Toxicological Consequences of Differential Regulation of Cytochrome P450 Isoforms in Rat Brain Regions by Phenobarbital

Archives of Biochemistry and Biophysics Vol. 399, No. 1, March 1, pp. 56 – 65, 2002 doi:10.1006/abbi.2001.2727, available online at http://www.idealibrary.com on

Toxicological Consequences of Differential Regulation of Cytochrome P450 Isoforms in Rat Brain Regions by Phenobarbital 1 Sudarshan C. Upadhya,* Shankar J. Chinta,† Harish V. Pai,* Michael R. Boyd,‡ and Vijayalakshmi Ravindranath* ,† ,2 *National Brain Research Centre, ICGEB Campus, Aruna Asaf Ali Marg, New Delhi 110 067, India; †Department of Neurochemistry, National Institute of Mental Health & Neurosciences, Hosur Road, Bangalore 560 029, India; and ‡Molecular Targets Drug Discovery Program, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Frederick, Maryland 21702-1201

Received October 5, 2001; published online February 7, 2002

Cytochrome P4502B is an isoform of cytochrome P450 (P450) that is induced by the anticonvulsant drug phenobarbital. Here, we demonstrate the constitutive expression and predominant localization of CYP2B in neurons of rat brain. Administration of phenobarbital to rats resulted in selective induction of P450 levels in cortex and midbrain, while other regions were unaffected. Immunohistochemical localization of P4502B in brains of phenobarbital treated rats revealed localization of P4502B in neuronal cells, most predominantly the reticular neurons in midbrain. The anticancer agent 9-methoxy-N 2-methylellipticinium acetate (MMEA) has been shown to exhibit preferential neuronal toxicity in vitro. Pretreatment of rats with phenobarbital potentiated the toxicity of intrathecally administered MMEA in vivo, as seen by the degeneration of reticular neurons. Thus, induction of P450 in selective regions of brain by phenobarbital would profoundly influence xenobiotic metabolism in these regions, especially in clinical situations where phenobarbital is coadministered with other psychoactive drugs/xenobiotics. © 2002 Elsevier Science (USA) Key Words: brain; drug metabolism; cytochrome P450; psychoactive drugs; monooxygenase; neurotoxicity.

Cytochrome P450 (EC 1.14.14.1; P450) 3 and associated monooxygenases constitute an important family of hemeproteins that are involved in the metabolism of xenobiotics (including drugs) and certain endogenous compounds. Multiple forms of P450, which are selectively induced or inhibited by a variety of drugs, are known to exist in liver, the major organ involved in P450 mediated metabolism (1). In recent years, P450mediated metabolism in extrahepatic organs (such as lung, kidney, skin, nasal epithelium) and the potentially far-reaching consequences of such metabolism, in situ, within specific cells in target organs have been recognized in laboratory animals (2) and humans (3). Over the last decade, our laboratory has demonstrated the presence of P450 enzymes in rat and human brain, and the capability of the enzyme to metabolize a variety of xenobiotic substrates (4). We have demonstrated the constitutive presence of several forms of P450 including 1A1/1A2, 2B1/2B2, and P4502E in rat brain (4, 5). Since there is much regional and cellular heterogeneity within the brain, it is important to know the localization of specific forms of P450 within the brain. This will further enable an understanding of the potential consequences of metabolism of xenobiotics mediated by specific isoforms of P450. Most of the earlier studies relating to the localization of the multiple forms of P450 in brain were performed using immunohistochemistry (5– 8). However, several concerns have been raised about the use

1

This research was supported by National Institutes of Health Grant MH55494. 2 To whom correspondence should be addressed at National Brain Research Centre, ICGEB Campus, Aruna Asaf Ali Marg, New Delhi 110 067, India. Fax: 91 124 622 0237. E-mail: [email protected]. 56

3

Abbreviations used: P450, cytochrome P450; CNS, central nervous system; MMEA, 9-methoxy-N 2 -methylellipticinium acetate; PBS, phosphate-buffered saline. 0003-9861/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.

DIFFERENTIAL REGULATION OF RAT BRAIN CYTOCHROME P450 BY PHENOBARBITAL

of antibodies from different sources for determining the constitutive expression of isoforms of P450, particularly CYP2B, in brain (9). These concerns have generally been related to issues regarding the specificity of the antibody and the need for controls while interpreting results from immunohistochemistry. Localization of mRNA by in situ hybridization offers an excellent alternative to address this issue. We therefore studied the constitutive expression of CYP2B mRNA in rat brain by Northern blot analysis and have localized the transcript in the CNS by fluorescence in situ hybridization. Phenobarbital is an anticonvulsant that is used extensively for seizure control, often for prolonged periods and in combination with other drugs. Since brain exhibits considerable regional and cellular heterogeneity, the effect of phenobarbital administration on P450 levels in different brain regions was studied with a view toward understanding the toxicological and pharmacological consequences of long-term administration of phenobarbital. 9-Methoxy-N 2 -methylellipticinium acetate (MMEA), a potential anticancer drug, is metabolized by CYP2B1 to its active form 9-hydroxy-N-methylellipticinium, which selectively damages neurons (10, 11). We therefore examined the effect of phenobarbital pretreatment on MMEA toxicity in midbrain, a region where P450 levels are maximally induced by phenobarbital. MATERIALS AND METHODS Materials. CYP2B1 cDNAs were obtained as a gift from Dr. M. Adesnik and Dr. N. G. Avadhani. The DIG RNA Labeling and Detection Kit and antidigoxigenin Fab fragments linked to peroxidase were purchased from Boehringer Mannheim, USA. The Tyramide Signal Amplification (Indirect) Kit for in situ hybridization was obtained from New England Nuclear (Boston, MA). All other chemicals and reagents were of analytical grade and were obtained from Sigma Chemical Company (St. Louis, MO) or Qualigens, India. The antiserum to phenobarbital-inducible rat liver P450 (P4502B) was generated as described (5). Animals. Male Wistar rats (3– 4 months old, 200 –250 g) were obtained from the Central Animal Research Facility of the National Institute of Mental Health and Neurosciences. Animals had access to a pelleted diet (Lipton India, Calcutta) and water ad libitum. All animal experiments were carried out according to National Institutes of Health guidelines for the care and use of laboratory animals. All efforts were made to minimize animal suffering, to reduce the number of animals used, and to use alternatives to in vivo techniques, if available. Northern analysis and in situ hybridization. cDNA to CYP2B1 provided by Dr. M. Adesnik (12) was subcloned into pBluescript II SK⫹ (Stratagene, La Jolla, CA) and used for preparation of riboprobes, while the cDNA obtained from Dr. Avadhani in pGEM-7Zf(⫹) (Promega, Madison, WI) was used without further subcloning. The Northern blot analyses and the fluorescence in situ hybridization were carried out using the cDNA obtained from both of the above sources, and gave similar results. The figures depict the results obtained using the cDNA provided by Dr. Avadhani. Total RNA from rat brain and liver was extracted as described by Chomezynski (13) and poly(A) ⫹ RNA was isolated from total RNA using oligo(dT)-

57

cellulose chromatography. Both total RNA and poly(A) ⫹ RNA were separated electrophoretically and transferred onto positively charged nylon membranes by capillary transfer (14). After UV crosslinking, the membranes were hybridized overnight at 55°C with digoxigenin-labeled antisense riboprobe prepared using the cDNA to CYP2B1, washed, incubated with antibody to digoxigenin Fab fragments and conjugated with alkaline phosphatase, and the bands were visualized using nitroblue tetrazolium salt and 5-bromo-4chloro-3-indolyl phosphate as the chromogenic substrates for alkaline phosphatase. Rats were anesthetized with ether and perfused transcardially with 200 ml of phosphate-buffered saline followed by 200 ml of 4% paraformaldehyde solution (w/v) prior to removal of brain and spinal cord. The tissue was processed for paraffin embedding, and serial sections (8 –10 ␮m thick) were cut in the coronal plane under RNasefree conditions. Sections were dewaxed, hydrated in graded ethanol, acetylated, and treated with proteinase K. Sections were then rinsed in PBS and dehydrated using graded ethanol. Digoxigenin-labeled sense (for control sections) and antisense cRNA probes were synthesized from cDNA to CYP2B1 using T3 and T7 RNA polymerases for CYP2B1 cDNA in pBluescript and SP6 and T7 polymerases for the CYP2B1 cDNA in pGEM7Zf(⫹) vector. Sections were hybridized overnight at 45°C with the sense or antisense probes. After hybridization, sections were washed, blocked with 0.5% bovine serum albumin (w/v, NEN Life Sciences Products, USA), and incubated with antibody to digoxigenin Fab fragments conjugated to horseradish peroxidase. After washing, the sections were incubated with biotinylated tyramide (NEN Life Sciences Products, USA) followed by FITC-labeled streptavadin. Finally the sections were washed, dried, and mounted prior to examination under the fluorescence microscope. Treatment with phenobarbital. Male Wistar rats (3– 4 months old) were administered phenobarbital (80 mg/kg body weight, ip in saline, daily) for 10 days. Control rats were given saline. Animals were sacrificed 24 h after the last dose. This treatment protocol was followed for the estimation of total P450, reductase, immunoblot analysis, and immunohistochemistry. For Northern blot studies, rat liver total RNA was prepared 24 h after a single dose of phenobarbital. Rat brain mRNA was prepared after phenobarbital administration for 1, 2, and 3 days, respectively. Preparation of microsomes. Animals were anesthetized with ether and perfused transcardially with ice-cold Tris buffer (100 mM, pH 7.4) containing potassium chloride (1.15%, w/v) prior to decapitation and removal of the brain. For certain experiments, brain regions (cortex, cerebellum, hippocampus, olfactory, striatum, thalamus, and brain stem) were dissected using standard anatomical landmarks as described (15). Brain regions were pooled from 5–10 rats and homogenized using a Potter–Elvehjem homogenizer in 9 vol of ice-cold Tris buffer (0.1 M, pH 7.4) containing dithiothreitol (0.1 mM), EDTA (0.1 mM), potassium chloride (1.15%, w/v), phenylmethylsulfonyl fluoride (0.1 mM), butylated hydroxytoluene (22 ␮M), and glycerol (20%, v/v), previously bubbled with nitrogen (buffer A). The homogenate was centrifuged at 17,000g for 30 min. Thereafter, the supernatant was centrifuged at 100,000g for 1 h to give the microsomal pellet (16). The pellet was suspended in a small volume of buffer A, aliquoted, flash-frozen in liquid nitrogen, and stored at ⫺70°C. Protein concentration was measured by a dye-binding method (17). Estimation of P450 and reductase. The total P450 content was measured from the carbon monoxide reduced minus oxidized difference spectrum (18). NADPH cytochrome c reductase activity was determined according to Phillips and Langdon (19) with modifications as described by Guengerich (20). Immunoblotting studies. Microsomal protein from rat brain regions prepared from control and phenobarbital-treated rats were subjected to sodium dodecyl sulfate–polyacrylamide gel electro-

58

UPADHYA ET AL.

phoresis (21) and the separated proteins were electroblotted onto nitrocellulose membrane (22). The membrane was immunostained with the antiserum to purified rat liver P4502B1/2 (4), followed by incubation with anti-rabbit IgG conjugated to alkaline phosphatase (Vector Laboratories, USA). The immunostained bands were visualized using nitroblue tetrazolium salt and 5-bromo-4-chloro-3-indolyl phosphate. Immunohistochemical localization of P4502B in phenobarbital treated rat brain. Rats treated with phenobarbital for 10 days were anesthetized with ether and perfused transcardially with 200 ml of phosphate-buffered saline followed by 200 ml of 4% paraformaldehyde solution (w/v) prior to removal of brain. Serial sections (20 ␮m thick) were cut on a coronal plane using a vibratome and washed in phosphate buffered saline. Endogenous peroxidase activity was blocked using methanol containing hydrogen peroxide (3%, v/v), followed by incubation in normal goat serum. The sections were incubated in antiserum to P4502B diluted in normal goat serum. The sections were washed and incubated with biotinylated anti-rabbit IgG solution for 30 min at room temperature. After washing, the sections were incubated with Vectastain-Elite ABC reagent for 30 min at room temperature. The sections were then washed and the color was developed using diaminobenzidine containing hydrogen peroxide. Treatment of rats with MMEA. Male Wistar rats 2–3 months old were injected intrathecally with 5␮g of MMEA in 0.1 ml of normal saline. Control rats were given vehicle alone. Animals were sacrificed 7 days after MMEA treatment. To study the effect of phenobarbital induction, one group of rats were administered phenobarbital (80 mg/ kg body wt, ip, daily) for 7 days prior to MMEA injection and phenobarbital treatment was continued for 7 days after MMEA treatment to maintain the induction of P450. Prior to sacrifice, rats were anesthetized with ether and perfused transcardially with 200 ml of phosphate buffered saline followed by 200 ml of 4% buffered paraformaldehyde solution (w/v) prior to removal of brain. The brain was removed and left overnight in the fixative. Serial sections of 20-␮m thickness were cut in the coronal plane at the midbrain level using a vibratome. The sections were washed with phosphate-buffered saline and stained with 0.1% cresyl violet for 15 min at 60°C. Sections were washed with water, dehydrated with graded alcohol, cleared with xylene, air-dried, and mounted using Permount.

RESULTS

Effect of phenobarbital treatment on P4502B. Cytochrome P450 content was differentially induced in rat brain regions following phenobarbital administration. Maximal induction of total P450 content was observed in brain stem (1.7-fold induction) followed by cortex (1.3-fold), while P450 content was not significantly elevated in hippocampus, cerebellum, striatum, and thalamus (Fig. 2A). The reduced carbon monoxide binding spectra of brain microsomes from control and phenobarbital-treated rats are shown in Figs. 1A and 1B. The distinctive absorption maximum at 450 nm is seen predominantly in microsomes from phenobarbital treated rat brain cortex. NADPH cytochrome c reductase activity was induced nearly twofold in all the regions of phenobarbital-treated rat brain. NADPH cytochrome c reductase activity did not show any regional variation either in constitutive activity or in degree of induction following phenobarbital administration (Fig. 2B).

FIG. 1. Dithionite reduced carbon monoxide binding spectra of cortical microsomes from (A) control and (B) phenobarbital-treated rats. (A) The microsomal protein concentration was 0.41 mg/ml and the P450 content was calculated to be 61 pmol/mg protein. (B) The microsomal protein concentration was 0.35 mg/ml and the P450 content was calculated to be 77 pmol/mg protein.

Northern blot analysis of CYP2B expression in rat brain. Northern blot analysis of total RNA from rat liver and poly(A) ⫹ RNA from rat brain cortex using the cDNA to CYP2B1 revealed the constitutive expression of CYP2B mRNA in rat brain. The molecular mass of the transcript was approximately 1.6 kb which was similar to that seen in rat liver (Fig. 3A). The band hybridizing to CYP2B was absent in the Northern blots hybridized with the sense riboprobe (Fig. 3B). Localization of CYP2B mRNA in rat brain by fluorescence in situ hybridization. Fluorescence in situ hybridization (FISH) studies demonstrated the presence of CYP2B mRNA predominantly in neuronal cells in rat brain regions. High levels of CY2B mRNA were seen in the olfactory bulb, cerebral cortex, and mid-

DIFFERENTIAL REGULATION OF RAT BRAIN CYTOCHROME P450 BY PHENOBARBITAL

59

The anterior horn cells of the spinal cord were intensely fluorescent, indicating the presence of CYP2B mRNA (Fig. 6D), while the control section hybridized with the sense probe did not show any staining (Fig. 5C). Regulation of CYP2B expression by phenobarbital. The expression of CYP2B was studied by Northern blot analysis using total RNA from liver and poly(A) ⫹ RNA from rat brain cerebral hemisphere and cerebellum from vehicle and phenobarbital-treated rats. CYP2B1 mRNA levels were upregulated in liver 24 h after a single dose of 80 mg/kg body wt phenobarbital, while CYP2B1 mRNA level was upregulated in cerebral hemisphere only after four daily doses of phenobarbital over 96 h and no change was observed in the cerebellum at any of the time points examined (Figs. 7A and 7B).

FIG. 2. Microsomal P450 content and NADPH cytochrome c reductase activity in brain regions from control and phenobarbital treated rats. P450 content (A) and NADPH cytochrome P450 reductase activity (B) were measured in microsomes prepared from brain regions of vehicle-treated (filled bars) and phenobarbital-treated (empty bars) rats. The values are means ⫾ SD (n ⫽ 3 batches of microsomes prepared from pooled brains of 10 rats). Values significantly different from controls are indicated by asterisks (P ⬍ 0.05).

brain. The neuronal cells in the cerebral cortex showed intense cytosolic staining, indicating the presence of the CYP2B mRNA (Fig. 4B), while the sections hybridized with the sense probe showed no fluorescence (Fig. 4A). The laminar architecture of the different cortical layers was clearly discernible. In the cerebellum, the Purkinje cells and the granule cell layer showed intense fluorescence, while the interneurons of the molecular layer were relatively less intensely stained (Fig. 4D). In the midbrain, the reticular neurons were intensely labeled, indicating the predominant presence of CYP2B mRNA in these cell populations (Fig. 4F). In the hippocampus the granule cells of the dentate gyrus were labeled (Fig. 5D) and the pyramidal cells of CA1, CA2, and CA3 were also labeled. There was only sparse labeling of neurons in the striatum (Fig. 5B). In the olfactory bulb, the neuronal cells in the olfactory nucleus and glomeruli were intensely labeled (Fig 6B).

Immunoblot analysis of P4502B in brain regions from control and phenobarbital-treated rats. Immunoblot analysis was carried out using microsomes from untreated and phenobarbital-treated rat brain regions, and immunostained with antiserum to rat liver P4502B1/2B2 which revealed the constitutive expression of P4502B in all the brain regions examined (Fig. 8). In the cortex and midbrain the immunostained band was highly intense in the microsomes from phenobarbital-treated rat brain as compared with vehicletreated controls. No induction was seen in the immunostained band from other brain regions after phenobarbital treatment.

FIG. 3. Northern blot analysis of rat brain RNA using the cDNA to CYP2B1. (A) Total RNA from liver (12 ␮g, lane 1) and poly(A) ⫹ RNA from rat brain cortex (3 ␮g, lane 2) were subjected to electrophoresis, transferred to nylon membranes and hybridized with digoxigeninlabeled antisense riboprobe to CYP2B. (B) Lane 1: Poly(A) ⫹ RNA from rat brain cortex (3 ␮g) was hybridized with the sense riboprobe to CYP2B. The relative mobilities of the 18S and 28S RNAs are indicated.

60

UPADHYA ET AL.

FIG. 4. Localization of CYP2B mRNA in rat brain cortex, cerebellum, and midbrain using fluorescence in situ hybridization. The presence of CYP2B mRNA in the neurons of cerebral cortex is depicted in (B). The control section hybridized with the sense probe did not reveal any fluorescence in (A). Bar ⫽ 60 ␮m. (D) Intense fluorescent labeling of Purkinje cells (arrow) was seen in the rat cerebellum. Staining in the granule cell layer (double arrow) was less intense. Bar ⫽ 120 ␮m. Inset: Higher magnification of Purkinje cells. Bar ⫽ 60 ␮m. (C) The control section of rat cerebellum hybridized with the sense probe did not show any fluorescence. Bar ⫽ 240 ␮m. (F) The reticular neurons (arrowhead) in the midbrain expressed the CYP2B mRNA. Bar ⫽ 240 ␮m. Inset: Higher magnification of a giant reticular neuron. Bar ⫽ 60 ␮m. (E) Control section of midbrain hybridized with the digoxigenin-labeled sense probe. Bar ⫽ 240 ␮m.

Regulation of CYP2D and CYP3A by phenobarbital. Since phenobarbital has been shown to upregulate CYP3A in addition to CYP2B, we examined the effect of phenobarbital on the expression of this isoform in brain by Northern and immunoblotting (Fig. 9). Further, since CYP2D is present in relatively larger amounts in brain, we also examined the effect of phenobarbital on CYP2D by Northern and immunoblotting. While the expression of CYP2D and CYP3A was induced in rat liver after a single dose of phenobarbital, no change in the expression of these isoforms was seen in the brain even after four doses of phenobarbital given over 96 h (Fig. 9A). Immunoblotting experiments were in agreement with the above and the expression of the CYP2D and CYP3A protein levels was unaltered even after prolonged treatment with phenobarbital for 10 days (Fig. 9B). Localization of P4502B in phenobarbital treated rat brain by immunohistochemistry. Immunohistochemical studies demonstrated the presence of P4502B protein predominantly in neuronal cells in brain sections

FIG. 5. Localization of CYP2B mRNA in rat brain using fluorescence in situ hybridization in striatum and hippocampus of rat brain (B) Constitutive expression of CYP2B was seen in the neurons of the striatum (arrow) although the intensity of staining was less. Bar ⫽ 120 ␮m. (A) Control section of midbrain hybridized with the digoxigenin labeled sense probe. Bar ⫽ 240 ␮m. (D) Intense fluorescence was seen the granule cell layer of the dentate gyrus (arrow). Staining of the interneurons of the hilius (two arrowheads) was also observed. Bar ⫽ 120 ␮m. (C) The control section hybridized with the sense probe did not have any fluorescent staining. Bar ⫽ 240 ␮m.

FIG. 6. Localization of CYP2B mRNA in spinal cord and olfactory bulb of rat using fluorescence in situ hybridization. (B) In situ hybridization of section from rat brain olfactory bulb showing intense labeling of different subsets of neurons (arrowheads) in olfactory bulb. Bar ⫽ 240 ␮m. (A) Control section of the olfactory bulb hybridized with the sense probe. Bar ⫽ 240 ␮m. (D) The anterior horn cells of the lumbar spinal cord (arrowhead) were intensely fluorescent, indicating substantial expression of CYP2B. The position of the central canal is indicated by the two arrowheads. Bar ⫽ 240 ␮m. (C) The control section hybridized with CYP2B sense riboprobe did not show any staining. Bar ⫽ 240 ␮m.

DIFFERENTIAL REGULATION OF RAT BRAIN CYTOCHROME P450 BY PHENOBARBITAL

61

FIG. 8. Immunoblot analysis of microsomal protein prepared from brain regions of vehicle and phenobarbital-treated rats. Microsomal protein from rat brain regions was subjected to SDS–PAGE and transferred to nitrocellulose membrane. The blots were immunostained with antiserum to P4502B1. The lanes contained 40 ␮g of microsomal protein from control and phenobarbital-treated rat brain regions, respectively: olfactory bulb (lanes 1 and 2), brain stem (lanes 3 and 4), thalamus (lanes 5 and 6), hippocampus (lanes 7 and 8), cerebral cortex (lanes 9 and 10), cerebellum (lanes 11 and 12), and striatum (lanes 13 and 14). Arrow indicates P4502B-immunoreactive protein of molecular weight 52K.

FIG. 7. Regulation of CYP2B mRNA in rat liver, cerebral hemisphere (A), and cerebellum (B) following phenobarbital administration. (A) Poly(A) ⫹ RNA was prepared from the cerebral hemispheres of rats treated with vehicle (lane 1, 6 ␮g) and phenobarbital for 24 h (lane 2, 6 ␮g), 48 h (lane 3, 6 ␮g), and 96 h (lane 4, 6 ␮g) as described under Materials and Methods and subjected to Northern blot analysis. CYP2B expression was visualized at 1.6 kb. (B) Total RNA was prepared from control and phenobarbital-treated rat liver (lanes 1 and 2, 6 ␮g each) as described under Materials and Methods. Poly(A) ⫹ RNA was prepared from the cerebellum of rats treated with vehicle (lane 3, 6 ␮g) and phenobarbital for 24 h (lane 4, 6 ␮g), 48 h (lane 5, 6 ␮g), and 96 h (lane 6, 6 ␮g) as described under Materials and Methods and subjected to Northern blot analysis. CYP2B expression was visualized at 1.6 kb. The relative mobilities of the 18S and 28S RNAs are indicated.

from phenobarbital treated rats incubated with the antiserum to rat liver P4502B. Intense immunostaining was observed in rat brain cortical neurons, indicating the presence of P4502B (Fig. 10A). No immunostaining was seen in control sections pretreated with nonimmune serum (Fig. 10B). Significant immunostaining of cerebellar granular cell layer (GL) and Purkinje cells and sparse staining of cells in the molecular layer (ML) were observed (Fig. 10C). Intense immunostaining of neurons in the olfactory bulb was seen, indicating the presence of P4502B (Fig. 10E). The neurons in the olfactory glomeruli (arrow) were intensely stained. No staining was seen in the external plexiform layer (EPL), while the granule cell layer (GL) was intensely stained (Fig. 10E). Immunostaining of striatal neurons was observed, indicating the presence of

P4502B protein (Fig. 10F). Expression of P4502B was observed in the midbrain region of rat brain (Figs. 11A, 11C, 11D). Control sections of midbrain treated with nonimmune serum did not show any immunostaining (Fig. 11B). Intense immunostaining of reticular neurons (arrow) in the midbrain was observed, indicating the presence of P4502B (Figs. 11C, 11D). P4502B ex-

FIG. 9. Regulation of CYP2D and CYP3A in rat brain following phenobarbital treatment. (A) The upper and lower panels were loaded identically as described below. Northern blot analysis was performed using vehicle and phenobarbital-treated rat liver total RNA (lanes 1 and 2, 10 ␮g each). Total RNA was prepared from the cerebral hemispheres of rats treated with vehicle (lane 3, 15 ␮g) and phenobarbital for 24 h (lane 4, 15 ␮g), 48 h (lane 5, 15 ␮g), and 96 h (lane 6, 15 ␮g) as described under Materials and Methods and subjected to Northern blot analysis. Lanes 7–10 were similarly loaded but contained total RNA from the cerebellum. The blots in the upper and lower panels were hybridized with cDNA to CYP2D and CYP3A, respectively. The expression of CYP2D and CYP3A was visualized at 1.6 kb. (B) Microsomal protein (40 ␮g) from control and phenobarbital treated rat brain cortex (lanes 1 and 2), cerebellum (lanes 3 and 4), and brain stem (lanes 5 and 6) were subjected to immunoblot analysis using antisera to P4502D6 and P4503A4 (upper and lower panels, respectively). The immunoreactive band was observed at molecular weight approximately 52K.

62

UPADHYA ET AL.

FIG. 10. Localization of P4502B in cortex, cerebellum, olfactory bulb, and striatum in phenobarbital-treated rat brain by immunocytochemistry using antiserum to P4502B. (A) Rat brain section showing immunostaining of cortical neurons indicating the presence of P4502B. Bar ⫽ 200 ␮m. Inset: Higher magnification of cortical neurons showing intense immunostaining in the neuronal soma. (B) Control section incubated with nonimmune serum did not show immunostaining in the cerebral cortex. Bar ⫽ 200 ␮m. (C) Intense immunostaining of granular cell layer (GL) of cerebellum was seen. Sparse staining of the cells in the molecular layer (ML) was also noted. Bar ⫽ 100␮m. Inset: Higher magnification of cerebellum showing intense staining in Purkinje cells (arrow) and the granule cell layer, indicating the presence of P4502B. (D) No staining was seen in cerebellum section incubated with nonimmune serum. Bar ⫽ 200 ␮m. (E) Intense immunostaining of neurons in olfactory bulb was seen, indicating presence of P4502B. The neurons in the glomeruli (arrow) were intensely stained. No staining was seen in external plexiform layer (EPL) while the granule cell layer (GL) was intensely stained. Bar ⫽ 100 ␮m.

pression was also seen in the neurons of thalamus (Fig. 11E). Robust immunostaining was seen in the CA2 and CA3 subfields of hippocampus (Fig. 12A) and the gran-

FIG. 11. Localization of P4502B in midbrain and thalamus of phenobarbital-treated rat brain by immunocytochemistry. (A) Expression of P4502B was seen in midbrain region of rat brain. Bar ⫽ 200 ␮m. (B) Control section of midbrain treated with nonimmune serum did not show any immunostaining. Bar ⫽ 200 ␮m. (C) Intense immunostaining of reticular neurons (arrow) in the midbrain was observed, indicating the presence of P4502B. Bar ⫽ 100 ␮m. (D) Higher magnification of a reticular neuron (arrow) indicating intense immunostaining. Bar ⫽ 10 ␮m. (E) Localization of P4502B in the neurons of thalamus was observed. Bar ⫽ 50 ␮m. (F) Immunolabeling was not observed in the control section of thalamus treated with nonimmune serum. Bar ⫽ 100 ␮m.

ular cell layer of the dentate gyrus (arrow in Fig. 12B) in phenobarbital-treated rat brain, indicating the presence of P4502B. The interneurons of the hilius (arrowhead) were also immunostained (Fig. 12B). Intense staining was seen in the island of Calleja (arrow in Fig. 12C) and lateral septal nucleus (LS) and in the endothelial cells surrounding the ventricle (V), indicating the presence of P4502B in these regions.

FIG. 12. Localization of P4502B in hippocampus and island of Calleja in phenobarbital-treated rat brain by immunocytochemistry. (A) Intense immunostaining indicating the presence of P4502B was observed in granule cells of dentate gyrus (DG) and pyramidal neurons of CA2 and CA3 subfields of hippocampus. Bar ⫽ 200 ␮m. (B) Higher magnification of the granular cell layer of the dentate gyrus (arrow) showing intense immunostaining. Staining of the interneurons of the hilius (arrowhead) was also observed. Bar ⫽ 50 ␮m. (C) Intense staining was seen in the island of Calleja (arrow), lateral septal nucleus (LS), and endothelial cells surrounding the ventricle (V), indicating the presence of P4502B. Bar ⫽ 200 ␮m. (D) Higher magnification of neurons (arrow) in the island of Calleja. Bar ⫽ 50 ␮m. FIG. 13. Effect of phenobarbital administration on toxicity of intrathecally administered MMEA in rat midbrain. Rats were administered MMEA intrathecally (5 ␮g per rat) with (E, F) and without (C, D) phenobarbital pretreatment and sacrificed 7 days later. Control rats were administered vehicle alone (A, B). Sections were cut coronally at the level of the midbrain. Magnification: (A, C, E) 16⫻; (B, D, F) 400⫻. Arrow in (D) depicts a swollen axon, while the arrow in (F) shows the loss of Nissl stain in a degenerating neuron.

DIFFERENTIAL REGULATION OF RAT BRAIN CYTOCHROME P450 BY PHENOBARBITAL

63

64

UPADHYA ET AL.

Effect of phenobarbital treatment on MMEA toxicity in midbrain. Seven days after intrathecal administration of a single dose of 5 ␮g of MMEA to rats, decreased Nissl staining and swelling of axons of the large neurons in the midbrain were observed (Figs. 13C, 13D). When a similar dose of MMEA was administered to phenobarbital-treated animals, massive neuronal damage was seen in midbrain with almost complete loss of Nissl staining in the large neurons (Figs. 13E, 13F), indicating that phenobarbital treatment potentiated MMEA toxicity in neurons in midbrain, a region where maximal P450 induction is observed. DISCUSSION

The presence of P4502B in brain has been demonstrated earlier using immunoblot and immunohistochemistry (5), slot blot analysis (23), reverse transcription PCR (24, 25), and Southern blot analysis of the double-stranded cDNA from rat brain (26). However, concerns have been raised as to whether such observations irrevocably demonstrate the constitutive expression of P4502B in rat brain (9). In the present study, we have demonstrated the presence of full-length mRNA of CYP2B in rat brain and visualized the presence of the transcript in substantial amounts in different brain regions by in situ hybridization. Both Northern blot analyses and in situ hybridization studies were carried out with CYP2B1 cDNA obtained from two different sources and the results were identical. The mRNA to CYP2B was localized predominantly in the neurons. Significant labeling of neurons was seen in the cortex, cerebellum, hippocampus, thalamus, olfactory bulb, and spinal cord. In the striatum, the neurons were labeled sparsely as compared with other regions. The neurons in the midbrain were labeled intensely, indicating significant expression of CYP2B. The localization of CYP2B mRNA seen in the present study correlates well with the localization of P4502B protein determined immunohistochemically earlier (5, 7). Earlier studies on immunohistochemical analysis of cytochrome P450 revealed strong staining of neurons, including the cerebellar granule cells (7). In the present study, immunohistochemical analysis of phenobarbital-treated brain also revealed the expression of P4502B1/2 in granular cell layers in cerebellum, in addition to cortical neurons, granule cell layer of dentate gyrus, and reticular neurons of the midbrain. Overall, the immunohistochemical study presented herein correlates well with the localization of P4502B1 by in situ hybridization. The constitutive expression of CYP2B in different neuronal cell populations in the brain may have important implications since this isoform of P450 is inducible by the commonly used anticonvulsants drugs phenobarbital (5) and phenytoin (7, 27). The chronic use of

phenobarbital and phenytoin could presumably result in the induction of the constitutively expressed P4502B in the neurons, leading to alterations in both the toxicological consequences following exposure to protoxic xenobiotics and the pharmacological and pharmacodynamic profiles of drugs coadministered with phenobarbital at their site of action, in the neurons. We therefore examined the effect of phenobarbital treatment on P450 levels in brain regions and localized the protein in phenobarbital treated rat brain. Cytochrome P450 and NADPH cytochrome c reductase activity were detectable constitutively in all regions of rat brain that were examined, indicating the ubiquitous presence of P450 in brain. Following phenobarbital administration P450 levels were significantly induced in cortex and brain stem, while they were not significantly different from controls in cerebellum, thalamus, striatum, and hippocampus. Although the constitutive expression of CYP2B was detectable in all the brain regions examined, the induction was selective only to cortex and midbrain. Thus, the CYP2B expressed in regions like cerebellum seems refractory to induction by phenobarbital. This was also validated by immunoblotting (Fig. 8) wherein an increase in the intensity of the immunostained bands was clearly seen in microsomes from cortex and brain stem in phenobarbital-treated rats (Fig. 8). The selective induction of P450 content in certain rat brain regions following phenobarbital treatment reveals the differential regulation of P450 in different regions of rat brain. In rat liver increased trancription of CYP2B is seen 24 h after a single dose of phenobarbital, while in the brain increased transcription is seen in the cerebral hemispheres only after four doses of phenobarbital administered over 96 h (Fig. 7). In the cerebellum, no change in the transcription of CYP2B was seen even at this time. CYP2B protein levels were also unchanged in the cerebellum after 10 days of daily phenobarbital administration (Fig. 8), indicating that although CYP2B was expressed constitutively in this region, it was refractory to induction. Induction of P450 in brain stem following phenobarbital administration is of significance since levels of the endogenous nucleophile glutathione are known to be significantly low in this region (28). Thus, phenobarbital administration may render the reticular neurons in brain stem particularly vulnerable to damage by bioactivation mediated by P4502B. To test this hypothesis, we studied the effect of MMEA in normal and phenobarbital-treated rats. MMEA exhibits one of the most striking examples of selective cytotoxicity for the human brain tumor cell line subpanel within the National Cancer Institute’s 60-cell line screening panel (29). Accumulation of MMEA by sensitive brain tumor cell lines may account, in part, for the selective cytotoxicity (30, 31). Moreover, in vitro studies using sag-

DIFFERENTIAL REGULATION OF RAT BRAIN CYTOCHROME P450 BY PHENOBARBITAL

ittal slices of rat brain have shown that inhibitors of cytochrome P450 metabolism abolished MMEA-induced neurotoxicity in the brain slices, while phenobarbital treatment potentiated the toxicity, indicating that P450-mediated metabolic activation is responsible for MMEA cytotoxicity (11). Further morphological examination of the slices demonstrated that MMEA exhibited selective toxicity to neuronal cell populations (10) in which P450 is localized. Since MMEA is not known to cross the blood– brain barrier, it was administered directly into the brain. Intrathecal administration of a single dose of 5 ␮g of MMEA to rats followed by morphological examination 7 days later revealed selective damage to neuronal cell populations such as the large reticular neurons in the midbrain. In phenobarbital-treated animals, the neuronal damage was dramatically increased and loss of the neuronal cell population was clearly discernible even at low magnification (Fig. 13), indicating that phenobarbital treatment increased the vulnerability to neuronal damage mediated by MMEA. Phenobarbital treatment is known to induce P4503A levels in rat liver (32) and brain (33) as examined by reverse transcription PCR. However, in the present study, while we were able observe the induction of CYP3A expression in liver after a single dose of phenobarbital, no such induction was seen in the brain either in the cerebral hemisphere or in the cerebellum (Fig. 9). Even after 10 days of phenobarbital treatment, we were unable to detect increased expression of CYP3A protein. In a similar manner, CYP2D was also not induced by phenobarbital treatment in brain while it was induced in liver. These results indicate that phenobarbital selectively induces CYP2B in brain while not affecting other P450 isoforms. Phenobarbital and phenytoin are anticonvulsants that are used extensively in patient care for extended duration, often in combination with other drugs. The chronic use of phenobarbital and phenytoin could presumably result in induction of constitutively expressed P4502B in the neurons, leading to alterations in toxicological and pharmacological profiles of coadministered xenobiotics including drugs at their site of action, in the neurons. ACKNOWLEDGMENTS The authors thank Dr. M. Adesnik and Dr. N. Avadhani for providing the cDNA to CYP2B1. The technical assistance of Ms. Shailaja Hegde is acknowledged.

REFERENCES 1. de Montellano, O. (1986) Cytochrome P450 Structure, Mechanism and Biochemistry, Plenum, New York. 2. Gram, T. E., Okine, L. R., and Gram, R. A. (1986) Annu. Rev. Pharmacol. Toxicol. 26, 259 –291.

65

3. McLemore, T. L., Litterest, C. C., Coudert, B. P., Liu, M. C., Hubbard, W. C., Adelberg, S., Czerwinki, M., McMahon, J. A., Eggleston, J. C., and Boyd, M. R. (1990) J. Natl. Can. Inst. 82, 1420 –1426. 4. Ravindranath, V., and Boyd, M. R. (1995) Drug. Metabol. Rev. 27, 419 – 448. 5. Anandatheerthavarada, H. K., Shankar, S. K., and Ravindranath, V. (1990) Brain Res. 536, 339 –343. 6. Kapitulnik, J., Gelboin, H. V., Guengerich, F. P., and Jacobowitz, D. M. (1987) Neuroscience 20, 829 – 833. 7. Volk, B., Hettmannsperger, U., Papp, T., Amelizad, Z., Oesch, F., and Knoth, R. (1991) Neuroscience 42, 215–235. 8. Le Goascogne, C., Robel, P., Gouezou, M., Sananes, N., Baulieu, E. E., and Waterman, M. (1987) Science 237, 1212–1215. 9. Hedlund, E., Gustafsson, J. A., and Warner, M. (1998) Trends Pharmacol. Sci. 19, 82– 85. 10. Shankar, L., Ravindranath V., Boyd, M. R., Vistica, D. T., and Shankar, S. K. (1997) Neurotoxicology 18, 89 –96. 11. Sriram, K., Boyd, M. R., Vistica, D. T., and Ravindranath, V. (1987) Neurotoxicology 18, 97–104. 12. Fujii-Kuriyama, Y., Mizukami, Y., Kawajiri, K., Sogawa, K., and Muramatsu, M. (1982) Proc. Natl. Acad. Sci. USA 79, 2793–2797. 13. Chomczynski, P. (1993) Biotechniques 15, 532–537. 14. Kevil, C. G., Walsh, L., Laroux, S., Kalogeris, T., Grisham, M. B., and Alexander, J. S. (1997) Biochem. Biophys. Res. Commun. 238, 277–279. 15. Glowinsky, J., and Iversen, L. L. (1966) J. Neurochem. 13, 655– 669. 16. Ravindranath, V., and Anandatheerthavarada, H. K. (1990) Anal. Biochem. 187, 310 –313. 17. Bradford, M. M. (1976) Anal. Biochem. 72, 248 –256. 18. Mastubara, T., Koike, M., Touchi, A., Touchino, Y., and Sugeno, K. (1976) Anal. Biochem. 76, 596 – 603. 19. Phillips, A. H., and Langdon, R. G. (1962) J. Biol. Chem. 237, 2652–2660. 20. Guengerich, F. P. (1984) in Principles and Methods of Toxicology (Hayes, A. W., Ed.), pp. 609 – 634, Raven Press, New York. 21. Laemmli, U. K. (1970) Nature 227, 680 – 685. 22. Towbin, H., Staehlin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76, 4350 – 4354. 23. Strobel, H. W., Cattaneo, E., Adesnik, M., and Maggi, A. (1989) Pharmacol. Res. 21, 169 –175. 24. Hodgson, A. V., White, T. B., White, J. W., and Strobel, H. W. (1992) Mol. Cell. Biochem. 120, 171–179. 25. Schilter, B., and Omiecinski, C. J. (1993) Mol. Pharmacol. 44, 990 –996. 26. Tirumalai, P. S., Bhamre, S., Upadhya, S. C., Boyd, M. R., and Ravindranath, V. (1998) Biochem. Pharmacol. 56, 371–375. 27. Kempermann, G., Knoth, R., Gebicke-Haerter, P. J., Stolz, B. J., and Volk, B. (1994) Neurosci. Res. 39, 576 –588. 28. Ravindranath, V., Anandatheerthavarada, H. K., and Shivakumar, B. R. (1989) Neurosci. Lett. 101, 187–190. 29. Boyd, M. R., Niederhuber, J. E. (Eds.) (1993) Current Therapy in Oncology, pp. 11–22, B. C. Decker, Philadelphia. 30. Vistica, D. T., Kenney, S., Hursey, M. L., and Boyd, M. R. (1994) Biochem. Biophys. Res. Commun. 200, 1762–1768. 31. Kenney, S., Vistica, D. T., Linden, H., and Boyd, M. R. (1995) Biochem. Pharmacol. 49, 23–32. 32. Sueyoshi, T., and Negishi, M. (2001) Annu. Rev. Pharmacol. Toxicol. 41, 123–143. 33. Schilter, B., Andersen, M. K., Acharya, C., and Omiecinski, C. J. (2000) J. Pharmacol. Exp. Ther. 294, 916 –922.