SummarySafracin is an antibiotic with anti-tumour activity produced by Pseudomonas fluorescens A2-2. The entire safracin synthetic gene cluster spanning 17.5 kb has been identified, cloned and sequenced. The safracin cluster comprises 10 open reading frames (ORFs) encoding proteins for three non-ribosomal peptide synthetases (NRPS), three safracin precursor biosynthetic enzymes, two safracin tailoring enzymes, a safracin resistance protein and a small hypothetical protein of unknown function. These genes are organized in two divergent operons of eight and two genes respectively. This pathway exhibits unusual features when compared with other NRPS systems. We have demonstrated by heterologous expression of the cluster that it is able to direct the synthesis of safracin in other strains. Cross-feeding experiments have confirmed that 3-hydroxy-5-methyl-O -methyltyrosine is the precursor of two amino acids of the molecule. Genetic analyses have allowed us to demonstrate that the bicistronic operon encodes the hydroxylation and N-methylation activities of the pathway. The cloning and expression of the safracin cluster has settled the basis for the in vivo and in vitro production of a wide variety of compounds, such as the promising ecteinascidins anti-cancer compounds.

The phenylacetic acid transport system (PATS) of Pseudomonas putida U was studied after this bacterium was cultured in a chemically defined medium containing phenylacetic acid (PA) as the sole carbon source.Kinetic measurement was carried out, in vivo, at 300C in 50 mM phosphate buffer (pH 7.0). Under these conditions, the uptake rate was linear for at least 3 min and the value ofKm was 13 ttM. The PATS is an active transport system that is strongly inhibited by 2,4-dinitrophenol, 4-nitrophenol (l100), KCN (97%), 2-nitrophenol (90%Ye), or NaN3 (80%) added at a 1 mM final concentration (each). Glucose or D-lactate (10 mM each) increases the PATS in starved cells (140%o), whereas arsenate (20 mM), NaF, orNN'-dicyclohexylcarbodiimide (1 mM) did not cause any effect. Furthermore, the PATS is insensitive to osmotic shock. These data strongly suggest that the energy for the PATS is derived only from an electron transport system which causes an energy-rich membrane state. The thiol-containing compounds mercaptoethanol, glutathione, and dithiothreitol have no significant effect on the PATS, whereas thiol-modifying reagents such as N-ethylmaleimide and iodoacetate strongly inhibit uptake (100 and 93%, respectively). Molecular analogs of PA with a substitution (i) on the ring or (ii) on the acetyl moiety or those containing (iii) a different ring but keeping the acetyl moiety constant inhibit uptake to different extents. None of the compounds tested significantly increase the PA uptake rate except adipic acid, which greatly stimulates it (163%). The PATS is induced by PA and also, gratuitously, by some phenyl derivatives containing an even number of carbon atoms on the aliphatic moiety (4-phenylbutyric, 6-phenylhexanoic, and 8-phenyloctanoic acids). However, similar compounds with an odd number of carbon atoms (benzoic, 3-phenylpropionic, 5-phenylvaleric, 7-phenylheptanoic, and 9-phenylnonanoic acids) as well as many other PA derivatives do not induce the system, suggesting that the true inducer molecule is phenylacetyl-coenzyme A (PA-CoA). Furthermore, after P. putida U is cultured in the same medium containing other carbon sources (glucose or octanoic, benzoic, or 4-hydroxyphenylacetic acid) in the place of PA, the PATS and PA-CoA are not detected; neither the PATS nor PA-CoA is found in cases in which mutants (PA-and PCL-) lacking the enzyme which catalyzed the initial step of the PA degradation (phenylacetyl-CoA ligase) are used. PA-CoA has been extracted from bacteria and identified as a true PA catabolite by high-performance liquid chromatography and also enzymatically with pure acyl-CoA:6-aminopenicillanic acid acyltransferase from Penicillium chrysogenum.

Pseudomonas putida U does not degrade D-glucose through the glycolytic pathway but requires (i) its oxidation to D-gluconic acid by a peripherally located constitutive glucose dehydrogenase (insensitive to osmotic shock), (ii) accumulation of D-gluconic acid in the extracellular medium, and (iii) the induction of a specific energy-dependent transport system responsible for the uptake of D-gluconic acid. This uptake system showed maximal rates of transport at 30 "C in 50 mM potassium phosphate buffer, pH 7.0. Under these conditions the Km calculated for D-gluconic acid was 6 7 pM. Furthermore, a different transport system, specific for the uptake of glucose, was also identified. It is active and shows maximal uptake rates at 35 "C in 50 mM potassium phosphate buffer, pH 60, with a Km value of 8 3 pM.

The polyketide pederin family are cytotoxic compounds isolated from insects, lichen, and marine sponges. During the past decade, different uncultivable bacteria symbionts have been proposed as the real producers of these compounds, such as those found in insects, lichen, and marine sponges, and their trans-AT polyketide synthase gene clusters have been identified. Herein we report the isolation and biological activities of a new analogue of the pederin family, compound 1, from the culture of a marine heterotrophic alphaproteobacterium, Labrenzia sp. PHM005. This is the first report of the production of a pederin-type compound by a free-living marine bacteria that could be cultured in the laboratory.

The complete genome of the strain Labrenzia sp. PHM005, a free-living producer of a pederin analog 18-O-demethyl pederin, hereinafter labrenzin, has been sequenced. This strain contains two replicons comprising a circular chromosome of 6,167,349 bp and a circular plasmid (named p1BIR) of 19,450 bp. A putative gene cluster responsible for the synthesis of labrenzin (lab cluster) has been identified showing that it encodes a trans-AT mixed type PKS/NRPS biosynthetic pathway that is responsible for the synthesis of pederin and possibly an onnamide analog. The putative boundaries of the lab gene cluster were determined by genetic comparisons with other related strains, suggesting that the cluster consists of a 79-kb region comprising 3 genes encoding multidomain hybrid polyketide synthase/non-ribosomal peptide synthetase (PKS/NRPS) proteins (PKS4, PKS/NRPS13, and PKS/NRPS15), and 16 auxiliary enzymes. Transcriptomic analyses suggest that all the genes of the cluster are expressed in our culture conditions (i.e., in minimal medium in the absence of any specific inducer) at detectable levels. We have developed genetic tools to facilitate the manipulation of this strain and the functional characterization of the cluster genes. We have created a site-directed mutant unable to produce pederin, demonstrating experimentally for the first time the role of the cluster in the synthesis of pederin. This work paves the way to unravel the clues of the biosynthesis of pederin family compounds and opens the door to modify and overproduce these anticancer drugs for industrial and pharmaceutical purposes.