Skip to main content
NCBI home page
JACS Au logoLink to JACS Au
. 2026 Feb 5;6(2):801–812. doi: 10.1021/jacsau.5c00949

Structural Diversification of 14-Membered Macrolides by Chemoenzymatic Synthesis

Brian J Curtis †, Hannah A Boesger †,‡, Jennifer J Schmidt †, Maria L Adrover-Castellano †, Areeba N Momin †, Carolyn A Glasser †, David H Sherman †,‡,□,¶,*
PMCID: PMC12933321  PMID: 41755848

Abstract

The pikromycin polyketide synthase (PKS) catalyzes formation of 12-membered macrolactone 10-deoxymethynolide, and 14-membered macrolactone narbonolide. Herein, we show the efficient diversification of novel 14-membered macrolactones and identification of 6-membered δ-lactones from a series of unnatural pentaketides using the PikAIII/PikAIV PKS in vitro system. New macrocycles were further elaborated by the addition of D-desosamine and late-stage C–H hydroxylation. Molecular dynamics (MD) simulations and density functional theory (DFT) calculations were conducted to probe the reactivity and selectivity of this terminal catalytic step on the assembled unnatural macrolides. This approach highlights the flexibility of the PikAIII/PikAIV bimodule system in processing non-native substrates and demonstrates the utility of sequential biocatalytic steps for the chemoenzymatic synthesis of complex antibiotic scaffolds.

Keywords: chemoenzymatic synthesis, C−H functionalization, density functional theory, macrocyclic scaffolds, molecular dynamics simulations, pikromycin pathway, polyketide synthase

graphic file with name au5c00949_0010.jpg

graphic file with name au5c00949_0008.jpg

Introduction

Macrolide antibiotics are assembled by multimodular proteins called polyketide synthases (PKSs) that are responsible for sequential extension by two carbon units to produce linear chain elongation intermediates. These mega-proteins consist of three main catalytic domains; the acyl transferase (AT) that selects the appropriate acyl-CoA extender unit derivative and transfers it to the acyl carrier protein (ACP). The PKS is then primed for the ketosynthase (KS) domain to accept the growing polyketide chain from the upstream module for decarboxylative Claisen condensation, which incorporates this new unit into the growing chain. Multiple optional domains produce an array of chemical diversity in these systems through sequential processing of the β-carbonyl group. , These include a ketoreductase (KR) that catalyzes the stereospecific reduction of the ketone, a dehydratase (DH) that generates either a cis or trans double bond, and an enoyl reductase (ER) that catalyzes a final reduction giving the saturated carbon chain. Once the catalytic cycle of the terminal module is completed, macrocycles are formed via a regio- and stereospecific cyclization by the terminal thioesterase (TE) domain. −

The pikromycin (Pik) biosynthetic pathway contains a unique PKS that is able to produce both a 12-membered macrolactone, 10-deoxymethynolide (10-DML, 2) and a 14-membered macrolactone, narbonolide (7) (Figure A). Both rings are further glycosylated and oxidized by post-PKS tailoring enzymes to produce macrolide antibiotics. Post-PKS tailoring of 12-membered 10-DML (2) produces methymycins (3–6) and tailoring of 14-membered narbonolide (7) affords pikromycin (8) and small amounts of neopikromycin (9) and novapikromycin (10), demonstrating the inherent flexibility and selectivity of this metabolic system. This feature makes it uniquely suited for in-depth mechanistic dissection studies, and as a platform to develop biocatalytic syntheses of natural and unnatural macrolides. Previous work demonstrated that the modular nature of PKS proteins can be manipulated to modify product structures. Despite some success, PKS engineering has been frequently met with significantly attenuated yields, shunt products, , or failure to produce the desired products altogether. , To overcome some of these challenges, recent engineering efforts have focused on using evolution-guided approaches to enable production and increase titers of non-native polyketides from module-exchanged assembly lines. − Additionally, scalable and modular synthetic methods have been employed to generate macrolides that are otherwise inaccessible through these biocatalytic approaches. ,

1.

1

Open in a new tab

(A) Chemoenzymatic synthesis of native 12- and 14-membered ring products (2–10) from the pikromycin pathway. (B) Chemoenzymatic production of unnatural 12-membered macrolactones from synthetic pentaketide analogs. (C) The current study focused on an expanded set of pentaketide analogs to generate structurally diverse 14-membered macrolides.

We have developed a chemoenzymatic strategy to interrogate PKS terminal modules and tailoring enzymes using natural and unnatural late-stage polyketide intermediates. This approach enables a thorough analysis of structural variations and their functional impact on each step of the biosynthetic pathway, such as substrate selection, extension, β-processing, cyclization, and post PKS tailoring. ,− Utilizing chemoenzymatic methods to generate new macrolides is especially critical for achieving transformations that are synthetically challenging, including stereoselective chain elongation, terminal cyclization, and late-stage, selective C–H functionalization steps. Early in vitro work on the Pik biosynthetic system revealed that the terminal two modules (PikAIII/PikAIV) can be reconstituted using a synthetic pentaketide (1) to produce both the 12-membered aglycone, 10-DML (2), and the 14-membered aglycone, narbonolide (7). ,, The macrolactones are subsequently converted to their respective macrolides (3–6 and 8–10) using an engineered Streptomyces venezuelae biotransformation strain to append the desosamine sugar, and conduct P450 mediated C–H oxidations (Figure A). This work was further expanded by generating a series of unnatural chain elongation intermediates with altered side chain substituents and stereochemistry on the distal region of the pentaketides relative to the activated thioester. Although PikAIII-TE accepted each unnatural substrate, yield of 12-membered macrolactone products diminished with increasing changes to the native pentaketide molecule (Figure B). This investigation suggested that the PikAIII-TE module was able to process the unnatural substrates, with the TE limiting effective formation of macrocyclic products. Notably, a substrate bearing the unnatural C9–OH stereochemistry was elongated to the hexaketide (or C11–OH of the synthetic hexaketide) and exclusively off-loaded to hydrolyzed byproducts. The barriers to macrolactonization of this epimerized hexaketide were overcome through mutagenesis of the active site residue Ser148 to Cys148, resulting in production of the epimerized macrolactone in a yield comparable to that of the native functionality. Quantum mechanical (QM) studies revealed a lower-energy, concerted mechanism for this gain-of-function TE mutation, in contrast to the stepwise addition–elimination mechanism observed for the wild-type (WT) TE. Additionally, the Pik TE (S148C) mutation was recently found to markedly increase the macrolactonization yield of an amide-containing hexaketide compared to the WT Pik TE.

In this work, we report an expanded analysis of the terminal Pik PKS modules that probes substrate flexibility and selectivity by the PikAIII/PikAIV monomodules to catalyze 14-membered ring formation. A series of unnatural pentaketides bearing altered side chain substituents and heteroatom exchanges were tested as chain elongation intermediates. The unnatural macrocycles were then glycosylated by DesVII/DesVIII glycosyltransferase and analyzed for differential C–H oxidation via the late-stage PikC P450 monooxygenase (Figure C). Furthermore, we conducted molecular dynamics simulations and density functional theory calculations (DFT) to probe the basis for C–H oxidation selectivity and reactivity of the new macrolide structures.

Results and Discussion

Synthesis of Unnatural Pik Pentaketides 15–18

In an earlier study, we investigated the ability of a set of unnatural pentaketides bearing structural variations distal to the thiophenol ester linkage to be converted by PikAIII-TE to new 12-membered ring macrocycles. In the current study, we sought to expand this set of analogs and pursue the production of 14-membered macrolactones via PikAIII/PikAIV for subsequent conversion to their corresponding macrolides. In addition to previously synthesized pentaketides 11–14, we envisioned synthesizing analogs containing α/β unsaturated amides (15, 16) and an ester (17) in place of the natural α/β unsaturated ketone (Figure , Scheme A,B). Processing of these substrates by the Pik PKS proteins would result in hybrid macrolactone/lactam and macrodiolide products, which are intriguing as related analogs exhibit a variety of biological activities. − The ester and amide containing pentaketides were generated via a coupling strategy with TBS protected acid 21, which is produced via cross metathesis of 2-pentenoic acid (19) with 20. , Protecting group manipulation of either commercially available (S)-Roche ester or (S)-3-(Boc-amino)-2-methylpropionic acid afforded 22–24, which were subsequently coupled with 21 to furnish protected pentaketides 25–27. The methyl esters in 25 and 26 were hydrolyzed via LiOH/H2O, while selective hydrolysis of the methyl ester in 27 required use of Me3SnOH. Thiophenylation using Ph2S2/PBu3, followed by TBS deprotection using aq. HF furnished the desired ester and amide-containing pentaketides 15–17.

1. Synthesis of Pentaketide Analogs 15–18 .

1

Open in a new tab

a (A) Synthetic pentaketide substrates used in this study. (B) Synthesis of amide 15–16 and ester 17 containing pentaketides. (C) Synthesis of gem-dimethyl containing pentaketide 18.

Many therapeutics (e.g., anticancer, antibacterial, neuroprotective) contain gem-dimethyl moieties including the epothilones, which are derived from a mixed NRPS/PKS pathway. , Thus, compound 18 was generated containing an α,α gem-dimethyl moiety (Scheme A,C). We sought to determine if the KS of PikAIII can accept and elongate this analog, which has more steric bulk relative to the native Pik pentaketide. , For this analog, we pursued a cyclic anhydride ring opening strategy to produce precursor 30 that could be functionalized to the desired pentaketide 18 via cross metathesis with alkene 20. Ring opening of 2,2-dimethylglutaric anhydride (28) using N,O-dimethylhydroxylamine hydrochloride 29 in pyridine/DCM afforded the Weinreb amide intermediate, which was subsequently methylated and then reacted with vinyl MgBr2 to generate vinyl ketone 30 in three steps. Subsequent cross metathesis with 20 produced the protected pentaketide 31, which was functionalized in a similar manner to the amide and ester analogs to generate the desired gem-dimethyl pentaketide 18.

PikAIII-TE Selectively Catalyzes 12-Membered Ring Formation

With the new set of pentaketide analogues in hand, substrate loading, extension, and cyclization of these non-native chain elongation intermediates by PikAIII-TE (WT and S148C) was explored using previously established conditions for efficient substrate processing by the Pik modules. ,, Despite the success of the altered chain length substrates 11–14 with PikAIII-TE, only the N-methylated pentaketide 16 from this set of pentaketides 15–18 displayed evidence of a macrocyclization product as a major component from the reaction (Figures A, S1–S3). The N-methylated 12-membered ring (33) was isolated in 15% yield using the PikAIII-TE (S148C) mutant. Byproducts from this reaction include seco-acids resulting from hydrolysis of activated penta- and hexaketide intermediates and their corresponding NAC thioesters (Figure S2) derived from the (R)-methylmalonyl N-acetylcysteamine (MM-SNAC) extender unit. , The N–H amide and ester-containing pentaketides 15 and 17 appeared to be converted to extended, hydrolyzed hexaketides based on liquid chromatography–mass spectrometry (LC-MS) analysis (Figures S1 and S3), providing evidence for the KS-loading, chain extension and β-keto processing of these analogs by PikAIII and supporting the Pik module 5 substrate flexibility. Moreover, the KS appears to chain extend ester-containing pentaketide 17 more efficiently than both amide-containing pentaketides 15 and 16, as mainly the hexaketide seco-acid was observed from reaction of 17 and PikAIII-TE (WT), whereas an abundance of pentaketide seco-acids were observed relative to hexaketide seco-acids for reactions of both amide-containing substrates. The appended TE (WT or S148C), however, catalyzed the formation of only trace amounts of 12-membered macrolactone/lactam 32, and no macrodiolide 34 was detected. While pentaketides 15–17 were fully consumed by PikAIII, reactions with gem-dimethyl 18 displayed no turnover of starting material, evident from comparative analysis of the no enzyme control reaction to PikAIII-TE (Figure B). Contrary to the previous substrates 15–17, compound 18 contains additional steric bulk proximal to the site of KS-loading and extension, which is unable to be processed by the PikAIII-TE fusion protein. To further probe if this failure was due to KS active site loading or Claisen condensation, intact protein MS was employed to assess substrate loading to the KS5 Ser-active site of PikAIII-derived KS5-AT5 didomain. Ions corresponding to the mass of the acylated protein were detected for 18, indicating the unnatural substrate can be loaded onto the PikAIII KS5–Cys residue, but the Claisen condensation step likely fails to proceed based on the lack of observed chain-extension products (Figure S4).

2.

2

Open in a new tab

Evaluation of reactions of PikAIII-TE with heteroatom exchanged (A) and sterically bulky (B) pentaketides (15–18).

PikAIII/PikAIV Reactions Exhibit Greater Flexibility in 14-Membered Ring Formation

Next, we pursued reactions of pentaketide analogs 11–17 with the PikAIII/PikAIV tandem monomodule system, which catalyzes two extensions prior to cyclization by the TE to produce 14-membered macrolactones (Figure ). Initial analysis via LC-MS of the PikAIII/PikAIV-TE (WT) pair with substrates 11–14, that contain alkyl modifications distal to the thioester, demonstrated complete consumption of starting materials and putative 14-membered macrocyclic products were detected (Figures S5–S8). Notably, reactions of 11–14 with the corresponding PikAIII/PikAIV-TE (S148C) mutant produce the putative macrocyclic products, but in lower amounts relative to PikAIII/PikAIV-TE (WT) based on AUC values. Thus, scale-up reactions (up to 20 mg of substrate) were conducted using PikAIII/PikAIV-TE (WT) to characterize the presumed macrocyclic products. Macrolactones 35–38 were isolated in variable ratios of the C9 keto form and hemiketal form between the C5 hydroxyl and the C9 keto group or as a C2 epimerized product caused by tautomerization (Figure , Tables S3–S9). The 14-desmethyl narbonolide product 35 was isolated in 44% yield while the 12-desmethyl narbonolide (36) and 12,14-didesmethyl narbonolide (37) products were isolated in lower yields of 24% and 11%, respectively. Finally, the 13-desethyl narbonolide derivative 38 was isolated in a 20% yield. Attenuated yields for these transformations were a result of failed cyclization of intermediates following chain extension as well as substrate decomposition or transthioesterification via reactions with free NAC that is derived from the methyl malonyl-NAC extender unit. ,

3.

3

Open in a new tab

Evaluation of 14-membered macrolactone formation (with isolated yields provided) from pentaketides 11–18 and tandem PikAIII/PikAIV (WT) biocatalysis. Isolated yields for 35–39 and 41 were determined after reaction time = 2 h whereas the yield for 40 was determined after reaction time = 16.5 h.

For reactions of PikAIII/PikAIV with N–H amide (15), N-methylated amide (16), and ester (17) pentaketides, a similar product profile was observed compared to reactions with PikAIII-TE, which includes penta-, hexa-, and in this case, heptaketide shunt products (Figures S9–S11). In addition, trace amounts of putative decarboxylation products were assigned from reactions of amide pentaketides 15–16 based on observed m/z values and predicted molecular formulas, which were previously described for reactions of epimeric pentaketides with PikAIII-TE. Surprisingly, two major products were observed via LC-MS in a ∼ 1:1 ratio bearing the anticipated m/z values. Although both products resulted from double extension to the heptaketide, the terminal cyclization produced two distinct ring systems. The expected 14-membered macrolactones 39 (22%), 40 (13%), and 41 (21%), formed via intramolecular cyclization of the C13 hydroxyl group within the TE domain, were identified based on 2D NMR spectroscopic analysis and characteristic chemical shifts and multiplicities that are similar to other 14-membered macrolactones (Tables S10, S12, S15). The unexpected products, comprising 6-membered δ-lactones 43, 44, and 45, were characterized via 2D NMR as well as a diagnostic downfield shift of H-5 from ∼3.6 ppm for the 14-membered rings to ∼4.5–4.7 ppm for the 6-membered δ-lactones (Figure , Tables S11, S13–S14, S16–S17). We reason that the 6-membered δ-lactones are formed via the competing intramolecular cyclization of the C5 hydroxyl group (Figure ). Recognizing inefficiencies in the PikAIII KS based on the abundance of pentaketide seco-acid in PikAIII/PikAIV reactions (Figure S10) and to also produce more of the 14-membered ring 40 for tailoring reactions, the corresponding N-methylated hexaketide was reacted with PikAIV-TE (Figure S12). The major products were 14- and 6-membered rings 40 and 44 in 27% and 28% yields, respectively, along with 12-membered macrolactone 33, observed in 14% yield, resulting from KS6 skipping and direct loading onto the TE. We further note that reactions of N–H amide (15), N-methylated amide (16), and ester (17) pentaketides with the corresponding PikAIII/PikAIV-TE (S148C) mutant do not increase 14-membered ring formation. In addition, we evaluated if reactions of N–H pentaketide (15) with gain-of-function Pik TE variants from our previously reported directed evolution campaign increase formation of 14-membered ring (39). The PikAIII/PikAIV-TE (S148C, A217T, G222 V) and PikAIII/PikAIV-TE (S148C, A217T, G222 V, M271 V, Y25C, L126 V) variants instead produced mainly the corresponding 6-membered δ-lactone product (43, Figure S13).

4.

4

Open in a new tab

Pik TE catalyzes formation of 6-membered δ-lactone byproducts. (A) Proposed mechanism for the formation of 6-membered (43–45) vs 14-membered (39–41) macrocyclic products from pentaketide substrates 15–17. (B) Isomerization of narbonolide (7) to 6-membered δ-lactone 46 using Pik TE (WT). The isolated yield (12%) of 46 was determined after a 42 h reaction. (C) Extracted ion chromatograms (EICs, ESI+, m/z = 375.2142) for the reaction of narbonolide (7) with Pik TE (WT). Displayed traces include narbonolide (7) standard and 24 and 42 h reaction time points.

Further analysis of this unexpected product profile was conducted to address whether 6-membered δ-lactone formation occurs spontaneously or is TE-mediated. Incubation of the 6- (43–45) and 14-membered macrolactone ring products (39–41) in buffer alone, as well as with buffer and either PikAIV or standalone Pik TE were conducted, and the product profiles were analyzed after 2 and 24 h incubation periods (Figures S14–S16). The 14-membered macrolactone ring products (39–41) and 6-membered δ-lactones (43–45) were stable in buffer solution, though trace isomerization of 14-membered macrolactones to 6-membered δ-lactones was observed. However, the 14-membered rings (39 and 41) were completely transformed to the 6-membered δ-lactone counterparts (43 and 45) when incubated with PikAIV or standalone Pik TE for 24 h. In contrast, the N-methylated 14-membered ring (40) was found to only partially isomerize to its 6-membered δ-lactone counterpart (44) after 24 h. These data are consistent with the reactions of pentaketides 15–17 and PikAIII/PikAIV, where extended reaction times (monitored up to 24 h) afford only 6-membered δ-lactones for 15 and 17, while 2 and 24 h reactions for the N-methylated amide pentaketide (16) both afford roughly a 1:1 ratio of 14- to 6-membered rings. Conversion of the 14- to 6-membered ring species indicates that the TE Ser active site nucleophile catalyzes rebound macrolactone ring opening, which can again catalyze cyclization by either pathway A to the 14-membered rings 39–41 or by pathway B to the 6-membered δ-lactones 43–45 (Figure A). With extended reaction times, the equilibrium may shift to the thermodynamically favored 6-membered ring. Interestingly, the isomerization of these heteroatom-exchanged 14-membered rings to 6-membered rings was observed to a much more limited extent when challenged with the Pik TE Cys active site mutation. A hydrolytic form of this process has been reported in studies of the epothilone pathway, where epothilone TE catalyzes ring opening of epothilone C followed by hydrolysis from the active site Ser-ester intermediate to generate seco-epothilone C. In contrast to epothilone C, we observed only trace amounts of the putative heptaketide seco-acids via LC-MS following rebound macrolactone ring opening of 39–41, suggesting minimal competition with water in the acyl cavity, as seen in the Pik TE crystal structure containing covalently bound native heptaketide intermediate. In addition, incubation of the corresponding 12-membered macrolactone/lactams 32 and 33 with Pik TE also produces trace amounts of seco-acid products, suggesting that the 12-membered analogs are less prone to ring opening (Figures S17–S18).

Next, we reanalyzed the product profile resulting from reactions of native Pik pentaketide (1) and truncated analogs 11–14 with PikAIII/PikAIV to address if 6-membered δ-lactones are also observed. For all tested substrates, trace amounts of earlier eluting isomers with the same m/z as the 14-membered macrolactones were observed. To determine whether these are also 6-membered δ-lactones, a scale-up reaction of native 14-membered ring, narbonolide (7), with Pik TE was first performed (Figure B). Slow isomerization to an earlier eluting isomer, with the same retention time as that observed from reaction of native Pik pentaketide (1) with PikAIII/PikAIV, was observed over time, which was isolated and characterized as the 6-membered δ-lactone 46 in 14% yield following 42 h incubation (Figures C and S18, Table S18). These 6-membered δ-lactones were also observed in reactions using truncated analogs 11–14, which were proposed based on retention time, in-source fragmentation, and/or MS/MS data (Figures S20–S22) that are consistent with 6-membered δ-lactone 46 derived from narbonolide (7). Motivated by our findings that 6-membered δ-lactones are produced in biocatalytic reactions and that their formation is catalyzed by Pik TE, we next addressed if 6-membered δ-lactone 46 is observed in the macrolactone-producing strain S. venezuelae ATCC 15439 ΔdesI (DHS8708). Indeed, we observed trace amounts of 46 relative to narbonolide (7) via LC-MS in the supernatant following 48 h incubation (Figure S23).

Generation of New Macrolides via Glycosylation and Late-Stage C–H Oxidation

To generate the corresponding macrolide structures from the set of new macrolactones (35–41), we investigated the efficiency of the PikB-encoded glycosyl transferases (DesVII/DesVIII) to append the desosamine sugar (Figure A). We employed the engineered S. venezuelae strain DHS316, which retains the genes necessary for glycosylation (but lacks the PKS (pikAI–pikAIV) and the cytochrome P450 (pikC)) to append desosamine to our series of macrolactones. Macrolactones 35–41 along with the acetyl-narbonolide activator were added to actively growing cultures for a 12–24 h biotransformation prior to analytical assessment of product formation via LC-MS. Compounds 35–38 were completely consumed and their macrolide counterparts (47–50) were isolated in yields between 48–68% when conducted on a 10 mg scale (Tables S19–S22). For hybrid macrolactone/lactam (39) and macrodiolide (41) aglycones, neither biotransformation went to completion, though a significant fraction was converted to the anticipated masses of the glycosylated products. The glycosylated compounds corresponding to the hybrid macrolactone/lactam (51) and macrodiolide (53) were isolated in 38% yield and 37% yield, respectively (Tables S23–S24). Despite the overall lower isolated yields from these reactions, the aglycones are recovered in high purity and are recyclable. In contrast to the hybrid macrolactone/lactam (39) and macrodiolide (41) aglycones, nearly all N-methylated aglycone (40) was consumed, similar to the distal end modified aglycones 35–38, where the glycosylated product 52 was isolated in 69% yield (Table S25).

5.

5

Open in a new tab

Generation of macrolides from macrolactones. (A) Biocatalytic pipeline for tailoring of narbonolide analogs (35–41). (B) Macrolides (54, 56, and 57) were isolated and characterized following installation of desosamine and terminal C–H hydroxylation. Isolated yields for all characterized macrolides are shown in parentheses.

With new macrolides 47–53 in hand, we further explored late-stage C–H oxidation using the PikC monooxygenase (Figure B). This enzyme preferentially installs an allylic hydroxyl group at the C12 position of narbomycin (58) and has been studied extensively, leading to an engineered variant PikCD50N with improved turnover. − CYP450 PikC WT and its D50N variant selectively catalyzed C–H hydroxylation of C12 methyl group bearing 14-desmethyl (47) and 13-desethyl (50) narbomycin, both exhibiting consumption of starting material to oxidized products (Figures S24–S25). In contrast, 12-desmethyl (48) and 12,14-didesmethyl (49) narbomycin showed no evidence of hydroxylated products. The expected 12-hydroxy macrolide 54 was isolated in 71% yield (5.1% from (R)-Roche ester) from 14-desmethyl macrolide 47 (Table S26). Despite evidence of some turnover to an oxidized product by PikCD50N for 13-desethyl macrolide 49, low yield hindered isolation, which suggests a need for further PikC engineering to optimize catalytic efficiency toward this unnatural substrate. The heteroatom exchanged amide-containing macrolides 51 and 52 showed only minimal turnover of starting material, with only small amounts of oxidation of N-methylated macrolide 52 observed (Figure S26). Interestingly, oxidation of macrodiolide 53 resulted in partial conversion to two isomeric products that were isolated in near identical amounts (Figure S27, Tables S27–S28), 12-hydroxy macrodiolide 56 (12%, 0.14% from 19) and unexpected 14-hydroxy (neo) macrodiolide 57 (12%, 0.14% from 19). The observed 1:1 ratio of C12 to C14 oxidation of macrodiolide 53 contrasts sharply with the ∼ 40:1 ratio of pikromycin (8, C14) to neopikromycin (9, C12) observed in vivo following narbomycin (58) oxidation.

Molecular Dynamics Simulations Provide Rationale for Selectivity of Biocatalytic Oxidation

Intrigued by the altered oxidation selectivity and reactivity of narbomycin analogs 47–53 via PikCD50N, we next performed a series of molecular dynamics simulations to provide structural insight on these observations (Figure ). ,, In particular, we compared conformational dynamics of heteroatom exchanged substrates (51–53) with the native 14-membered narbomycin (58) and 12-membered YC-17 in the PikCD50N active site. , Molecular dynamics simulations of narbomycin (58) support the site selectivity of oxidation, displaying a preference for C–H hydroxylation at C12 over C14 (Figure A,D). In addition, simulations of macrodiolide 53 in the PikCD50N active site suggested a much greater structural preference for oxidation at the C14 position (Figure B,E), supporting the shift in hydroxylation selectivity that was experimentally observed for macrodiolide (53) relative to narbomycin (58). The simulation results for macrodiolide 53 were consistent with those observed for YC-17, which is hydroxylated in near equal amounts at positions C10 and C12, where MD trajectories also suggest preferential C–H oxidation poses for position C12 (position C14 in 14-membered analogs) over position C10 (Figure S28). ,, While these simulations focus on the substrate conformational dynamics in the active site, transition state energies of the rate-limiting hydrogen abstraction are an important aspect for substrate reactivity. In addition to conformational dynamics of the enzyme-ligand complex, density functional theory (DFT) calculations found that the energetic barrier to hydrogen abstraction at the model substrate C12 position is 2.5 kcal/mol lower than that of the C14 position, contributing to preferential C12 oxidation (Figure S29, Table S1). Therefore, relative sampling of structural conformations provides only a snapshot of PikCD50N reactivity, and the 1:1 ratios of C12:C14 oxidation of 53 and C10:12 oxidation of YC-17 can be explained based on both substrate reactivity and conformation in the active site. By scaling substrate near attack conformations (NACs) to relative rates of hydrogen abstraction, predicted ratios from 1 μs MD simulations were found to correlate to the experimental oxidation ratios of reactions catalyzed by PikCD50N (Figure S30, Table S2).

6.

6

Open in a new tab

Molecular dynamics simulations of 52, 53, and 58 in the PikCD50N active site. Highest-occupied pose of a MD trajectory of PikCD50N bound to narbomycin (58, A), 53 (B), and 52 (C). Representative frames of MD trajectories are shown as deviations from the idealized transition state for narbomycin (58, D), 53 (E), and 52 (F) from 200 ns simulations. Site selectivity was determined by relative orientation of the ligand for H-atom abstraction by the heme-iron oxo species (compound I). See Supporting Information (Figure S30) for the corresponding 1 μs simulations.

For N-methylated macrolide 52, molecular dynamics simulations were performed prior to the in vitro reactions. In this case, minimal oxidation was predicted to occur due to limited reactive poses in the PikCD50N active site. Our experimental observations aligned with the molecular dynamics data, with only trace amounts of oxidation product observed (Figure C,F). Interestingly, while N–H amide 51 consistently occupied a reactive pose during the MD trajectory, we did not observe oxidation via P450D50N. A conformational search of 51 and 52 showed that the N–H amide 51 exhibits intramolecular hydrogen bonding, leading to a significantly stabilized macrocyclic conformation (Figure S31). We hypothesize that this low-energy conformation may yield an energetic barrier to correct substrate binding, providing rationale for lack of reactivity of 51.

While previous work has investigated the interaction of PikCD50N with synthetic substrate mimics, ,, this work is the first to describe the interaction of PikCD50N with narbomycin (58) via molecular dynamics simulations. The C3′ dimethylamino group of desosamine forms a salt bridge with Glu85, anchoring the substrate for catalysis. , During the MD simulations of narbomycin (58) in the active site, we found that the macrocycle undergoes a conformational change that results in formation of an anchoring interaction with Asn381. The hydrogen bond contact between narbomycin (58) and the Asn381 side chain, while not present within the ligand bound PikCD50N crystal structure, remains throughout the MD simulation (72% of 1 μs simulation <3 Å bond distance, Figure S32). In contrast, macrodiolide 53 does not interact with this residue (0% of 1 μs simulation <3 Å bond distance). Therefore, we generated a PikCD50NN381A mutant, both in silico and in vitro, to probe the reactivity and selectivity-control of this residue for narbomycin (58) and macrodiolide 53. The in silico 1 μs PikCD50NN381A MD trajectories resulted in reduced NAC sampling for narbomycin (58) and macrodiolide (53), which were supported by the observed decrease in substrate consumption from in vitro biocatalytic reactions of PikCD50NN381A with narbomycin (58) and macrodiolide 53 as compared to PikCD50N (Figures S30, S33–S34). These data further support the importance of Asn381 for substrate positioning in the PikC active site, providing a deeper understanding to PikC reactivity and a key starting point for future P450 engineering efforts.

Conclusions

In this work, we evaluated the substrate flexibility of the PikAIII-TE, PikAIII/PikAIV, DesVII/DesVIII and PikC biocatalysts for development of 14-membered ring macrolides bearing non-native functionalities including altered side chain substituents and unnatural heteroatom exchanges in the polyketide chain. Initially, four new pentaketides, including newly developed amides (15 and 16), ester (17), and gem-dimethyl (18) substrates, were queried with PikAIII-TE (to generate 12-membered ring macrolactones), with preferential conversion of the N-methylated analog 16 to 12-membered macrolactone/lactam (33). The N–H amide- and ester-containing pentaketide substrates were accepted and extended by PikAIII-TE to varying degrees; however, penta- and hexaketide byproducts were almost exclusively identified. These data suggest early catalytic domain (e.g., KS domain) flexibility for chain extension, though failure in the cyclization step via the TE domain, further supporting the Pik TE as a highly selective catalytic bottleneck. , The selective cyclization of the N-methylated hexaketide intermediate, compared to the N–H amide and ester analogs, is likely due to the interconversion between the s-trans and s-cis isomeric forms, as well as enhanced hydrophobic interactions within the Pik TE active site resulting from the presence of the N-methyl group. , While the reaction of PikAIII-TE (WT and S148C) with N–H pentaketide (15) shows inefficiency in generating the 12-membered macrolactone/lactam (33) due to the preference for hydrolysis of the acylated pentaketide and hexaketide intermediates, we recently discovered that the Pik TE domain can be engineered via stepwise directed evolution to favor efficient formation (up to 79% yield) of the 12-membered macrolactone (32) from the N–H hexaketide thioester. We also demonstrated that the reaction yields of pentaketides with PikAIII-TE differ significantly from those of hexaketides and the excised Pik TE domain. Specifically, reactions of PikAIII-TE (S148C) with substrates 15 and 16 produced compounds 32 and 33 in trace amounts and 15% isolated yield, respectively, compared to 13% and 75% yields when using the corresponding hexaketide and excised Pik TE (S148C). These data support the presence of biocatalytic inefficiencies in domains upstream of the thioesterase in the PikAIII module (Figures S1 and S2). , Finally, the gem-dimethyl pentaketide (18) displayed minimal substrate consumption, indicating that the KS5 domain was insufficiently flexible to accept and process intermediates bearing additional steric bulk proximal to the thioester bond.

Subsequently, we expanded the substrate scope analysis with PikAIII/PikAIV and identified products corresponding to the 14-membered rings derived from unnatural pentaketides (11–14) bearing altered side chain substituents. We also observed 14-membered macrocycles derived from all three hetereoatom exchanged pentaketides, though minimal substrate consumption was observed for gem-dimethyl pentaketide, consistent with the PikAIII-TE reactions. Our studies revealed that the 12- and 14-methyl and 13-ethyl groups are important for recognition by the PKS enzymes, as indicated by the variations in macrolactone yields. For pentaketides 11–14, the observed yields for 14-membered ring formation follow a similar trend compared to our previously reported yields for 12-membered ring formation. These data demonstrate the importance of the branched and terminal alkyl substituents for KS processing and TE-mediated cyclization. Pik TE exhibited moderate tolerance for amide (15 and 16) and ester (17) containing pentaketides following chain elongation and β-keto group processing through PikAIII/PikAIV. The observed reaction profiles of non-native substrates with PikAIII-TE and the PikAIII/PikAIV tandem modules demonstrate that these enzymes display significant flexibility for extension, processing and catalyzing formation of novel macrocyclic products when challenged with unnatural substrates, including those that contain heteroatom replacements. However, it is evident that the cyclization selectivity differs based on ring size. While WT TE is mainly selective for cyclization of the N-methylated hexaketide intermediate to the 12-membered ring 33, 14-membered ring formation was observed for all analyzed hetereoatom exchanged analogs and these macrocycles were isolated in similar yields.

We also demonstrated that Pik TE catalyzes ring opening of 14-membered macrolactones followed by ring closure to afford the corresponding thermodynamically stable 6-membered δ-lactones, an unexpected activity for the TE domain. The 6-membered δ-lactones derived from Pik pentaketide (1) and truncated analogs (11–14) were only observed in trace amounts from reactions with PikAIII/PikAIV. Conversely, the 6-membered δ-lactones (43–45) derived from heteroatom-exchanged analogs (15–17) were observed up to 20% in reactions using PikAIII/PikAIV. Our findings align with a recent report of a 6-membered δ-lactone produced by a refactored pikromycin synthase via module exchange. Compounds 43–46 are similar to the polyketide natural product discodermolide, a potent microtubule-stabilizing agent in cancer cells that contains a fully substituted 6-membered δ-lactone ring, posing a challenge for chemical synthesis. In addition, the plant natural product, camptothecin, contains a critical δ-lactone moiety which is essential for its topoisomerase I inhibitory activity. Finally, lovastatin, produced by Aspergillus terreus and used to treat hypercholesterolemia, contains a δ-lactone in its prodrug form. , Our work demonstrates the ability to generate two distinct ring sizes mediated by a TE-catalyzed process (e.g., rebound hydrolysis and cyclization), thereby opening avenues for further production of lactone-containing molecules. −

Finally, macrolide products were generated by an engineered strain of S. venezuelae (DHS316), and in vitro application of the P450 monooxygenase PikC as a late-stage C–H functionalization biocatalyst. Although the tailoring steps uniformly occurred for the glycosyltransferases with yields between 37–69% for 35–41, the PikC P450 reactions proceeded with a wide range of yields from trace amounts to 71%. This spectrum of reactivity helps clarify the functionalities required in the macrolactone scaffold for late-stage C–H oxidation. While both amide substrates (51 and 52) contain the key C12 methyl group, oxidation was not observed in either case, suggesting that modifications to the native C-Me functionality also affect oxidization preference. Unexpectedly, macrodiolide (53) was partially converted to the canonical 12-hydroxy product (56) in equal amounts to the energetically disfavored 14-hydroxy product (57). This surprising oxidation profile contrasts significantly with the ∼40:1 ratio of pikromycin (8) to neopikromycin (9) from oxidation of narbomycin (58). However, these results align with the 1:1 ratio obtained of methymycin (3) and neomethymycin (4) from 12-membered YC-17, which is supported by molecular dynamics simulations (Figures and S28). Utilizing MD simulations in combination with DFT calculations provides a quantitative model for these oxidation profiles, which will be especially useful for future engineering efforts of P450 monooxygenases for generation of novel macrolides.

Overall, this work has provided deeper insight into the mechanism, limitations, and flexibility of the PikAIII-TE, PikAIII/PikAIV, DesVII/DesVIII and PikC biocatalysts toward non-native substrates. We expect rational engineering or directed evolution of individual domains, including the KS, TE and P450 monooxygenases, will improve catalytic efficiencies and product yields. Identification and analysis of heterologous P450s from alternative macrolide pathways will be of continuing interest to further diversify macrolide structures.

Supplementary Material

au5c00949_si_001.pdf (19.9MB, pdf)

Acknowledgments

We are grateful for support from NIH grant R35 GM118101, a F31 Fellowship GM143769 (to M.L.A.-C.), a Michigan Pioneer Fellowship (to B.J.C.), a T32 GM140223 PSTP training grant (to H.A.B.), and the Hans W. Vahlteich Professorship (to D.H.S.). We are grateful to Nicole Rivera-Fuentes and Rachel L. O’Rourke for their assistance in substrate synthesis during the early phase of this work.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacsau.5c00949.

  • Detailed experimental and analytical methodologies, compound synthesis and characterizations, expression and purification of Pik proteins, computational methods, additional figures and tables (PDF)

§.

B.J.C. and H.A.B. contributed equally to this work.

The authors declare no competing financial interest.

References

  1. Weissman K. J.. Polyketide Biosynthesis: Understanding and Exploiting Modularity. Philos. Trans. R. Soc., A. 2004;362(1825):2671–2690. doi: 10.1098/rsta.2004.1470. [DOI] [PubMed] [Google Scholar]
  2. Mortison J. D., Sherman D. H.. Frontiers and Opportunities in Chemoenzymatic Synthesis. J. Org. Chem. 2010;75(21):7041–7051. doi: 10.1021/jo101124n. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Weissman K. J.. Uncovering the Structures of Modular Polyketide Synthases. Nat. Prod. Rep. 2015;32(3):436–453. doi: 10.1039/C4NP00098F. [DOI] [PubMed] [Google Scholar]
  4. Adrover-Castellano M. L., Schmidt J. J., Sherman D. H.. Biosynthetic Cyclization Catalysts for the Assembly of Peptide and Polyketide Natural Products. ChemCatChem. 2021;13(9):2095–2116. doi: 10.1002/cctc.202001886. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Weissman K. J.. Genetic Engineering of Modular PKSs: from Combinatorial Biosynthesis to Synthetic Biology. Nat. Prod. Rep. 2016;33(2):203–230. doi: 10.1039/C5NP00109A. [DOI] [PubMed] [Google Scholar]
  6. Du L., Lou L.. PKS and NRPS Release Mechanisms. Nat. Prod. Rep. 2010;27(2):255–278. doi: 10.1039/B912037H. [DOI] [PubMed] [Google Scholar]
  7. Nivina A., Yuet K. P., Hsu J., Khosla C.. Evolution and Diversity of Assembly-Line Polyketide Synthases. Chem. Rev. 2019;119(24):12524–12547. doi: 10.1021/acs.chemrev.9b00525. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Xue Y., Zhao L., Liu H. W., Sherman D. H.. A Gene Cluster for Macrolide Antibiotic Biosynthesis in Streptomyces venezuelae: Architecture of Metabolic Diversity. Proc. Natl. Acad. Sci. U.S.A. 1998;95(21):12111–12116. doi: 10.1073/pnas.95.21.12111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Lee S. K., Park J. W., Kim J. W., Jung W. S., Park S. R., Choi C. Y., Kim E. S., Kim B. S., Ahn J. S., Sherman D. H., Yoon Y. J.. Neopikromycin and Novapikromycin from the Pikromycin Biosynthetic Pathway of Streptomyces venezuelae . J. Nat. Prod. 2006;69(5):847–849. doi: 10.1021/np060026p. [DOI] [PubMed] [Google Scholar]
  10. Hansen D. A., Rath C. M., Eisman E. B., Narayan A. R., Kittendorf J. D., Mortison J. D., Yoon Y. J., Sherman D. H.. Biocatalytic Synthesis of Pikromycin, Methymycin, Neomethymycin, Novamethymycin, and Ketomethymycin. J. Am. Chem. Soc. 2013;135(30):11232–11238. doi: 10.1021/ja404134f. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gerlinger, P. D. A. G. ; Paczia, N. ; Erb, T. J. . Ketosynthase engineering enhances activity and shifts specificity towards non-native extender units in type I linear polyketide synthase. bioRxiv, 2023, 10.1101/2023.02.16.528826. [DOI]
  12. Chiang C. Y., Ohashi M., Tang Y.. Deciphering chemical logic of fungal natural product biosynthesis through heterologous expression and genome mining. Nat. Prod. Rep. 2023;40(1):89–127. doi: 10.1039/D2NP00050D. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hansen D. A., Koch A. A., Sherman D. H.. Identification of a Thioesterase Bottleneck in the Pikromycin Pathway through Full-Module Processing of Unnatural Pentaketides. J. Am. Chem. Soc. 2017;139(38):13450–13455. doi: 10.1021/jacs.7b06432. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Koch A. A., Hansen D. A., Shende V. V., Furan L. R., Houk K. N., Jimenez-Oses G., Sherman D. H.. A Single Active Site Mutation in the Pikromycin Thioesterase Generates a More Effective Macrocyclization Catalyst. J. Am. Chem. Soc. 2017;139(38):13456–13465. doi: 10.1021/jacs.7b06436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Peng H., Ishida K., Sugimoto Y., Jenke-Kodama H., Hertweck C.. Emulating evolutionary processes to morph aureothin-type modular polyketide synthases and associated oxygenases. Nat. Commun. 2019;10(1):3918. doi: 10.1038/s41467-019-11896-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Ray K. A., Lutgens J. D., Bista R., Zhang J., Desai R. R., Hirsch M., Miyazawa T., Cordova A., Keatinge-Clay A. T.. Assessing and harnessing updated polyketide synthase modules through combinatorial engineering. Nat. Commun. 2024;15(1):6485. doi: 10.1038/s41467-024-50844-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Huang Z., Xie S., Liu R. Z., Xiang C., Yao S., Zhang L.. Plug-and-play engineering of modular polyketide synthases. Nat. Chem. Biol. 2025;21(9):1361–1367. doi: 10.1038/s41589-025-01878-4. [DOI] [PubMed] [Google Scholar]
  18. Mabesoone M. F. J., Leopold-Messer S., Minas H. A., Chepkirui C., Chawengrum P., Reiter S., Meoded R. A., Wolf S., Genz F., Magnus N.. et al. Evolution-guided engineering of trans-acyltransferase polyketide synthases. Science. 2024;383(6689):1312–1317. doi: 10.1126/science.adj7621. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Keatinge-Clay A., Miyazawa T.. Refactoring the Pikromycin Synthase for the Modular Biosynthesis of Macrolide Antibiotics in E. coli . Res. Sq. 2025:rs.3.rs-5640596. doi: 10.21203/rs.3.rs-5640596/v1. [DOI] [Google Scholar]
  20. Seiple I. B., Zhang Z., Jakubec P., Langlois-Mercier A., Wright P. M., Hog D. T., Yabu K., Allu S. R., Fukuzaki T., Carlsen P. N.. et al. A platform for the discovery of new macrolide antibiotics. Nature. 2016;533(7603):338–345. doi: 10.1038/nature17967. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Chen L., Quan H., Xu Z., Wang H., Xia Y., Lou L., Yang W.. A modular biomimetic strategy for the synthesis of macrolide P-glycoprotein inhibitors via Rh-catalyzed C-H activation. Nat. Commun. 2020;11(1):2151. doi: 10.1038/s41467-020-16084-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Malico A. A., Nichols L., Williams G. J.. Synthetic Biology Enabling Access to Designer Polyketides. Curr. Opin. Chem. Biol. 2020;58:45–53. doi: 10.1016/j.cbpa.2020.06.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hwang S., Lee N., Cho S., Palsson B., Cho B. K.. Repurposing Modular Polyketide Synthases and Non-ribosomal Peptide Synthetases for Novel Chemical Biosynthesis. Front. Mol. Biosci. 2020;7:87. doi: 10.3389/fmolb.2020.00087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Paulsel T. Q., Williams G. J.. Current State-of-the-Art Toward Chemoenzymatic Synthesis of Polyketide Natural Products. ChemBioChem. 2023;24(21):e202300386. doi: 10.1002/cbic.202300386. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Aldrich C. C., Beck B. J., Fecik R. A., Sherman D. H.. Biochemical Investigation of Pikromycin Biosynthesis Employing Native Penta- and Hexaketide Chain Elongation Intermediates. J. Am. Chem. Soc. 2005;127(23):8441–8452. doi: 10.1021/ja042592h. [DOI] [PubMed] [Google Scholar]
  26. Aldrich C. C., Venkatraman L., Sherman D. H., Fecik R. A.. Chemoenzymatic Synthesis of the Polyketide Macrolactone 10-Deoxymethynolide. J. Am. Chem. Soc. 2005;127(25):8910–8911. doi: 10.1021/ja0504340. [DOI] [PubMed] [Google Scholar]
  27. Adrover-Castellano M. L., Curtis B. J., Schmidt J. J., Boesger H. A., Glasser C. A., Olukorede D. E., Qu F., Sherman D. H.. Directed Evolution of a Modular Polyketide Synthase Thioesterase for Generation of a Hybrid Macrocyclic Ring System. ACS Catal. 2025;15(4):3405–3417. doi: 10.1021/acscatal.4c07922. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Hansen D. A., Koch A. A., Sherman D. H.. Substrate Controlled Divergence in Polyketide Synthase Catalysis. J. Am. Chem. Soc. 2015;137(11):3735–3738. doi: 10.1021/ja511743n. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Pavlović D., Mutak S.. Synthesis and Structure-Activity Relationships of Novel 8a-aza-8a-Homoerythromycin A Ketolides. J. Med. Chem. 2010;53(15):5868–5880. doi: 10.1021/jm100711p. [DOI] [PubMed] [Google Scholar]
  30. Stimac V., Alihodzic S., Lazarevski G., Mutak S., Istuk Z. M., Fajdetic A., Palej I., Paljetak H. C., Padovan J., Marsic N., Tavcar B., Padovan J.. Synthesis and Biological Properties of 4″-O-Acyl Derivatives of 8a-Aza-8a-Homoerythromycin. J. Antibiot. 2009;62(3):133–144. doi: 10.1038/ja.2009.1. [DOI] [PubMed] [Google Scholar]
  31. Hutinec A., Derek M., Lazarevski G., Sunjic V., Paljetak H. C., Alihodzic S., Erakovic Haber V., Dumic M., Marsic N., Mutak S.. Novel 8a-Aza-8a-Homoerythromycin--4″-(3-Substituted-Amino)­Propionates with Broad Spectrum Antibacterial Activity. Bioorg. Med. Chem. Lett. 2010;20(11):3244–3249. doi: 10.1016/j.bmcl.2010.04.062. [DOI] [PubMed] [Google Scholar]
  32. Pavlović D., Fajdetic A., Mutak S.. Novel Hybrids of 15-Membered 8a- and 9a-Azahomoerythromycin A Ketolides and Quinolones as Potent Antibacterials. Bioorg. Med. Chem. 2010;18(24):8566–8582. doi: 10.1016/j.bmc.2010.10.024. [DOI] [PubMed] [Google Scholar]
  33. Kimishima A., Ando H., Sennari G., Noguchi Y., Sekikawa S., Kojima T., Ohara M., Watanabe Y., Inahashi Y., Takada H.. et al. Chemical Degradation-Inspired Total Synthesis of the Antibiotic Macrodiolide, Luminamicin. J. Am. Chem. Soc. 2022;144(50):23148–23157. doi: 10.1021/jacs.2c10856. [DOI] [PubMed] [Google Scholar]
  34. Lacey H. J., Booth T. J., Vuong D., Rutledge P. J., Lacey E., Chooi Y. H., Piggott A. M.. Conglobatins B-E: Cytotoxic Analogues of the C(2)-Symmetric Macrodiolide Conglobatin. J. Antibiot. 2020;73(11):756–765. doi: 10.1038/s41429-020-0332-3. [DOI] [PubMed] [Google Scholar]
  35. Zhou Y., Prediger P., Dias L. C., Murphy A. C., Leadlay P. F.. Macrodiolide Formation by the Thioesterase of a Modular Polyketide Synthase. Angew. Chem., Int. Ed. 2015;54(17):5232–5235. doi: 10.1002/anie.201500401. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Hunter T. J., O’Doherty G. A.. Enantioselective Syntheses of Colletodiol, Colletol, and Grahamimycin A. Org. Lett. 2002;4(25):4447–4450. doi: 10.1021/ol0269502. [DOI] [PubMed] [Google Scholar]
  37. Salvatore M. M., DellaGreca M., Nicoletti R., Salvatore F., Vinale F., Naviglio D., Andolfi A.. Talarodiolide, a New 12-Membered Macrodiolide, and GC/MS Investigation of Culture Filtrate and Mycelial Extracts of Talaromyces pinophilus . Molecules. 2018;23(4):950. doi: 10.3390/molecules23040950. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Yamada T., Kogure H., Kataoka M., Kikuchi T., Hirano T.. Halosmysin A, a Novel 14-Membered Macrodiolide Isolated from the Marine-Algae-Derived Fungus Halosphaeriaceae sp . Mar. Drugs. 2020;18(6):320. doi: 10.3390/md18060320. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kohda Y., Sakamoto S., Umekita M., Kimura T., Kubota Y., Arisaka R., Shibuya Y., Muramatsu H., Sawa R., Dan S.. et al. Isolation of new derivatives of the 20-membered macrodiolide bispolide from Kitasatospora sp. MG372-hF19. J. Antibiot. 2022;75(2):77–85. doi: 10.1038/s41429-021-00492-5. [DOI] [PubMed] [Google Scholar]
  40. Nicolaou K. C., Estrada A. A., Zak M., Lee S. H., Safina B. S.. A Mild and Selective Method for the Hydrolysis of Esters with Trimethyltin Hydroxide. Angew. Chem., Int. Ed. 2005;44(9):1378–1382. doi: 10.1002/anie.200462207. [DOI] [PubMed] [Google Scholar]
  41. Talele T. T.. Natural-Products-Inspired Use of the gem-Dimethyl Group in Medicinal Chemistry. J. Med. Chem. 2018;61(6):2166–2210. doi: 10.1021/acs.jmedchem.7b00315. [DOI] [PubMed] [Google Scholar]
  42. Ray L., Moore B. S.. Recent Advances in the Biosynthesis of Unusual Polyketide Synthase Substrates. Nat. Prod. Rep. 2016;33(2):150–161. doi: 10.1039/C5NP00112A. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Poust S., Phelan R. M., Deng K., Katz L., Petzold C. J., Keasling J. D.. Divergent Mechanistic Routes for the Formation of gem-Dimethyl Groups in the Biosynthesis of Complex Polyketides. Angew. Chem., Int. Ed. 2015;54(8):2370–2373. doi: 10.1002/anie.201410124. [DOI] [PubMed] [Google Scholar]
  44. Prasad G., Borketey L. S., Lin T. Y., Schnarr N. A.. A Mechanism-Based Fluorescence Transfer Assay for Examining Ketosynthase Selectivity. Org. Biomol. Chem. 2012;10(33):6717–6723. doi: 10.1039/c2ob26008e. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Murphy A. C., Hong H., Vance S., Broadhurst R. W., Leadlay P. F.. Broadening Substrate Specificity of a Chain-Extending Ketosynthase through a Single Active-Site Mutation. Chem. Commun. 2016;52(54):8373–8376. doi: 10.1039/C6CC03501A. [DOI] [PubMed] [Google Scholar]
  46. Jo W. S., Curtis B. J., Rehan M., Adrover-Castellano M. L., Sherman D. H., Healy A. R.. N-to-S Acyl Transfer as an Enabling Strategy in Asymmetric and Chemoenzymatic Synthesis. JACS Au. 2024;4(5):2058–2066. doi: 10.1021/jacsau.4c00257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Ackenhusen S. E., Wang Y., Chun S. W., Narayan A. R. H.. Understanding and Circumventing the Requirement for Native Thioester Substrates for alpha-Oxoamine Synthase Reactions. ACS Chem. Biol. 2022;17(9):2389–2395. doi: 10.1021/acschembio.2c00365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Ogura H., Furuhata K., Kuwano H., Harada N.. Stereochemistry of Macrolides. I. 1a-Conformation of Aglycones of Pikromycin and Narbomycin and their Derivatives. J. Am. Chem. Soc. 1975;97(7):1930–1934. doi: 10.1021/ja00840a054. [DOI] [PubMed] [Google Scholar]
  49. Boddy C. N., Schneider T. L., Hotta K., Walsh C. T., Khosla C.. Epothilone C Macrolactonization and Hydrolysis are Catalyzed by the Isolated Thioesterase Domain of Epothilone Polyketide Synthase. J. Am. Chem. Soc. 2003;125(12):3428–3429. doi: 10.1021/ja0298646. [DOI] [PubMed] [Google Scholar]
  50. McCullough T. M., Choudhary V., Akey D. L., Skiba M. A., Bernard S. M., Kittendorf J. D., Schmidt J. J., Sherman D. H., Smith J. L.. Substrate Trapping in Polyketide Synthase Thioesterase Domains: Structural Basis for Macrolactone Formation. ACS Catal. 2024;14(16):12551–12563. doi: 10.1021/acscatal.4c03637. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Lowell A. N., DeMars M. D. 2nd, Slocum S. T., Yu F., Anand K., Chemler J. A., Korakavi N., Priessnitz J. K., Park S. R., Koch A. A.. et al. Chemoenzymatic Total Synthesis and Structural Diversification of Tylactone-Based Macrolide Antibiotics through Late-Stage Polyketide Assembly, Tailoring, and C-H Functionalization. J. Am. Chem. Soc. 2017;139(23):7913–7920. doi: 10.1021/jacs.7b02875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Sherman D. H., Li S., Yermalitskaya L. V., Kim Y., Smith J. A., Waterman M. R., Podust L. M.. The Structural Basis for Substrate Anchoring, Active Site Selectivity, and Product Formation by P450 PikC from Streptomyces venezuelae . J. Biol. Chem. 2006;281(36):26289–26297. doi: 10.1074/jbc.M605478200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Li S., Ouellet H., Sherman D. H., Podust L. M.. Analysis of Transient and Catalytic Desosamine-Binding Pockets in Cytochrome P-450 PikC from Streptomyces venezuelae . J. Biol. Chem. 2009;284(9):5723–5730. doi: 10.1074/jbc.M807592200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Narayan A. R. H., Jimenez-Oses G., Liu P., Negretti S., Zhao W., Gilbert M. M., Ramabhadran R. O., Yang Y. F., Furan L. R., Li Z.. et al. Enzymatic Hydroxylation of an Unactivated Methylene C-H Bond Guided by Molecular Dynamics Simulations. Nat. Chem. 2015;7(8):653–660. doi: 10.1038/nchem.2285. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Li S., Chaulagain M. R., Knauff A. R., Podust L. M., Montgomery J., Sherman D. H.. Selective Oxidation of Carbolide C-H bonds by an Engineered Macrolide P450 Mono-Oxygenase. Proc. Natl. Acad. Sci. U.S.A. 2009;106(44):18463–18468. doi: 10.1073/pnas.0907203106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Negretti S., Narayan A. R., Chiou K. C., Kells P. M., Stachowski J. L., Hansen D. A., Podust L. M., Montgomery J., Sherman D. H.. Directing Group-Controlled Regioselectivity in an Enzymatic C-H Bond Oxygenation. J. Am. Chem. Soc. 2014;136(13):4901–4904. doi: 10.1021/ja5016052. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Wang Z. C., Stegall H., Miyazawa T., Keatinge-Clay A. T.. A CRISPR-Cas9 system for knock-out and knock-in of high molecular weight DNA enables module-swapping of the pikromycin synthase in its native host. Microb. Cell Fact. 2025;24(1):125. doi: 10.1186/s12934-025-02741-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Yang S., DeMars M. D. 2nd, Grandner J. M., Olson N. M., Anzai Y., Sherman D. H., Houk K. N.. Computational-Based Mechanistic Study and Engineering of Cytochrome P450 MycG for Selective Oxidation of 16-Membered Macrolide Antibiotics. J. Am. Chem. Soc. 2020;142(42):17981–17988. doi: 10.1021/jacs.0c04388. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Akey D. L., Kittendorf J. D., Giraldes J. W., Fecik R. A., Sherman D. H., Smith J. L.. Structural Basis for Macrolactonization by the Pikromycin Thioesterase. Nat. Chem. Biol. 2006;2(10):537–542. doi: 10.1038/nchembio824. [DOI] [PubMed] [Google Scholar]
  60. Xue Y., Wilson D., Zhao L., Liu H., Sherman D. H.. Hydroxylation of Macrolactones YC-17 and Narbomycin is Mediated by the PikC-Encoded Cytochrome P450 in Streptomyces venezuelae . Chem. Biol. 1998;5(11):661–667. doi: 10.1016/S1074-5521(98)90293-9. [DOI] [PubMed] [Google Scholar]
  61. Caddell Haatveit K., Garcia-Borras M., Houk K. N.. Computational Protocol to Understand P450 Mechanisms and Design of Efficient and Selective Biocatalysts. Front. Chem. 2019;6:663. doi: 10.3389/fchem.2018.00663. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Bvumbi M. V., Zimuwandeyi M., Nemudzivhadi A. I., Kabanda M. M.. Rotational barriers in N-methyl-cinnamoyl-chloroquine analogues: Variable temperature NMR spectroscopy and DFT studies. J. Mol. Struct. 2025;1335:141943. doi: 10.1016/j.molstruc.2025.141943. [DOI] [Google Scholar]
  63. Huang J., Chen L., Ikeda H., Hyodo T., Tang Y., Otani Y., Yamaguchi K., Inagaki S., Ohwada T.. Switching Conjugation Is the Predominant Factor Contributing to Complete Reversal of Amide cis–trans (Z-E)­Preference through N-Methylation. J. Org. Chem. 2025;90:4854–4872. doi: 10.1021/acs.joc.4c02929. [DOI] [PubMed] [Google Scholar]
  64. Smith A. B., Freeze B. S.. (+)-Discodermolide: Total Synthesis, Construction of Novel Analogues, and Biological Evaluation. Tetrahedron. 2008;64(2):261–298. doi: 10.1016/j.tet.2007.10.039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Redinbo M. R., Stewart L., Kuhn P., Champoux J. J., Hol W. G.. Crystal Structures of Human Topoisomerase I in Covalent and Noncovalent Complexes with DNA. Science. 1998;279(5356):1504–1513. doi: 10.1126/science.279.5356.1504. [DOI] [PubMed] [Google Scholar]
  66. Alberts A. W., Chen J., Kuron G., Hunt V., Huff J., Hoffman C., Rothrock J., Lopez M., Joshua H., Harris E.. et al. Mevinolin: a Highly Potent Competitive Inhibitor of Hydroxymethylglutaryl-Coenzyme A Reductase and a Cholesterol-lowering Agent. Proc. Natl. Acad. Sci. U.S.A. 1980;77(7):3957–3961. doi: 10.1073/pnas.77.7.3957. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Subhan M., Faryal R., Macreadie I.. Exploitation of Aspergillus terreus for the Production of Natural Statins. J. Fungi. 2016;2(2):13. doi: 10.3390/jof2020013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Martí-Centelles V., Pandey M. D., Burguete M. I., Luis S. V.. Macrocyclization Reactions: The Importance of Conformational, Configurational, and Template-Induced Preorganization. Chem. Rev. 2015;115(16):8736–8834. doi: 10.1021/acs.chemrev.5b00056. [DOI] [PubMed] [Google Scholar]
  69. Parenty A., Moreau X., Niel G., Campagne J. M.. Update 1 of: Macrolactonizations in the Total Synthesis of Natural Products. Chem. Rev. 2013;113(1):PR1–40. doi: 10.1021/cr300129n. [DOI] [PubMed] [Google Scholar]
  70. Fürstner A.. Lessons from Natural Product Total Synthesis: Macrocyclization and Postcyclization Strategies. Acc. Chem. Res. 2021;54(4):861–874. doi: 10.1021/acs.accounts.0c00759. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

au5c00949_si_001.pdf (19.9MB, pdf)

Articles from JACS Au are provided here courtesy of American Chemical Society

RESOURCES