The lincomycin (LM)-production gene cluster of the overproducing strain Streptomyces lincolnensis 78-11 was cloned, analysed by hybridization, as well as by DNA sequencing, and compared with the respective genome segments of other lincomycin producers. The lmb/lmr gene cluster is composed of 27 open reading frames with putative biosynthetic or regulatory functions (lmb genes) and three resistance (lmr) genes, two of which, lmrA and lmrC, flank the cluster. A very similar overall organization of the lmb/lmr cluster seems to be conserved in four other LM producers, although the clusters are embedded in non-homologous genomic surroundings. In the wild-type strain (S. lincolnensis NRRL2936), the lmb/lmr-cluster apparently is present only in single copy. However, in the industrial strain S. lincolnensis 78-11 the non-adjacent gene clusters for the production of LM and melanin (melC) both are duplicated on a large (0.45-0.5 Mb) fragment, accompanied by deletion events. This indicates that enhanced gene dosage is one of the factors for the overproduction of LM and demonstrates that large-scale genome rearrangements can be a result of classical strain improvement by mutagenesis. Only a minority of the putative Lmb proteins belong to known protein families. These include members of the gamma-glutamyl transferases (LmbA), amino acid acylases (LmbC), aromatic amino acid aminotransferases (LmbF), imidazoleglycerolphosphate dehydratases (LmbK), dTDP-glucose synthases (LmbO), dTDP-glucose 4,6-dehydratases (LmbM) and (NDP-) ketohexose (or ketocyclitol) aminotransferases (LmbS). In contrast to earlier proposals on the biosynthetic pathway of the C-8 sugar moiety (methylthiolincosaminide), this branch of the LM pathway actually seems to be based on nucleotide-activated sugars as precursors.

The alpha-glucosidase inhibitor acarbose, O-[4,6-dideoxy-4[1 s-(1,4,6/5)-4,5,6-trihydroxy-3-hydroxymethyl-2-cyclohexen-1-yl]-amino-alpha-D-glucopyranosyl]-(1-->4)- O-alpha-D-glucopyranosyl-(1-->4)-D-glucopyranose, is produced in large-scale fermentation by the use of strains derived from Actinoplanes sp. SE50. It has been used since 1990 in many countries in the therapy of diabetes type II, in order to enable patients to better control blood sugar contents while living with starch-containing diets. Thus, it is one of the latest successful products of bacterial secondary metabolism to be introduced into the pharmaceutical world market. Cultures of Actinoplanes sp. also produce various other acarbose-like components, of which component C is hard to separate during downstream processing, which is one of the most modern work-up processes developed to date. The physiology, genetics and enzymology of acarbose biosynthesis and metabolism in the producer have been studied to some extent, leading to the proposal of a new pathway and metabolic cycle, the "carbophore". These data could give clues for further biotechnological developments, such as the suppression of side-products, enzymological or biocombinatorial production of new metabolites and the engineering of production rates via genetic regulation in future.

The putative biosynthetic gene cluster for the ␣-glucosidase inhibitor acarbose was identified in the producer Actinoplanes sp. 50/110 by cloning a DNA segment containing the conserved gene for dTDP-D-glucose 4,6-dehydratase, acbB. The two flanking genes were acbA (dTDP-D-glucose synthase) and acbC, encoding a protein with significant similarity to 3-dehydroquinate synthases (AroB proteins). The acbC gene was overexpressed heterologously in Streptomyces lividans 66, and the product was shown to be a C 7 -cyclitol synthase using sedo-heptulose 7-phosphate, but not ido-heptulose 7-phosphate, as its substrate. The cyclization product, 2-epi-5-epi-valiolone ((2S,3S,4S,5R)-5-(hydroxymethyl)cy-

Three streptomycin (SM) production genes from Streptomyces griseus clustered around aphD, the major resistance gene, have been sequenced: strB, coding for an aminocyclitol amidinotransferase, ORF5 (strR), a putative regulatory gene, and ORF1 (strD), possibly coding for a hexose nucleotidylating enzyme. Three promoters and at least five, partially overlapping, transcripts have been identified by S1 mapping and Northern blot experiments. aphD, the resistance gene, is transcribed from two promoters. One of them, located inside the strR gene, seems to be constitutive and the other is switched on later in the growth phase. The late transcripts cover the resistance gene (aphD) and a regulatory gene (strR) which controls the expression of strB.

No abstract

Glycosylation represents an attractive target for protein engineering of novel antibiotics, because specific attachment of one or more deoxysugars is required for the bioactivity of many antibiotic and antitumour polyketides. However, proper assessment of the potential of these enzymes for such combinatorial biosynthesis requires both more precise information on the enzymology of the pathways and also improved Escherichia coli-actinomycete shuttle vectors. New replicative vectors have been constructed and used to express independently the dnmU gene of Streptomyces peucetius and the eryBVII gene of Saccharopolyspora erythraea in an eryBVII deletion mutant of Sac. erythraea. Production of erythromycin A was obtained in both cases, showing that both proteins serve analogous functions in the biosynthetic pathways to dTDP-L-daunosamine and dTDP-L-mycarose, respectively. Over-expression of both proteins was also obtained in S. lividans, paving the way for protein purification and in vitro monitoring of enzyme activity. In a further set of experiments, the putative desosaminyltransferase of Sac. erythraea, EryCIII, was expressed in the picromycin producer Streptomyces sp. 20032, which also synthesises dTDP-D-desosamine. The substrate 3-alpha-mycarosylerythronolide B used for hybrid biosynthesis was found to be glycosylated to produce erythromycin D only when recombinant EryCIII was present, directly confirming the enzymatic role of EryCIII. This convenient plasmid expression system can be readily adapted to study the directed evolution of recombinant glycosyltransferases.

The genes lmbA,B1,B2 in the lincomycin A production gene cluster of Streptomyces lincolnensis were shown to form a common transcription unit with the promoter located directly upstream of lmbA. The proteins LmbB1 (mol. mass, 18 kDa) and LmbB2 (mol. mass 34 kDa), when over-produced together in Escherichia coli, brought about enzyme activities for the specific conversion of both L-tyrosine and L-3,4-dihydroxyphenylalanine (L-DOPA) to a yellow-colored product. The LmbB1 protein alone catalyzed the conversion of L-DOPA, but not of L-tyrosine. The purified LmbB1 protein showed a Km for L-DOPA of 258.3 microM. The L-tyrosine converting activity could not been demonstrated in vitro. The preliminary interpretation of these data suggests that the protein LmbB1 is an L-DOPA extradiol-cleaving 2,3-dioxygenase and that the protein LmbB2, either alone or in accord with LmbB1, represents an L-tyrosine 3-hydroxylase. This sequence of putative oxidation reactions on L-tyrosine seems to represent a new pathway different from the ones catalyzed by mammalian L-tyrosine hydroxylases or the wide-spread tyrosinases. The protein LmbA seemed not to be involved in this process. The labile, yellow-colored product from L-DOPA could not be converted to a picolinic acid derivative [3-(2-carboxy-5-pyridyl)alanine] in the presence of ammonia. Therefore, it probably is not a derivative of a cis, cis-3-hydroxymuconic acid semialdehyde; instead, its speculative structure represents a heterocyclic precursor of the propylhygric acid moiety of lincomycin A.

We have previously demonstrated that the biosynthesis of the C 7 -cyclitol, called valienol (or valienamine), of the ␣-glucosidase inhibitor acarbose starts from the cyclization of sedo-heptulose 7-phosphate to 2-epi-5-epivaliolone (Stratmann, A., Mahmud, T., Lee, S., Distler, J., Thus, a phosphotransferase activity was identified modifying 2-epi-5-epi-valiolone by ATP-dependent phosphorylation. This activity could be attributed to the AcbM protein by verifying this activity in S. lividans strain TK64/pCW4123M, expressing His-tagged AcbM. The Histagged AcbM protein was purified and subsequently characterized as a 2-epi-5-epi-valiolone 7-kinase, presumably catalyzing the first enzyme reaction in the biosynthetic route, leading to an activated form of the intermediate 1-epi-valienol. The AcbK protein could not catalyze the same reaction nor convert any of the other C 7 -cyclitol monomers tested. The 2-epi-5-epi-valiolone 7-phosphate was further converted by the AcbO protein to another isomeric and phosphorylated intermediate, which was likely to be the 2-epimer 5-epi-valiolone 7-phosphate. The products of both enzyme reactions were characterized by mass spectrometric methods. The product of the AcbM-catalyzed reaction, 2-epi-5-epi-valiolone 7-phosphate, was purified on a preparative scale and identified by NMR spectroscopy. A biosynthetic pathway for the pseudodisaccharidic acarviosyl moiety of acarbose is proposed on the basis of these data.