Introduction

In order to get new microbial resources that are capable of producing new metabolites, it is useful to isolate untapped actinomycete strains. We have a particular interest in non-Streptomyces actinomycetes as sources of new bioactive compounds. The majority of actinomycetes isolated from soil samples belong to the genus Streptomyces,1 while few non-Streptomyces actinomycetes are typically isolated. We have isolated rare actinomycetes from plant roots, including members of a proposed novel genus, Phytohabitans,2 and a novel species, Streptosporangium oxazolinicum,3 which produces the new bioactive compounds known as spoxazomicins.4 Therefore, we focused one line of our search for novel actinomycetes on plant roots.

Pozzi et al.5 reported that Actinoallomurus strains possess the ability to produce various bioactive compounds. Many Actinoallomurus strains have been found among our isolates. In the process of the screening search for new bioactive compounds in these strains, two strains among our isolates, K10-0485T and K10-0528T, were classified as novel species of Actinoallomurus. At present, the genus Actinoallomurus comprises 12 species:6 Actinoallomurus acaciae,7 A. amamiensis, A. caesius, A. coprocola, A. fulvus, A. iriomotensis, A. luridus, A. oryzae,8 A. purpureus, A. radicium,9 A. spadix and A. yoronensis. Some type strains of these species, such as A. acaciae, A. oryzae and A. radicium, were isolated from plants. In this paper, we report the taxonomic characteristics of the strains K10-0485T and K10-0528T.

Materials and methods

Samples of the perennial plant Ophiopogon japonicus, which is native to East Asia, were collected in Kanagawa Prefecture in May 2009. In accordance with the method described by Inahashi et al.,2 actinomycete strains were isolated from the roots on CMC agar (carboxymethyl cellulose 1.0%, ISP (International Streptomyces Project) medium 510 1.7%, distilled water, pH 7.0) and water proline agar (proline 1.0%, agar 1.5%, tap water, pH 7.0). Genomic DNA from all isolates was prepared by sonication of the cell suspension11 and 16S rRNA gene sequences were analyzed as described previously.2 The phylogenetically closest neighbors were identified by BLAST search using DDBJ database (http://blast.ddbj.nig.ac.jp/top-j.html). Evolutionary distances12 were estimated by SeaView version 4.2.13 Multiple alignments with selected sequences were calculated using the ClustalW2 program. The phylogenetic tree was constructed based on the neighbor-joining method,14 maximum-likelihood method15 and the maximum-parsimony method.16 Data were re-sampled with 1000 bootstrap replications.17 The values of sequence similarities with the closest strains were determined using the EzTaxon server.18 The strains K10-0485T and K10-0528T, as well as Actinoallomurus luridus NBRC 103683T (=TT02-15T), were cultured for 3 weeks at 27 °C in order to observe cultural and morphological characteristics on ISP media 2, 3, 4 and 7, and HV agar19 and YS agar (yeast extract 2.0%, starch 1.0%, agar 1.5%, pH 7.0). The color of aerial and vegetative mycelia and soluble pigments were determined using the Color Harmony Manual.20 Morphological characteristics were observed by light microscopy and scanning electron microscopy (JSM-5600, JEOL, Tokyo, Japan). The temperature range, pH range and the NaCl tolerance for growth were determined using ISP medium 2. Utilization of carbohydrates as the sole carbon source was tested using ISP medium 9.21 Utilization of urea was tested using the method of Gordon et al.22 ISP medium 4 was used for starch hydrolysis, ISP medium 8 was used for nitrate reduction, glucose–peptone–gelatin medium (glucose 2.0%, peptone 0.5%, gelatin 20%, pH 7.0) was used for gelatin hydrolysis, 10% skim milk was used for coagulation and peptonization of milk, and skim milk agar was used for casein hydrolysis.22 Enzyme activities were determined using the API ZYM system (BioMérieux, Lyon, France) according to the manufacturer's instructions. Biomass for the molecular systematics and chemotaxonomic studies was obtained after cultivation on a rotary shaker in yeast extract-glucose broth (yeast extract 1.0%, glucose 1.0%, pH 7.0) for 1 week at 27 °C. Isoprenoid quinones extracted as described by Collins et al.23 were analyzed by LC/MS (JMS-T 100LP, JEOL) using a CAPCELL PAK C18 UG120 (Shiseido, Tokyo, Japan) with methanol/2-propanol (7:3). Purified cell wall was obtained using the method of Kawamoto et al.,24 and the amino-acid composition of hydrolyzed cell walls was determined by thin layer chromatography. The N-acyl types of muramic acid were determined using the method of Uchida and Aida.25 Phospholipids in cells were extracted and identified using the method of Minnikin et al.26 The presence of mycolic acids was examined by thin layer chromatography in accordance with Tomiyasu.27 Whole-cell sugar composition was analyzed according to the methods of Becker et al.28 Methyl esters of cellular fatty acids were prepared by direct transmethylation with methanolic hydrochloride, and were analyzed on a GLC system (HP 6890, Hewlett Packard, Palo Alto, CA, USA). Identification and quantification of the fatty acid methyl esters, as well as numerical analysis of fatty acid profiles, were performed according to the instructions for the Microbial Identification System I with ACTIN1 method.29 For G+C content and DNA–DNA hybridization, chromosomal DNA was prepared in accordance with the procedure of Saito and Miura.30 DNA G+C content was determined by HPLC according to the method of Tamaoka and Komagata,31 and DNA–DNA hybridization was performed using the photobiotin-labeling method of Ezaki et al.32

Results and Discussion

Isolation of actinomycetes from plant roots

A total of 135 actinomycete strains were isolated from the roots of a single specimen of Ophiopogon japonicus. The 16S rRNA gene partial sequences (approximately 500 bp) showed that the isolates comprised the genera Planotetraspora (23%), Acrocarpospora (18%), Actinoallomurus (13%), Streptomyces (13%) and others (33%) (Table 1). This suggests that non-Streptomyces strains were isolated with high frequency. There were 17 strains of the genus Actinoallomurus among the isolates. Phylogenetic analysis by the almost full-length sequences showed that these strains fell into three clusters; the first cluster comprised strains K10-0467 and K10-0474, the second consisted of strains K10-0475 and K10-0528T, and the third cluster comprised strain K10-0485T and 12 similar strains (Figure 1). This indicates that various strains belonging to the genus Actinoallomurus were isolated from a single plant. Furthermore, the two clusters including K10-0485T and K10-0528T were not clustered with previously described species, so a taxonomic study of K10-0485T and K10-0528T was carried out.

Table 1 Classification of isolates by 16S rRNA gene sequences
Full size table
Figure 1
figure 1

Neighbor-joining tree, based on 16S rRNA gene sequences, showing relationship between isolates from plant roots and members of the genus Actinoallomurus. Only bootstrap values above 70% (percentages of 1000 replications) are indicated. Bar, 0.01 nucleotide substitutions per site.

Full size image

Taxonomic study of strains K10-0485T and K10-0528T

Morphological, cultural and physiological characteristics

Strain K10-0485T grew well on ISP media 2 and 3, and on YS agar. Spores on aerial mycelia were only produced on HV agar after 3 weeks at 27 °C. Vegetative mycelia were branched but not fragmented. The colony color of the strain was white to pale yellow, and soluble pigment was not produced (Figure 2, Table 2). The temperature and pH range for growth were 20–40 °C and pH 5–8, with optimal growth at 28–36 °C and pH 5–7. The strain did not grow on 3% NaCl medium. Casein was degraded. Starch was hydrolyzed. Gelatin was weakly liquefied. Milk was not peptonized or coagulated. Nitrate was not reduced to nitrite. Melanin was not produced. The strain utilized L-arabinose, D-fructose, D-galactose, D-glucose, myo-inositol, L-rhamnose, D-sucrose, D-xylose and cellulose, but did not utilize dulcitol, maltose, D-mannitol, melibiose, raffinose or D-sorbitol. Using the API ZYM system, N-acetyl-β-glucosaminidase, acid phosphatase, cystine allylamidase, α-chymotrypsin, esterase (C4), α-glucosidase, β-glucosidase, β-glucuronidase, leucine allylamidase, α-mannosidase, naphthol-AS-BI-phosphohydrolase and valine allylamidase were positive, while alkaline phosphatase, esterase lipase (C8), α-galactosidase, β-galactosidase and trypsin were weakly positive, and α-fucosidase and lipase (C14) were negative (Table 3).

Figure 2
figure 2

Scanning electron micrograph of strain K10-0485T grown on HV agar for 3 weeks at 27 °C.

Full size image
Table 2 Cultural characteristics of strains K10-0485T, K10-0528T and A. luridus TT02-15T
Full size table
Table 3 Differential physiological characteristics of strains K10-0485T, K10-0528T and A. luridus TT02-15T
Full size table

Strain K10-0528T grew well on ISP media 2 and 3, and on YS agar. Aerial mycelia were only produced on ISP medium 2 for 6 weeks at 27 °C, but spores were not observed. Vegetative mycelia were branched but not fragmented. The colony color of the strain was white to gray–purple and purple soluble pigment was produced (Table 2). The temperature and pH range for growth were 15–39 °C and pH 5–8, with optimal growth at 23–39 °C and pH 5–6. The strain did not grow on 2% NaCl medium. Casein was degraded. Starch was hydrolyzed. Gelatin was not liquefied. Milk was not peptonized or coagulated. Nitrate was not reduced to nitrite. Melanin was not produced. The strain utilized L-rhamnose and D-sucrose, but did not utilize L-arabinose, dulcitol, D-fructose, D-galactose, D-glucose, myo-inositol, maltose, D-mannitol, melibiose, raffinose, D-sorbitol, D-xylose or cellulose. Using the API ZYM system N-acetyl-β-glucosaminidase, acid phosphatase, alkaline phosphatase, cystine allylamidase, α-fucosidase, β-galactosidase, α-glucosidase, β-glucosidase, leucine allylamidase, naphthol-AS-BI-phosphohydrolase and valine allylamidase were positive, esterase (C4), esterase lipase (C8), α-galactosidase and α-mannosidase were weakly positive, and α-chymotrypsin, β-glucuronidase lipase (C14) and trypsin were negative (Table 3).

Chemotaxonomy

Strains K10-0485T and K10-0528T both contained meso-diaminopimelic acid, lysine, alanine and glutamic acid in the cell-wall peptidoglycan, and galactose, glucose, madurose, mannose and ribose were detected as whole-cell sugars. The N-acyl type of muramic acid was acetyl. Phosphatidylglycerol and diphosphatidylglycerol were detected, but phosphatidylethanolamine, phosphatidylcholine and phosphatidylinositol were not found. Mycolic acids were not detected. The major menaquinones were MK-9 (H8) (72%), MK-9 (H6) (25%) for strain K10-0485T, and MK-9 (H6) (60%) and MK-9 (H8) (40%) for strain K10-0528T. The major cellular fatty acids were iso-C16:0 (41.3%), 10-methyl C17:0 (17.2%) and anteiso-C17:0 (14.6%) for K10-0485T, and iso-C16:0 (30.9%), anteiso-C17:0 (18.5%) and ω9c C17:1 (15.4%) for K10-0528T (Table 4). The G+C content of the genomic DNA was 72 mol% in both strains. These chemotaxonomic properties of strains K10-0485T and K10-0528T, which contain both meso-diaminopimelic acid and lysine as diamino acids in the cell-wall peptidoglycan, madurose and galactose as whole-cell sugars, and MK-9 (H8) and MK-9 (H6) as the predominant menaquinones, indicated that they are members of the genus Actinoallomurus.

Table 4 Cellular fatty acid compositions (%) of strains K10-0485T, K10-0528T and A. luridus TT02-15T
Full size table

Phylogenetic analysis

Phylogenetic analysis, based on the 16S rRNA gene sequences, indicated that the two strains belong to the genus Actinoallomurus, and that strain K10-0485T is closest to A. luridus TT02-15T, while strain K10-0528T is closest to A. spadix JCM 3146T, A. purpureus TTN02-30T and A. luridus TT02-15T (Figure 3).

Figure 3
figure 3

Neighbor-joining tree, based on 16S rRNA gene sequences, showing relationship between strains K10-0485T, K10-0528T and members of the genus Actinoallomurus. Only bootstrap values above 70% (percentages of 1000 replications) are indicated. Solid circle branches also recovered in the maximum-likelihood tree; asterisks show branches also recovered in the maximum-parsimony tree; Bar, 0.01 nucleotide substitutions per site.

Full size image

The organisms showing the highest similarity values for strain K10-0485T were A. luridus TT02-15T (99.1%), A. purpureus TTN02-30T (98.3%) and A. spadix JCM 3146T (97.7%). The organisms showing the highest similarity values to strain K10-0528T were strains K10-0485T (98.1%), A. purpureus TTN02-30T (98.0%), A. spadix JCM 3146T (98.0%) and A. luridus TT02-15T (97.9%). As the highest similarity value to strain K10-0528 T was 98.1%, much lower than the boundary value of 98.7%33 for distinguishing organisms of different species, this strain should be recognized as a separate species. The DDBJ accession numbers of the 16S rRNA gene sequences of strains K10-0485T and K10-0528T are AB668306 and AB668307, respectively.

DNA–DNA hybridization

DNA–DNA relatedness values between strain K10-0485T and A. luridus TT02-15T were 53–63%, below the recommended DNA–DNA relatedness cutoff point for species delineation of 70%.34

Conclusion

We isolated 135 actinomycete strains from the roots of the Japanese plant Ophiopogon japonicus. The 16S rRNA gene partial sequences indicated that non-Streptomyces strains were isolated at high rate (87.4%) containing 17 Actinoallomurus strains, which possess the ability to produce various bioactive compounds (Table 1). Although the 17 strains were isolated from a sole plant root, they were separated phylogenetically into three groups based on the almost full-length 16S rRNA gene sequences (Figure 1). Two strains, K10-0485T and K10-0528T, showed low similarities with nearest previously described species, indicating the need for further taxonomic studies.

The morphological and chemotaxonomic properties of strains K10-0485T and K10-0528T supported the notion that these strains belong to the genus Actinoallomurus. DNA–DNA relatedness values between strain K10-0485T and the closest species (A. luridus TT02-15T) were below the critical value of 70%. Furthermore, strain K10-0485T can be distinguished from related species based on cultural and physiological characteristics, as shown in Tables 2 and 3. The 16S rRNA gene sequence similarities between strain K10-0528T and possibly related strains were low (less than 98.1%). These observations support the contention that strains K10-0485T and K10-0528T represent two novel species in the genus Actinoallomurus, for which the names Actinoallomurus liliacearum sp. nov. and Actinoallomurus vinaceus sp. nov., respectively, are proposed.

Description of Actinoallomurus liliacearum sp. nov.

Actinoallomurus liliacearum (li.li.a.ce.a'rum. N.L. pl. n. Liliaceae, a scientific family name; referring to the isolation of the strain from the roots of Liliaceae plant, Ophiopogon japonicas).

The color of colonies is white to gray. Vegetative mycelia are branched and not fragmented, and the color of aerial mycelia is white. Spore chains form spirals. Spores are smooth. Growth occurs at 20–40 °C and pH 5–8, with no growth in the presence of 3% NaCl. Casein and cellulose are degraded. Starch is hydrolyzed, but urea is not. Gelatin is weakly liquefied. Milk is not peptonized or coagulated. Nitrate is not reduced to nitrite. Melanin is not produced. L-arabinose, D-fructose, D-galactose, D-glucose, myo-inositol, L-rhamnose, D-sucrose and D-xylose are utilized, but dulcitol, maltose, D-mannitol, melibiose, raffinose and D-sorbitol are not. According to the API ZYM system, N-acetyl-β-glucosaminidase, acid phosphatase, cystine allylamidase, α-chymotrypsin, esterase (C4), α-glucosidase, β-glucosidase, β-glucuronidase, leucine allylamidase, α-mannosidase, naphthol-AS-BI-phosphohydrolase and valine allylamidase are positive, alkaline phosphatase, esterase lipase (C8), α-galactosidase, β-galactosidase and trypsin are weakly positive, α-fucosidase and lipase (C14) are negative. Major fatty acids are iso-C16:0, 10-methyl C17:0 and anteiso-C17:0. The G+C content of the genomic DNA of type strain is 72 mol%. The species description is based on a single strain. The type strain is K10-0485T (=JCM 17938T, BCC 49424T, NBRC 108762T), which was isolated from the root of Ophiopogon japonicus in Yokohama, Kanagawa, Japan.

Description of Actinoallomurus vinaceus sp. nov.

Actinoallomurus vinaceu (vi.na'ceus. L. masc. adj. vinaceus grape, referring to the color of the colonies).

The color of colonies is gray to purple. Vegetative mycelia are branched and not fragmented, while the color of aerial mycelia is white. Growth occurs at 15–39 °C and pH 5–8, with no growth in the presence of 2% NaCl. Casein is degraded, but cellulose is not. Starch is hydrolyzed, but urea is not. Gelatin is not liquefied. Milk is not peptonized or coagulated. Nitrate is not reduced to nitrite. Melanin is not produced. L-rhamnose and D-sucrose are utilized, but L-arabinose, dulcitol, D-fructose, D-galactose, D-glucose, myo-inositol, maltose, D-mannitol, melibiose, raffinose, D-sorbitol and D-xylose are not. According to the API ZYM system, N-acetyl-β-glucosaminidase, acid phosphatase, alkaline phosphatase, cystine allylamidase,α-fucosidase, β-galactosidase, α-glucosidase, β-glucosidase, leucine allylamidase, naphthol-AS-BI-phosphohydrolase and valine allylamidase are positive, esterase (C4), esterase lipase (C8), α-galactosidase and α-mannosidase are weakly positive, α-chymotrypsin, β-glucuronidase lipase (C14) and trypsin are negative. Major fatty acids are iso-C16:0, anteiso-C17:0 and ω9c C17:1. The G+C content of the genomic DNA of type strain is 72 mol%. The species description is based on a single strain. The type strain is K10-0528T (=JCM 17939T, BCC 49425T, NBRC 108763T), which was isolated from a root of Ophiopogon japonicus in Yokohama, Kanagawa, Japan.