HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 41, 32147-32156, October 13, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/41/32147    most recent M003725200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Komoto, J. Articles by Takusagawa, F. Search for Related Content PubMed PubMed Citation Articles by Komoto, J. Articles by Takusagawa, F. Social Bookmarking                      What's this?

Effects of Site-directed Mutagenesis on Structure and Function of Recombinant Rat Liver S-Adenosylhomocysteine Hydrolase

CRYSTAL STRUCTURE OF D244E MUTANT ENZYME^*

Junichi Komoto, Yafei Huang, Tomoharu Gomi^§, Hirofumi Ogawa^§, Yoshimi Takata^§, Motoji Fujioka^§, and Fusao Takusagawa^¶

From the  Department of Molecular Biosciences, University of Kansas, Lawrence, Kansas 66045-2106 and the ^§ Department of Biochemistry, Toyama Medical and Pharmaceutical University, Faculty of Medicine, Sugitani, Toyama 930-0194, Japan

Received for publication, May 2, 2000, and in revised form, June 23, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

A site-directed mutagenesis, D244E, of S-adenosylhomocysteine hydrolase (AdoHcyase) changes drastically the nature of the protein, especially the NAD⁺ binding affinity. The mutant enzyme contained NADH rather than NAD⁺ (Gomi, T., Takata, Y., Date, T., Fujioka, M., Aksamit, R. R., Backlund, P. S., and Cantoni, G. L. (1990) J. Biol. Chem. 265, 16102-16107). In contrast to the site-directed mutagenesis study, the crystal structures of human and rat AdoHcyase recently determined have shown that the carboxyl group of Asp-244 points in a direction opposite to the bound NAD molecule and does not participate in any hydrogen bonds with the NAD molecule. To explain the discrepancy between the mutagenesis study and the x-ray studies, we have determined the crystal structure of the recombinant rat-liver D244E mutant enzyme to 2.8-Å resolution. The D244E mutation changes the enzyme structure from the open to the closed conformation by means of a ~17° rotation of the individual catalytic domains around the molecular hinge sections. The D244E mutation shifts the catalytic reaction from a reversible to an irreversible fashion. The large affinity difference between NAD⁺ and NADH is mainly due to the enzyme conformation, but not to the binding-site geometry; an NAD⁺ in the open conformation is readily released from the enzyme, whereas an NADH in the closed conformation is trapped and cannot leave the enzyme. A catalytic mechanism of AdoHcyase has been proposed on the basis of the crystal structures of the wild-type and D244E enzymes.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

S-Adenosylhomocysteine hydrolase (AdoHcyase,¹ EC 3.3.1.1) catalyzes the hydrolysis of S-adenosylhomocysteine (AdoHcy) to form adenosine (Ado) and homocysteine (Hcy). The reaction is reversible, and the equilibrium lies far in the direction of AdoHcy synthesis. Under physiological conditions, however, the removal of both Ado and Hcy is sufficiently rapid, so that the net reaction proceeds in the direction of hydrolysis (1). Ado is removed by Ado deaminase and Ado kinase, and Hcy is used for the synthesis of cysteine and the regeneration of methionine. In mammals, Hcy is produced solely from AdoHcy, and it was reported recently that an elevated plasma Hcy level is one of the risk factors for coronary heart disease (2). Rat liver AdoHcyase is a tetramer consisting of chemically identical and functionally equivalent subunits (3, 4). Each subunit consisting of 431 amino acid residues contains 1 mol of tightly bound NAD⁺ (5).

The mechanism of reversible hydrolysis of AdoHcy catalyzed by AdoHcyase has been studied by Palmer and Abeles (6, 7), who showed that AdoHcy and Ado are first oxidized to 3'-keto derivatives by the enzyme-bound NAD⁺. This facilitates abstraction of the 4'-proton by an enzyme base, and the resulting carbanion eliminates the 5'-substituent, Hcy or water, to yield 3'-keto-4',5'-dehydroadenosine. Michael-type addition of water or Hcy to this central intermediate, and reduction of 3'-keto by the NADH formed results in the formation of the final product, Ado or AdoHcy.

From an analysis of amino acid sequence of rat liver AdoHcyase, Ogawa et al. (5) found that the enzyme has a "fingerprint" sequence of the ADP-binding domains of NAD⁺- and FAD-binding proteins at positions 213-244 containing the sequence GXGXXG at 219-224. To examine whether this region is indeed part of the NAD⁺-binding domain of AdoHcyase, Gomi et al. (8) have converted each of the glycines at positions 219, 221, and 224 of rat liver AdoHcyase to a valine by site-directed mutagenesis. The mutant enzymes obtained do not bind NAD⁺, are catalytically inactive, and are found to exist as monomers. These results suggest that the region of residues 213-244 is the NAD⁺-binding site of rat liver AdoHcyase. To further probe the NAD⁺-binding site of AdoHcyase, Gomi et al. (9) made the D244E mutant enzyme by site-directed mutagenesis. The mutant enzyme contained only 0.05 mol of NAD⁺ but contained ~0.6 mol each of NADH and adenine/mol of subunit. In the presence of a saturating concentration of NAD⁺, the mutant enzyme showed apparent K_m values for substrates similar to those of the WT enzyme. Moderate decreases of 8- and 15-fold were observed in V_max values for the synthetic and hydrolytic directions, respectively. From the appearance of activity as a function of NAD⁺ concentration, the enzyme was shown to bind NAD⁺ with a K_d of 23 µM (83 nM for WT) at 25 °C. These results were considered to suggest the importance of Asp-244 in binding NAD, consistent with the idea that the region of AdoHcyase from residues 213 to 244 is part of the NAD binding site (9).

The crystal structures of human placental AdoHcyase complexed with an inhibitor (10) and rat liver AdoHcyase (11) have been determined by an x-ray diffraction method. On the basis of the structures of Turner et al. (10) and themselves, Hu et al. (11) concluded that the substrate-free AdoHcyase has an open conformation, whereas the AdoHcyase complexed with an inhibitor has a closed conformation. Hu et al. (11) estimated that the open-closed conformational change requires ~18° rotation of the catalytic domain around the border between the catalytic and NAD⁺-binding domains. An NAD molecule lies in the cleft of the NAD-binding domain of each subunit and is connected to the protein by both polar and non-polar interactions. The adenosine moiety of the bound NAD is further covered by the C-terminal domain of the adjacent subunit. Lys-425 and Tyr-429 of the adjacent subunit participate in hydrogen bonding with adenosine ribose and pyrophosphate of NAD, respectively (10, 11). These NAD binding features explain the very tight binding of NAD to the enzyme.

In contrast to the site-directed mutagenesis studies, the crystal structures of AdoHcyase have shown that the carboxyl group of Asp-244 points in a direction opposite to the bound NAD molecule and does not participate in any hydrogen bonds with the NAD molecule, although Asp-244 is positioned near the NAD molecule (10, 11). To explain the discrepancy between the mutagenesis studies and the x-ray studies, we have determined the crystal structure of the D244E mutant enzyme. The D244E mutant structure indicates that the original D244E data open to an alternative interpretation, but it also illuminates several important features of the catalytic mechanism of AdoHcyase.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Purification and Crystallization Procedures-- Escherichia coli JM109 harboring the mutant plasmid encoding the Asp-244 to Glu mutation (D244E) was used to obtain the mutant enzyme (9). The enzyme was purified to homogeneity from E. coli extracts by gel filtration over Sephacryl S-300 and DEAE-cellulose chromatography as described previously (9).

The hanging-drop vapor-diffusion method was employed for crystallization of the enzyme. All crystallization experiments were conducted at 22 °C. Small crystals of the enzyme were grown for 1 week in a solution containing 22% (w/v) PEG 4000, 50 mM Tris/HCl buffer, pH 7.2, and 10% (v/v) isopropanol with a protein concentration of 10 mg/ml. The small crystals were put as seeds in a fresh solution containing 15% (w/v) PEG 4000, 50 mM Tris/HCl buffer, pH 7.2, 2% (v/v) glycerol, 5% (v/v) isopropanol, and 1 mM dithiothreitol with a protein concentration of 10 mg/ml. The plate-shaped crystals suitable for x-ray diffraction (~0.3 mm × 0.2 mm × 0.1 mm) were grown for 3 days. Note that any substrate/cofactors were not added in the crystallization solution.

Determination of Enzyme-bound Compounds-- The crystals of D244E were washed with 30% PEG 4000 solution, and were dissolved in Tris/HCl buffer. The protein was denatured by adding 2 volumes of absolute ethanol. The precipitate was washed once with 70% ethanol, and the combined supernatants were dried in vacuo. The residue was dissolved in 100 mM Tris/HCl, pH 7.2, and analyzed by HPLC on a column of C18 (1.0 × 30 cm), using a gradient between 100 mM Tris/HCl, pH 7.2, and the same solution containing 15% acetonitrile. The flow rate was 2 ml/min. The effluent was monitored by absorbance at 260 nm. Three significant peaks were observed at 9.0, 9.8, and 11.9 min, respectively. The peaks corresponded to NADH, Ade, and Ado, respectively. The peak area indicated the ratio of NADH:Ade:Ado to be about 3:1:2.

Data Measurement-- The crystals in a hanging drop were scooped by a nylon loop and were dipped into a cryoprotectant solution containing 30% ethylene glycol, 50 mM Tris/HCl buffer, pH 7.2, and 15% (w/v) PEG 4000 for 30 s before they were frozen in liquid nitrogen. The frozen crystals were transferred onto a Rigaku RAXIS IIc imaging plate x-ray diffractometer with a rotating anode x-ray generator as an x-ray source (CuK radiation operated at 50 kV and 100 mA). The diffraction data were measured up to 2.8-Å resolution at -180 °C. The data were processed with the program DENZO (12). The data statistics are given in Table I.

Structure Determination-- The crystal structure was determined by a molecular replacement procedure using X-PLOR. The structure of the rat liver WT enzyme (open conformation) and the structure of the human placental enzyme (closed conformation) were used as search models. The closed structure model gave significantly better PC refinement results than the open structure model. At this stage, the open structure model was abandoned.

The crystal structure was refined by a standard refinement procedure in the X-PLOR protocol (13). During the later stages of refinement, difference maps (F_o  F_c maps) showed a large significant residual electron density peak in the region of active site. The shape of the electron density peak suggested that the individual subunit contains an Ado or Ado* rather than adenine itself. Four Ado molecules were placed into the electron density peaks in the four subunits. Refinement of isotropic temperature factors for individual atoms was carried out by the individual B-factor refinement procedure of X-PLOR using bond and angle restraints. The temperature factors of NAD and Ado were significantly higher than those of the atoms surrounding the NAD and Ado. Thus, the temperature factors were re-refined with different occupancy factors (0.75, 0.5, and 0.25) of NAD and Ado. Since the occupancy factors, 0.50, of NAD and Ado gave the lowest R-factor as well as reasonable temperature factors of NAD and Ado, the occupancy factors of NAD and Ado were set to 0.5. Although 3'-ketoadenosine molecules were placed in the active sites instead of Ado molecules and the crystal structure was refined, no significant changes were observed in the refinement parameters. During the final refinement stage, well defined residual electron density peaks in difference maps were assigned to water molecules if peaks were able to bind the protein molecules with hydrogen bonds. The final crystallographic R-factor is 0.197 for all data (no  cut-off) from 8.0- to 2.8-Å resolution. The free R for randomly selected 5% data is 0.248 (14).

Rigid-body Domain Analysis-- There are three known AdoHcyase structures (human placenta, rat liver, and D244E mutant enzyme). In order to compare the structures, the individual structures were divided into three domains. At first, the catalytic and NAD binding are approximately defined as follows: catalytic domain, residues 4-175 and 360-385; NAD-binding domain, residues 200-350. The two domains of WT were individually aligned on those of D244E by a least-squares method. The C-C distances between the two structures were plotted against residue numbers. The two plotted curves of the catalytic domain and NAD-binding domain across each other around residue numbers 183-184, 356-357, and 390-391. From the plot, the catalytic, NAD-binding, and C-terminal domains are defined as follows: catalytic domain, residues 4-183 and 357-390; NAD-binding domain, residues 184-356; C-terminal domain, residues 391-431.

The r.m.s.d. listed in Table II were calculated based on the above domain assignment.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Overall Structure-- The crystallographic refinement parameters (Table I), final (2F_o  F_c) maps and conformational analysis by PROCHECK (15) indicate that the crystal structure of D244E has been successfully determined.

The mutant enzyme consists of four subunits related by a pseudo-222 symmetry (Fig. 1). The four subunits are denoted subunits A, B, C, and D. Although the four subunits are crystallographically independent, the subunits are structurally very similar. The r.m.s.d. of the C among the subunit structures related by a pseudo 222-fold symmetry is less than 0.38 Å. For simplicity, therefore, the following description mainly refers to one subunit (A). When subunit-subunit interactions are referred to, subunits A and B are described.

View larger version (63K): [in this window] [in a new window]   Fig. 1.   Tetrameric structure of AdoHcyase. Subunits A, B, C, and D related by a pseudo 222 symmetry are marked by letters and are yellow, cyan, magenta, and white, respectively. Four tightly bound NAD+ molecules and four Ado* are illustrated as red and green sticks, respectively. The following color codes are used in all figures: catalytic domain (yellow), NAD-binding domain (blue), C-terminal domain (cyan), NAD (red), and Ado* (green).

The subunit is composed of three domains with the peptide chain organized into 17  helices and 15  strands. The three domains are denoted the catalytic domain (residues 1-183 and 357-390; topology: 0-1-1-2-2-3-3-4-4-5-5-6-6 and 14-15), the NAD-binding domain (residues: 184-356; topology: 7-7-8-8-9-9-10-10-11-11-12-12-13-14-13), and the C-terminal domain (residues 391-431; topology: 15-16). The catalytic and NAD-binding domains are each folded into an ellipsoid with a typical / twisted open-sheet structure. The two domains are connected at the ellipsoidal pole sections by a reciprocal penetration of a pair of relatively long -helices. Consequently, the subunit is a "fat U" in shape (Fig. 2). In contrast to the structure of WT enzyme (open conformation), a cleft between the catalytic domain and the NAD-binding domain is closed, indicating that the D244E mutant enzyme adopts a closed conformation. The small C-terminal domain, composed of (random coil)-(two-turn -helix)-(random coil), is separated from the main body of the subunit and penetrates into the adjacent subunit B. The C-terminal domain is tightly connected to the NAD-binding domain of subunit B by both polar and non-polar interactions, and participates in forming the NAD-binding site of subunit B (Fig. 2).

View larger version (55K): [in this window] [in a new window]   Fig. 2.   Ribbon drawing of a single subunit of AdoHcyase showing three domains: catalytic domain, NAD+-binding domain, and C-terminal domain. The C-terminal domain from the subunit B is illustrated with white. The bound NAD molecule and Ado* are illustrated by ball-and-stick modes. A, WT structure (open conformation); B, D244E structure (closed conformation).

In the tetrameric organization, the four NAD-binding domains are located near the center of the tetramer, and are tightly connected to each other by both polar and non-polar interactions. Formation of the tetramer creates a unique channel structure (~10 × 10 × 50 Å) that passes through the center of tetramer. The polar C-terminal ends go into the channel. The catalytic domains are placed far from the center of the tetramer and, therefore, interact little with each other.

Cofactor and Substrate-- An NADH molecule lies in a crevice between the tips of 7 and 10 with the pyrophosphate group straddling the  sheet and the two ends on the opposite sides of the  sheet. The NADH molecule is held by hydrophobic interactions and by hydrogen bonds (Fig. 3). The nicotinamide moiety of NADH is positioned near the bottom of the fat U and faces the  strands of the catalytic domain.

View larger version (38K): [in this window] [in a new window]   Fig. 3.   The NAD binding sites observed in the WT structure (A) and the D244E structure (B). It is noted that both NAD binding sites are quite similar to each other.

An Ado molecule or Ado* is found in the active site of the catalytic domain (Fig. 4). The ribose moiety of Ado lies in a crevice between the tips of 1 and 4, and the adenine ring occupies a hydrophobic pocket located in the catalytic domain. The Ado is connected by hydrogen bonds not only to the catalytic domain, but also to the NAD-binding domain (Fig. 5). The ribose ring is approximately parallel to the nicotinamide ring of NADH. The distance between the C3' of Ado and the C4 of NADH is 3.3 Å (Fig. 3B), indicating that the H3' can be directly transferred between the C3' of Ado and the C4 of NAD⁺ with only minor structural adjustments. The details of hydrogen-bond networks are given in Fig. 6.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   A (Fo  c) map showing the Ado*-like electron density peak. It notes that one of four subunits showed an Ade-like electron density peak. The contour is drawn at 3 level.

View larger version (39K): [in this window] [in a new window]   Fig. 5.   The Ado-binding sites observed in the WT structure (A) and the D244E structure (B). It is noted that the WT structure has a well developed substrate binding site.

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Schematic diagrams of interactions of NADH and Ado* in the active site of the D244E structure. The possible hydrogen bonds are indicated by dashed lines.

The temperature factors of the bound NADH and Ado were relatively high in comparison to those of the surrounding protein atoms, indicating that the NADH and Ado molecules do not fully occupy the sites. A simple analysis between the temperature factor and the occupancy factor indicated that D244E contains ~0.5 mol of each NADH and Ado/mol of subunit.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Effect of D244E Mutation on the Subunit Structure-- The D244E mutation shifts the enzyme conformation from open to closed. However, as summarized in Table II, there is no significant structural difference among the catalytic domains and the NAD-binding domains of the three known crystal structures of AdoHcyase (rat liver WT enzyme, human placenta enzyme, and D244E mutant enzyme). The small C-terminal domains (residues 390-431) of the rat WT and human enzymes are quite similar to each other (r.m.s.d. = 0.33 Å), whereas the C-terminal domain of the D244E mutant is different from those of the WT enzyme and the human enzyme (r.m.s.d. = 1.11 and 1.03 Å). An especially significant difference is seen around residues 420-427 (Fig. 7). Thus, the D244E mutation in the NAD-binding domain produces its effects in the structure of the C-terminal domain rather than in the structure of the NAD-binding domain. The electron density of residues 420-427 is relatively weak, indicating that these amino acid residues are disordered. The disordering of the amino acid residues is apparently caused by the D244E mutation, since no such disordering is observed in the WT structure. The carboxylic group of Glu-244 apparently expels Phe-424B and Lys-425B of subunit B from their original sites. Consequently, Phe-424 and Lys-425 and several amino acid residues connected to them are disordered. A hydrogen bond between Lys-425B and the phosphate of NAD⁺ observed in the structure of WT enzyme is not seen in the D244E structure. The structural changes by the D244E mutation is summarized in Fig. 8.

View larger version (35K): [in this window] [in a new window]   Fig. 7.   A movement of the section of C-terminal domain (residues: 420B-430B) of subunit B. A, WT structure (open conformation); B, D244E structure (closed conformation). The NAD-binding domains are aligned by a least-squares method.

View larger version (26K): [in this window] [in a new window]   Fig. 8.   Summary of D244E mutation effects on the enzyme structure. A, WT and human enzymes; B, D244E mutant enzyme. Possible hydrogen bonds are indicated by dashed lines. Upon the mutation, the C-terminal domain of subunit B changes the conformation and the catalytic domain moved from the open to the closed site.

Open Versus Closed Structures-- The structures of the D244E and human placental enzyme complexed with 2'-hydroxy-3'-ketocyclopent-4'-enyladenine (10) have a closed conformation, whereas the structure of the substrate-free rat-liver WT enzyme (11) has an open conformation (Fig. 9). However, the tetrameric cores composed of the four NAD-binding domains of the D244E structure (closed conformation), the WT structure (open conformation), and the human placenta structure (closed conformation) are quite similar to each other (Table II). Although the four small C-terminal domains are separated from the main body of the subunit, they are tightly bound to the NAD-binding domain of the adjacent subunit. Even when the four C-terminal domains are included with the tetrameric core formation, the structures of the three AdoHcyases remain quite similar to each other. These observations indicate that the open-closed conformational changes in AdoHcyase are mainly due to movement of the relatively rigid individual catalytic domains. A rigid body analysis indicates three possible hinge sections (residues 183-184, 356-357, and 390-391) between the catalytic domain and the NAD-binding domain or the C-terminal domain. These hinges are located in the -helix sections (7, 14, 15) rather than the loop sections, and form approximately a straight line. A simple analysis indicates that the individual catalytic domain rotates by 16.9° around an axis passing through the molecular hinges (Fig. 9). After rotating the four catalytic domains of the WT structure by 16.9°, the r.m.s.d. between the WT and D244E structures is reduced to 0.73 Å from 4.50 Å.

View larger version (44K): [in this window] [in a new window]   Fig. 9.   "Open" and "closed" structures of the AdoHcyase. The NAD-binding domains of the WT and D244E structures are aligned by a least-squares method. The WT structure (open) and D244E structure (closed) are illustrated by thin and thick coils, respectively.

The rotation of the catalytic domain brings the bound substrate/inhibitor to the bound NAD⁺ into proximity so that the substrate/inhibitor can be oxidized by the bound NAD⁺. In the D244E structure, the distance between the C3' of Ado and the nicotinamide C4 of NADH is 3.3 Å, and the C3'-H3' is pointed toward the C4 (H3' ... C4 = 2.3 Å), indicating that a direct hydride transfer is possible. As shown in Fig. 5A, the open structure observed in the substrate-free WT enzyme has a substrate/inhibitor binding site quite similar to that seen in the D244E closed structure. Thus, a substrate/inhibitor can readily enter or leave the substrate binding site when the enzyme adopts an open structure since there is a large cleft between the catalytic and NAD-binding domains (Fig. 2). On the other hand, a substrate/inhibitor cannot freely enter or leave when the enzyme adopts a closed structure since the cleft is closed. Thus, it is reasonable to assume that the large cleft between the two domains is closed upon binding of substrate to bring NAD⁺ and substrate in proximity, and is opened upon completion of the catalytic reaction to release products. The open-closed mechanism is therefore an important aspect of AdoHcyase catalysis.

Probability of Open and Closed Conformation in the Wild-type and D244E Mutant Enzymes-- It is necessary to define a probability that the wild-type enzyme and the D244E mutant enzyme purified with the procedures described under "Experimental Procedures" have an open conformation and a closed conformation, respectively. There are two pieces of evidence that support the open-closed conformational changes upon the D244E mutation. Those are as follows. 1) The crystals of wild-type enzyme and D244E enzyme have quite different topologies and have quite different unit cell dimensions. The unit cell dimensions of the five different crystals of wild-type enzyme have been determined and agreed to each other within 1% error. Similarly, the unit cell dimensions of the five different crystals of D244E enzyme have agreed to each other within 1% error. If the enzyme changes its conformation from open to closed and vice versa, then the unit cell dimensions will be drastically changed since the 17° movement of the catalytic domain changes the molecular volume. It is noted that the Matthews volumes (V_M) of the open (wild-type enzyme) and closed (D244E mutant enzyme) conformers are 3.10 and 2.44 Å³/Da, respectively. 2) The crystals of wild-type enzyme are cracked and dissolved by adding a drop of Ado solution into the mother liquor, indicating that a large conformational change (open to closed) occurs inside the crystals. On the other hand, the crystals of D244E mutant enzyme are reasonably stable in the mother liquor containing Ado. Considering the two evidences mentioned above, the probability that the wild-type enzyme and the D244E mutant enzyme have an open conformation and a closed conformation, respectively, should be very high.

Open-Closed Mechanism-- In the D244E structure, there are no systematic differences between the temperature factors of the catalytic domain and the NAD-binding domain. On the other hand, for the open wild-type structure (11), the temperature factors of the catalytic domain are much higher than those of the NAD-binding domain. These observations indicate that the catalytic domain is relatively mobile in the open conformation. The high mobility of the catalytic domain is due to the unique architecture of the tetramer. As described in the previous section, a ~50 × 10 × 10-Å channel is formed in the center of the tetramer. This channel structure not only provides a strong core framework for the enzyme structure, but also provides for mobility of the catalytic domains. The activation energy for the transition between the open and closed conformations should be quite small so that the catalytic domain can swing constantly when the substrate-binding site is empty, i.e. the molecular hinges should be quite flexible. When the enzyme adopts a closed conformation without the presence of a substrate/inhibitor, the affinity between the catalytic and NAD⁺-binding domain is relatively weak since there is a large substrate-binding space between them. For this reason, the substrate-free enzyme has an open structure as observed in the crystal structure of the WT enzyme (11). On the other hand, when the space is occupied by a substrate/inhibitor, the affinity between the two domains becomes strong enough to stabilize a closed-conformation structure since the substrate/inhibitor mediates interactions between the two domains. In the D244E structure, the bound Ado molecule forms hydrogen bonds with both the catalytic domain (His-54, Thr-56, Glu-58, Glu-155) and the NAD-binding domain (Lys-185, Asp-189, His-300, His-352) (Fig. 5B). Furthermore, repulsions between the catalytic domain and the NAD-binding domain would be reduced when the bound NAD⁺ becomes electrically neutral (NADH). For these reasons, the structures of D244E mutant enzyme and the human enzyme have closed conformations because both structures contain substrates or inhibitors and NADH.

Effect of the D244E Mutation on the NAD⁺ Affinity-- NAD molecules in AdoHcyase are tightly bound to the protein and are unable to be released from their binding sites under physiological condition (1). Indeed, NAD molecules in the crystal structures of AdoHcyase are tightly connected to the NAD-binding domain. The adenosine moiety of NAD is further covered by the C-terminal domain of the adjacent subunit (Fig. 2). In the structures of the WT enzyme (open conformation) and the human enzyme (closed conformation), Lys-425 and Tyr-429 of the adjacent subunit (subunit B) participate in hydrogen bonding with the Ado ribose and pyrophosphate of NAD, respectively (Fig. 3A). On the other hand, in the D244E structure, residues 420-427 are disordered and Lys-425 is no longer involved in hydrogen bonding with the Ado ribose of NAD (Fig. 3B). For this reason, the NAD affinity to the mutant enzyme should be reduced substantially. The dissociation constant for NAD, K_d = 23 µM at 25 °C, is 280-fold greater than that of the WT enzyme (K_d = 83 nM at 25 °C) (9). A difference of a few hundredfold in the K_d values for NAD⁺ translates to a binding energy in the D244E enzyme that is lower by ~14 kJ·mol¹ than for the WT enzyme. The magnitude of this binding energy difference falls in the range expected for hydrogen bonding involving a charged hydrogen bond donor/acceptor (16). Thus the difference between the dissociation constants for NAD⁺ of the WT and D244E enzymes is consistent with the difference between the NAD binding geometries found in the WT and D244E crystal structures.

D244E Mutant Enzyme Contains ~0.5 mol of Each NADH and Ado*-- Gomi et al. reported that the mutant enzyme had only 0.05 mol of NAD⁺ but contained ~0.6 mol of each NADH and adenine/mol of subunit (9). As described under "Experimental Procedures," the D244E crystals contained both Ade and Ado with a 1:2 ratio. The present x-ray study indicates that the mutant enzyme contains ~0.5 mol of each NADH and Ado*/mol of subunit. Except for the adenine-adenosine discrepancy, the overall agreement between the data of Gomi et al. and this study is excellent. It is noted that it is impossible to distinguish between NAD⁺ and NADH by an x-ray analysis and it is also impossible to determine the structure of the Ado* from a 2.8-Å resolution map. From a consideration of the reaction mechanism of AdoHcyase, the mutant enzyme might be expected to contain either NADH and oxidized Ado (3'-ketoadenosine) or NAD⁺ and Ado. As already mentioned, the structure of WT enzyme has an open conformation and contains four NAD⁺ molecules and no substrate/inhibitor, whereas the structure of the human enzyme has a closed conformation and contains NADH and an oxidized inhibitor. Thus, the contents, NADH and 3'-keto-adenosine, would be consistent with the results of the other two crystal structures.

As first reported by Hershfield (17), and by Chiang et al. (18), AdoHcyase undergoes a mechanism-based inactivation in vitro upon incubation with Ado or a number of analogues. The mechanism of this inactivation was studied by Abeles et al. (19), who reported that 2'-dAdo reduces enzyme-bound NAD⁺ to NADH with the concomitant oxidation of 2'-dAdo to its 3'-keto derivative. The 3'-keto compound is unstable and readily eliminates adenine. The Ado derivative found in the D244E structure might be stabilized by the active site environment. When the bound intermediate (Ado*) was extracted with ethanol, the intermediate would be converted to Ade or Ado depending on the experimental conditions. Thus, the contents in the extracted solution and the crystal structure might have some discrepancy.

Two independent studies have suggested that AdoHcyase has a half-site reactivity. 1) Complete inactivation can be observed with reduction of only two of the four enzyme-bound NAD⁺ (19). 2) Reactivation of the apo-enzyme prepared by acid-ammonium sulfate treatment of rat liver AdoHcyase proceeds in a biphasic fashion; two of four subunits are almost instantaneously occupied by NAD⁺ molecules, and the other two subunits are much slowly occupied by NAD⁺ molecules (20). In light of these observations, 0.5 mol each of NADH⁺ and Ado*/mol of subunit found in the D244E structure could be interpreted as follows. The D244E mutagenesis might increase a half-site reactivity in AdoHcyase, i.e. the D244E mutant enzyme might be composed of two active subunits (₂) and two inactive subunits (₂). Specifically one of subunits A and B has high NAD⁺ affinity and thus traps 1 mol each of NADH and Ado* whereas the other has low NAD⁺ affinity and thus traps neither NADH nor Ado*. Similarly one of subunits C and D traps 1 mol each of NADH and Ado* whereas the other has neither NADH nor Ado*. Since the ligand-and-substrate-bound subunits are randomly distributed in the individual enzymes, the tetrameric enzyme elucidated in the crystal structure is composed of four equivalent subunits, which are an averaged structure of the ligand-and-substrate-bound subunit and apo subunit.

The Mutant Enzyme Has a Weak NAD⁺ Affinity but a Strong NADH Affinity-- A significant affinity difference between NAD⁺ and NADH was observed in the D244E mutant enzyme, i.e. the mutant enzyme has a relatively weak NAD⁺ affinity (K_d = 23 µM), but has a strong NADH affinity (K_d = 22 nM) (9). As described in the previous section, there is no significant geometrical difference between the NAD⁺ binding sites in the open structure (as seen in the WT enzyme structure) and in the closed structure (as seen in the human enzyme complexed with an inhibitor). Therefore, the significant affinity difference is mainly due to the enzyme conformation. The NAD⁺ molecule in the open structure is readily released from the D244E enzyme since the C-terminal domain of the adjacent subunit is disordered and consequently is less participates in the NAD⁺ binding. On the other hand, the NADH molecule in the closed structure is completely trapped in the enzyme and cannot escape from the enzyme.

The D244E Mutation Shifts the Catalytic Reaction from Reversible to Irreversible-- Since the reaction catalyzed by AdoHcyase is reversible and the WT and D244E mutant enzyme were purified and crystallized without adding any Ado or AdoHcy, their active sites should be empty because the Ado concentration in the crystallization solution is extremely low.

However, as described above, the D244E mutant enzyme (E') contains ~0.5 mol of Ado*/subunit. This indicates that the D244E mutation shifts the catalytic reaction from reversible to irreversible.

E. coli contains a relatively high level of Ado so that the following processes would occur in an early stage of the purification; an Ado binds to the active site of D244E subunit, the subunit changes conformation from open to closed, the bound NAD⁺ oxidizes the bound Ado, and the reaction pauses. Since the enzyme adopts a closed conformation, the NADH and the reduced Ado are trapped in the enzyme during the purification and crystallization. However, occasionally, the catalytic reaction would be reversed so that the subunit changes to the open conformation, and both Ado and NAD⁺ are released from the active site. For this reason, the D244E mutant enzyme contains ~0.5 mol each NADH and Ado*/mol of subunit.

It should be noted that two of the four subunits of the D244E mutant enzyme would have neither substrate nor NAD⁺ in their active sites. Nevertheless the D244E mutation shifts the conformational stability from the open to closed conformation. In both the closed structure of the human enzyme and the open structure of the WT enzyme, His-428B in the C-terminal domain of adjacent subunit has two short contacts with Asp-181 in the catalytic domain located near the molecular hinge section (CD2-OD2 (Asp-181), NE2-OD2 (Asp-181)). As described above, the end section (residues 420-427) of the small C-terminal domain in the D244E structure is moved and disordered upon the D244E mutation. Although His-428B is slightly disordered, it apparently interacts with not only Asp-181 but also Tyr-164 (Figs. 7B and 8). The imidazole ring of His-428B is apparently rotated by 180° so that ND2 of His-428B participates in a hydrogen bond with the OH of Tyr-164 of the catalytic domain. This His-428B ··· Tyr-164 interaction would stabilize the closed conformation of both holo and apo subunits.

Proposed Catalytic Mechanism-- On the basis of the crystal structures of the WT enzyme and the D244E mutant enzyme, the mechanism of oxidation of Ado could be visualized as shown in Fig. 10. The structure of the open conformation has an intact Ado/AdoHcy binding site, and its NZ (NH₃⁺) of Lys-185 is involved in three hydrogen-bonds (O (Glu-155), OD1 (Asn-180), and OD1 (Asp-189)). An Ado binds to the substrate-binding site and is held to the enzyme by polar and non-polar interactions. Upon the binding of Ado, the catalytic domain moves toward the NAD-binding domain so as to place the C3'-H near the C4 of nicotinamide of NAD⁺. Asp-189 abstracts a proton from NZ of Lys-185, and forms a hydrogen bond with O₂' (Ado) as a hydrogen bond donor. The activated NZ (NH₂) of Lys-185 serves as a base to accept the proton from O3'-H. The basicity of NZ of Lys-185 would be increased by approaching Asn-190 to Lys-185. Asp-130 is near the C4' of Ado (C4' ··· OD1 (Asp-130) = 3.1 Å, C4' ··· OD2 (Asp-130) = 3.3 Å). Thus, it is highly likely that Asp-130 acts as a base to directly abstract the proton from C4' of the substrate. The resulting carbanion then releases H₂O to form the 3'-keto-4',5'-dehydroadenosine intermediate. Although Turner et al. (10) suggested that a disordered water molecule near C4' participates in abstracting the proton of C4', such a water molecule has not been observed in this structure.

View larger version (17K): [in this window] [in a new window]   Fig. 10.   A proposed catalytic mechanism based on the crystal structures of the WT enzyme and the D244E mutant enzyme. The proton movements are indicated by arrow lines.

	ACKNOWLEDGEMENTS

We thank Professors Richard H. Himes and Richard L. Schowen for a critical reading of this manuscript and very valuable comments.

	FOOTNOTES

^* The work carried out at the University of Kansas was supported by National Institutes of Health Grant GM37233.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The atomic coordinates and the structure factors (code 1D4F) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).

^¶ To whom all correspondence should be addressed: Dept. of Molecular Biosciences, 3042 Haworth Hall, University of Kansas, Lawrence, KS 66045-2106. Tel.: 785-864-4727; Fax: 785-864-5321; E-mail: xraymain@ukans.edu.

Published, JBC Papers in Press, July 26, 2000, DOI 10.1074/jbc.M003725200

	ABBREVIATIONS

The abbreviations used are: AdoHcyase, S-adenosyl-L-homocysteine hydrolase; Ado, adenosine; AdoHcy, S-adenosyl-L-homocysteine; Hcy, L-homocysteine; AdoMet, S-adenosyl-L-methionine; NAD, NAD⁺ or NADH; Ado*, adenosine-like intermediate (probably 3'-ketoadenosine); WT, wild-type enzyme; PEG, poly(ethylene glycol); r.m.s.d., root-mean-square deviation.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				P. S.C.F. Rocha, M. Sheikh, R. Melchiorre, M. Fagard, S. Boutet, R. Loach, B. Moffatt, C. Wagner, H. Vaucheret, and I. Furner The Arabidopsis HOMOLOGY-DEPENDENT GENE SILENCING1 Gene Codes for an S-Adenosyl-L-Homocysteine Hydrolase Required for DNA Methylation-Dependent Gene Silencing PLANT CELL, February 1, 2005; 17(2): 404 - 417. [Abstract] [Full Text] [PDF]

				Y. Takata, T. Yamada, Y. Huang, J. Komoto, T. Gomi, H. Ogawa, M. Fujioka, and F. Takusagawa Catalytic Mechanism of S-Adenosylhomocysteine Hydrolase. SITE-DIRECTED MUTAGENESIS OF ASP-130, LYS-185, ASP-189, AND ASN-190 J. Biol. Chem., June 14, 2002; 277(25): 22670 - 22676. [Abstract] [Full Text] [PDF]

				Y. Huang, J. Komoto, Y. Takata, D. R. Powell, T. Gomi, H. Ogawa, M. Fujioka, and F. Takusagawa Inhibition of S-Adenosylhomocysteine Hydrolase by Acyclic Sugar Adenosine Analogue D-Eritadenine. CRYSTAL STRUCTURE OF S-ADENOSYLHOMOCYSTEINE HYDROLASE COMPLEXED WITH D-ERITADENINE J. Biol. Chem., February 22, 2002; 277(9): 7477 - 7482. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/41/32147    most recent M003725200v1