ARTICLE pubs.acs.org/crystal
Crystallization of Adenylylsulfate Reductase from Desulfovibrio gigas: A Strategy Based on Controlled Protein Oligomerization Published as part of the Crystal Growth & Design virtual special issue on the 13th International Conference on the Crystallization of Biological Macromolecules (ICCBM13). Jou-Yin Fang,§,† Yuan-Lan Chiang,§,† Yin-Cheng Hsieh,§,† Vincent C.-C. Wang,‡ Yen-Chieh Huang,§ Phimonphan Chuankhayan,§ Ming-Chi Yang,§ Ming-Yih Liu,§ Sunney I. Chan,‡,^ and Chun-Jung Chen*,§,||,# §
)
Life Science Group, Scientific Research Division, National Synchrotron Radiation Research Center, Hsinchu 30076, Taiwan Department of Physics and †Institute of Bioinformatics and Structural Biology, National Tsing Hua University, Hsinchu 30013, Taiwan # Institute of Biotechnology, National Cheng Kung University, Tainan 701, Taiwan ‡ Institute of Chemistry, Academia Sinica, Taipei 11529, Taiwan ^ Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, United States ABSTRACT: Adenylylsulfate reductase (adenosine 50 -phosphosulfate reductase, APS reductase or APSR, E.C.1.8.99.2) catalyzes the conversion of APS to sulfite in dissimilatory sulfate reduction. APSR was isolated and purified directly from massive anaerobically grown Desulfovibrio gigas, a strict anaerobe, for structure and function investigation. Oligomerization of APSR to form dimersR2β2, tetramersR4β4, hexamersR6β6, and larger oligomers was observed during purification of the protein. Dynamic light scattering and ultracentrifugation revealed that the addition of adenosine monophosphate (AMP) or adenosine 50 -phosphosulfate (APS) disrupts the oligomerization, indicating that AMP or APS binding to the APSR dissociates the inactive hexamers into functional dimers. Treatment of APSR with β-mercaptoethanol decreased the enzyme size from a hexamer to a dimer, probably by disrupting the disulfide Cys156—Cys162 toward the C-terminus of the β-subunit. Alignment of the APSR sequences from D. gigas and A. fulgidus revealed the largest differences in this region of the β-subunit, with the D. gigas APSR containing 16 additional amino acids with the Cys156—Cys162 disulfide. Studies in a pH gradient showed that the diameter of the APSR decreased progressively with acidic pH. To crystallize the APSR for structure determination, we optimized conditions to generate a homogeneous and stable form of APSR by combining dynamic light scattering, ultracentrifugation, and electron paramagnetic resonance methods to analyze the various oligomeric states of the enzyme in varied environments.
1. INTRODUCTION Sulfate-reducing bacteria (SRB), capable of metabolizing sulfate, are prokaryotes in a special group that are found in various sulfate-rich environments as well as anaerobic environments, such as soil, oil fields, the sea, or the innards of animals or human beings.15 Desulfovibrio gigas (D. gigas), a strict anaerobe, is a representative of SRB well-studied to elucidate metabolic pathways under many diverse conditions.6,7 Sulfate reduction is either assimilatory or dissimilatory. The former reduction occurs in archaebacteria, bacteria, fungi, and plants via various pathways.8 In the assimilatory reduction of Escherichia coli, sulfate is initially transformed by ATP sulfurylase (ATPS) to adenosine 50 -phosphosulfate (APS), which is then phosphorylated by APS kinase to 30 -phosphate adenosine 50 -phosphosulfate (PAPS). PAPS is in turn reduced to sulfite by PAPS reductase, and eventually the sulfite is reduced by sulfite reductase (SiR) to sulfide. In dissimilatory sulfate reduction of SRB, sulfate is first catalyzed by ATPS to APS, which is then directly reduced by APS reductase r 2011 American Chemical Society
to sulfite. Sulfite is subsequently reduced by dissimilatory sulfite reductase (Dsr) to three observed products: trithionite (S3O62), thiosulfate (S2O32), or sulfide (S2). Adenylylsulfate reductase, also called adenosine 50 -phosphosulfate reductase, APS reductase or APSR, plays an important role in catalyzing APS to adenosine monophosphate (AMP) and sulfite in the dissimilatory sulfate reduction. The catalytic mechanism of APSR is divided into stages consisting of the transport of electrons and the cleavage of APS by FAD. Electron input to FAD catalyzes the cleavage of APS to release AMP and sulfite. Multiple forms of APSR in D. vulgaris have been observed in buffers under varied conditions.9 APS reductase from D. gigas, first purified by Lampreia,10 consists of several R- and β-subunits, molecular masses 70 and 23 kDa, respectively, with a total molecular Received: October 15, 2010 Revised: April 1, 2011 Published: April 05, 2011 2127
dx.doi.org/10.1021/cg1013818 | Cryst. Growth Des. 2011, 11, 2127–2134
Crystal Growth & Design mass of 400 kDa. According to analysis by electron paramagnetic resonance (EPR) and M€ossbauer spectroscopy, APSR contains one flavin adenine dinucleotide (FAD) and two [4Fe-4S] clusters. The enzyme from D. gigas is an R2β complex with one FAD and two [4Fe-4S] clusters.11 However, APSR from D. vulgaris is an R2β2 complex with a total molecular mass of 186 kDa and only one [4Fe-4S] cluster in each Rβheterodimer.12 The quarternary structures of APSR and their constitution of cofactors in terms of FAD and ironsulfur clusters in the subunits are still under investigation. The crystal structure of APSR from Archaeoglobus fulgidus (A. fulgidus) indicates that the functional unit is a 1:1 Rβheterodimer, containing two ironsulfur clusters and one FAD.13 However, the asymmetric unit is an R2β2 heterotetramer due to crystal packing. We have determined the crystal structure of APSR from D. gigas;14 in contrast with the R2β2 heterotetramer of A. fulgidus, the overall structure of APSR from D. gigas comprises six Rβ-heterodimers that form a hexameric structure. The FAD is noncovalently attached to the R-subunit, and the two [4Fe-4S] clusters are enveloped by cluster-binding motifs. Alignment of the APSR sequences from D. gigas and A. fulgidus shows the largest differences toward the C-terminus of the β-subunit, in which the D. gigas APSR contains 16 additional amino acids. The C-terminal segment of the β-subunit wraps around the R-subunit to form a functional unit, with the C-terminal loop inserted into the active-site channel of the R-subunit from another Rβheterodimer. The two cysteine residues Cys156 and Cys162 at the C-terminal end form a disulfide linkage. In preparing suitable crystals of APSR from D. gigas for structure investigation, we encountered great difficulty in maintaining APSR in a homogeneous and stable form. Oligomerization of APSR with the dimerR2β2, tetramersR4β4, hexamersR6β6, and larger multimer forms were observed during purification of the protein. The multimer forms of APSR were suggested to be related to the enzymatic function.14 In this study, investigations of the influence of AMP, reducing agent β-mercaptoethanol, and pH gradient on the oligomerization of APSR from D. gigas have been carried out in conjunction with the dynamic light scattering, ultracentrifugation, and EPR methods to seek optimum conditions for crystallization from solutions containing the various multimeric states.
2. EXPERIMENTAL SECTION 2.1. Materials. D. gigas (ATCC 19364) was grown at 37 C for 22 h in a lactate/sulfate medium modified from the previously described procedures.6 APSR protein was purified by chromatography (DE52, Macro-DEAE, HTP and mono Q columns) as described previously.14 The purified APSR protein was dialyzed, desalted, and concentrated before biochemical assays and crystallization. 2.2. Activity Assay of APSR. The activity of APSR was assayed in reverse as the rate of ferricyanide reduction to obtain APS from AMP and sulfite.15 Ferricyanide accepts electrons from the reduced APSR to form ferrocyanide with decreasing absorption at 420 nm (spectrophotometer Beckman Acta V). The reaction mixture containing AMP (3.3 mM), β-mercaptoethanol (0.5 mM) in Tris buffer (100 mM, pH 7.6) served first as the control, and then was incubated with the oxidized APSR and ferricyanide (K3Fe(CN)6, 1.3 mM) for ∼100 s at 25 C. After sodium sulfite (Na2SO3, 3 mM) in EDTA (5 mM) was added into the mixture, the absorption at 420 nm decreased. The specific activity (unit) is expressed as μmol of APS formed per min and per mg of APSR.
ARTICLE
2.3. Electron Paramagnetic Resonance (EPR) Spectroscopy. The EPR spectra were recorded on a Bruker E580 spectrometer equipped with a continuous liquid helium cryostat (Oxford Instruments) outfitted with a turbo-pump to decrease the vapor pressure of the liquid helium (at Academia Sinica, Institute of Chemistry). APSR proteins were prepared at concentrations ∼100 μM in Tris buffer (50 mM, pH 7.6). EPR measurements (microwave frequency 9.635 GHz, microwave power 2 mW, modulation frequency 100 kHz, modulation amplitude 4 G, and time constant 327 ms) were performed at 5 K. 2.4. Analytical Ultracentrifugation. The sedimentation velocity experiments were performed on an analytical ultracentrifuge (Beckman Optima XL-A, equipped with UVVIS optical detection, An60Ti rotor and standard double sector cells). The molecular masses of the APSR multimers were measured at 20 C with the protein concentration at 5 μM in the absence or presence of AMP, APS, and Na2SO3. Data analysis and the distribution of sedimentation coefficients were determined with the SEDFIT85 program.16 2.5. Dynamic Light Scattering (DLS). DLS analysis (Malvern Instruments, Zetasizer Nano-S) was performed at 25 C. APSR (1 mg/mL) was analyzed in Tris buffer (500 mL, 20 mM, pH 7.5). To examine the effects of AMP, APS, and SO32 on the polymerization of APSR, AMP, APS or Na2SO3 (3.5 mM) was added to the APSR protein solution in separate experiments. All sample solutions were centrifuged and filtered through a membrane (porosity 0.2 mm) to remove dust before addition to the sampling cuvette (DTS0012disposable). All the measurements were made in triplicate. 2.6. Crystallization. The protein was crystallized with the hangingdrop vapor-diffusion method at 18 C using drops (2 μL) of the purified protein (∼4.6 mg/mL) in Tris buffer (0.1 M, pH 7.4) mixed with a reservoir solution (equal volume) containing PEG 6000 (20% mass/ volume) and ammonium sulfate (60 mM) in Tris buffer (0.1 M, pH 7.0), and equilibrated against the reservoir solution (500 μL) in an ADX plate (24 wells, Hampton Research). Yellowish crystals of APSR appeared after five days and continued to grow to a final size of 0.12 0.12 0.2 mm3 after three weeks. Crystals of quality satisfactory for diffraction were used to collect data. 2.7. X-ray Diffraction and Data Collection. A single yellowish crystal of APSR was mounted (Cryoloop, 0.10.2 mm), dipped briefly in glycerol (20%) as cryoprotectant, and frozen in liquid nitrogen. X-ray diffraction data were collected at 110 K under a nitrogen stream provided by a cryo-system (X-Stream, Rigaku/MSC, Inc.) using synchrotron radiation on beamlines BL12B2 at SPring-8 (Taiwan contract, Harima, Japan) and BL13B1 at NSRRC (Hsinchu, Taiwan). The diffraction data were obtained by a total rotation of 150 with 0.5 oscillation and exposure durations of 2 s per frame using X-ray wavelength 1.0 Å at a distance of 460 mm from the crystal to the detector at 100 K in a nitrogen stream provided by the cryo-system. The data were indexed and processed with program HKL2000.17
3. RESULTS AND DISCUSSION 3.1. Growth of Bacteria. According to Peck (1982),18 phos-
phate plays an important role in the growth of D. gigas. In this work, we used a phosphate-free medium also to compare the rate of bacterial growth. The visible absorption at 600 nm was used to determine the growth density of the bacterial cells. The results showed that a phosphate-free medium initially gave a smaller rate of growth of bacteria than the original condition with phosphate, but the bacteria in both media eventually grew to the same density (1 g of bacteria L1 medium), implying that phosphate might affect bacterial growth at only the early stage. 3.2. Purification and Characterization of APSR. In each purification step, the purity of the eluted proteins was examined 2128
dx.doi.org/10.1021/cg1013818 |Cryst. Growth Des. 2011, 11, 2127–2134
Crystal Growth & Design
ARTICLE
Table 1. Activity Assays of APSR during the Process of Purification activity (μmol of
specific activity (μmol of
fractions after column
APS/min/mL)
APS/min/mg)
crude extract DE-52
1 1.98
0.048 1.64
Macro-DEAE
2.01
3.23
HTP
2.28
2.53
Mono Q
2.28
4.85
Figure 2. X-band EPR spectra of APSR. Both the P1 (in red) and P2 (in gray) fractions exhibit the same signals near g ∼2, corresponding to [4Fe-4S] clusters.
Figure 1. Purification and separated fractions of APSR. (a) Analysis (12% SDSPAGE) of various collections eluded from the DE-52 column (lanes 1 and 2); Macro-DEAE (lane 3); HTP (lane 4); the last Micro-DEAE (lane 5). The molecular weight markers are shown in lane M. (b) Two fractions from Mono Q ion-exchange column. P1 and P2 represent the major and minor fractions, respectively.
with an activity assay and UV absorption (A278/A392). The ratio of the UV absorption of the proteins from the final step was about
4.87 units, comparable with earlier work.10 Measurement of the activity during protein purification showed a decrease after the third column (Table 1). For the HTP column, phosphate buffer was used to elute the protein, which might be the reason for the decreased protein activity because phosphate has a chemical structure similar to that of sulfite, the substrate used in the APSR activity assay, and might behave as an inhibitor. The SDSPAGE showed two subunits of molecular masses 70 kDa and 23 kDa, respectively (Figure 1A). The final ion-exchange column (Mono Q) further separated the protein into two peaks with one major (P1) and one minor fraction (P2) that were eluted with concentrations of Tris buffer from 200 and 250 mM (Figure 1B). Both fractions exhibited different protein activities but showed the same two subunits on SDSPAGE. EPR experiments at 5 K showed that both the P1 and P2 fractions exhibited the same signals at about g = 2.01, corresponding to [4Fe-4S] clusters at the same oxidation state (Figure 2).10,19 Oligomerization of APSR from D. vulgaris had been reported during purification of the protein under low salt conditions.12 In our ion-exchange column chromatography, the gradient concentration of Tris-HCl buffer might affect the composition of APSR from D. gigas. The molecular masses of the APSR P1 and P2 fractions were determined by analytical ultracentrifugation. The data showed that the P1 fraction contained proteins of molecular mass 586, 708, and 219 kDa for the major and the two minor components, respectively (Figure 3A), whereas the P2 fraction contained one major and two minor components with molecular masses of 742, 854, and 223 kDa, respectively (Figure 3B). Thus, the purified APSR exists in multiple forms under varied ion concentrations. 3.3. Effects of AMP, APS, and SO32 on APSR Polymerization. The polymerization of APSR from D. vulgaris had been examined in the presence of various substrates.11,15 Adding AMP (10 mM) decreased the apparent molecular mass of the APSR to 220 kDa. The structure of the APSR soaked with substrate AMP also revealed the binding residues that held the AMP in the active-site channel.20 2129
dx.doi.org/10.1021/cg1013818 |Cryst. Growth Des. 2011, 11, 2127–2134
Crystal Growth & Design
ARTICLE
Figure 3. Molecular masses measured by ultracentrifugation. (a) The P1 fraction contained proteins of molecular masses 586 (major), 708 (minor), and 219 kDa (minor) for the major and minor components, respectively. (b) The P2 fraction contained major and minor components with molecular masses of 742 (major), 854 (minor) and 223 kDa (minor), respectively. (c) The AMP-treated P1 fraction gave a major component with molecular mass of 235 kDa and a minor component of molecular mass of 468 kDa. (d) The AMP-treated P2 fraction gave a major component with molecular mass of 668 kDa and two minor components with molecular masses of 219 and 843 kDa.
In this study, we have examined the polymerization of APSR from D. gigas by analytical ultracentrifugation and DLS. After the protein was treated with excess AMP, the P1 fraction gave a major component with molecular mass of 235 kDa (dimer form) and a minor component with molecular mass of 468 kDa (tetramer form) (Figure 3C), a decrease from the larger molecular masses of 586 and 708 kDa for the AMP-free enzymes noted earlier. Thus, the major component with molecular mass of 586 kDa was transformed into a lower molecular-mass protein species of 235 kDa; and the minor component at 708 kDa was similarly converted to the 235 kDa protein species. The molecular masses of the major and minor components of the P2 fraction changed from 742 and 854 kDa to 219 (minor), 668 (major), and 843 kDa (minor) (Figure 3D). Interestingly, while the major component in this fraction was partially converted to a protein species with the lower molecular mass of 219 kDa, the minor fraction with the higher molecular mass of 843 kDa did not
seem to change its state of oligomerization. Since the majority of the APSR oligomers in the P1 fraction (but apparently not in the P2 fraction) were dissociated by AMP, we surmise that substrate binding is triggering some conformational alteration in the structures of the subunits to affect their propensity toward protein oligomerization. The same experiments were repeated using APS and Na2SO3. The addition of APS was found to induce a similar transformation of APSR from hexamers to dimers. However, SO32 did not produce noticeable effects. The effect of AMP on the APSR polymerization was investigated also by DLS to measure changes in the particle sizes of the APSR aggregates. We focused on the P1 fraction of the APSR solution as this was the major fraction with the higher activity. After AMP (3.5 mM) was added to the APSR solution, the average diameter (2 Stokes radius) of the protein in the P1 fraction decreased from 12.81 to 8.11 nm (Figure 4). The same experiment was repeated using APS and Na2SO3. The results 2130
dx.doi.org/10.1021/cg1013818 |Cryst. Growth Des. 2011, 11, 2127–2134
Crystal Growth & Design
ARTICLE
Figure 4. Oligomerization of APSR P1 fraction observed by DLS analysis. The diameter of APSR in the P1 fraction (1.0 mg/mL) decreased from 12.8 nm (black line) to 8.1 nm (gray line) and 8.3 nm (dashed line) upon the addition of AMP (3.5 mM) and β-mercaptoethanol (186 mM), respectively.
showed that the addition of APS led to a similar size decrease of the protein particles, whereas SO32 did not change the size notably. To delineate additional factors that might influence the APSR polymerization, we also examined the effects of temperature (4, 25, and 37 C) and the concentration (100, 200, and 400 mM) of salts (NaCl, MgCl2, and CaCl2). These experiments showed no notable change in the size distribution of the protein under these conditions. The DLS experiments reinforced our interpretation of the ultracentrifugation data that the P1 fraction of the APSR solution changed from hexamers to dimers in the presence of AMP and APS. From the crystal structures of APSR from D. gigas, the sizes of the APSR hexamer and dimer were measured to be 135 120 100 Å3 and 100 70 65 Å3 based on the molecular structures (Figure 5), and the Stokes radii of the hexamers and dimers were predicted by the program HYDROPRO21 to be ca. 6.61 and 4.23 nm, respectively, in agreement with the DLS measurements. On the other hand, the APSR remained aggregated in the P2 fraction as larger oligomers, which might explain the lower activity of the APSR in this fraction compared with the P1 fraction. 3.4. pH Effects on APSR Oligomerization. The oligomerization of APSR from D. vulgaris in the presence of various substrates and buffers had also been reported.9,12 In a phosphate buffer, Tris-HCl or other buffers at low concentrations, a large aggregate of molecular mass about 440 kDa was observed for the D. vulgaris enzyme. However, changing the buffer to Tris-maleate decreased the apparent molecular mass of the APSR to 220 kDa. Because the pH of a buffer system is a major factor affecting the stability of a homogeneous protein solution for crystallization, we have also investigated the effect of pH on the APSR polymerization. The sizes of the proteins in the P1 fraction were measured with DLS at varied pH and in varied buffers, including Tris, phosphate, HEPES, sodium citrate, and sodium acetate with the solution pH ranging from 8 to 4. The results of the pH studies showed that the diameters of APSR gradually decreased from 12.8 nm in the pH range of 7.08.0 to 9.8 nm at pH 4.0 (Figure 6). The size distribution decreased progressively as the pH is lowered, indicating that APSR existed in various oligomeric forms in an acidic environment. Thus, it appears that a solution pH in the range of 78 might be optimal to stabilize the APSR as a homogeneous hexamer. Furthermore, when we investigated the effects of different buffers at pH 7.5 with 20 mM phosphate and Tris, there was no change in the oligomerization. A similar diameter of ∼12.8 nm was obtained for the APSR in the P1 fraction between these two conditions.
Figure 5. Structure of APSR from D. gigas. (a) Structure of the R2β2heterotetramer with dimensions 100 70 65 Å3. (b) Structure of the APSR hexamer with three R2β2-heterotetramers with dimensions 135 120 100 Å3. (c) Interactions between two Rβ-heterodimers. The C-terminal segment of the β-subunit wraps around the R-subunit to form a functional Rβ-heterodimer unit and introduces an anchor-like hook to plug into the active site of another Rβ-heterodimer.
3.5. Crystallization of APSR and X-ray Data Collection. Because of the unstable character of the protein solutions, fresh APSR protein was prepared right before the crystallization trial. Early observations indicated that APSR exhibited multimeric forms: dimers, tetramers, hexamers, and larger oligomers. Since a homogeneous protein sample is very critical for successful crystallization, and our study of pH effects on the APSR polymerization 2131
dx.doi.org/10.1021/cg1013818 |Cryst. Growth Des. 2011, 11, 2127–2134
Crystal Growth & Design
ARTICLE
Figure 6. pH effects on the oligomerization of APSR. The sizes of the protein species in the P1 fraction were measured with DLS at varied pH and in varied buffers, including Tris, phosphate, HEPES, sodium citrate, and sodium acetate with the solution pH ranging from 8.0 to 4.0. The diameters of the APSR proteins gradually decreased from 12.8 nm at pH 7.08.0 to 9.8 nm at pH 4.0.
suggested that pH 78 was the optimal pH range for stabilizing homogeneous APSR as the hexameric form, the P1 fraction protein was prepared in Tris buffer (0.1 M, pH 7.5) for subsequent crystallization experiments. The initial crystallization conditions were screened with the screen kits (Hampton Research) using the screening robot Hydra-II (MATRIX). Crystals appeared with the screen kit, Index #90, #92, and #94 from the first screening. The qualities of crystals obtained under these three conditions, even with further modifications, were still poor in terms of the size, shape, and X-ray diffraction (∼10 Å resolution) (Figure 7A). We then further screened with the precipitant PEG6000, which gave crystals of a different shape. This condition was further modified, and the resolution of the X-ray diffraction on the crystals was improved from 8 to 6 Å (Figure 7B). In order to obtain higher quality crystals for structure determination, we tested the effects of various additives with the Additive Screen (Hampton Research). Ammonium sulfate proved to be the best additive to improve the crystals, with better shapes and dimensions. The final crystallization condition was as follows: protein (∼4.6 mg/mL) in Tris buffer (0.1 M, pH 7.5); the precipitant PEG 6000 (20%); and ammonium sulfate (60 mM) in Tris buffer (0.1 M, pH 7.5). The yellowish crystals of APSR appeared after five days and continued to grow to a final size 0.12 0.12 0.2 mm3 after three weeks (Figure 7C). Crystals of quality satisfactory for diffraction were used for the data collection. Crystals of satisfactory quality were identified through a careful screening and selection for data collection, as they typically exhibited a fairly high mosaicity (>1.0). The best crystals diffracted to 3.1 Å resolution and the data were collected for structure determination. Overlapping diffraction occurred in the data processing because of the long c axis of the unit cell. Radiation damage was observed after the protracted exposure during data collection (30 s per frame with 0.5 oscillation), which caused a decrease of I/σI and increase of Rsym. Although data for a total rotation of 270 were collected, after an inspection of the data statistics with regard to the decay of the crystals, we selected only the first 150 range for data processing. Analysis of the diffraction pattern indicated that these crystals exhibited trigonal symmetry; systematic absences indicated the space group to be P3121, with unit cell dimensions a = b = 199.63 Å, c = 317.42 Å. Assuming the presence of six APSR molecules per
Figure 7. Crystallization of APSR. (a) The crystals after initial screening. (b) The crystals grew in PEG6000 as the precipitant. (c) The best crystals grown in PEG6000 as the precipitant and ammonium sulfate as the additive for the data collection.
asymmetric unit, the Matthew’s coefficient was estimated to be 3.27 Å3 Da1, corresponding to a solvent content of 62.43% (Matthews, 1968), which is within the general range for protein crystals. 3.6. Structure Features of the APSR and the Nature of the Oligomerization. The structure of APSR from D. gigas was solved by the molecular replacement method using the monomer structure of APSR of A. fulgidus (PDB code: 1JNR) as a search model with the sequence similarities of 64% for the R-subunit and 82% for the β-subunit.7 Structure refinement of APSR yielded an R-factor 19.3% and Rfree 24.6%, respectively, to 3.1 Å resolution.14 Different from the R2β2 heterotetramer of the A. fulgidus, the overall structure of APSR from D. gigas 2132
dx.doi.org/10.1021/cg1013818 |Cryst. Growth Des. 2011, 11, 2127–2134
Crystal Growth & Design
ARTICLE
3.7. The Effect of β-Mercaptoethanol. Our crystallographic analysis indicates that there is a disulfide bond between Cys156 and Cys162 in the β-subunit C-terminus, which stabilizes the structure of the C-terminal loop that is responsible for the oligomerization of the enzyme. To confirm this, we have undertaken a study of the effects of β-mercaptoethanol on the APSR polymerization. DLS experiments revealed that the average diameter of APSR in the P1 fraction decreased from 12.8 nm (hexamer) to 8.3 nm (dimer) in the presence of 186 mM β-mercaptoethanol (Figure 8A). The change in the average size of the APSR protein species as the concentration of the β-mercaptoethanol is gradually increased from none to saturation in the protein solution is shown in Figure 8B. 3.8. Attempts To Crystallize the APSR Dimer. We have also attempted to prepare the dimer form of the APSR for crystallization by going to acidic pH and the addition of AMP or β-mercaptoethanol. So far, only small and poor diffracting crystals were obtained by these approaches. These dimers are homogeneous and stable according to analysis of the size and size distribution as well as solubility. However, it appears that the flexible C-terminus of the β-subunit might affect the crystallization and decrease the crystal quality.
Figure 8. β-mercaptoethanol effects on the oligomerization of APSR. (a) The diameters of the APSR proteins decreased from 12.8 to 8.3 nm as the concentration of β-mercaptoethanol (β-me) was gradually increased from none (red) to saturation (186 mM β-mercaptoethanol) (blue) in the protein solution. (b) The size of the APSR became stable and remained unchanged after the β-mercaptoethanol concentration reached 0.2 M.
comprises six Rβ-heterodimers that form a hexameric structure. In the Rβ-heterodimers, the C-terminal segment of the β-subunit wraps around the R-subunit, with the globular domain of the β-subunit embedded in a shallow hollow of the R-subunit to form a functional unit. The Rβ-heterodimer performs a rotation about a pseudo-2-fold axis with another Rβ-heterodimer to form a tightly contacted R2β2-heterotetramer (Figure 5A). The β-subunit C-terminii of the R2β2-heterotetramer introduces anchor-like hooks on the R-subunits of another two R2β2heterotetramers. Three R2β2-heterotetramers connect to each other through the C-terminii of the β-subunits to form a hexamer structure containing six Rβ-heterodimers (Figure 5B). Each C-terminus of the two β-subunits of the R2β2-heterotetramer introduces an anchor-like hook to interact with the R-subunits of a contiguous R2β2-heterotetramer. A loop at the C-terminus contains the Cys156Cys162 disulfide which is inserted into the active-site channel of the R-subunit from another Rβ-heterodimer (Figure 5C). These structural features suggest that the APSR in the hexameric form is inactive because all the active sites are blocked by a β-subunit C-terminus from a contagious Rβheterodimer. In an environment where AMP or APS is not concentration limiting, however, these substrates would compete for the active site, and the β-subunit C-terminus blocking the active site entrance would be displaced by an AMP or APS. This scenario predicts that the APSR hexamer will dissociate partially or fully into the active dimer or tetramer form in solution in the presence of AMP and APS, as we have observed in our sedimentation and light scattering studies of the APSR solutions.
4. CONCLUSION APSR exhibits several multimeric forms including dimers R2β2, tetramersR4β4, hexamersR6β6, and larger oligomers in the solution, a complication that renders it difficult to relate its biological function to the three-dimensional structure. To obtain high-quality crystals of APSR for successful crystallographic analysis, we have optimized the conditions to stabilize the APSR as a hexamer to obtain a homogeneous protein species. This strategy of optimization of the solution conditions for crystallization of the APSR as well as the methods we have developed here to characterize the colligative properties of the protein solution may be useful for structural studies of other proteins with similar aggregation problems. For example, Bacillus cereus NCTU2 chitinase exists as a monomer under neutral pH but exhibits the dimer or tetramer form in more acidic media. As in APSR, formation of the multimeric form(s) covers up the active site and obstructs access of the substrates to the active site of the enzyme. The complex structures with substrates were successfully determined only after the protein monomer was obtained.22 Another system is the nucleoside diphosphate kinase, which also exhibits an oligomerization equilibrium (dimer, tetramer, and hexamer) upon binding nucleotide substrates.2325 Association of dimers to form the hexamer is induced by the presence of a nucleotide substrate ADP or ATP.22 Other systems such as the rice Bowman-Birk inhibitor,26 Pseudomonas aeruginosa histidine phosphotransfer protein B (unpublished result from this laboratory), and Dioscorea alta dioscorin (Ong, P.-L., unpublished) also undergo oligomerization in buffer solution, and the crystallographic analysis of these proteins may benefit from the strategy described here. ’ AUTHOR INFORMATION Corresponding Author
*Address: Life Science Group, Scientific Research Division, National Synchrotron Radiation Research Center, 101 Hsin-Ann Road, Hsinchu, 30076, Taiwan; e-mail:
[email protected]; tel: 886-3-578-0281 ext. 7330; fax: 886-3-578-3813. 2133
dx.doi.org/10.1021/cg1013818 |Cryst. Growth Des. 2011, 11, 2127–2134
Crystal Growth & Design
’ ACKNOWLEDGMENT We are indebted to Yuch-Cheng Jean and the supporting staff at beamlines BL13B1 and BL13C1 at the National Synchrotron Radiation Research Center (NSRRC) and Masato Yoshimura, Jeyakanthan Jeyaraman, and Hirofumi Ishii at the Taiwan contracted beamline BL12B2 at SPring-8 for technical assistance. Portions of this research were carried out at the NSRRC-NCKU Protein Crystallography Laboratory at National Cheng Kung University (NCKU). We thank Prof. Rong-Long Pan and Prof. Ping-Ling Ong for discussions and suggestions. This work was supported in part by grants from the National Science Council [NSC 95-2313-B-213-001, 95-2313-B-009-001-MY, and 98-2311-B-213-MY3] and the National Synchrotron Radiation Center [NSRRC 983RSB02 and 993RSB02] to C.-J.C. ’ REFERENCES (1) Gibson, G. R.; Cummings, J. H.; Macfarlane, G. T. Growth and activities of sulfate-reducing bacteria in gut contents of healthy-subjects and patients with ulcerative-colitis. FEMS Microbiol. Ecol. 1991, 86, 103–111. (2) Pochart, P.; Dore, J.; Lemann, F.; Goderel, I.; Rambaud, J. C. Interrelations between populations of methanogenic Archaea and sulfate-reducing bacteria in the human colon. FEMS Microbiol. Lett. 1992, 98, 225–228. (3) Hamilton, W. A. Bioenergetics of sulphate-reducing bacteria in relation to their environmental impact. Biodegradation 1998, 9, 201–212. (4) Matias, P. M.; Pereira, I. A.; Soares, C. M.; Carrondo, M. A. Sulphate respiration from hydrogen in Desulfovibrio bacteria: a structural biology overview. Prog. Biophys. Mol. Biol. 2005, 89, 292–329. (5) Kuang, F.; Wang, J.; Yan, L.; Zhang, D. Effects of sulfate-reducing bacteria on the corrosion behavior of carbon steel. Electrochim. Acta 2007, 52, 6084–6088. (6) Le Gall, J. A new species of Desulfovibrio. J. Bacteriol. 1963, 86, 1120. (7) Traore, A. S.; Hatchikian, C. E.; LeGall, J.; Belaich, J. P. Microcalorimetric studies of the growth of sulfate-reducing bacteria: comparison of the growth parameters of some Desulfovibrio species. J. Bacteriol. 1982, 149, 606–611. (8) Kopriva, S.; Koprivova, A. Plant adenosine 5 0 -phosphosulphate reductase: the past, the present, and the future. J. Exp. Bot. 2004, 55, 1775–1783. (9) Bramlett, R. N.; Peck, H. D., Jr. Some physical and kinetic properties of adenylylsulfate reductase from Desulfovibrio vulgaris. J. Biol. Chem. 1975, 250, 2979–2986. (10) Lampreia, J.; Moura, I.; Teixeira, M.; Peck, H. D.; LeGall, J.; Huynh, B. H.; Moura, J. J. G. The active-centers of adenylylsulfate reductase from Desulfovibrio gigas - characterization and spectroscopic studies. Eur. J. Biochem. 1990, 188, 653–664. (11) Lampreia, J.; Pereira, A. S.; Moura, J. J. G. Adenylylsulfate reductases from sulfate-reducing bacteria. Inorg. Microb. Sulfur Metab. 1994, 241–260. (12) Verhagen, M.; Kooter, I. M.; Wolbert, R. B. G.; Hagen, W. R. On the iron-sulfur cluster of adenosine phosphosulfate reductase from Desulfovibrio vulgaris (Hildenborough). Eur. J. Biochem. 1994, 221, 831–837. (13) Fritz, G.; Roth, A.; Schiffer, A.; Buchert, T.; Bourenkov, G.; Bartunik, H. D.; Huber, H.; Stetter, K. O.; Kroneck, P. M. H.; Ermler, U. Structure of adenylylsulfate reductase from the hyperthermophilic Archaeoglobus fulgidus at 1.6-Å resolution. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 1836–1841. (14) Chiang, Y.-L.; Hsieh, Y.-C.; Fang, J.-Y.; Liu, E.-H.; Huang, Y.-C.; Chuankhayan, P.; Jeyaraman, J.; Liu, M.-Y.; Chan, S. I.; Chen, C.-J. Crystal structure of adenylylsulfate reductase from Desulfovibrio gigas suggests a potential self-regulation mechanism involving the C-terminus of the β-subunit. J. Bacteriol. 2009, 191, 7597–7608.
ARTICLE
(15) Peck, H. D., Jr.; Deacon, T. E.; Davidson, J. T. Studies on adenosine 50 -phosphosulfate reductase from Desulfovibrio desulfuricans and Thiobacillus thioparus. I. the assay and purification. Biochim. Biophys. Acta 1965, 96, 429–446. (16) Lebowitz, J.; Lewis, M. S.; Schuck, P. Modern analytical ultracentrifugation in protein science: a tutorial review. Protein Sci. 2002, 11, 2067–2079. (17) Otwinowski, Z.; Minor, W. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 1997, 276, 307–326. (18) Peck, H. D., Jr. Biochemistry of dissimilatory sulphate reduction. Phil. Trans. R. Soc. Lond. 1982, B298, 443–466. (19) Kopriva, S.; B€uchert, T.; Fritz, G.; Suter, M.; Weber, M.; Benda, R.; Schaller, J.; Feller, U; Sch€urmann, P.; Sch€unemann, V.; Trautwein, A. X.; Kroneck, P. M. H.; Brunold, C. Plant adenosine 5-phosphosulfate reductase is a novel iron-sulfur protein. J. Biol. Chem. 2001, 276, 42881–42886. (20) Schiffer, A.; Fritz, G.; Kroneck, P. M. H.; Ermler, U. Reaction mechanism of the iron-sulfur flavoenzyme adenosine-50 -phosphosulfate reductase based on the structural characterization of different enzymatic states. Biochemistry 2006, 45, 2960–2967. (21) Garcia de la Torre, J.; Huertas, M. L.; Carrasco, B. Calculation of hydrodynamic properties of globular proteins from their atomic-level structure. Biophys. J. 2000, 78, 719–730. (22) Hsieh, Y.-C.; Wu, Y.-J.; Chiag, T.-Y.; Kuo, C.-Y.; Shrestha, K. L.; Chao, C.-F.; Huang, Y.-C.; Chuankhayan, P.; Wu, W.; Li, Y.-K.; Chen, C.-J. Crystal structures of Bacillus cereus NCTU2 Chitinase complexes with chitooligomers reveal novel substrate binding for catalysis: a Chitinase without chitin-binding and insertion domains. J. Biol. Chem. 2010, 285, 31603–31615. (23) Chen, C.-J.; Liu, M.-Y.; Chang, C.; Payne, W. J.; Wang, B.-C.; LeGall, J. Crystal structure of nucleoside diphosphate kinase from Bacillus halodenitrificans at 2.2 Å resolution, an enzyme co-expressed of its activity with a Mn-superoxide dismutase. J. Struct. Biol. 2003, 142, 247–255. (24) Huang, J.-Y.; Chang, T; Chang, T.; Chen, C.-J. Crystal structure of a nucleoside diphosphate kinase required for coleoptile elongation in rice (Oryza sativa L.). J. Strut. Biol. 2005, 150, 309–318. (25) Mesnildrey, S.; Agou, F.; Karlsson, A.; Bonne, D. D.; Veron, M. Coupling between catalysis and oligomeric structure in nucleoside diphosphate kinase. J. Biol. Chem. 1998, 273, 4436–4442. (26) Lin, Y.-H.; Lee, H.-T.; Huang, Y.-C.; Liu, M.-Y.; Chang, T.; Wang, A. H.; Chen, C.-J. Purification, crystallization and preliminary X-ray crystallographic analysis of rice Bowman-Birk inhibitor from Oryza sativa. Acta Crystallogr. 2006, F62, 522–524.
2134
dx.doi.org/10.1021/cg1013818 |Cryst. Growth Des. 2011, 11, 2127–2134
