Song et al. (2021) diversified fusidane-type antibiotics — fungal triterpenoids including fusidic acid, the only clinical antibiotic targeting bacterial elongation factor G — by randomly combining post-tailoring genes from the helvolic acid, fusidic acid and cephalosporin P1 pathways in Aspergillus oryzae: 24 of 27 designed gene combinations were productive, yielding 58 compounds (54 new) and an 80-member library, yet no derivative exceeded fusidic acid against Staphylococcus aureus. The study left four interlinked issues open: (i) the structural and mechanistic determinants of the observed substrate-dependent shifts in the regioselectivity of the P450 enzyme CepB4 (7α/7β hydroxylation) and the specificity of the acetyltransferase CepD2 (6α-OH vs 7β-OH acetylation); (ii) why three combinations failed and many titers were low, and how expression and yields of the engineered transformants might be improved; (iii) whether backbone-diversified fusidanes can yield antibacterial leads superior to fusidic acid or effective against alternative pathogens; (iv) whether the stochastic combinational strategy generalizes beyond related pathways sharing a common intermediate. No later work has resolved any of these strands.

Background

Fusidane-type antibiotics are a small family of fungi-derived triterpenoid antimicrobials — represented by helvolic acid, fusidic acid and cephalosporin P1 — built on the protostadienol framework (a tetracyclic sterol-like skeleton with a 4α-methyl group, 16β-acetoxy group and 20-carboxylic acid). Fusidic acid is the only clinically used antibiotic that inhibits bacterial elongation factor G (EF-G, the translocase of the ribosome during protein synthesis), which gives this class little cross-resistance with other antibiotics. Because resistance to front-line drugs is spreading, ways to diversify the fusidane backbone and improve on fusidic acid are of renewed interest. The complete biosynthetic gene clusters for all three representative compounds have been characterized: the hel (helvolic acid), fus (fusidic acid) and cep (cephalosporin P1) clusters share six conserved genes that build a common intermediate, 16β-acetyloxy-29-norprotosta-17(20)Z,24-dien-3-one-21-oic acid, after which pathway-specific tailoring enzymes — cytochrome P450 monooxygenases (heme-thiolate enzymes that insert oxygen into unactivated C–H bonds), short-chain dehydrogenases/reductases, and acetyltransferases — diverge to the individual end products. Song et al. (2021) exploited this architecture with a 'stochastic combinational strategy': in an Aspergillus oryzae NSAR1 host carrying the six conserved genes, they introduced all possible random combinations of the post-tailoring genes from the three clusters (27 combinations in total). 24 combinations produced expected products; 58 compounds were isolated (54 new), which together with 22 previously known analogues formed an 80-member library for structure–activity analysis, quantified as MIC values (minimum inhibitory concentration, the lowest antibiotic concentration preventing visible bacterial growth) against S. aureus 209P and Micrococcus luteus. Two unanticipated enzymatic behaviors emerged. The multifunctional P450 CepB4, which natively catalyzes dual 6α- and 7β-hydroxylation in cephalosporin P1 biosynthesis, additionally gave both 7α- and 7β-monohydroxylated epimers, alongside its original hydroxylation products, when the substrate carried a C1/C2 double bond and an 11α-hydroxyl; the authors inferred — but did not establish — that the ferryl-oxo reactive intermediate sits at equal distance from the 7α- and 7β-hydrogens. The acetyltransferase CepD2, natively site-selective for 6α-OH acetylation, acetylated 7β-OH on substrates processed by the enzyme HelE; the authors proposed, without structural or mechanistic evidence, that the 3-keto group and/or C1/C2 double bond enables this expanded activity. No library member was more potent than fusidic acid (MIC 0.125 µg/mL against S. aureus 209P), and three designed combinations produced no target product. The strategy itself was justified by enzyme compatibility within related pathways sharing a common intermediate; its behavior on distantly related clusters or other natural-product classes was never tested.

Problem Statement

The common scientific objective is to convert the stochastic combinational strategy of Song et al. (2021) from a demonstrated one-off diversification concept into a mechanistically understood, practically productive, and general platform for fusidane-antibiotic discovery. Four interlinked questions, all left open by the source and unresolved by later literature, constitute the problem. (1) Specificity-shift mechanism: identify, at the structural and mechanistic level, the determinants that allow the multifunctional P450 CepB4 — natively a 6α/7β dual hydroxylase on the cephalosporin P1 pathway — to additionally produce both 7α- and 7β-monohydroxylated epimers, alongside its original hydroxylation products, when the substrate carries the C1/C2 double bond and the 11α-hydroxyl, and that allow the acetyltransferase CepD2 — natively site-selective for 6α-OH — to acetylate 7β-OH on HelE-processed substrates; the explanations offered by the source — altered substrate positioning in the active site, a ferryl-oxo intermediate inferred to sit at equal distance from the 7α- and 7β-hydrogens, and expansion of CepD2 activity enabled by the 3-keto group and/or the C1/C2 double bond — remain untested hypotheses, and other mechanistic accounts are not excluded. (2) Production: explain why 3 of the 27 designed post-tailoring gene combinations failed to give target products and why many productive combinations give only low titers, and improve expression, yields and productivities of the engineered transformants to support downstream chemical modification and activity screening. (3) Lead translation: determine whether backbone-diversified fusidanes — the 80-member library or its further chemical modifications — can be translated into antibacterial potency superior to fusidic acid, or into efficacy against alternative pathogens, given that no library member exceeded fusidic acid (MIC 0.125 µg/mL) against S. aureus 209P. (4) Strategy scope: test the generalizability of the stochastic combinational strategy — random combination of post-tailoring genes — to natural-product classes beyond related pathways sharing a common intermediate, and characterize the limits of enzyme compatibility across more distantly related biosynthetic clusters.

References

  1. Song X, Lv J, Cao Z, Huang H, Chen G, Awakawa T, Hu D, Gao H, Abe I, Yao X. Extensive expansion of the chemical diversity of fusidane-type antibiotics using a stochastic combinational strategy. Acta Pharmaceutica Sinica B 11(6):1676-1685, 2021. DOI: 10.1016/j.apsb.2020.12.007. https://doi.org/10.1016/j.apsb.2020.12.007
  2. Cao Z-Q, Lv J-M, Liu Q, Qin S-Y, Chen G-D, Dai P, Zhong Y, Gao H, Yao X-S, Hu D. Biosynthetic Study of Cephalosporin P1 Reveals a Multifunctional P450 Enzyme and a Site-Selective Acetyltransferase. ACS Chemical Biology 15(1):44-51, 2020. DOI: 10.1021/acschembio.9b00863. https://doi.org/10.1021/acschembio.9b00863
  3. Cao Z, Li S, Lv J, Gao H, Chen G, Awakawa T, Abe I, Yao X, Hu D. Biosynthesis of clinically used antibiotic fusidic acid and identification of two short-chain dehydrogenase/reductases with converse stereoselectivity. Acta Pharmaceutica Sinica B 9(2):433-442, 2019. DOI: 10.1016/j.apsb.2018.10.007. https://doi.org/10.1016/j.apsb.2018.10.007
  4. Lv J-M, Hu D, Gao H, Kushiro T, Awakawa T, Chen G-D, Wang C-X, Abe I, Yao X-S. Biosynthesis of helvolic acid and identification of an unusual C-4-demethylation process distinct from sterol biosynthesis. Nature Communications 8:1644, 2017. DOI: 10.1038/s41467-017-01813-9. https://doi.org/10.1038/s41467-017-01813-9
  5. Skellam E, Rajendran S, Li L. Combinatorial biosynthesis for the engineering of novel natural products. Communications Chemistry 7:89, 2024. DOI: 10.1038/s42004-024-01172-9. https://doi.org/10.1038/s42004-024-01172-9
  6. Kang Y-S, Silva SC, Smith K, Sumida K, Wang Y, Chiaraviglio L, Donthiri RR, Aljahdali AZ, Kirby JE, O'Doherty GA. Exploration of the Fusidic Acid Structure Activity Space for Antibiotic Activity. Molecules 30(3):465, 2025. DOI: 10.3390/molecules30030465. https://doi.org/10.3390/molecules30030465
  7. Honma S, Kimishima A, Matsui H, Nishitomi A, Negami S, Kamo T, Ujie Y, Wasuwanich P, Honsho M, Kojima H, Tokiwa T, Sugawara A, Asami Y. Identification and evaluation of helvolic acid as an antibacterial compound against Neisseria gonorrhoeae. Bioscience, Biotechnology, and Biochemistry 90(6):780-784, 2026. DOI: 10.1093/bbb/zbag043. https://doi.org/10.1093/bbb/zbag043
  8. Tian C, Wang K, Zhang X, Li G, Lou H-X. Old fusidane-type antibiotics for new challenges: Chemistry and biology. Chinese Journal of Natural Medicines 20(2):81-101, 2022. DOI: 10.1016/S1875-5364(21)60114-0. https://doi.org/10.1016/S1875-5364(21)60114-0
  9. Li S, Hao X, Li Y, Gan M. Emericellopsic acid, a helvolic acid derivative with a 6/5/7/5 tetracyclic skeleton from sponge-derived Emericellopsis maritima. The Journal of Antibiotics 78(10):580-585, 2025. DOI: 10.1038/s41429-025-00852-5. https://doi.org/10.1038/s41429-025-00852-5
  10. Cao Z-Q, Wang G-Q, Luo R, Gao Y-H, Lv J-M, Qin S-Y, Chen G-D, Awakawa T, Bao X-F, Mei Q-H, Yao X-S, Hu D, Abe I, Gao H. Biosynthesis of Enfumafungin-type Antibiotic Reveals an Unusual Enzymatic Fusion Pattern and Unprecedented C-C Bond Cleavage. Journal of the American Chemical Society 146(18):12723-12733, 2024. DOI: 10.1021/jacs.4c02415. https://doi.org/10.1021/jacs.4c02415
  11. Gao Y, Lv J, Zhong Y, Cao Z, Luo R, Qi Y, Wang G, Li S, Chen G, Hu D, Gao H, Yao X. Engineered Fusion Enzyme-Mediated Non-Consecutive Cyclization-Glycosylation Enables Heterologous Synthesis of Antifungal Enfumafungin. Advanced Science 12:e07531, 2025. DOI: 10.1002/advs.202507531. https://doi.org/10.1002/advs.202507531
  12. Li L, Ali T, Goralczyk J, Jayasundara S, Paul A, de Amorim MR, Beemelmanns C, Skellam E. Investigating the versatility of cytochalasan cytochrome P450 monooxygenases using combinatorial biosynthesis reveals stereochemical restrictions. Fungal Biology and Biotechnology 13:9, 2026. DOI: 10.1186/s40694-026-00214-2. https://doi.org/10.1186/s40694-026-00214-2
  13. Lin P, Tian C, Peng X-P, Lou H-X, Li G. Metabolic profiling for the discovery of two rare fusidane-type heterodimers from the fungal endophyte Acremonium pilosum F47. Steroids 192:109188, 2023. DOI: 10.1016/j.steroids.2023.109188. https://doi.org/10.1016/j.steroids.2023.109188
  14. Yuan W, Wang Y, Jing S, Wang J, Guo L, Gao W. Promiscuity of Cytochrome P450 Enzymes: Discovery Strategies, Mechanistic Insights, and Applications in Terpenoid Biosynthesis. Journal of Agricultural and Food Chemistry 74(29):22353-22376, 2026. DOI: 10.1021/acs.jafc.6c00134. https://doi.org/10.1021/acs.jafc.6c00134
  15. Luo P, Huang J-H, Lv J-M, Wang G-Q, Hu D, Gao H. Biosynthesis of fungal terpenoids. Natural Product Reports 41(5):748-783, 2024. DOI: 10.1039/d3np00052d. https://doi.org/10.1039/d3np00052d
  16. Tian C, Gao H, Peng X-P, Li G, Lou H-X. Fusidic acid derivatives from the endophytic fungus Acremonium pilosum F47. Journal of Asian Natural Products Research 23(12):1148-1155, 2021. DOI: 10.1080/10286020.2020.1866559. https://doi.org/10.1080/10286020.2020.1866559
  17. Chang Z, Li Y, Lu Y, Xiao H. Towards crucial post-modification in biosynthesis of terpenoids and steroids: C3 oxidase and acetyltransferase. Enzyme and Microbial Technology 162:110148, 2023. DOI: 10.1016/j.enzmictec.2022.110148. https://doi.org/10.1016/j.enzmictec.2022.110148