Introduction

Taxonomic relationships within the genus Streptomyces have been clarified by the application of genotypic and phenotypic methods (Goodfellow et al. 1992; Manfio et al. 1995; Anderson and Wellington 2001). It is now apparent that many streptomycete type strains can be assigned to distinct multimembered species-groups, as exemplified by species classified in the Streptomyces albidoflavus (Lanoot et al. 2005), Streptomyces griseus (Liu et al. 2005), Streptomyces violaceoruber (Duangmal et al. 2005), Streptomyces violaceusniger (Sembiring et al. 2000) and Streptomyces yeochonensis (Xu et al. 2006) 16S rRNA gene clades. Polyphasic taxonomic studies based on representatives of validly described species show the genus to be overspeciated (Hatano et al. 2003; Lanoot et al. 2002, 2004; Liu et al. 2005) though complementary studies on unknown streptomycetes show that the taxon as a whole is underspeciated (Manfio et al. 2003; Xu et al. 2006).

Members of the S. violaceusniger 16S rRNA gene clade form a gray aerial spore mass and a grayish-yellow substrate mycelium on oatmeal agar, and produce aerial hyphae that differentiate into spiral chains of rugose ornamented spores (Sembiring et al. 2000; Ward and Goodfellow 2004). The clade currently contains 13 validly described species that include S. hygroscopicus (Jensen 1931) Labeda and Lyons 1991, S. malaysiensis Al-Tai et al. 1999, S. melanosporofaciens Arcamone et al. 1959, S. sporoclivatus Preobrazhenskaya et al. 1986, S. violaceusniger (Waksman and Curtis 1916) Labeda and Lyons 1991 and S. yatensis Saintpierre et al. 2003. The remaining species, S. asiaticus, S. cangkringensis, S. indonesiensis, S. javensis, S. rhizosphaericus and S. yogyakartensis, were proposed following a taxonomic study of S. violaceusniger-like strains isolated from the rhizosphere of Paraserianthes falcataria (Sembiring et al. 2000), a tropical leguminous tree used for improvement of poor soils and the supply of wood pulp for paper. Phenotypic tests are available to distinguish between species classified in the S. violaceusniger 16S rRNA gene clade (Saintpierre et al. 2003).

Putative members of the S. violaceusniger 16S rRNA gene clade are considered to be relatively easy to recognize on media designed to be selective for streptomycetes as they form colonies with a gray aerial spore mass that subsequently turns black and mucilaginous. However, in a pilot study strains previously identified as S. hygroscopicus and S. violaceusniger were shown to be misclassified as they fell outside the S. violaceusniger 16S rRNA gene clade (Ward and Goodfellow 2004), a finding in line with DNA:DNA relatedness (Labeda and Lyons 1991) and numerical taxonomic (Kämpfer et al. 1991) data which show that strains assigned to these species form a heterogeneous group. The organisms outside the S. violaceusniger 16S rRNA gene clade have been considered to merit species status as S. auranticolor, S. phaeogriseichromogenes, S. phaeoluteichromogenes and S. phaeoluteigriseus (Sembiring et al. 2000; Ward and Goodfellow 2004) though these names do not have any formal nomenclatural standing. This also applies to species assigned to the S. violaceusniger 16S rRNA gene clade and designated as S. albiflaviniger, S. griseiniger and S. geldanamycinus (Sembiring et al. 2000; Ward and Goodfellow 2004).

It is important to clarify the taxonomy of S. hygroscopicus, S. violaceusniger and related strains, not least because members of the S. violaceusniger 16S rRNA gene clade are a source of antibacterial and antifungal metabolites (DeBoer et al. 1970; Lam et al. 1990; Tripathi et al. 2004), biological control agents (Trejo-Estrada et al. 1998a, b; Chamberlain and Crawford 1999), enantioselective biocatalysts (Molinari et al. 2005) and immunosuppressants, including rapamycin (Vezina et al. 1975). Authenticated members of the S. violaceusniger 16S rRNA gene clade show the same pattern of HPLC-detected secondary metabolites, namely eliaophylin, geldanamycin, nigericin and a polyene (Ward and Goodfellow 2004), a result consistent with those of earlier studies (Allen and Ritchie 1994; Fang et al. 2000).

The present study was designed to establish the taxonomic status and relationships of 16 strains assigned to species classified in the S. violaceusniger 16S rRNA gene clade and 5 strains received as S. violaceusniger that fall outside the clade (Sembiring et al. 2000; Ward and Goodfellow 2004). These organisms were the subject of a polyphasic study which confirmed that five of them merited species status outside the S. violaceusniger 16S rRNA gene clade, the remaining ones were assigned to three novel and three established species within the clade.

Materials and methods

Organisms and cultural conditions

Sixteen strains received as S. hygroscopicus NRRL B-1477, NRRL 2339 and NRRL 2387T, S. hygroscopicus subsp. geldanus NRRL 3602T, S. melanosporofaciens NRRL B-12234T, S. sparsogenes NRRL 2940T, S. sporoclivatus NRRL B-24330T and S. violaceusniger ISP 5182, ISP 5563, NRRL B-1356, NRRL B-1476, NRRL B-1478T, NRRL B-1865, NRRL B-5799, NRRL 2834 and NRRL 8097 were studied. The organisms were maintained on glucose-yeast extract-malt extract agar (DSMZ Catalogue 1998) at 4°C and as glycerol suspensions (20%, v/v) at −20°C. Biomass for the 16S rRNA gene sequencing analyses was prepared by growing the test organisms on a non-sporulating agar medium (Sanglier et al. 1992) for 14 days at 28°C.

Sequencing of 16S rRNA genes

Extraction of genomic DNA and PCR-amplification of 16S rRNA genes from the 16 strains, and from the type strain of S. indonesiensis, were carried out as described by Pitcher et al. (1989), using the modifications of Sembiring (2000). The amplified fragments were purified with Nucleospin Extraction kits (Biogen Ltd.) and sequenced directly using ABI PRISMR BigDyeTM Terminator Cycle Sequencing kits (Applied Biosystems) and previously described oligonucleotide primers (Lane 1991; Chun and Goodfellow 1995). Sequencing gel electrophoresis was carried out and the nucleotide sequences automatically obtained by using an Applied Biosystems DNA sequencer (model 377) and software provided by the manufacturer. The 16S rRNA gene sequences were aligned manually with available streptomycete nucleotide sequences retrieved from the DDBJ/EMBL/GenBank databases, using the pairwise alignment option and 16S rRNA secondary structure information held in the PHYDIT program (available at http://plaza.snu.ac.kr/∼jchun/phydit/).

Unrooted phylogenetic trees based on almost complete nucleotide sequences were inferred by using the least-squares (Fitch and Margoliash 1967), maximum-likelihood (Felsenstein 1981), maximum-parsimony (Fitch 1972) and neighbor-joining (Saitou and Nei 1987) tree-making algorithms from the PHYLIP software package (Felsenstein 1993). Evolutionary distance matrices were generated for the neighbor-joining and least-squares methods, as described by Jukes and Cantor (1969) and the resultant tree topologies evaluated in a bootstrap analysis (Felsenstein 1985) based on 1000 resamplings from the neighbor-joining dataset, using the SEQBOOT and CONSENSE programs from the PHYLIP package (Felsenstein 1993). The root position of the unrooted neighbor-joining tree was estimated using Kitasatospora kifunensis NBRC 15206T (accession number AJ 852052) as the outgroup.

Determination of DNA: DNA relatedness

Total genomic DNA for estimation of DNA-DNA relatedness values between nine pairs of organisms known to share high 16S rRNA gene sequences and two pairs of control strains was isolated from strains grown in Tryptone Soy broth for 7 days at 28°C. Cells were resuspended in 500 μl of 0.5 × TE buffer (pH8), approximately 400 μg of sterile glass beads added, and the preparations ribolised for 20 s at speed setting 4.0; 2.5 μl of lysozyme (50 mg/ml) and 5 μl of proteinase K (20 mg/ml) were added prior to incubation at 37°C for 2 h. The resultant preparations were centrifuged, the supernatants separated and incubated at 75°C for 15 min then cooled to room temperature; 20 μl of RNAse (20 mg/ml) solution was added before incubation at room temperature for 2 min. DNA was precipitated from the samples by adding 200 μl of cold 99.5% ethanol and purified using a GenEluteTM Bacterial Genomic DNA Extraction kit (Sigma-Aldrich, St. Louis, U.S.A); the resultant DNA samples were eluted with 200 μl of sterile distilled water (pH8). DNA-DNA relatedness experiments were carried out between the selected pairs of strains (Table 1) by measuring the divergence between the thermal denaturation midpoint of homoduplex DNA and heteroduplex DNA (ΔT m), as described by Gonzalez and Saiz-Jimenez (2005).

Table 1 Relationships found between closely related members of the S. violaceusniger 16S rRNA clade based on ΔT m (°C) values together with approximate corresponding figures expressed as percentage DNA: DNA relatedness value
Full size table

Partial RNA polymerase β-subunit gene sequencing

Genomic DNA for sequencing of a ∼300 base pair segment of the rpoB (RNA polymerase β subchain) gene was isolated from strains grown on yeast extract-malt extract agar plates (ISP medium 2; Shirling and Gottlieb 1966) using UltraCleanTM Microbial DNA Isolation kits (MoBio Laboratories Inc., Solana Beach, CA). The partial rpoB genes were amplified by using a modification of the procedure described by Kim et al. (2004) with the omission of loading buffer from the reaction mixture and amplicon clean-up performed using UltraCleanTM PCR kits (MoBio Laboratories Inc.). Sequencing followed the procedure of Kim et al. (2004), the resultant partial rpoB sequences were deposited in GenBank under accession numbers DQ241992 through to DQ242018. The sequences were aligned against published sequences for taxa classified in the family Streptomycetaceae within ARB (Ludwig et al. 2004) and a phylogenetic tree constructed using the neighbor-joining method (Saitou and Nei 1987); the stability of the resultant groupings were estimated by bootstrap analysis (Felsenstein 1989). A maximum-likelihood phylogenetic tree was prepared after Felsenstein (1989), as modified by Olsen et al. (1994).

Morphology and cultural characteristics

The strains were grown on oatmeal (ISP medium 3; Küster 1959) and peptone-yeast extract-iron agar plates (ISP medium 6; Shirling and Gottlieb 1966) at 28°C for 14 and 4 days, respectively. The oatmeal agar plates were examined by eye to determine aerial spore mass color, substrate mycelial pigmentation and the color of any diffusible pigments; colors were recorded using National Bureau of Standards (NBS) Color Name Charts (Kelly 1958; NBS 1964). The peptone-yeast extract-iron agar plates were examined to see whether the strains produced melanin pigments.

Spore-chain morphology was determined by light and scanning electron microscopy (SEM) of 14-day-old cultures grown at 28°C on inorganic salts/starch agar (ISP medium 4; Difco). Spore-chain morphology was observed using a Nikon Optiphot binocular light microscope fitted with long working distance objectives; spore-chains were assigned to the morphological categories proposed by Pridham et al. (1958). Spore-surface ornamentation was determined on SEM preparations using a Cambridge Stereoscan 240 instrument; ornaments were assigned to either the categories of Tresner et al. (1961) or to the rugose section of Dietz and Mathews (1971). Light and SEM preparations were also examined for the presence of sclerotia, fragmentation of the substrate mycelium and for the formation of spores on the substrate mycelium. Aerial spore mass color, substrate mycelial pigmentation and the production of diffusible pigments were recorded on several agar media (Tables 2 and 3) after incubation for 14 days at 28°C.

Numerical taxonomy

The 24 tested strains consisted of the 16 organisms and 8 type strains of Streptomyces species known to produce rugose spores. The organisms were examined for 142 unit characters using the media and methods described by Williams et al. (1983). In addition, all the strains assigned to the S. violaceusniger 16S rRNA gene clade were examined for 19 enzymes using API® ZYM kits (bioMérieux, Inc, Durham, USA), following the manufacturer’s guidelines. Unit characters were scored either plus (1) or minus (0) and the resultant data examined using the NTSYSpc program (version 2.0; Numerical Taxonomy and Multivariate Analysis System: Rohlf 1998) and the simple matching coefficient (SSM; Sokal and Michener 1958), which includes both positive and negative matches. Clustering was achieved using the unweighted pair group method with arithmetic averages algorithm (UPGMA; Sneath an Sokal 1973). The final data matrix contained differential information on 24 strains and 92 unit characters. A SSM, UPGMA analysis was carried out on the 19 strains assigned to the S. violaceusniger clade using 56 differential unit characters and the results presented as a dendrogram (Fig. 3).

Results

16S rRNA gene sequences

Eleven out of the 16 strains were assigned to the S. violaceusniger 16S rRNA gene clade, including the type strain of S. sporoclivatus (Fig. 1). The taxonomic integrity of the clade is supported by results from all of the tree-making algorithms and by a bootstrap value of 88% in the neighbor-joining analysis. The members of the clade shared 16S rRNA gene similarities within the range 97.8–100%.

Fig. 1
figure 1

Neighbor-joining tree showing relationships between the tested strains and between them and representatives of the genus Streptomyces based on almost complete 16S rRNA gene sequences. The asterisks denote branches that were recovered using the least- squares (Fitch and Margoliash 1967), maximum-likelihood (Felsenstein 1981) and maximum-parsimony (Fitch 1972) tree-making algorithms. The numbers at the nodes indicate the levels of bootstrap support (%) based on a neighbor-joining analysis of 1,000 resampled datasets. The tested strains are given in bold. The arrow gives the estimated root position of the tree. The scale bar indicates 0.02 nucleotide substitutions per nucleotide position. GenBank accession numbers are given in parentheses

Full size image

The almost complete 16S rRNA gene sequences obtained for S. violaceusniger strains NRRL B-1476 and ISP 5563 were identical. Streptomyces violaceusniger NRRL B-1356 shared a 16S rRNA gene similarity with each of these strains of 99.7%, a value that corresponds to 4 nucleotide (nt) differences at 1449 locations. Similarly, S. hygroscopicus NRRL 1477, NRRL 2339 and NRRL 2387T shared 16S rRNA gene similarities within the range 99.8% and 99.9%, values equivalent to up to two nt differences at 1465 sites. Streptomyces violaceusniger NRRL B-1478T shared 16S rRNA gene similarities with the S. hygroscopicus strains within the range 99.7% to 99.9%, values which corresponded to between 2 and 4 nt differences at 1449 locations.

Other organisms which shared high 16S rRNA gene similarities included the type strains of S. asiaticus and S. cangkringensis (99.7% similarity; 5 nt differences), S. cangkringensis and S. rhizosphaericus (99.6% similarity; 6 nt differences) and S. melanosporofaciens and S. sporoclivatus (99.8% similarity, 3 nt differences). Streptomyces hygroscopicus subsp. geldanus NRRL 3602T showed its closest 16S rRNA gene similarities with the type strains of S. melanosporofaciens (99.7%, 9 nt differences) and S. sporoclivatus (99.3%, 11 nt differences). High 16S rRNA gene similarities were also found between S. violaceusniger NRRL B-1865 and S. cangkringensis NRRL 41769T (99.6%, 6nt differences), S. indonesiensis DSM 41759T (99.7%, 4 nt differences) and S. rhizosphaericus DSM 41670T (99.6%, 6 nt differences). The remaining members of the S. violaceusniger 16S rRNA gene clade had relatively low similarities both with one another and with the above mentioned strains. The strains which fell outside the S. violaceusniger 16S rRNA gene clade, namely, S. sparsogenes NRRL 2940T and S. violaceusniger ISP 5182, NRRL 2834, NRRL B-5799 and NRRL 8097 shared relatively low 16S rRNA gene similarities with one another (range 95.7–98.3%) and with members of the S. violaceusniger clade (range 95.3–97.6%).

DNA:DNA relatedness studies

With a single exception the DNA:DNA relatedness studies carried out by thermal denaturation on phylogenetically close strains revealed differences in melting temperature above the cut-off point recommended for the delineation of genomic species (ΔT m > 5.0°C; Wayne et al. 1987), as shown in Table 1. It is evident from these results that the type strains of S. melanosporofaciens and S. sporoclivatus belong to a single genomic species (ΔT m 2.0°C; corresponding to approximately 90% relatedness) through the remaining pairs are quite closely related with ΔT m values within the range 5.6–7.0°C which corresponds to approximately 62–55% relatedness.

Partial RNA polymerase β-subunit gene sequences

The results based on the partial rpoB sequences are shown in Fig. 2. It is apparent that S. violaceusniger strains ISP 5182, NRRL B-5799 and NRRL B-8097 form single membered clusters. These organisms were found to be even more distant and separate from the other strains shown in Fig. 2 in the analysis of partial rpoB sequences available for the family Streptomycetaceae (data not shown). The remaining strains under study form two major clades, one of which encompassed S. hygroscopicus strains NRRL B-1477 and NRRL 2339 and S. violaceusniger strains NRRL B-1478 and NRRL 2834 and the other the rest of the tested strains, apart from those that formed single membered clusters. It is not clear why the partial rpoB sequence of S. hygroscopicus NRRL 2387T is so different from those of S. hygroscopicus (formerly S. endus) NRRL 2339 and S. hygroscopicus NRRL B-1477.

Fig. 2
figure 2

Neighbor-joining tree showing relationships between the tested strains and representatives of the genus Streptomyces based on partial nucleotide sequences (301 bp) of the RNA polymerase β-subunit gene (rpoB). Percentages at the nodes represent levels of bootstrap support from 1,000 resampled datasets; values less than 50% are not shown. Asterisks indicate the corresponding branches that were also recovered in the maximum-likelihood tree. The tested strains are given in bold. The scale bar indicates 0.01 nucleotide substitutions per nucleotide position. GenBank accession numbers are given in parentheses

Full size image

It was determined experimentally that the third codon position of the rpoB sequences accounted for most of the variability and that the distance tree calculated without this position was very close to that determined from the amino acid translation of those sequences (data not shown). With the elimination of the third codon position, S. violaceusniger strains ISP 5182, NRRL 8097 and NRRL B-5799 are still recovered as single membered clusters, the remaining strains form clades corresponding to the results obtained for all of the codon positions.

Morphological and cultural characteristics

Most of the 16 tested strains grew well on a range of agar media, but only S. violaceusniger ISP 5182 produced melanin pigments on peptone-yeast extract-iron agar (Tables 2, 3). Eleven of the strains formed rugose ornamented spores in spiral chains on inorganic salts/starch agar; three out of the nine S. violaceusniger strains produced flexuous spore chains composed of either smooth or ridged ornamented spores. Streptomyces sparsogenes NRRL 2940T and S. violaceusniger NRRL 8097 produced ridged and smooth surfaced spores, respectively in spiral chains.

Table 2 Color grouping and morphological characteristics of the tested strains*
Full size table
Table 3 Growth and cultural characteristics of the tested strains
Full size table

Numerical taxonomy

The 24 tested strains produced catalase; degraded DNA, elastin, gelatin, RNA, starch and tributyrin, but not keratin or pectin; grew on meso-erythritol, d-trehalose and d-xylose as sole sources of carbon, but not on aesculin, arbutin, d-arabinose, glucosamine, inulin, malonic acid, l-sorbose or pectin (all at 1%, w/v); grew in the presence of phenol (0.01%, w/v) and sodium chloride (4, 7 and 10%, w/v), but not at 4°C; and were resistant to ampicillin (32 and 64 μg ml−1), cephalosporin (32 and 64 μg ml−1), penicillin (10 and 20 μg ml−1) and tetracycline hydrochloride (32 and 64 μg ml−1), but sensitive to gentamicin sulphate, kanamycin sulphate, neomycin sulphate, streptomycin sulphate and tobramycin sulphate (all at 8 and 32 μg ml−1).

It was clear from the initial numerical taxonomic analysis that the organisms fell into two aggregate groups, one encompassed all of the strains assigned to the S. violaceusniger 16S rRNA gene clade and the other organisms that fell outside the clade (data not shown). All of the strains assigned to the S. violaceusniger 16S rRNA gene clade hydrolyzed urea; produced hydrogen sulphide; degraded arbutin, but not chitin, xanthine or xylan; used d-mannitol and d-mannose as sole carbon sources (all at 1%, w/v), and l-valine (0.1%, w/v) as a sole carbon and nitrogen source. They also produced acid phosphatase, N-acetyl-β-glucosaminidase, alkaline phosphatase, esterase (C4), esterase lipase (C8), naphthol-AS-BI-phosphohydrolase, but not cystine arylamidase, α-glucosidase, β-glucuronidase, α-fucosidase, lipase, α-mannosidase or valine arylamidase. The all positive and all negative properties were removed from the initial phenotypic database and the remaining 58 unit characters (Table 4) examined in a SSM, UPGMA analysis.

Table 4. Phenotypic properties that separate members of the Streptomyces violaceusniger 16S rRNA gene clade
Full size table

It is evident from Fig. 3 that the strains classified in the S. violaceusniger 16S rRNA gene clade can be assigned to 4 multimembered and 10 single-membered clusters defined at or above the 90% similarity (S-) level. Multimembered clusters were formed between the type strains of S. asiaticus and S. yogyakartensis at the 88% S- level, between S. hygroscopicus strains NRRL 1477, NRRL 2339 and NRRL 2387T also at the 88% S- level, between S. violaceusniger strains NRRL B-1476 and ISP 5563 at the 100% S-level, and between S. melanosporofaciens NRRL B-12234T and S. sporoclivatus NRRL B-24330T just above the 95% S-level. The differential phenotypic properties of the strains assigned to the S. violaceusniger 16S rRNA gene clade are shown in Table 4; those marked with an asterisk distinguish between members of putatively novel and established species.

Fig. 3
figure 3

Dendrogram showing relationships between members of the Streptomyces violaceusniger 16S rRNA gene clade based on an analysis of phenotypic properties using the SSM,UPGMA analysis

Full size image

Discussion

The circumscription of Streptomyces species needs to be based on a judicious combination of genotypic and phenotypic data (Manfio et al. 1995, 2003; Atalan et al. 2000; Liu et al. 2005). It is, therefore, encouraging that good congruence was found between the phenetic and phylogenetic data as all of the organisms assigned to the S. violaceusniger 16S rRNA gene clade formed a distinct multimembered cluster based on the numerical taxonomic analysis of the differential phenotypic properties. In contrast, the five organisms found outside the S. violaceusniger 16S rRNA gene clade formed a heterogeneous group. These results provide further evidence that members of the S. violaceusniger 16S rRNA gene clade can be distinguished from all other streptomycetes using a range of phenotypic properties, notably by their ability to form spiral chains of rugose ornamented spores and distinctive pigments on oatmeal agar. The colonial properties are significant as streptomycete groups based on aerial spore mass color, colony reverse and diffusible pigment colors on oatmeal agar, and the ability to form melanin pigments on peptone-yeast extract-iron agar, are predictive as representatives of streptomycete color groups key out to either established or novel species based on chemotaxonomic and computer-assisted identification procedures (Goodfellow and Haynes 1984; Williams and Vickers 1988; Atalan et al. 2000; Manfio et al. 2003; Liu et al. 2005).

The taxonomic integrity of the S. violaceusniger 16S rRNA gene clade was supported by results from all of the tree-making algorithms and by a high bootstrap value in the neighbor-joining analysis, a result that confirms and extends those from previous studies (Sembiring et al. 2000; Saintpierre et al. 2003; Ward and Goodfellow 2004). The phylogeny of the strains under study based on partial rpoB sequences was not completely congruent with the tree computed from 16S rRNA gene sequences, a similar finding was reported by Rickert et al. (2007) in a molecular analysis of Aureobacterium and Microbacterium strains. It appears that partial rpoB sequences have value for distinguishing between members of phylogenetically distant Streptomyces, but care must be taken in interpreting data for closely related streptomycetes.

It is evident from the 16S rRNA gene tree that some of the organisms have either identical or almost identical sequences; representatives of those with almost identical 16S rRNA gene sequences were examined in DNA:DNA relatedness studies using the thermal denaturation procedure of Gonzalez and Saiz-Jimenez (2005). The results obtained in comparisons between S. violaceusniger NRRL B-1865 and S. hygroscopicus subsp. geldanus NRRL 3602T and S. melanosporofaciens NRRL B-12234T were in good agreement with those reported by Labeda and Lyons (1991) who used a spectrophotometric procedure. Congruent results of a similar nature have been reported for Phyllobacterium strains (Jurado et al. 2005).

Streptomyces violaceusniger ISP 5563 and NRRL B-1476, its reported parent strain, gave identical 16S rRNA and partial rpoB gene sequences and an identical phenotypic profile. These results are in stark contrast to those reported by Labeda and Lyons (1991) who found that strains designated NRRL B-1476 and ISP 5563 shared a DNA: DNA relatedness value of only 46%. They also found that S. violaceusniger NRRL B-1476 and ISP 5563 shared DNA:DNA relatedness values of 39% and 97% with a strain designated S. violaceusniger NRRL B-1478 which they considered to be the type strain of the species. Labeda and Lyons (1991) attributed the apparently anomalous results between S. violaceusniger strains NRRL B-1476 and ISP 5563 to strains NRRL B-1476 and NRRL B-1478 having been switched when they were lyophilised on the same day in 1958. The ampoule submitted to E.B. Shirling, which was designated ISP 5563, was from this set of lyophilized cultures. However, this interpretation is not supported by the present data which show a commonality of properties between S. violaceusniger NRRL B-1476 and ISP 5563. These results are consistent with the original view that strain ISP 5563 is derived from strain NRRL B-1476. This means that the reference material sent directly from the International Streptomyces Project to the American Type Culture Collection, the Deutsche Sammlung von Mikroorganismen und Zellkulturen and the Japanese Collection of Microorganisms, and subsequently designated ATCC 27477, DSM 40563 and JCM 4092, respectively was derived from S. violaceusniger NRRL B-1476. It also means that S. violaceusniger NRRL B-1476 is the type strain of the species.

Streptomyces hygroscopicus strains NRRL B-1477, NRRL 2339 and NRRL 2387T have almost identical 16S rRNA gene sequences and similar phenotypic profiles, results in good agreement with DNA:DNA relatedness data (Labeda and Lyons 1991). These workers found that S. hygroscopicus strains NRRL B-1477 and NRRL 2339 shared DNA relatedness values of 83% and 92%, respectively with DNA from S. hygroscopicus NRRL 2387T. The minimum level of DNA relatedness between strains required to define a genomic species is recommended as 70% (Wayne et al. 1987) though extensive DNA:DNA studies of Streptomyces strains implied that genomic homologies greater than 80% may correspond to species level relatedness in this genus (Labeda 1993, 1998; Labeda and Lyons 1992). These data underpin the taxonomic integrity of S. hygroscopicus (Jensen 1931) Labeda and Lyons 1991 and provide further evidence that members of the S. violaceusniger clade can be assigned to species using a combination of 16S rRNA gene sequence and phenotypic data (Sembiring et al. 2000; Saintpierre et al. 2003).

The type strains of S. melanosporofaciens and S. sporoclivatus formed a distinct multimembered cluster in the numerical taxonomic study; S. hygroscopicus subsp. geldanus NRRL 3602T was loosely associated with this taxon. These organisms formed a subclade in the 16S rRNA gene tree, and grouped together in the rpoB tree together with the type strains of S. javensis and S. yatensis. DNA:DNA relatedness studies are needed to resolve relationships between such phylogenetically close streptomycetes, as exemplified by studies on members of the S. griseus 16S rRNA gene clade (Liu et al. 2005). The type strains of S. hygroscopicus subsp. geldanus and S. melanosporofaciens belong to different genomic species though they share the relatively high DNA:DNA relatedness value of 64% (Labeda and Lyons 1991); these organisms can also be distinguished from one another and from the type strain of S. javensis using a range of phenotypic characters (Table 4). In contrast, it is clear from the present study that S. melanosporofaciens NRRL 12234T and S. sporoclivatus NRRL B-24330T belong to the same genomic species; these organisms were grouped together using 16S-ITS RFLP fingerprint data (Lanoot et al. 2004). It is, therefore, proposed that S. sporoclivatus Preobrazhenskaya et al. 1986 should become a subjective synonym of S. melanosporofaciens Arcamone et al. 1957. It is evident from the genotypic and phenotypic data that S. hygroscopicus subsp. geldanus NRRL 3602T merits recognition as a new species.

The type strains of S. malaysiensis and S. yatensis were loosely associated with members of the S. melanosporofaciens 16S rRNA gene subclade and with one another in the numerical phenetic study, but were sharply separated in the rpoB gene tree. Streptomyces malaysiensis NRRL B-24313T had identical partial rpoB sequences with the type strains of S. asiaticus, S. indonesiensis and S. rhizosphaericus whereas S. yatensis NRRL B-24116T was most closely related to the type strain of S. melanosporofaciens. The two strains can be separated from one another and from representatives of other species classified in the S. violaceusniger 16S rRNA gene clade using a range of phenotypic properties (Table 4). The type strains of S. malaysiensis and S. melanosporofaciens have been distinguished using 16S-ITS RFLP fingerprint data (Lanoot et al. 2004). The results of the present study underpin the taxonomic integrity of S. malaysiensis and S. yatensis.

Another group of closely related organisms include the type strains of S. asiaticus, S. cangkringensis, S. indonesiensis and S. rhizosphaericus. These organisms shared 16S rRNA gene similarities within the range of 99.2–99.7%, values that correspond to between 4 and 6 nt differences at 1449 locations. Apart from S. cangkringensis NRRL B-24177T, these strains had identical partial rpoB sequences. However, DNA:DNA homology analyses performed by thermal denaturation revealed differences in melting temperature between representatives of these strains that were above the cut-off for the delineation of genomic species (ΔT m > 5°C; Wayne et al. 1987; Roselló-Mora and Amann 2001). The taxonomic integrity of these species is supported by the numerical taxonomic data as the type strains were recovered as distinct single membered clusters and can be separated using a combination of phenotypic properties.

The present genotypic and phenotypic data confirm and extend those of Labeda and Lyons (1991) in showing that strains identified as S. violaceusniger form a markedly heterogeneous group. Indeed, the recognition of S. violaceusniger NRRL B-1476T as the type strain of the species necessitates the reclassification of strains received as S. violaceusniger. Strain NRRL B-1478 shares its highest 16S rRNA gene similarities with S. hygroscopicus NRRL 2339, NRRL 2387T and NRRL B-1477, but has DNA:DNA similarities with the latter within the range 39–42% (Labeda and Lyons 1991). The sharp separation from the S. hygroscopicus strains is underpinned by the numerical taxonomic and the partial rpoB sequence data. It is evident, therefore, that S. violaceusniger NRRL B-1478 merits recognition as a new species within the S. violaceusniger 16S rRNA gene clade. The name proposed for this taxon is Streptomyces demainii sp. nov.

It is evident from 16S rRNA sequence data that S. violaceusniger NRRL B-1865 belongs to the S. rhizosphaericus subclade. It shares 16S rRNA gene similarities with the members of this taxon within the range 99.4–99.7, values that correspond to between 4 and 6 nt differences at 1449 locations. However, DNA:DNA homology analyses revealed ΔT m differences (5.6–6.9°C; which corresponds to approximately 56–62% DNA:DNA relatedness) against representatives of the S. rhizosphaericus subclade well above the species level cut-off point (ΔT m > 5°C). The organism was most closely associated with the type strain of S. hygroscopicus in the partial rpoB analysis, but it is known that these strains belong to different genomic species as they share a DNA:DNA similarity of only 48% (Labeda and Lyons 1991). These workers found that the strain NRRL B-1865 shared its highest DNA: DNA homology value, 78%, with S. violaceusniger NRRL B-1356, albeit below the 80% cut-off point mentioned previously; these strains also have 16S rRNA gene sequences which differ by 20 nt. The combined genotypic and phenotypic data are consistent with the classification of S. violaceusniger NRRL B-1865 as a separate species.

It is evident from the 16S rRNA gene sequence data that S. violaceusniger NRRL B-1356 is most closely related to S. violaceusniger NRRL B-1476T. The two strains share a 16S rRNA gene similarity of 99.7%, but belong to different, albeit closely related, species on the basis of DNA:DNA pairing data (Labeda and Lyons 1991). In addition, S. violaceusniger NRRL B-1356 was sharply distinguished from all of the other members of the S. violaceusniger 16S rRNA gene clade in the numerical taxonomic and partial rpoB analyses. These results are consistent with S. violaceusniger NRRL B-1356 being described as a new species.

The present investigation provides further evidence that strains classified in the S. violaceusniger 16S rRNA gene clade have a combination of colonial and morphological properties that distinguish them from all other streptomycetes. Members of the clade share high 16S rRNA gene similarities but can be classified into species using a combination of genotypic and phenotypic data. The present data support the taxonomic status of all but one of the validly described species assigned to the clade, the exception, S. sporoclivatus Preobrazhenskaya et al. 1986 should be seen as a subjective synonym of S. melanosporofaciens Arcamone et al. 1959. It is also clear that S. hygroscopicus subsp. geldanus NRRL 3602 and S. violaceusniger NRRL B-1356 and NRRL B-1865 should be recognized as new species. It is proposed that these organisms be recognized as S. geldanamycininus sp. nov., S. albiflaviniger sp. nov., and S. griseiniger sp. nov., respectively, names used, or in case of the former epithet in line with that used, in previous studies (Sembiring et al. 2000; Ward and Goodfellow 2004). Members of these and the established species classified in the S. violaceusniger 16S rRNA gene clade can be distinguished from one another using a range of genotypic and phenotypic data. Descriptions of the new species assigned to the S. violaceusniger 16S rRNA gene clade are given below.

Description of Streptomyces albiflaviniger sp. nov.

Streptomyces albiflaviniger (al. bi. fla.vi. ni’ger. L. adj. albus, white; L. adj. flavus, yellow; L. adj. niger, black., N.L. adj. albiflaviniger, white, yellow and black colors).

The description is based on data taken from this and from a previous study (Labeda and Lyons 1991). Spore chains in Spirales; the spore surface is rugose. On oatmeal agar the aerial spore mass color is white becoming black and moist when mature; the reverse side of colonial growth is yellow. Brown, orange and yellow diffusible pigments are formed, but not melanin pigments. Additional morphological and pigmentation properties are shown in Table 3. The phenotypic properties of the organism are cited in the text and in Table 4. The G + C content of the DNA of the type strain is 70.5 mol% (as determined by the T m method). The type strain is NRRL B-1356T (=DSM 41598T).

Description of Streptomyces demainii sp. nov.

Streptomyces demainii (de. mai. ni’ i. N.L. gen. n. demainii, of Demain, named in honour of Arnold Demain, a celebrated actinomycete biologist).

The description is based on data taken from this and from a previous study (Labeda and Lyons 1991). Spore chains are Spirales: the spore surface is rugose. On oatmeal agar the aerial spore mass color is gray, becoming black and moist when mature; the reverse side of colony growth is grayish-yellow. Melanin pigments are not formed. Additional morphological and pigmentation properties are shown in Table 3. The phenotypic properties of the organism are cited in the text and in Table 4. The G + C content of the DNA is 71.2 mol%, as determined by the T m method. The type strain is NRRL B-1478T ( = DSM 41600T).

Description of Streptomyces geldanamycininus sp. nov.

Streptomyces geldanamycininus (gel. da. na. my. ci. ni’nus N.L. neut.n. geldanamycinum, geldanamycin; L. suff. -inus, adjectival suffix used with the sense of belonging to or related to: N.L. masc. adj. geldanamycininus, related to geldanomycin, producing the antibiotic geldanamycin).

The description is based on data taken from this and from a previous study (Labeda and Lyons 1991). Spore chains are Spirales; the spore surface is rugose. On oatmeal agar the aerial spore mass color is grayish-brown and the substrate mycelium grayish-yellow. Melanin pigments are not produced. Additional morphological and pigmentation properties are shown in Table 3. The phenotypic properties of the organism are cited in the text and in Table 4. The G + C content of the DNA is 70.2 mol %, as determined by the T m method. The type strain is NRRL 3602T (=DSM 41894T).

Description of Streptomyces griseiniger sp. nov.

Streptomyces griseiniger (gri’se.i.ni’ger. N.L. adj. griseus, gray; L. adj. niger, black; N.L. adj. griseiniger, gray-black).

The description is based on data taken from this and from a previous study (Labeda and Lyons 1991). Spore chains are Spirales; the spore surface is rugose. On oatmeal agar the aerial spore mass color is gray, becoming black and moist when mature; the reverse side of colonial growth is grayish-yellow. Melanin pigments are not formed. Additional morphological and pigmentation properties are shown in Table 3. The phenotypic properties of the organism are cited in the text and in Table 4. The G + C content of the DNA of the type strain is 70.2 mol%, as determined by the T m method. The type strain is NRRL B-1865T (=DSM 41895T).

The five strains that fell outside the S. violaceusniger 16S rRNA gene clade formed distinct phyletic lines within the evolutionary radiation occupied by the genus Streptomyces. These organisms were also outside the zone of evolutionary radiation occupied by members of S. violaceusniger 16S rRNA tree in the rpoB tree, apart from the S. sparsogenes strain which was loosely associated with the S. melanosporofaciens rpoB subclade. The type strain of S. sparsogenes was assigned to the S. violaceusniger cluster by Williams et al. (1983), who recorded that it produced rugose ornamented spores. In contrast, Kämpfer et al. (1991) found that the organism was physiologically different from authentic S. hygroscopicus and S. violaceusniger strains. In the present study, S. sparsogenes NRRL 2940T was shown to produce spiral chains of ridged ornamented spores and formed a single membered cluster outside the S. violaceusniger aggregate cluster that was circumscribed in the numerical taxonomic study. These results are in good agreement with those of Labeda and Lyons (1991) who found low levels of DNA:DNA relatedness between S. sparsogenes NRRL 2940T and strains now known to belong to the S. violaceusniger 16S rRNA gene clade. The S. sparsogenes strain is most closely, albeit quite distantly, related to S. fulvissimus NBRC 3717T; these organisms share a 16S rRNA gene similarity of 96.5%, a value that corresponds to 35 nt differences at 1443 locations. They can also be distinguished using morphological and pigmentation properties as the S. fulvissimus strain forms straight to flexuous chains of smooth surfaced spores and melanin pigments on peptone-yeast extract-iron agar (Shirling and Gottlieb 1969). It is timely to emend the description of S. sparsogenes in light of these results.

Emended description of Streptomyces sparsogenes

Streptomyces sparsogenes Owen et al. 1963, 772AL (spar. so. gen’nes. L. part. adj. sparsus, scattered; Gr. v. genmao, to produce; N.L. part. adj. sparsogenes (sic), scattered producing, probably referring to the sparse formation of aerial mycelium).

The description is based on data taken from this and from previous studies (Shirling and Gottlieb 1969; Labeda and Lyons 1991). Spore chains are Spirales and spore surfaces ridged. Aerial hyphae when present may emerge from coremia-like structures. On oatmeal agar the aerial spore mass color is gray and the reverse side of colonial growth grayish-yellow. Melanin pigments are not formed. Additional morphological and pigmentational properties are shown in Table 3. Aesculin is hydrolyzed, but not urea. Nitrate is reduced, but hydrogen sulphide is not produced. Adenine, arbutin, chitin, Tween 80 and xylan are degraded, but not guanine, hypoxanthine, tyrosine, uric acid or xanthine. Grows at pH 6.0 and pH 8.0 and at 25°C and 40°C, but not at 10°C. l-arabinose, d- and l-arabitol, d-maltose, d-mannose, d-raffinose and d-sucrose are used as sole carbon sources, but not butane- 1, 4 diol, d-cellobiose, citric acid, dextrin, d-fructose, l-fucose, d-galactose, d-glucose, meso-inositol, α-lactose, d-mannose, d-melezitose, methanol, propanol, pyruvic acid, l-rhamnose, d-ribose or salicin. α-Alanine, l-histidine, l-proline, l-serine and l-threonine are used as sole carbon and nitrogen sources, but not l-alanine, l-aminobutyric acid, l-arginine, l-glutamic acid, l-glycine, l-leucine, dl-methionine, dl-norleucine, l-phenylalanine, l-tryptophane or l-valine. Grows in the presence of 13%, w/v NaCl. Resistant (μg ml−1) to carbenicillin (32, 64), cephalosporin (32), cefoxitin (32, 64), cephaloridine (32, 64), chlortetracycline hydrochloride (4, 8), doxycycline hydrochloride (4, 8), tetracycline hydrochloride (64), rifampicin (32, 64), fusidic acid (8, 16), lincomycin hydrochloride (32, 64) and novobiocin (4), but is sensitive to erythromycin (32, 64), oleandomycin phosphate (32, 64) and novobiocin (8). Additional phenotypic properties are cited in the text. The G + C content of the DNA is 71.9 mol%, as determined by the T m method. The type strain is NRRL 2940T (=DSM 40356T).

The four remaining strains received as S. violaceusniger are clearly misclassified as they do not form rugose ornamented spores, and fell outside the S. violaceusniger aggregate group defined in the initial numerical taxonomic study. These findings are in excellent agreement with DNA:DNA relatedness data which show that these organisms bear little relationship either to one another or to streptomycetes which form rugose spores (Labeda and Lyons 1991); these results are underpinned by the 16S rRNA gene sequence data and in the main by those from the partial rpoB studies.

Streptomyces violaceusniger ISP 5182 is most closely related to S. bobili JCM 4624T and S. galilaeus JCM 4757T; it shares 16S rRNA gene similarities of 99.42% and 99.45% with the latter, values which correspond to 9 nt differences at 1445 and 1446 locations, respectively. The S. violaceusniger and S. bobili strains can be can be readily distinguished as the former produces a gray aerial spore mass and a yellow-brown reverse colonial color on oatmeal agar, flexuous spore chains, but does not form melanin pigments whereas the latter may or may not form a sparse white aerial mycelium, but it does produce a grayish-yellow reverse color and melanin pigments (Williams et al. 1989). The S. galilaeus strain forms a brownish yellow to brownish red reverse color on oatmeal agar, but unlike the S. violaceusniger strain strain, it bears spores in spiral chains (Gauze et al. 1983).

Streptomyces violaceusniger NRRL 2834 is most closely related to S. murinus NBRC 14802T and S. graminearus NBRC 15420T; it shares 16S rRNA gene similarities of 98.7%and 98.8% with these organisms, values which correspond to to 18 nt differences at 1487 and 1444 locations, respectively. The S. violaceusniger and S. murinus strains can be distinguished readily as the former produces a white aerial mycelium and a yellow-reverse colonial color on oatmeal agar and flexuous spore chains whereas the latter is characterized by a gray or red aerial spore mass, grayish-yellow substrate mycelial pigments and spiral spore chains (Shirling and Gottlieb 1968). It can also be distinguished readily from the S. graminearus strain as the latter produces a gray aerial spore mass and a colorless to light lemon-yellow undersurface on oatmeal agar and spiral spore chains (Gauze et al. 1983).

Streptomyces violaceusniger NRRL B-5799 is most closely related to S. misionensis NBRC 13063T; these organisms share a 16S rRNA gene similarity of 99.1%, which corresponds to 13 nt differences at 1443 locations. The two strains have many physiological properties in common, but were assigned to different subclusters within the S. albidoflavus/S. griseus/S. antibioticus cluster in the numerical taxonomic study of Kämpfer et al. (1991). They can be distinguished using colonial and morphological properties as the S. violaceusniger strain forms a brown aerial spore mass, a yellow reverse colonial color and a yellow diffusible pigment on oatmeal agar and flexuous chains of smooth ornamented spores whereas the S. misionensis strain produces a gray-brown or light grayish red aerial spore mass and olive-brown or yellow-brown reverse colors and spiral chains of smooth or slightly warty spores (Shirling and Gottlieb 1969).

The remaining strain, S. violaceusniger NRRL 8097, is most closely related to the type strain of S. roseoviolaceus. These organisms share a 16S rRNA gene similarity of 98.8%, which is equivalent to 18nt differences at 1448 positions. They can also be distinguished by their colonial and morphological properties as the S. violaceusniger strain produces a gray aerial spore mass, grayish-orange reverse colonial pigments and an orange diffusible pigment on oatmeal agar, forms spiral chains of smooth ornamented spores, but does not produce melanin pigments. In contrast, S. roseoviolaceus ISP 5277T forms a red aerial spore mass, a red to blue substrate mycelium and a violet–blue diffusible pigment, melanin pigments and spiral chains of spiny spores (Shirling and Gottlieb 1969).

It is evident that the four misclassified S. violaceusniger strains can be distinguished from one another using a combination of genotypic and phenotypic data thereby underpinning the DNA:DNA relatedness data of Labeda and Lyons (1991). In addition, they can be distinguished from their closest phylogenetic neighbors using predictive colonial and morphological characteristics. It is evident that these organisms should be recognized as novel Streptomyces species. It is proposed that S. violaceusniger NRRL 8097, NRRL 2834, NRRL B-5799 and ISP 5182 be classified as S. auratus sp. nov., S. phaeogriseichromatogenes sp. nov., S. phaeoluteichromatogenes sp. nov. and S. phaeoluteigriseus sp. nov., the latter three names were used information in earlier studies (Sembiring et al. 2000; Ward and Goodfellow 2004).

Description of Streptomyces auratus sp. nov.

Streptomyces auratus (au.ra’ tus. L. masc. adj. auratus, gold colored).

The description is based on data taken from this and from an earlier study (Labeda and Lyons 1991). Spore chains are Spirales and the spore surface is smooth. On oatmeal agar the aerial spore mass color is gray and the substrate mycelium grayish-yellow; an orange diffusible pigment is produced, but not melanin pigments. Additional cultural and pigmentation properties are given in Table 3. The organism produces hydrogen sulphide, degrades adenine, arbutin, chitin, hypoxanthine, Tweens 40, 60 and 80, uric acid and xanthine, but not casein, guanine, tyrosine or xylan. It does not reduce nitrate or hydrolyze aesculin or urea. It grows from pH 5 to pH 10, at 25°C and 37°C, but not at 10°C or 40°C. Butan-1, 4 diol, d-cellobiose, citric acid, dextrin, d-fructose, l-fucose, d-galactose, d-glucose, meso-inositol, d-maltose, d-mannitol, d-mannose, d-melezitose, methanol, propanol, pyruvic acid, d-raffinose, l-rhamnose, d-ribose, l-salicin and d-sucrose are used as sole carbon sources, but not α-lactose. α- and l-alanine, l-aminobutyric acid, l-glutamic acid, l-glycine, l-histidine, l-proline, l-serine and l-threonine are used as sole carbon and nitrogen sources, but not aspartic acid, l-leucine, dl-methionine, dl-norleucine, l-phenylalanine, l-tryptophane or l-valine. Growth occurs in the presence of 13%, w/v NaCl. Resistant (μg ml-1) to carbenicillin (32, 64), cefoxitin (32), cephaloridine (32), chlortetracycline hydrochloride (4, 8), doxycycline hydrochloride (4), rifampicin (32, 64) and novobiocin (4, 8), but is sensitive to cefoxitin (64), cephaloridine (64), doxycycline hydrochloride (8), fusidic acid (8, 16), lincomycin hydrochloride (32, 64) and oleandomycin phosphate (32, 64). Additional phenotypic properties are cited in the text. The G + C content of the DNA is 65.6 mol %, as determined by the T m method. The type strain is NRRL 8097T (=DSM 41897T).

Description of Streptomyces phaeogriseichromatogenes sp. nov.

Streptomyces phaeogriseichromatogenes (phae.o.gri.se’i.chro.ma’to.ge.nes. Gr.adj. phaeus, brown; N.L. adj. griseus, gray; Gr. n. chroma-atos, color; Gr. v. genmao, to produce; N.L. part. adj. phaeogriseichromatogenes, producing brown and gray colors).

The description is based on data taken from this and from an earlier study (Labeda and Lyons 1991). Spore chains are Rectiflexibiles; the spore surface is ridged. Aerial spore mass colors range from grayish-white to grayish-brown and the substrate mycelium is yellow to yellowish-brown, a yellow diffusible pigment is formed on yeast extract-malt extract agar, but melanin pigments are not produced. Additional cultural and pigmentation properties are shown in Table 3. Hydrogen sulphide is produced, aesculin is hydrolyzed, but not urea. Nitrate is not reduced. Casein, chitin, hypoxanthine, tyrosine, Tween 60, uric acid and xanthine are degraded, but not adenine, arbutin, guanine or xylan. Grows from pH 5 to pH 10, at 25 and 37°C, but not at 10°C or 40°C. l-arabinose, d-arabitol, butan-1, 4 diol, d-cellobiose, citric acid, dextrin, d-fructose, d-galactose, d-glucose, α-lactose, d-maltose, d-mannitol, d-mannose, d-ribose and d-sucrose are used as sole carbon sources, but not l-arabitol, l-fucose, meso-inositol, d-melezitose, methanol, propanol, pyruvic acid, d-raffinose, l-rhamnose or salicin. l-alanine, l-arginine, l-glutamic acid, l-histidine, l-leucine, dl-methionine, dl-norleucine, l-proline and l-valine are used as sole carbon and nitrogen sources, but not α-alanine, l-aminobutyric acid, l-glycine, l-phenylalanine, l-threonine or l-tryptophane. Does not grow in the presence of 13%, w/v NaCl. Resistant (μg ml−1) to cephalosporin (32), cefoxitin (32), cephaloridine (33, 64), doxycycline hydrochloride (4) and rifampicin (64), but not to cefoxitin (64), carbenicillin (32, 64), erythromycin (32, 64), oleandomycin phosphate (32, 64), chlortetracycline hydrochloride (4, 8), doxycycline hydrochloride (8), rifampicin (32, 64), fusidic acid (8, 16), lincomycin hydrochloride (32, 64) or novobiocin (4, 8). Additional phenotypic properties are cited in the text. The G + C content of the DNA is 71.2 mol %, as determined by the T m method. The type strain is NRRL 2834T (=DSM 40710T).

Description of Streptomyces phaeoluteichromatogenes sp. nov.

Streptomyces phaeoluteichromatogenes (phae.o.lu.te.i.chro.ma.to.ge.nes. Gr. adj. phaeos, dark, brown; L. adj. luteus, yellow; Gr. n. chroma-atos, color; Gr. v. genmao, to produce, N.L. part. adj. phaeoluteichromatogenes, producing brown and yellow colors).

The description is based on data taken from this and from an earlier study (Labeda and Lyons 1991). Spore chains are Rectiflexibiles; the spore surface is smooth. On oatmeal agar the aerial spore mass color is brown and the substrate mycelium yellow; a yellow diffusible pigment is produced. Melanin pigments are not formed. Additional cultural and pigmentation properties are shown in Table 3. Aesculin and urea are hydrolyzed. Does not produce hydrogen sulphide or reduce nitrate. Adenine, casein, chitin, hypoxanthine, tyrosine and Tween 80 are degraded, but not casein, guanine, uric acid, xanthine or xylan. Grows from pH 4 to pH 9 and at 25°C and 40°C, but not at 10°C. Uses l-arabinose, d- and l-arabitol, butan-1, 4 diol, d-cellobiose, citric acid, d-fructose, d-galactose, d-glucose, meso-inositol, d-mannitol, d-mannose, d-melezitose, methanol, propanol and pyruvic acid as sole carbon sources, but not dextrin, l-fucose, α-lactose, d-maltose, d-raffinose, l-rhamnose, d-ribose, d-salicin or d-sucrose. α-Alanine, l-glutamic acid, l-glycine, l-leucine, l-proline, l-serine and l-threonine are used as sole carbon and nitrogen sources, but not l-alanine, l-aminobutyric acid, l-histidine, dl-methionine, dl-norleucine, l-phenylalanine, l-tryptophane or l-valine. Growth occurs in the presence of 13%, w/v NaCl. Resistant (μg ml−1) to carbenicillin (32, 64), cephalosporin (32), cefoxitin (32, 64), cephaloridine (32, 64), chlortetracycline hydrochloride (4,8), doxycycline hydrochloride (4, 8), tetracycline hydrochloride (64), rifampicin (32, 64) and novobiocin (4), but is sensitive to erythromycin (32, 64), oleandomycin phosphate (32, 64), fusidic acid (8, 16) and lincomycin hydrochloride (32, 64). Additional phenotypic properties are cited in the text. The G + C content of the DNA is 69.8 mol %, as determined by the T m method. The type strain is NRRL B-5799T (=DSM 41898T).

Description of Streptomyces phaeoluteigriseus sp. nov.

Streptomyces phaeoluteigriseus (phae.o.lu.te.i.gri. se’ us. Gr. adj. phaeos, dark, brown; L. adj. luteus, yellow; N.L. adj. griseus, gray. N.L. adj. phaeoluteigriseus, brown, gray and yellow-colored).

The description is based on data taken from this and from an earlier study (Labeda and Lyons 1991). Spore chains are Rectiflexibiles; the spore surface is smooth. On oatmeal agar the aerial spore mass color is gray and the substrate mycelium yellowish-brown; a yellow diffusible pigment is produced, as are melanin pigments. Additional cultural and morphological properties are given in Table 3. Aesculin is hydrolyzed but not urea. Hydrogen sulphide is produced, but nitrate is not reduced. Adenine, arbutin, chitin, hypoxanthine, Tween 80 and xanthine are degraded, but not arbutin, guanine, tyrosine, uric acid or xylan. Grows from pH 4 to pH 10, and at 25°C and 37°C, but not at 10°C or 40°C. Adonitol, l-arabitol, citric acid and dextrin are used as sole carbon sources, but not l-arabitol, butan-1, 4 diol, d-cellobiose, d-fructose, l-fucose, d-galactose, d-glucose, meso-inositol, α-lactose, d-mannitol, d-mannose, d-melezitose, methanol, propanol, pyruvic acid, d-raffinose, l-rhamnose, d-ribose, salicin or d-sucrose. α-Alanine, l-histidine and l-threonine are used as sole carbon and nitrogen sources, but not l-alanine, l-aminobutyric acid, l-arginine, l-glutamic acid, l-leucine, dl-methionine, dl-norleucine, l-phenylalanine, l-serine, l-threonine, l-tryptophane or l-valine. Growth occurs in the presence of 13%, w/v NaCl. Resistant (μg ml−1) to carbenicillin (32, 64), cefoxitin (32, 64), cephaloridine (32), oleandomycin phosphate (32), chlortetracycline hydrochloride (4), tetracycline hydrochloride (64), fusidic acid (8, 16) and novobiocin (4), but is sensitive to cephalosporin (32), cephaloridine (64), erythromycin (32, 64), oleandomycin phosphate (64), chlortetracycline hydrochloride (8), doxycycline hydrochloride (4, 8), rifampicin (32, 64), lincomycin hydrochloride (32, 64) and novobiocin (8). Additional phenotypic properties are cited in the text. The G + C content of the DNA is 72.2 mol%, as described by the T m method. The type strain is ISP 5182T (=DSM 41896T).