Crystal Structure and Biochemical Studies of the Trans-Acting Polyketide Enoyl Reductase Lovc from Lovastatin Biosynthesis

Home , Mannitol dehydrogenase

Crystal structure and biochemical studies of the trans-acting polyketide enoyl reductase LovC from lovastatin biosynthesis

Brian D. Amesa, Chi Nguyena, Joel Brueggera, Peter Smitha, Wei Xub, Suzanne Mab, Emily Wonga, Steven Wonga, Xinkai Xieb, Jesse W.-H. Lic, John C. Vederasc, Yi Tangb,1, and Shiou-Chuan Tsaia,1

aDepartments of Molecular Biology and Biochemistry, Chemistry, and Pharmaceutical Sciences, University of California, Irvine, CA 92697; bDepartment of Chemical and Biomolecular Engineering, University of California, Los Angeles, CA 90095; and cDepartment of Chemistry, University of Alberta, Edmonton, AB, Canada T6G 2G2

Edited by Gregory A. Petsko, Brandeis University, Waltham, MA, and approved May 1, 2012 (received for review August 10, 2011)

Lovastatin is an important statin prescribed for the treatment and bassiana (14). These trans-acting ERs are highly homologous (38– prevention of cardiovascular diseases. Biosynthesis of lovastatin 80% sequence identity; Fig. S1A) and are observed to interact with fi – uses an iterative type I polyketide synthase (PKS). LovC is a trans- megasynthases through speci cproteinprotein interactions (4). fi acting enoyl reductase (ER) that specifically reduces three out of However, how LovC speci cally interacts with LovB and why it eight possible polyketide intermediates during lovastatin biosyn- reduces only at the tetraketide, pentaketide, and heptaketide stages is not well understood. thesis. Such trans-acting ERs have been reported across a variety of Herein, we report the crystal structure of LovC, which repre- other fungal PKS enzymes as a strategy in nature to diversify poly- fi sents the structure of a polyketide ER from an iterative reducing ketides. How LovC achieves such speci city is unknown. The 1.9-Å Type I PKS. LovC is a 39.5-kDa protein (363 residues) and, sur- structure of LovC reveals that LovC possesses a medium-chain de- prisingly, is shown to be a unique monomeric member of the me- hydrogenase/reductase (MDR) fold with a unique monomeric as- dium-chain dehydrogenase/reductase (MDR) superfamily (15). sembly. Two LovC cocrystal structures and enzymological studies Two cocrystal structures are presented: the LovC–crotonoyl help elucidate the molecular basis of LovC specificity, define stereo- CoA complex and the LovC-NADP+ binary complex. To study the chemistry, and identify active-site residues. Sequence alignment trans-acting ERs and their unique substrate specificity, an in vitro indicates a general applicability to trans-acting ERs of fungal PKSs, assay is essential, and we report the successful development of as well as their potential application to directing biosynthesis. LovC enzyme assays that may be applied to other trans-acting ERs. We elucidated the structural basis of LovC substrate specificity and found that LovC mutations can alter its substrate specificity. The therosclerosis is the current leading cause of death for adults in results can be leveraged with our previous work using trans-acting Athe western world (1). Statins inhibit cholesterol biosynthesis ERs to biosynthesize new statin analogs such as the compactin and are the most widely prescribed drugs for the prevention and precursor (5), and the chemical diversity can potentially be further treatment of atherosclerosis (2). Lovastatin (compound 1) is the fi expanded by using other trans-acting ERs working in conjugation rst statin approved by the Food and Drug Administration and is with the megasynthases for directing biosynthesis. the direct precursor for manufacture of simvastatin, the second most-prescribed drug worldwide (1). The high impact of statins to Results and Discussion human health, including their possible use for cancer and neuro- degenerative diseases (3), has prompted vigorous efforts toward Overall Structure: LovC Is a Unique Monomeric ER and MDR Member. synthesis of lovastatin (1). However, the structural complexity of Unlike previously reported ER domains [Protein Data Bank lovastatin has prevented its commercial production by total chem- (PDB) ID codes 1GUF and 1ZSY], LovC exists as a monomer as ical synthesis (1). In nature, lovastatin is biosynthesized by the detected from the crystal structure (Fig. 2) and size-exclusion fungus Aspergillus terreus (4) using a gene cluster that contains two chromatography (Fig. 3). The monomeric state of LovC is highly polyketide synthases (PKSs), LovB and LovF (Fig. 1A)(5).Based unique in the MDR superfamily (15), the members of which are on past biochemical studies of lovastatin biosynthesis, our previous functionally dimeric or tetrameric, except for the monomeric bioengineering efforts led to rational generation of lovastatin ana- mannitol dehydrogenase (16). The MDR superfamily includes the zinc-containing alcohol dehydrogenases (ADHs), leukotriene logs with enhanced activities and reduced side effects (6). Despite dehydrogenases, quinone oxidoreductases, the eukaryotic the above advances, many questions about lovastatin biosynthesis ERs, and membrane-sensing proteins (15). Because none of the remain unanswered. Understanding lovastatin biosynthesis repre- MDR proteins of known structures shares significant sequence sents an opportunity to leverage our knowledge for future efforts in homology to LovC, we could not solve the LovC structure by directing biosynthesis of statin analogs. molecular replacement. Subsequently, the LovC structure was During lovastatin biosynthesis, the iterative type I PKS LovB solved by multiwavelength anomalous dispersion (MAD) tech- builds the nonaketide main framework with the aid of the trans- nique using selenomethionine-derived protein (Table S1). The acting enoyl reductase (ER) LovC (4, 7). A second type I iterative PKS, LovF, constructs the 2-methylbutyryl side chain. LovB and LovF are both large multidomain megasynthases that are highly homologous in sequence and domain arrangements (Fig. 1A). Author contributions: B.D.A., P.S., J.C.V., Y.T., and S.-C.T. designed research; B.D.A., C.N., When coupled with LovC, LovB iteratively catalyzes more than 30 P.S., W.X., E.W., S.W., and J.W.-H.L. performed research; B.D.A., S.M., X.X., and J.W.-H.L. contributed new reagents/analytic tools; B.D.A., J.B., and S.-C.T. analyzed data; and B.D.A., precisely synchronized reactions to yield a 19-carbon intermediate C.N., and S.-C.T. wrote the paper. dihydromonacolin L (DmL) (compound 2) (Fig. 1A). In compari- fl son, LovF only catalyzes one round of Claisen condensation to The authors declare no con ict of interest. produce a 2-methylbutyryl intermediate, which is attached to the C8 This article is a PNAS Direct Submission. hydroxyl of monacolin J (compound 3) to produce lovastatin. LovC Freely available online through the PNAS open access option. specifically interacts with only LovB, but not LovF, and accepts only Data deposition: The atomic coordinates and structure factors reported in this paper have three out of eight possible LovB intermediates as its substrates (Fig. been deposited in the Protein Data Bank, www.pdb.org (PDB ID codes 3B6Z, 3B70, and 1A; tetra-, penta-, and heptaketides). Similar trans-acting ERs have 3GQV). also been reported in other fungi such as MokE in Monascus pilosus 1To whom correspondence may be addressed. E-mail: [email protected] or [email protected]. (8), MlcG in Penicillium citrinum (9), ApdC in Aspergilus nidulans This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. (10–12), cytochalasans (12, 13), and tenellin_ORF3 in Beauveria 1073/pnas.1113029109/-/DCSupplemental.

11144–11149 | PNAS | July 10, 2012 | vol. 109 | no. 28 www.pnas.org/cgi/doi/10.1073/pnas.1113029109 Downloaded by guest on September 27, 2021 Fig. 1. (A) Proposed lovastatin biosynthetic pathway with 1-3 shown in their acidic forms. (B) Three putative substrates (tetra-, penta- and hepta-ketides) of LovC. (C) In vitro substrates. NADPH (compound 4) and NADH (compound 5) are discussed in the text.

LovC structure contains two domains: the catalytic domain (res- The principal difference between LovC and these homologs lies idues 1–134, 283–363) and cofactor-binding domain (Rossmann in two transition regions between the catalytic and cofactor- fold; residues 135–282) (Fig. 2A). The structural comparison binding domains, consisting of α2-αA-loop and α3-loop-α4. LovC between LovC and selected MDR proteins (18–23% sequence has an additional 10 or more residues in each loop (xL1 and xL2; identity) shows 1.6–2.4 Å rms deviation (RMSD) between the Fig. 2A and Fig. S1B). The extended lengths of these two loops aligned Cα coordinates (Figs. S1B and S2). All are dimers except are a unique conserved feature of trans-acting ERs associated LovC. The least conserved region is located between α2 and βA. with fungal PKSs (Fig. S1A), and, as discussed below, the

Fig. 2. (A) Overall fold and quaternary structure of LovC. Two extended loops, xL1 and xL2, are unique to the LovC structure. (B) Comparison of LovC-NADP+

(blue) vs. LovC-CO7 (yellow) structures showing the major change at Y296. (C) Comparison of both models (rendered as electrostatic surface) reveals that the BIOCHEMISTRY protein surface has signiﬁcant changes. (D) Zoomed view shows that the active site does not change.

Ames et al. PNAS | July 10, 2012 | vol. 109 | no. 28 | 11145 Downloaded by guest on September 27, 2021 Fig. 3. Gel-filtration interaction study indicative of LovB-LovC complex formation. Inset, SDS/ PAGE confirmed interaction: M, marker; 1, purified LovB (335 kDa, doublet resulting from the loss of an N-terminal KS-containing fragment); 2, purified LovC (40 kDa); F9, fraction 9 from the combined run.

extended loops may influence the oligomeric state and protein– (E)-2-butenoyl-acyl-N-acetylcysteamine [(E)-2-butenoyl-NAC] protein interactions of LovC during lovastatin biosynthesis. (compound 8), (2E,4E)-hexadienoyl-NAC (compound 9), and The LovC catalytic domain has three β-sheets: the anti-parallel (E)-2-octenoyl-NAC (compound 10) to mimic the α,β unsaturated Beta1 (β1-β2), the anti-parallel Beta2 (β5-β4-β6), and the six- di-, tri, and tetraketide intermediates, respectively, and (E)-2- stranded Beta3 (β7-β5-β4-β3-β9-β8) (Fig. 2A). The LovC cofactor- octenoic acid (compound 11) as a “NAC-less” analog of com- binding domain has two copies of the β−α−β−α−β Rossmann fold, pound 10 (Fig. 1C). The fact that compound 10 can be reduced but in which βA−αB−βB−αC−βC forms a mirror image with βD−αD not compound 11 supports the importance of the NAC moiety for −βE−βF (Fig. 2A). In dimeric MDR proteins such as the porcine substrate binding (Table 1 and Fig. S4). For the NCI compounds 6 fatty acid synthase (FAS) ER (17) and quinone oxidoreductase and 7, the t-butyl group may serve a similar anchoring/orienting (18), the dimer interface includes the backbone hydrogen bonding purpose as the NAC moiety. The fast turnover rate of compounds across βF from each monomer to form a characteristic 12-stranded 6 and 7 provides a facile assay for ERs in general. These results β-sheet, as well as intermonomer interactions between αF, βE, and show that recombinant LovC is enzymatically active as a stand- βF(Fig. S2). Importantly, computer modeling to create the LovC- alone protein and support that NADPH is its cofactor. LovC homodimer (based on other MDR dimer structures) showed that the presence of the extended xL2 loop (Fig. 2A and Fig. S2) Open and Closed Conformations of LovC. The cocrystal structures of generates significant steric clash with its dimeric partner. The xL2 LovC-isomerized 2′-phosphate crotonoyl-CoA (referred to as loop does not exist in dimeric MDR proteins but is conserved in LovC-CO7; detailed in SI Text) and LovC-NADP+ are highly trans-acting ERs associated with iterative fungal PKSs such as similar, sharing an RMSD of 0.45 Å for Cα atoms (Fig. 2B). MlcG (9), MokE (8), ApdC (10), and tenellin synthase ORF3 (14) Interestingly, upon binding with LovC, the 3′-phosphate of cro- (Fig. S1A). Therefore, the steric hindrance caused by the con- tonyl-CoA is isomerized to 2′-phosphate (Fig. S5A), which pro- served xL2 helps explain why LovC is isolated as a monomeric vides a ligand-binding mode that is remarkably similar to the protein in vitro, and the monomeric state may be a conserved cofactor, with near perfect overlap for the 2′-phosphoadenosine feature shared among the trans-acting ER domains of fungal PKSs. moiety of both the bound CO7 ligand and NADP+. There are LovC is also the first monomeric MDR structure reported (19). many previous reports of acid/base catalyzed 3′ to 2′-phosphate isomerization of an acyl-CoA (22–25), and it is likely that Interactions Between LovC and the LovB Megasynthase. LovC, charged residues near the cofactor ribose binding site may fa- a trans-acting ER, may need to remain monomeric to interact with cilitate the isomerization of crotonyl-CoA (Fig. S5). the LovB megasynthase, which has an inactivated ER0 domain The major conformational difference between the two cocrys- (residues 1,851–2,250) attributable to the deletion of key active site tal structures lies at the xL2 loop (residues 290–301), where the residues (Fig. S3) (4). Complex formation between LovC and LovB LovC-NADP+ structure adopts an open conformation with Y296 has been hypothesized previously (4). We show by size-exclusion pointing toward the solvent, whereas the LovC-CO7 structure chromatography that in the absence of an in vitro substrate, LovB adopts a closed conformation with Y296 pointing toward the and LovC coelute, confirming the formation of LovB-LovC com- active site (Fig. 2B). The conformational change also significantly plex (Fig. 3). LovC likely interacts with LovB by forming a heter- alters the surface potential between the two structures around the odimer with LovB-ER0. However, no complex formation was xL2 region (Fig. 2C) but does not alter the interior portion of the detected when ER0 was cloned, expressed, purified, and incubated putative substrate-binding pocket (Fig. 2D). The open-closed with LovC. Therefore, to form the LovC-LovB complex, LovC may conformations serve as an indicator for the flexible nature of xL2, need to interact with not just the ER0 but multiple domains of so that it can be adaptive for docking with the upstream ACP for LovB [such as the LovB acyl-carrier protein (ACP), dehydratase subsequent substrate transfer. Indeed, a docking simulation be- (DH), MT (methyl transferase), or KR (ketoreductase) domains]. tween a homology model of LovB ACP and LovC identified the Nevertheless, Fig. 3 represents direct evidence for the association LovB ACP-LovC docking site near this region (Fig. S6 C–E). The between a trans-acting ER with its megasynthase partner. docking simulation also indicates that multiple residues are in- volved for ACP-LovC interactions. Supporting this prediction, Cofactor-Binding and Enzyme-Activity Assays Show That LovC Is an single mutations of xL2, including W292A, Y296A and R298A, NADPH-Dependent ER. To determine the cofactor-binding speci- do not change the LovB-LovC interactions, as detected by size ficity of LovC, we used fluorometric titration with NADPH exclusion chromatography (Fig. S7E). Both xL2 and other regions (compound 4) or NADH (compound 5) (Fig. S4 A and B) (20), and predicted for LovB ACP-LovC docking are conserved among we found a strong preference to bind NADPH. To characterize trans-acting ERs from fungal PKSs (Fig. S1A). LovC activity, we searched for in vitro substrates resembling the three putative LovC substrates (Fig. 1B) by ChemDB (21), and we NADP+ Binding. The LovC-NADP+ cocrystal structure identifies identified NCI-636688 (compound 6) and NCI-636689 (compound the cofactor-binding cleft in between the catalytic and Rossmann 7) that can be reduced by LovC in the presence of NADPH but not fold domains (Fig. 2A). Structure and sequence analysis of NAD NADH (Fig. S4). To explore substrate specificity, we synthesized (P)-binding enzymes has identified a “fingerprint” region

11146 | www.pnas.org/cgi/doi/10.1073/pnas.1113029109 Ames et al. Downloaded by guest on September 27, 2021 Table 1. Summary of fluorometric activity assay kinetic intermediate (Fig. 4A). Because K54 is highly conserved (Fig. parameters for WT and mutant LovC S1A), the ACP docking point and the PPT docking entrance may μ −1 −1 μ −1 also be conserved among trans-acting ERs in fungal PKSs. Sample Km ( M) kcat (s ) kcat/Km (s M ) To dissect the possible substrate-binding modes, we conducted α β Substrate: NCI-636688 docking simulations of the putative LovC substrates ( , un- − WT 540 0.048 8.9 × 10 5 saturated tetra-, penta-, and heptaketide) to the LovC active site S51A 1,255 0.12 9.5 × 10−5 using Gold (34). The results for all three substrates are highly −5 consistent, placing the polyketide moiety in the pocket defined K54S 63 0.002 3.1 × 10 − by N49, S51, A93, A135, S138, T139, L142, G282, P283, I285, S51A/K54S 142 0.009 6.5 × 10 5 × −5 and F286, whereas the PPT group loops toward the outer surface A93M 10 0.023 3.8 10 of the pocket to dock the PPT-phosphate over K54 (Fig. 4 A and S138M 55 0.0007 1.1 × 10−5 − B). The putative substrate-binding pocket is large enough to T139V 21 0.0014 6.4 × 10 5 β − accommodate all putative LovC substrates, and places the C of N263A 254 0.037 14.6 × 10 5 each within hydride-transfer distance to the C4 of NADP+. The N263S 247 0.018 7.1 × 10−5 docking simulation also identifies a possible oxyanion hole to Substrate: octenoyl-SNAC accommodate the negative charge developed on C=O after hy- − WT 452 0.0034 0.75 × 10 5 dride transfer. This can be either the side chains of N49, S51, − S51A 290 0.0055 1.89 × 10 5 T68, T139, K54, N263, or the NH of G282 (Fig. 4 A and B). K54S 1,023 0.0009 0.088 × 10−5 These residues are conserved among trans-acting ERs, implying S51A/K54S 446 0.0044 0.98 × 10−5 a conserved catalytic mechanism. − T68V 466 0.012 2.58 × 10 5 Interestingly, MDR sequence or structure alignment does not − A93M 50 0.0012 2.40 × 10 5 reveal a conserved catalytic motif among MDR proteins. The − active site Tyr proposed for MDR homologs such as PDB ID S138M 137 0.0006 0.44 × 10 5 − codes 1GUF (35), 1QOR (36), and 2OBY (37) corresponds to T139V 111 0.0013 1.17 × 10 5 −5 F60 in LovC (Fig. S6A). Therefore, based on docking result and N263A 24 0.0008 3.33 × 10 sequence conservation in trans-acting ERs associated with fungal −5 N263S 102 0.0012 1.18 × 10 PKS, we proposed that S51, K54, T68, and N263 may serve as the NCI, National Cancer Institute. SNAC, S-N-acetylcysteamine. oxyanion hole. Mutations Identified Key Catalytic and Substrate-Binding Residues. (detailed in SI Results and Discussion) (26). For LovC, this region To determine the roles of substrate pocket residues, we generated contains the motif GXXTXXA (Fig. S1B), as well as the βBtoαC two sets of mutants: S51A, K54S, S51A/K54S, T68V, T139V, loop, both are important for NADPH binding and are highly N263A, and N263S to identity catalytic site(s), and A93M and conserved for trans-acting ERs. There is substantial interaction S138M to decipher substrate-binding residues (Table 1). The mutant proteins were expressed, purified, and subjected to the between LovC, the adenosine-phosphate, and nicotinamide-ri- fl bose moieties (Fig. S5B). The 2′-phosphate of NADP+ is situated assays of cofactor binding, uorometric activity, and in vitro reconstitution to screen for the production of DmL. All mutants in an electropositive pocket, whereas the adenine is sandwiched fl between the highly conserved residues Y215 and I239. Two res- displayed an increase in uorescence when titrated with NADPH, indicating that they are all competent in cofactor binding. idues, K54 and N263, form a bridge directly above the nicotin- The steady-state kinetic data obtained from the fluorometric amide ribose, with the side chains of each forming two hydrogen activity assay using NCI-636688 or (E)-2-octenoyl-NAC as sub- bonds with the ribose hydroxyls. K54 is highly conserved in the + strates are summarized in Table 1. With NCI-636688, the kcat of trans-acting ERs, and the LovC-NADP structure showed that it K54S and T139V are 24- and 35-fold less than WT, respectively, is important for cofactor binding. The highly conserved nature of whereas the kcat of S51A is 2.5-fold greater than WT. Although the NADP-interacting residues indicates that the cofactor-bind- less drastic, the data from (E)-2-octenoyl-NAC show a similar ing motif should be conserved among fungal trans-acting ERs. trend in activity for the mutants compared with WT, supporting that the observed kinetic constants are substrate independent NADPH Stereochemistry. In previous studies of the FAS ER do- and mutation dependent. mains, a transfer of the NADPH pro-4R hydride can result in The increased k of S51A came as a surprise because this is the either syn- (in animal and E. coli) or anti- (in yeast) addition to cat – polar residue nearest the position of the thioester carbonyl in the produce [3R, 2S] or [3R, 2R] reduced fatty acids (27 29). It is docked substrates (and, therefore, a likely candidate for stabili- interesting to note that the cryptic stereochemistry of fatty acid zation of the oxyanion following hydride transfer) (Fig. 4B). The ER and PKS ER can be opposite in the same fungal organism rate increase may be attributable to enhancement of a step other (30, 31). Because the si-face of the cofactor NADPH is blocked than the catalytic step, such as product release. For the pocket by highly conserved active site residues (Fig. S5 B and C), only mutants N263A, and N263S, A93M and S138M, we observed the pro-R hydride of NADPH can be transferred to the sub- a change of relative substrate specificity (Table 1). The result strate. This conclusion is consistent with a recent stereochemical supports that altering the substrate pocket topology by mutagen- study of crotonyl-CoA carboxylase/reductase (32), as well as esis can change substrate specificity (detailed in SI Text)andis previous analyses of nicotinamide-binding proteins (33), that a proof of principle for future endeavor to apply synthetic sub- when NADPH adopt the anti-conformation, the re face of the strate mimics to biosynthesize new analogs. Finally, the in vitro cofactor is exposed and the pro-R hydride is transferred. Thus, reconstitution of DmL biosynthesis by combining LovB, LovC, and the LovC structure provides a structural basis for the stereo- substrates demonstrates that all mutants are enzymatically active specificity of the trans-acting ERs. and capable of interacting productively with LovB (Fig. S6B). The greatly diminished, but not abolished, activity of K54S and Active Site and Substrate-Binding Pocket. The putative LovC sub- T139V is of particular interest. A structurally equivalent lysine to strate-binding pocket can be identified from the LovC-CO7 K54 in SDR enzymes is crucial for cofactor binding (38) and ca- structure, in which the four-carbon crotonoyl group extends into talysis (39), whereas the side chain hydroxyl of T139 lies within 3.2 a hydrophobic pocket that is adjacent to the position of the Å from the C4 of the nicotinamide (the site of hydride transfer) nicotinamide ring of LovC-NADP+ (Fig. 2D and Fig. S5A). The (detailed in SI Text). As a control, we solved the 1.7-Å crystal binding pocket is defined by residues from α1, α3, and the loops structure of LovC-K54S bound with NADP+ and generated the between α1−β4 and β5−β6. At the entrance of the pocket, the S51A/K54S mutant. The LovC K54S structure confirms no sig-

K54-N263 electropositive bridge could serve as a docking point niﬁcant conformational change from that of the wild-type BIOCHEMISTRY for the phosphate in PPT group of the ACP-tethered polyketide (RMSD Cα of 0.21 Å; detailed in SI Results and Discussion and

Ames et al. PNAS | July 10, 2012 | vol. 109 | no. 28 | 11147 Downloaded by guest on September 27, 2021 Fig. 4. (A) Docking results for the α,β unsaturated heptaketide intermediate attached to PPT. (B) Stereoview of the polyketide-binding pocket. (C) General hydride transfer mechanism proposed for LovC.

Fig. S7 A–D). Furthermore, the turnover rate and binding affinity studied to associate with fungal PKSs, we also found eight se- of S51A/K54S displayed additive effect of the single mutants, quence homologs (32–59% identity) in Aspergillus clavatus, six S51A and K54S. Therefore, the low K54S activity is attributed homologs (36–58% identity) in Magnaporthe grisea, twelve directly to the loss of functionality provided by the K54 side chain. homologs (29–46% identity) in Aspergillus niger,andfive homologs The above result agrees with the recent mutational study by the (32–36% identity) in Gibberella zeae. Note that the fungal fatty Leadlay group on the modular type I PKS ER domain, which also acid ER domain has a completely different fold (42). Further- concludes that no single residue serves as oxyanion hole or more, an extensive sequence comparison of LovC with other Brønsted acid (40). Rather, a combination of multiple side- MDR family members confirms that no known alcohol de- chains, waters, or the backbone NH of G282 may all contribute to hydrogenase or quinone reductase share > 20% sequence identity enzyme catalysis, a unique feature observed for PKS ER domains. with LovC. Namely, the trans-acting ER domains form a subfamily Proposed Mechanism for the Observed LovC Substrate Specificity. that is distinct from other MDR family members, as analyzed in details by a recent paper on the MDR family (15). Therefore, Based on the above analyses, we propose the following mechanism – to interpret the observed LovC substrate specificity for the puta- these LovC homologs, with 30 60% of sequence identity to LovC, tive α,β unsaturated tetra-, penta-, and heptaketide intermediates: most likely also serve as trans-acting ER domains that are used for For an ER, the cofactor NADPH is proposed to bind before polyketide biosynthesis. If this is true, these fungi should have not substrate binding (41). LovC then gains proximity to the LovB only possess the trans-acting ER domains, but also the corre- megasynthase through protein–protein interaction (Fig. 3 and sponding PKS megasynthases whose ERs are inactivated. As Figs. S6 and S8), which may occur by forming a heterodimer with presented in the SI Text, the inactivated ER0 domain has a short- the LovB ER0 domain and extra help from interactions with other ened linker region between β3andβ4, resulting in the absence α1 LovB domains, such as ACP. The LovB-ACP is then docked to the which contains important active site residues of MDR proteins electropositive surface of LovC near K54, and the polyketide (including K54 of LovC). This can be used as a diagnostic property substrate is delivered from ACP to the LovC substrate pocket. The when we search for homologs of the LovB-ER0 using the BLAST pro-R hydride of NADPH is then transferred to C3 of the poly- server. As a positive control, the BLAST search identifies known ketide alkene group (Fig. 4C), and the resulting enolate oxide is ER0 domains from the monacolin synthase in Monascus pilosus stabilized by the oxyanion hole (candidate residues include K54 and the compactin synthetase in Penicillium citrinum, both syn- and amide NH of G282). The enolate collapses and proton thases have well-characterized trans-acting ER domains. Signifi- transfer from a Brønsted acid (likely water) to C2 completes enoyl 0 reduction. Substrate docking indicates that because of the pocket cantly, the BLAST search also reveals the existence of ER size limitation, LovC cannot accommodate polyketides larger than domains in PKS or PKS/nonribosomal peptide synthetase megasynthases whose functions are not known yet, such as the six PKSs the heptaketide intermediate (Figs. 1B and 4 A and B). On the 0 other hand, although the smaller di- and triketide intermediates with ER in Aspergillus clavatus (25–30% identity), five in Mag- do fit in the active site, the shorter di- and triketide may adopt naporthe grisea (27–33% identity), two in Aspergillus niger (21–23% multiple, nonproductive orientations, as demonstrated by the identity), and one in Gibberella zeae (21% identity). In all ER0 docking result of these shorter ketides, the double bonds of which homologs, the key helix–loop region between β3andβ4 is short- restrict their flexibility and prevent a proper orientation in the ened, similar to the LovB-ER0 domain. The above analyses sup- LovC active site, thus discouraging productive binding for enoyl port that these megasynthases have inactivated ER0 domains, as reduction. Furthermore, chain elongation by KS may compete for well as corresponding trans-acting ER domains that can produce the shorter ketides. In comparison, the longer tetra- and penta- partially reduced polyketides such as brefeldin, cytochalasin, and ketide intermediates (the first two putative substrates for LovC) equisetin from Aspergillus clavatus, Ace1 from Magnaporthe grisea, can be extended to the back of substrate pocket, promoting more fi fi and zearalenone from Gibberella zeae. In conclusion, the signi - signi cant hydrophobic contact to anchor the polyketide, thus cance of the LovC crystal structure lies in its uniqueness as a trans- resulting in the productive-binding mode. In addition, the hex- acting, monomeric ER domain, and its general applicability to aketide intermediate is strongly driven to undergo Diels–Alder other trans-acting ER domains that react with corresponding cyclization rather than enoyl reduction. In conclusion, LovC structural analyses help explain substrate specificity, provide in- megasynthases with inactivated ER domains. In the future, this sight into the stereochemistry of enoyl reduction, and identify work will facilitate the elucidation of biosynthetic pathways in candidate active site residues important for catalysis. these ER-inactivated megasynthases, as well as an in vitro assay to characterize other trans-acting ER domains. Biological Significance. The above work represents the structural and functional characterization of a trans-acting ER from an it- Methods erative fungal PKS. The LovC structures help identify the possible Protein Expression and Purification. Recombinant MatB, LovC, and LovB were ACP docking site, as well as residues important for substrate expressed and purified as detailed in SI Methods. Selenomethionine- binding and enzyme catalysis. Looking beyond lovastatin bio- substituted LovC protein was produced in E. coli BL21(DE3) using metabolic synthesis, when we submitted the LovC sequence to the BLAST inhibition of the methionine pathway in M9 minimal medium detailed in server, in addition to the trans-acting ER domains that are well SI Methods.

11148 | www.pnas.org/cgi/doi/10.1073/pnas.1113029109 Ames et al. Downloaded by guest on September 27, 2021 Site-Directed Mutagenesis. Site-directed mutagenesis was achieved using the Gel-Filtration Interaction Studies. Samples of purified LovB (10 μM; 100 μL) QuikChange II kit (Stratagene). Mutagenic oligonucleotides are listed in and LovC (10 μM; 100 μL) were incubated for 20 min and then run on SI Methods. a Superdex 200 10/300 GL column (equilibrated with Buffer C), and the 8.9- mL peak fraction representing the LovB-LovC complex was concentrated to μ Cofactor-Binding Assay. An Hitachi 4500 fluorescence spectrometer (50-nm 250 L before SDS/PAGE. slit; 700V) was used to record an emission scan (350–600 nm) with excitation fi wavelength set to 340 nm (slit width 5.0 nm). NADH or NADPH (0.1–50 μM) Crystallization, Data Collection, Phasing, Model Building, and Re nement. was added to 0.1 M potassium phosphate buffer (pH 7.0; 100 μL) with or Cocrystals of LovC with crotonoyl CoA (LovC-CO7), SeMet LovC with crotonoyl CoA (SeMet LovC-CO7), and LovC WT or mutant K54S with cofactor without 50 μM LovC. The fluorescence titration plots (Fig. S3A) were gen- NADP+ (LovC-NADP, LovC K54S-NADP) were grown at 25 °C using the sitting erated from the λ emission at 455 nm. max drop vapor diffusion method. The condition of crystallization and data collection is detailed in SI Methods. Multiwavelength anomalous dispersion Fluorometric Activity Assay. Potential in vitro substrates of LovC were (MAD) methods were used to solve the structure of the SeMet LovC-CO7. obtained from the Developmental Therapeutics Program/National Cancer Details of phasing, model building, and refinement are provided in SI Institute (DTP/NCI) small molecule repository (43) or synthesized as NACs. Methods. Detailed statistics are listed in Table S1. The synthesis and assay conditions are detailed in SI Methods. ACKNOWLEDGMENTS. S.-C.T. thanks the National Institute of General Reconstitution of DmL Biosynthesis. LovB (25 μM) was incubated with 25 μM Medicinal Sciences (NIGMS Grant R01GM076330), the Pew Foundation, and MatB, 25 μM LovC (WT or mutant), 100 mM malonate, 5 mM CoA, 20 mM American Heart Association for support. J.C.V. thanks the Natural Sciences and Engineering Council of Canada and the Canada Research Chair in Bioorganic ATP, 2 mM NADPH, 2 mM S-(5′-adenosyl)-L-methionine chloride (SAM), and and Medicinal Chemistry for support. We appreciate the experimental support 7 mM MgCl2 in buffer [100 mM NaH2PO4 (pH 7.4), 10% (vol/vol) glycerol, provided by Robyn Kaake, Lili Kolozian, Nam Ho, John Leong, BaoChuong Le, 2 mM DTT] at ∼25 °C overnight. Processing of reactions and product de- and Jessica Yang. Portions of this research were performed at the Stanford tection is detailed in SI Methods. Synchrotron Radiation Laboratory and the Advanced Light Source.

1. Tobert JA (2003) Lovastatin and beyond: The history of the HMG-CoA reductase in- 24. Lonnberg T, Kralikova S, Rosenberg I, Lonnberg H (2006) Kinetics and mechanisms for hibitors. Nat Rev Drug Discov 2:517–526. the isomerization of internucleosidic 3 ’-O-P-CH2-5 ’ and 3 ’-O-P-CH(OH)-5 ’ linkages 2. Young F, Capewell S, Ford ES, Critchley JA (2010) Coronary mortality declines in the to their 2 ’,5 ’-counterparts. Collect Czech Chem Commun 71:859–870. U.S. between 1980 and 2000 quantifying the contributions from primary and sec- 25. Jarvinen P, Oivanen M, Lonnberg H (1991) Interconversion and phosphoester hy- – ondary prevention. Am J Prev Med 39:228 234. drolysis of 2’,5′-dinucleoside and 3′,5′-dinucleoside monophosphates - kinetics and 3. Gauthaman K, Fong CY, Bongso A (2009) Statins, stem cells, and cancer. J Cell Biochem mechanisms. J Org Chem 56:5396–5401. 106:975–983. 26. Bellamacina CR (1996) The nicotinamide dinucleotide binding motif: A comparison of 4. Kennedy J, et al. (1999) Modulation of polyketide synthase activity by accessory nucleotide binding proteins. FASEB J 10:1257–1269. proteins during lovastatin biosynthesis. Science 284:1368–1372. 27. Saito K, Kawaguchi A, Seyama Y, Yamakawa T, Okuda S (1981) Steric course of re- 5. Ma SM, et al. (2009) Complete reconstitution of a highly reducing iterative polyketide action catalyzed by the enoyl acyl-carrier-protein reductase of Escherichia coli. Eur synthase. Science 326:589–592. – 6. Xie X, Tang Y (2007) Efficient synthesis of simvastatin by use of whole-cell biocatalysis. J Biochem 116:581 586. Appl Environ Microbiol 73:2054–2060. 28. Anderson VE, Hammes GG (1984) Stereochemistry of the reactions catalyzed by – 7. Auclair K, Kennedy J, Hutchinson CR, Vederas JC (2001) Conversion of cyclic non- chicken liver fatty acid synthase. Biochemistry 23:2088 2094. aketides to lovastatin and compactin by a lovC deficient mutant of Aspergillus ter- 29. Sedgwick B, Morris C (1980) Stereochemical course of hydrogen transfer catalyzed by reus. Bioorg Med Chem Lett 11:1527–1531. the enoyl reductase enzyme of the yeast fatty-acid synthetase. J Chem Soc Chem 8. Chen Y-P, et al. (2008) Cloning and characterization of monacolin K biosynthetic gene Commun (3):96–97. cluster from Monascus pilosus. J Agric Food Chem 56:5639–5646. 30. Rawlings BJ, Reese PB, Ramer SE, Vederas JC (1989) Comparison of fatty-acid and 9. Abe Y, et al. (2002) Molecular cloning and characterization of an ML-236B (com- polyketide biosynthesis - stereochemistry of cladosporin and oleic-acid formation in pactin) biosynthetic gene cluster in Penicillium citrinum. Mol Genet Genomics 267: Cladosporium-Cladosporioides. J Am Chem Soc 111:3382–3390. 636–646. 31. Arai K, Rawlings BJ, Yoshizawa Y, Vederas JC (1989) Biosyntheses of antibiotic 10. Bergmann S, et al. (2007) Genomics-driven discovery of PKS-NRPS hybrid metabolites A26771b by Penicillium-Turbatum and dehydrocurvularin by Alternaria-Cinerariae - – from Aspergillus nidulans. Nat Chem Biol 3:213 217. comparison of stereochemistry of polyketide and fatty-acid enoyl thiol ester reduc- 11. Xu W, Cai X, Jung ME, Tang Y (2010) Analysis of intact and dissected fungal poly- tases. J Am Chem Soc 111:3391–3399. ketide synthase-nonribosomal peptide synthetase in vitro and in Saccharomyces 32. Erb TJ, Brecht V, Fuchs G, Müller M, Alber BE (2009) Carboxylation mechanism and cerevisiae. J Am Chem Soc 132:13604–13607. stereochemistry of crotonyl-CoA carboxylase/reductase, a carboxylating enoyl-thio- 12. Qiao K, Chooi YH, Tang Y (2011) Identification and engineering of the cytochalasin – gene cluster from Aspergillus clavatus NRRL 1. Metab Eng 13:723–732. ester reductase. Proc Natl Acad Sci USA 106:8871 8876. 13. Scherlach K, Boettger D, Remme N, Hertweck C (2010) The chemistry and biology of 33. Benner SA (1982) The stereoselectivity of alcohol dehydrogenases: A stereochemical – cytochalasans. Nat Prod Rep 27:869–886. imperative? Experientia 38:633 636. 14. Eley KL, et al. (2007) Biosynthesis of the 2-pyridone tenellin in the insect pathogenic 34. Jones G, Willett P, Glen RC, Leach AR, Taylor R (1997) Development and validation of fungus Beauveria bassiana. ChemBioChem 8:289–297. a genetic algorithm for flexible docking. J Mol Biol 267:727–748. 15. Riveros-Rosas H, Julian-Sanchez A, Villalobos-Molina R, Pardo JP, Pina E (2003) Di- 35. Airenne TT, et al. (2003) Structure-function analysis of enoyl thioester reductase in- versity, taxonomy and evolution of medium-chain dehydrogenase/reductase super- volved in mitochondrial maintenance. J Mol Biol 327:47–59. family. Eur J Biochem270:3309–3334. 36. Thorn JM, Barton JD, Dixon NE, Ollis DL, Edwards KJ (1995) Crystal structure of Es- 16. Stoop JM, Williamson JD, Conkling MA, Pharr DM (1995) Purification of NAD-dependent cherichia coli QOR quinone oxidoreductase complexed with NADPH. J Mol Biol 249: mannitol dehydrogenase from celery suspension cultures. Plant Physiol 108:1219–1225. 785–799. 17. Maier T, Leibundgut M, Ban N (2008) The crystal structure of a mammalian fatty acid 37. Polyak K, Xia Y, Zweier JL, Kinzler KW, Vogelstein B (1997) A model for p53-induced – synthase. Science 321:1315 1322. apoptosis. Nature 389:300–305. 18. Shimomura Y, Kakuta Y, Fukuyama K (2003) Crystal structures of the quinone oxi- 38. Parikh S, Moynihan DP, Xiao G, Tonge PJ (1999) Roles of tyrosine 158 and lysine 165 in doreductase from Thermus thermophilus HB8 and its complex with NADPH: Impli- the catalytic mechanism of InhA, the enoyl-ACP reductase from Mycobacterium tu- cation for NADPH and substrate recognition. J Bacteriol 185:4211–4218. berculosis. Biochemistry 38:13623–13634. 19. Persson B, Hedlund J, Jörnvall H (2008) Medium- and short-chain dehydrogenase/re- 39. Tanaka N, et al. (1996) Crystal structures of the binary and ternary complexes of 7 ductase gene and protein families: The MDR superfamily. Cell Mol Life Sci 65: α-hydroxysteroid dehydrogenase from Escherichia coli. Biochemistry 35:7715–7730. 3879–3894. 40. Kwan DH, Leadlay PF (2010) Mutagenesis of a modular polyketide synthase enoylreductase 20. Hart GJ, Dickinson FM (1983) The coenzyme-binding characteristics of highly purified domain reveals insights into catalysis and stereospecificity. ACS Chem Biol 5:829–838. preparations of sheep liver cytoplasmic aldehyde dehydrogenase. Biochem J 211:363–371. 41. Liu N, Cummings JE, England K, Slayden RA, Tonge PJ (2011) Mechanism and in- 21. Chen JH, Linstead E, Swamidass SJ, Wang D, Baldi P (2007) ChemDB update—full-text search and virtual chemical space. Bioinformatics 23:2348–2351. hibition of the FabI enoyl-ACP reductase from Burkholderia pseudomallei. J Anti- 22. Lonnberg H, Stromberg R, Williams A (2004) Compelling evidence for a stepwise mechanism microb Chemother 66:564–573. of the alkaline cyclisation of uridine 3′-phosphate esters. Org Biomol Chem 2:2165–2167. 42. Jenni S, et al. (2007) Structure of fungal fatty acid synthase and implications for it- 23. Maki E, Oivanen M, Poijarvi P, Lonnberg H (1999) Buffer-catalyzed interconversion of erative substrate shuttling. Science 316:254–261. ribonucleoside 2’/3’-methylphosphonates and 2’/3’-alkylphosphates. J Chem Soc Per- 43. Monga M, Sausville EA (2002) Developmental therapeutics program at the NCI: Mo- kin Trans 2 2493–2499. lecular target and drug discovery process. Leukemia 16:520–526. BIOCHEMISTRY

Ames et al. PNAS | July 10, 2012 | vol. 109 | no. 28 | 11149 Downloaded by guest on September 27, 2021