified by partial 16S rRNA gene sequences, and were preserved in a glycerol solution (25 %, w/v) at −70 °C. Strain DG7AT was deposited into the Korean Collection for Type Cultures (KCTC 32554T) and the Japan Collection of Microorganisms (JCM 30007T).

Phenotypic and Biochemical Characteristics

The Gram reaction was performed using the non-staining method [2]. Cell morphology was observed using a light microscope (Nikon, Japan) and a transmission electron microscope (LIBRA 120, Carl Zeiss, Germany). Oxidase activity was evaluated via the oxidation of 1 % (w/v) tetramethyl-p-phenylene diamine. Catalase activity was determined by measurement of bubble production after the application of a 3 % (v/v) H2O2 solution. Growth at various pHs (5–11 at pH 1 intervals) was assessed on R2A broth at 25 °C. To determine the NaCl tolerance, R2A agar containing different salt concentrations [0–10 % (w/v %), 1 % intervals] were used. Growth on different media was also assessed, sing tryptic soy agar (TSA; Difco), nutrient agar (NA; Difco), Luria–Bertani (LB; Difco) agar, and Ancylobacter–Spirosoma Medium (ASM; glucose 1 g, peptone 1 g, yeast extract 1 g, agar 15 g). All of the above growth tests were performed at 25 °C. Growth at different temperatures (4, 12, 17, 20, 25, 30, 37, and 42 °C) was assessed on R2A agar plates for 3–7 days. The API 20NE, API 50CH, and API ZYM microtest systems were employed according to the manufacturer’s instructions (bioMérieux), for studying carbon source utilization and the enzyme activities of the strains.

Pigments were extracted using 95 % ethanol, and the absorption spectrum between 200 and 800 nm was measured with a UV spectrophotometer (UV-2450, Shimazu). Flexirubin-type pigments were tested for based on a color shift after exposure to 0.1 N NaOH solution [7, 35].

Survival Assays to Ultraviolet

Cells were grown up to early stationary phase on R2A broth (Difco) and adjusted to ~108 CFU/m [11, 27]. Cells were then serially diluted with 0.85 % NaCl and spotted on R2A agar. After taking off the lids, cells on the R2A agar plates were exposed to UV at room temperature using a UVC ultraviolet crosslinker (UVP, CX-2000, CA, USA) at 254 nm. The applied dose rate was 20 J/m2/s, and different doses of radiation were achieved by adjusting the total exposure time. After exposure, cells were incubated at 30 °C for 2 days prior to enumeration of the colonies.

16S rRNA Gene Sequencing and Phylogenetic Analysis

The 16S rRNA gene of strain DG7AT was amplified from chromosomal DNA using the universal bacterial primer set, 9F, and 1492R [36]. Sequence analysis was performed using the 27F, 785F, 805R, and 1492R universal primers from Genotech (Daejeon, South Korea). The full sequences of the 16S rRNA gene were compiled with SeqMan software (DNASTAR Inc.).

For phylogenetic analysis, the nearly full sequence of strain DG7AT obtained (1461 bp) was compared with other taxa using the EzTaxon-e service [15]. The 16S rRNA gene sequences of related taxa were obtained from GenBank and were edited with the BioEdit program [8]. Multiple alignments were performed with the CLUSTAL X program [32]. Pairwise distances for the neighbor-joining algorithm [24] were calculated according to Kimura two-parameter model [16], and the phylogenetic tree was constructed using the MEGA 5 Program [31]. Bootstrap analysis with 1,000 replicates was also conducted to obtain confidence levels for the branches [6]. The close-neighbor-interchange (CNI) on random trees method with a search factor of one and the number of initial trees (random addition) of ten was applied in maximum parsimony analysis, and maximum likelihood analysis was performed with the kimura 2-parameter model in the MEGA 5 Program.

Chemotaxonomic Analyses

To analyze the composition of polar lipids, total lipids were extracted following the procedures of Minnikin et al. [21]. The total lipids were separated using two-dimensional thin layer chromatography (TLC) with the first and second developing solvents, chloroform/methanol/water (65:25:4), and chloroform/methanol/acetic acid/water (80:12:15:4), respectively. After separation of the lipids, various lipids were tested for by spraying with the appropriate detection reagents [19]. The total lipids were verified by heating for 10 min at 150 °C after spraying with a molybdophosphoric acid solution (Sigma-Aldrich, St. Louis, MO, USA). In order to perform fatty acid methyl ester analysis, cells were allowed to grow on R2A agar for 3 days at 25 °C, and then two loops of well-grown cells were harvested. Fatty acids were purified by saponification, methylation, and extraction procedures, as described previously [25]. Fatty acid methyl esters (FAME) were prepared, separated, and identified with the Sherlock Microbial Identification System V6.01 (MIS, data base TSBA6, MIDI Inc., Newark, DE, USA).

Genomic Analysis

DNA–DNA hybridization was performed fluorometrically, according to the method reported by Ezaki et al. [5]. For the determination of the G+C mol content of strain DG7AT and relative taxa, genomic DNA was extracted using the standard method, which employs a CTAB/NaCl solution. The purified genomic DNA was enzymatically degraded into nucleosides by nuclease P1 and alkaline phosphatase, and the nucleosides were then analyzed using reverse-phase HPLC as previously described in the literature [20, 30].

Results and Discussion

Morphological and Phenotypic Characteristics

Cells of strain DG7AT were Gram-negative, non-motile, and rod-shaped without flagella (Supplementary Fig. S1), exhibiting a red-pink color when cultured on R2A agar at 25 °C. The red-pink pigment produced by strain DG7AT had the spectrum of a carotenoid pigment with absorption maxima at 264.5, 319.0, 368.0, and 480.0 (pentosyl-2′-hydroxyflexixanthin [17]) nm (Supplementary Fig. S2). This result differs from that of H. saemangeumensis GSR0100T (453.0, 483.0 (hexosyl-2′-hydroxyflexixanthin or methyl-2′-hydroxyflexixanthin [17]), and 508.0 nm) [13]. Alkalization with 0.1 volume of 0.1 M NaOH did not lead to any shift in peak positions, indicating that strain DG7AT did not have flexirubin pigment. Cells were able to grow at temperatures between 12 and 30 °C with an optimum of 25 °C but did not grow at 4, 37, and 42 °C. The optimum pH of strain DG7AT was found to be between 6.0 and 9.0, and weak growth was observed at pH 5 and pH 10–11. Strain DG7AT did not grow in the examined NaCl (w/v) range (0.5–10 %). Results regarding the physiological characteristics of strain DG7AT were summarized in the species description, and the differential characteristics with type strains of closely related species are shown in Table 1.

Table 1 Differential characteristics between strain DG7AT and closely related species
Full size table

Phylogenetic Analysis

16S rRNA gene (1461 bp) sequence of strain DG7AT was compared with those of closely related taxa. The phylogenetic tree constructed by the neighbor-joining method revealed that strain DG7AT belongs to the genus Hymenobacter (Fig. 1). Strain DG7AT had the highest sequence similarity with Hymenobacter soli PB17T (98.35 %) [14], H. glaciei VUG-A130T (96.78 %) [18], H. antarcticus VUG-A42aaT (96.66 %) [18], and H. saemangeumensis GSR0100T (96.57 %) [13]. Levels of sequence similarity to other genera (Adhaeribacter and Nibribacter) were less than 89.32 %.

Fig. 1
figure 1

A neighbor-joining tree based on the 16S rRNA gene sequences showing the phylogenetic relationship between strain DG7AT and other closely related taxa. A bar represents 0.02 substitutions per nucleotide position. Bootstrap values (expressed as percentages of 1,000 replications) greater than 50 % are shown at the branch points. Filled circles indicate the common nodes recovered from either the maximum parsimony algorithm or the maximum likelihood tree. Filled double circles indicate that the corresponding nodes were recovered in both the maximum parsimony tree and the maximum likelihood tree

Full size image

Chemotaxonomic Characteristics

The major cellular fatty acids of strain DG7AT were iso-C15:0 (24.87 %), anteiso-C15:0 (21.42 %), and summed feature 3 (C16:1 ω6c and/or C16:1 ω7c) (15.01 %), which are predominant in most Hymenobacter species. The minor fatty acids of strain DG7AT were iso-C14:0, iso-C15:0 3OH, iso-C15:1 G, C16:0, iso-C16:0, iso-C16:0 3OH, iso-C16:1 H, C16:1 ω5c, iso-C17:0, antesio-C17:0, C17:0 2OH, iso-C17:0 3OH, and Summed Feature 5 (anteiso-C17:1 B and/or iso-C17:1 I). The major cellular fatty acids of strain DG7AT are common to closely related species. However, some qualitative and quantitative differences in the fatty acid composition were observed between the novel strain and the other closely related Hymenobacter species. Strain DG7AT had smaller amounts of iso-C15:1 G (1.64 %), whereas other Hymenobacter species (H. soli KACC 13040T and H. glaciei JCM 17225T) had larger amounts of the corresponding fatty acids. In addition, the fatty acids iso-C14:0 and anteiso-C17:0 composed of more than 1 % of the fatty acids in strain DG7AT, but were present at less than 1 % or not detected at all in the closely related species. Strain DG7AT also had larger amounts of C16:0 (5.31 %), whereas other closely related Hymenobacter species had smaller amounts of the corresponding fatty acids (Table 2). The major polar lipid found in strain DG7AT was phosphatidylethanolamine (PE), which is similar to other Hymenobacter species [3, 13]. Minor amounts of an unknown aminolipid (AL), unknown aminophospholipid (APL), unknown glycolipid (GL), and unknown polar lipid (L) were also found (Supplementary Fig. S3).

Table 2 Cellular fatty acid profiles of strain DG7AT and closely related reference strains
Full size table

Genomic Characteristics

When examining DNA, strain DG7AT exhibited low DNA–DNA relatedness values with the closely related species H. soli KACC13040T (5.3 ± 1.8 %; reciprocal analysis, 36.3 ± 2.2 %), H. glaciei JCM17225T (5.6 ± 0.8 %), H. antarcticus JCM17213T (3.2 ± 0.8 %), and H. saemangeumensis KACC16452T (8.1 ± 0.8 %) (Supplementary Table 1). DNA–DNA hybridization levels between strain DG7AT and other type strains were determined to be less than 70 %, which is the threshold for delineating a genomic species [28, 34]. Thus, our results support the placement of strain DG7AT as a representative of a separate and previously unrecognized genomic species.

The G+C contents of genomic DNA from strain DG7AT and H. soli KACC13040T were 63.5 and 57.1 mol%, respectively (58.8 mol% for H. soli PB17T from Kim et al. [14]). Their G+C contents differ by over 5 %, providing additional support that strain DG7AT is not the same species, but the same genus as H. soli KACC13040T. It is now generally accepted that bacteria with DNA differing by more than 5 % G+C content should not be assigned to the same species, and those differing by more than 10 % should not be classified in the same genus [4, 23, 26, 29, 33].

UV Radiation Resistance Analysis

Survival of strain DG7AT against UV radiation was examined. Strain DG7AT showed the characteristic shoulder of resistance in survival curves, as shown by UV radiation-resistant Deinococcus radiodurans [1] (Supplementary Fig. 4). Whereas E. coli could not survive even at the lowest dose tested (300 J/m2), 60.0 % survival was observed at this dose for strain DG7AT (87.5 % for D. radiodurans R1T), and 0.3 % of DG7AT cells were able to survive after exposure to UV doses as high as 600 J/m2.

Taxonomic Conclusion

Strain DG7AT has similar characteristics as other Hymenobacter species, defined by the presence of phosphatidylethanolamine (PE) as the major polar lipid, and iso-C15:0, anteiso-C15:0, and summed feature 3 (C16:1 ω7c and/or C16:1 ω6c) as the abundant fatty acids. The result of DNA–DNA hybridization implied that strain DG7AT was not the same species as the closest related species, as the DNA–DNA relatedness value was lower than 70 %. The physiological characters differentiating the strain DG7AT from the other reference species are shown in Table 1.

Description of Hymenobacter terrae sp. nov

Hymenobacter terrae (ter’ra. L. gen. n. terrae of the earth)

Cells are 0.6–1.1 μm wide and 2.1–0.3.0 μm long, Gram-negative, non-motile, and rod-shaped, when grown on R2A agar (Difco) at 25 °C for 3 days. Colonies on R2A agar are pink to red-colored, circular, smooth and slimy. They are positive for both oxidase and catalase activities. The cells are able to grow at a temperature range of 12–30 °C (optimum 25–30 °C), but not at 4, 37, and 42 °C. Cells grow on R2A and ASM, but not on LB, TSA, and NA media at 25 °C. Efficient growth occurs at pH 6–9, with weak growth occuring at pH 5 and 10–11. Cells grow optimally in the absence of NaCl. Cells cannot reduce nitrate to nitrite or nitrogen. Glucose fermentation and indole production are also negative (API 20NE).

In tests with the API Zym system, cells tested positive for N-acetyl-β-glucosaminidase, acid phosphatase, alkaline phosphatase, cysteine arylamidase, esterase (C4), esterase (C8), α-galactosidase, β-galactosidase (OPNG), α-glucosidase (weakly), leucine arylamidase, naphtol-AS-BI-phosphohydrolase, and valine arylamidase. Cells tested negative for α-chymotrypsin, α-fucosidase, β-glucosidase, β-glucuronidase, lipase (C14), α-mannosidase, and trypsin.

Acid is produced weakly (API 50CH) with the use of l-arabinose, esculin ferric citrate, d-fructose, d-galactose, glycogen, 5-ketogluconate (positive), d-lactose, d-maltose, d-raffinose, d-saccharose (sucrose), d-trehalose, and d-xylose. Acid is not produced with N-acetyl-glucosamine, d-adonitol, amidon, amygdalin, d-arabinose, d-arabitol, l-arabitol, arbutin, d-cellobiose, dulcitol, erythritol, d-fucose, l-fucose, gentiobiose, gluconate, d-glucose, glycerol, inositol, inulin, 2-ketogluconate, d-lyxose, d-mannitol, d-mannose, d-melezitose, d-melibiose, methyl-α-d-mannopyranoside, methyl-α-d-glucopyranoside, methyl-β-d-xylose, l-rhamnose, d-ribose, salicin, d-sorbitol, l-sorbose, d-tagatose, d-turanose, xylitol, or l-xylose. The predominant cellular fatty acids of strain DG7AT are iso-C15:0 (24.87 %), anteiso-C15:0 (21.42 %), and summed feature 3 (C16:1 ω7c and/or C16:1 ω6c) (15.01 %). Strain DG7AT has phosphatidylethanolamine (PE) as the major polar lipid and a G+C content of 63.5 mol%. The type strain DG7AT (= KCTC 32554T = KEMB 9004-164T = JCM 30007T) was isolated from a soil sample collected in Seoul, (GPS; N 37° 34′ 30″ E 127° 00′ 30″), South Korea.