S-Adenosylhomocysteine ... - ACS Publications

Sep 1, 2017 - Tris (pH 8.0) buffer and then loaded onto a Q anion-exchange column (HiTrap-Q, GE Healthcare) for .... and thermal unfolding experiments...
0 downloads 9 Views 5MB Size
Article pubs.acs.org/biochemistry

Structural and Functional Analyses of Periplasmic 5′-Methylthioadenosine/S‑Adenosylhomocysteine Nucleosidase from Aeromonas hydrophila Yongbin Xu,*,†,‡ Lulu Wang,§ Jinli Chen,† Jing Zhao,†,‡ Shengdi Fan,†,‡ Yuesheng Dong,§ Nam-Chul Ha,∥ and Chunshan Quan*,†,‡ †

Department of Bioengineering, College of Life Science, Dalian Minzu University, Dalian 116600, Liaoning, China Key Laboratory of Biotechnology and Bioresources Utilization (Dalian Minzu University), Ministry of Education, Dalian, China § School of Life Science and Biotechnology, Dalian University of Technology, No. 2 Linggong Road, Dalian 116024, Liaoning, China ∥ Department of Agricultural Biotechnology, College of Agriculture and Life Sciences, Seoul National University, Gwanak-gu, Seoul 151-742, Republic of Korea ‡

S Supporting Information *

ABSTRACT: The Gram-negative, rod-shaped bacterium Aeromonas hydrophila has two multifunctional 5′-methylthioadenosine/S-adenosylhomocysteine nucleosidase (MTAN) enzymes, MtaN-1 and MtaN-2, that differ from those in other bacteria. These proteins are essential for several metabolic pathways, including biological methylation, polyamine biosynthesis, methionine recycling, and bacterial quorum sensing. To gain insight into how these two proteins function, we determined four high-resolution crystal structures of MtaN-1 in its apo form and in complex with the substrates S-adenosyl-L-homocysteine, 5′-methylthioadenosine, and 5′-deoxyadenosine. We found that the domain structures were generally similar, although slight differences were evident. The crystal structure demonstrates that AhMtaN-1 has an extension of the binding pocket and revealed that a tryptophan in the active site (Trp199) may play a major role in substrate binding, unlike in other MTAN proteins. Mutation of the Trp199 residue completely abolished the enzyme activity. Trp199 was identified as an active site residue that is essential for catalysis. Furthermore, biochemical characterization of AhMtaN-1 and AhMtaN-2 demonstrated that AhMtaN-1 exhibits inherent trypsin resistance that is higher than that of AhMtaN-2. Additionally, the thermally unfolded AhMtaN-2 protein is capable of refolding into active forms, whereas the thermally unfolded AhMtaN-1 protein does not have this ability. Examining the different biochemical characteristics related to the functional roles of AhMtaN-1 and AhMtaN-2 would be interesting. Indeed, the biochemical characterization of these structural features would provide a structural basis for the design of new antibiotics against A. hydrophila.

T

macromolecules, including proteins, carbohydrates, lipids, and small molecules such as sterols and nucleosides.4 MTAN catalyzes the irreversible cleavage of the N9−C1′ bond in SAH, MTA, and 5′-DOA to yield S-methyl-5′-thioribose (MTR), S-ribosyl-L-homocysteine (SRH), and 5′-deoxyribose, respectively.7 A recent study revealed that MTAN is also involved in the futalosine pathway and is crucial for the biosynthesis of the essential prokaryotic respiratory compound menaquinone.8 Thus, during this process, highly reactive product inhibitors of these reactions could cause the accumulation of MTA, SAH, and 5′-DAO within bacterial cells and result in a wide range of physiological consequences, such as the inhibition of radical SAM enzymes.9 Therefore, because inhibition of this enzyme is critical for the regulation of cellular processes in bacteria, it is a

he Gram-negative bacterium Aeromonas hydrophila is a facultatively anaerobic, opportunistic pathogen in hosts with an impaired local or general defense mechanism. This pathogen readily infects humans, fish, and other lower vertebrates and is widely distributed in stagnant and flowing fresh water, at the interface between seawater and fresh water, and in sewage.1 According to the literature, A. hydrophila infections can take the following four forms: acute diarrheal disease, cellulitis, sepsis, and other infections.2,3 Therefore, identifying new drug targets and understanding the underlying mechanisms of these targets are crucial for the development of new and effective anti-A. hydrophila agents. A. hydrophila has an important methylthioadenosine/ S-adenosylhomocysteine (MTA/SAH) nucleosidase (MTAN) that is closely associated with the metabolism of 5′-methylthioadenosine (MTA), S-adenosyl-L-homocysteine (SAH), and 5′-deoxyadenosine (5′-DAO) generated from various S-adenosylmethionine (SAM) utilization pathways,4−6 which are the primary methyl donors for the methylation of DNA and other © XXXX American Chemical Society

Received: July 20, 2017 Revised: August 30, 2017 Published: September 1, 2017 A

DOI: 10.1021/acs.biochem.7b00691 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

and 2 mM β-mercaptoethanol; and disrupted by sonication. The cell debris was removed by centrifugation at 45000g for 30 min at 277 K. The resulting supernatant fractions were loaded onto a nickel-nitrilotriacetic acid (Ni-NTA) affinity resin (GE Healthcare) that had been preincubated with Tris buffer, and the mixture was then stirred for 30 min at 277 K. The resin was washed with lysis buffer supplemented with 20 mM imidazole and eluted with 30 mL of lysis buffer containing 250 mM imidazole. The fractions containing MtnN-1 were pooled, and 2-mercaptoethanol was added to a final concentration of 10 mM. To remove the hexahistidine tag, this solution was incubated overnight with a recombinant TEV protease at 298 K. The reaction mixture was diluted 4-fold with 20 mM Tris (pH 8.0) buffer and then loaded onto a Q anion-exchange column (HiTrap-Q, GE Healthcare) for further purification. The fractions containing the target recombinant proteins were further purified using a HiLoad Superdex 200 gel-filtration column (GE Healthcare) pre-equilibrated with lysis buffer. The purities of the final proteins were confirmed by sodium dodecyl sulfate−polyacrylamide gel electrophoresis (SDS−PAGE) on a 15% gel and stained with Coomassie blue. The purified proteins were concentrated using Centriprep (Millipore) and stored at 277 K for use within a week or frozen at 211 K until they were used. Site-directed mutagenesis was performed by two subsequent PCRs,13 and the proteins were expressed and purified under the same conditions as the AhMtaN-2 protein. Crystallization, Data Collection, and Structure Determination and Refinement. The crystallization and preliminary X-ray analysis of AhMtaN-1 (residues 31−278, lacking the signal sequence) have been published previously.12 The structure of the AhMtaN-1 protein was determined by molecular replacement (MR) using the Collaborative Computational Project, Number 4 (CCP4) software package.14 We used the coordinates of the structures of H. pylori MTAN (9.4% sequence identity, PDB entry 3NM415) as the search model. To build our protein model, we first removed the model bias via rounds of simulated annealing performed with the PHENIX software16 and then calculated the differences using Fourier maps. To obtain a crystal of AhMtaN-1 in complex with its substrates, high-quality crystals of AhMtaN-1 were gently removed from the crystallization drop and incubated for 3 s in a solution consisting of 6% (w/v) polyethylene glycol 4000, 0.1 M sodium acetate trihydrate (pH 4.6), and 0.5 mM SAH, MTA, or 5′-DAO. High-resolution data for refinement were collected on Pohang Light Source (PLS, Pohang, South Korea) beamline 5C and Shanghai Synchrotron Radiation Facility (SSRF, Shanghai, China) beamline BL17U1 at 100 K with an ADSC Quantum 315r charge-coupled device (CCD) detector.17 The structures of AhMtaN-1-SAH, AhMtaN-1-MTA, and AhMtaN-1-DOA were determined by an MR approach using AhMtaN-1 as a search model. Iterative model building and density modification were conducted to place the remaining residues by refinement in PHENIX.16 Ligands and water molecules were also added by manually inspecting a 2F0 − Fc electron density map as a guide in Crystallographic ObjectOriented Toolkit (COOT).18 The final refined model had crystallographic Rfactor and Rfree values that were within the range of average values for structures refined at these resolutions.19 Model Quality. The PROCHECK5 program in the CCP4 software package was used to check the geometry of the refined model. The Ramachandran statistics showed that 98.19% of the model residues were located in the favored regions and 1.81% in the allowed regions for apo AhMtaN-1, that 98.58% of the

potential antibiotic target. A. hydrophila has two MTAN subfamily proteins, MtaN-1 and MtaN-2, unlike other bacteria. The MtaN-1 protein appears to be a periplasmic protein because of the presence of a predicted N-terminal signal sequence for secretion, whereas AhMtaN-2 appears to be a cytosolic protein due to the lack of the predicted signal sequence (highlighted by the blue box in Figures S1 and S2). The presence of periplasmic AhMtaN-1 is interesting because MTAN should function in the cytosol to metabolize radical products. Despite accumulating evidence that MTAN plays a critical role in SAH metabolism, why A. hydrophila has two MtnN subfamily proteins remains unknown. The structures of MTAN homologues in various organisms, such as Helicobacter pylori [Protein Data Bank (PDB) entry 4YO8], Vibrio cholerae, Borrelia burgdorferi (PDB entry 4L0M), and Escherichia coli (PDB entry 4WKP), have been previously determined. These reported structures are dimeric, and the overall folding of MTAN is strongly structurally similar with that of the nucleoside phosphorylase-1 (NP-1) family of enzymes.10,11 The crystal structure of the MTAN from A. hydrophila has not yet been published. The functional and structural features of AhMtaN proteins provide a basis for the design of novel antibiotics for diseases caused by A. hydrophila. Although considerable effort has been invested in understanding the mechanisms of action of MTAN proteins, the crystal structures and molecular functions of AhMtaN proteins remain largely uncharacterized. Moreover, no studies have investigated the structure or function of periplasmic MTAN. In this study, we present the first crystal structure of AhMtaN-1 and its complex structures with the substrates SAH (AhMtaN1-SAH), MTA (AhMtaN-1-MTA), and 5′-DAO (AhMtaN-1DAO) at high resolution. These structural snapshots provide insight into the conformational flexibility of AhMtaN-1 during catalysis. In addition, to compare the biochemical features of these two proteins, AhMtaN-1 and AhMtaN-2, we employed circular dichroism (CD) spectroscopy and limited trypsin digestion and observed significant differences between these homologous proteins.



MATERIALS AND METHODS Protein Preparation. The recombinant AhMtaN-1 (residues 21−278, lacking the signal sequence) protein was produced and purified as described previously.12 To express AhMtaN-2 (residues 1−230), we amplified DNA fragments encoding A. hydrophila MtaN-2 from the genomic DNA of an A. hydrophila strain by polymerase chain reaction (PCR) using standard methods. The resulting PCR product was inserted between the EcoRI and HindIII sites of the pProEx-HTa vector (Invitrogen) to generate a protein containing a hexahistidine tag and tobacco etch virus (TEV) protease cleavage site at the N-terminus of AhMtaN-2. The primers were MtaN-2_his6_forward (5′-ggggaattcatgaaagtaggtattatcggcgc-3′) and MtaN2_his-6_reverse (5′-gggaagctttcacagcttgccgagcatcttgatgat-3′), and the resulting plasmid was named pPROTEXHTA-MtaN-2. The plasmid was transformed into E. coli BL21 (DE3) competent cells to produce the AhMtaN-2 protein. The cells were grown in 1 L of lysogeny broth (LB) medium containing 0.5 μg mL−1 ampicillin at 310 K to an optical density at 600 nm (OD600) of 0.6−0.8, when protein overexpression was induced by adding 0.5 mM isopropyl D-thiogalactoside (IPTG). The temperature was then decreased to 303 K, and growth was continued for 16 h. The cells were harvested by centrifugation; suspended in a lysis buffer containing 20 mM Tris (pH 8.0), 150 mM NaCl, B

DOI: 10.1021/acs.biochem.7b00691 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

AhMtaN-2, AhMtaN-1 W199A, and AhMtaN-1 W199L were examined using a xanthine oxidase-coupled spectrophotometric assay.21 For each molecule of adenine oxidized by xanthine oxidase, two molecules of 2-(4-iodophenyl)-3-(4-nitrophenyl)5-phenyl tetrazolium chloride (INT) are reduced to formazan, which absorbs at 470 nm.22 The reaction mixtures (800 μL) consisted of 100 mM HEPES, 50 mM potassium phosphate (pH 7.0), 5 μg of AhMtaN-1 or AhMtaN-2, 100 μM SAH, 0.28 unit of xanthine oxidase (Sigma-Aldrich Chemicals), and 1 mM INT (Sigma-Aldrich Chemicals) in a final volume of 800 μL.

model residues were located in the favored regions and 1.42% in the allowed regions for AhMtaN-1-SAH, that 98.56% of the model residues were located in the favored regions and 1.44% in the allowed regions for AhMtaN-1-MTA, and that 98.58% of the model residues were located in the favored regions and 1.42% in the allowed regions for AhMtaN-1-DOA. Pictures of the structural models were generated using PyMOL.20 PDB Entries. The atomic coordinates and structure factors were deposited in the PDB with the following identification codes: 5B7G for AhMtaN-1, 5B7N for AhMtaN-1-SAH, 5B7P for AhMtaN-1-MTA, and 5B7Q for AhMtaN-1-DOA. Size-Exclusion Chromatography. To determine the molecular sizes of AhMtaN-1 and AhMtaN-2, size-exclusion chromatography was performed. To this end, 200 μL of the AhMtaN-1 protein (1.5 mg/mL) or AhMtaN-2 protein (0.8 mg/mL) was injected into a Superdex 200 10/300 GL column (GE Healthcare) at room temperature (295 K) at a flow rate of 0.5 mL/min equilibrated with 20 mM Tris buffer (pH 8.0) containing 150 mM NaCl and 2 mM 2-mercaptoethanol. Limited Trypsin Digestion. The AhMtaN-1 and AhMtaN-2 proteins were mixed at a concentration of 1.3 mg/mL in an assay buffer consisting of 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) (pH 8.0) and 150 mM NaCl with 0.05 mg/mL trypsin. The reaction mixture was incubated for 2−16 min at 310 K in a water bath. The negative control was the reaction mixture without trypsin. After the trypsin treatment, Loading Buffer Blue (2X) [0.5-M Tris-HCl (pH 6.8), 4.4% (w/v) SDS, 20% (v/v) glycerol, 2% (v/v) 2-mercaptoethanol, and bromophenol blue in distilled/ deionized water] was added to stop the reaction. After the samples had been boiled for 5 min, the limited trypsin digestion was verified by 15% SDS−PAGE and Coomassie brilliant blue staining. CD Spectroscopy. To evaluate the structural integrity of the AhMtaN-1 and AhMtaN-2 proteins, CD spectroscopy was performed using a JASCO-J1500 spectropolarimeter (JASCO). Each purified protein was centrifuged at 14000g for 10 min at 277 K, and the supernatant was diluted to 0.1 mg/mL in 20 mM HEPES (pH 7.5) buffer containing 150 mM NaCl. The protein concentration was measured on the basis of the ultraviolet (UV) absorption of the Tyr and Trp residues of the protein at 280 nm or by Coomassie brilliant blue staining of the protein bands after SDS−PAGE; bovine serum albumin was used as the standard for both assays. To prepare heat-treated protein samples, the two proteins were heated for 5 min at 373 K in a water bath, followed by cooling from 296 K to room temperature. The absorption spectra were recorded on a JASCO-J1500 spectropolarimeter at 298 K. Three consecutive scans were performed and averaged, and the solvent signal was subtracted from all of the spectra. Thermal Denaturation Studies (Tm) Performed Using CD Spectroscopy. The thermostability of the secondary structures of AhMtaN-1 and AhMtaN-2 was determined by CD spectroscopy using a JASCO-J1500 spectropolarimeter. Samples were prepared in 20 mM HEPES (pH 7.5) containing 150 mM NaCl, and thermal unfolding experiments were performed by monitoring the CD signal at 222 nm between 298 and 370 K at a heating rate of 1 °C/min and a concentration of 0.2 mg/mL. Enzyme Activities of AhMtaN-1, AhMtaN-2, AhMtaN-1 W199A, and AhMtaN-1 W199L. To investigate the functional abilities of these two proteins, the activities of AhMtaN-1,



RESULTS Structural Determination of AhMtaN-1 and AhMtaN-1 in Complex with SAH, MTA, and DOA. Although several MTAN subfamily proteins have been extensively studied, why A. hydrophila has two MtnN subfamily proteins remains unknown. To clarify this issue, we determined the structure of AhMtaN-1. We initially attempted to determine the crystal structure of AhMtaN-2 but were unsuccessful because of a crystal diffraction problem. We eventually successfully determined the crystal structures of AhMtaN-1 alone and in complex with SAH, MTA, and DOA at high resolution using synchrotron radiation. For all four structures, all 249 residues of the protein were clearly visible in the electron density map and were modeled. The apo and SAH-, MTA-, and DOA-complexed structures are members of the trigonal space group (P3121) and were refined to values of Rfactor of 14.87% (Rfree = 16.36%), 17.54% (Rfree = 19.82%), 13.40% (Rfree = 15.55%), and 15.31% (Rfree = 17.68%), respectively. Analysis of the structures in PROCHECK5 revealed that none of the nonglycine residues fell into the disallowed region of the Ramachandran plot. Further details regarding the data processing and structural determination statistics are listed in Table 1. Overall Structure of A. hydrophila MtaN-1 and Comparison with the Most Homologous Structures. The crystal structure of AhMtaN-1 demonstrates that this protein’s overall structure is folded into a single domain consisting of 249 residues (Ala27−Phe275), and significant electronic density was obtained for all residues, except the N-terminally located Ala2-Ala3-Thr4-Pro5 residues and the C-terminally located Asn276-Lys277-Ala-278 residues, similar to previously deposited structures. The N-terminus and C-terminus are thought to be very mobile in these proteins, which is why the models deposited to date are missing these regions. The core structure is the α−β−α sandwich fold, in which nine β-strands form a central parallel β-sheet surrounded by six α-helices and a small 3 helix (Figure 1B). Two short β-strands, β8 (two residues) and β9 (two residues), are part of a long loop that connects α5 and α6. Two monomers in the crystallographic asymmetric unit form a canonical dimer by interacting with a neighboring molecule that is related by a crystallographic 2-fold axis. The dimeric structure has a heart symbol shape, as observed for other MTAN proteins (Figure 1A). These two molecules are similar to each other, and the rootmean-square deviation (RMSD) for the position 214 Cα atom is 0.10 Å. The interactions and hydrophobic residues in the interfaces are well-conserved in AhMtaN-1 and were involved in dimer formation in the crystals. The interface is formed by three α-helices (α2, α4, and α5) and two β-strands (β8 and β9) (Figure 1B). The crystal structure strongly suggests that the dimeric unit is the functional form of AhMtaN-1. To confirm whether AhMtaN-1 is a dimer in solution, we measured the molecular size of AhMtaN-1 using size-exclusion chromatography C

DOI: 10.1021/acs.biochem.7b00691 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry Table 1. Diffraction Statisticsa source PDB entry wavelength (Å) space group resolution (Å) parameters (Å) Rsym (%) completeness (%) redundancy average I/σ(I) Rfactor (%) Rfree (%) root-mean-square deviation for bonds (Å) root-mean-square deviation for angles (deg) Ramachandran plot (%) favored regions allowed regions disallowed regions number of atoms protein water ligand average B factor (Å2)

apo

SAH

MTA

5′-DOA

beamline 5C at PLS 5B7G 0.985 P3121 29.64−1.40 a = b = 102.7 Å, c = 118.8 Å, α = β = 90°, γ = 120° 6.4 (33.0) 99.8 (99.3) 8.8 (6.8) 40.46 (3.95) 14.87 16.36 0.006

beamline 5C at PLS 5B7N 0.985 P3121 49.49−1.40 a = b = 103.1 Å, c = 118.9 Å, α = β = 90°, γ = 120° 9.9 (44.5) 99.2 (96.5) 6.5 (3.6) 14.79 (1.59) 17.54 19.82 0.006

BL17U1 at SSRF 5B7P 0.903 P3121 24.16−1.49 a = b = 102.7 Å, c = 118.7 Å, α = β = 90°, γ = 120° 7.2 (25.3) 99.6 (93.4) 16.8 (11.5) 37.59 (7.41) 13.40 15.55 0.009

BL17U1 at SSRF 5B7Q 0.903 P3121 24.25−1.49 a = b = 103.1 Å, c = 119.0 Å, α = β = 90°, γ = 120° 10.4 (49.7) 99.7 (98.4) 12.7 (7.8) 19.90 (3.95) 15.31 17.68 0.009

1.088

1.066

1.268

1.268

98.19 1.81 0.00

98.58 1.42 0.00

98.56 1.44 0.00

98.58 1.42 0.00

3863 783 20 15.41

3863 749 52 14.29

7522 759 70 15.95

7546 711 62 17.24

Rmerge = ∑hkl∑i|Ii(hkl) − ⟨I(hkl)⟩|∑hkl∑i|Ii(hkl), where I(hkl) is the intensity of reflection hkl, ∑hkl is the sum over all reflections, and ∑i is the sum over i measurements of reflection hkl. Rwork = ∑hkl}F0 − Fc|/∑hkl|F0| for all data with F0 >2σ(F0), excluding data used to calculate Rfree. Rfree = ∑hkl|F0 − Fc|/∑hkl|F0| for all data with F0 >2σ(F0) that were excluded from refinement. a

On the basis of the amino acid sequence analysis of AhMtaN-1 relative to other MTANs, only one residue (Trp199 of AhMtaN-1) is conserved among the AhMtaN-1 proteins (Figure 2A). To determine whether Trp199 of AhMtaN-1 is responsible for substrate binding, we analyzed the crystal structure of AhMtaN-1 complex structures with the substrates SAH (AhMtaN-1-SAH), MTA (AhMtaN-1-MTA), and 5′-DAO (AhMtaN-1-DAO) by the MR method using the AhMtaN-1 structure as a search model. The resulting models of AhMtaN1-SAH, AhMtaN-1-MTA, and AhMtaN-1-DAO were refined to data sets with resolutions of 1.4, 1.49, and 1.49 Å, respectively. As expected, we observed that an asymmetric unit of the crystal contains two AhMtaN-1 molecules and two substrate molecules and that each AhMtaN-1 promoter binds one substrate molecule. Additionally, the initial σA-weighted F0 − Fc map revealed strong electron density corresponding to SAH, MTA, and 5′-DAO in the crystal structures of complexed AhMtaN-1. In the crystal structure, we observed that the indole ring of Trp199 has a base-stacking interaction with the adenine ring and that the substrate binding site is located in the region surrounded by the loop between α7 and α8 and the loop between β10 and α7. This site is composed of residues Met37, Val78, Thr104, and Met220 in the key conserved regions of AhMtaN-1 across species, and these residues are inserted into the active site cleft of the substrate (Figure 2B−D). The values of the B factors in these residues are higher than those for other residues in β1, β4, β5, α1, and α2, suggesting that these residues are structurally heterogeneous and conformationally flexible. Residues Met37, Val78, and Met220 are also conserved in all other proteins in this family (Figure S3).

(Figure S3). The calculated molecular weight from the sizeexclusion column was located in the middle of the monomer and dimer, indicating that this protein may exist as a dimer in solution, which is consistent with the crystal structure. We used the DaliLite server23 to conduct a search to identify unique proteins that are structurally similar to AhMtaN-1, revealing that EcMTAN (PDB entry 3DF9) is highly similar to AhMtaN-1. Indeed, EcMTAN was indicated as the top match with a Z-score of 32.0 and a sequence identity of ≤27%. This protein has been annotated as a bacterial MTAN protein.21 The structural superposition of AhMtaN-1 on EcMTAN revealed that these proteins have remarkably structurally similar but are not identical; some minor structural differences were observed, as shown in the superimposed structures (Figure 1C). In particular, sequence and structural analyses revealed that the major difference between these two proteins is that AhMtaN-1 has a longer extension region with a variable amino acid sequence in the protruding loops (L1 and L2) that consist of 38 residues (residues 125−162) (Figure 1C,D). Additionally, these loops are not found in other MTAN subfamily proteins and seem to be a unique feature of AhMtaN-1. Further sequence features unique to the MTAN subfamily proteins are highlighted in the red box in Figure S3. Active Site Identification. Interestingly, in the crystal structure of AhMtaN-1, we found a deep pocket formed by the loop between α7 and α8 and the loop between β10 and α7. Furthermore, we observed that the electron density map around Trp199 clearly shows dual electron density throughout the refinement process; thus, the crystal structure represents an alternative conformation (Figure 2A). This finding suggests that Trp199 may be required for substrate binding and catalysis. D

DOI: 10.1021/acs.biochem.7b00691 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Figure 1. Overall structure of AhMtaN-1. (A) Structure of AhMtaN-1 in the trigonal crystal form displayed as ribbons. The asymmetric unit contains two protomers. (B) Secondary structural elements of AhMtaN-1 are numbered. (C) Structural comparison of the cartoon traces of AhMtaN-1 and EcMTAN (PDB entry 3DF9). AhMtaN-1 is colored green, and EcMTAN is colored cyan. (D) Close-up of part of AhMtaN-1. Sequence alignment of AhMtaN-1 with homologous proteins. Strictly conserved residues are outlined in black, and α3 and α4 of AhMtaN-1 are highlighted in light pink. The residues of α3 and α4 in AhMtaN-1 are strictly conserved in the AhMtaN-1 protein only but not in AhMtaN-2, E. coli MTAN (EcMtaN), or H. pylori MTAN (HpMtaN).

differences between AhMtaN-1 and AhMtaN-2 affect their susceptibility to proteolysis, both proteins were subjected to limited proteolysis with trypsin under identical conditions (as described in Materials and Methods). The digested samples were analyzed by SDS−PAGE. After trypsin treatment, for AhMtaN-2, we observed several discrete bands with molecular masses lower than that of the full-length protein (Figure 3A), whereas for AhMtaN-1, no such bands were noted. Thus, AhMtaN-1 is more resistant to trypsin digestion than AhMtaN-2 is. No intact AhMtaN-2 was detected after digestion for 5 min, but most of the AhMtaN-1 proteins remained intact, even after proteolysis for 16 min. AhMtaN-2 Undergoes a Thermal Unfolding/Refolding Process. Our most surprising finding was that when AhMtaN-1 and AhMtaN-2 were subjected to high-temperature treatment, the AhMtaN-1 protein precipitated but the AhMtaN-2 protein did not (Figure S4). We performed activity assays to investigate whether AhMtaN-1 and AhMtaN-2 retain their activities after high-temperature treatment. Intriguingly, under identical experimental conditions, heat-treated AhMtaN-2 still had activity whereas AhMtaN-1 did not (Figure 3B). To examine whether the heat-treated proteins have secondary structural integrity, we then analyzed recorded CD spectra. The CD spectra of the two proteins without heat treatment were very similar, with the patterns of mixed α-helix and β-sheet suggesting that both are properly folded because the spectra are

To further verify our conclusion, we designed a site-directed mutant of AhMtaN-1 in which the Trp199 residue was substituted with an alanine residue or a leucine residue. We produced a truncated form of the mutant proteins (W199A and W199L) in E. coli using the same procedure that was used for the wild-type (WT) AhMtaN-1 protein. The activities of WT AhMtaN-1 and its mutants were quantified spectrophotometrically by monitoring the production of formazan at 470 nm.21 As expected, the catalytic efficiency of the AhMtaN-1 mutants toward MTA was lower than that of the WT AhMtaN-1 protein (Figure 2E). In particular, the AhMtaN-1 W199A protein exhibited the largest decrease in catalytic efficiency. However, the W199L mutant displayed only partially decreased activity. These results demonstrate that the unconserved Trp199 residue plays a crucial role in substrate binding and that the AhMtaN-1 protein has a hydrophobic cavity and the substrate is binding by ϕ−ϕ interaction between the imidazole ring of Trp199 and the ring of the MTA substrate. Thus, this deep pocket might be a candidate target site for an inhibitor to abolish the activity of AhMtaN-1. Comparison of the Results of Biochemical Characterization of AhMtaN-1 and AhMtaN-2. In this study, we observed a significant difference between these homologous proteins according to the results of an in vitro assay. AhMtaN-1 Is More Strongly Protected from Proteolysis Than AhMtaN-2 Is. To investigate how the structural E

DOI: 10.1021/acs.biochem.7b00691 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Figure 2. Active sites in AhMtaN-1. (A) The alternative conformation of the Trp199 residue is colored yellow. The 2F0 − Fc map of the ligand is contoured at the 1σ level (gray). (B) Active site of AhMtaN-1 in complex with the substrate SAH. The 2F0 − Fc map around the ligand is contoured at the 1σ level (gray). (C) Active site of AhMtaN-1 in complex with the substrate MTA. The 2F0 − Fc map around the ligand is contoured at the 1σ level (gray). (D) Active site of AhMtaN-1 in complex with the substrate DOA. The 2F0 − Fc map around the ligand is contoured at the 1σ level (gray). (E) Activities of the wild-type enzyme and its mutants (W199A and W199L). The graph represents the mean of three independent experiments, and the standard deviation is indicated by the error bars.

have coincident values, with melting temperatures of 55 and 54 °C, respectively. These data suggest that both proteins undergo a similar thermal unfolding transition (Figure 3D).

consistent with the structure of MTAN subfamily proteins. The samples were heated to 373 K, followed by cooling from 373 K to room temperature, and a further spectrum was obtained (Figure 3B). The results indicate that the heated AhMtaN-2 protein was still able to properly fold but that AhMtaN-1 was not (Figure 3C). To provide additional information, we also analyzed the changes in the secondary structural characteristics by performing CV-based melting temperature studies using a JASCO-J1500 spectropolarimeter. As the temperature increased, the loss of the secondary structure (or unfolding) was monitored at 222 nm to generate melting curves and calculate the melting temperature. The thermal denaturation curves suggest that both proteins (AhMtaN-1 and AhMtaN-2)



DISCUSSION To date, most of the molecular functions and the mechanism of action of MTAN have been determined. However, why A. hydrophila contains two MTAN subfamily proteins (AhMtaN-1 and AhMtaN-2) remains unknown. This paper reports the first three-dimensional structural insight into the functional and structural properties of MtaN-1 from A. hydrophila. The high-resolution structure of AhMtaN-1 and substrate binding mechanism may provide clues for engineering of the substrate and F

DOI: 10.1021/acs.biochem.7b00691 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Figure 3. Biochemical comparison of AhMtaN-1 and AhMtaN-2. (A) Comparison of the resistance to trypsin digestion of AhMtaN-1 and AhMtaN-2. Trypsin and AhMtaN-1 or AhMtaN-2 were incubated at 310 K for 2−16 min in a water bath (see Materials and Methods). (B) Activities of the native (AhMtaN-1 and AhMtaN-2) and heat-treated proteins (HT-AhMtaN-1 and HT-AhMtaN-2). The graph represents the mean of three independent experiments, and the standard deviation is indicated by error bars. (C) CD spectra of native (AhMtaN-1 and AhMtaN-2) and heattreated proteins (HT-AhMtaN-1 and HT-AhMtaN-2). The CD spectrum of AhMtaN-1 is shown in the top panel, and that of AhMtaN-2 is presented in the bottom panel. (D) Comparison of the melting temperatures (Tm) of AhMtaN-1 and AhMtaN-2. Tm values were measured using CD spectroscopy. The arrows indicate the Tm value of each protein.

MTAN sequences from various bacterial pathogens produced degrees of sequence identity that ranged from 30 to 100%. Strikingly, this loop is not found in other MTAN subfamily proteins and seems to be a unique feature of AhMtaN-1 (Figure 1D and Figure S3). In the crystal structure of AhMtaN-1, we observed two protruding loops (L1 and L2) in close contact with another molecule (Figure 1A), which suggests that the conformation of the two protruding loops may be critical for the molecular mechanism of AhMtaN-1 dimerization (Figure 1A,C). The active sites of MTAN homologues have been identified from substrate-bound MTAN complexes in a variety of bacteria.10,29 The structure of MTAN shows that the architectures of its active site and the nucleoside−enzyme interactions are identical.21,30 To find the substrate binding site of AhMtaN-1, we determined the crystal structures of

preventing inhibition of MTAN family proteins. Comparing the architectures of AhMtaN-1 and EcMTAN revealed striking similarity, despite their different domain orientations. We attempted to confirm that AhMtaN-1 exists as a dimer in solution, but our result suggested that AhMtaN-1 does not have a propensity to form a dimer; however, this result is inconsistent with the crystal structure. The asymmetric units in every case consisted of the AhMtaN-1 homodimer; the crystal structure of AhMtaN-1 also strongly suggests that the functional form is a dimer.8,10,21,24−28 Taken together, these results indicate that the dimeric structure of AhMtaN-1 reflects the solution state of AhMtaN-1 and is functionally important in vivo. The crystal structure of AhMtaN-1 clearly showed that AhMtaN-1 has a protruding loop (L1) consisting of 38 residues (residues 125−162) (Figure 1C). Alignments of available primary G

DOI: 10.1021/acs.biochem.7b00691 Biochemistry XXXX, XXX, XXX−XXX

Biochemistry



AhMtaN-1 complexed with SAH, MTA, and DOA and found that each AhMtaN-1 subunit binds one substrate molecule in a hydrophobic pocket formed by the loop between α7 and α8 and the loop between β10 and α7. Importantly, Trp199 forms ϕ−ϕ interaction with the substrates in the complex structure. The Trp199 residue is also conserved in A. hydrophila MtaN-1 but is variable in all other enzymes of this family (Figure 2A). As this residue is replaced with Phe (residue 150) in EcMtaN, we investigated whether the proposed Trp199 residue is essential for substrate processing by mutating Trp199 to alanine and leucine. This mutational study suggested that the Trp199 residue of AhMtaN-1 is involved in substrate recognition and the nonconserved amino acid Trp199 plays a crucial role in substrate binding (Figure 2E). In this study, we also observed a significant difference between these homologous proteins (AhMtaN-1 and AhMtaN-2) in an in vitro assay. Our limited proteolysis experiment and thermal denaturation studies suggest that AhMtaN-1 and AhMtaN-2 differ significantly. According to the data from the limited proteolysis experiments of AhMtaN-1 and AhMtaN-2 using trypsin, AhMtaN-1 was largely protected from proteolysis, whereas AhMtaN-2 exhibited no inherent trypsin resistance (Figure 3A). Thermal denaturation studies performed under identical experimental conditions revealed that the heated AhMtaN-2 protein was still able to properly fold but that AhMtaN-1 was not (Figure 3C). We also observed that high-temperature-treated AhMtaN-2 retained its activity (Figure 3D). These results indicate that the thermally unfolded AhMtaN-2 protein is capable of refolding to active forms, whereas the thermally unfolded AhMtaN-1 protein does not have this ability. No MTAN subfamily protein has been previously reported to undergo a thermal unfolding/refolding process. Although we observed a significant difference between these homologous proteins during extensive structural and biochemical studies, we did not fully clarify some key issues. Therefore, further experiments are required to clarify why these two proteins have such different biochemical characteristics and elucidate their relationship with protein function.



Article

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biochem.7b00691. Additional figures and tables of collected crystallographic data and refinement statistics and structural characteristics (PDF) Accession Codes

The structural factors and atomic coordinates of apo AhMtaN-1, AhMtaN-1-SAH, AhMtaN-1-MTA, and AhMtaN-1-DOA have been deposited as PDB entries 5B7G, 5B7N, 5B7P, and 5B7Q, respectively.



AUTHOR INFORMATION

Corresponding Authors

*Phone: +86-411-8765-6045. Fax: +86-411-8765-6219. E-mail: [email protected]. *Phone: +86-411-8765-6219. E-mail: [email protected]. ORCID

Yongbin Xu: 0000-0002-6591-9320 Yuesheng Dong: 0000-0001-5010-6426 Author Contributions

Y.X., L.W., and J.C. contributed equally to this work. Funding

This work was supported by the National Natural Science Foundation of China (Grant 31200556 to Y.X., Grant 21272031 to C.-S.Q., and Grant 21172028 to S.F.), the Program for Liaoning Excellent Talents in University (Grant LJQ2015030 to Y.X.) and the Fundamental Research Funds for the Central Universities (Grant DC201502020203 to Y.X.). Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS We thank the staff of the 5C beamline at the PLS and the BL17U1 beamline at the SSRF for their help with data collection. REFERENCES

(1) Park, S. Y., Nam, H. M., Park, K., and Park, S. D. (2011) Aeromonas hydrophila Sepsis Mimicking Vibrio vulnificus Infection. Ann. Dermatol. 23 (Suppl. 1), S25−29. (2) Davis, W. A., 2nd, Kane, J. G., and Garagusi, V. F. (1978) Human aeromonas infections: a review of the literature and a case report of endocarditis. Medicine (Philadelphia, PA, U. S.) 57, 267−277. (3) Janda, J. M., and Abbott, S. L. (2010) The genus Aeromonas: taxonomy, pathogenicity, and infection. Clin. Microbiol. Rev. 23, 35− 73. (4) Parveen, N., and Cornell, K. A. (2011) Methylthioadenosine/Sadenosylhomocysteine nucleosidase, a critical enzyme for bacterial metabolism. Mol. Microbiol. 79, 7−20. (5) Longshaw, A. I., Adanitsch, F., Gutierrez, J. A., Evans, G. B., Tyler, P. C., and Schramm, V. L. (2010) Design and synthesis of potent ″sulfur-free″ transition state analogue inhibitors of 5′methylthioadenosine nucleosidase and 5′-methylthioadenosine phosphorylase. J. Med. Chem. 53, 6730−6746. (6) Silva, A. J., Parker, W. B., Allan, P. W., Ayala, J. C., and Benitez, J. A. (2015) Role of methylthioadenosine/S-adenosylhomocysteine nucleosidase in Vibrio cholerae cellular communication and biofilm development. Biochem. Biophys. Res. Commun. 461, 65−69. (7) Duerre, J. A. (1962) Preparation and properties of S-adenosyl-Lhomocysteine, S-adenosyl-L-homocysteine sulfoxide and S-ribosyl-Lhomocysteine. Arch. Biochem. Biophys. 96, 70−76.

CONCLUSION

In conclusion, we determined the high-resolution dimeric crystal structures of AhMtaN-1 alone and AhMtaN-1 in complex with SAH, MTA, and DOA. These structural snapshots provided insight into the conformational flexibility of AhMtaN-1 during catalysis. The overall folding of AhMtaN-1 is highly conserved in this superfamily. A consensus substrate binding site has been identified in the MTAN homologues that bind these substrates.8,10,31 Structural analysis results revealed differences suggesting that the Trp199 residue in AhMtaN-1 is involved in substrate binding and that the protein has a hydrophobic cavity that binds ligands. We were unable to obtain a crystal structure of AhMtaN-2, but we compared the results of biochemical characterization of AhMtaN-1 and AhMtaN-2. This comparison revealed that AhMtaN-1 is more resistant to trypsin digestion than AhMtaN-2 is. More importantly, AhMtaN-2 is more stable against heat treatment than AhMtaN-1 is. Examining the difference in the results of biochemical characterization relevant to the functional roles of AhMtaN-1 and AhMtaN-2 would be interesting. H

DOI: 10.1021/acs.biochem.7b00691 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry (8) Kim, R. Q., Offen, W. A., Davies, G. J., and Stubbs, K. A. (2014) Structural enzymology of Helicobacter pylori methylthioadenosine nucleosidase in the futalosine pathway. Acta Crystallogr., Sect. D: Biol. Crystallogr. 70, 177−185. (9) Beeston, A. L., and Surette, M. G. (2002) pfs-dependent regulation of autoinducer 2 production in Salmonella enterica serovar Typhimurium. J. Bacteriol. 184, 3450−3456. (10) Lee, J. E., Cornell, K. A., Riscoe, M. K., and Howell, P. L. (2003) Structure of Escherichia coli 5′-methylthioadenosine/ S-adenosylhomocysteine nucleosidase inhibitor complexes provide insight into the conformational changes required for substrate binding and catalysis. J. Biol. Chem. 278, 8761−8770. (11) Lee, J. E., Singh, V., Evans, G. B., Tyler, P. C., Furneaux, R. H., Cornell, K. A., Riscoe, M. K., Schramm, V. L., and Howell, P. L. (2005) Structural rationale for the affinity of pico- and femtomolar transition state analogues of Escherichia coli 5′-methylthioadenosine/ S-adenosylhomocysteine nucleosidase. J. Biol. Chem. 280, 18274− 18282. (12) Xu, Y., Quan, C. S., Jin, X., Jin, X., Zhao, J., Jin, L., Kim, J. S., Guo, J., Fan, S., and Ha, N. C. (2015) Crystallization and preliminary X-ray diffraction analysis of the interaction of Aeromonas hydrophila MtaN-1 with S-adenosylhomocysteine. Acta Crystallogr., Sect. F: Struct. Biol. Commun. 71, 393−396. (13) Landt, O., Grunert, H. P., and Hahn, U. (1990) A general method for rapid site-directed mutagenesis using the polymerase chain reaction. Gene 96, 125−128. (14) McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C., and Read, R. J. (2007) Phaser crystallographic software. J. Appl. Crystallogr. 40, 658−674. (15) Ronning, D. R., Iacopelli, N. M., and Mishra, V. (2010) Enzymeligand interactions that drive active site rearrangements in the Helicobacter pylori 5′-methylthioadenosine/S-adenosylhomocysteine nucleosidase. Protein Sci. 19, 2498−2510. (16) Adams, P. D., Afonine, P. V., Bunkoczi, G., Chen, V. B., Davis, I. W., Echols, N., Headd, J. J., Hung, L. W., Kapral, G. J., GrosseKunstleve, R. W., McCoy, A. J., Moriarty, N. W., Oeffner, R., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T. C., and Zwart, P. H. (2010) PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr., Sect. D: Biol. Crystallogr. 66, 213−221. (17) Wang, Z., Pan, Q., Yang, L., Zhou, H., Xu, C., Yu, F., Wang, Q., Huang, S., and He, J. (2016) Automatic crystal centring procedure at the SSRF macromolecular crystallography beamline. J. Synchrotron Radiat. 23, 1323−1332. (18) Emsley, P., and Cowtan, K. (2004) Coot: model-building tools for molecular graphics. Acta Crystallogr., Sect. D: Biol. Crystallogr. 60, 2126−2132. (19) Chen, V. B., Arendall, W. B., 3rd, Headd, J. J., Keedy, D. A., Immormino, R. M., Kapral, G. J., Murray, L. W., Richardson, J. S., and Richardson, D. C. (2010) MolProbity: all-atom structure validation for macromolecular crystallography. Acta Crystallogr., Sect. D: Biol. Crystallogr. 66, 12−21. (20) DeLano, W. L. (2004) Use of PYMOL as a communications tool for molecular science. Abstracts of Papers of the American Chemical Society, Vol. 228, 030-CHED, American Chemical Society, Washington, DC. (21) Siu, K. K. W., Lee, J. E., Smith, G. D., Horvatin-Mrakovcic, C., and Howell, P. L. (2008) Structure of Staphylococcus aureus 5 ′-methylthioadenosine/S-adenosylhomocysteine nucleosidase. Acta Crystallogr., Sect. F: Struct. Biol. Cryst. Commun. 64, 343−350. (22) de Groot, H., and Noll, T. (1985) Enzymic determination of inorganic phosphates, organic phosphates and phosphate-liberating enzymes by use of nucleoside phosphorylase-xanthine oxidase (dehydrogenase)-coupled reactions. Biochem. J. 230, 255−260. (23) Holm, L., and Rosenstrom, P. (2010) Dali server: conservation mapping in 3D. Nucleic Acids Res. 38, W545−W549. (24) Siu, K. K., Lee, J. E., Sufrin, J. R., Moffatt, B. A., McMillan, M., Cornell, K. A., Isom, C., and Howell, P. L. (2008) Molecular

determinants of substrate specificity in plant 5′-methylthioadenosine nucleosidases. J. Mol. Biol. 378, 112−128. (25) Lee, J. E., Cornell, K. A., Riscoe, M. K., and Howell, P. L. (2001) Structure of E. coli 5′-methylthioadenosine/S-adenosylhomocysteine nucleosidase reveals similarity to the purine nucleoside phosphorylases. Structure (Oxford, U. K.) 9, 941−953. (26) Singh, V., Lee, J. E., Nunez, S., Howell, P. L., and Schramm, V. L. (2005) Transition state structure of 5′-methylthioadenosine/Sadenosylhomocysteine nucleosidase from Escherichia coli and its similarity to transition state analogues. Biochemistry 44, 11647−11659. (27) Singh, V., Luo, M., Brown, R. L., Norris, G. E., and Schramm, V. L. (2007) Transition-state structure of neisseria meningitides 5′methylthioadenosine/S-adenosylhomocysteine nucleosidase. J. Am. Chem. Soc. 129, 13831−13833. (28) Singh, V., and Schramm, V. L. (2007) Transition-state analysis of S. pneumoniae 5′-methylthioadenosine nucleosidase. J. Am. Chem. Soc. 129, 2783−2795. (29) Lee, J. E., Smith, G. D., Horvatin, C., Huang, D. J., Cornell, K. A., Riscoe, M. K., and Howell, P. L. (2005) Structural snapshots of MTA/AdoHcy nucleosidase along the reaction coordinate provide insights into enzyme and nucleoside flexibility during catalysis. J. Mol. Biol. 352, 559−574. (30) Singh, V., Shi, W., Almo, S. C., Evans, G. B., Furneaux, R. H., Tyler, P. C., Painter, G. F., Lenz, D. H., Mee, S., Zheng, R., and Schramm, V. L. (2006) Structure and inhibition of a quorum sensing target from Streptococcus pneumoniae. Biochemistry 45, 12929− 12941. (31) Kang, X. S., Zhao, Y., Jiang, D. H., Li, X. M., Wang, X. P., Wu, Y., Chen, Z. L., and Zhang, X. C. (2014) Crystal structure and biochemical studies of Brucella melitensis 5 ′-methylthioadenosine/Sadenosylhomocysteine nucleosidase. Biochem. Biophys. Res. Commun. 446, 965−970.

I

DOI: 10.1021/acs.biochem.7b00691 Biochemistry XXXX, XXX, XXX−XXX