The isolation of organisms from the natural environment synthesizing commercially useful metabolites is an expensive and laborious process. Therefore, other means of producing novel compounds which may be of some industrial significance have been attempted. Probably the most successful alternative approach has been the semi-synthetic one where microbial products have been chemically modified, for example, the semisynthetic penicillins. Precursor feeding has also met with some success in this context; by incorporating a precursor of a natural product into the fermentation medium the level of the end product may be increased, for example, the use of phenylacetic acid in the production of penicillin G. The feeding of an analogue of the normal precursor of a natural product frequently results in the production of an analogue of the natural product. Hamill et al. (1970) demonstrated that if 5-, 6-or 7-substituted tryptophan replaced tryptophan, the normal precursor of the antifungal agent, pyrrolnitrin, in cultures of Pseudomonas aureofaciens, then a series of substituted pyrrolnitrins were obtained, some of which had improved antifungal activity. However, the disadvantages of the analogue-precursor technique are that the end product tends to be very similar to the natural product and the new product will be contaminated with the normal product which the organism may still synthesize from its self-produced natural precursors. Birch (1963) suggested that the problem of mixed products may be overcome by isolating mutants which would not produce the normal precursor but could convert it into the end product. Thus, the analogue precursor could be converted into the novel end product without competition from the normal endogenous precursor. Shier et al. (1973) succeeded in applying Birch's idea in the study of neomycin production by Strep-
si ivces fradiae from the precursor, deoxystreptamine. llT'iising a replica plating technique, the survivors of a nutation treatment were screened for the ability to ' ilubit the growth of the test organism, Bacillus subtilis, onlv in the presence of deoxystreptamine. By feeding the mutant isolated by this method, different antibiotics were synthesized.
Nagaoka and Demain (1975) described this technique for the isolation of new products as 'mutational biosynthesis' and the mutants isolated as 'idiotrophs'. Ik-ides the aminoglycoside-aminocyclitol antibiotics (of which neomycin is an example), mutational biosynthesis has been applied to the macrolide antibiotics, the novobiocin antibiotics and the /3-lactam antibiotics (Daum and Lemk, 1979).
Ihe advent of readily available recombination techniques resulted in attempts to produce novel compounds from recombinants, particularly streptomycetes. The rationale behind these experiments was that by mixing the genotypes of two organisms synthesizing different metabolites then new combinations of biosyn-thetic genes, and hence pathways, may be produced. Little progress was achieved using protoplast fusion but the exploitation of recombinant DNA technology has yielded some significant successes.
Streptomyces coelicolor produces the polyketide acti-norhodin (Fig. 3.34) whilst Streptomyces sp. AM 7161 produces medermycin. Hopwood et al. (1985b) transformed some of the cloned genes coding for acti-norhodin into Streptomyces sp. AM7161. The recombinant produced another antibiotic, mederrhodin A (Fig.
3.34). The modified strain contained the actW gene from S. coelicolor coding for the /;-hydroxylation of actinorhodin; the enzyme was also capable of hydroxy-lating medermycin. Hopwood et al. (1985b) also introduced the entire actinorhodin gene cluster into Streptomyces violaceoruber which produces granaticin or di-hydrogranaticin (Fig. 3.35). The recombinant synthesized the novel antibiotic, dihydrogranatirhodin which has the same structure as dihydrogranaticin apart from the stereochemistry at one of its chiral centres (Fig.
The enzymes responsible for polyketide biosynthesis have been extensively studied in recent years and significant advances have been made in the understanding of both their biochemistry and genetics. Polyketide synthases (PKSs) are multifunctional enzymes which catalyse repeated decarboxylative condensations between coenzyme A thioesters and are very similar to the fatty acid synthases. An enormous range of microbial polyketides are known and the variation is due to the chain length, the nature of the precursors and
Was this article helpful?