Transient-phase kinetic studies on the nucleotide binding to 3a-hydroxysteroid dehydrogenase from Pseudomonas sp. B-0831 using fluorescence stopped-flow procedures Shigeru Ueda 1 , Masayuki Oda 2 , Shigeyuki Imamura 1 and Masatake Ohnishi 2 1 Department Diagnostics Research and Development, Division of Fine Chemicals and Diagnostics, Asahi Kasei Pharma Corporation, Shizuoka, Japan; 2 Department of Cellular Macromolecule Chemistry, Graduate School of Agriculture, Kyoto Prefectural University, Kyoto, Japan The dual nucleotide cofactor-specific enzyme, 3a-hydroxy- steroid dehydrogenase (3a-HSD) from Pseudomonas sp. B-0831, is a member of the short-chain dehydrogenase/ reductase (SDR) superfamily. Transient-phase kinetic stud- ies using the fluorescence stopped-flow method were con- ducted with 3a-HSD to characterize the nucleotide binding mechanism. The binding of oxidized nucleotides, NAD + , NADP + and nicotinic acid adenine dinucleotide (NAAD + ), agreed well with a one-step mechanism, while that of reduced nucleotide, NADH, showed a two-step mechanism. This difference draws attention to previous characteristic findings on rat liver 3a-HSD, which is a member of the aldo-keto reductase (AKR) superfamily. Although functionally similar, AKRs are structurally dif- ferent from SDRs. The dissociation rate constants associated with the enzyme–nucleotide complex formation were larger than the k cat values for either oxidation or reduction of substrates, indicating that the release of cofac- tors is not rate-limiting overall. It should also be noted that k cat for a substrate, cholic acid, with NADP + was only 6% of that with NAD + , and no catalytic activity was detectable with NAAD + , despite the similar binding affinities of nucleotides. These results suggest that a certain type of nucleotide can modulate nucleotide-binding mode and fur- ther the catalytic function of the enzyme. Keywords: aldo-keto reductase superfamily; nucleotide- binding; fluorescence stopped-flow method; 3a-hydroxyster- oid dehydrogenase; short-chain dehydrogenase/reductase superfamily. The NAD(P) + -dependent enzyme, 3a-hydroxysteroid dehydrogenase (3a-HSD), catalyzes the reversible intercon- version of hydroxy and oxo groups at position 3 of the steroid nucleus, and has been found in many mammalian cells and microorganisms [1,2]. Prokaryotic 3a-HSDs have been described in Eubacterium lentum [3], Clostridium perfringens [4], Pseudomonas putida [5], Comamonas (Pseu- domonas) testosteroni ATCC11996 [6–9], and Pseudomonas sp. B-0831 [10]. Despite similar substrate specificities, eucaryotic and prokaryotic 3a-HSDs belong to two differ- ent protein superfamilies: viz, the eucaryotic and prokary- otic enzymes, respectively, belonging to the aldo-keto reductase (AKR) (EC 1.1.1.213) and the short-chain dehy- drogenase/reductase (SDR) superfamilies (EC 1.1.1.50) [11–14]. Although structurally different, both HSDs are functionally similar; the AKRs are monomeric and have an a/b-barrel fold, whereas the SDRs are dimeric or tetrameric and contain a Rossmann nucleotide-binding fold. We have previously cloned 3a-HSD from Pseudomonas sp. strain B-0831 and expressed it in Escherichia coli [10]. The SDS/PAGE and gel-filtration analyses revealed that this enzyme forms a homodimer comprised of two 25 kDa monomer proteins. The amino acid sequence shares about 50% identity with that of Comamonas (Pseudomonas) testosteroni ATCC11996 [15], and contains two motifs common to SDRs, the Gly-X-X-X-Gly-X-Gly cofactor- binding and Tyr-X-X-X-Lys substrate-binding motifs. Because 3a-HSD from C. testosteroni can catalyze not only the oxidoreduction at position 3 of the steroid nucleus but also carbonyl reduction of a variety of nonsteroidal aldehydes and ketones, the enzyme was named as 3a-hydroxysteroid dehydrogenase/carbonyl reductase (3a-HSD/CR) [15,16]. In a similar manner, P. sp. B-0831-derived 3a-HSD also shows carbonyl reduc- tase activity [17]. Different from the NAD + -dependent 3a-HSD/CR, P. sp. B-0831-derived 3a-HSD has the ability to use not only NAD + but also NADP + .In clinical diagnostics, the enzyme has been used for the measurement of total bile acids in serum [18]. Further- more, a highly sensitive and unique enzyme cycling method for total bile acids assay has been developed Correspondence to M. Oda, Graduate School of Agriculture, Kyoto Prefectural University, 1–5, Shimogamo Nakaragi-cho, Sakyo-ku, Kyoto 606–8522, Japan. Fax: + 81 75 7035673, Tel.: + 81 75 703 5673, E-mail: oda@kpu.ac.jp Abbreviations:3a-HSD, 3a-hydroxysteroid dehydrogenase; AKR, aldo-keto reductase superfamily; CR, carbonyl reductase; NAAD + , nicotinic acid adenine dinucleotide; SDR, short-chain dehydrogenase/ reductase superfamily. Enzyme:3a-hydroxysteroid dehydrogenase (EC 1.1.1.213, EC 1.1.1.50). (Received 3 February 2004, revised 5 March 2004, accepted 15 March 2004) Eur. J. Biochem. 271, 1774–1780 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04089.x using this B-0831 enzyme in the presence of excessive thio-NAD + (NAD + analogue) and NADH, which are commercially available reagents [19]. Rat liver 3a-HSD, the first HSD to be assigned to the AKRs, is involved in bile acid biosynthesis, and participates in the inactivation of circulating steroid hormones such as androgens, progestins and glucocorticoids [20]. Its nucleo- tide cofactor-binding mechanisms have been extensively studied [21–27]. This enzyme shows NADP + preference and follows an ordered bi-bi reaction mechanism with nucleotide binding first. The binding kinetics of NADP(H) by the fluorescence stopped-flow method showed a two-step mechanism which consists of fast formation of a loose complex followed by slow formation of a tightly bound complex [26], and the dissociation of NADP(H) is not rate- limiting. The latter finding discriminates the rat enzyme from other AKRs, such as human and pig muscle aldose reductases [28,29]. As well as rat liver enzyme, 3a-HSD/CR from C. testo- steroni ATCC11996 follows an ordered bi-bi mechanism with pyridine nucleotide binding first and dissociation last [7]. In contrast to 3a-HSDs in AKRs, however, little is known about the nucleotide-binding mechanism of 3a-HSD in SDRs. In terms of nucleotide binding to the enzyme, the nicotinamide ring adopts syn conformation in SDRs while it adopts anti conformation in AKRs, exhibiting the opposite stereospecificity of hydride transfer [27]. Therefore, it is of great interest to elucidate the mechanism of the nucleotide binding to 3a-HSD from P. sp. B-0831 as a member of SDRs. In the present study, we analysed transient-phase kinetics of nucleotide binding to bacteria-derived 3a-HSD by fluorescence stopped-flow measurements for the first time. Transient-phase kinetics can distinguish between the for- mation and decay of individual complexes on either one- or multiple-step reaction pathways. Application of this method to 3a-HSD revealed that the reaction mechanism depends on the type of nucleotides, and is different from that of rat liver 3a-HSD. In addition to the unique features of the kinetic mechanism, nucleotide preference of this enzyme could be elucidated in comparison with the functional and structural information of other SDRs and AKRs, such as C. testosteroni 3a-HSD/CR and rat liver 3a-HSD. Materials and methods Materials NAD + ,NADP + , NAAD + , NADH, and NADPH (Ori- ental Yeast Co., Ltd in Japan), steroids (Sigma Chem. Co.) and other chemicals were of the highest commercial quality available. Recombinant 3a-HSD from Pseudomonas sp. strain B-0831 was expressed in E. coli and purified as described previously [10]. The protein purity was deter- mined to be over 95% by SDS/PAGE analysis, and the protein concentration was determined by the method of Lowry et al. [30]. Steady-state kinetic study Assays were carried out at 37 °Cor15°Cin40m M Tris/ HCl (pH 8.5), in the presence of 1 m M NAD(P) + , 0.025% nitrotetrazolium blue (NTB), 0.4% Triton X-100, and 2.5 units/mL diaphorase (Asahi Kasei) for oxidation reaction, and in the presence of 0.3 m M NAD(P)H for reduction reaction [10]. In the assay for substrate specificity, various substrate concentrations were used for the determination of K m and k cat . To determine K m for the NAD(P) + and NAD(P)H binding, 1 m M cholic acid and 0.8 m M dehydro- cholic acid were used, respectively. The reaction was initiated by adding the enzyme, and terminated by adding 2 mL of 0.5% SDS after incubation for 5 min. The K m and k cat values for oxidation reaction were determined by measuring the formation of formazan dye at 550 nm, and those for reduction reaction were determined by measuring the decrease of NAD(P)H at 340 nm, using a Shimadzu UV-2200 spectrophotometer. In the inhibition study by NAAD + ,1.0m M cholic acid (substrate) was used in the presence of NAD + with concentration ranging from 5.0 l M to 200 l M ,and2.0m M NAAD + as an inhibitor. The K m and K i values were calculated by the nonlinear least squares method using the Taylor expansion [31]. Fluorescence titration measurements The nucleotide binding to 3a-HSD was measured by monitoring the quenching of intrinsic enzyme fluorescence upon incremental addition of nucleotides. Emission spectra (300–500 nm) were recorded on a JASCO Corporation FP-777 fluorescence spectrophotometer at 280-nm excita- tion. All titrations were carried out in 3-mL volumes with 0.49 l M of the enzyme in 50 m M Tris/HCl (pH 8.5) at 15 °C. The total volume change due to the addition of nucleotide was less than 2%. The K d value was determined by the nonlinear least squares method [31]. Stopped-flow kinetic study The time course of fluorescence intensity caused by nucleo- tide binding to 3a-HSD was monitored through cut-off filters (Toshiba Kasei Kogyo), UV-31 with 50% transmittance at 310 nm for oxidized nucleotides and UV-42 with 50% transmittance at 420 nm for reduced nucleotides, equipped with an Otsuka Electronics RA-401 stopped-flow apparatus at 280-nm excitation by a 200 W D 2 lamp at 15 °C [32]. Using a quartz cell (inner diameter: 2 mm), the dead time of the apparatus was determined as 1.3 ms under the experimental conditions described in a previous study [33]. In order to determine the reliable kinetic range in the present experi- ments, we measured the reduction rate of 2,6-dichloro- phenol-indophenol by ascorbate, and confirmed the linearity of the apparent first-order rate constant, k app , within the range of 0–400 s )1 in the Guggenheim plot [34]. Various concentrations of nucleotide solutions and 3a-HSD in 50 m M Tris/HCl (pH 8.5) were introduced into the stopped-flow apparatus by means of a separate syringe. The k app value for each nucleotide concentration was determined in triplicate. Data analysis of rapid reaction The apparent first-order rate constant, k app ,wasdetermined from the Guggenheim plot. When the plots followed the linear dependence of k app against initial concentration of nucleotide, Eqn (1) was adopted: Ó FEBS 2004 Transient kinetics of nucleotide binding to 3a-HSD (Eur. J. Biochem. 271) 1775 E þ Nuc À! k þ1 À k À1 E-Nuc ð1Þ where E is the enzyme, Nuc is the nucleotide, E-Nuc is the enzyme–nucleotide complex, and k +1 , k )1 are the transient- phase kinetic rate constants. On the condition that the initial concentration of nucleotide, [Nuc], is greater than E, k app is thus expressed as Eqn (2) [35]: k app ¼ k þ1 ½Nucþk À1 ð2Þ The values of k +1 and k )1 were determined from the slope and the intercept of the plot of [Nuc] vs. k app , respectively. The dissociation constant of E-Nuc, K )1 , can be calculated from Eqn (3): K À1 ¼ k À1 =k þ1 ð3Þ In contrast, when the plots of [Nuc] vs. k app were nonlinear, Eqn (4) was applied: E þ Nuc À! k þ1 À k À1 E-Nuc À! k þ2 À k À2 E Ã -Nuc ð4Þ where E-Nuc is an intermediate (loosely bound) form and E*-Nuc is a more tightly bound isomerized form, and k +1 , k )1 , k +2 and k )2 are the transient-phase kinetic rate constants. In this mechanism, the reciprocal of the slower relaxation time, k app , is expressed as Eqn (5) when [Nuc] >>[E] [36]. k app ¼ k þ2 [NADH] K À1 þ [NADH] þ k À2 ð5Þ where K )1 (¼ k )1 /k +1 ) is the dissociation constant for the intermediate E-Nuc complex. The k )2 value can be estima- ted by linear extrapolation in the low [Nuc] region of k app vs. [Nuc] plot. The k +2 and K )1 values were obtained with the k )2 value determined above, by the reciprocal of the intercept and the slope/intercept of the secondary plot of 1/(k app ) k )2 ) vs. 1/[Nuc], respectively [35]. In the two-step mechanism, the overall dissociation constant, K d ,forthe E-NuccomplexcanbecalculatedfromEqn(6). K d ¼ K À1 =½1 þðk þ2 =k À2 Þ ð6Þ Results Substrate specificity The K m and k cat values of recombinant 3a-HSD from P. sp. B-0831 toward typical substrates for both forward and reverse reactions were determined (Table 1). There was no difference between native and recombinant enzymes with respect to these kinetic values (data not shown). Compared with the enzyme from C. testosteroni ATCC11996, the K m value for androsterone was 10 times higher (210 to 31.1 l M ), while that for 5a-androstan-3,17- dione in the reverse reaction was of the same magnitude (44 to 42.2 l M ) [15]. For cholic acid as a substrate, the k cat value with NADP + was only 6% of that with NAD + , whereas both of the K m values were similar. In the case of dehydrocholic acid as a substrate, the k cat value with NADPH was 4.2% of that with NADH whereas the K m value with NADPH was of the same order in magnitude as that with NADH. The k cat values at 15 °C were also measured for comparison with the dissociation constants of nucleotide binding obtained by fluorescence stopped-flow procedure, as described below. Nicotinic acid adenine dinucleotide (NAAD + ), a precur- sor of NAD + , does not work as cofactor in the reaction [37]. However, the oxidation reaction of cholic acid in the presence of NAD + was inhibited by NAAD + with 373 l M of the inhibitor constant (K i ) in a competitive manner, indicating that NAAD + also binds to the enzyme. The NAAD + binding to the enzyme, together with the binding of other nucleotides, was also investigated by both fluores- cence titration and stopped-flow measurements, as des- cribed below. Fluorescence titration with nucleotides P. sp. B-0831-derived 3a-HSD irradiated an intrinsic fluor- escence emission spectrum of 336 nm k max at 280 nm excitation. The incremental addition of NAD + and NADH quenched the fluorescence emission signal (Fig. 1). Plots of the degree of decrease in fluorescence intensity against NAD(P) + ,NAAD + , and NAD(P)H examined here were fitted to a saturation absorption isotherm, yielding the K d value by the nonlinear least squares method. The K d values for NAD + and NADP + approximated well, while the K d value for NADPH was 16-fold larger than that of NADH (Table 2). As a reference, steady-state kinetic assays were performed. The K m value of NADP + was slightly larger than that of NAD + ,andtheK m value of NADPH was 23-fold higher than that of NADH (Table 2). Transient-phase kinetics on oxidized nucleotide binding To determine whether nucleotide binding to 3a-HSD from P. sp. B-0831 constituted a one- or two-step mechanism, we used fluorescence stopped-flow measurement. The time course for oxidized nucleotide binding demonstrated that binding of not only NAD + but also NADP + is apparent by the fluorescence kinetic transients. The k app values for NAD + and NADP + were 84–225 s )1 and 62–129 s )1 , respectively. From the relationship of k app against the respective initial concentrations of NAD + and NADP + Table 1. Steady-state kinetic parameters for 3a-HSD from Pseudo- monas sp. B-0831 with steroid substrates. n.d., Not determined. Substrate K m a (l M ) k cat a (s )1 ) k cat b (s )1 ) Nucleotide Androsterone 210 ± 8 c 134 ± 24 c 16.4 NAD + Cholic acid 31 ± 2 c 75 ± 1 c n.d. NAD + Deoxycholic acid 54 ± 3 89 ± 6 n.d. NAD + Cholic acid 72 ± 2 4.5 ± 0.5 n.d. NADP + 5a-Androstan-3, 17-dione 64 ± 3 98 ± 2 11.6 NADH 5a-Androstan- 17b-ol-3-one 44 ± 4 98 ± 3 11.6 NADH Dehydrocholic acid 17 ± 1 78 ± 4 9.3 NADH Dehydrocholic acid 85 ± 3 3.3 ± 0.3 n.d. NADPH a Determined at 37 °C. b Determined at 15 °C. c Data were taken from Ueda et al. [37]. 1776 S. Ueda et al. (Eur. J. Biochem. 271) Ó FEBS 2004 (Fig. 2), the linear dependence of k app against both [NAD + ] and [NADP + ] indicated that the oxidized nucleotide cofactor binding is in accordance with a simple one-step mechanism. The NAAD + binding also showed a one-step mechanism (data not shown). The kinetic rate constants, k +1 and k )1 , and the dissociation constant, K )1 ,forthe nucleotide binding to 3a-HSD (Table 3) are of the same order to the corresponding K m value indicated in Table 2, although the measurements were conducted at different temperatures. Fig. 1. Fluorescence emission spectra from binary complexes of 3a-HSD with NAD + or NADH, where an intrinsic fluorescence emis- sion spectrum of 336 nm k max for 3a-HSD was portrayed at 280 nm excitation. (A) Spectra of 6.43 l M 3a-HSD in the absence (solid line) or presence of 6.7 l M (dotted line), 23.3 l M (broken line), and 90 l M (dot-dashed line) NAD + .(B)Spectraof0.49l M 3a-HSD in the absence (solid line) or presence of 1.3 l M (dotted line), 2.6 l M (broken line), and 6.7 l M (dot-dashed line) NADH. Table 2. K d values obtained from fluorescence titration and K m values obtained from steady-state kinetics for each nucleotide binding. Nucleotide Fluorescence titration K d a (l M ) Steady-state kinetics K m b (l M ) NAD + 173 ± 24 29 NADP + 192 ± 23 114 NAAD + 258 ± 26 373 c NADH 7.6 ± 0.9 11.4 NADPH 120 ± 4 268 a Determined at 15 °C. b Determined at 37 °C. c K i value, deter- mined in the inhibition assay. Fig. 2. Dependence of k app on the initial concentration of NAD + (A) and NADP + (B). All reactions were carried out in 50 m M Tris/HCl (pH 8.5) at 15 °C with 0.56 l M enzyme and various concentrations of NAD + (from 2.5 to 75 l M ) or NADP + (from 10 to 100 l M ).Values were expressed as the mean ± SD. The lines were drawn according to Eqn (2), using k +1 ¼ 1.79, k )1 ¼ 84.8 for NAD + and k +1 ¼ 6.34, k )1 ¼ 54.5 for NADP + . Ó FEBS 2004 Transient kinetics of nucleotide binding to 3a-HSD (Eur. J. Biochem. 271) 1777 Transient-phase kinetics on reduced nucleotide binding In the case of reduced nucleotide binding, the increase of fluorescence intensity in NADH binding could be monit- ored through a UV-42 cut-off filter. The k app value was 56–145 s )1 , and the dependence of k app on the initial concentration of NADH showed a hyperbolic curve (Fig. 3). Different from oxidized nucleotides, the kinetic feature of the NADH binding was consistent with a two- step mechanism, which involves a fast bimolecular associ- ation process followed by a slow unimolecular isomerization process. The kinetic rate constants, k +2 and k )2 ,andthe dissociation constants, K )1 and K d , are summarized in Table 3. The binding of NADPH to 3a-HSD was also investigated by stopped-flow measurement. Although the quench in intrinsic enzyme fluorescence emission spectrum following NADPH binding was observed here as well, there were no kinetic transients over a wide range of nucleotide concen- trations. These results suggest different binding modes toward NADPH and NADH. Discussion The transient-phase kinetics revealed that the binding of oxidized nucleotide cofactors, NAD + and NADP + ,to 3a-HSD from P. sp. B-0831 follows a one-step mechanism, while the binding of reduced nucleotide cofactor, NADH, follows a two-step mechanism. The binding of NAAD + , not a cofactor but a competitive inhibitor, also follows a one-step mechanism, as is the case of the oxidized cofactors. The validity of the proposed mechanism was supported by similar dissociation constants between the K )1 or K d values determined by stopped-flow measurements and the K d values determined by the fluorescence titration (Tables 2 and 3). For binding of oxidized nucleotides, the K d values by fluorescence titration ranged from 2.2- to 3.6-fold the corresponding K )1 values. The K d value for the NADH binding obtained by fluorescence titration was 2.5-fold that obtained by stopped-flow analysis. In HSD-related enzymes, whether or not the binding or release of nucleotide cofactors is the rate-limiting step in the reaction remains an issue [26,28,29]. The rate constant k )1 for release of NAD + in a one-step mechanism (84.8 s )1 ) was larger than the k cat value (11.6 s )1 ) for 5a-androstan- 3,17-dione reduction (Tables 1 and 3), indicating that the dissociation of NAD + from the B-0831 3a-HSD is not rate-limiting overall, a finding that coincides well with dissociation reported on the enzyme from rat liver [26]. The rate constant k )2 for release of NADH from the B-0831 3a-HSD in a two-step reaction (24 s )1 ) was slightly larger than the k cat value for androsterone oxidation (16.4 s )1 ). While the dissociation of NADH in this reaction could contribute to rate limiting, those in the oxidation of other substrates examined in this study would not be rate limiting as the k cat values are smaller than that for androsterone oxidation (Table 1). The rapid kinetic transients were not observed in the NAD(H) binding to rat liver 3a-HSD, suggesting different modes of binding toward NAD(H) and NADP(H) [26]. The K d values of NADP(H) binding were much smaller (c., three order in magnitude) than those of NAD(H) [27]. It was concluded the interaction between the 2¢-phosphate of NADP + and Arg276 was essential for the observation of kinetic transients. In contrast, rapid kinetic transients were observed in the binding of NAD + ,NADP + and NADH to P. sp. B-0831-derived 3a-HSD, albeit such was the case in NADPH binding. The oxidized nucleotide cofactors NAD + and NADP + bound to the enzyme with similar affinity, following a one-step catalysis. In the reduced nucleotide cofactors, the respective binding affinities were different, and the relatively higher binding affinity nucleo- tide, NADH, only showed a two-step catalysis. These results indicate that the reaction mechanism depends on the type of nucleotides, and is different from that of rat liver 3a-HSD. In order to discuss the nucleotide cofactor preference, the amino acid sequence in the nucleotide binding region was compared with those of other SDRs. Similar to 3a-HSD from P. sp. B-0831, 3a-HSD/CR from C. testosteroni and Table 3. Transient-phase kinetic parameters for the formation and decay of enzyme-nucleotide complexes. Nucleotide k +1 ( M )1 Æs )1 ) k )1 (s )1 ) K )1 (l M ) k +2 (s )1 ) k )2 (s )1 ) K )1 (l M ) K d (l M ) NAD + 1.79 ± 0.1 · 10 6 84.8 ± 0.8 47.6 ± 2.6 NADP + 6.34 ± 1.0 · 10 5 54.5 ± 3.3 87.2 ± 8.0 NAAD + 3.67 ± 0.3 · 10 5 41.5 ± 1.9 113.1 ± 4.7 NADH 278 ± 26 23.9 ± 3.1 39.2 ± 10.9 3.1 ± 0.9 Fig. 3. Dependence of k app on the initial concentration of NADH. All reactions were carried out in 50 m M Tris/HCl (pH 8.5) at 15 °C, with 0.84 l M enzyme and various concentrations of NADH (from 5 to 37.5 l M ). Values were expressed as the mean ± SD. The line was drawn according to Eqn (5), using k +2 ¼ 278, k )2 ¼ 24, K )1 ¼ 39. Insert: plot of 1/(k app – k )2 ) vs. 1/[NADH]. The k )2 value was estimated by linear extrapolation in the low concentration region of NADH. 1778 S. Ueda et al. (Eur. J. Biochem. 271) Ó FEBS 2004 7a-HSD from E. coli have common Asp residues at position 32 (numbering according to the B-0831 3a-HSD), which is highly conserved among enzymes preferring NAD + [14,16,38,39]. It is noteworthy that the substitution of Asp for Thr38 in mouse lung CR changed the nucleotide preference from NADP + to NAD + [40]. The crystal structure of C. testosteroni-derived 3a-HSD/CR complexed with NAD + has shown that Ile33, located next to Asp32, impedes NADP + with a 2¢-phosphate group [39]. The corresponding residue in 3a-HSD from P. sp. B-0831 is Arg33, which is conserved in the NADP + preferring enzymes, such as mouse liver 11b-HSD and mouse lung CR [16,41]. The crystal structure of mouse lung CR complexed with NADPH displays a pair of basic residues, Lys17 and Arg39, which correspond, respectively, to Ser11 and Arg33 in P. sp. B-0831-derived 3a-HSD, causing electrostatic interaction with the 2¢-phosphate group of NADP + [41]. These results indicate that the Asp32 and Arg33 residues are, respectively, critical for NAD + and NADP + preferences, resulting in the unique dual nucleotide cofactor specificity of B-0831 3a-HSD, with the binding affinity relatively lower than that of the NAD + -or NADP + -preferring enzyme [7,26,42]. The ordered bi-bi reaction mechanism can be explained by structural analyses: the substrate-binding loop of 3a-HSD is ordered when nucleotide cofactor binds to the site next to the substrate-binding site. The conformational change of C. testosteroni-derived 3a-HSD/CR induced by nucleotide binding is more subtle than that of rat liver 3a-HSD [25,39]. The conformational change observed in rat liver 3a-HSD is in good correlation with the slow formation of a tightly bound complex via a two-step reaction [26]. The NADH binding to P. sp. B-0831-derived 3a-HSD, which is a two-step reaction, may induce a conformational change similar to that of rat liver 3a-HSD. In contrast, the binding of NAD + or NADP + ,whichisa one-step reaction, may induce little conformational change similar to that of C. testosteroni-derived 3a-HSD/CR. The relatively lower binding affinity of these nucleotides supports the notion that only a loose complex is formed in the one-step reaction. It should also be noted that the k cat value for cholic acid as a substrate with NADP + is only 6% of that for the same substrate with NAD + , although both nucleotides bind to B-0831 3a-HSD with similar affinity. Additionally, no catalytic activity was detectable with NAAD + . In short, the different types of nucleotides can modulate the dynamic conformation and the catalytic function of the enzyme. References 1. Cheng, K.C., White, P.C. & Qin, K.N. (1991) Molecular cloning and expression of rat liver 3a-hydroxysteroid dehydrogenase. Mol. Endocrinol. 5, 823–828. 2. Jez, J.M., Flynn, T.G. & Penning, T.M. (1997) A new nomen- clature for the aldo-keto reductase superfamily. Biochem. Phar- macol. 54, 639–647. 3. MacDonald, I.A., Jellet, J.F., Mahony, D.E. & Holdeman, L.V. (1979) Bile salt 3a-and12a-hydroxysteroid dehydrogenases from Eubacterium lentum and related organisms. Appl. Environ. Microbiol. 37, 992–1000. 4. MacDonald, I.A., Meier, E.C., Mahony, D.E. &Costain,G.A.(1976) 3a-, 7a-and12a-hydroxysteroid dehydrogenase activities from Clostridium perfringens. Biochim. Biophys. Acta 450, 142–153. 5. Uwajima, T., Takayama, K. & Terada, O. (1978) Production, purification and crystallization of 3a-hydroxysterid dehydro- genase from Pseudomonas putida. Agric. Biol. Chem. 42, 1577– 1583. 6. Boyer, J., Baron, D.N. & Talalay, P. (1965) Purification and properties of 3a-hydroxysteroid dehydrogenase from Pseudo- monas testosteroni. Biochemistry 4, 1825–1833. 7. Ska ˚ lhegg, B.A. (1975) 3a-hydroxysteroid dehydrogenase from Pseudomonas testosteroni: kinetic properties with NAD and its thionicotinamide analogue. Eur. J. Biochem. 50, 603–609. 8.Maser,E.,Oppermann,U.C.,Bannenberg,G.&Netter,K.J. (1992) Functional and immunological relationships between metyrapone reductase from mouse liver microsomes and 3a-hy- droxysteroid dehydrogenase from Pseudomonas testosteroni. FEBS Lett. 297, 196–200. 9. Oppermann, U.C. & Maser, E. (1996) Characterization of a 3a-hydroxysteroid dehydrogenase/carbonyl reductase from the gram-negative bacterium Comamonas testosteroni. Eur. J. Bio- chem. 241, 744–749. 10. Suzuki, K., Ueda, S., Sugiyama, M. & Imamura, S. (1993) Cloning and expression of a Pseudomonas 3a-hydroxysteroid dehydrogenase-encoding gene in Escherichia coli. Gene 130, 137– 140. 11. Persson, B., Krook, M. & Jo ¨ rnvall, H. (1991) Characteristics of short-chain alcohol dehydrogenases and related enzymes. Eur. J. Biochem. 200, 537–543. 12. Jo ¨ rnvall, H., Persson, B., Krook, M., Atrian, S., Gonzalez- Duarte, R., Jeffery, J. & Ghosh, D. (1995) Short-chain dehy- drogenases/reductases (SDR). Biochemistry 34, 6003–6013. 13. Jez, J.M., Bennett, M.J., Schlegel, B.P., Lewis, M. & Penning, T.M. (1997) Comparative anatomy of the aldo-keto reductase superfamily. Biochem. J. 326, 625–636. 14. Kallberg, Y., Oppermann, U., Jo ¨ rnvall, H. & Persson, B. (2002) Short-chain dehydrogenases/reductases (SDRs). Eur. J. Biochem. 269, 4409–4417. 15. Maser, E., Mo ¨ bus, E. & Xiong, G. (2000) Functional expression, purification, and characterization of 3a-hydroxysteroid dehydro- genase/carbonyl reductase from Comamonas testosteroni. Biochem. Biophys. Res. Commun. 272, 622–628. 16. Mo ¨ bus, M. & Maser, E. (1998) Molecular cloning, over- expression, and characterization of steroid-inducible 3a-hydroxy- steroid dehydrogenase/carbonyl reductase from Comamonas testosteroni. A novel member of the short-chain dehydrogenase/ reductase superfamily. J. Biol. Chem. 273, 30888–30896. 17. Ueda, S., Oda, M., Imamura, S. & Ohnishi, M. (2004) Carbonyl reductase activity of a pluripotent enzyme, 3a-hydroxysteroid dehydrogenase from Pseudomonas sp. B-0831. J. Biol. Macromol. 4, 29–32. 18. Oda, K., Yoshida, S. & Takeda, T. (1989) Determination of total 3a-hydroxy bile acids in serum by a bioluminescent flow injection system using a hollow-fiber reactor. Anal. Chim. Acta 225, 273–282. 19. Takahashi, M., Ueda, S., Misaki, H., Sugiyama, N., Mastumoto, K., Matsuo, N. & Murao, S. (1994) Carnitine determination by an enzymatic cycling method with carnitine dehydrogenase. Clin. Chem. 40, 817–821. 20. Penning, T.M., Smithgall, T.E., Askonas, L.J. & Sharp, R.B. (1986) Rat liver 3a-hydroxysteroid dehydrogenase. Steroids 47, 221–247. 21. Askonas, L.J., Ricigliano, J.W. & Penning, T.M. (1991) The kinetic mechanism catalysed by homogeneous rat liver 3a- hydroxysteroid dehydrogenase: evidence for binary and ternary Ó FEBS 2004 Transient kinetics of nucleotide binding to 3a-HSD (Eur. J. Biochem. 271) 1779 dead-end complexes containing non-steroidal anti-inflammatory drugs. Biochem. J. 278, 835–841. 22. Pawlowski, J.E. & Penning, T.M. (1994) Overexpression and mutagenesis of the cDNA for rat liver 3a-hydroxysteroid/ dihydrodiol dehydrogenase. Role of cysteines and tyrosines in catalysis. J. Biol. Chem. 269, 13502–13510. 23. Jez, J.M., Schlegel, B.P. & Penning, T.M. (1996) Characterization of the substrate binding site in rat liver 3a-hydroxysteroid/ dihydrodiol dehydrogenase: the roles of tryptophans in ligand binding and protein fluorescence. J. Biol. Chem. 271, 30190– 30198. 24. Bennett, M.J., Schlegel, B.P., Jez, J.M., Penning, T.M. & Lewis, M. (1996) Structure of 3a-hydroxysteroid/dihydrodiol dehydro- genase complexed with NADP + . Biochemistry 35, 10702– 10711. 25. Bennett, M.J., Albert, R.H., Jez, J.M., Ma, H., Penning, T.M. & Lewis, M. (1997) Steroid recognition and regulation of hormone action: crystal structure of testosterone and NADP + bound to 3a-hydroxysteroid/dihydrodiol dehydrogenase. Structure 5, 799–812. 26. Ratnam, K., Ma, H. & Penning, T.M. (1999) The arginine 276 anchor for NADP(H) dictates fluorescence kinetic transients in 3a-hydroxysteroid dehydrogenase, a representative aldo-keto reductase. Biochemistry 38, 7856–7864. 27. Ma, H., Ratnam, K. & Penning, T.M. (2000) Mutation of nicotinamide pocket residues in rat liver 3a-hydroxysteroid dehydrogenase reveals different modes of cofactor binding. Biochemistry 39, 102–109. 28. Kubiseski, T.J., Hyndman, D.J., Morjana, N.A. & Flynn, T.G. (1992) Studies on pig muscle aldose reductase: kinetic mechanism and evidence for a slow conformational change upon coenzyme binding. J. Biol. Chem. 267, 6510–6517. 29. Grimshaw, C.E., Bohren, K.M., Lai, C.J. & Gabbay, K.H. (1995) Human aldose reductase: rate constants for a mechanism includ- ing interconversion of ternary complexes by recombinant wild- type enzyme. Biochemistry 34, 14356–14365. 30. Lowry, O.H., Rosebrough, N.J., Farr, A.L. & Randall, R.J. (1951) Protein measurement with the folin phenol reagent. J. Biol. Chem. 193, 265–275. 31. Sakoda, M. & Hiromi, K. (1976) Determination of the best-fit values of kinetic parameters of the Michaelis-Menten equation by the method of least squares with the Taylor expansion. J. Biochem. 80, 547–555. 32. Tanaka, A., Ohnishi, M. & Hiromi, K. (1982) Stopped-flow kinetic studies on the binding of gluconolactone and maltose to glucoamylase. Biochemistry 21, 107–113. 33. Tonomura, B., Nakatani, H., Ohnishi, M., Yamaguchi-Ito, J. & Hiromi, K. (1978) Test reactions for a stopped-flow apparatus: reduction of 2,6-dichlorophenolindophenol and potassium ferri- cyanide by 1-ascorbic acid. Anal. Biochem. 84, 370–383. 34. Guggenheim, E.A. (1926) On the determination of the velocity constant of a unimolecular reaction. Phil. Mag. 2, 538–543. 35. Strickland, S., Palmer, G. & Massey, V. (1975) Determination of dissociation constants and specific rate constants of enzyme-sub- strate (or protein–ligand) interactions from rapid reaction kinetic data. J. Biol. Chem. 250, 4048–4052. 36. Halford, S.E. (1975) Stopped-flow fluorescence studies on sac- charide binding to lysozyme. Biochem. J. 149, 411–422. 37. Ueda,S.,Oda,M.,Imamura,S.&Ohnishi,M.(2004)Steady-state kinetic properties of 3a-hydroxysteroid dehydrogenase from Pseudomonas sp. B-0831: Steroid substrate specificity and nuc- leotide cofactor dependency. J. Biol. Macromol. 4, 23–28. 38. Tanaka,N.,Nonaka,T.,Tanabe,T.,Yoshimoto,T.,Tsuru,D.& Mitsui, Y. (1996) Crystal structures of the binary and ternary complexes of 7a-hydroxysteroid dehydrogenase from Escherichia coli. Biochemistry 35, 7715–7730. 39. Grimm, C., Maser, E., Mo ¨ bus,E.,Klebe,G.,Reuter,K.&Ficner, R. (2000) The crystal structure of 3a-hydroxysteroid dehydro- genase/carbonyl reductase from Comamonas testosteroni shows a novel oligomerization pattern within the short chain dehydro- genase/reductase family. J. Biol. Chem. 275, 41333–41339. 40. Nakanishi, M., Matsuura, K., Kaibe, H., Tanaka, N., Nonaka, T., Mitsui, Y. & Hara, A. (1997) Switch of coenzyme specificity of mouse lung carbonyl reductase by substitution of threonine 38 with aspartic acid. J. Biol. Chem. 272, 2218–2222. 41. Tanaka, N., Nonaka, T., Nakanishi, M., Deyashiki, Y., Hara, A. & Mitsui, Y. (1996) Crystal structure of the ternary complex of mouse lung carbonyl reductase at 1.8 A ˚ resolution: the structural origin of coenzyme specificity in the short-chain dehydrogenase/ reductase family. Structure 4, 33–45. 42. Wilson, D.K., Nakano, T., Petrash, J.M. & Quiocho, F.A. (1995) 1.7 A ˚ structure of FR-1, a fibroblast growth factor-induced member of the aldo-keto reductase family, complexed with coen- zyme and inhibitor. Biochemistry 34, 14323–14330. 1780 S. Ueda et al. (Eur. J. Biochem. 271) Ó FEBS 2004 . studies on the nucleotide binding to 3a-hydroxysteroid dehydrogenase from Pseudomonas sp. B-0831 using fluorescence stopped-flow procedures Shigeru Ueda 1 , Masayuki. Kyoto Prefectural University, Kyoto, Japan The dual nucleotide cofactor-specific enzyme, 3a-hydroxy- steroid dehydrogenase (3a-HSD) from Pseudomonas sp. B-0831,