Comparative Study with Modified Phenobarbitals with Different Functional Groups
Abstract
To study the specific target to which phenobarbital (PB) binds, resulting in the induction of cytochrome P450, we prepared two azido-PBs (AZPBs) as photoaffinity ligands. The azido substituent was introduced at the para- or meta-position of the PB aromatic ring. In this study, we estimated the utility of these compounds by examining their inducing activities in vivo in rats. Induction was assessed by immunoblotting with anti-CYP2B1/2 antibody and measuring testosterone-metabolizing activity, using hepatic microsomes. Administration of p-AZPB to rats increased hepatic CYP2B1/2 protein and testosterone 16β-hydroxylase activity, although the effects were less than those of unmodified PB.m-AZPB showed no effect in the induction of CYP2B1/2. To assess the specificity of the effects of substituents, we compared the inducing activities of p/m-nitro-PBs,p/m-amino-PBs, andp/m-hydroxy-PBs with those of AZPBs. The results showed that p-nitro-PB, m-amino-PB, andp-hydroxy-PB were also potent inducers for CYP2B1/2, with lower activity than that of unmodified PB, whereas the other three isomers had no effect. These results suggest that 1) the absence of any substituents on the aromatic ring of PB is needed for maximal inducing activity and 2) substitution at the meta-position of the PB aromatic ring tends to reduce effectiveness as an inducer more than does substitution at the para-position. Becausep-amino-PB and p-acetylamino-PB, the minor and major metabolites of p-AZPB, respectively, were without effect in the induction of CYP2B1/2, the effect of p-AZPB was considered to be due to the unchanged compound itself. The present study demonstrates that, based on the weak but positive ability to induce CYP2B1/2, p-AZPB may be a useful tool for identifying the putative PB receptor.
As is widely known, PB1 induces a number of liver proteins, including P450, epoxide hydrolase, UDP-glucuronosyltransferase, and glutathione S-transferase (1). Of the several forms of P450 that are inducible by PB, the CYP2B subfamily is increased to the greatest degree, at least in rats (2). The CYP2B isozymes catalyze the activation and/or inactivation of many compounds, including drugs, pesticides, and carcinogens (3, 4). The activities of a series of barbiturates in CYP2B induction and their potencies as liver tumor promoters are correlated (5). Induction of the CYP2B subfamily is, therefore, important from a toxicological point of view.
The PB-mediated increase in the CYP2B subfamily was first observed three decades ago (6, 7), but the mechanism remains largely unknown, despite many investigations. A number of recent studies focused on the interaction between the region 5′-upstream of the gene and transcription regulatory factors. These studies have provided evidence for identification and/or localization of barbiturate-responsive DNA elements that regulate gene transcription and for determination of the presence of trans-acting factors that bind to the aforementioned elements (8-14). However, there remains a fundamental question, namely, “What is the initial site capable of interacting with PB?” Receptor-dependent induction has been considered as one of the possible mechanisms by which PB activates gene transcription (1). However, the validity of the consideration remains to be clarified, and identification of the receptor has not yet been achieved. To address this problem, we prepared photoaffinity PBs in which aromatic rings were substituted by an azido group at the para- ormeta-position. In this study, we estimated the utility of these compounds by examining their inducing activities for hepatic CYP2B1/2. In addition, to determine whether the effect is specific for azido substituents, we compared the inducing effects of modified PBs containing a nitro, amino, hydroxy, or azido substituent on the phenyl group (fig. 1).
Materials and Methods
Chemicals.
The following chemicals were purchased from the sources indicated: sodium PB, palladium carbon (5%), and ammonium sulfamate, Wako Pure Chemical Industries (Osaka, Japan); sodium azide, sodium nitrite, testosterone, polyoxyethylene sorbitan monolaurate (Tween 20), and 4-androstene-3,17-dione, Nacalai Tesque Co. (Kyoto, Japan); 6β-, 7α-, 16α-, and 16β-hydroxytestosterone, Steraloids Inc. (Wilton, NH); nitronium tetrafluoroborate (85%), Aldrich. 2α-Hydroxytestosterone was kindly donated by Dr. M. Nakano (Shionogi Pharmaceutical Co., Osaka, Japan). Rabbit anti-CYP2B1/2 IgG was prepared by a method described previously (15). Recognition of CYP2B1 and CYP2B2 by this antibody was confirmed using purified isozymes. The antibody also recognized purified CYP2B3 (16). All other materials were of the highest quality commercially available.
Synthesis of Modified PBs.
p-NTPB, m-NTPB, p-AMPB, andm-AMPB were synthesized by the method of Bright et al. (17), with minor modification. p-OHPB andm-OHPB were prepared according to the method of Butler (18).p-AAPB was obtained by acetylating p-AMPB with acetic anhydride. The reaction was carried out at room temperature for 10 min in the absence of pyridine (19). The purity of these compounds was ascertained by measuring the melting points, mass spectra, and1H-NMR spectra (not shown).
A new compound, m-AZPB, was synthesized as described below.m-AMPB (120 mg) was dissolved in 16.8 ml of 1.2 M HCl and placed on ice. To this solution 0.7 ml of 1.04 M sodium nitrite was slowly added over 30 min, and the mixture was stirred for an additional 1 hr. After inactivation of the excess nitrite with 0.12 ml of 8 M urea, 0.24 ml of 4.15 M sodium azide was added. The reaction solution was stirred in the dark for 2 hr at room temperature, and the precipitate formed was removed. This was washed with 10% HCl and then water and was dried in vacuo. The crude product was purified by silica gel column chromatography (1.5 × 10 cm, chloroform/methanol, 9:1, v/v) to give a white powder [yield, 124 mg (94%); m.p. 176–178°C (uncorrected); analysis, calculated for C12H11N5O3: C, 52.74; H, 4.06; N, 25.63; found: C, 52.70; H, 4.08; N, 25.45; EI-MS,m/z 273 (M+, 20%), 245 (100%), 216 (44%), 202 (31%); 1H-NMR (500 MHz, CDCl3), δ 7.91 (bs, 2H), 7.36 (t,J = 8.0 Hz, 1H), 7.11 (d, J = 0.9 Hz, 1H), 7.05 (d, J = 0.9 Hz, 1H), 7.04 (m, 1H), 2.47 (dd,J = 14.7, 7.3 Hz, 2H), 1.01 (t, J = 7.3 Hz, 3H)].
p-AZPB, another new compound, was prepared fromp-AMPB in a manner similar to that described above [yield, 118 mg (89%); m.p. 180–182°C (uncorrected); analysis, calculated for C12H11N5O3: C, 52.74; H, 4.06; N, 25.63; found: C, 52.70; H, 4.09; N, 25.20; EI-MS,m/z 273 (M+, 20%), 245 (100%), 216 (44%); 1H-NMR (500 MHz, CDCl3), δ 7.93 (bs, 2H), 7.35 (dd,J = 11.9, 2.3 Hz, 2H), 7.03 (dd, J = 11.9, 2.3 Hz, 2H), 2.45 (dd, J = 14.7, 7.3 Hz, 2H), 1.00 (t, J = 7.3 Hz, 3H)].
Animal Treatment and Preparation of Liver Microsomes.
Male Sprague-Dawley rats weighing 78–80 g were purchased from Charles River, Inc. (Yokohama, Japan). Four or five animals were used in each treatment group described below. They were acclimatized for 1 week and allowed free access to food (CE-2; Nippon Clea. Inc., Tokyo, Japan) and water before the start of the treatment. Rats were injected ip with PB and its derivatives, at a dose of 80 mg/kg/2 ml 20% Tween 20, daily for 4 days. Control animals were given drug-free Tween 20 solution (2 ml/kg body weight). After treatment, rats were fasted overnight and then their livers were removed. Preparation of hepatic microsomes was performed as described elsewhere (20).
Assays.
Testosterone hydroxylase activity was determined by HPLC analysis in which a Hitachi L-7100 pump and 7420 UV-visible detector were used. Incubation of testosterone and liver microsomes and extraction of metabolites were carried out by the method described elsewhere (21). The extracts were dissolved in 200 μl of methanol, and the aliquot (20 μl) was injected into the HPLC apparatus. Operating conditions for HPLC were virtually the same as those reported previously (21), but modified as follows: column, TOSOH TSK gel ODS-80TM (21.5 mm i.d. × 30 cm) with a guard column (15 × 3.2 mm) containing the same gel as the main column; column temperature, 40°C; flow rate, 0.7 ml/min; mobile phase, linear gradient from solvent A (methanol-acetonitrile-water = 36:2:62, v/v/v) to solvent B (methanol-acetonitrile-water = 64:6:30, v/v/v) (0–40 min), solvent B (40–50 min), change from solvent B to solvent A (50–51 min), and solvent A (51–61 min); detection, UV at 254 nm. Under these conditions, the retention times of 7α-, 6β-, 16α-, 16β-, and 2α-hydroxytestosterone and 4-androstene-3,17-dione were 25, 26, 29, 32, 34, and 40 min, respectively. The calibration curves for six testosterone metabolites were liner, ranging between 0.65 and 3 nmol/incubation mixture. The content of microsomal P450 was measured by an established method (22). Significant differences were analyzed by using unpaired Student’s t tests. Protein was determined by the method of Lowry et al. (23), using bovine serum albumin as a standard.
Immnoblotting.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (7% acrylamide) of microsomal proteins was performed according to the procedure of Laemmli (24). After electrophoresis, the protein bands were transferred to a polyvinylidene difluoride membrane and immunostained with rabbit anti-CYP2B1/2 antibody using reported methods (25, 26). Briefly, CYP2B1/2-antibody complex was treated with anti-rabbit IgG antibody conjugated with peroxidase and was visualized with the peroxidase reaction, using 3,3′-diaminobenzidine as a substrate.
Results and Discussion
The effects of eight modified PBs on the hepatic content of P450 and activity of testosterone metabolism are shown in fig.2. The activity of testosterone 16β-hydroxylation is one of the useful markers for the CYP2B P450 subfamily. p-AZPB significantly increased this activity, whereas m-AZPB showed no effect (fig. 2A). This result suggested that only the para-isomer of the two AZPBs is a potent inducer of the CYP2B subfamily, although the effect was less than that of a reference inducer, PB (fig. 2). Of other modified PBs examined, p-NTPB (fig. 2B), m-AMPB (fig. 2C), and p-OHPB (fig. 2D) significantly increased 16β-hydroxylase activity. p-AZPB and p-OHPB showed a greater ability to induce this activity than did the other two derivatives. The activity of 16α-hydroxylase was also increased by the compounds capable of increasing 16β activity, except for p-NTPB, but the induction of 16β-hydroxylase was more pronounced than that of 16α-hydroxylase. Two modified PBs, m-AZPB and p-AMPB, increased testosterone 16α- but not 16β-hydroxylase activity (fig. 2,A and C). The former reaction is catalyzed by not only CYP2B1 but also CYP2C11 (27-29). The enzyme catalyzes the 2α-hydroxylation of testosterone as well as the 16α-hydroxylation. Consistent with this, both m-AZPB and p-AMPB increased testosterone 2α-hydroxylase activity. From these data, it seems that these two derivatives are specific inducers of CYP2C11.p-NTPB enhanced only testosterone 16β-hydroxylase activity. This was thought to be due to inhibition of and/or a reduction in CYP2C11, because testosterone 2α-hydroxylase activity was reduced to 25% of the control level by p-NTPB treatment (fig. 2B). The same was also seen with p-AZPB treatment (fig. 2A). The modified PBs tested here exhibited no effect or only a slight effect on the activity of testosterone 7α- and 6β-hydroxylation, the markers for CYP2A and CYP3A, respectively.
When the hepatic microsomes were electrophoresed and immunostained with anti-CYP2B1/2 antibody, increased levels of CYP2B1 and CYP2B2 proteins were detected in rats treated with p-AZPB,p-NTPB, m-AMPB, or p-OHPB (fig.3). These results, together with the increased P450 content and activity of testosterone metabolism, strongly suggested that these four modified PBs are the inducers of the CYP2B subfamily. However, it should be noted that there is partial discordance between the metabolism and immunoblot data presented; whereas p-AZPB increased testosterone 16β-hydroxylase activity to more than half the level of PB-treated rats, immunoblot bands of CYP2B1/2 showed much greater intensity in PB-treated rats than in modified PB-treated animals. The reason for this inconsistency is unknown. Conceivably, p-AZPB might modify the CYP2B1/2in vivo, resulting in reduction of the affinity for the specific antibody used in immunoblotting. A band with greater mobility than CYP2B1/2 was seen in all immunoblots and was thought to be CYP2B3, a constitutive form (16, 30).
Because the azido group is labile, it is conceivable that a metabolite or degradation product of p-AZPB, but not the unchanged drug, contributes to the induction of CYP2B1/2 described above. To address this possibility, we examined in vivo metabolism ofp-AZPB and assessed the inducing activity of metabolites. Urinary extracts from rats given a single ip injection ofp-AZPB (80 mg/kg) were analyzed by HPLC, and a major metabolite was purified to measure the EI-MS spectrum. The results indicated that the major and minor metabolites of p-AZPB in rat urine were p-AAPB and p-AMPB, respectively (data not shown). The proposed metabolic pathway of p-AZPB is schematically shown in fig. 4.p-AAPB was administered ip to rats at a dose of 80 mg/kg for 4 days, and the effects on hepatic testosterone-metabolizing activity were compared with those of p-AZPB and p-AMPB (data not shown). As already described (fig. 2), p-AZPB enhanced 16α/β-hydroxylation. However, no increase in this activity was observed after administration of p-AAPB, similarly top-AMPB (fig. 2C). An immunoblot analysis with anti-CYP2B1/2 antibody showed immunolabeled bands of CYP2B1 and CYP2B2 proteins (fig. 5, arrows) in microsomes of rats treated with p-AZPB, but no, or faint, bands were observed in microsomes from p-AAPB-treated rats. These results strongly suggest that the metabolites ofp-AZPB have no ability or only a very weak ability to induce CYP2B.
This study indicates that substitution with different functional groups on the PB aromatic ring causes reduced activity of CYP2B induction. That is, p/m-azido,p/m-nitro, p/m-amino, andp/m-hydroxy substituents reduced or diminished the inducing activity. It is, therefore, suggested that the phenyl group without these substitutions is needed for maximal induction of the CYP2B subfamily. The reduced activity of modified PBs may be due to a change in pharmacokinetics. The observation that the degree of induction by barbiturates is correlated with the plasma half-life (31) seems to support this view. However, as reported here, the inducing activities of different isomers, i.e. p-AZPB andm-AZPB, were quite different. It is conceivable that the reduced activity of modified PBs is due to a reason apart from altered pharmacokinetics, such as a change in affinity for the putative receptor. The polarity of substituents introduced into PB would alter the lipophilicity. However, this effect is suggested to be minor insofar as the efficacy of induction is concerned, becausep-OHPB, a relatively polar compound, was a more potent inducer than p/m-NTPB, lipophilic compounds. This was also reported in another study; secobarbital (32, 33) and thiopental (31, 33), highly lipophilic compounds, are weaker inducers than PB in rats. These observations support the view that the induction of the CYP2B subfamily by barbiturates is independent of the lipophilicity of the latter (33). Except for m-AMPB, the compounds tested here that are capable of inducing CYP2B1/2 have substituents at the para-position of the phenyl ring. This may mean that the absence of a substituent at themeta-position is more important than that at thepara-position as far as inducing activity is concerned. The reason why m-AMPB did not follow this trend is unknown, but it may be that the basic features of the substituent as well as the substitution position play a role.
Our unpublished examination indicated that p-AZPB andm-AZPB modify bovine serum albumin to similar extents under irradiation with UV light. Furthermore, the time course and the products of photodegradation of p/m-AZPBs were also the same for the compounds. In spite of the common characteristics, their abilities to increase the CYP2B subfamily were quite different, as reported here. From these findings,p-AZPB, a possible photoaffinity ligand, is expected to be a useful probe for searching for the “PB receptor,” although its presence has not yet been established. This is now being investigated in our laboratory.
Footnotes
-
Send reprint requests to: Kazuta Oguri, Ph.D., Faculty of Pharmaceutical Sciences, Kyushu University 62, 3–1-1 Maidashi, Higashi-ku, Fukuoka 812–82, Japan.
-
Received March 11, 1997; accepted August 28, 1997.
-
↵1 Abbreviations used are: PB, phenobarbital; P450 or CYP, cytochrome P450; AZPB, azidophenobarbital; NTPB, nitrophenobarbital; AMPB, aminophenobarbital; OHPB, hydroxyphenobarbital; p-AAPB,p-(acetylamino)phenobarbital.
- The American Society for Pharmacology and Experimental Therapeutics