In our screening program for anthracycline producers from soil, a Streptomyces sp. was isolated from soil. The isolate was identified as Streptomyces purpurascens by 16S rRNA gene sequencing and deposited in Microbial Type Culture Collection (MTCC), Chandigarh, India (http://mtcc.imtech.res.in/) with the accession number MTCC 8547. The isolate was found to produce rhodomycins which are reported for S. purpurascens. Apart from this, some rhodomycin analogues which were not reported for S. purpurascens before were also obtained. In the present study at least one new rhodomycin analogue, was obtained from this isolate.
The isolate when grown in liquid culture produced red colored pigments extracellularly. Preliminary testing using the crude cell extract of the isolate showed it to be antibacterial against Gram positive bacteria. However, the crude extracts were not found to be active against Gram negative bacteria nor were they antifungal.
Studies on optimization of the growth medium indicated Starch as the best carbon source for growth as well as antibiotic production whereas yeast extract and peptone served as the best nitrogen source for antibiotic production and for growth respectively (For more details refer to Additional files). Therefore for all the further antibiotic production procedures the Starch-Yeast extract medium was preferred. Solvent extraction of the spent broth and subsequent purification using preparative TLC helped in obtaining the purified compounds. Seven purified compounds thus obtained were named A, B, C, D, E, F and G.
The UV visible spectra of all the fractions showed a peak in the 230-290 nm region indicating the presence of an aromatic ring in the structure. Also the FT-IR spectra of the compounds indicated the presence of hydroxyl groups, ketonic groups and hydrogen- bonded carbonyl groups which are characteristic of anthracyclines. Hence it was speculated that the isolate was an anthracycline producer. Identification of our isolate as S. purpurascens further strengthened the supposition since it is well known to produce anthracyclines belonging to the rhodomycin group (Brockmann & Bauer 1950). On acid hydrolysis each compound yielded an aglycone and deoxy- sugars. The identity of the aglycone and sugars (by TLC) confirmed that these compounds might be rhodomycin and/or its anaologues (Johdo et al. 1991a).
Compound A appeared very similar to α2-Rhodomycin II and compounds B and E to that of Rhodomycins, which perhaps could be Rhodomycin A or B depending upon whether there are two rhodosamine moieties or one. All these compounds have been previously reported for S. purpurascens by Brockman (1955) and Shoji (1968; Arcamone 1984).
From our study, compound C appeared to be similar to an ε-RMN glycoside, Epelmycin and compound F resembled to an γ-isoRMN glycoside, Obelmycin. To the best of our knowledge, these compounds have not yet been reported for S. purpurascens. They have, however been reported for mutants of S. violaceus (Johdo et al. 1991c), which is very closely related and a phylogenetic neighbor of S. purpurascens as is evident from the Figure 1.The biosynthetic pathway for rhodomycin synthesis has been elucidated by Oki (1977), according to which during the synthesis of rhodomycin (a β-RMN glycoside), ε-RMN is an intermediate. This ε-RMN can be glycosylated by a glycosyl transferase to form ε-RMN glycosides, epelmycins. It has been suggested that biosynthetic conversion from ε-RMN to β-RMN takes place at glycoside level (Johdo et al. 1991b; Johdo et al. 1991c). Therefore formation of Epelmycins which are ε-RMN glycosides can be expected to occur in S. pupurascens too. Similarly, Compound D identified as an aklavinone glycoside, Aclacinomycin A which is the main product in S. galilaeus has also not been reported for S. purpurascens. Aklavinone is the first aglycone during the synthesis of rhodomycin (Connors et al. 1990). Aklavinone can be glycosylated by a glycosyl transferase to form Aclacinomycin (Oki et al. 1979). Hence the possibility of obtaining aclacinomycins in the fermentation broth of S. purpurascens cannot be denied.
Compound G was identified as Rhodosaminyl-2-Deoxyfucosyl-α2 Rhodomycinone. α2-RMN glycosides have been designated as Alldimycins by Johdo (Johdo et al. 1991d). All Alldimycins have been reported to contain only one sugar, Rhodosamine. Our studies consistently showed the presence of 2-deoxy fucose in the chromatogram of compound G. This leads us to believe that compound G may be a new rhodomycin analogue which has not been previously reported.
Antibacterial studies showed that compound E which appeared similar to Rhodomycin B was the most potent compound. Its MIC was found to be 2 μg/ml which was comparable to the reported value of 0.5-1 μg/ml (Shoji et al. 1968). However, the IC50 value, around 15 μM (~8 μg/ml) indicating much less toxicity than the reported value of 1 μg/ml (Saito et al. 1995).
Compound A, identified as α2-Rhodomycin II and compound F, identified as Obelmycin were found to be less potent. The MIC of both A and F was found to be > 20 μg/ml and IC50 was found to be around 15 μM (~8 μg/ml), which is much higher than the reported values. Compound F appeared similar to Obelmycins which are γ-isoRMN glycosides. These are reported to have low bioactivity because the sugar rhodosamine is bound at C-10 of the carbon skeleton (Johdo et al. 1991e). According to Vanek et al.(1977), the position of sugar moiety on the carbon skeleton is important. Anthracycline glycosides having C-7 sugar are more active than those having it at C-10.
Thus this study resulted in the isolation of a rhodomycin producer from soil and its identification as Streptomyces purpurascens. The isolate was found to produce one novel Rhodomycin analogue and two known analogues being reported however, first time from S. purpurascens.