FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2020779013> ?p ?o ?g. }

• W2020779013 endingPage "7482" @default.
• W2020779013 startingPage "7477" @default.
• W2020779013 abstract "d-Eritadenine (DEA) is a potent inhibitor (IC50 = 7 nm) ofS-adenosyl-l-homocysteine hydrolase (AdoHcyase). Unlike cyclic sugar Ado analogue inhibitors, including mechanism-based inhibitors, DEA is an acyclic sugar Ado analogue, and the C2′ and C3′ have opposite chirality to those of the cyclic sugar Ado inhibitors. Crystal structures of DEA alone and in complex with AdoHcyase have been determined to elucidate the DEA binding scheme to AdoHcyase. The DEA-complexed structure has been analyzed by comparing it with two structures of AdoHcyase complexed with cyclic sugar Ado analogues. The DEA-complexed structure has a closed conformation, and the DEA is located near the bound NAD+. However, a UV absorption measurement shows that DEA is not oxidized by the bound NAD+, indicating that the open-closed conformational change of AdoHcyase is due to the substrate/inhibitor binding, not the oxidation state of the bound NAD. The adenine ring of DEA is recognized by four essential hydrogen bonds as observed in the cyclic sugar Ado complexes. The hydrogen bond network around the acyclic sugar moiety indicates that DEA is more tightly connected to the protein than the cyclic sugar Ado analogues. The C3′-H of DEA is pointed toward C4 of the bound NAD+ (C3′···C4 = 3.7 Å), suggesting some interaction between DEA and NAD+. By placing DEA into the active site of the open structure, the major forces to stabilize the closed conformation of AdoHcyase are identified as the hydrogen bonds between the backbone of His-352 and the adenine ring, and the C3′-H···C4 interaction. DEA has been believed to be an inactivator of AdoHcyase, but this study indicates that DEA is a reversible inhibitor. On the basis of the complexed structure, selective inhibitors of AdoHcyase have been designed. d-Eritadenine (DEA) is a potent inhibitor (IC50 = 7 nm) ofS-adenosyl-l-homocysteine hydrolase (AdoHcyase). Unlike cyclic sugar Ado analogue inhibitors, including mechanism-based inhibitors, DEA is an acyclic sugar Ado analogue, and the C2′ and C3′ have opposite chirality to those of the cyclic sugar Ado inhibitors. Crystal structures of DEA alone and in complex with AdoHcyase have been determined to elucidate the DEA binding scheme to AdoHcyase. The DEA-complexed structure has been analyzed by comparing it with two structures of AdoHcyase complexed with cyclic sugar Ado analogues. The DEA-complexed structure has a closed conformation, and the DEA is located near the bound NAD+. However, a UV absorption measurement shows that DEA is not oxidized by the bound NAD+, indicating that the open-closed conformational change of AdoHcyase is due to the substrate/inhibitor binding, not the oxidation state of the bound NAD. The adenine ring of DEA is recognized by four essential hydrogen bonds as observed in the cyclic sugar Ado complexes. The hydrogen bond network around the acyclic sugar moiety indicates that DEA is more tightly connected to the protein than the cyclic sugar Ado analogues. The C3′-H of DEA is pointed toward C4 of the bound NAD+ (C3′···C4 = 3.7 Å), suggesting some interaction between DEA and NAD+. By placing DEA into the active site of the open structure, the major forces to stabilize the closed conformation of AdoHcyase are identified as the hydrogen bonds between the backbone of His-352 and the adenine ring, and the C3′-H···C4 interaction. DEA has been believed to be an inactivator of AdoHcyase, but this study indicates that DEA is a reversible inhibitor. On the basis of the complexed structure, selective inhibitors of AdoHcyase have been designed. S-Adenosyl-l-homocysteine hydrolase (AdoHcyase, 1AdoHcyaseS-adenosyl-l-homocysteine hydrolaseAdoHcyS-adenosyl-l-homocysteineHcyl-homocysteinerWTrat liver AdoHcyaserD244Erat liver D244E mutant AdoHcyaseAdo*3′-keto-adenosinerD244E:Ado*rD244E complexed with Ado*ADC2′,3′-dihydroxycyclopenten-4′-yl-adeninehWThuman placenta AdoHcyasehWT:ADChWT complexed with ADCDEAd-eritadeninerWT:DEArWT complexed with ADCPEGpolyethylene glycolrmsdroot mean square deviation EC 3.3.1.1) catalyzes the hydrolysis of S-adenosyl-l-homocysteine (AdoHcy) to form adenosine (Ado) and homocysteine (Hcy) (1Cantoni G.L. Scarano E. J. Am. Chem. Soc. 1954; 76: 4744Crossref Scopus (41) Google Scholar). The enzymes from all sources are oligomeric proteins with subunits ofM r 45,000–50,000. Each subunit contains one mole of tightly bound NAD+ (2Fujioka M. Takata Y. J. Biol. Chem. 1981; 256: 1631-1635Abstract Full Text PDF PubMed Google Scholar, 6Palmer J.L. Abeles R.H. J. Biol. Chem. 1979; 254: 1217-1226Abstract Full Text PDF PubMed Google Scholar). 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 that the net reaction proceeds in the direction of hydrolysis (3Richards H.H. Chiang P.K. Cantoni G.L. J. Biol. Chem. 1978; 253: 4476-4480Abstract Full Text PDF PubMed Google Scholar). 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 has been reported that an elevated plasma Hcy level is one of the risk factors for coronary heart disease (4Nygard O. Nordrehaug J.E. Refsum H. Ueland P.M. Farstad M. Vollset S.E. N. Engl. J. Med. 1997; 337: 230-236Crossref PubMed Scopus (1662) Google Scholar). S-adenosyl-l-homocysteine hydrolase S-adenosyl-l-homocysteine l-homocysteine rat liver AdoHcyase rat liver D244E mutant AdoHcyase 3′-keto-adenosine rD244E complexed with Ado* 2′,3′-dihydroxycyclopenten-4′-yl-adenine human placenta AdoHcyase hWT complexed with ADC d-eritadenine rWT complexed with ADC polyethylene glycol root mean square deviation The mechanism of reversible hydrolysis of AdoHcy catalyzed by AdoHcyase has been studied by Palmer and Abeles (5Palmer J.L. Abeles R.H. J. Biol. Chem. 1976; 251: 5817-5819Abstract Full Text PDF PubMed Google Scholar, 6Palmer J.L. Abeles R.H. J. Biol. Chem. 1979; 254: 1217-1226Abstract Full Text PDF PubMed Google Scholar). On the basis of crystal structures of rat liver AdoHcyase (rWT) and rD244E:Ado* (3-keto-adenosine) determined in our laboratory, we have proposed a detailed catalytic mechanism (7Komoto J. Huang Y. Gomi T. Ogawa H. Takata Y. Fujioka M. Takusagawa F. J. Biol. Chem. 2000; 275: 32147-32156Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). The substrate-free enzyme takes mainly an open conformation. Once the substrate enters the active site, Asp-189 that hydrogen bonds to the ε-NH3+ of Lys-185 moves away, taking a proton from the Lys residue. The catalytic domain closes a large cavity between it and the NAD-binding domain by rotating 17° around the molecular hinge section so that the substrate is brought closer to the bound NAD+. The resultant neutral Lys-185 and the bound NAD+ remove protons from 3′-OH and 3′-CH, respectively. Subsequently, the 4′-CH proton is abstracted by Asp-130. The resulting carbanion then releases H2O/Hcy to form the 3′-keto-4′, 5′-dehydroadenosine intermediate. AdoHcy is a potent inhibitor ofS-adenosyl-l-methionine (AdoMet)-dependent methyltransferases (8Hurwitz J. Gold M. Anders M. J. Biol. Chem. 1964; 239: 3474-3482Abstract Full Text PDF PubMed Google Scholar, 9Deguchi T. Barchas J. J. Biol. Chem. 1971; 246: 3175-3181Abstract Full Text PDF PubMed Google Scholar, 10Coward J.K. Slisz E.P. Wu F.Y.-H. Biochemistry. 1973; 12: 2291-2297Crossref PubMed Scopus (81) Google Scholar, 11Pugh C.S.G. Borchardt R.T. Stone H.O. Biochemistry. 1977; 16: 3928-3932Crossref PubMed Scopus (61) Google Scholar, 12Hasebe M. Mckee J.G. Borchardt R.T. Antimicrob. Agents Chemother. 1989; 33: 828-834Crossref PubMed Scopus (52) Google Scholar). Because AdoHcyase is the only enzyme involved in AdoHcy metabolism, and because the reaction it catalyzes is reversible, the activity of AdoHcyase is thought to play a critical role in the control of tissue levels of AdoHcy and hence to modulate the activities of various methyltransferases (13Cantoni G.L. Borchardt R.T. Creveling C.R. Ueland P.M. Biological Methylation and Drug Design. Humana Press, Clifton, NJ1986: 227-238Crossref Google Scholar). Inhibition of AdoHcyase in vivoelevates the AdoHcy level, and consequently the AdoMet-dependent transmethylation is suppressed. Therefore, AdoHcyase has been an attractive target for the design of antiviral agents because most viruses require a methylated cap structure at the 5′-terminus of their mRNA for viral replication (Ref. 14Banerjee A.K. Microbiol. Rev. 1980; 44: 175-205Crossref PubMed Google Scholar and references therein and Ref. 15Morch M.-D. Joshi R.L. Denial T.M. Haenni A.L. Nucleic Acids Res. 1987; 15: 4123-4130Crossref PubMed Scopus (41) Google Scholar), and the virus-encoded methyltransferases that are involved in the formation of this methylated cap structure are inhibited by AdoHcy. A number of inhibitors of AdoHcyase have been identified. These are classified into two groups. One group includes Ado or 3-deazaadenosine analogues with carbocyclic ribose moieties such as aristeromycin (16Guranowski A. Montgomery J.A. Cantoni G.L. Chiang P.K. Biochemistry. 1981; 20: 110-115Crossref PubMed Scopus (139) Google Scholar,17Hill D.L. Straight S. Allan P.W. Bennett L.L., Jr. Mol. Pharmacol. 1971; 7: 375-380PubMed Google Scholar), neplanocin A (18Borchardt R.T. Keller B.T. Patel-Thombre U. J. Biol. Chem. 1984; 259: 4353-4358Abstract Full Text PDF PubMed Google Scholar, 19Yaginuma S. Muto N. Tsujino M. Sudate Y. Hayashi M. Otani M. J. Antibiot. (Tokyo). 1981; 34: 359-366Crossref PubMed Scopus (281) Google Scholar, 20De Clercq E. Antimicro. Agents Chemother. 1985; 28: 84-89Crossref PubMed Scopus (209) Google Scholar), and 2′,3′-dihydroxycyclopenten-4′-yl-adenine (ADC) (21Narayanan S.R. Keller B.T. Borcherding D.R. Scholtz S.A. Borchardt R.T. J. Med. Chem. 1988; 31: 500-508Crossref PubMed Scopus (37) Google Scholar). The other group contains Ado analogues with acyclic sugar moieties such asd-eritadenine (DEA) (22Merta A. Votruba I. Vesely J. Holy A. Collect. Czech. Chem. Commun. 1983; 48: 2701-2708Crossref Google Scholar, 23Holy A. Votruba I. De Clercq E. Collect. Czech. Chem. Commun. 1982; 47: 1392-1407Crossref Scopus (40) Google Scholar, 24De Clercq E. Bergstrom D.E. Holy A. Montgomery J.A. Antiviral Res. 1984; 4: 119-133Crossref PubMed Scopus (64) Google Scholar, 25Votruba I. Holy A. Collect. Czech. Chem. Commun. 1982; 47: 167-172Crossref Scopus (38) Google Scholar), 9-(S)-(2,3-dihydroxypropyl)adenine (26De Clercq E. Descamps J., De Somer P. Holy A. Science. 1978; 200: 563-565Crossref PubMed Scopus (226) Google Scholar, 27De Clercq E. Holy A. J. Med. Chem. 1979; 22: 510-513Crossref PubMed Scopus (94) Google Scholar, 28Votruba I. Holy A. Collect. Czech. Chem. Commun. 1980; 45: 3039-3044Crossref Google Scholar), and (R,S)-3-adenine-9-yl-2-hydroxypropanoic acid (29De Clercq E. Holy A. J. Med. Chem. 1985; 28: 282-287Crossref PubMed Scopus (51) Google Scholar,30De Clercq E. Microbiologica (Pavia). 1990; 13: 165-178PubMed Google Scholar). Cyclic sugar Ado analogues, including mechanism-based inhibitors, are oxidized by the bound NAD+, and thus the inhibited AdoHcyase contains NADH rather than NAD+ in the active site. On the other hand, most acyclic sugar Ado analogues except for DEA are relatively weak reversible inhibitors. DEA is a potent inhibitor of AdoHcyase (IC50 = 7 nm) and is believed to be an inactivator of AdoHcyase (31Schanche J.-S. Schanche T. Ueland P.M. Holy A. Votruba I. Mol. Pharmcol. 1984; 26: 553-558PubMed Google Scholar). As illustrated in Scheme FS1, the chiral centers (C2′ and C3′) of DEA have opposite chirality to those of Ado and cyclic sugar Ado analogues. Therefore, if the adenine ring of DEA binds to the enzyme in a similar fashion as observed in the structures of rD244E:Ado* and hWT:ADC, the acyclic sugar moiety must have quite a different interaction with the enzyme. To elucidate the binding scheme of DEA, we have determined the crystal structures of rWT inhibited by DEA and pure sodium DEA salt and analyzed the oxidation state of the bound NAD cofactor by UV measurement. Here we report why DEA is a potent inhibitor of AdoHcyase despite having an acyclic sugar moiety. The enzyme crystals of rWT and rWT:DEA were washed with 30% PEG 4000 solution and were subsequently dissolved in Tris/HCl buffer. The proteins were denatured by adding two volumes of absolute ethanol. The precipitates were washed once with 70% ethanol, and the combined supernatants were driedin vacuo. The residues were dissolved in 100 mm Tris/HCl (pH 7.2) and analyzed by UV absorption spectrometry using a JASCO V560 spectrophotometer. The spectra of rWT and rWT:DEA gave 0.038 and 0.042 absorbances at 260 nm, respectively. Absorption spectra of 1.0 μm concentrations of NAD+ + DEA (1:1) and NADH + DEA (1:1) were also recorded for references. The absorbances at 260 nm were 0.037 and 0.046, respectively, indicating that the concentrations of the extractions and the references were similar. Sodium DEA (Na+·[C9H10N5O4]−·2.5H2O) was provided by the Tanabe Research Laboratory (San Diego, CA) and was recrystallized from aqueous solution. A colorless plate-shaped crystal of dimensions 0.43 × 0.40 × 0.04 mm was selected for structural analysis. X-ray diffraction data for this compound were collected using a Bruker SMART APEX CCD area detector using graphite-monochromated MoKα radiation (λ = 0.71073 Å). The data were collected in the range 3.57 < θ < 30.50° at −173 °C (0.70-Å resolution). Coverage of unique data was 99.8% complete to 26.0° in θ. The unit cell parameters (a = 7.873(1), b = 8.363(1),c = 20.557(1) Å, β = 97.99(1)°) were determined from a least squares fit of 4542 reflections. From 117 reflections that were measured both at the beginning and end of data collections, the crystal showed no decay during the data measurement. The data were corrected for absorption by a semiempirical method from equivalent reflections giving minimum and maximum transmission factors of 0.9342 and 0.9936. Lorentz and polarization corrections were applied. The data were merged to form a set of 3799 independent data with R(int) = 0.0131. The space group C2 was determined by systematic absences in the data. The structure was solved by direct methods and refined by full-matrix least squares methods onF o2. The absolute configuration of the molecule was determined by using the anomalous dispersion effects from sodium and oxygen atoms in the crystal. Non-hydrogen atoms were refined with anisotropic thermal factors, whereas hydrogen atoms were refined with isotropic thermal factors. A total of 195 parameters were refined against the 3799 data to givewR(F o2) = 0.0923 andS = 1.062. The finalR(F o) was 0.0342 for the 3735 observed data (Fo > 4ς(Fo)). The coordinates have been deposited in the Cambridge Crystallographic Data Center (deposition number CCDC 171285). AdoHcyase used in this study is the recombinant rat enzyme produced in Escherichia coli JM109 transformed with a pUC118 plasmid that contains the coding sequence of rat AdoHcyase cDNA (32Gomi T. Date T. Ogawa H. Fujioka M. Aksamit R.R. Backlund P.S. Cantoni G.L. J. Biol. Chem. 1989; 264: 16138-16142Abstract Full Text PDF PubMed Google Scholar). The enzyme was purified to homogeneity from E. coli extracts by gel filtration over Sephacryl S-300 and DEAE-cellulose chromatography as described previously (32Gomi T. Date T. Ogawa H. Fujioka M. Aksamit R.R. Backlund P.S. Cantoni G.L. J. Biol. Chem. 1989; 264: 16138-16142Abstract Full Text PDF PubMed Google Scholar). Recombinant AdoHcyase lacks the N-terminal acetyl group but exhibits other structural features similar to those of the liver enzyme (32Gomi T. Date T. Ogawa H. Fujioka M. Aksamit R.R. Backlund P.S. Cantoni G.L. J. Biol. Chem. 1989; 264: 16138-16142Abstract Full Text PDF PubMed Google Scholar). 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 1 mm sodium DEA, 22% (w/v) PEG 4000, 50 mm Tris/HCl buffer, pH 7.2, and 10% (v/v) isopropyl alcohol 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 weeks. A crystal with dimensions of 0.3 × 0.2 × 0.1 mm in a hanging drop was scooped out with a nylon loop and 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 it was frozen in liquid nitrogen. The frozen crystal was transferred onto a Rigaku RAXIS IIc imaging plate x-ray diffractometer with a rotating anode x-ray generator as an x-ray source (CuKα radiation operated at 50 kV and 100 mA). The diffraction data were measured up to 3.0 Å resolution at −180 °C. The unit cell dimensions and the space group were uniquely determined from the observed data set. The data were processed with the program DENZO and SCALEPACK (33Otwinowski Z. Minor W. Methods Enzymol. 1997; 276: 307-326Crossref PubMed Scopus (38702) Google Scholar). The data statistics are given in Table I.Table IExperimental details and refinement parameters of crystal structure analysesExperimental details: Resolution20.0–3.0 Å No. crystals1 No. reflections measured241,990 No. unique reflections1-aUnique reflections in the range between 10.0 Å and highest resolution.59,048 % complete91.5 R sym1-bRsym = Σ‖I − 〈I〉‖/Σ‖I‖.0.085 I/ς(I)1-cI/ς(I) in 3.1–3.0 Å resolution range.3.8R.m.s.d. from the ideal valuesRefinement parameters No. residues3448 No. NAD+ molecules8Bond (Å)0.007 No. DEA molecules8Angle (°)1.12 No. water molecules381Torsion angle (°)25.8 R1-dR = Σ‖F o −F c‖/Σ‖F o‖.0.208Luzzati coord. errors (Å)0.034 R free0.265Ramachandran plotResidues in most favored region88.8%Residues in additional allowed region11.2%Space group: P21: cell dimension (Å):a = 89.87 Å, b = 177.37 Å,c = 112.16 Å, β = 107.6 °;M r of subunit: 47,410; no. subunits in the unit cell: 16; V M = 2.25 Å3; percentage of solvent content: 45%.1-a Unique reflections in the range between 10.0 Å and highest resolution.1-b Rsym = Σ‖I − 〈I〉‖/Σ‖I‖.1-c I/ς(I) in 3.1–3.0 Å resolution range.1-d R = Σ‖F o −F c‖/Σ‖F o‖. Open table in a new tab Space group: P21: cell dimension (Å):a = 89.87 Å, b = 177.37 Å,c = 112.16 Å, β = 107.6 °;M r of subunit: 47,410; no. subunits in the unit cell: 16; V M = 2.25 Å3; percentage of solvent content: 45%. The unit cell dimensions and the assigned space group indicated two tetrameric enzymes (eight subunits) in the asymmetric unit. The crystal structure was determined by a molecular replacement procedure using X-PLOR (34Brünger A.T. X-PLOR: A System for X-ray Crystallography and NMR, Version 3.1. Yale University Press, New Haven and London1993Google Scholar, 35Brünger A.T. Acta Crystallogr. 1990; A46: 46-57Crossref Scopus (361) Google Scholar). The structure of the rWT enzyme (open conformation) (36Hu Y. Komoto J. Huang Y. Gomi T. Ogawa H. Takata Y. Fujioka M. Takusagawa F. Biochemistry. 1999; 38: 8323-8333Crossref PubMed Scopus (108) Google Scholar) and the structure of the D244E mutant enzyme (closed conformation) (7Komoto J. Huang Y. Gomi T. Ogawa H. Takata Y. Fujioka M. Takusagawa F. J. Biol. Chem. 2000; 275: 32147-32156Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar) were used as search models. The closed structure model gave significantly better Patterson correlation-refinement results than the open structure model (i.e. the best refined Patterson correlation values of the closed and open structure models were 0.2418 and 0.1297, respectively). At this stage, the open structure model was abandoned. The crystal structure was refined by a standard refinement procedure in the X-PLOR protocol (34Brünger A.T. X-PLOR: A System for X-ray Crystallography and NMR, Version 3.1. Yale University Press, New Haven and London1993Google Scholar) with the noncrystallographic symmetry restraint. 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 the active site of each subunit. The shape of the electron density peak suggested that each individual subunit contains DEA. Initially the DEA molecule found in the crystal structure of sodium DEA was placed into the electron density peak in each subunit. The DEA molecule fit well into the electron density peak except for its O3′-hydroxyl group and C4′-carboxyl group. Once the torsion angle of the C2′−C3′ bond was changed from gauche − to transconformation (−60° → 180°), these two groups also fit into the electron density peak. All eight subunits were tightly restrained to have the similar conformation (i.e. rmsd < 0.09 Å). Refinement of isotropic temperature factors for individual atoms was carried out by the individual B-factor refinement procedure of X-PLOR (34Brünger A.T. X-PLOR: A System for X-ray Crystallography and NMR, Version 3.1. Yale University Press, New Haven and London1993Google Scholar) using bond and angle restraints. 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 was 0.208 for all data (2 ς cut off) from 8.0 to 3.0 Å resolution. TheR free for 10% randomly selected data is 0.265. The coordinates have been deposited in the Protein Data Bank (code number 1K0U). The UV absorption spectra of the ethanol extract of rWT:DEA crystals and of a pure NAD+:DEA (1:1) solution are very similar. The characteristic peak around 340 nm from NADH is not observed, indicating that the rWT:DEA crystals contain NAD+ rather than NADH, and thus the bound DEA is not oxidized by the bound NAD+. The crystal structure of sodium DEA hydrate (Na+·[C9H10N5O4]−·2.5H2O) has been determined at 0.7 Å resolution (Fig.1). The DEA molecule has a semiring-forming conformation. The torsion angles of the N9–C1′, C1′–C2′, and C3′–C4′ bonds in the C8–N9–C1′–C2′–C3′–C4′ chain are −110°, −165°, and −64°, respectively. A sodium ion coordinates to three oxygen atoms (O2′, O3′, O4a'). All nitrogens and oxygens participate in hydrogen bonding either as H-bond donors or H-bond acceptors. The adenine rings are stacked on top of each other with hydrophobic interactions. The crystallographic refinement parameters (Table I), final (2F o − F c) maps, and conformational analysis by PROCHECK (37Laskowski R.A. MacArthur M.W. Moss D.S. Thornton J.M. J. Appl. Crystallogr. 1993; 26: 283-291Crossref Google Scholar) indicate that the crystal structure of rWT:DEA has been successfully determined. The crystal contains two crystallographically independent tetrameric AdoHcyase molecules where each subunit of the tetrameric AdoHcyase molecule is related by 222 symmetry. Because the eight subunits are identical at a resolution of 3.0 Å, they have been tightly restrained to have the same structure (rmsd ≤ 0.09 Å) to increase the quality of the structure. The quality of the structure is even higher, although it was obtained at a moderate resolution. The AdoHcyase structure found in the rWT:DEA complex is very similar to those found in rD244E:Ado* and hWT:ADC complexes, although the rWT:DEA, rD244E:Ado*, and hWT:ADC complexes were crystallized in different crystal systems. The rmsd values between the Cα positions of the subunits of rWT:DEA and rD244E:Ado* and between those of rWT:DEA and hWT:ADC are 0.67 Å and 0.49 Å, respectively. The substrate-binding site is remarkably similar to those of the NADH-bound enzymes, indicating that the oxidation state of NAD does not affect the active site geometry. Furthermore, the geometry of the substrate-binding site is not changed by the binding of quite differently shaped molecules, indicating that the substrate-binding site is quite rigid. The complexed structures along the rWT structure indicate that the open-closed conformational change depends upon the substrate/inhibitor binding to the enzyme and not the NAD+/NADH oxidation state. The DEA molecule in the protein has an extended conformation and the torsion angles of the backbone chain, C8–N9–C1′–C2′–C3′–C4′, are −120°, 180°, and 180° for N9–C1′, C1′–C2′ and C2′–C3′ bonds, respectively (Fig.1 B). All oxygen atoms (O2′, O3′, O4a′, O4b′) participate in hydrogen bonding with the protein (Fig.2 A). The 3′-CH hydrogen is pointed toward C4 of NAD+ at a distance of 2.7 Å, indicating a weak interaction with the bound NAD+ molecule. The adenine ring of DEA binds at the same site as observed in the rD244E:Ado* and hWT:ADC structures. All nitrogen atoms except for N3 are involved in hydrogen bonding. The –CH2–S–CH3 moiety of Met-357 lies on the adenine ring. The bound NAD+ molecule has a similar conformation as the NAD+ molecule in the rWT structure and NADH molecules in the rD244E:Ado* and hWT:ADC structures. The NAD+ molecule is tightly bound to the protein with the similar interactions as seen in the rD244E:Ado* and hWT:ADC structures. Four crystal structures of AdoHcyase, including the one in this study, have been determined (7Komoto J. Huang Y. Gomi T. Ogawa H. Takata Y. Fujioka M. Takusagawa F. J. Biol. Chem. 2000; 275: 32147-32156Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar, 36Hu Y. Komoto J. Huang Y. Gomi T. Ogawa H. Takata Y. Fujioka M. Takusagawa F. Biochemistry. 1999; 38: 8323-8333Crossref PubMed Scopus (108) Google Scholar, 38Turner M.A. Yuan C.S. Borchardt R.T. Hershfield M.S. Smith G.D. Howell P.L. Nat. Struct. Biol. 1998; 5: 369-376Crossref PubMed Scopus (141) Google Scholar). Each structure represents a different conformation of the AdoHcyase structure. The structure of the substrate-free rWT, which has a large open cleft between the NAD-binding domain and the catalytic domain, represents an open conformation structure of the enzyme (36Hu Y. Komoto J. Huang Y. Gomi T. Ogawa H. Takata Y. Fujioka M. Takusagawa F. Biochemistry. 1999; 38: 8323-8333Crossref PubMed Scopus (108) Google Scholar). The bound NAD is in an oxidized state (i.e. NAD+). The rD244E:Ado* structure contains a trapped Ado intermediate (Ado*, i.e.3-keto-adenosine) in the active site and a reduced NAD (i.e.NADH) (7Komoto J. Huang Y. Gomi T. Ogawa H. Takata Y. Fujioka M. Takusagawa F. J. Biol. Chem. 2000; 275: 32147-32156Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). The cleft between the NAD-binding domain and the catalytic domain is closed (closed conformation). The structure of hWT:ADC contains the oxidized inhibitor (i.e. the inhibitor ADC is oxidized to 2′-hydroxy-3′-keto-cyclopenten-4′-yl-adenine) and NADH (38Turner M.A. Yuan C.S. Borchardt R.T. Hershfield M.S. Smith G.D. Howell P.L. Nat. Struct. Biol. 1998; 5: 369-376Crossref PubMed Scopus (141) Google Scholar). The enzyme has a closed conformation. The rWT:DEA structure in this study has a closed conformation but contains unmodified DEA and NAD+. By comparing the rWT:DEA structure with the other AdoHcyase structures, several unique features of DEA and AdoHcyase have been revealed. The adenine rings of the substrate and the inhibitors bind at the same site with the same interactions. All the nitrogen atoms except for N3 are involved in hydrogen bonding with the protein as either a hydrogen bond acceptor or a hydrogen bond donor. It is noted that 3-deazaadenosine analogues bind to the protein as strongly as Ado analogues (39Chiang P.K. Cantoni G.L. Bader J.P. Shannon W.M. Thomas H.J. Montgomery J.A. Biochem. Biophys. Res. Commun. 1978; 82: 417-423Crossref PubMed Scopus (50) Google Scholar). The hydrogen bonding apparently recognizes the adenine moiety of Ado/AdoHcy. Such an adenine recognition scheme would allow various compounds having adenine rings to bind to the active site of AdoHcyase. On the other hand, other nucleoside analogues such as compounds having guanine, hypoxanthine, and xanthine rings would be limited in their ability to bind to AdoHcyase. The ribose-binding site is composed of relatively hydrophilic amino acid residues. Therefore, the hydrogen-bonding capability of the sugar moiety determines the binding strength of Ado analogues, i.e. if an Ado analogue's sugar moiety is able to make many hydrogen bonds to the protein, then the molecule can bind tightly to the protein. Several different hydrogen bond networks are possible depending on the structure of the sugar moiety. As shown in Fig. 1, the DEA structures in the crystal and in the protein are superimposable except for O3′, C4′, O4a′, and O4b′, but the torsion angles of C2′–C3′ bonds are significantly different (−64° versus 180°). Indeed, the structure of the DEA molecule found in the crystal did not fit into the residual electron density peak in the active site. This observation indicates that the C2′–C3′ bond is twisted by −120° after a sodium ion is released. The carboxyl group of DEA in the crystal structure interacts strongly with a sodium ion. In the protein structure, the carboxyl group hydrogen bonds to the positively charged Lys-185. As illustrated in Scheme FS1, the chiralities of C2′ and C3′ of DEA are different from those of Ado and ADC. Nevertheless all of these compounds bind tightly to AdoHcyase. As shown in Fig.3, the hydrogen-bonding schemes of the sugar moiety are quite different. The 2′-OH of DEA hydrogen bonds exclusively to Asp-189, whereas that of Ado hydrogen bonds to Asp-189 and Glu-155. The 3′-OH of DEA hydrogen bonds solely to Asp-130, whereas that of Ado involves a hydrogen bond to Lys-185. Roughly speaking, O2′, O3′, and O4a′ of DEA are positioned at the O2′, C4′, and O3′ sites of Ado*, respectively. Except for two histidine residues (His-54 and His-300), all hydrophilic amino acid residues on the active site surface are involved in hydrogen bonding with DEA. Interestingly the 3′-CH hydrogen of DEA points to C4 of the bound NAD+ with a distance of C3′···C4 = 3.7 Å. However, the bound DEA is not oxidized by the bound NAD+. It is noted that the C3′···C4 distance in the rD244E:Ado* structure is 3.3 Å. In the rWT:DEA complex, the oxidation reaction does not occur because there is no strong nucleophilic group near the hydrogen of 3′-OH. We have proposed that the hydrogen of 3′-OH of Ado/AdoHcy is abstracted by the neutral Lys-185. When the catalytic domain of the rWT:DEA structure is superimposed on that of the rWT structure (open conformation structure), and the DEA molecule in rWT:DEA is placed in the rWT structure, the introduced DEA molecule can make all the hydrogen bonds to AdoHcyase that are observed in the rWT:DEA structure (Fig. 2 B). This simple modeling indicates that DEA can bind to the active site of the open conformation and can make most of the hydrogen bonds with the protein observed in the rWT:DEA complex. In comparison to rD244E:Ado*, DEA is involved in more hydrogen bonding than Ado* because DEA has more hydrogen bond-forming oxygens than Ado*. Once the catalytic reaction occurs (i.e.Ado is converted to 3′-keto-4′,5′-dehydroadenosine), only the 2′-OH and 3′-O of Ado* can participate in hydrogen bonding with AdoHcyase, whereas all four oxygens (2′-OH, 3′-OH, 4′-OH, and 4′-O) of DEA are involved in hydrogen bonding. Therefore, it appears that DEA binds to AdoHcyase more tightly than the substrate Ado does. When the enzyme closes the cleft between the catalytic domain and the NAD-binding domain, the bound DEA is buried within the protein body and cannot easily leave the active site. The DEA-mediated hydrogen bond networks between the catalytic domain and the NAD-binding domain apparently stabilize the closed structure. In the closed structure, the backbone of His-352 participates in hydrogen bonding with DEA (N6···O (His-352) = 3.3 Å; N7···N (His-352) = 3.0 Å), whereas in the open structure, His-352 is too far away to make hydrogen bonds with DEA (N6···O (His-352) = 5.7 Å; N7···N (His-352) = 5.2 Å). Similarly, 3′-CH hydrogen of DEA points to C4 of NAD+ and has an interaction with it (C3′···C4 = 3.7 Å) in the closed structure, whereas in the open structure 3′-CH hydrogen is too far to have an interaction with C4 of NAD+ (C3′···C4 = 6.8 Å). Therefore, the modeling indicates that the hydrogen bonds between the adenine ring and the backbone of His-352 and the C3′-H···C4 interaction are the major forces (through this lock mechanism) that stabilize the closed conformation of AdoHcyase. DEA is a very potent inhibitor because it can fit into the active site of the intact AdoHcyase (i.e. open conformation), it can make all possible hydrogen bonds with the protein, and it has a lock mechanism to stabilize the closed conformation structure. Once DEA binds to AdoHcyase and stabilizes the closed conformation, the reverse process would be very slow. Therefore, DEA behaves as an inactivator of AdoHcyase, but DEA is a reversible inhibitor because the bound DEA is intact. On the basis of the rWT:DEA structure, we can design specific inhibitors of AdoHcyase. As discussed above, Ado analogues could bind not only to AdoHcyase but also to various nucleotide/nucleoside-binding proteins (such as ATP-, ADP-, and AMP-binding proteins, Ado kinase, and Ado deaminase) and would disrupt their biological functions. Therefore, it is important to modify Ado analogues to bind solely to AdoHcyase and not to other nucleotide/nucleoside-binding proteins. There are three pieces of useful structural information. First, in the rWT:DEA, rD244E:Ado*, and hWT:ADC structures, N3 of the adenine ring is not involved in any hydrogen bonding, and there is a relatively large space in front of N3 that is large enough to put an –H or –O. Replacing the adenine ring with a 3-deazaadenine ring or adenine-3-oxide ring would severely limit the analogue's ability to bind to some nucleotide/nucleoside-binding proteins, i.e.3-deazaadenosine and adenosine-3-oxide analogues would have higher inhibitory selectivity than unmodified Ado analogues. Secondly, the 2′-OH and 3′-OH of DEA participate in hydrogen bonding with the negatively charged carboxyl groups of Asp-189 and Asp-130, respectively. Substituting the hydroxyl groups with positively charged amino groups would increase the binding affinity due to a charge-charge interaction. However, it is noted that introducing a charge-carrying amino group would reduce its membrane diffusion rate. Thirdly, there is a large open space on the tip of the carboxyl group in the rWT:DEA structure. Because a carboxylic acid ester would diffuse into cells faster than the carboxylic acid itself, formation of an ester with short chain monohydroxyl alcohols would improve the inhibitory activityin vivo. By considering these three factors, certain compounds are predicted to be potent and selective reversible inhibitors of AdoHcyase; these compounds are displayed below in Scheme 2. A*A*A*‖‖‖A*=3­deazaadenine or adenine­3­oxideCH2CH2CH2‖‖‖X­CH2X­CHX­CHX=−OH;−NH2‖‖Y=−OH;−NH2Y­CH2Y­CH‖R=−H;−CH3;−C2H5COOR SCHEME2Synthesis and characterization of the above compounds are underway in our laboratory. Schanche et al. (31Schanche J.-S. Schanche T. Ueland P.M. Holy A. Votruba I. Mol. Pharmcol. 1984; 26: 553-558PubMed Google Scholar) have characterized the AdoHcyase inhibitory activity of acyclic Ado analogues (DEA, l-eritadenine,l-threoeritadenine, and 9-(S)-(2,3-dihydroxypropyl)adenine). They have found that DEA is a potent inhibitor, whereas the others are moderate inhibitors. The acyclic Ado inhibitors were placed in the active site by superimposing their adenine rings on that of DEA and changing the backbone conformational angles of their acyclic sugar to maximize the number of hydrogen bonds with the protein. The moderate inhibitors can form fewer hydrogen bonds than can DEA, suggesting that the observed inhibitory activities by Schanche et al. (31Schanche J.-S. Schanche T. Ueland P.M. Holy A. Votruba I. Mol. Pharmcol. 1984; 26: 553-558PubMed Google Scholar) are quite consistent with the rWT:DEA structure. Okumura et al. (40Okumura K. Matsumoto K. Fukamizu M. Yasuo H. Taguchi Y. Sugihara Y. Inoue I. Seto M. Sato Y. Takamura N. Kanno T. Kawazu M. Mizuguchi T. Saito S. Takashima K. Takeyama S. J. Med. Chem. 1974; 17: 846-855Crossref PubMed Scopus (17) Google Scholar) have synthesized more than 100 derivatives of DEA because DEA has a significant hypocholesterolemic activity (40Okumura K. Matsumoto K. Fukamizu M. Yasuo H. Taguchi Y. Sugihara Y. Inoue I. Seto M. Sato Y. Takamura N. Kanno T. Kawazu M. Mizuguchi T. Saito S. Takashima K. Takeyama S. J. Med. Chem. 1974; 17: 846-855Crossref PubMed Scopus (17) Google Scholar). They have found that the carboxylic acid ester analogues with short chain monohydroxyl alcohols are 50 times more active than DEA and effective in lowering serum cholesterol of rats at the dose of 0.0001% in the diet. The intact adenine, the carboxyl functional group, and at least one hydroxyl group are essential for hypocholesterolemic activity. Although no direct correlation has been reported between the hypocholesterolemic activity and the AdoHcyase inhibitory activity of DEA, the findings by Okumura et al.(40Okumura K. Matsumoto K. Fukamizu M. Yasuo H. Taguchi Y. Sugihara Y. Inoue I. Seto M. Sato Y. Takamura N. Kanno T. Kawazu M. Mizuguchi T. Saito S. Takashima K. Takeyama S. J. Med. Chem. 1974; 17: 846-855Crossref PubMed Scopus (17) Google Scholar) are consistent with the DEA binding mode to AdoHcyase found in this study. It is necessary to accumulate more data to conclude whether it is the AdoHcyase inhibitory activity of DEA that lowers the serum cholesterol. We thank Professor Richard H. Himes for critical reading of this manuscript and very valuable comments and Tanabe Research Laboratory for providing DEA." @default.
• W2020779013 created "2016-06-24" @default.
• W2020779013 creator A5011643007 @default.
• W2020779013 creator A5024159668 @default.
• W2020779013 creator A5024715249 @default.
• W2020779013 creator A5027596703 @default.
• W2020779013 creator A5062145659 @default.
• W2020779013 creator A5064890753 @default.
• W2020779013 creator A5085606147 @default.
• W2020779013 creator A5089173245 @default.
• W2020779013 date "2002-03-01" @default.
• W2020779013 modified "2023-10-16" @default.
• W2020779013 title "Inhibition of S-Adenosylhomocysteine Hydrolase by Acyclic Sugar Adenosine Analogue d-Eritadenine" @default.
• W2020779013 cites W1505130887 @default.
• W2020779013 cites W1520194152 @default.
• W2020779013 cites W1531774225 @default.
• W2020779013 cites W1532382821 @default.
• W2020779013 cites W1536187233 @default.
• W2020779013 cites W1539796472 @default.
• W2020779013 cites W1541052928 @default.
• W2020779013 cites W1605480781 @default.
• W2020779013 cites W1852004395 @default.
• W2020779013 cites W1985874627 @default.
• W2020779013 cites W1986191025 @default.
• W2020779013 cites W1988404624 @default.
• W2020779013 cites W1992837175 @default.
• W2020779013 cites W2006969273 @default.
• W2020779013 cites W2007744928 @default.
• W2020779013 cites W2023804937 @default.
• W2020779013 cites W2027752643 @default.
• W2020779013 cites W2028214561 @default.
• W2020779013 cites W2028243355 @default.
• W2020779013 cites W2031308714 @default.
• W2020779013 cites W2033172930 @default.
• W2020779013 cites W2038675286 @default.
• W2020779013 cites W2056897523 @default.
• W2020779013 cites W2071677633 @default.
• W2020779013 cites W2084283372 @default.
• W2020779013 cites W2084763593 @default.
• W2020779013 cites W2084789101 @default.
• W2020779013 cites W2093316814 @default.
• W2020779013 cites W2094109442 @default.
• W2020779013 cites W2105925046 @default.
• W2020779013 cites W2145929111 @default.
• W2020779013 cites W2313801590 @default.
• W2020779013 cites W2317172136 @default.
• W2020779013 cites W2949910565 @default.
• W2020779013 cites W964490185 @default.
• W2020779013 doi "https://doi.org/10.1074/jbc.m109187200" @default.
• W2020779013 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/11741948" @default.
• W2020779013 hasPublicationYear "2002" @default.
• W2020779013 type Work @default.
• W2020779013 sameAs 2020779013 @default.
• W2020779013 citedByCount "37" @default.
• W2020779013 countsByYear W20207790132012 @default.
• W2020779013 countsByYear W20207790132013 @default.
• W2020779013 countsByYear W20207790132014 @default.
• W2020779013 countsByYear W20207790132015 @default.
• W2020779013 countsByYear W20207790132016 @default.
• W2020779013 countsByYear W20207790132017 @default.
• W2020779013 countsByYear W20207790132019 @default.
• W2020779013 countsByYear W20207790132020 @default.
• W2020779013 crossrefType "journal-article" @default.
• W2020779013 hasAuthorship W2020779013A5011643007 @default.
• W2020779013 hasAuthorship W2020779013A5024159668 @default.
• W2020779013 hasAuthorship W2020779013A5024715249 @default.
• W2020779013 hasAuthorship W2020779013A5027596703 @default.
• W2020779013 hasAuthorship W2020779013A5062145659 @default.
• W2020779013 hasAuthorship W2020779013A5064890753 @default.
• W2020779013 hasAuthorship W2020779013A5085606147 @default.
• W2020779013 hasAuthorship W2020779013A5089173245 @default.
• W2020779013 hasBestOaLocation W20207790131 @default.
• W2020779013 hasConcept C181199279 @default.
• W2020779013 hasConcept C185592680 @default.
• W2020779013 hasConcept C2776991684 @default.
• W2020779013 hasConcept C2777108408 @default.
• W2020779013 hasConcept C2780120296 @default.
• W2020779013 hasConcept C55493867 @default.
• W2020779013 hasConcept C71240020 @default.
• W2020779013 hasConceptScore W2020779013C181199279 @default.
• W2020779013 hasConceptScore W2020779013C185592680 @default.
• W2020779013 hasConceptScore W2020779013C2776991684 @default.
• W2020779013 hasConceptScore W2020779013C2777108408 @default.
• W2020779013 hasConceptScore W2020779013C2780120296 @default.
• W2020779013 hasConceptScore W2020779013C55493867 @default.
• W2020779013 hasConceptScore W2020779013C71240020 @default.
• W2020779013 hasIssue "9" @default.
• W2020779013 hasLocation W20207790131 @default.
• W2020779013 hasOpenAccess W2020779013 @default.
• W2020779013 hasPrimaryLocation W20207790131 @default.
• W2020779013 hasRelatedWork W1498415432 @default.
• W2020779013 hasRelatedWork W1753865293 @default.
• W2020779013 hasRelatedWork W1997666966 @default.
• W2020779013 hasRelatedWork W2016119250 @default.
• W2020779013 hasRelatedWork W2056897523 @default.
• W2020779013 hasRelatedWork W2069001564 @default.
• W2020779013 hasRelatedWork W2110336949 @default.
• W2020779013 hasRelatedWork W2398205385 @default.

• next