Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2020 Aug 6;5(32):20548–20557. doi: 10.1021/acsomega.0c02776

Nonspecific Heme-Binding Cyclase, AbmU, Catalyzes [4 + 2] Cycloaddition during Neoabyssomicin Biosynthesis

Qinglian Li †,*, Wenjuan Ding †,‡, Jiajia Tu †, Changbiao Chi §, Hongbo Huang †, Xiaoqi Ji †,‡, Ziwei Yao †,‡, Ming Ma §, Jianhua Ju †,‡,*
PMCID: PMC7439702  PMID: 32832808

Abstract

graphic file with name ao0c02776_0006.jpg

Diels–Alder (DA) [4 + 2]-cycloaddition reactions rank among the most powerful transformations in synthetic organic chemistry; biosynthetic examples, however, are few and far between. We report here a heme-binding cyclase, AbmU, that catalyzes an essential [4 + 2] cycloaddition during neoabyssomicin scaffold assembly. In vivo genetic and in vitro biochemical analyses strongly suggest that AbmU catalyzes an intramolecular and stereoselective [4 + 2] cycloaddition to form a spirotetronate skeleton from an acyclic substrate featuring both a terminal 1,3-diene and an exo-methylene group. Biochemical assays and X-ray diffraction analyses reveal that AbmU binds nonspecifically to a heme b cofactor and that this association does not play a catalytic role in AbmU catalysis. A detailed study of the AbmU crystal structure reveals a unique mode of substrate binding and reaction catalysis; His160 forms a H-bond with the C-1 carbonyl O-atom of the acyclic substrate, and the imidazole of the same amino acid directs the tetronate moiety of acyclic substrate toward the terminal Δ10,11, Δ12,13-diene moiety, thereby facilitating intramolecular DA chemistry. Our findings expand upon what is known about mechanistic diversities available to biosynthetic [4 + 2] cyclases and help to lay the foundation for the use of AbmU in possible industrial applications.

Introduction

[4 + 2]-Cycloaddition, in which a 1,3-diene and a dienophile react to form a cyclohexene unit, is a major reaction widely used to regio- and stereoselectively construct carbon–carbon bonds in organic synthesis.1 This reaction has also long been postulated to be a key transformation in the biosynthesis of many natural products. However, only a small number of standalone natural enzymes, whose sole function is to catalyze [4 + 2] cycloadditions like the Diels–Alder (DA) reaction, have been identified.2 These DA enzymes (DAases) include SpnF in spinosyn,3,4 VstJ in versipelostatin,5 PyrI4 and PyrE3 in pyrroidomycin,6,7 AbyU in abyssomicin C,8 and MycB in the myceliothermophin biosynthetic pathways.9 Single enzymes catalyzing hetero-Diels–Alder reactions (HDAs) have also been reported in thiopeptide pathways (e.g., TclM in thiocillins10 and TbtD in thiomuracin11,12). Among these standalone DAases, only PyrE3 and SpnF are known to contain a cofactor; PyrE3 contains a noncovalently bound FAD as a cofactor, playing an essential structural role in catalysis.6,13 Although structural studies have revealed SpnF to bind S–adenosyl-l-homocysteine (SAH) in vitro, S-adenosyl-l-methionine (SAM) is believed to be a structural factor of SpnF in vivo.4 In addition, a multifunctional enzyme in the leporine biosynthetic pathway, LepI,14 catalyzes a stereoselective dehydration followed by three pericyclic transformations including intramolecular DA and HDA reactions and has been reported to contain SAM as a functional cofactor.

The abyssomicin class of natural products belong to the class I spirotetronate polyketides containing a signature spirotetronate moiety within a macrocycle of variable size. Thus far, ∼40 abyssomicins have been isolated from both Verrucosispora and Streptomyces. The stereochemical configuration of the spirotetronate C-15 center forms the basis for classification of these compounds as either type I abyssomicins or their “enantiomeric” type II counterparts (Figures 1A and S1).15,16 The antibacterial type I abyssomicin derivative, abyssomicin C (Figure 1B), was isolated from Verrucosispora maris AB-18-032;17 functional and structural studies have revealed AbyU to be a cofactor-independent, standalone DAase that catalyzes a stereoselective intramolecular [4 + 2] cycloaddition to install the spirotetronate moiety during abyssomicin C biosynthesis.8 Ten type II abyssomicin derivatives, including neoabyssomicin B (1), abyssomicins 2 (2), and 4 (3) (Figures 1A and S1), have been isolated from the deep sea-derived Streptomyces koyangensis SCSIO 5802 by our group.15,16 The biosynthetic gene cluster of these compounds (termed abm, Figure 1C) has been reported and genetically verified.18 We have demonstrated that AbmV, a P450 monooxygenase, triggers a Domino reaction sequence, leading to installation of both bridged ether and C-11 hydroxylation moieties seen in abyssomicin 2 (2),19 and that a luciferase-like monooxygenase AbmE2 and a flavin reductase AbmZ constitute a type II Baeyer–Villiger (BV) monooxygenase pair catalyzing BV oxidation as a post-PKS tailoring step during neoabyssomicin/abyssomicin biosynthesis (Figure 1D).20 In this work, we demonstrate that the formation of the spirotetronate moiety in type II abysomicins 1–3 is catalyzed by a heme-binding [4 + 2] cyclase, AbmU (Figure 1D). Although AbmU was unexpectedly found to bind a heme b cofactor, biochemical assays and elucidation of the crystal structure of AbmU suggest that the heme binds nonspecifically to AbmU, and this association does not play a catalytic role in AbmU-catalyzed chemistry. Furthermore, a detailed study of the AbmU crystal structure reveals a unique mode of substrate binding and reaction catalysis.

Figure 1.

Figure 1

Open in a new tab

Chemical structures of representatives of abyssomicins and the role of AbmU in the biosynthetic pathway of neoabyssomicins. (A) Chemical structures for representatives of type II abyssomicins; 1–3 were isolated from S. koyangensis SCSIO 5802. (B) Chemical structures of representatives of type I compounds. (C) Genetic organization of the neoabyssomicin/abyssomicin biosynthetic gene cluster in S. koyangensis SCSIO 5802. The [4 + 2] cyclase is shaded in brown. (D) Proposed pathway for neoabyssomicin/abyssomicin construction in S. koyangensis SCSIO 5802; the AbmU-catalyzed transformation has been validated in this work.

Results and Discussion

Functional Assignment of abmU

To identify the putative gene responsible for [4 + 2] cycloaddition in neoabyssomicin biosynthesis, we first performed homology searching of the abm gene cluster using AbyU as a probe by BLASTP. These searches unveiled the hypothetical abmU gene (657 bp) as an excellent candidate. The abmU gene encodes a protein showing 37% sequence identity to AbyU (Figure S2). Furthermore, structural homology searching for AbmU using the online program HHpred revealed that AbmU showed structural homology to the characterized enzymes, AbyU and PyrI4. Both AbyU and PyrI4 are known to catalyze [4 + 2] chemistry.7,8 These bioinformatic analyses suggested that AbmU is likely involved in catalyzing [4 + 2] cycloaddition during neoabyssomicin biosynthesis.

In Vivo Gene Inactivation of abmU

To explore the function of abmU in neoabyssomicin biosynthesis, we inactivated this gene in wild-type (WT) S. koyangensis SCSIO 5802 using established λ-RED-mediated PCR-targeting methodology (Figure S3). HPLC analysis of the ΔabmU mutant strain fermentation revealed that abmU inactivation abolished the production of all relevant neoabyssomicins/abyssomicins (1–3) (Figure 2, trace ii). Complementation of the ΔabmU mutant with the abmU in trans restored production of the full set of metabolites characteristic of the WT strain (Figure 2, trace iii). These results confirmed the essential role of abmU in neoabyssomicin/abyssomicin biosynthesis. In addition, fermentation of the ΔabmU mutant and subsequent extract analysis revealed a new metabolite (5) (Figure 2, trace ii). Purification of 5 from a 10 L culture broth of the ΔabmU mutant followed by HRMS and extensive NMR analyses (Table S1 and Figures S4–S6) revealed that 5 is a 19-carbon linear polyene polyketide (Figure 1D). This compound bears a terminal conjugated Δ8,9, Δ10,11, Δ12,13-triene and a Δ14,15exo-methylene group; the C8–C9, C10–C11, and C12–C13 olefins of the Δ8,9, Δ10,11, Δ12,13-triene were all identified as bearing the E configuration (Figure 1D). The accumulation of 5 in the ΔabmU mutant strain firmly supported the notion that AbmU catalyzes stereoselective intramolecular [4 + 2] cycloaddition between the terminal conjugated diene and the conjugated exo-methylene moiety in 5 to install the spirotetronate moiety during neoabyssomicin/abyssomicin biosynthesis.

Figure 2.

Figure 2

Open in a new tab

HPLC analyses of fermentation broths: (i) WT S. koyangensis SCSIO, (ii) ΔabmU mutant, and (iii) ΔabmU::abmU.

In Vitro Production of Spirocyclic Core of Neoabyssomicins Using Purified AbmU

To further confirm the role of abmU in neoabyssomicin/abyssomicin biosynthesis, AbmU was overexpressed as an N-terminal His6-tagged protein in Escherichia coli and purified to near-homogeneity (Figure S7). The activity of purified AbmU was then assayed using linear intermediate 5 as a substrate. Incubation of 5 with purified AbmU at 30 °C for 10 min led to rapid generation of sole product 4 (Figure 3C, trace vi); this newly formed species was found to have the same HPLC retention time and molecular mass as those of authentic abyssomicin 6 (4) (see Figure S8), prepared from fermentation of the ΔabmV mutant. We had previously demonstrated that the cytochrome P450 enzyme, AbmV, is responsible for converting spirocyclic intermediate abyssomicin 6 (4) into abyssomicin 2 (2) via tandem ether installation and hydroxylation reactions (Figure 1D).19 It should be noted that product 4 was never detected in the control reaction lacking AbmU (Figure 3C, trace ii). These results demonstrate that AbmU is responsible for the [4 + 2] cycloaddition of linear intermediate 5. Steady-state kinetic characterization of the AbmU-catalyzed [4 + 2] cycloaddition gives a KM of 51.24 ± 8.61 μM and a kcat of 37.60 ± 1.42 min–1 (Figure S9).

Figure 3.

Figure 3

Open in a new tab

AbmU is a heme-associated enzyme. (A) UV–vis spectra of purified (nonreduced) AbmU, dithionite-reduced AbmU, and dithionite-reduced AbmU treated with carbon monoxide. (B) UV–vis spectra of purified AbmU and its H160A mutant (200 μM of each enzyme). (C) HPLC analyses of in vitro reactions of substrate 5 with AbmU as well as its H160A mutant, λ = 254 nm.

Purified AbmU Binds a Heme b Cofactor

Unexpectedly, the purified AbmU appeared light brown in color (Figure S7), consistent with the presence of heme. Notably, no such observation was ever reported during characterization of AbyU.8 To specifically assay for a putative heme cofactor, purified AbmU was first subjected to UV–visible (UV–vis) spectroscopic analyses, which revealed an absorbance maximum (λmax) at 413 nm (Figure 3A), typical of a heme cofactor.21,22 To further identify the heme type, we then conducted a pyridine hemochrome assay on AbmU and found that its cofactor has the characteristic spectrum of heme b (Figure S10A).23 To determine the oxidation state of the heme iron in purified AbmU, we added the oxidizing reagent potassium ferricyanide and the reducing reagent sodium dithionite to the protein solution, respectively, and observed changes in the absorption spectra. Addition of ferricyanide did not alter the spectrum of AbmU (Figure S10B), whereas addition of dithionite resulted in a shift of λmax from 413 to 425 nm (Figure 3A). Collectively, these data demonstrated that the purified AbmU binds an oxidized, ferric heme b cofactor.

Although the purified AbmU was demonstrated to contain a heme b, we noted that the brown color of purified AbmU is very light, consistent with its low UV–vis absorbance reading (Figure S11A). The heme content of AbmU was determined to be only ∼1.5% by pyridine hemochrome assay even when the heme precursor 5-aminolevulinic acid (5-ALA) was supplemented into the LB medium during AbmU overexpression (see Experimental Section, Figure S11A). This low level of heme incorporation for AbmU is inconsistent with that of the typical heme-dependent enzymes for which heme binding is biologically relevant.22 To determine whether the heme cofactor functions during the AbmU-catalyzed [4 + 2] cycloaddition, we first tested whether the heme is critical to AbmU [4 + 2] cycloaddition activity. We compared the activity of AbmU purified from minimal LB medium (possessing 0.1% heme content) with that purified from LB medium supplemented with the heme precursor 5-ALA (possessing 1.5% heme content); no obvious differences in catalytic efficiencies were noted between the two proteins (Figure S11B).

We next sought to interrogate whether the heme iron oxidation state is important in the AbmU-mediated cycloaddition. We prepared a reduced ferrous version of AbmU by subjecting purified AbmU to 1 mM sodium dithionite in Tris-HCl buffer (pH 8.0). Linear species 5 was then added to the putatively reduced version of AbmU, and the reaction was incubated at 30 °C for 10 min. HPLC analyses of the reaction mixture revealed that substrate 5 was completely consumed and that a new product 6, distinct from spirocylic 4, was generated (Figure S12A, trace iv). Given that linear substrate 5 was stable to the reducing conditions of the AbmU reaction (Figure S12A, trace iii), we speculated that 6 results from dithionite-mediated sulfation of AbmU-catalyzed cycloadduct 4. To test this hypothesis, we subjected standard spirocyclic 4 to the same reducing reaction conditions (devoid of AbmU) and found that, indeed, 4 was converted to 6 (Figure S12A, trace vii) as evidenced by identical HPLC and MS data for the two differently prepared samples of 6. Adduct 6 was proposed to be spontaneously derived from the AbmU-catalyzed cycloadduct 4 via Michael-type addition of sulfite to the C8–C9 olefin. This logic was supported by the chemical formula C19H24O8S as assigned to 6 on the basis of LC-HRESIMS data (m/z 413.1270 [M + H]+, calcd 413.1265) (Figure S12B) and 1H NMR spectral data (Figure S12C–E). Consequently, these data revealed that AbmU is able to employ either heme iron oxidation state (+2 or +3) during the course of [4 + 2] cycloaddition catalysis.

Because many heme-dependent enzymes require O2 for activity, we next evaluated the prospect that AbmU requires O2 to ensure [4 + 2] cycloaddition activity. Incubation of linear intermediate 5 with purified AbmU (nonreduced) under anaerobic conditions resulted in the generation of spirocyclic product 4 (Figure S13, trace ii). Conversely, incubation of 5 with dithionite-reduced AbmU under anaerobic conditions afforded sulfite adduct 6 (Figure S13). That both anaerobic reaction systems readily afforded [4 + 2] cycloaddition adducts highlights the O2 independence of AbmU-catalyzed [4 + 2] cycloaddition chemistry.

We next generated a series of point mutations in efforts to identify which amino acid likely acts as the axial heme ligand. Generally speaking, the most common biologically relevant axial ligating atoms for iron in heme proteins are histidine (His) N atoms, methionine (Met) or cysteine (Cys) sulfur groups, and the phenolic oxygen of tyrosine (Tyr).24 Therefore, a total of 17 residues, including seven His residues (H10, H48, H61, H99, H160, H166, and H176), three Met residues (M101, M102, and M146), and seven Tyr residues (Y53, Y69, Y92, Y103, Y104, Y143, and Y163), were each mutated to nonpolar residues (alanine, phenylalanine, or leucine). All 17 mutant proteins were overexpressed and purified from E. coli. UV–vis absorption spectroscopic analysis of these purified proteins revealed that the purified H160A protein was colorless (Figure S7) and displayed sharply reduced heme-binding activity (Figure 3B), whereas the other 16 mutant proteins appeared to bind heme in a manner similar to WT AbmU. These results suggest that His160 acts as the axial ligand involved in heme binding. Given our biochemical observations that AbmU activity is independent of heme content, heme iron oxidation state, and O2 availability, we hypothesized that heme binding plays a nonessential role in AbmU catalysis. However, in vitro enzymatic activity assays of the AbmU mutants revealed that the H160A mutant (devoid of heme binding) was dramatically impaired in its ability to convert linear 5 to cycloadduct 4 (Figure 3C, trace viii). The 16 other AbmU mutants, relative to unmodified AbmU, showed no such signs of catalytic inhibition (Figure S14).

Crystal Structure of AbmU

In efforts to better understand AbmU structure and catalysis, we solved the crystal structure of AbmU to 2.5 Å resolution (Table S2). The crystal structure of AbmU was found to contain four molecules in one asymmetric unit, with each set of two molecules forming a homodimer (Figure 4A). The overall structure of AbmU possesses an antiparallel β barrel core containing eight β strands (β1−β8 with β1 split into β1a and β1b) and one α helix (α2 between β5 and β6) (Figure 4B). The structural similarity survey based on the β barrel structure of AbmU revealed that AbmU’s closest structural homologues are PyrI4 (PDB ID: 5BU3)7 and AbyU (PDB ID: 5DYV)8 known to catalyze [4 + 2] chemistry (Figure 4C,D). Although the overall β barrel core of AbmU is similar to those found in PyrI4 and AbyU, there are a number of critical structural departures from these proteins. In addition to the β barrel core, AbmU contains one N-terminal α helix (α1, Pro8-Ser29), one C-terminal α helix (α3, Asp180–Phe194), and a long C-terminal loop. In the complex structure of PyrI4 with its product, the N-terminal α0 of PyrI4 has been shown to constitute part of the active channel by stacking on the concave rim of the β barrel (Figure 4D).7 However, the N-terminal α1 of AbmU is stabilized by interactions with β1a, β6−β8, and the C-terminal α3 (Figure 4B), excluding the possibility that α1 of AbmU interacts with its ligand/substrate at the active channel as the α0 helix of PyrI4 does.7 The C-terminal α3 of AbmU, which is absent in AbyU8 and PyrI4,7 interacts with the N-terminal α1, β5 and β6 (Figure 4B). Following α3, a long C-terminal loop takes part in continuous interactions from the middle (β5 and β6) to the top (the loop between β3 and β4) of the β barrel (Figure 4B). This C-terminal loop appears to function like a “hand” to bend the β3/β4 loop back, thus creating sufficient space at the top of the β barrel to accommodate substrates (Figure 4B). The N-terminal α1 and C-terminal α3 also play key roles in the dimerization of AbmU, as exemplified by interactions with their counterparts of an adjacent monomer (Figure 4A). This dimerization motif is distinct from those of AbyU and PyrI4, whose dimerizations are driven by β strand → β strand or β strand → α helix interactions, respectively.

Figure 4.

Figure 4

Open in a new tab

Crystal structure of AbmU and modeling of substrate and product. (A) The crystal structure of AbmU shows a homodimer. The two monomers were colored with red and yellow. (B) The overall structure of AbmU shows the β barrel, the N-terminal α1, C-terminal α3, and the C-terminal loop. (C) Structural comparison of AbmU (red) and AbyU (cyan). (D) Structural comparison of AbmU (red) and PyrI4 (gray). (E) The salt bridge formed by Glu47 and Arg150 (cyan) at the bottom of the active site, and the hydrophobic cavity formed by nonpolar residues (yellow) inside the β barrel. (F) The entrance rim of β barrel formed by nonpolar residues (yellow) from different secondary structures. (G) Modeling of substrate 5 (cyan) into the active site; H-bonds were labeled with yellow dashed lines. (H) Modeling of product 4 (magenta) in the active site relative to substrate 5 (cyan). Notably, we were unable to obtain AbmU crystals with detectable heme associations under various conditions.

The channel of the β barrel in AbmU is sealed by a salt bridge between Glu47 and Arg150. At the top of this salt bridge, a large hydrophobic cavity constituting the AbmU active site is formed by the side chains of Val49, Leu51, Tyr69, Leu84, Met102, Tyr104, Val121, Ile126, Gln132, Phe134, Phe152, and Tyr163 (Figure 4E). The entrance rim of the β barrel is formed by hydrophobic interactions between the side chains of Leu51 and Tyr53 from the β1b strand, Phe55 and His61 from the β1/β2 loop, Leu100 from the β4 strand, Leu127 from the α2 helix, and His160 from the β7/β8 loop (Figure 4F). Notably, there is no detectable electron density for heme b around His160, despite earlier evidence suggesting its role as the axial ligand involved in heme binding; this was the case despite extensive efforts to soak and crystallize AbmU in the presence of heme under various conditions. Moreover, a close inspection of the AbmU crystal structure revealed that the region proximal His160 is insufficient to accommodate occupancy by heme b. These results are consistent with our biochemical analyses and support the notion that heme–AbmU associations are nonspecific and nonessential for DA chemistry. Heme has been reported to nonspecifically associate with many proteins including skeletal muscle proteins,25 circadian clock proteins PER2,26 and even bovine serum albumin (BSA). These nonspecific interactions are driven by the low solubility of free heme and often take place at His and Cys residues; they also often involve surface exposed hydrophobic domains.26 Importantly from the AbmU perspective, these facts help explain our experimental observations since His160, and its placement within AbmU, meets these criteria.

Substrate and Product Modeling Reveals Possible Catalytic Mechanism of AbmU

DA catalysis by AbmU was further investigated by modeling substrate 5 and product 4 into the AbmU active site with AutoDock 4.2.27 Docking of 5 into the active site reveals that the tetronate moiety likely interacts with several residues; notably, H-bonds are clearly evident between the C-1 carbonyl group and His160, the C-1 lactone oxygen and Tyr163, and the C-16 hydroxy group and Asp67/Tyr104 (Figures 4G and S15A). Other elements of 5 appear to be in alignment with the large hydrophobic cavity of the β barrel, with the terminal Δ8,9, Δ10,11, Δ12,13-triene moiety stabilized tightly by the hydrophobic interactions with Tyr69, Ile90, Met102, Val121, Ile126, Gln132, Phe134, and Phe152 (Figures 4G and S15B). Among the residues that interact with 5, His160 appears to play a key role. Besides forming a H-bond with the C-1 carbonyl O atom, the imidazole moiety of His160 closely approaches the dienophilic tetronate moiety of 5, pushing it toward the terminal Δ10,11, Δ12,13-diene moiety. These His160–substrate interactions appear to induce a “U” turn at the C5/C6 bond of 5 and effectively bury 5 into the active channel (Figure 4G). This observation is consistent with the results of Figure 3C (trace viii), wherein conversion of His to Ala at site 160 almost completely abolished AbmU’s cycloaddition activity. Notably, neither AbyU nor PyrI4 contain His residues in the β7/β8 loops. Instead, the dynamic N-terminal α0 of PyrI4 traps the flexible substrate in the rigid β barrel core, thus inducing a similar substrate orientation to that seen in the β7/β8 loops of AbmU; it turns out that Arg9 from α0 of PyrI4 is essential for the [4 + 2] cycloaddition activity of Pyr14. Thus, although AbmU and Pyr14 share some similarities with respect to substrate positioning strategies, the two systems use clearly different overall approaches to set up respective cycloaddition reactions.

Modeling of product 4 into the AbmU active site showed that the C-6 through C-13 fragment of 4 overlays well with that of 5, but that the tetronate moiety of 4 is rotated about 60° clockwise relative to the analogous portion of 5 (Figure 4H). This further supports the idea that [4 + 2] cycloaddition is enabled by movement of the tetronate moiety toward the highly stabilized Δ8,9, Δ10,11, Δ12,13-triene moiety in the hydrophobic cavity. In light of these observations, we propose an overall mechanism for AbmU-catalyzed cycloaddition as follows: (i) the Δ8,9, Δ10,11, Δ12,13-triene moiety of 4 first enters the active channel of AbmU and is stabilized by tight hydrophobic interactions with multiple nonpolar residues, (ii) the tetronate moiety of 4 is then projected toward the Δ10,11, Δ12,13-diene moiety by interactions dominated by His160; in particular, H-bonds between the substrate C-1 carbonyl group and His160, the C-1 lactone oxygen and Tyr163, and C-16 hydroxy group with Asp67/Tyr104 ensure that facial alignment of the tetronate ring with the Δ10,11, Δ12,13-diene moiety enables stereoselective [4 + 2] cycloaddition, giving rise to 4.

Conclusions

We have demonstrated herein that AbmU is a cyclase that catalyzes the stereoselective [4 + 2]-cycloaddition responsible for assembling the spirotetronate skeleton of the neoabyssomicins/abyssomicins in S. koyangensis SCSIO 5802. The tandem accumulation of a 19-carbon linear polyketide 5 (bearing both a diene and dienophile) and the absence of neoabyssomicins/abyssomicins in fermentations of the engineered ΔabmU mutant strain suggested early on the role of AbmU as a DAase. This hypothesis was validated in in vitro experiments where linear 5 was enzymatically converted to spirotetronate abyssomicin 6 (4) using recombinant AbmU. Although the purified AbmU was unexpectedly found to bind a heme b cofactor, both structural and biochemical analyses revealed the heme–AbmU association to be nonspecific partly dictated by His160 and the hydrophobic core of the AbmU β barrel. Interestingly, the nonspecifically bound heme plays no direct role in catalysis. X-ray diffraction analyses revealed that although the overall β barrel core of AbmU is similar to previously reported spirotetronate cyclases PyrI47 and AbyU,8 AbmU displays clearly different motifs for dimerization, substrate binding, and catalysis. Compared to AbyU and PyrI4, AbmU contains a distinct N-terminal α helix (α1, Pro8-Ser29), a C-terminal α helix (α3, Asp180-Phe194), and a long C-terminal loop; these distinct features very likely endow this DAase the structural stabilization necessary for substrate binding and/or AbmU dimerization. AbmU and PyrI4 appear to differ radically in how they impose substrate geometries needed for cycloaddition. Whereas PyrI4 employs a substrate trapping mechanism in which the N-terminal tail acts as a form of lid to impose conformational constraints on the β barrel catalytic core,7 the substrate localizing/trapping mechanism employed by AbmU uses the imidazole side chain of His160 to project the tetronate moiety of substrate 5 toward the terminal Δ10,11, Δ12,13-diene of the same substrate molecule. We posit that the AbmU-mediated proximity of each reactive moiety within 5 is critically enabling with respect to the ensuing intramolecular [4 + 2] cycloaddition. In total, our findings with AbmU dramatically expand our insights into Diels–Alderase enzymes and also lay an excellent foundation for future applications of AbmU and related enzymes as useful biocatalysts with industrial potential.

Experimental Section

General Procedures

Strains and plasmids used and generated in this study are listed in Table S3. The culture conditions for S. koyangensis SCSIO 5802 and its mutants have been previously described.15E. coli including BW25113/pIJ790, ET12567/pUZ8002, as well as culture conditions have also been previously described.28

The enzymes for genetic manipulation were purchased from New England Biolabs Inc. (MA). Plasmid, gel extraction, and cycle-pure kits were acquired from Omega Bio-Tek, Inc. (GA). Sodium dithionite and potassium ferricyanide were purchased from Sigma-Aldrich. Unless otherwise stated, other biochemical reagents were purchased from standard commercial sources.

All DNA isolation and manipulation procedures in E. coli and Streptomycetes were performed according to standard procedures29 or the manufacturer’s protocol. Primers were synthesized at Sangon Biotech Co., Ltd. (Shanghai, China) and are listed in Table S4. DNA sequencing was carried out at IGE Biotech Co. Ltd. (Guangzhou, China).

High-performance liquid chromatography (HLPC) analysis was performed using an Agilent Technologies 1260 Infinity system equipped with a DAD detector. Preparative HPLC was performed on an Agilent 1260 system. HPLC electrospray ionization MS (HPLC-ESI-MS) was performed on the amaZon SL ion trap mass spectrometer (Bruker). NMR spectra were recorded on an Avance 500 MHz spectrometer (Bruker). UV spectra were measured with a UV-2600 spectrometer (Shimadzu). ECD spectra were recorded with a Chirascan spectrometer (Applied Photophysics Limited). High-resolution mass spectra were obtained on a Maxis quadrupole-time-of-flight mass spectrometer (Bruker).

In-Frame Deletion of abmU

The inactivation of abmU in S. koyangensis SCSIO 5802 was performed by in-frame deletion to exclude polar effects on downstream gene expression. The in-frame deletion of abmU in S. koyangensis was achieved by a conjugative transfer of the disruption plasmid from E. coli ET12567/pUZ8002 and Streptomyces according to general protocols.30 The disruption plasmid was constructed using λ-RED-mediated PCR-targeting technology.31 The SuperCos I-based S. koyangensis SCSIO 5802 genomic cosmid library has been constructed and reported previously,18 and cosmid 7-6F (containing abmU gene) was selected and introduced into E. coli BW25113/pIJ790. The apramycin resistance gene cassette (aac(3)IV-oriT) was amplified by PCR from plasmid pIJ773 with designed primers (with SpeI restriction site between the 19 or 20 nt matching the right or left end of the disruption cassette and the 39 nt homologous to abmU gene) listed in Table S4. The apramycin resistance cassette was then introduced into E. coli BW25113/pIJ790/7-6F by electroporation to replace 459 bp of the coding region of abmU in the cosmid 07-6H using λ-RED-mediated recombination, yielding mutagenized plasmid 7-6F-UKO. To remove the apramycin resistance gene cassette, mutagenized plasmid 7-6F-UKO was digested with SpeI and then religated; the ligated mixtures were transformed into E. coli DH5α. The resulting transformants that were apramycin-sensitive were subjected to PCR amplification to examine the loss of apramycin resistance gene cassette, generating mutagenized plasmid 7-6F-UKOIF. To enable the conjugative transfer of plasmid 7-6F-UKOIF from E. coli ET12567/pUZ8002 into Streptomyces, 7-6F-UKOIF was modified by replacing the kanamycin resistance gene neo with the apramycin resistance gene cassette (aac(3)IV-oriT) using PCR targeting and λ-RED-mediated recombination, generating 7-6F-UKOIF-Apr. The conjugative transfer of plasmid 7-6F-UKOIF-Apr into S. koyangensis SCSIO 5802 was carried out as described previously.18 The single-crossover mutant strain was obtained from antibiotic selection (apramycinR) and was cultured for two rounds on the antibiotic-free medium. The double-crossover mutant strain was first selected by showing the apramycinS phenotypes and further confirmed by PCR verification, as judged by a 904 bp desired product using the primers abmU-TF and abmU-TR (Figure S3).

In Trans Complementation of abmU into the ΔabmU Mutant Strain

The coding regions of abmU were amplified by PCR using the genomic DNA of S. koyangensis SCSIO 5802 WT strain and the primers abmU-comF and abmU-comR. The PCR products were digested with NdeI/SpeI and then cloned into the same digested sites of a pSET152-derived expression plasmid, pL646,32 under the control of a strong constitutive promoter ermE*p. The recombinant plasmids pL646-abmU were transformed into E. coli ET12567/pUZ8002 and then transferred into the ΔabmU mutant strain by conjugation. The exconjugants were selected on the basis of phenotypes showing apramycin resistance and then confirmed by PCR to give the recombinant strain ΔabmU::abmU.

Metabolite Analyses of WT S. koyangensis SCSIO 5802 and Related Derivative Strains

A portion of mycelium and spores (1 cm2) for each strain was added to 250 mL flasks containing 50 mL of RA medium15 and 1% XAD-16 macroporous resin. Fermentations were then carried out at 28 °C on rotary shakers (200 rpm) for 8 days. After fermentation, each fermentation culture was centrifuged to yield the supernatant and pellet. Pellets, including the mycelium and macroporous resin, were extracted with 50 mL of MeOH, and the MeOH solvent was removed under reduced pressure to afford an oily residue. Residues were each dissolved in 1 mL of MeOH and centrifuged at 13 000 g for 10 min; supernatants were subjected to HPLC-UV analyses, each of which was performed using an Agilent Technologies 1260 Infinity system using a Phenomenex ODS column (150 × 4.6 mm2, 5 μm), eluting with a linear gradient of 0–65% solvent B (solvent B: CH3CN + 0.1% trifluoroacetic acid (TFA); solvent A: H2O + 0.1% TFA) over 20 min, followed by 65–100% solvent B in 2 min, and then 100% solvent B for 5 min, at a flow rate of 1 mL/min. In all cases, chromatograms employed UV detection at 254 nm.

Large-Scale Fermentation, Isolation, and Structural Elucidation of Compound 5

To isolate the metabolite 5, large-scale fermentation of the ΔabmU mutant was carried out using a two-step fermentation process. First, a suitable portion of spore and mycelium (∼1 cm3) was inoculated into a 250 mL flask containing 50 mL of RA medium. After incubation for 36 h at 28 °C on a rotary shaker at 200 rpm, each of the seed cultures was trans-inoculated into a 1 L flask containing 200 mL of RA medium with supplemental 2% XAD-16 resin. Fermentations were then carried out at 28 °C on rotary shakers (200 rpm) for another 8 days. The fermentation cultures (10 L) were centrifuged at 4000 rpm for 10 min to yield the supernatant and pellet. HPLC analyses revealed that the supernatant extract did not contain the target compound. Therefore, only the pellet (including XAD-16 resin and mycelia cake) was extracted with 2.5 L of acetone three times and evaporated to dryness. The pellet extract was subjected to normal-phase silica gel column chromatography (100–200 mesh) using a gradient elution with a CHCl3/CH3OH mixture (100:0, 98:2, 96:4, 94:6, 92:8, 90:10, 80:20, 50:50, and 0:100, v/v, each solvent combination used was 800 mL in total volume) to yield nine fractions (Fr.1–Fr.9). Fr.3 and Fr.4 were further purified by preparative HPLC using an Innoval ODS column (250 × 21.2 mm, 5 μm), eluted with a linear gradient of 20–70% solvent B (solvent B: CH3CN; solvent A: ddH2O) over 30 min at a flow rate of 10 mL/min and using UV detection at 254 nm, to give compound 5 (yield: ∼8 mg/L). Compound 5, white solid; high-resolution-electrospray ionization-mass spectrometry (HR-ESI-MS) (m/z): 329.1396 [M – H]− (calculated: 329.1394), see Figure S4; 1H and 13C NMR data, see Table S1; NMR spectra, see Figures S5 and S6.

Cloning and Expression of Recombinant Proteins

The abmU gene was amplified from S. koyangensis SCSIO 5802 genomic DNA by PCR using primers abmU-expF and abmV-expR (Table S4). The gel-purified PCR product was digested with NdeI/BamHI and then ligated into the same sites of pET28a(+) to yield the recombinant expression plasmid pET28a-abmU. The construct was confirmed by DNA sequencing and transformed into E. coli BL21(DE3) for protein expression.

The site-directed mutations of AbmU were conducted using the Fast Mutagenesis System (Transgen, Beijing, China) according to the manual protocol with primers listed in Table S4. The specific residues mutated in this study include seven His residues (H10, H48, H61, H99, H160, H166, and H176), three Met residues (M101, M102, and M146), and seven Tyr residues (Y53, Y69, Y92, Y103, Y104, Y143, and Y163). For the seven His residue and three Met residues, H176 and M101 were mutated to the nonpolar residues phenylalanine and leucine, respectively; the other six His residues and two Met residues were each mutated to alanine. For the seven Tyr residues, they were each mutated to phenylalanine. Each mutation was confirmed by DNA sequencing.

Expression of AbmU and its variants for enzymatic assays followed the same general procedure detailed as follows. The cell culture was grown at 37 °C in LB medium containing 50 μg/mL kanamycin to an OD600 of ≈0.4. The cells were then cooled on ice for 10 min and induced with 0.1 mM isopropyl β-d-thiogalactopyranoside (IPTG). The cells were grown for an additional 20 h at 16 °C. To support efficient expression of holo-form AbmU enzyme, 1 mM heme precursor 5-aminolevulinic-acid (5-ALA) and 1 mM (NH4)2Fe(SO4)2 were added to the culture with IPTG. The cells were harvested by centrifugation, and the resulting cell pellet was stored at −80 °C prior to purification.

Expression of AbmU for X-ray crystallization was achieved as follows. The cell culture was grown at 37 °C in LB medium containing 50 μg/mL kanamycin to an OD600 of ≈0.6. IPTG was then added to the culture to a final concentration of 1 mM to induce protein expression. The culture was grown for a further 20 h at 16 °C. The cells were harvested by centrifugation, and the resulting cell pellet was stored at −80 °C prior to purification.

For expression of selenomethionine (SeMet)-labeled AbmU for X-ray crystallization, the culture of E. coli BL21(DE3) cells harboring plasmid pET28a-abmU was first grown in 30 mL of LB medium containing 50 μg/mL kanamycin at 37 °C for 16 h. Then, the cell culture was centrifuged and the resulting cell pellet was suspended in 30 mL of M9 medium.29 This suspension was centrifuged, and the resulting cell pellet was resuspended in 30 mL of M9 medium, which was used to inoculate 600 mL of M9 medium supplemented with 50 μg/mL kanamycin. The culture was grown at 37 °C until it reached an OD600 of 0.6. After Supplementation of 0.015 g SeMet, IPTG was then added to the culture to a final concentration of 1 mM to induce protein expression. The culture was grown for a further 20 h at 16 °C. The cells were harvested by centrifugation, and the resulting cell pellet was stored at −80 °C prior to purification.

Protein Purification

Purification of AbmU and its variants for enzymatic assay followed the same general procedure as mentioned below. The frozen cell pellet was thawed on ice, suspended in 30 mL of binding buffer (50 mM Tris-HCl, 150 mM NaCl, and 10 mM imidazole, 10% glycerol, pH 8.0), and sonicated on ice. Cellular debris was removed by centrifugation (12 000 rpm, 40 min, 4 °C), and the supernatant was loaded onto a Ni-NTA His-Bind Resin (Novagen, CA) column, washed with washing buffer (50 mM Tris-HCl, 150 mM NaCl, and 35 mM imidazole, 10% glycerol, pH 8.0), and then eluted with eluting buffer (50 mM Tris-HCl, 150 mM NaCl, and 250 mM imidazole, 10% glycerol, pH 8.0). The fractions containing target protein of interest were pooled, desalted by passage through a PD-10 column (GE Healthcare), and concentrated by ultrafiltration (Millipore membrane, 3 kDa cutoff size). The purified protein was analyzed by SDS-PAGE (Figure S7) and finally stored in storage buffer (50 mM Tris-HCl, 150 mM NaCl, 10% glycerol, pH 8.0) for further experiments at −80 °C. The protein concentration was determined by Bradford assays using bovine serum albumin as a standard.

To purify AbmU for protein X-ray crystallization, the frozen cell pellet was thawed on ice, suspended in lysis buffer (50 mM Tris-HCl, 300 mM NaCl, pH 8.0), and sonicated on ice. Cellular debris was removed by centrifugation (12 000 rpm, 40 min, 4 °C). The supernatant was loaded onto a 5 mL Ni-affinity HisTrap HP column (GE Healthcare), washed with washing buffer (50 mM Tris-HCl, 300 mM NaCl, and 20 mM imidazole, pH 8.0), and then eluted with eluting buffer (50 mM Tris-HCl, 300 mM NaCl, and 250 mM imidazole, pH 8.0). The fractions containing AbmU were pooled and concentrated to 5 mL by microfiltration and applied to a Superdex-200 gel-filtration column, preequilibrated in 10 mM Tris-HCl, 100 mM NaCl, pH 8.0. Fractions containing the target protein were pooled and concentrated to 10 mg/mL by microfiltration for crystallization.

Purification of SeMet-AbmU for protein X-ray crystallization used the same procedures as described above, except that all of the buffers were supplemented with 1 mM β-ME.

In Vitro Biochemical Assays

Assays of AbmU were performed in 100 μL of reaction mixtures containing 1 μM AbmU and 200 μM compound 5 in 50 mM Tris-HCl (pH 8.0) at 30 °C for 10 min. The reaction in the absence of the enzyme was used as the negative control. For the AbmU reaction under reducing conditions, the freshly prepared dithionite was added to a final concentration of 1 mM in the assay (100 μL) and preincubated with AbmU (1 μM) for 5 min at 30 °C; to initiate the reaction, substrate 5 (200 μM final concentration) was added, and the assay mixtures were further incubated at 30 °C for 10 min. All reactions were quenched by adding equal volumes of prechilled methanol. The precipitated proteins were removed by centrifugation, and 80 μL of each supernatant was subjected to HPLC analyses following the same method as described above for metabolite analyses of WT S. koyangensis SCSIO 5802 and related derivative strains.

The enzymatic activity assays for all purified AbmU mutant proteins were performed as described above for the native WT protein AbmU.

Kinetic Analysis of the AbmU Reaction

To determine the kinetics of AbmU, the assays were performed in 50 μL reaction mixtures containing 50 mM Tris-HCl (pH 8.0), 0.3 μM AbmU, and varying concentrations of substrate 5 (0, 25, 50, 100, 200, 400, 600, 800, or 1000 μM) at 30 °C for 6 min. The reactions were quenched by addition of 50 μL of prechilled methanol. The precipitated proteins were removed by centrifugation, and 80 μL of each supernatant was analyzed by HPLC, as mentioned above. All assays were performed in triplicate. Data fitting was performed using Origin Pro 9.0 software (OriginLab, CO).

Pyridine Hemochrome Assay

The concentration of heme in AbmU was determined using the pyridine hemochrome assay.33 The percent of heme in AbmU was calculated based on a comparison of the heme concentration from the pyridine assay and the concentration of protein determined by Bradford assays. The percentage of heme in the AbmU purified from the minimal LB medium was determined to be ∼0.1% based on the above assay. When the heme precursor 5-ALA and (NH4)2Fe(SO4)2 was added to the LB medium, the measured heme% increased to ∼1.5%.

Anaerobic AbmU Reaction

AbmU (1 μM) in Tris-HCl buffer (50 mM, pH 8.0) was gently bubbled with Ar for 20 min before being placed into an anaerobic chamber. Substrate 5 (20 mM) was prepared with Ar-saturated dimethyl sulfoxide (DMSO) in the anaerobic chamber; sodium dithionite solution (0.1 M) was prepared with Ar-saturated water in the anaerobic chamber. The reaction mixtures of compound 5 (200 μM) with purified AbmU (1 μM) or dithionite-reduced AbmU (1 μM) were incubated at 30 °C for 10 min in the anaerobic chamber, then quenched by the addition of equal volumes of Ar-treated methanol, and subjected to HPLC analyses, as described above.

Crystallization and Structure Determination

Crystallization of AbmU and SeMet-AbmU was performed using hanging-drop vapor diffusion method at 18 °C. To obtain AbmU crystals, 1 μL of protein (10 mg/mL in 10 mM Tris-HCl pH 8.0, 100 mM NaCl) was mixed with 1 μL of reservoir solution containing 20% PEG 3350, 0.1 M HEPES pH 7.0. The SeMet-AbmU crystals were harvested under the crystallization condition with 20% PEG 3350, 0.1 M MES pH 6.8. Crystals were rapidly soaked in the reservoir solution supplemented with 20% glycerol as a cryo-protectant, mounted on loops, and flash-cooled in a nitrogen gas cryo-stream.

Crystal diffraction data were collected from a single crystal at Shanghai Synchrotron Radiation Facility BL19U beamline, China, with a wavelength of 0.97593 Å. The diffraction data were processed and scaled with HKL-3000.34 The structure was determined by the Se-SAD method, using PHENIX AutoSol.35 The model was automatically built using PHENIX AutoBuild, followed by manual model building using COOT36 and structural refinement using PHENIX.refine and Refmac5.37 Relevant statistics are summarized in Table S2.

Docking Studies

Substrate modeling was performed using AutoDock 4.2.27 Before the docking, water molecules were removed from AbmU. AutoDockTools 1.5.6 was used to prepare the substrate 5 and product 4, AbmU protein, and grid map files. The default parameters were used to set the tortion constraints for 4 and 5, and charges and hydrogen atoms were added to 4, 5, and AbmU protein. A grid box of 20 × 16 × 26 nm3 (x × y × z) with the search space center of x = 0.175, y = 1.887, and z = −55.848 was created (grid spacing: 0.375 Å, smoothing: 0.5 Å). All dockings were performed using a Lamarkian genetic algorithm (LGA). The independent flexible docking runs were set as 400, and other search parameters were used with default values. For the modeling of substrate 5, the modeling hit with the lowest binding energy among those with terminal conjugated Δ8,9, Δ10,11, Δ12,13-triene close to the Δ14,15 exo-methylene group was selected for substrate binding analysis. For the modeling of product 4, the modeling hit with lowest binding energy among those possessing similar orientations to the selected hit in the modeling of 5 was used for product binding analysis.

Acknowledgments

This study was supported in part by the China NSF (31670087, U1706206, and 41676151), the Key Special Project for Introduced Talents Team of Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou, GML2019ZD0406), the Natural Key Research and Development Program of China (2019YFC0312500), the Guangdong NSF (2016A030312014), Guangdong Local Innovation Team Program (2019BT02Y262), the Pearl River S&T Nova Program of Guangzhou (201806010109), the Special Support Program for Training High-Level Talents in Guangdong (No. 2016TQ03R288), and the CAS (XDA13020302-2).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.0c02776.

  • NMR and HRESIMS spectra of compound 5, data collection and refinement statistics of AbmU, and all confirmatory data (tables and figures) for in vivo and in vitro enzyme assays (PDF)

Author Contributions

Q.L. designed the study, performed the experiments, analyzed the data, and wrote the manuscript. W.D., J.T., X.J., and Z.Y. performed the experiments. H.H. analyzed the NMR data. C.C. and M.M. analyzed the structural data. J.J. designed the study, directed the research, and revised the manuscript.

The authors declare no competing financial interest.

Notes

Atomic coordinates and structure factors for the reported crystal structures are deposited in the PDB under accession number PDB: 6LE0.

Supplementary Material

ao0c02776_si_001.pdf (2.1MB, pdf)

References

  1. Corey E. J. Catalytic enantioselective Diels–Alder reactions: Methods, mechanistic fundamentals, pathways, and applications. Angew. Chem., Int. Ed. 2002, 41, 1650–1667. . [DOI] [PubMed] [Google Scholar]
  2. Jeon B.-S.; Wang S.-A.; Ruszczycky M. W.; Liu H.-W. Natural [4 + 2]-cyclases. Chem. Rev. 2017, 117, 5367–5388. 10.1021/acs.chemrev.6b00578. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Kim H. J.; Ruszczycky M. W.; Choi S.-H.; Liu Y.-N.; Liu H.-W. Enzyme-catalysed [4 + 2] cycloaddition is a key step in the biosynthesis of spinosyn A. Nature 2011, 473, 109–112. 10.1038/nature09981. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Fage C. D.; Isiorho E. A.; Liu Y.; Wagner D. T.; Liu H.-W.; Keatinge-Clay A. T. The structure of Spnf, a standalone enzyme that catalyzes [4 + 2] cycloaddition. Nat. Chem. Biol. 2015, 11, 256–258. 10.1038/nchembio.1768. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Hashimoto T.; Hashimoto J.; Teruya K.; Hirano T.; Shin-Ya K.; Ikeda H.; Liu H.-W.; Nishiyama M.; Kuzuyama T. Biosynthesis of versipelostatin: Identification of an enzyme-catalyzed [4 + 2]-cycloaddition required for macrocyclization of spirotetronate-containing polyketides. J. Am. Chem. Soc. 2015, 137, 572–575. 10.1021/ja510711x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Tian Z.; Sun P.; Yan Y.; Wu Z.; Zheng Q.; Zhou S.; Zhang H.; Yu F.; Jia X.; Chen D. An enzymatic [4 + 2] cyclization cascade creates the pentacyclic core of pyrroindomycins. Nat. Chem. Biol. 2015, 11, 259–265. 10.1038/nchembio.1769. [DOI] [PubMed] [Google Scholar]
  7. Zheng Q.; Guo Y.; Yang L.; Zhao Z.; Wu Z.; Zhang H.; Liu J.; Cheng X.; Wu J.; Yang H.; Jiang H.; Pan L.; Liu W. Enzyme-dependent [4 + 2] cycloaddition depends on lid-like interaction of the N-terminal sequence with the catalytic core in PyrI4. Cell Chem. Biol. 2016, 23, 352–360. 10.1016/j.chembiol.2016.01.005. [DOI] [PubMed] [Google Scholar]
  8. Byrne M. J.; Lees N. R.; Han L.-C.; van der Kamp M. W.; Mulholland A. J.; Stach J. E.; Willis C. L.; Race P. R. The catalytic mechanism of a natural Diels–Alderase revealed in molecular detail. J. Am. Chem. Soc. 2016, 138, 6095–6098. 10.1021/jacs.6b00232. [DOI] [PubMed] [Google Scholar]
  9. Li L.; Yu P.; Tang M.-C.; Zou Y.; Gao S.-S.; Hung Y.-S.; Zhao M.; Watanabe K.; Houk K.; Tang Y. Biochemical characterization of a eukaryotic decalin-forming Diels–Alderase. J. Am. Chem. Soc. 2016, 138, 15837–15840. 10.1021/jacs.6b10452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Wever W. J.; Bogart J. W.; Baccile J. A.; Chan A. N.; Schroeder F. C.; Bowers A. A. chemoenzymatic synthesis of thiazolyl peptide natural products featuring an enzyme-catalyzed formal [4 + 2] cycloaddition. J. Am. Chem. Soc. 2015, 137, 3494–3497. 10.1021/jacs.5b00940. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Hudson G. A.; Zhang Z.; Tietz J. I.; Mitchell D. A.; van der Donk W. A. In vitro biosynthesis of the core scaffold of the thiopeptide thiomuracin. J. Am. Chem. Soc. 2015, 137, 16012–16015. 10.1021/jacs.5b10194. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cogan D. P.; Hudson G. A.; Zhang Z.; Pogorelov T. V.; van der Donk W. A.; Mitchell D. A.; Nair S. K. Structural insights into enzymatic [4 + 2] aza-cycloaddition in thiopeptide antibiotic biosynthesis. Proc. Natl. Acad. Sci. U.S.A. 2017, 114, 12928–12933. 10.1073/pnas.1716035114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Zheng Q.; Gong Y.; Guo Y.; Zhao Z.; Wu Z.; Zhou Z.; Chen D.; Pan L.; Liu W. Structural insights into a flavin-dependent [4 + 2] cyclase that catalyzes trans-decalin formation in pyrroindomycin biosynthesis. Cell Chem. Biol. 2018, 25, 718–727. 10.1016/j.chembiol.2018.03.007. [DOI] [PubMed] [Google Scholar]
  14. Ohashi M.; Liu F.; Hai Y.; Chen M.; Tang M.-C.; Yang Z.; Sato M.; Watanabe K.; Houk K.; Tang Y. SAM-dependent enzyme-catalysed pericyclic reactions in natural product biosynthesis. Nature 2017, 549, 502–506. 10.1038/nature23882. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Song Y.; Li Q.; Qin F.; Sun C.; Liang H.; Wei X.; Wong N.-K.; Ye L.; Zhang Y.; Shao M.; Ju J. Neoabyssomicins A–C, polycyclic macrolactones from the deep-sea derived Streptomyces koyangensis SCSIO 5802. Tetrahedron 2017, 73, 5366–5372. 10.1016/j.tet.2017.07.034. [DOI] [Google Scholar]
  16. Huang H.; Song Y.; Li X.; Wang X.; Ling C.; Qin X.; Zhou Z.; Li Q.; Wei X.; Ju J. Abyssomicin monomers and dimers from the marine-derived Streptomyces koyangensis SCSIO 5802. J. Nat. Prod. 2018, 81, 1892–1898. 10.1021/acs.jnatprod.8b00448. [DOI] [PubMed] [Google Scholar]
  17. Bister B.; Bischoff D.; Ströbele M.; Riedlinger J.; Reicke A.; Wolter F.; Bull A. T.; Zähner H.; Fiedler H. P.; Süssmuth R. D. Abyssomicin C—a polycyclic antibiotic from a marine Verrucosispora Strain as an inhibitor of the P-aminobenzoic acid/tetrahydrofolate biosynthesis pathway. Angew. Chem., Int. Ed. 2004, 43, 2574–2576. 10.1002/anie.200353160. [DOI] [PubMed] [Google Scholar]
  18. Tu J.; Li S.; Chen J.; Song Y.; Fu S.; Ju J.; Li Q. Characterization and heterologous expression of the neoabyssomicin/abyssomicin biosynthetic gene cluster from Streptomyces koyangensis SCSIO 5802. Microb. Cell Fact. 2018, 17, 28 10.1186/s12934-018-0875-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Li Q.; Ding W.; Yao Z.; Tu J.; Wang L.; Huang H.; Li S.; Ju J. AbmV catalyzes tandem ether installation and hydroxylation during neoabyssomicin/abyssomicin biosynthesis. Org. Lett. 2018, 20, 4854–4857. 10.1021/acs.orglett.8b01997. [DOI] [PubMed] [Google Scholar]
  20. Ji X.; Tu J.; Song Y.; Zhang C.; Wang L.; Li Q.; Ju J. A Luciferase-like monooxygenase and flavin reductase pair AbmE2/AbmZ catalyzes Baeyer-Villiger oxidation in neoabyssomicin biosynthesis. ACS Catal. 2020, 10, 2591–2595. 10.1021/acscatal.9b05488. [DOI] [Google Scholar]
  21. Beltrán J.; Kloss B.; Hosler J. P.; Geng J.; Liu A.; Modi A.; Dawson J. H.; Sono M.; Shumskaya M.; Ampomah-Dwamena C.; et al. Control of carotenoid biosynthesis through a Heme-based cis-trans isomerase. Nat. Chem. Biol. 2015, 11, 598–605. 10.1038/nchembio.1840. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Du Y.-L.; He H.-Y.; Higgins M. A.; Ryan K. S. A Heme-dependent enzyme forms the nitrogen–nitrogen bond in piperazate. Nat. Chem. Biol. 2017, 13, 836–838. 10.1038/nchembio.2411. [DOI] [PubMed] [Google Scholar]
  23. Berry E. A.; Trumpower B. L. Simultaneous determination of Hemes a, b, and c from pyridine hemochrome spectra. Anal. Biochem. 1987, 161, 1–15. 10.1016/0003-2697(87)90643-9. [DOI] [PubMed] [Google Scholar]
  24. Zoppellaro G.; Bren K. L.; Ensign A. A.; Harbitz E.; Kaur R.; Hersleth H. P.; Ryde U.; Hederstedt L.; Andersson K. K. Review: Studies of ferric heme proteins with highly anisotropic/highly axial low spin (S = 1/2) electron paramagnetic resonance signals with bis-histidine and histidine-Methionine axial iron coordination. Biopolymers 2009, 91, 1064–1082. 10.1002/bip.21267. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Liu S.-C.; Zhai S.; Lawler J.; Palek J. Hemin-mediated dissociation of erythrocyte membrane skeletal proteins. J. Biol. Chem. 1985, 260, 12234–12239. [PubMed] [Google Scholar]
  26. Airola M. V.; Du J.; Dawson J. H.; Crane B. R. Heme binding to the mammalian circadian clock protein period 2 is nonspecific. Biochemistry 2010, 49, 4327–4338. 10.1021/bi901945w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Morris G. M.; Huey R.; Lindstrom W.; Sanner M. F.; Belew R. K.; Goodsell D. S.; Olson A. J. Autodock4 and AutoDockTools4: Automated docking with selective receptor flexiblity. J. Comput. Chem. 2009, 16, 2785–2791. 10.1002/jcc.21256. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Li Q.; Song Y.; Qin X.; Zhang X.; Sun A.; Ju J. Identification of the biosynthetic gene cluster for the anti-infective desotamides and production of a new analogue in a heterologous host. J. Nat. Prod. 2015, 78, 944–948. 10.1021/acs.jnatprod.5b00009. [DOI] [PubMed] [Google Scholar]
  29. Green M. R.; Sambrook J.. Molecular Cloning: A Laboratory Manual, 4th Ed.; Cold Spring Harbor Laboratory Press: New Year, 2012. [Google Scholar]
  30. Kieser T.; Bibb M. J.; Buttner M. J.; Chater K. F.; Hopwood D. A.. Practical Streptomyces Genetics; John Innes Foundation: Norwich, 2000; Vol. 291. [Google Scholar]
  31. Gust B.; Kieser T.; Chater K. F.. Redirect Technology: PCR-Targeting System in Streptomyces coelicolor A3 (2); John Innes Centre: Norwich, 2002. [Google Scholar]
  32. Hong B.; Phornphisutthimas S.; Tilley E.; Baumberg S.; McDowall K. J. Streptomycin production by Streptomyces griseus can be modulated by a mechanism not associated with change in the adpA component of the A-factor cascade. Biotechnol. Lett. 2006, 29, 57–64. 10.1007/s10529-006-9216-2. [DOI] [PubMed] [Google Scholar]
  33. Barr I.; Guo F. Pyridine hemochromagen assay for determining the concentration of heme in purified protein solutions. Bio-Protoc. 2015, 5, e1594 10.21769/BioProtoc.1594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Minor W.; Cymborowski M.; Otwinowski Z.; Chruszcz M. Hkl-3000: The integration of data reduction and structure solution–from diffraction images to an initial model in minutes. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2006, 62, 859–866. 10.1107/S0907444906019949. [DOI] [PubMed] [Google Scholar]
  35. Weinert T.; Olieric V.; Waltersperger S.; Panepucci E.; Chen L.; Zhang H.; Zhou D.; Rose J.; Ebihara A.; Kuramitsu S.; et al. Fast native-sad phasing for routine macromolecular structure determination. Nat. Methods. 2015, 12, 131–133. 10.1038/nmeth.3211. [DOI] [PubMed] [Google Scholar]
  36. Emsley P.; Cowtan K. Coot: Model-building tools for molecular graphics. Acta Crystallogr., Sect. D Biol. Crystallogr. 2004, 60, 2126–2132. 10.1107/S0907444904019158. [DOI] [PubMed] [Google Scholar]
  37. Murshudov G. N.; Skubák P.; Lebedev A. A.; Pannu N. S.; Steiner R. A.; Nicholls R. A.; Winn M. D.; Long F.; Vagin A. A. Refmac5 for the refinement of macromolecular crystal structures. Acta Crystallogr., Sect. D Biol. Crystallogr. 2011, 67, 355–367. 10.1107/S0907444911001314. [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

ao0c02776_si_001.pdf (2.1MB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES