Balhimycin

Secondary metabolites such as antibiotics are typically produced by actinomycetes as a response to growth limiting stress conditions. Several studies have shown that secondary metabolite production is correlated with changes observed in actinomycete pellet morphology. Therefore, we investigated the correlation between the production of balhimycin and the spatio-temporal distribution of live and dead cells in pellets of Amycolatopsis balhimycina in submerged cultures. To this end, we used laser scanning confocal microscopy to analyze pellets from balhimycin producing and nonproducing media containing 0.2 and 1.0 g l(-1) of potassium di-hydrogen phosphate, respectively. We observed a substantially higher fraction of live cells in pellets from cultures yielding larger amounts of balhimycin. Moreover, in media that resulted in no balhimycin production, the pellets exhibit an initial death phase which commences from the centre of the pellet and extends in the radial direction. A second growth phase was observed in these pellets, where live mycelia are seen to appear in the dead core of the pellets. This secondary growth was absent in pellets from media producing higher amounts of balhimycin. These results suggest that distribution of live and dead cells and its correlation with antibiotic production in the non-sporulating A. balhimycina differs markedly than that observed in Streptomycetes.

Glycopeptides, such as vancomycin, are powerful antibiotics against methicillin-resistant Staphylococcus aureus. Balhimycin, a glycopeptide antibiotic isolated from Amycolatopsis balhimycina, is similarly effective as vancomycin. Side effects of vancomycin include triggering of platelet apoptosis, which is characterized by cell shrinkage and by cell membrane scrambling with phosphatidylserine exposure at the cell surface. Stimulation of apoptosis may involve increase of cytosolic Ca(2+) activity, ceramide formation, mitochondrial depolarization and/or caspase activation. An effect of balhimycin on apoptosis has, however, never been reported. The present study thus tested whether balhimycin triggers platelet apoptosis. Human blood platelets were treated with balhimycin and cell volume was estimated from forward scatter, phosphatidylserine exposure from annexin V-binding, cytosolic Ca(2+) activity from fluo-3AM fluorescence, ceramide formation utilizing antibodies, mitochondrial potential from DiOC6 fluorescence, and caspase-3 activity utilizing antibodies. As a result, a 30 min exposure to balhimycin significantly decreased cell volume (≥1 μg/ml), triggered annexin V binding (≥1 μg/ml), increased cytosolic Ca(2+) activity (≥1 μg/ml), stimulated ceramide formation (≥10 μg/ml), depolarized mitochondria (≥1 μg/ml) and activated caspase-3 (≥1 μg/ml). Cell membrane scrambling and caspase-3 activation were virtually abrogated by removal of extracellular Ca(2+). Cell membrane scrambling was not significantly blunted by pancaspase inhibition with zVAD-FMK (1μM). In conclusion, balhimycin triggers cell membrane scrambling of platelets, an effect dependent on Ca(2+), but not on activation of caspases.

Infection caused by methicillin-resistant Staphylococcus aureus (MRSA) is an increasing societal problem. Typically, glycopeptide antibiotics are used in the treatment of these infections. The most comprehensively studied glycopeptide antibiotic biosynthetic pathway is that of balhimycin biosynthesis in Amycolatopsis balhimycina. The balhimycin yield obtained by A. balhimycina is, however, low and there is therefore a need to improve balhimycin production. In this study, we performed genome sequencing, assembly and annotation analysis of A. balhimycina and further used these annotated data to reconstruct a genome-scale metabolic model for the organism. Here we generated an almost complete A. balhimycina genome sequence comprising 10,562,587 base pairs assembled into 2,153 contigs. The high GC-genome (~ 69%) includes 8,585 open reading frames (ORFs). We used our integrative toolbox called SEQTOR for functional annotation and then integrated annotated data with biochemical and physiological information available for this organism to reconstruct a genome-scale metabolic model of A. balhimycina. The resulting metabolic model contains 583 ORFs as protein encoding genes (7% of the predicted 8,585 ORFs), 407 EC numbers, 647 metabolites and 1,363 metabolic reactions. During the analysis of the metabolic model, linear, quadratic and evolutionary programming algorithms using flux balance analysis (FBA), minimization of metabolic adjustment (MOMA), and OptGene, respectively were applied as well as phenotypic behavior and improved balhimycin production were simulated. The A. balhimycina model shows a good agreement between in silico data and experimental data and also identifies key reactions associated with increased balhimycin production. The reconstruction of the genome-scale metabolic model of A. balhimycina serves as a basis for physiological characterization. The model allows a rational design of engineering strategies for increasing balhimycin production in A. balhimycina and glycopeptide production in general.