Short Communication Inhibitory Effects of Benzoate on Chiral Inversion and Clearance of N-Nitro-Arginine in Conscious Rats

N-nitro-arginine (NNA) is known to exhibit stereoselective pharmacokinetics in which N-nitro-D-arginine (D-NNA) has a faster clearance rate than N-nitro-L-arginine (L-NNA) in anesthetized rats, and D-NNA undergoes unidirectional chiral inversion. It was postulated that chiral inversion of D-NNA was performed in a twostep pathway by D-amino acid oxidase (DAAO) followed by an unidentified transaminase. Such chiral inversion contributes (at least partially) to the pharmacokinetic stereoselectivity of NNA. This study used the selective inhibitor of DAAO, sodium benzoate, to test the above hypothesis. An i.v. bolus injection of D-NNA (32 mg/kg) and L-NNA (16 mg/kg) in conscious rats exhibited biphasic disposition with different pharmacokinetic parameters in a stereospecific manner (approximately 5–10-fold differences). Unidirectional chiral inversion of D-NNA but not L-NNA was found from these animals. In addition to its similar inhibitory effects on the D-NNA conversion and DAAO activity in kidney homogenates, sodium benzoate completely blocked chiral inversion of D-NNA and led to a smaller stereospecific difference, reflected by a nearly 50% reduction of D-NNA clearance and a 2-fold increase in t1/2 and area under the curve of D-NNA in benzoate-pretreated rats. The results suggest that DAAO plays an essential role in chiral inversion of D-NNA and chiral inversion contributes mostly to the pharmacokinetic stereospecificity of NNA. N-nitro-arginine (NNA) is an inhibitor of nitric oxide synthase, exhibiting stereospecificity such that the Lbut not the D-enantiomer inhibits nitric oxide synthesis in vitro (Moncada et al., 1991; Wang et al., 1999). However, in vivo administration of the inactive D-enantiomer (D-NNA) produced the same biological effects as its L-enantiomer (L-NNA), such as increasing blood pressure that was blocked by administration of L-arginine but not D-arginine (Wang et al., 1991, 1993), leading to a notion that D-NNA was converted into L-NNA in vivo (Wang et al., 1993). Wang et al. (1999) from another laboratory showed that D-NNA was indeed unidirectionally converted into LNNA in vivo by using chiral high-performance liquid chromatography. Furthermore, the kidney was confirmed to be the major organ, accounting for approximately 80% of chiral inversion of D-NNA in vivo (Xin et al., 2005). It was hypothesized that D-NNA unidirectionally converts to LNNA by a two-step pathway involving the oxidation of D-NNA to N-nitro-5-gunidino-2-oxopentanoic acid by D-amino acid oxidase (DAAO), followed by amination to L-NNA by an unidentified transaminase (Wang et al., 1999; Xin et al., 2006). Several lines of evidence support the above hypothesis: 1) DAAO incubation with D-NNA in vitro reduced D-NNA content (Wang et al., 1999); 2) the chiral inversion rate of D-NNA was parallel with the DAAO activity in tissue homogenates (Xin et al., 2005); and 3) injection of the inhibitor of DAAO benzoate completely blocked the pressor response to naive D-NNA in conscious rats (Xin et al., 2005). However, further studies, such as testing DAAO inhibitors on chiral inversion of DNNA in vitro and in vivo, are needed to confirm the above hypothesis. It is known that chiral drugs may exhibit stereoselective pharmacokinetics. Stereoselective disposition of NNA was observed in anesthetized rats in which D-NNA had a faster clearance than that of L-NNA (Wang et al., 1999). Chiral inversion is one type of drug metabolism that contributes to drug clearance. However, it is not yet known whether chiral inversion of D-NNA accounts for pharmacokinetic stereoselectivity of NNA. This study aimed to test whether the DAAO inhibitor sodium benzoate blocks the chiral inversion of D-NNA in rat kidney homogenates and conscious rats. This study also intended to confirm pharmacokinetic stereospecificity of NNA in conscious rats, as well as deduce whether chiral inversion of D-NNA contributes to this stereospecificity. Our results showed that benzoate completely blocked chiral inversion of D-NNA in vitro and in vivo, suggesting that D-NNA conversion is via the metabolism of DAAO. Moreover, NNA exhibited its pharmacokinetics in a stereospecific manner in which the D-enantiomer cleared much faster (5-fold higher) as a result of its chiral inversion because administration of sodium benzoate mostly reduced the clearance of D-NNA. Materials and Methods Drugs and Reagents. D-NNA was obtained from Bachem Bioscience Inc. (King of Prussia, PA), and L-NNA and aspartame were purchased from Acros Organics N.V. (Geel, Belgium). All the other reagents were purchased from Shanghai Chemical Reagents Co. (Shanghai, China). D-NNA and L-NNA were This work was supported by grants from the National Natural Sciences Foundation of China (30472052) and the Science Foundation of Shanghai Municipal Commission of Sciences and Technology (04DZ19214). Article, publication date, and citation information can be found at http://dmd.aspetjournals.org. doi:10.1124/dmd.106.011429. ABBREVIATIONS: NNA, N-nitro-arginine; D-NNA, N-nitro-D-arginine; L-NNA, N-nitro-L-arginine; DAAO, D-amino acid oxidase; CEC, capillary electrochromatography; AUC, area under the curve; CL, clearance. 0090-9556/07/3503-331–334$20.00 DRUG METABOLISM AND DISPOSITION Vol. 35, No. 3 Copyright © 2007 by The American Society for Pharmacology and Experimental Therapeutics 11429/3181864 DMD 35:331–334, 2007 Printed in U.S.A. 331 at A PE T Jornals on Jne 1, 2017 dm d.aspurnals.org D ow nladed from dissolved in 0.9% saline solution, and the dissolution required 20 min of ultrasonication. Animal Preparation. Male Sprague-Dawley rats (350–400 g) from Fudan University Medical Animal Center (Shanghai, China) were anesthetized with sodium pentobarbital (65 mg/kg, i.p.). A polyethylene cannula (PE50, Becton Dickinson Co., Franklin Lakes, NJ) was inserted into the left femoral vein for the collection of blood samples. The vascular cannula were tunneled s.c. and exteriorized at the back of the neck. The rats were given at least a 24-h recovery from anesthesia and surgery before use. Preparation of Tissue Homogenates. Rats were sacrificed, and the kidneys were removed and washed with Tris-HCl buffer (4°C, 0.1 M, pH 8.2). Tissue samples (1 g each) were homogenized in Tris-HCl buffer (3 ml, pH 8.2) at 4°C with a homogenizer (Fluko Equipment Co., Shanghai, China). The homogenates were centrifuged at 1500 rpm for 10 min, and the supernatants were used for the measurements of the DAAO activity and D-NNA conversion. Determination of DAAO Activity. The activity of DAAO was determined according to the “keto acid method” (D’Aniello et al., 1993; Sarower et al., 2003). The supernatants (200 l) of the homogenates were incubated with 0.1 M D-alanine (200 l, dissolved in the above Tris-HCl buffer) plus additional sodium benzoate solution (100 l, concentration range 0.00025–500 mg/ml) for 30 min in a shaking bath (700 rpm) at 37°C. The control samples were treated similarly, except the benzoate solution (100 l) was replaced by Tris-HCl buffer (100 l). Afterward, trichloroacetic acid (200 l, 25%) was added to the incubated mixture, mixed again, and centrifuged at 12,000 rpm for 10 min. The supernatant (400 l) was mixed with 2,4-dinitrophenylhydrazine (400 l of 1 mM in 1 M HCl) and incubated at 37°C for 10 min. NaOH (800 l of 1.5 M) was subsequently added and mixed. The mixture was kept at room temperature for 10 min, and the absorbance was read at 445 nm against a blank sample consisting of the same mixture without D-alanine. The activity of DAAO in the homogenates was quantified against the standard curve of pyruvic acid (from 0.35 nM to 2.80 nM, r 0.99). Tissue Homogenate Incubation with D-NNA. The effects of kidney homogenates (n 4) on the conversion of D-NNA into L-NNA were determined. The incubation systems of D-NNA and kidney homogenates for conversion were treated similarly to that for DAAO activity detection, except that in the D-alanine solution (200 l of 0.1 M), the substrate of DAAO was replaced by D-NNA solution (200 l of 50 mM, dissolved in the same Tris-HCl buffer). In some experiments, the homogenates were first denatured through heating (10 min at 100°C). The incubation supernatants were used for ex vivo determination of D-NNA and L-NNA by capillary electrochromatography (CEC). Measurement of D-NNA and L-NNA in Biological Samples. D-NNA and L-NNA were measured by CEC in the chiral ligand exchange mode (Xin et al., 2005). Plasma samples (100 l) or incubation samples (100 l) of D-NNA or L-NNA were deproteinated through mixing for 20 min with methanol/acetone (1000 l; v/v 1:1). After centrifugation at 10,000 rpm for 20 min, the organic layer was dried at 20°C, and the residue was dissolved in acetate buffer (pH 5.0, plasma in 100 l and incubation samples in 5 ml). After mixing for 1 min, a 10-nl aliquot of the sample was injected into the CEC system (Unimicro Technologies, Inc., Pleasanton, CA). Cupric acetate (1 mM) and aspartame (2 mM) were dissolved in the methanol/acetate buffer (pH 5.0, v/v 1:20) to allow the formation of diastereomeric pairs with D-NNA or L-NNA. The diastereomeric pairs were separated through a reverse-phase C18 column (75 m 20 cm; Unimicro Tech. Ltd, Shanghai, China) at a detection wavelength of 280 nm. The retention times for L-NNA and D-NNA were 17 and 19 min, respectively. The samples were quantified against standard curves of L-NNA and D-NNA that ranged from 0.025 to 0.75 mM (r 0.98). The between-day (n 5) precision values measured by the time and the area under curve were 3.5 and 3.9% for L-NNA and D-NNA, respectively. The within-day precision of the quality control samples was 2% for both L-NNA and D-NNA. Pharmacokinetic Studies. Conscious rats were divided into four groups (n 5 in each group). Two groups of rats were injected with D-NNA (32 mg/kg) or L-NNA (16 mg/kg), respectively. The other two groups of rats were pretreated with injection of sodium benzoate (400 mg/kg), a selective inhibitor of DAAO that effectively blocks DAAO activity at this dose (Moses et al., 1996; Xin et al., 2005, 2006). Twenty minutes later, the benzoate-pretreated group was injected with D-NNA (32 mg/kg) or L-NNA (16 mg/kg), respectively. All the drugs were injected as an i.v. bolus via the indwelling cannula. The cannula was flushed with saline solution (600 l, 0.9%) to ensure no contamination during blood sampling. Blood samples (200 l) were obtained at 5, 20, 40, 60, 90, 120, 180, 240, 300, and 420 min after administration. Blood volume was replaced by injection of an equal volume of saline solution (0.9% NaCl) after each sampling. Plasma samples obtained were stored at 20°C for later analysis by CEC. Pharmacokinetic Analyses. Pharmacokinetic parameters were obtained from the plasma concentration-time curves using a noncompartmental model. Data were weighted 1/c. The elimination half-life (t1/2) was calculated by 0.693/Kel. The chiral inversion rate of D-NNA to L-NNA was calculated using the following formula (Pang and Kwan, 1993): FD-NNA3L-NNA AUCL-NNA from D-NNA/DoseD-NNA AUCL-NNA/DoseL-NNA The area under the curve (AUC) was calculated by the trapezoidal rule. Data Analyses. IC50 values for benzoate to block DAAO and D-NNA conversion were determined by using a curve fitting program, DoseResp of Origin 7.5 (Northampton, MA). All the results were expressed as mean S.E.M. and analyzed by the analysis of variance followed by Duncan’s multiple range test. Differences with values of P 0.05 were considered statistically significant. Results and Discussion Previous findings suggested that chiral inversion of D-NNA was metabolized by kidney DAAO (Wang et al., 1999; Xin et al., 2005, 2006). However, further experiments such as those using DAAO inhibitors are needed to confirm the above hypothesis. Sodium benzoate was selected as a DAAO inhibitor for this study to examine whether DAAO inhibitors block chiral inversion of D-NNA in vitro and in vivo. Our previous study indicated that D-NNA but not L-NNA was the substrate of DAAO, and the Kcat/Km of DAAO to D-NNA was approximately 10% of that to D-alanine, the optimum substrate for kidney DAAO (Xin et al., 2006). In kidney homogenates, DAAO activity was measured by using the “keto acid method,” in which 0.1 M D-alanine was used (D’Aniello et al., 1993; Sarower et al., 2003). Sodium benzoate inhibited DAAO activity in a sigmoid concentration-response manner (Fig. 1), with the IC50 value of 6.7 mM (95% confidence interval, 5.72–7.68 mM). Our results, together with a previous report that the Kd for benzoate to inhibit DAAO was approximately 0.2 mM (Pilone, 2000), suggest that benzoate is a weak inhibitor of DAAO. Moreover, the chiral inversion of D-NNA in this study was observed at 50 mM D-NNA, in contrast to the measurement of DAAO activity in which 0.1 M D-alanine was used. Benzoate also blocked the chiral inversion of D-NNA in kidney homogenates, with the concentration-response curve correlating very well with that of the FIG. 1. Inhibitory effects of sodium benzoate on the DAAO activity (f) and the chiral inversion of D-NNA (F) in rat kidney homogenates (n 4 samples in each point). All the readings are mean S.E.M. 332 YAN-FEI ET AL. at A PE T Jornals on Jne 1, 2017 dm d.aspurnals.org D ow nladed from inhibition of DAAO activity (Fig. 1). The IC50 value for benzoate to block D-NNA disposition was 2.1 mM (95% confidence interval, 1.31–2.88 mM), which was close to the inhibition of the DAAO activity. These results suggest benzoate blocks D-NNA conversion via the inhibition of the DAAO activity in kidney homogenates. Pharmacokinetics of D-NNA was studied in two groups of conscious rats (n 5 in each group) pretreated with the vehicle or benzoate. An i.v. bolus injection of D-NNA (32 mg/kg) exhibited a biphasic disposition curve, whereas D-NNA was almost undetectable in plasma after 3 h of administration (Fig. 2A), with calculated clearance (CL) and t1/2 of 0.98 0.07 l/h/kg and 0.55 0.33 h, respectively (Table 1). Meanwhile, L-NNA was immediately detected in the plasma following D-NNA injection, and the plasma concentration of L-NNA exceeded that of D-NNA at 1.5 h and reached the peak at 2 h after D-NNA dosage (Fig. 2A). To observe whether inhibition of the DAAO activity blocked the chiral inversion of D-NNA in vivo, conscious rats were pretreated with benzoate at 400 mg/kg, a dosage selected based on the biological effects of D-dopa (Moses et al., 1996), as well as results from a previous study that showed benzoate led to the inhibition of DAAO activity (Williams and Lock, 2005). As seen in Fig. 2B, benzoate completely blocked D-NNA conversion in vivo, in agreement with our previous study that showed benzoate abolished the pressor response to naive D-NNA (Xin et al., 2005). In contrast to control rats, in which nearly no D-NNA was detected in plasma samples at 3 h after administration, plasma D-NNA was kept at 10 g/ml at 3 h and was still detectable at 7 h after D-NNA dosing in benzoate-pretreated rats. As a result of the blockade of chiral inversion of D-NNA, the mean clearance of D-NNA was significantly reduced (P 0.05) by nearly 50% from 0.98 0.07 l/h/kg to 0.49 0.12 l/h/kg on coadministration (Table 1). Table 1 also shows that benzoate significantly increased t1/2 of D-NNA and approximately doubled the AUC of D-NNA. The effectiveness of benzoate to block chiral inversion of D-NNA and partially reduce clearance of D-NNA is in agreement with the previous report that pretreatment of sodium benzoate (1000 mg/kg, i.p.) led to a 40% reduction of D-ethionine disposition in 4 h in rats (Brada and Bulba, 1980). Similar results have also been shown in the study of D-methionine metabolism, in which i.p. treatment with sodium benzoate (0.9 g/kg) reflected a gradual clearance of D-methionine from the blood and a decrease of hepatic metabolism to D-methionine (London and Gabel, 1988). To find out whether the effect of benzoate on pharmacokinetics of D-NNA was specific on chiral inversion blockade, the effect of benzoate on L-NNA pharmacokinetics was tested in two groups of conscious rats (n 5 in each group). An i.v. bolus injection of L-NNA (16 mg/kg) also exhibited a biphasic disposition curve in control rats (Fig. 3A), with calculated CL and t1/2 of 0.19 l/h/kg and 5.21 h, respectively (Table 1). These parameters clearly showed a stereoselectivity in pharmacokinetics of NNA, in which CL and t1/2 of L-NNA had approximately 5and 10-fold differences, respectively, from those of D-NNA (Table 1). The results are consistent with previous findings that showed the stereoselective disposition of NNA in anesthetized rats (Wang et al., 1999). Unlike in D-NNA injection, in which L-NNA was immediately detected, the D-enantiomer of L-NNA was not detected in plasma samples during the period observed after L-NNA administration, confirming that chiral inversion of D-NNA was unidirectional (Wang et al., 1999). On the other hand, benzoate pretreatment, as shown in Fig. 3B, did not significantly affect pharmacokinetic parameters of L-NNA (Table 1), excluding the possibility that the effect of benzoate on D-NNA chiral inversion and clearance was nonspecific. Moreover, in the presence of benzoate for complete blockade of chiral inversion, the difference in