Abstract

Streptomyces leeuwenhoekii C34T produces the polyketides chaxamycins A–D and chaxalactins A–C. The former metabolites exhibit antibiotic activity and antitumoral properties. In this study, we selected a subset of overexpression targets, predicted using the FSEOF algorithm, to increase chaxamycin production based on the iVR1007 genome-scale model (GEM). Eight genes were individually cloned under strong constitutive promoters and introduced into the genome of S. leeuwenhoekii C34T, including ugp, tkt, tal, and the acc complex, giving rise to strains S1–S8. The production of chaxamycins by these engineered strains was compared to the wild-type and to a strain overexpressing the pcc complex (S9) (not predicted as target by the model). Chaxamycins were detected in all the samples, with higher amounts observed in strains S1 through S8. In contrast, strain S9 did not show improved chaxamycin A–C production. The best results were obtained with strain tkt (S2), tal (S3), and mcm/mce (S8), which demonstrated 7.41-, 3.67-, and 3.33-fold increases in total chaxamycin production, respectively. These strains overexpress genes associated with AHBA synthesis, pentose phosphate pathway or (2S)-methyl-malonyl-CoA biosynthesis. Extracts from higher producers also showed higher bioactivity against Micrococcus luteus. Thus, the predictions made using the iVR1007 model, were successfully applied to increase chaxamycin production, further validating the utility of the FSEOF algorithm for predicting metabolic engineering targets.