Abstract
Aldehyde oxidase 1 (AOX1) is a major member of the xanthine oxidase family belonging to the class of complex molybdo-flavoenzymes and plays an important role in the nucleophilic oxidation of N-heterocyclic aromatic compounds and various aldehydes. The enzyme has been well known to show remarkable species differences. Comparing the rabbit and monkey enzymes, the former showed extremely high activity toward cinchonidine and methotrexate, but the latter exhibited only marginal activities. In contrast, monkey had several times greater activity than did rabbit toward zonisamide and (+)-4-(4-cyanoanilino)-5,6-dihydro-7-hydroxy-7H-cyclopenta[d]-pyrimidine [(S)-RS-8359]. In this report, we tried to confer high cinchonidine oxidation activity comparable with that of rabbit AOX1 to monkey AOX1. The chimera proteins prepared by restriction enzyme digestion and recombination methods between monkey and rabbit AOX1s indicated that the sequences from Asn993 to Ala1088 of rabbit AOX1 are essential for the activity. The kinetic parameters were then measured using monkey AOX1 mutants prepared by site-directed mutagenesis. The monkey V1085A mutant acquired the high cinchonidine oxidation activity. Inversely, the reciprocal rabbit A1081V mutant lost the activity entirely: amino acid 1081 of rabbit AOX1 corresponding to amino acid 1085 of monkey AOX1. Thus, cinchonidine oxidation activity was drastically changed by mutation of a single residue in AOX1. However, this might be true for bulky substrates such as cinchonidine but not for small substrates. The mechanism of substrate-dependent species differences in AOX1 activity toward bulky substrates is discussed.
Aldehyde oxidase (AO, EC 1.2.3.1) is a major member of the xanthine oxidase family belonging to the molybdo-flavoenzymes (MFEs) together with xanthine oxidoreductase (XOR; xanthine dehydrogenase form, EC 1.1.1.204; xanthine oxidase form, EC 1.1.3.22) (Beedham, 1985, 1987, 1998, 2002; Kitamura et al., 2006; Garattini et al., 2008; Schumann et al., 2009). Aldehyde oxidase 1 (AOX1) consists of a homodimer with a subunit molecular mass of approximately 150 kDa. Each subunit is made up of a 20-kDa N-terminal domain containing two different 2Fe-2S clusters in which reducing equivalents necessary for catalysis are stored, a 40-kDa central domain containing a flavin adenine dinucleotide (FAD), and an 85-kDa C-terminal domain containing a molybdenum cofactor (MoCo) in which a substrate binding site is located (Garattini et al., 2008). Substrates are oxidized via Mo-OH by base-assisted hydroxylation at the MoCo center, the Mo being reduced from Mo(VI) to the MO(IV) state. MO(IV) is reoxidized via rapid one-electron transfer to the Fe-S cluster and further intramolecular electron transfer to FAD. FADH2 is finally reoxidized by molecular oxygen to produce a superoxide anion and hydrogen peroxide via either a one-electron or a two-electron transfer. The enzyme catalyzes the nucleophilic, but not electrophilic, oxidation of a wide range of endogenous and exogenous aldehydes and N-heterocyclic aromatic compounds. In addition, the enzyme can also catalyze the reductive-ring cleavage metabolism of the atypical antipsychotic drug ziprasidone in humans (Prakash et al., 1997; Beedham et al., 2003).
Being fundamentally different from XOR, AOX1-catalyzed drug metabolism results in marked species differences, which have been well documented for methotrexate (Jordan et al., 1999; Kitamura et al., 1999), famciclovir (Rashidi et al., 1997), cinchona alkaloids (Beedham et al., 1992; Itoh et al., 2006), zonisamide (Kitamura et al., 2001), and a monoamine oxidase A inhibitor RS-8359 (Takasaki et al., 1999, 2005; Itoh et al., 2005, 2006). As a rule, AOX1 activity is high in monkeys and humans, moderate to low in rabbits, rats, and mice, and deficient in dogs. This might be because of the amount of enzyme present in the liver. However, this order happens to change depending on the substrate structure used and might be difficult to explain based on the quantitative difference of the enzyme. For example, when monkey and rabbit are compared, the following considerably complicated situation seems to arise. Approximately 2 orders of magnitude higher methotrexate 7-hydroxylase activity was shown in rabbit compared with that of monkey (Kitamura et al., 1999). Likewise, rabbit exhibited 2 to 3 orders of magnitude higher activity toward cinchonidine 2′-oxidation than did that of monkey (Itoh et al., 2006). In contrast, roughly 1 order of magnitude higher AOX1-catalyzed zonisamide reduction activity was observed in monkey compared with that of rabbit (Kitamura et al., 2001). Monkey had more than several times greater (S)-RS-8359 2-oxidation activity than did rabbit (Itoh et al., 2006). Thus, the two species show remarkably different catalytic properties from each other depending on the substrate structures used as already documented by Beedham (2002), Kitamura et al. (2006), and Garattini et al. (2008). The reversal of order in AOX1 activities depending on substrate structure might be primarily the result of structural differences in the active sites. MFEs have a large molecular mass with the active site buried inside (approximately 10–15 Å from the surface), which substrates can access through a funnel-shaped cavity (Garattini et al., 2003). The tunnel is wider at the surface (15–20 Å in diameter) and narrower at mid-height in close proximity to the Mo atom (2 Å in diameter). Substrate specificity and inhibition studies suggest that the tunnel may vary even in different AO isozymes (Garattini et al., 2008). Accordingly, it is expected that some amino acid in the tunnel might be responsible for the reversed species difference seen between monkey and rabbit.
Enzyme manipulation using recombinant or site-directed mutagenesis (SDM) techniques could be an excellent way of characterizing important properties of targeted enzymes. In the present study, we constructed mutant AOX1 cDNAs by substituting nucleotides of monkey AOX1 with relevant ones of rabbit AOX1 and prepared AOX1 mutant proteins by expressing them in Escherichia coli. Analyses of their catalytic characteristics revealed that a single amino acid substitution of V1085A in monkey AOX1 resulted in a drastic change of substrate specificity toward cinchonidine. Inversely, the rabbit AOX1 A1081V mutant completely lost its high cinchonidine oxidation activity. The mechanism of the interesting substrate-dependent species differences will be discussed in detail.
Materials and Methods
Chemicals and Reagents.
(S)-RS-8359 and the 2-keto metabolite were supplied by Ube Kosan (Yamaguchi, Japan). Acetanilide, an internal standard of high-performance liquid chromatography (HPLC) analysis, and kanamycin sulfate were purchased from Sigma-Aldrich (St. Louis, MO). Cinchonidine sulfate was obtained from Tokyo Chemical Industry (Tokyo, Japan). Ampicillin, imidazole, sodium molybdic acid, and isopropyl-β-d-(−)-thiogalactopyranoside were purchased from Wako Pure Chemical Industries (Tokyo, Japan). All the other reagents were of reagent grade. The cinchonidine 2′-oxide derivative was prepared according to the method previously reported (Itoh et al., 2006). The chemical structures of (S)-RS-8359 and cinchonidine are shown in Fig. 1.
Chemical structures of (S)-RS-8359 and cinchonidine. Arrows indicate the oxidation sites. The volume of substrate molecule was computed using the Hartree-Fock/6-31G* basis set implemented in Spartan'06 software (Wavefunction Inc.).
Enzyme Activity Assay.
The oxidation activities of cinchonidine and (S)-RS-8359 were measured according to our previous methods (Itoh et al., 2006). In brief, (S)-RS-8359 (2.5–400 μM) or cinchonidine (2.0–200 μM) was incubated at 37°C for 30 min with enzyme solution (0.10 ml) in a reaction mixture (0.25 ml) consisting of 80 mM phosphate buffer, pH 7.4. Similar to those by liver cytosol, the reactions were linear for all the preparations at least until 120 min. The reaction was stopped by the addition of methanol (0.25 ml) containing 0.2 mg/ml hydrocortisone for the metabolite of RS-8359 and 0.2 μg/ml acetanilide for the metabolite of cinchonidine as an internal standard. The mixture was then centrifuged at 5000g for 3 min. Aliquots (25 μl) of the supernatant were analyzed for quantification of the respective oxidation product by reverse-phase HPLC on an YMC Pak Pro C18 AS-302 column (6.0 mm i.d. × 150 mm; YMC Co., Ltd., Kyoto, Japan). The mobile phase was composed of acetonitrile/0.5% ammonium acetate (14:86) for the 2-keto metabolite of RS-8359 and acetonitrile/2.5% acetic acid for the 2-keto metabolite of cinchonidine; the flow rate was 1.0 ml/min. The HPLC instrument consisted of a Shimadzu model 6A HPLC System (Shimadzu Seisakusho Ltd., Kyoto, Japan). The peaks were monitored for absorbance at 315 nm for the RS-8359 metabolite and at 235 nm for the cinchonidine metabolite. The retention times were 7.5 min for the 2-keto metabolite of the (S)-RS-8359 and 10.0 min for the oxidation product of cinchonidine.
Construction of Expression Plasmids.
The expression experiments were basically conducted according to the methods described by Huang et al. (1999) and Hoshino et al. (2007). Chimeric cDNAs were constructed by the ligation of monkey AOX1 cDNA fragments corresponding with those of rabbit AOX1 generated using the restriction enzymes SpeI, BspHI, BglII, BsrGI, and MluI. A schematic diagram of the constructs is shown in Fig. 2. Approximately half was exchanged in chimeras 1 and 2, one-sixth in chimeras 3, 4, and 5, one-fifteenth in chimeras 6 and 8, and one-thirtieth in chimera 7. For the preparation of the last three chimeras, the restriction sites for BsrGI and MluI were newly introduced by SDM, and the accompanying amino acid substitutions were corrected by SDM after ligation. The SDM was conducted with a QuikChange SDM Kit (Stratagene, La Jolla, CA) according to the manufacturer's specifications. The sets of forward and reverse primers used are listed in Table 1. In addition, 12 monkey and 1 rabbit mutant AOX1 cDNA were prepared using the synthetic oligonucleotides listed in Table 2 as the mutagenic primers. The resulting chimeric and mutant constructs were verified by complete sequence analysis. They were then subcloned in the expression vector pQE-30Xa (QIAGEN GmbH, Hilden, Germany) and used to transform competent E. coli M15 cells (QIAGEN GmbH).
Schematic models of amino acid sequences of wild and chimeric AOX1 proteins designed for the study of structural analyses of active sites of AOX1. The black and gray columns represent the amino acid sequences of monkey and rabbit AOX1, respectively. Chimeric cDNAs were constructed by digestion and recombination between monkey and rabbit AOX1 cDNAs, and chimeric AOX1 proteins were expressed in E. coli, which were used for catalytic activity measurement after purification on a nickel column.
Primer sets used for introduction of restriction enzyme sites by mutagenesis: introduction of restriction site accompanied nucleotide substitutions that were corrected
Primer sets used for preparation of monkey and rabbit AOX1 mutant cDNA
Volume of Substrate Molecule.
For predicting the bulkiness of the substrate molecules, the volume of the substrate molecule was computed using the Hartree-Fock/6-31G* basis set implemented in Spartan'06 software (Wavefunction Inc., Tokyo, Japan).
Statistical Analysis.
All the expression experiments were conducted three times, and data are presented as the mean ± S.D. The kinetic parameters were determined by nonlinear regression analysis using SigmaPlot 11.0 software (Systat Software, Inc., San Jose, CA). Significant difference was evaluated by Student's t test using SPSS 11.5 software (SPSS Inc., Chicago, IL); a p value of <0.05 was considered statistically significant.
Results
Expression of Chimera and Mutant Proteins of Monkey AOX1.
In this study, the chimera proteins refer to those derived from the cDNA generated by restriction enzyme digestion and recombination methods and the mutant proteins to those generated by SDM techniques. The expressed proteins in E. coli were purified by Ni column chromatography. Several hundred micrograms of each purified enzyme were yielded throughout all the expression experiments. Each preparation retained substantial (S)-RS-8359 oxidation activities, as mentioned below. The results indicated a certain expression of chimera and mutant proteins of AOX1s in E. coli.
Kinetic Parameters for Cinchonidine and (S)-RS-8359 by Monkey and Rabbit Wild-Type AOX1.
The oxidation activities of cinchonidine and (S)-RS-8359 were measured using monkey and rabbit wild-type AOX1 proteins expressed in E. coli. The kinetic parameters analyzed by Eadie-Hofstee plots are shown in Table 3. Rabbit AOX1 had an extremely high Vmax value toward cinchonidine, whereas no substantial activity was detected in monkey enzyme. When (S)-RS-8359 served as a substrate, the Vmax value observed in monkey AOX1 was approximately twice that in rabbit AOX1. Because the kinetic parameters were calculated by the use of purified enzymes, the results were considered to suggest that the remarkable species differences in cinchonidine oxidation activity were essentially the result of qualitative differences in the structure of the active site of AOX1 rather than quantitative differences in the enzyme that is generally thought to be a major mechanism of species differences. Thus, the AOX1 chimera proteins between monkey and rabbit are considered useful for the study of the structural analysis of the active site.
Kinetic parameters for oxidation activities of cinchonidine and (S)-RS-8359 by monkey and rabbit AOX1 chimera proteins
Several areas of the latter C-terminal half of monkey AOX1 were replaced by those of rabbit AOX1 with restriction enzyme treatment and recombination. Each value is the mean ± S.D. for three separate experiments independently expressed in E. coli.
Kinetic Parameters for Cinchonidine and (S)-RS-8359 by Monkey AOX1 Chimera Proteins.
The chimera proteins were prepared by exchanging the sequences of monkey AOX1 with those corresponding to rabbit AOX1, as shown in Fig. 2, and their kinetic parameters were determined (Table 3). Chimera 1, in which the N-terminal half of monkey AOX1 was substituted with that of rabbit AOX1, had no activity. In contrast, chimera 2, in which the C-terminal half of monkey AOX1 was replaced with that of rabbit AOX1, showed a high degree of activity similar to that of wild-type rabbit AOX1. To narrow down the regions responsible for the catalytic activity toward cinchonidine, three chimera proteins were produced by substituting approximately one-third of the C-terminal half of monkey AOX1 with those of rabbit AOX1. The substituted amino acid sequences were from Thr616 to Phe859 in chimera 3, from Phe859 to Ala1092 in chimera 4, and from Ala1092 to the C terminus in chimera 5 (Fig. 2). As is shown in Table 3, only chimera 4 was found to exhibit exceedingly high activity, whereas the other two chimera proteins showed no activity. Furthermore, the amino acid areas from Phe859 to Ala1092 of monkey AOX1 were divided into three parts to which those of rabbit AOX1 were introduced. The substituted amino acid sequences were from Phe859 to Met955 in chimera 6, from Met955 to Asn997 in chimera 7, and from Asn997 to Ala1092 in chimera 8. Only chimera protein 8 exhibited remarkably high cinchonidine oxidation activity; the other two did not (Table 3).
The kinetic parameters for the oxidation activities of (S)-RS-8359 are also listed in Table 3. All the chimera proteins were found to have substantial (S)-RS-8359 oxidation activities, which was clearly different from the results for cinchonidine. The chimera proteins possessing cinchonidine oxidation activity had a tendency to show larger Km values for (S)-RS-8359 than did those lacking the activity.
Kinetic Parameters for Cinchonidine and (S)-RS-8359 by Monkey Mutant AOX1.
The amino acid areas from Asn997 to Ala1092 of rabbit AOX1 were revealed to be of primary importance for exertion of its catalytic activity toward cinchonidine. Within the areas, 11 amino acid residues were different from monkey AOX1, and four of them were different from those of human and rat AOX1s as well (Fig. 3) (Wright et al., 1993, 1999; Calzi et al., 1995; Huang et al., 1999). Accordingly, the following four sets of mutant proteins were prepared by SDM as being the most likely ones to endow monkey AOX1 with high cinchonidine oxidation activity: mutant 1 with the K1004Q mutation, mutant 2 with the M1009I mutation, mutant 3 with the A1023T mutation, and mutant 4 with the I1032V mutation (Table 2). Unexpectedly, those mutants did not have any detectable activity toward cinchonidine, although their (S)-RS-8359 oxidation activities were comparable with those of wild-type monkey AOX1 (Table 4).
Alignments of amino acid sequences of AOX1 of rabbit, monkey, human, and rat. Eleven amino acid residues from Asn993 to Ala1088 of rabbit AOX1 are different from the corresponding areas of other animals, which are displayed in boxes with the amino acid number.
Kinetic parameters for oxidation activities of cinchonidine and (S)-RS-8359 by mutant proteins of monkey AOX1
Eleven amino acids in the range of Ala997 to Ala1092 of monkey AOX1 were substituted with those of rabbit AOX1 with SDM. The number of substituted amino acids was single in mutants 1 to 4 and plural in mutants 5 to 10. Value is the mean ± S.D. of two experiments.
Nevertheless, the amino acid sequences from Asn997 to Ala1092 of rabbit AOX1 should certainly contain the key amino acid(s) responsible for the superb substrate specificity toward cinchonidine. Therefore, the mutant proteins were prepared by continuous substitutions of every two amino acids from the N-terminal end of the crucial 11 amino acids (Table 2). Mutants 5 to 9 substituted from Lys1004 to Ile1067 did not express any cinchonidine oxidation activity, but mutant 10 with both A1083T and V1085A mutations showed high activity (Table 4), suggesting that the solution questioned here ultimately narrowed down to the two amino acid residues of Ala1083 and Val1085. To learn which one was the determinant or whether both were needed, the two mutants were produced as follows: mutant 11 with the A1083T mutation and mutant 12 with the V1085A mutation. Mutant 11 revealed a similar extent of high cinchonidine oxidation activity as that of wild-type rabbit AOX1; there was no detectable activity in mutant 12 (Table 5). On the contrary, (S)-RS-8359 oxidation activity was confirmed in each mutant protein, and a larger Km value was seen for mutant 11 retaining high cinchonidine oxidation activity than that of mutant 12, which lacked such activity. The results clearly indicated that a single substitution of valine with alanine at position 1085 of monkey AOX1 can dramatically change monkey AOX1 to rabbit-type AOX1 as to substrate specificity toward cinchonidine.
Kinetic parameters for oxidation activity of cinchonidine and (S)-RS-8359 by mutant proteins of monkey and rabbit AOX1
Kinetic Parameters for Cinchonidine and (S)-RS-8359 by Rabbit Mutant AOX1 A1081V.
It was supposed that the reciprocal mutation of rabbit AOX1 by substituting Ala1081 with valine converts it to an enzyme that resembles monkey AOX1 as far as cinchonidine oxidation activity is concerned. Table 5 shows that mutant 13 with mutation A1081V completely lost the high cinchonidine oxidation activity.
Discussion
As a rule, AOX1 activity is higher in monkeys and humans, medium to low in rabbits, guinea pigs, rats, and mice, and deficient in dogs. However, rabbit shows approximately 2 orders of magnitude greater oxidation activity toward cinchonidine and methotrexate than do the other species. The species differences in the substrates might be mostly caused by differences in the structures of the active site of AOX1 and not by differences in the amount of enzyme present in the liver. To verify this hypothesis in the current report, we aimed to change the substrate specificity toward cinchonidine of monkey AOX1 to that of rabbit type AOX1 using genetic engineering methods. Two hundred twenty-one amino acids of monkey and rabbit AOX1s are different from each other. Therefore, the chimera proteins were first prepared to focus on the region of interest by the method of restriction enzyme digestion and recombination followed by preparation of mutant proteins by SDM to isolate the amino acid responsible for the phenomenon.
The results of measuring cinchonidine oxidation activity by the monkey chimera AOX1 proteins clearly displayed that the range from Asn993 to Ala1088 of rabbit AOX1, which corresponds to Asn997 to Ala1092 of monkey AOX1 because of the shortage of four amino acids of rabbit AOX1 in the hinge between the FAD and MoCo domains (Huang et al., 1999; Hoshino et al., 2007), is central to the exertion of cinchonidine oxidation activity. Among the 95 amino acid sequences present in the region, 11 are different from each other in monkey and rabbit AOX1s. The one that is essential for cinchonidine oxidation activity was determined using monkey mutant AOX1 proteins. Consequently, substituting valine at position 1085 of monkey AOX1 with alanine was shown to dramatically endow monkey AOX1 with exceedingly high cinchonidine oxidation activity. That means that Ala1081 of rabbit AOX1 is essential for the oxidation activity. In fact, cinchonidine oxidation activity was no longer confirmed in the reciprocal rabbit mutant A1081V protein. The results strongly support the special significance of alanine at the relevant position of AOX1 to exert its catalytic enzyme function toward cinchonidine.
XOR is a major member of MFEs along with AOX1. According to the three-dimensional structure of rabbit AOX1 predicted by SWISS-MODEL analysis (Guex and Peitsch, 1997; Arnold et al., 2006) using Rhodobacter capsulatus xanthine dehydrogenase (Truglio et al., 2002) as a template, Ala1081 is likely to be positioned in a tunnel leading into an active site buried approximately 10 to 15 Å from the surface. The tunnel is a funnel-shaped cavity dominated by the presence of hydrophobic residues able to accommodate the ring structures of substrates. In bovine XOR, Leu648, Phe649, Leu873, Phe914, Phe1009, Val1011, Phe1013, and Leu1014 are important for the stabilization of substrates in the channel (Garattini et al., 2003). The structural difference in alanine and valine is simply derived from the distinction between the methyl and isopropyl groups that may not function to form the hydrogen bonds or hydrophobic interactions necessary for fitting substrates to the enzyme. The bulkiness of the substrate molecules was predicted from its volume, which was computed using the Hartree-Fock/6-31G* basis set implemented in Spartan'06 software (Wavefunction Inc.). The volumes of cinchonidine and (S)-RS-8359 were 316 and 251 Å3, respectively (Fig. 1). Alanine might be small enough so that a bulky cinchonidine substrate could approach and bind to the active site. In contrast, the isopropyl group of valine might be thought to interfere with fitting better into the active site. A similar example has recently been reported for aldehyde dehydrogenase 3 by Ho et al. (2008). They found that removing the methyl group of threonine in the T186S mutant of aldehyde dehydrogenase 3 enables the bulky aldophosphamide to bind better and to be a poorer enzyme when small substrates such as benzaldehyde derivatives were used. In rabbit AOX1, the small size of Ala1081 plays an important role in determining a suitable shape for bulky substrates, and the substitution of valine with alanine in monkey AOX1 dramatically improved the oxidation activity toward bulky cinchonidine.
Compared with cinchonidine, when a little smaller and more planar molecule (S)-RS-8359 was used as a substrate, the kinetic parameters were slightly complicated. The monkey mutant V1085A showed a similar large Vmax value as that of the wild-type monkey protein, but the Km value increased from 12.3 ± 1.94 to 149 ± 10.1 μM, which is almost equal to that of the wild-type rabbit AOX1 (149 ± 10.1 μM). The results suggest that Val1085 of monkey AOX1 might be related to the stabilization of (S)-RS-8359 in the tunnel leading into the active site. The (S)-RS-8359 oxidation activity is likely to increase with the increasing content of the amino acid sequences of monkey AOX1 in the chimera proteins, suggesting that monkey AOX1 may possess some desirable factors influencing catalytic activity, such as a more efficient electron transfer system compared with that of rabbit. Although the precise reasons are still unknown, it is obvious that the sizes of the amino acid at position 1085 of monkey AOX1 and 1081 of rabbit AOX1 play important roles in the determination of species differences toward bulky substrates. Humans have generally high AOX1 activity along with monkeys, but this might not be true with regard to bulky substrates as expected from the same kind of amino acid at position 1085 of monkeys and indeed confirmed by the expressed enzyme (data not shown). Thus, it can be said that substrate specificity of human AOX1 is highly similar to that of monkey. However, a wide range of substrates, including small molecules such as benzaldehyde and vanillin in addition to (S)-RS-8359 and large molecules such as methotrexate and retinal in addition to cinchonidine, should be investigated to clarify the role of the amino acid at the relevant position. Results for those substrates might be useful for a better understanding of the structure of the active site of AOX1. Such investigation will be an interesting subject for future study.
In conclusion, the monkey mutant V1085A AOX1 protein provided high cinchonidine oxidation activity, whereas the reciprocal rabbit mutant A1081V AOX1 protein completely lost the activity. The results suggest that the size of the amino acid at the relevant positions is critical for oxidation toward bulky substrates and that this is the reason for species differences toward those substrates.
Footnotes
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
doi:10.1124/dmd.109.030064.
-
- AO
- aldehyde oxidase
- MFE
- molybdo-flavoenzyme
- XOR
- xanthine oxidoreductase
- AOX1
- aldehyde oxidase 1
- FAD
- flavin adenine dinucleotide
- MoCo
- molybdenum cofactor
- RS-8359
- 4-(4-cyanoanilino)-5,6-dihydro-7-hydroxy-7H-cyclopenta[d]pyrimidine
- (S)-RS-8359
- (+)-4-(4-cyanoanilino)-5,6-dihydro-7-hydroxy-7H-cyclopenta[d]-pyrimidine
- SDM
- site-directed mutagenesis
- HPLC
- high-performance liquid chromatography.
- Received August 31, 2009.
- Accepted November 9, 2009.
- Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics
