Synthesis and reduction of ( S )-(-)-nicotine-N ́-oxide and N , N ́-dioxides by rat liver S9 fraction

cisAnd trans-Nicotine-N’-oxide (5) and N,N’-dioxide (7) diastereomers were synthesized in good yields from nicotine (1) using m-CPBA. Asymmetric reductions of nicotine N-oxides with rat liver S-9 fraction were investigated.


Introduction
Nicotine (1) is one of the main pharmacologically active constituents of tobacco, and when nicotine (1) is administered it is well absorbed by digestive and respiratory organs as is metabolized to cotinine (2), 5'-hydroxycotinine (3) and 3'-hydroxycotinine (4) by CYP2A6.Nicotine (1) is also metabolized by FMO to trans-nicotine-N'-oxide (5a) (Scheme 1). 1,2The Noxidation of nicotine (1) shows an interesting stereoselectivity in the formation of cis-and transnicotine-N'-oxides (5a and 5b). 3 The kinetic properties of the formation of these diastereoisomeric nicotine-N'-oxides (5a and 5b) in rat liver microsomes were only recently reported. 4Interestingly, in humans, the trans-isomer was the major excreted urinary metabolite, while cis-nicotine-N'-oxide (5b), nicotine-N-oxide (6), and the further oxidized product, nicotine-N,N'-dioxide (7) are not found in the urine.To understand these observations nicotine-N-oxides were prepared using m-CPBA and the transition state energies of the reaction intermediates were investigated by an ab-initio MO method.The metabolic reduction of nicotine N-oxides using the rat liver S-9 fraction was also investigated.
Table 1.Yields of products and E a in the reactions of (S)-nicotin4e (1) and its N-oxides with m-CPBA, using Gaussian 98 at the RHF/6-31++G(d,p)//RHF/6-31G(d) level  In the energy diagrams of those reactions, Ea for the formation of cis and trans nicotine-N'oxides (5a and 5b) were 29.05 and 27.01 kcal/mol, respectively.Our calculations suggest that the trans isomer is more stable than the cis isomer, and that the formation rate of the cis isomer is much smaller than that of the trans isomer.The highest Ea value (38.03 kcal/mol) was for forming nicotine-N-oxide (6) (Table 1), which is consistent with the experimental result that nicotine-N-oxide (6) was not obtained in the reaction of (S)-nicotine (1) with m-CPBA.The Ea for the oxidation of the N of cis-and trans-nicotine-N'-oxides (5a and 5b) were 36.59 and 40.79 kcal/mol, respectively, and these values were much higher than those corresponding to N'.

Reduction of the nicotine N-oxides by rat liver S-9 fraction
Scheme 3 shows the results from the metabolic reductions of nicotine N-oxides.Diastereomeric mixture of nicotine-N'-oxide (5) was transformed to nicotine (1), and nicotine-N,N'-dioxide (7)  was converted into nicotine-N-oxide (6).In these reactions, the aliphatic N-oxide was reduced faster than the aromatic N-oxide.
Experiments performed to determine the apparent K m and V max for the reduction of the diastereo mixture of nicotine-N'-oxides (5) and nicotine-N,N'-dioxides (7) were conducted at a final protein concentration of 6.0 mg/mL (30 min incubation) and in the presence of increasing concentrations of nicotine-N'-oxide (5.6-262.2mM) or nicotine-N,N'-dioxide (0.86-240.55 mM).Under these conditions, the metabolite formation was linear with respect to the S-9 concentration and time of incubation.Kinetic analyses of the enantioselective reductive metabolism of nicotine-N-oxides with rat liver S-9 fraction were conducted using Lineweaver-Burk plots.The reductions of the diastereomeric mixture of nicotine-N'-oxides ( 5) and nicotine-N,N'-dioxides (7) were characterized by V max of 0.037 and 0.271 nmol/mg protein/min, respectively, whereas the estimated apparent K m for these reactions were 114.07 and 81.73 mM, respectively (Fig. 2).Kinetic analyses of the enantioselectivity of the reductive metabolism of nicotine-N-oxides with rat liver S-9 fraction showed that V max /K m values were 0.3 nL/mg protein/min for reducing of nicotine-N'-oxide (5) to nicotine (1) and 3.3 nL/mg protein/min for the reduction of nicotine-N,N'-dioxide (7) to nicotine N-oxide.These results suggest that the reduction of nicotine-N,N'-dioxide ( 7) is much faster than that of nicotine-N'-oxide (5) and that reduction of aromatic N-oxide does not proceed under these experimental conditions.When a diastereomeric mixture of nicotine-N'-oxides (5) was treated with rat liver S-9 fraction, cis-nicotine-N'-oxide (5b) was reduced faster than trans-nicotine-N'-oxide (5a).(Fig. 3).

Acute toxicity
The acute toxicities (LD 50 ) of nicotine-N-oxides (diastereomeric mixtures of 5 and 7, and 6) in mice was determined by the Litchfield-Wilcoxon method. 10The results demonstrated that, unlike nicotine (LD 50 = 9.5 mg/kg), these nicotine N-oxides did not show an acute toxicity even at 200 mg/kg after intraperitoneal injection (Table 2).The acute toxicities of the nicotine-Noxides clearly show that when the N atoms are oxidized, the toxicity decreases.Since the oxidation of the aliphatic amine was more effective in reducing the toxicity than that of the pyridine nitrogen, in humans, metabolization of nicotine (1) to nicotine-N'-oxide (5a) accomplishes two goals.First nicotine ( 1) is transformed into a non-toxic metabolite.Second, excretion is facilitated.

Calculation of activation energies
The oxidation reactions of nicotine and its N-oxides with m-CPBA are considered to proceed as follows: The reactant molecules interact, forming an equilibrium complex (Eq).Then, the final compound is produced through a transition state (TS).The structures of Eq and TS were optimized using Gaussian 98 at the RHF/6-31++G (d,p)//RHF/6-31G (d) level. 7The solvent effect was not considered.The activation energy was calculated as the energy difference between the TS and the Eq.After optimizing the TS structure, the vibrational frequencies were calculated and it was confirmed that the TS had a single imaginary vibrational frequency.

Animals
Male Wistar rats (150-180 g) were supplied by Japan SLC (Hamamatsu, Japan) and housed in an air-conditioned room (23 ± 1 o C and 55.0 ± 5% relative humidity) with a 12 h light cycle, and allowed to feed (CE-2 obtained from Japan Crea Co., Japan) and drink water ad libitum.

Preparation of rat liver 9000 g supernatant fraction
Rat liver S-9 fraction was prepared from untreated rats as previously described. 8Briefly, the liver was removed and perfused with saline, and the liver sample was homogenized in an ice-cold 0.1 M K, Na-phosphate buffer.The homogenates were centrifuged (9000 g, 4 o C) for 20 min, and the supernatant (S-9) was used in the experiments.The protein concentration was determined using the method of Bradford et al. 9 with bovine serum albumin as a standard.

in vitro Reduction of diastereomeric nicotine-N'-oxide (5) and nicotine-N,N'-dioxide (7).
A mixture (final volume, 6 mL) consisting of 10 mM Na,K-phosphate buffer (pH 7.4), NADPHgenerating system (1.3 mM NADP, 10 mM glucose 6-phosphate, 5 mM MgCl 2 , 0.4 U/mL glucose 6-phosphate dehydrogenase), 6.0 mg/mL S-9 fraction and nicotine-N-oxides (5 or 7) was incubated at 37 o C for 30 min.Adding a 2.0 M solution of aqueous HClO 4 (3 mL) followed by caffeine as an internal standard terminated the reaction.After centrifugation (800 g, 15 min) to remove the protein, a 1N solution of aqueous NaOH (2 mL) and CHCl 3 (15 mL) was added to the supernatant, which was subsequently shaken for 15 min.After the phases separated, 10 mL of the organic fraction was concentrated in vacuo and the residue was redissolved in 200 µL of MeOH and a 20 µL portion of the sample was subjected to HPLC.
HPLC analyses were performed using a Shodex DEGAS KT-27 degasser (Showa Denko, Tokyo, Japan), and a Waters 510 HPLC Pump (Waters, Milford MA, USA) equipped with a CAPCELL PAK SCX column (4.6 × 150 mm, 5 µm; Shiseido, Tokyo, Japan).The eluent was monitored at 220 nm using a Waters 486 Tunable Absorbance Detector (Waters).The mobile phase was 100 mM K, Na-phosphate buffer and acetonitrile (1:1, v/v).The flow rate was 1.0 mL/min and the column temperature was 40 o C. The metabolites were quantified by comparing the HPLC peak areas with those of authentic standards with an internal standard reference.

Kinetic analysis
The kinetic studies were performed using rat liver S-9 fraction incubated with nicotine-N'-oxide and nicotine-N,N'-dioxide concentrations ranging from 5.6 to 262 and 0.86 to 240 mM, respectively.The maximum velocity (V max ) and Michaelis-Menten constant (K m ) were evaluated by non-linear regression analysis of the untransformed kinetic data.

Acute toxicity
Male mice of the ddy strain (25-26 g) were purchased from Nihon SLC Co. (Hamamatsu, Japan).The mice were housed in groups of 10 per cage (30 × 30 × 16 cm), and kept in an airconditioned room (22 ± 2 o C and 55.5 ± 5% relative humidity) with 12 h light cycle, and allowed to feed (F-2 obtained from Funabashi Farm Co., Funabashi, Japan) and drink water ad libitum.Samples were suspended in physiological saline containing 0.5% Tween 80 and intraperitoneally administered in geometrically increasing doses from 0 to 100% lethal dose (10 mL/kg body weight) 5-10 mice were used for each dose.The LD 50 values were estimated according to the Litchfield-Wilcoxon method. 10

Conclusions
Nicotine-N-oxides were successfully synthesized using m-CPBA and the transition state energies of the reaction intermediates were investigated by an ab-initio MO method.Consequently, it was found that oxidation occurred first at the N' position, followed by the N position both in theory and practice.The asymmetric reduction of nicotine-N-oxides with the rat liver S-9 fraction showed that the aliphatic N-oxide was reduced faster than the aromatic one, and cis-nicotine-N'oxide (5b) was easier to reduce than trans-nicotine-N'-oxide (5a).These results suggest that even if cis-nicotine-N'-oxide (5b) is produced in vivo, the reduction-oxidation process should exclusively result in the urinary excretion of trans-nicotine-N'-oxide (5a).The acute toxicities of the nicotine-N-oxides were also investigated and suggest a reason of oxidation of nicotine (1).
Although the mechanism of the oxidation of the nicotine in mammalian metabolic system has been previous reported, little is known about the reduction of nicotine-N-oxides.The present study sheds light on the reductive properties of the mammalian reductive system.Further studies are currently underway in our laboratory, including a detailed mechanistic study of the metabolism of these compounds.

Table 2 .
Acute toxicity of the nicotine N-oxides