Natural compounds are a valuable source for therapeutically relevant natural compounds ¹. Between the years 1981 to 2019, around 33% of all FDA‑approved small molecule drugs were natural products or derivatives thereof. Their high medicinal relevance emerges from the molecular scaffolds that evolved during hundreds of millions of years to interact with biomacromolecules ². However, most of the natural compounds are derivatized for its use in therapy, as a result of tuning the natural compound in the binding to the biomacromolecule, in pharmacokinetic properties and in bioavailability ¹. Here, fluorine plays a pivotal role, because of its unique properties, such as small size and highest electronegativity ^3,4. 45% of small-molecule pharmaceuticals approved by the FDA in 2018, and 41% of all small-molecule drugs approved in 2019 contained at least one fluorine atom ⁵.

The polyketide natural products feature a wealth of medicinally important activities, including antibiotic, anticancer, antifungal, antiparasitic and immunosuppressive properties. The polyketide macrolide erythromycin was one of the first antibiotics introduced to the market. In addition to erythromycin, derivatives are now in use whose antibiotic properties are modulated and improved by chemical modifications; clarithromycin, roxithromycin and azithromycin ⁶. Seeking to improve its properties, macrolides have also been fluorinated. The fluorinated antibiotics flurithromycin and solithromycin have been produced by chemical fluorination of the erythromyicin with perchloryl fluoride ⁷ and SelectFluor ⁸, respectively. Solithromycin gained attention as fourth-generation macrolide antibiotic ⁹, and is currently evaluated for its use in therapeutic treatment (2021 Antibacterial agents in clinical and preclinical development: an overview and analysis, WHO).

Polyketides spurred the interest of researchers around the globe for many years, because their natural synthesis involves a highly modular, assembly-line like synthesis ^10,11. Small carbonic acid precursors, usually derivates of acetic acids, are stepwise assembled by the polyketide synthases (PKSs) to elaborate compounds of molecular masses of several hundred grams. A multitude of reports demonstrates the susceptibility of particularly multienzyme (type I)  PKSs for reprogramming to directly biosynthesize polyketide derivatives ¹².

The interest of my lab in engineering the multienzyme PKS complexes arose from an initial research focus on the fatty acid synthases (FASs), with several contributions to first yeast and then mammalian multienzyme FASs. FASs and PKSs are evolutionarily strongly related, and perform similar chemistry. Both syntheses are highly compartmentalized, due to the multienzyme scaffold and the acyl carrier protein (ACP) mediated substrates shuttling ^13,14. During the past years, we were focusing on the transacylation reaction catalyzed by the acyl transferases. The acyl transferase domains are the gatekeepers of the reaction compartments, selecting for the acyl moieties that are finally condensed and processed by the synthases. Clearly, they are of high interest for engineering, because the control of its function is crucial to achieve the production of new-to-nature compounds.

As one main result of many years of lab work, we were able describing in detail the differences in acyl transferases of PKSs and FASs. While the malonyl-acetyl-transferase (MAT) of mammalian FASs are extremely substrates tolerant, transferring native and non-native substrates with high rates, the acyl transferase domains of PKSs are generally restrictive in the choice of substrates. This finding triggered our interest to harness the MAT for producing new-to-nature compounds in PKSs. We envisioned that combining the substrate tolerant MAT of mammalian FAS with the natively rather tolerant KS from PKSs would suspend natural substrate selectivity of PKSs and allow loading and condensing of non-native extender substrates.

In the collaborative work with the laboratory of David Sherman (University Michigan), we applied this approach for incorporating fluorinated units into polyketides during their enzymatic synthesis. In doing so, we could produce macrolactones and macrolides that harbor a fluoro or fluoro-methyl motifs in a-position to the internal ester group (Figure). The fluoro-methylated compounds feature a configuration exactly as present in solithromycin. When the MAT replaces other acyl transferases of PKSs, various positions in the polyketide would be susceptible to derivatization.

Our chemoenzymatic approach for the introduction of fluorine units into a polyketide during its synthesis allows alternative access to the synthetic-chemical fluorination of a given polyketide. It is worthwhile, because the introduction of fluorine units that modifies the polyketide is achieved via simple polyketide building blocks and the regioselectivity in the introduction is high.

The presented data give also remarkable new insight into the condensation reaction of PKSs and FASs. We demonstrate that disubstituted malonyl-moieties can be condensed by KS domains, opening new avenues for directed polyketide synthesis. Disubstitutions would constrain conformational properties in polyketides, which is in general attractive for drug design. 

Finally, I should note that our approach is not restricted to macrolides, but is broadly applicable to polyketide biosynthesis. It is able to deliver fluorinated variants of therapeutically relevant polyketides in sufficient amounts for activity screening. Thus, our approach opens up the prospect to bring polyketides to applications that have not yet been exhausted in their therapeutic potential.

Figure. Design of a hybrid protein by attaching the substrate loading acyl transferase domain from mouse fatty acid synthase (FAS) to the catalytic core of a bacterial polyketide synthase (PKS). We demonstrate that the hybrid is able to synthesize fluorinated macrolactones (and macrolides, not shown in the figure for clarity) from small fluorinated substrates.

1. Newman, D. J. & Cragg, G. M. Natural Products as Sources of New Drugs over the Nearly Four Decades from 01/1981 to 09/2019. J. Nat. Prod. 83, 770–803 (2020).
2. Breinbauer, R., Vetter, I. R. & Waldmann, H. From protein domains to drug candidates-natural products as guiding principles in the design and synthesis of compound libraries. Angew. Chem. Int. Ed Engl. 41, 2879–2890 (2002).
3. Hagmann, W. K. The many roles for fluorine in medicinal chemistry. J. Med. Chem. 51, 4359–4369 (2008).
4. Purser, S., Moore, P. R., Swallow, S. & Gouverneur, V. Fluorine in medicinal chemistry. Chem Soc Rev 37, 320–330 (2008).
5. Inoue, M., Sumii, Y. & Shibata, N. Contribution of Organofluorine Compounds to Pharmaceuticals. ACS Omega 5, 10633–10640 (2020).
6. Fernandes, P., Martens, E. & Pereira, D. Nature nurtures the design of new semi-synthetic macrolide antibiotics. J. Antibiot. (Tokyo) 70, 527–533 (2017).
7. Toscano, Luciano. Process for the preparation of (8s)-8-fluoroerythromycins. (1987).
8. Zhu, Y. et al. Modern Approaches for Asymmetric Construction of Carbon–Fluorine Quaternary Stereogenic Centers: Synthetic Challenges and Pharmaceutical Needs. Chem. Rev. 118, 3887–3964 (2018).
9. Fernandes, P., Martens, E., Bertrand, D. & Pereira, D. The solithromycin journey-It is all in the chemistry. Bioorg. Med. Chem. 24, 6420–6428 (2016).
10. Katz, L. Manipulation of Modular Polyketide Synthases. Chem. Rev. 97, 2557–2576 (1997).
11. Khosla, C. Harnessing the Biosynthetic Potential of Modular Polyketide Synthases. Chem Rev 97, 2577–2590 (1997).
12. Klaus, M. & Grininger, M. Engineering strategies for rational polyketide synthase design. Nat Prod Rep 35, 1070–1081 (2018).
13. Maier, T., Leibundgut, M., Boehringer, D. & Ban, N. Structure and function of eukaryotic fatty acid synthases. Q Rev Biophys 43, 373–422 (2010).
14. Grininger, M. Perspectives on the evolution, assembly and conformational dynamics of fatty acid synthase type I (FAS I) systems. Curr Opin Struct Biol 25, 49–56 (2014).