A strain described as ‘Streptomyces levoris’ K-3056 (=ATCC 3351; Naumova et al. 1980) has been reported to contain a poly(glycerophosphate)-lipoteichoic acid (PGP-LTA; Potekhina et al. 1983). Actinomycetes typically synthesize lipoglycans rather than LTA (Sutcliffe 1994) and reports of LTA are rare, with members of the genera Agromyces and Thermobifida, the only other reported examples (Gnilozub et al. 1994; Rahman et al. 2009). We, therefore, wished to confirm the work of Potekhina et al. (1983) as this finding would be of considerable interest with regard to cell envelope biology in Streptomyces and could be of chemotaxonomic significance.
As the strain ‘S. levoris’ K-3056 was not available and ATCC 3351 is not currently listed in the American Type Culture Collection online catalogue, we obtained a corresponding strain, Streptomyces sp. DSM 40537 (=ATCC 3351 = ISP 5537). The strain Streptomyces sp. ATCC 3351 (=ISP 5537) was deposited in 1947 by Selman Waksman as Streptomyces albus, although it has also been considered to belong to ‘Streptomyces saprophyticus’ nom. dub. (Shirling and Gottlieb 1972). Streptomyces sp. DSM 40537 (=ATCC 3351 = ISP 5537) has never been fully characterized and little studied, although it has been shown to belong to the numerically circumscribed Streptomyces albidoflavus cluster in numerical phenotypic studies (Williams et al. 1983; Kämpfer et al. 1991). To extend these observations, we have also undertaken a phylogenetic analysis of Streptomyces sp. DSM 40537.
Freeze dried biomass of Streptomyces sp. DSM 40537 (obtained directly from Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Braunschweig, Germany) was rehydrated into yeast extract–malt extract (YEME) medium and inoculated into liquid YEME medium and onto ‘sporulating’ Mannitol Soya Agar plates (Hobbs et al. 1989; Kieser et al. 2000). Surprisingly, two phenotypically distinct but stable variants were recovered after these first passages. The isolate recovered on Mannitol Soya Agar plates, herein referred to as Streptomyces sp. DSM 40537.1, exhibited a colony morphology closely resembling that described by Shirling and Gottlieb (1972), i.e. aerial hyphae in the yellow colour series and no obvious release of pigment to the growth medium (see Fig. S1 in Supplementary material). In contrast, the culture originally subcultured into liquid YEME medium (herein referred to as Streptomyces sp. DSM 40537.2) yielded colonies with grey spores on Mannitol Soya Agar plates and released an orange pigment into the growth medium. This pigment release was even more pronounced on nonsporulating agar plates (see Fig. S2 in Supplementary material).
To resolve this unexpected variation, near complete 16S rRNA gene sequences were obtained for both Streptomyces sp. DSM 40537.1 and Streptomyces sp. DSM 40537.2 using previously described methods (Humphry et al. 2007). In total, 1404 and 1398 bp of good quality nucleotide sequence (forward and reverse strand) were obtained for strain DSM 40537.1 and strain DSM 40537.2, respectively. Significantly, the 16S rRNA gene sequences for each of the strains were 100% identical over 1404 nucleotides, strongly suggesting each to be derived the same species. The sequence obtained has been deposited in Genbank as accession number EU847429. The cellular fatty acid profiles for each strain were both near identical and consistent with the assignment of strain DSM 40537 to the genus Streptomyces, as the major fatty acids were iC14:0, iC15:0, aiC15:0, iC16:0, C16:0, iC17:0, aiC17:0 and an unknown tentatively identified as a branched chain C12:0 fatty acid (data not shown). The distinct phenotypic characteristics observed for strain DSM 40537.2, therefore, appear to have been appeared during the first subculturing from the DSMZ stock onto different media, possibly as a result of overgrowth of a developmental mutant in the broth culture. Streptomyces sp. are extensively studied because of their complex lifecycles and the ability of mutations to cause developmental phenotypes, including changes in antibiotic and/or pigment production (Kieser et al. 2000). The phenotype of Streptomyces sp. DSM 40537.2 was stable and we did not observe revertants even on other media (data not shown), suggesting this was not a switching event but a mutational event.
The 16S rRNA gene sequence obtained was analysed by an NCBI Blast search (http://www.ncbi.nlm.nih.gov/blast/Blast.cgi), which clearly showed that the organism belongs to the genus Streptomyces and that the closest matches to sequences from validly described species were from members of the Streptomyces albidoflavus cluster (e.g. S. albidoflavus NBRC 13010T, 1390/1400 nucleotide identity; 99·29%) and Streptomyces koyangensis (1390/1400 nucleotide identity; 99·29%). Among the closest matches to this sequence were the 16S rRNA gene sequences deposited as originating from S. albus sub sp. albus (NBRC 3711 = ATCC 3351 = DSM 40537; accession number AB184782) and ‘S. saprophyticus’ (NBRC 13440 = ATCC 3351 = DSM 40537; accession number AB184404), both of which exhibited 1395/1400 (99·64%) nucleotide identity with the sequence determined in the present study. These data confirm that both cultures recovered herein (Streptomyces sp. DSM 40537.1 and Streptomyces sp. DSM 40537.2) are bona fide representatives of Streptomyces sp. DSM 40537 and that the orange-pigmented phenotype is likely caused by the acquisition of a developmental/regulatory mutation in Streptomyces sp. DSM 40537.2.
To further explore the taxonomic position of this strain, the phylogenetic position of the 16S rRNA gene sequence, was compared with those of other type strains of Streptomyces obtained from the Ribosomal Database Project-II (http://rdp.cme.msu.edu/). The tree (see Fig. S3 in Supplementary material) shows the clustering of the Streptomyces sp. DSM 40537 strain with the type strains of S. albidoflavus, Streptomyces canescens, Streptomyces coelicolor, Streptomyces felleus, Streptomyces limosus, Streptomyces odorifer and Streptomyces sampsonii (i.e. the organism belongs to the S. albidoflavus 16S rRNA gene clade). The phylogenetic tree is, thus, consistent with previous numerical taxonomic analyses (Williams et al. 1983; Kämpfer et al. 1991) and also supports the previously reported relationship between this group and S. koyangensis (Lee et al. 2005). DNA–DNA hybridization studies are now needed to confirm the relationship of Streptomyces sp. DSM 40537 with the type strains within this cluster, in order to determine if this strain merits status as a separate species. Our recovery of a stable, pigment releasing variant of Streptomyces sp. DSM 40537 is notable, as it has been previously reported that production of diffusible pigments is unusual for members of the S. albidoflavus cluster (Williams et al. 1983).
To confirm whether Streptomyces sp. DSM 40537 produced PGP-LTA, as previously reported (Potekhina et al. 1983) for ‘S. levoris’ K-3056 (=ATCC 3351), we extracted and purified material from both isolates described here, using the standard methodology of hot phenol–water extraction and hydrophobic interaction chromatography for the recovery of PGP-LTA and/or lipoglycans (Sutcliffe 2000; Rahman et al. 2009). Western blot analysis using a monoclonal antibody to Staphylococcus aureus PGP-LTA (BSYX-A110, Pagibaximab; http://biosynexus.com/productsbsyx.html) gave a clear positive cross-reaction with both preparations (Fig. 1). The fatty acid profiles of these extracts were very similar to those of the whole cells (data not shown). These data confirm the findings of Potekhina et al. (1983) that Streptomyces sp. can produce PGP-LTA, although further chemical analyses suggested that the extracts may contain additional non-PGP-LTA components (data not shown). We have also recently obtained similar PGP-LTA containing extracts from ‘S. coelicolor’ M145 (data not shown) and characterized a PGP-LTA from the thermophilic actinomycete Thermobifida fusca (Rahman et al. 2009). These data and previously published studies (Potekhina et al. 1983; Gnilozub et al. 1994) thus confirm that actinomycetes are capable of synthesizing PGP-LTA. Interestingly, all actinomycetes shown to date as synthesizing PGP-LTA also contain cell wall teichoic acids, consistent with the apparent conserved distribution of these two polyanionic cell envelope components (Weidenmaier and Peschel 2008).
To conclude, these data confirm that Streptomyces sp. DSM 40537 (=ATCC 3351) belongs to the S. albidoflavus cluster within the genus Streptomyces and we present the novel observation that orange pigment producing derivatives of this strain can be obtained. This suggests that further investigation of Streptomyces sp. DSM 40537 as a source of novel metabolites is warranted. Most importantly, the ability of actinomycetes to synthesize PGP-LTA is re-emphasized. This challenges previous assumptions that actinomycetes typically produce lipoglycans (Sutcliffe 1994) and indicates that a broader survey of LTA distribution in actinomycetes is necessary. The apparent correlation between the presence of LTA and cell wall teichoic acids suggests a rational basis for selecting further actinomycetes for study for the presence of LTA, which may be of chemotaxonomic value.