Functional characterization of CYP107W1 from Streptomyces avermitilis and biosynthesis of macrolide oligomycin A

Archives of Biochemistry and Biophysics 575 (2015) 1–7

Contents lists available at ScienceDirect

Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi

Functional characterization of CYP107W1 from Streptomyces avermitilis and biosynthesis of macrolide oligomycin A Songhee Han a,1, Tan-Viet Pham a,1, Joo-Hwan Kim a, Young-Ran Lim a, Hyoung-Goo Park a, Gun-Su Cha b, Chul-Ho Yun b, Young-Jin Chun c, Lin-Woo Kang a,⇑, Donghak Kim a,⇑ a b c

Konkuk University, Department of Biological Sciences, Seoul 143-701, Republic of Korea Chonnam National University, School of Biological Sciences and Technology, Gwangju 500-757, Republic of Korea Chung-Ang University, College of Pharmacy, Seoul 156-756, Republic of Korea

a r t i c l e

i n f o

Article history: Received 23 February 2015 and in revised form 26 March 2015 Available online 4 April 2015 Keywords: P450 CYP CYP107W1 Streptomyces avermitilis Oligomycin X-ray crystal structure

a b s t r a c t Streptomyces avermitilis contains 33 cytochrome P450 genes in its genome, many of which play important roles in the biosynthesis process of antimicrobial agents. Here, we characterized the biochemical function and structure of CYP107W1 from S. avermitilis, which is responsible for the 12-hydroxylation reaction of oligomycin C. CYP107W1 was expressed and purified from Escherichia coli. Purified proteins exhibited the typical CO-binding spectrum of P450. Interaction of oligomycin C and oligomycin A (12-hydroxylated oligomycin C) with purified CYP107W1 resulted in a type I binding with Kd values of 14.4 ± 0.7 lM and 2.0 ± 0.1 lM, respectively. LC–mass spectrometry analysis showed that CYP107W1 produced oligomycin A by regioselectively hydroxylating C12 of oligomycin C. Steady-state kinetic analysis yielded a kcat value of 0.2 min1 and a Km value of 18 lM. The crystal structure of CYP107W1 was determined at 2.1 Å resolution. The overall P450 folding conformations are well conserved, and the open access binding pocket for the large macrolide oligomycin C was observed above the distal side of heme. This study of CYP107W1 can help a better understanding of clinically important P450 enzymes as well as their optimization and engineering for synthesizing novel antibacterial agents and other pharmaceutically important compounds. Ó 2015 Elsevier Inc. All rights reserved.

Introduction Streptomyces are the largest genus of Actinobacteria and produce a variety of clinically and industrially useful secondary metabolites, including antibacterial, antifungal, antiparasitic, anticancer, and cardiovascular agents, immunosuppressants, and veterinary products [1]. Approximately 65% of all antibiotics of microbial origin are from this genus [2]. Many of these products belong to the group of polyketide macrolides. They are synthesized by polyketide synthases and subsequently modified via ring decoration [3]. The genomes of Streptomyces strains encode numerous cytochrome P450 (P450, CYP)2 enzymes, which are involved in the ⇑ Corresponding authors at: Dept. of Biological Sciences, Konkuk University, 120 Neungdong-ro, Gwangjjn-gu, Seoul 143-701, Republic of Korea. Fax: +82 2 3436 5432. E-mail addresses: [email protected] (L.-W. Kang), [email protected] (D. Kim). 1 These authors contributed equally to this work. 2 Abbreviations used: P450, cytochrome P450; CYP, cytochrome P450; NPR, NADPHP450 reductase; IPTG, isopropyl-b-D-thiogalactopyranoside; RMSD, root-mean-square deviation. http://dx.doi.org/10.1016/j.abb.2015.03.025 0003-9861/Ó 2015 Elsevier Inc. All rights reserved.

modification step of macrolide synthesis, generating novel antibiotics [4]. Since the genome of Streptomyces coelicolor, a prototypic actinobacterium, was sequenced and eighteen P450 genes were identified across its genome [5–7], increasing numbers of Streptomyces P450 genes have been discovered [8–11]. Numerous studies have been attempted to identify gene clusters involved in the synthesis of pharmaceutically important molecules and increase the product yields of these compounds. Streptomyces avermitilis produces avermectins, antiparasitic agents widely used in human and veterinary medicine [12], and contains as many as 33 P450 enzyme genes in its genome [4]. Among these, CYP171A1 participates in the catalysis of furan ring formation at C6–C8 of the avermectin aglycone, a part of the avermectins biosynthesis pathway [13,14]. CYP105P1 and CYP105D6 catalyze hydroxylation at C26 and C10 positions of filipin during filipin biosynthesis [15]. The CYP107W1 encoding gene is located in the gene cluster of oligomycin biosynthesis, and its putative activity is hydroxylation at position C12 of oligomycin C, a precursor of oligomycin A (Fig. 1) [13]. In this study, we describe the molecular cloning, purification, and biochemical characterization of CYP107W1 from S. avermitilis

2

S. Han et al. / Archives of Biochemistry and Biophysics 575 (2015) 1–7

OH OH

OH

OH

O

OH

OH

O

O

OH

CYP107W1

O

O

O

O

O

O

O O

O

OH

Oligomycin C (12-deoxy-oligomycin A)

OH

Oligomycin A

Fig. 1. Bioconversion of oligomycin C by CYP107W1.

and its crystal structure was determined at 2.1 Å resolution. The purified CYP107W1 successfully provided the catalytic turnover of oligomycin C to produce the metabolite, oligomycin A. Materials and methods Chemicals and enzymes Oligomycin A, econazole, miconazole, ketoconazole, itraconazole, fluconazole, sodium dithionite, glucose-6-phosphate, glucose-6-phosphate dehydrogenase, spinach ferredoxin, ferredoxin reductase, and NADP+ were purchased from Sigma–Aldrich (St. Louis, MO). Oligomycin C was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Crystal screening kits of Index, Crystal Screen, Crystal Screen Cryo, Crystal Screen Lite, PEGRx, SaltRx, and PEG/Ion were obtained from Hampton Research (Aliso Viejo, CA). Crystal screening kits of Wizard and Wizard precipitant synergy were purchased from Emerald Biosystems (Bainbridge Island, WA). All chemicals were of the highest grade commercially available. Construction of CYP107W1 expression plasmids The general approach has been described previously [16,17]. Genomic DNA of S. avermitilis was kindly provided by Professor Byung-Gee Kim of Seoul National University. The open reading frame for CYP107W1 was amplified using PCR with forward and reverse primers (50 -ATGGCCGAAGCGCCTTCCGAG-30 , 50 -TCACCA GGTGATGGGCAGGCT-30 ), and the amplified PCR fragments were reamplified with the second forward and reverse primers including the NdeI and EcoI restriction sites (50 -GGTGGTCATATGGCCGAA GCGCCTTCCGAG-30 , 50 -GGTGGTGAATTCTCACCAGGTGATGGGCAG GCT-30 ). The reamplified PCR fragments were cloned into the pET 28a(+) vector (EMD Millipore, Billerica, MA) containing a 6His tag and a thrombin cleavage site at the N-terminus, using the NdeI and EcoI restriction enzymes. The constructed mutant pET vector was confirmed by nucleotide sequencing analysis. Expression and purification of recombinant CYP107W1 enzyme Expression and purification of the CYP107W1 enzyme were carried out as previously described with some modifications [18]. Briefly, the Escherichia coli BL21(DE3)pLysS strains transformed

with pET/107W1 vectors were inoculated in TB medium containing 20 lg/mL kanamycin. Expression cultures were grown at 37 °C and 200 rpm until an OD600 of 0.4–0.6. After the addition of IPTG (1 mM), d-aminolevulinic acid (d-ALA; 0.5 mM), thiamine (1.0 mM), and trace elements, cultures were grown at 28 °C with shaking at 200 rpm for 26 h. Bacterial soluble fractions containing CYP107W1 were prepared from expression cultures and isolated by ultracentrifugation. CYP107W1 proteins were purified using a Ni2+-NTA column (Qiagen, Valencia, CA) and Superdex 200 10/ 300 GL column (GE Healthcare, Pittsburgh, PA). After sizeexclusion chromatography, the eluted fraction containing P450 enzyme was pooled and dialyzed against 25 mM Tris–HCl buffer (containing 50 mM sodium chloride and 3 mM b-mercaptoethanol) at 4 °C. The dialyzed fraction was concentrated using Amicon Ultra-15 Centrifugal Filter Units (EMD Millipore, Billerica, MA) for characterization and crystallization. Spectroscopic characterization and spectral binding titrations Sodium dithionite was added to reduce the purified ferric CYP107W1 enzymes. CO–ferrous P450 complexes were generated by passing CO gas through solutions of ferrous P450. UV–visible spectra were collected on a CARY Varian spectrophotometer (Agilent Technologies, Santa Clara, CA) in 100 mM potassium phosphate buffer (pH 7.4) at room temperature. For spectral binding titration, purified CYP107W1 enzyme was diluted to 3 lM in 100 mM potassium phosphate buffer (pH 7.4) and divided between two glass cuvettes. Spectra (350–500 nm) were recorded while subsequently adding various ligands, using a CARY Varian spectrophotometer [19]. The difference in absorbance between wavelength maximum and minimum was plotted against ligand concentration [20]. Oligomycin C hydroxylation assay Oligomycin C hydroxylation by CYP107W1 was determined using LC–mass spectrometry analysis. The reaction mixture included 200 pmol purified P450 enzyme, 40 lg/mL spinach ferredoxin, and 0.04 U/mL spinach ferredoxin reductase, in 0.50 mL of 100 mM potassium phosphate buffer (pH 7.4), along with a specified amount of oligomycin C. Reactions were initiated by adding 50 lL of an NADPH-generating system of 10 mM glucose 6phosphate, 0.5 mM NADP+, and 1.0 IU glucose 6-phosphate dehydrogenase mL1. The reactions were terminated after 30 min

3

S. Han et al. / Archives of Biochemistry and Biophysics 575 (2015) 1–7

Protein crystallization The sitting-drop vapor-diffusion method was employed for initial crystallization at 14 °C in 96-well Intelli plates (Art Robbins, Sunnyvale, CA), using a Hydra II e-Drop automated pipetting system (Matrix) and the screening kits of Index, Crystal Screen, Crystal Screen Cryo, Crystal Screen Lite, PEGRx, SaltRx, PEG/Ion, Wizard, and Wizard precipitant synergy. A concentration of 12 mg/mL CYP107W1 protein was used for crystallization: 0.5 lL protein solution was mixed with 0.5 lL reservoir solution and equilibrated against 50 lL reservoir solution. After 30 days, CYP107W1 protein crystals were obtained in the condition H12 (containing 0.15 M potassium bromide and 30% (w/v) polyethylene glycol monomethyl ether 2000) of the Index screening kit. Data collection Initial crystals were reproduced in the same condition by the hanging-drop method, in which drops consisted of 1.0 lL protein solution mixed with 1.0 lL reservoir solution. Each hanging drop was positioned over 1 mL reservoir solution. Fully grown needleshaped crystals (0.1  0.1  0.4 mm) were flash-cooled to 100 K in liquid nitrogen, using a solution of 0.15 M potassium bromide, Table 1 Data collection and refinement statistics.

30% (w/v) polyethylene glycol monomethyl ether 2000, and 6% (v/v) glycerol as cryoprotectant. X-ray diffraction data were collected from the cryoprotected crystal (at 100 K) with 1° rotation at a crystal-to-detector distance of 250 mm using an ADSC Q270 detector on beamline 5C-SBII of the Pohang Light Source (PLS, Pohang, Korea). Crystals diffracted to the resolution of 2.1 Å. Diffraction data were integrated and scaled using the HKL-2000 program package [21]. Structure determination The structure of CYP107W1 was determined by molecular replacement with Molrep [22], using the structure of P450 MycG from Micromonospora griseorubida (PDB ID: 2Y5N, 43% sequence identity) [23] as a search model. It was crystallized in the tetragonal space group P43212. The monomer structure placement was defined by rigid-body refinement in Refmac5 [24], and models were subjected to restrained refinement. One monomer of CYP107W1 was present in the asymmetric unit. After refinement of the protein model, the resulting map was checked thoroughly. The overall structure and volume of the substrate binding pocket for CYP107W1 were calculated and generated by PyMol [25]. Data collection statistics are provided in Table 1. Results Expression and purification of CYP107W1 Expression of recombinant CYP107W1 in E. coli cells was spectrally determined as 480 nmol P450 holoenzyme per liter culture. After ultracentrifugation of cell lysates, most P450 proteins were

A

kDa 240 140 100

Absorbance

of incubation at 37 °C by adding 1 mL of CH2Cl2, followed by vortex mixing and centrifugation. The reaction products were recovered from the organic phase after drying under N2. For the analysis of oxidized products, LC–mass spectrometry was performed on a Shimadzu LCMS-2010 EV system (Shimadzu, Kyoto, Japan), using an LC–MS solution software. Products were separated on a Shimpack VP-ODS column (2.0 mm i.d.  250 mm; Shimadzu), using CH3CN–MeOH–H2O (60:16:24, v/v/v) mobile phase at a flow rate of 0.15 mL/min. To identify metabolites, mass spectra were recorded by electrospray ionization in negative mode. Interface and detector voltages were 4.4 and 1.5 kV, respectively. Nebulization gas flow was set to 1.5 mL/min. Interface, curve desolvation line, and heat block temperatures were 250, 230, and 200 °C, respectively.

70 50 35 25

CYP107W1 structure

Total No. of reflections No. of unique reflections Completeness (%) Molecules per asymmetric unit Solvent content (%) Average I/r (I) Rsym  (%) Multiplicity Refinement Resolution range (Å) Rwork/Rfree B value RMSD Bond Angle Waters

Pal-5C (Pohang, Korea) 0.96418 50.00–2.10 (2.14–2.10) P43212 a = 114.9, b = 114.9, c = 79.0 a = c = b = 90.0 209,396 31,242 99.1 (98.1) 1 58.59 40.0 (7.4) 8.4 (40.4) 6.7 (6.7)

B

Fe³⁺

Fe²⁺ Fe²⁺-CO

Fe²⁺

Fe³⁺

Fe²⁺-CO

35.54–2.10 17.77/23.41 27.53 0.008 1.408 304

Values in parentheses are for highest-resolution shell. P P P P P Rsym = h i|I(h)i  | h iI(h)i, where I(h) is the intensity of reflection h, h P is the sum over all reflections, and i is the sum over i measurements of reflection h.  

Wavelength, nm

Absorbance

Data collection Beamline Wavelength (Å) Resolution range (Å) Space group Unit-cell parameters (Å)

Wavelength, nm Fig. 2. Spectral analysis of purified CYP107W1. (A) CO-binding difference spectra of purified CYP107W1. The inset shows a single band of purified protein at 46 kDa, as expected for the open reading frame of the CYP107W1 gene with a His-tag. (B) Absolute spectra are shown for CYP107W1 (3 lM) in the ferric (—), the ferrous (– –), and the ferrous CO-bound form (- - - -). All spectra were recorded at room temperature in 100 mM potassium phosphate (pH 7.4).

4

S. Han et al. / Archives of Biochemistry and Biophysics 575 (2015) 1–7

located in the soluble fraction. Subsequent Ni2+-NTA and size-exclusion columns produced a single band of purified CYP107W1 protein of 46 kDa on SDS–PAGE, as expected from the size of its open reading frame (Fig. 2A). The CO difference spectrum of the reduced form of CYP107W1 showed a maximum absorption at 449 nm (Fig. 2A). The absolute spectra indicated that the ferric form of CYP107W1 was in the low spin state, with a Soret band at 418 nm and smaller a-, and b-bands at 568 and 536 nm, respectively (Fig. 2B). The ferrous form of CYP107W1 with reduction by sodium dithionite, showed a broadened peak at around 417 nm and shifts of a- and b-bands to 558 and 530 nm, respectively (Fig. 2B).

Binding of oligomycin A and C to CYP107W1 Interaction of the substrate oligomycin C with purified CYP107W1 resulted in a shift from low-spin to high-spin spectra, with an increase at 385 nm and a decrease at 420 nm (type I binding mode, Fig. 3B). This spectral change by substrate binding indicated the displacement of an iron-bound water molecule by oligomycin C to yield a low spin hexa-coordinated heme for a productive oxidation reaction by P450. Binding constants (Kd) of P450 ligands were obtained from alterations of heme spectra as a function of titrant concentration [20]. A Kd value of 14.4 ± 0.7 lM were calculated (Fig. 3A). Oligomycin A (12-hydroxylated oligomycin C) was also found to display a similar type I binding spectral change with a Kd value of 2.0 ± 0.1 lM (Fig. 3B). Binding of various azole agents to purified CYP107W1 displayed the typical type II spectra with an increase at 435 nm and a decrease

Fig. 3. Binding of oligomycin A and C to CYP107W1. (A) Titration of CYP107W1 with oligomycin C. Kd value was calculated for oligomycin C: 14.4 ± 0.7 lM. (B) Titration of CYP107W1 with oligomycin A. Kd value was calculated for oligomycin A: 2.0 ± 0.1 lM.

at 413 nm, suggesting an inhibitor binding mode to the heme of P450 enzyme (Supplementary Fig. S1). For binding affinities of azole agents (econazole, miconazole, and ketoconazole) to CYP107W1, Kd values of 3.9–4.5 lM were calculated (Supplementary Fig. S1). Catalytic activities of CYP107W1 CYP107W1 is responsible for the catalytic reaction of regioselective hydroxylation of oligomycin C at C12 position. LC– mass spectrometric analysis showed that the P450 reaction of CYP107W1 selectively produced a 12-hydroxylated metabolite of the substrate (oligomycin A) (Fig. 4). A retention time of 30.3 min indicated a molecular weight of 790 of hydroxylated product (oligomycin A), corresponding to an increase of 16 compared to the parental oligomycin C (Fig. 4). The product was confirmed as oligomycin A by comparing it with an authentic compound (in mass fragmentation and retention time). Steady-state kinetic analysis of the oligomycin C hydroxylation was performed, and kinetic parameters were obtained using the Michaelis–Menten equation. The kcat value was approximately 0.21 ± 0.01 min1, with a calculated Km value of 17.9 ± 1.3 lM (Fig. 5). This catalytic turnover number is similar to that of the previously reported pentalenene oxidation by CYP183A1 (PtlI) from S. avermitilis [26]. Crystal structure of CYP107W1 A crystal structure of CYP107W1 was determined at 2.1 Å resolution by molecular replacement using the MycG structure from M. griseorubida (PDB ID: 2Y5N) (Table 1). We attempted to obtain

Fig. 4. Oligomycin C hydroxylation by purified CYP107W1. LC–mass spectrometry analysis of CYP107W1 oligomycin C hydroxylation reaction (upper panel). Retention time of oligomycin C, 44.4 min; retention time of oligomycin A, 30.3 min. MS scan of oligomycin C from CYP107W1 reaction extract (middle panel). MS scan of oligomycin A from CYP107W1 reaction extract (lower panel).

pmol 12-hydroxyoligomycin C/min/pmol P450

S. Han et al. / Archives of Biochemistry and Biophysics 575 (2015) 1–7

kcat = 0.21 ± 0.01 min -1 Km = 17.9 ± 1.3 µM

Oligomycin C, µM Fig. 5. Steady-state kinetic analysis of oligomycin C hydroxylation. Michaelis– Menten plot of oligomycin C hydroxylation by purified CYP107W1. Each point is presented as mean ± SD (range) of duplicate assays.

a substrate-bound complex structure of CYP107W1–oligomycin C by soaking or cocrystallizing CYP107W1 with oligomycin C. However, the obtained crystals showed no distinct extra density for oligomycin C. CYP107W1 appeared as a monomer in the asymmetric unit and showed a well conserved folding conformation of P450 structure, consisting of 16 a-helices and three long loops (Fig. 6A and Supplementary Fig. S3). The first long loop between helices A and B contains two b-strands in a hair-pin form. The

5

second long loop between helices K and K0 contains four b-strands in two hair-pin forms. The third long loop between helices K0 and L contains the heme binding domain. Heme is sandwiched between helices I and L in a conserved way of P450 structure (Fig. 6A). I-helix crosses the center of CYP107W1 in a slightly bent form over the heme structure and helices F and G are stacked onto I-helix (Fig. 6A). The heme iron was coordinated with Cys353 in the proximal side of heme and a water molecule in the distal side with a nice geometry (Fig. 7A). The substrate binding site above the distal side of heme is partially covered by I-helix and the substantial part is exposed to solvent through the wide open access pocket (Fig. 7B). The surface of the putative substrate binding pocket was formed with I-helix and three nearby loops. Interestingly, this surface are surrounded by hydrophobic residues (Ala239 and Thr243 from I-helix; Met85, Ser88, Leu89, and Val90 from the loop between helices B and C; Leu285, Gly289, Gly290, Ile291, and Ile292 from the loop between helices K and K0 ; Ser392, Ile393, and Ile394 from the C-terminal loop) and therefore, the hydrophobic pocket can be rendered for the better interaction with hydrophobic chain of oligomycin C (Fig. 7B). Discussion Macrolide-class antibiotics are among the biggest clinically used antibiotics. The detailed mechanisms of synthesis, recognition, and modification of such large molecules through a series of

Fig. 6. Overall structure of CYP107W1. (A) Cylindrical helices viewed from the top (left) and side (right) of the barrel. Secondary structure elements including a-helices, bstrands and loops are colored in red, yellow, and green, respectively. (B) Surface structures of CYP107W1, CYP105N1, and MycG. P450 structures involving in biosynthesis of macrolide or bulky molecule. The arrows indicate the addition side holes. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

6

S. Han et al. / Archives of Biochemistry and Biophysics 575 (2015) 1–7

A

B F G

I A239

B’

S88

T243

S392 I394

C-term loop

I393

M85

L285 V90

I291 I292

β2

β1

Fig. 7. Substrate binding pocket of CYP107W1. (A) The 2Fo–Fc electron density map of CYP107W1 heme structure. Heme coordinated Cys residue and water molecule are shown. (B) Hydrophobic pocket for the substrate binding. The hydrophobic residues in the pocket are indicated.

enzymes are largely unknown. Oligomycins have been known to inhibit ATP synthase by blocking its proton channel, which is necessary for the oxidative phosphorylation of ADP [27,28]. Structural analysis of oligomycin A-bound ATP synthase indicated that oligomycin binds to the surface of the C10 ring, making contact with two neighboring molecules at a position important for the inhibitory effect [29]. Oligomycins have been found in various Streptomyces strains including Streptomyces bottropensis, Streptomyces libani, Streptomyces griseolus, Streptomyces diastaticus as well as S. avermitilis, while oligomycin A was first isolated from Streptomyces diastatochromogenes[30–33]. Among these, S. avermitilis yielded high quantities of oligomycin A, and CYP107W1 from S. avermitilis was the P450 enzyme to finalize the C12-hydroxylation of oligomycin by converting oligomycin C to oligomycin A [32]. Three-dimensional structures are essential to understand the molecular basis of substrate recognition and specificity of P450s. Structures of two S. avermitilis P450 enzymes (CYP105P1 and CYP105D6) were determined, revealing hydrophobic and numerous water molecules-involving interactions with the macrolide substrate [15,34]. This hydrophobic pocket for binding of macrolide substrate was also conserved in CYP107W1 structure (Fig. 7B). CYP105P1 appears to take advantage of a small sub-pocket for regioselective C26 hydroxylation [15]. Structural characterization of P450 MycG, another P450 enzyme involved in multifunctional oxidation of macrolide compounds, from M. griseorubida revealed the conformational advance of the bulky macrolide substrate for the catalytic productive mode [23]. Previously, the structure of CYP105N1 from S. coelicolor was characterized by our own study and Zhao’s et al. [18,35]. These studies indicated that CYP105N1 possesses a wide, open binding pocket above the heme group, generating a special architecture

of the enzyme to fit the large substrate coelibactin in the active site for easy sequential access during biosynthesis [18,35]. A similar large open substrate-binding pocket was identified in the structure of CYP107W1 of this study (Fig. 6B). The size difference between oligomycin C (MW, 775) and camphor (MW, 153, a substrate for CYP101) is more than five times. In order to accommodate bigger substrates such as oligomycin C, CYP107W1 should have a special structural architecture, or it is possible that some part of the large substrate may be exposed to solvent. CYP107W1 shares a common feature of structure with CYP105N1 and MycG in that they have similar large-sized substrates such as coelibactin or mycinamicin. All these P450 enzymes bind large size of substrates such as macrolides and show wide open access pockets, and their heme prosthetic groups can be seen through the upside wide pockets (Fig. 6B) [18,23,35]. Interestingly, many of them show the additional smaller holes in the wall of the upside open access pocket and these additional side holes are connected to the upside open access pocket (Fig. 6B). In CYP105N1 and MycG structures, the additional holes are located at the left side, which are parallel to I helix (Fig. 6B). However, CYP107W1 has a hole at approximately 30° rotated position (Fig. 6B). This altered location of the smaller hole in CYP107W1 is because the B0 helix in CYP107W1 occupies the space for the left hole positions formed with loops in CYP105N1 and MycG structures (Fig. 7B). Although the detailed mechanism needs further study, we speculate that the wide open upside pocket with the additional side hole in CYP107W1 might help the bulky hydrophobic substrate to slide into the active site heme of P450 enzyme through the hydrophobic substrate binding pocket. We also compared the CYP107W1 structure with other resolved structures of known P450s particularly utilizing macrolides as substrates, such as CYP154C1 from S. coelicolor A3, CYP105D6 from S. avermitilis, PikC from Streptomyces venezuelae, and PimD from Streptomyces natalensis as well as MycG P450 (Supplementary Fig. S4). These P450 enzymes also share the conserved P450 folds and displayed wide open substrate access pockets (Supplementary Fig. S4), which are common features found in CYP107W1 and MycG structures (Fig. 6B). However, the structure of CYP105P1 possesses an altered location of substrate binding channel, which is located at diagonal position rather than top position [15]. Meanwhile, P450eryF from Saccharopolyspora erythraea uses the smaller substrate of 6-deoxyerythronolide B and its substrate binding pocket is totally surrounded by enzyme residues and therefore it is not exposed to solution [36]. The multiple sequence alignment of these macrolide-metabolizing P450 enzymes did not show any specific sequence for macrolide substrates other than the conserved P450 signature sequences (Supplementary Fig. S5). In conclusion, we have characterized the biochemical and structural aspects of CYP107W1 from S. avermitilis. This enzyme efficiently catalyzes oligomycin C 12-hydroxylation and has a large substrate access pocket with a conserved P450 folding. This study can extend the existing knowledge about Streptomyces P450s in general and its macrolide biosynthesis in particular. Accession numbers Coordinate and structure factors of CYP107W1 have been deposited in the Protein Data Bank with accession number of 4WPZ. Acknowledgments This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korea government (Nos. 20110016509, 2009-0088150) and a grant from the Cooperative

S. Han et al. / Archives of Biochemistry and Biophysics 575 (2015) 1–7

Research Program for Agriculture Science & Technology Development (PJ01127901), Rural Development Administration, Korea. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.abb.2015.03.025. References [1] P. Dyson, Streptomyces: Molecular Biology and Biotechnology, Caister Academic Press, Norfolk, UK, 2011. [2] D.C. Lamb, F.P. Guengerich, S.L. Kelly, M.R. Waterman, Expert Opin. Drug Metabol. Toxicol. 2 (2006) 27–40. [3] V.B. Urlacher, M. Girhard, Trends Biotechnol. 30 (2012) 26–36. [4] S.L. Kelly, D.E. Kelly, C.J. Jackson, A.G.S. Warrilow, D.C. Lamb, in: P.R. Ortiz de Montellano (Ed.), Cytochrome P450: Structure, Mechanism, and Biochemistry, Plenum Press, New York, 2005, pp. 585–617. [5] S.D. Bentley, K.F. Chater, A.M. Cerdeno-Tarraga, G.L. Challis, N.R. Thomson, K.D. James, D.E. Harris, M.A. Quail, H. Kieser, D. Harper, A. Bateman, S. Brown, G. Chandra, C.W. Chen, M. Collins, A. Cronin, A. Fraser, A. Goble, J. Hidalgo, T. Hornsby, S. Howarth, C.H. Huang, T. Kieser, L. Larke, L. Murphy, K. Oliver, S. O’Neil, E. Rabbinowitsch, M.A. Rajandream, K. Rutherford, S. Rutter, K. Seeger, D. Saunders, S. Sharp, R. Squares, S. Squares, K. Taylor, T. Warren, A. Wietzorrek, J. Woodward, B.G. Barrell, J. Parkhill, D.A. Hopwood, Nature 417 (2002) 141–147. [6] D.C. Lamb, L. Lei, B. Zhao, H. Yuan, C.J. Jackson, A.G. Warrilow, T. Skaug, P.J. Dyson, E.S. Dawson, S.L. Kelly, D.L. Hachey, M.R. Waterman, Appl. Environ. Microbiol. 76 (2010) 1975–1980. [7] D.C. Lamb, T. Skaug, H.L. Song, C.J. Jackson, L.M. Podust, M.R. Waterman, D.B. Kell, D.E. Kelly, S.L. Kelly, J. Biol. Chem. 277 (2002) 24000–24005. [8] H. Ikeda, J. Ishikawa, A. Hanamoto, M. Shinose, H. Kikuchi, T. Shiba, Y. Sakaki, M. Hattori, S. Omura, Nat. Biotechnol. 21 (2003) 526–531. [9] N. Parajuli, D.B. Basnet, H. Chan, Arch. Biochem. Biophys. 425 (2004) 233–241. [10] M. Oliynyk, M. Samborskyy, J.B. Lester, T. Mironenko, N. Scott, S. Dickens, S.F. Haydock, P.F. Leadlay, Nat. Biotechnol. 25 (2007) 447–453. [11] H. Wu, S. Qu, C. Lu, H. Zheng, X. Zhou, L. Bai, Z. Deng, BMC Genomics 13 (2012) 337. [12] R.W. Burg, B.M. Miller, E.E. Baker, J. Birnbaum, S.A. Currie, R. Hartman, Y.L. Kong, R.L. Monaghan, G. Olson, I. Putter, J.B. Tunac, H. Wallick, E.O. Stapley, R. Oiwa, S. Omura, Antimicrob. Agents Chemother. 15 (1979) 361–367.

7

[13] D.C. Lamb, B. Zhao, F.P. Guengerich, S.L. Kelly, M.R. Waterman, in: P. Dyson (Ed.), Streptomyces Molecular Biology and Biotechnology, Caister Academic Press, Norfolk, UK, 2011, pp. 233–253. [14] H. Ikeda, T. Nonomiya, M. Usami, T. Ohta, S. Omura, Proc. Natl. Acad. Sci. U.S.A. 96 (1999) 9509–9514. [15] L.H. Xu, S. Fushinobu, S. Takamatsu, T. Wakagi, H. Ikeda, H. Shoun, J. Biol. Chem. 285 (2010) 16844–16853. [16] D. Kim, F.P. Guengerich, Biochemistry 43 (2004) 981–988. [17] D. Kim, E.T. Yukl, P. Moenne-Loccoz, P.R. Ortiz de Montellano, Biochemistry 45 (2006) 14772–14780. [18] Y.R. Lim, M.K. Hong, J.K. Kim, T.T. Doan, D.H. Kim, C.H. Yun, Y.J. Chun, L.W. Kang, D. Kim, Arch. Biochem. Biophys. 528 (2012) 111–117. [19] S. Choi, S. Han, H. Lee, Y.J. Chun, D. Kim, Biomol. Ther. 21 (2013) 487–492. [20] J.B. Schenkman, H. Remmer, R.W. Estabrook, Mol. Pharmacol. 3 (1967) 113– 123. [21] Z. Otwinowski, W. Minor, Methods Enzymol. 276 (1997) 307–326. [22] A. Vagin, A. Teplyakov, J. Appl. Crystallogr. 30 (1997) 1022–1025. [23] S. Li, D.R. Tietz, F.U. Rutaganira, P.M. Kells, Y. Anzai, F. Kato, T.C. Pochapsky, D.H. Sherman, L.M. Podust, J. Biol. Chem. 287 (2012) 37880–37890. [24] G.N. Murshudov, A.A. Vagin, E.J. Dodson, Acta. Crystallogr. D Biol. Crystallogr. 53 (1997) 240–255. [25] L. Schrodinger, The PyMOL Molecular Graphics System, version 1.3r1, 2010, http://www.pymol.org/citing. [26] R. Quaderer, S. Omura, H. Ikeda, D.E. Cane, J. Am. Chem. Soc. 128 (2006) 13036–13037. [27] M. Jastroch, A.S. Divakaruni, S. Mookerjee, J.R. Treberg, M.D. Brand, Essays Biochem. 47 (2010) 53–67. [28] H. Kioka, H. Kato, M. Fujikawa, O. Tsukamoto, T. Suzuki, H. Imamura, A. Nakano, S. Higo, S. Yamazaki, T. Matsuzaki, K. Takafuji, H. Asanuma, M. Asakura, T. Minamino, Y. Shintani, M. Yoshida, H. Noji, M. Kitakaze, I. Komuro, Y. Asano, S. Takashima, Proc. Natl. Acad. Sci. U.S.A. 111 (2014) 273–278. [29] J. Symersky, D. Osowski, D.E. Walters, D.M. Mueller, Proc. Natl. Acad. Sci. U.S.A. 109 (2012) 13961–13965. [30] R.M. Smith, W.H. Peterson, C.E. Mc, Antibiot. Chemother. 4 (1954) 962–970. [31] P.W. Yang, M.G. Li, J.Y. Zhao, M.Z. Zhu, H. Shang, J.R. Li, X.L. Cui, R. Huang, M.L. Wen, Folia Microbiol. 55 (2010) 10–16. [32] X. Lin, Y. Wen, M. Li, Z. Chen, J. Guo, Y. Song, J. Li, Appl. Microbiol. Biotechnol. 81 (2009) 839–845. [33] M. Yamazaki, T. Yamashita, T. Harada, T. Nishikiori, S. Saito, N. Shimada, A. Fujii, J. Antibiot. 45 (1992) 171–179. [34] L.H. Xu, S. Fushinobu, H. Ikeda, T. Wakagi, H. Shoun, J. Bacteriol. 191 (2009) 1211–1219. [35] B. Zhao, S.C. Moody, R.C. Hider, L. Lei, S.L. Kelly, M.R. Waterman, D.C. Lamb, Int. J. Mol. Sci. 13 (2012) 8500–8513. [36] J.R. Cupp-Vickery, Nat. Struct. Biol. 2 (1995) 144–153.