Cannabinoid-induced alterations in brain disposition of drugs of abuse.

Marijuana contains a complex mixture of compounds including tetrahydrocannabinol (THC), the major psychoactive constituent, and cannabidiol (CBD), a nonpsychoactive constituent. We have shown previously that CBD pretreatment of mice increases brain levels of THC and have now further characterized this effect and determined whether the brain pharmacokinetics of other drugs are also affected. CBD pretreatment of mice (30-60 min) increased brain levels of THC nearly 3-fold, whereas CBD co-administration did not. Because marijuana is often consumed with other drugs, the influence of cannabinoids on the brain levels of several other drugs of abuse was also determined. CBD pretreatment of mice increased brain levels (2- to 4-fold) of subsequently administered cocaine as well as phencyclidine (PCP). Although CBD pretreatment increased blood and brain levels of cocaine comparably, blood levels of PCP were only modestly elevated (up to 50%). Behavioral tests indicated that the CBD-mediated increases in the brain levels of THC, cocaine, and PCP correlated with increased pharmacological responses. Pretreatment with THC instead of CBD could similarly increase brain levels of cocaine, PCP, and CBD, although with a lower potency than CBD. On the other hand, pretreatment of mice with CBD had no effect on the brain levels of several other drugs of abuse including morphine, methadone, or methylenedioxyphenyl-methamphetamine. These findings demonstrate that cannabinoids can increase the brain concentrations and pharmacological actions of several other drugs of abuse, thereby providing a biochemical basis for the common practice of using marijuana concurrently with such drugs.

The effects of phencyclidine pretreatment on cocaine-mediated hepatotoxicity in mice.

Cocaine-mediated hepatotoxicity (CMH) requires cocaine (CCN) bioactivation by microsomal monooxygenase enzymes that results in cell death. Proposed mechanisms of toxicity involve reactive metabolites that covalently bind to hepatocellular proteins, depletion of cellular reducing equivalents through redox cycling, and/or the generation of reactive oxygen and nitrogen species that alter lipids and proteins. We have previously shown that phencyclidine (PCP) pretreatment potentiated CMH in CF-1 mice without increasing in vitro N-demethylation or N-hydroxylation of CCN. We have now further characterized PCP-potentiated CMH and determined that it is a dose- and time-dependent process, with PCP doses as low as 2.5 mg/kg for 3 days significantly increasing CMH. Immunohistochemistry and histology of livers from mice pretreated with PCP before CCN administration revealed a marked correlation between the regions of CCN metabolite binding and that of necrosis, whereas there was little binding or necrosis in vehicle-pretreated mice. Although hepatic GSH levels were not altered after repetitive PCP treatment alone, a sustained decrease (at least 6 h) in these levels was observed following CCN administration. Inhibitors of inducible nitric oxide synthase (NOS) abrogated PCP-potentiated CMH, although repetitive PCP treatment alone did not increase nitric oxide synthesis systemically or locally in hepatic tissue nor did lipopolysaccharide induction of NOS (without PCP) directly potentiate CMH. The precise mechanisms of PCP potentiation of CMH and involvement of NOS in CMH remain unclear, however, sustained depletion of GSH levels and increased hepatocellular binding of reactive cocaine metabolites have been demonstrated.

Effects of unsaturated side-chain analogs of tetrahydrocannabinol on cytochromes P450.

The ability of unsaturated side-chain analogs of Delta(8)-tetrahydrocannabinol (THC) to selectively inactivate mouse hepatic cytochromes P450 3A11 and 2C29 was examined. THC side-chain analogs were preincubated with mouse hepatic microsomes and NADPH for various times before dilution and determination of Delta(9)-THC metabolism specific for P450s 3A11 and 2C29. THC-enyl analogs had little or no effect on P450 3A11 but inactivated P450 2C29 in a time-dependent manner, with approximately 50% inactivation observed after a 30-min preincubation. THC-ynyl analogs were less selective in their P450 inactivation but appeared to be more effective than their corresponding enyl analogs. THC-ynyl analogs inactivated P450s 3A11 and 2C29 in a time-dependent manner and could inactive 40-80% of their activities after a 30-min preincubation. The THC-ynyl analogs were nearly as effective as cannabidiol, a well-characterized inactivator of these mouse P450s. Despite their ability to inactivate P450 in vitro, neither the THC-enyl nor the THC-ynyl analogs were very effective after in vivo administration. Unsaturated side-chain THC analogs may be useful in the development of specific P450 inactivators.

Ethynylestradiol-mediated induction of hepatic CYP3A9 in female rats: implication for cyclosporine metabolism.

Repeated treatment of female rats with the synthetic estrogen ethynylestradiol (EE(2)) increases the formation of the cyclosporine A (CyA) metabolites AM1c and AM9 by 3-fold, whereas the formation of AM1 and AM4N is not significantly enhanced. The formation of all four CyA metabolites was inhibited by greater than 80% by the CYP3A-selective substrate midazolam or polyclonal anti-rat CYP3A IgGs in liver microsomes of untreated and EE(2)-induced rats. In contrast, anti-rat CYP2C6 IgGs had little effect, indicating the involvement of a CYP3A but not 2C6 in this EE(2)-stimulated CyA metabolism. Semiquantitative reverse-transcriptase polymerase chain reaction was used to determine the mRNA content for four CYP3A genes (CYP3A2, CYP3A9, CYP3A18, and CYP3A23) in livers of control and EE(2)-treated female rats. EE(2) selectively induced CYP3A9 by 3.3-fold whereas the expression of CYP3A18 and CYP3A23 was slightly decreased; neither CYP3A2 mRNA nor CYP3A1 mRNA was detectable in these EE(2)-treated livers. To determine whether rat liver microsomal CYP3A9 was indeed responsible for the EE(2)-stimulated CyA metabolism, a recombinant CYP3A9 was heterologously expressed in Escherichia coli. When functionally reconstituted, this enzyme was active in metabolizing CyA preferentially to its AM9 and AM1c metabolites as compared with CYP3A4. These findings thus support the notion that the increased CyA-metabolizing capacity of EE(2)-treated female rat liver microsomes is due to the induction of the CYP3A9 enzyme.

Identification of the heme-modified peptides from cumene hydroperoxide-inactivated cytochrome P450 3A4.

Cumene hydroperoxide-mediated (CuOOH-mediated) inactivation of cytochromes P450 (CYPs) results in destruction of their prosthetic heme to reactive fragments that irreversibly bind to the protein. We have attempted to characterize this process structurally, using purified, 14C-heme labeled, recombinant human liver P450 3A4 as the target of CuOOH-mediated inactivation, and a battery of protein characterization approaches [chemical (CNBr) and proteolytic (lysylendopeptidase-C) digestion, HPLC-peptide mapping, microEdman sequencing, and mass spectrometric analyses]. The heme-peptide adducts isolated after CNBr/lysylendopeptidase-C digestion of the CuOOH-inactivated P450 3A4 pertain to two distinct P450 3A4 active site domains. One of the peptides isolated corresponds to the proximal helix L/Cys-region peptide 429-450 domain and the others to the K-region (peptide 359-386 domain). Although the precise residue(s) targeted remain to be identified, we have narrowed down the region of attack to within a 17 amino acid peptide (429-445) stretch of the 55-amino acid proximal helix L/Cys domain. Furthermore, although the exact structures of the heme-modifying fragments and the nature of the adduction remain to be established conclusively, the incremental masses of approximately 302 and 314 Da detected by electrospray mass spectrometric analyses of the heme-modified peptides are consistent with a dipyrrolic heme fragment comprised of either pyrrole ring A-D or B-C, a known soluble product of peroxidative heme degradation, as a modifying species.

Characterization of cytochrome P450 3A inactivation by cannabidiol: possible involvement of cannabidiol-hydroxyquinone as a P450 inactivator.

Cannabidiol (CBD) is a major constituent of marijuana and a potent inhibitor of P450-mediated hepatic drug metabolism. Mouse P450 3A11 metabolism of [14C]CBD resulted in the formation of radiolabeled P450, which after digestion with lysyl endopeptidase C (Lys-C) and HPLC resolution of peptides, revealed one major broadly eluting peak of radioactivity. Electrophoresis/autoradiography of this peak identified several peptide bands, one of which was predominantly radiolabeled and had an apparent molecular mass of approximately 6 kDa. Amino-terminal sequence determination of this band revealed the presence of two peptides whose sequences identified them as Ala344-Lys379 and Gly426-Lys454. To characterize the reactive species that may be generated during P450 3A11-catalyzed CBD metabolism, reduced glutathione (GSH) was used as a trapping agent for possible electrophilic metabolites. Incubation of P450 3A11 in the presence of cofactors NADPH, CBD, and [3H]GSH resulted in the formation of a radiolabeled product which was absent in incubations lacking any of the cofactors. The UV absorption spectra of this compound indicated absorbances at approximately 220, 275, and 350 nm, and mass spectral analysis revealed prominent ions at m/z 634, 599, 505, 402, and 359, ions consistent with that of a GSH adduct of CBD-hydroxyquinone. A synthetic CBD-hydroxyquinone-GSH adduct was also prepared and had UV absorption and mass spectra nearly identical to that of the P450-mediated CBD-GSH adduct. CBD-hydroxyquinone formation may be the penultimate oxidative step involved in CBD-mediated modification and inactivation of P450 3A11.

Oleamide potentiates benzodiazepine-sensitive gamma-aminobutyric acid receptor activity but does not alter minimum alveolar anesthetic concentration.

UNLABELLED: A naturally occurring brain lipid, cis-9,10-octadeceamide--oleamide (OA), is found in increased concentrations in the cerebrospinal fluid of sleep-deprived cats, which suggests that it may be an endogenous sleep-inducing substance. We studied the effects of this fatty-acid derivative on the function of cloned gamma-aminobutyric acid (GABA(A)) receptors expressed in Xenopus oocytes. Oocytes were injected with cRNA synthesized in vitro to express simple GABA(A) receptors (alpha1beta1, alpha3beta1, alpha5beta1, and alpha1beta2 subunit combinations) and receptors in which the GABA-induced chloride currents were potentiated in the presence of benzodiazepines (alpha1beta1gamma2s and alpha1beta2gamma2s subunit combinations). OA only produced significant potentiation of the peak Cl- current when applied with GABA to benzodiazepine-sensitive GABA(A) receptors. The peak currents of the simple GABA(A) receptors in the presence of OA were either unaffected or slightly inhibited by OA, but the overall mean currents were not significantly altered. Oleic acid was also capable of potentiating benzodiazepine-sensitive GABA(A) receptor function. The function of other ligand-gated ion channels, such as the N-methyl-D-aspartate receptor (NR1 + NR2A or 2C) and the 5-HT3 receptor expressed in Xenopus oocytes, were unaffected by OA. Sprague-Dawley rats receiving intraperitoneal injections of oleamide (10, 20, or 100 mg/kg) showed no change in the minimum alveolar anesthetic concentration (MAC) of desflurane required to abolish movement in response to noxious (tail clamp) stimulation (control MAC 6.48% +/- 1.28% atm; 100 mg/kg OA MAC 7.05% +/- 0.42% atm). These results reinforce the view that oleyl compounds may be natural modulators of inhibitory ion channel function, but that these effects contribute little to the central nervous system depression produced by volatile anesthetics as measured by MAC. IMPLICATIONS: The putative sleep-inducing substance, oleamide, potentiates benzodiazepine-sensitive gamma-aminobutyric acid receptor function but does not alter desflurane minimum alveolar anesthetic concentration in rats.

Effect of cytochrome P450 inducers on cocaine-mediated hepatotoxicity.

The effect of several cytochrome P450 (P450) inducers on cocaine metabolism were examined in order to characterize the metabolic events contributing to cocaine-induced hepatotoxicity. Phenobarbital (PB)-pretreatment of mice induced P450s 3A and 2B and markedly increased serum alanine aminotransferase (ALT) activity after cocaine or norcocaine administration. Although dexamethasone (Dex) induced P450s 3A and 2B at least to the same extent as PB, no increase in serum ALT activity was observed after cocaine or norcocaine administration. Phencyclidine (PCP) pretreatment did not increase either P450s 3A or 2B, yet it markedly enhanced cocaine- or norcocaine-induced serum ALT activity. In contrast to the marked induction of P450s 3A and 2B, P450 2C was increased only 2.5-fold by PB and to an even lesser extent by Dex or PCP. Cannabidiol (CBD), which inactivates P450s 3A and 2C in mice, completely protected mice against cocaine- or norcocaine-induced hepatotoxicity irrespective of whether they were induced or not with PB or PCP. Both PB and Dex pretreatment increased the in vitro hepatic microsomal formation of the first two sequential oxidative metabolites of cocaine (norcocaine and N-hydroxynorcocaine), whereas PCP pretreatment did not. Hepatic esterase activity was also determined after pretreatment with P450 inducers, since this is the major detoxification pathway in cocaine metabolism. Dex pretreatment markedly increased (> 11-fold) total hepatic esterase activity, whereas PB pretreatment increased it more modestly (less than fourfold) and PCP pretreatment had little effect. This marked effect of Dex pretreatment may decrease liver cocaine concentrations and thus protect mice against cocaine-induced hepatotoxicity, despite their increased P450 2B and 3A contents.

Dual effects of anandamide on NMDA receptor-mediated responses and neurotransmission.

Anandamide is an endogenous ligand of cannabinoid receptors that induces pharmacological responses in animals similar to those of cannabinoids such as delta9-tetrahydrocannabinol (THC). Typical pharmacological effects of cannabinoids include disruption of pain, memory formation, and motor coordination, systems that all depend on NMDA receptor mediated neurotransmission. We investigated whether anandamide can influence NMDA receptor activity by examining NMDA-induced calcium flux (deltaCa2+NMDA) in rat brain slices. The presence of anandamide reduced deltaCa2+NMDA and the inhibition was disrupted by cannabinoid receptor antagonist, pertussis toxin treatment, and agatoxin (a calcium channel inhibitor). Whereas these treatments prevented anandamide inhibiting deltaCa2+NMDA, they also revealed another, underlying mechanism by which anandamide influences deltaCa2+NMDA. In the presence of cannabinoid receptor antagonist, anandamide potentiated deltaCa2+NMDA in cortical, cerebellar, and hippocampal slices. Anandamide (but not THC) also augmented NMDA-stimulated currents in Xenopus oocytes expressing cloned NMDA receptors, suggesting a capacity to directly modulate NMDA receptor activity. In a similar manner, anandamide enhanced neurotransmission across NMDA receptor-dependent synapses in hippocampus in a manner that was not mimicked by THC and was unaffected by cannabinoid receptor antagonist. These data demonstrate that anandamide can modulate NMDA receptor activity in addition to its role as a cannabinoid receptor ligand.

Determination of esterase-catalyzed cocaine metabolite formation by high-performance liquid chromatography with ultraviolet detection.

A reversed-phase high-performance liquid chromatographic (HPLC) method for the determination of cocaine metabolites produced in vitro by serum and liver esterases is described. Hydrolysis of cocaine at the methyl ester bond produces benzoylecgonine and methanol, whereas hydrolysis at the benzoyl ester bond produces ecgonine methyl ester and benzoic acid. This method quantitates benzoic acid (as a measure of ecgonine methyl ester formation) and benzoylecgonine, which can be determined simultaneously and with great sensitivity by HPLC and ultraviolet detection. Cocaine is found to be hydrolyzed to both ecgonine methyl ester and benzoylecgonine; hepatic microsomes exhibit the highest specific activity.

Inhibition of cyclosporine and tetrahydrocannabinol metabolism by cannabidiol in mouse and human microsomes.

1. The in vitro and in vivo effects of cannabidiol on mouse and human liver microsomal metabolism of the immunosuppressive drug cyclosporine and the psychoactive compound tetrahydrocannabinol have been examined. 2. Preincubation of mouse or human liver microsomes with cannabidiol decreased the formation of all detectable cyclosporine metabolites by 73-89%. 3. In vivo cannabidiol treatment of mouse similarly decreased the formation of all detectable cyclosporine metabolites by 60-86%. 4. Preincubation of human liver microsomes with cannabidiol selectively decreased the formation of tetrahydrocannabinol metabolites catalyzed by cytochrome P4503A by 60% but had no effect on P4502C9-catalyzed metabolites. 5. Cannabidiol has the potential to clinically affect cyclosporine metabolism which may result in increased cyclosporine blood levels and an increase in its toxic side effects, and likewise may also affect tetrahydrocannabinol and its metabolite levels in man.

Anandamide hydroxylation by brain lipoxygenase:metabolite structures and potencies at the cannabinoid receptor.

Anandamide (arachidonyl ethanolamide) is a compound that was identified from porcine brain lipids by its ability to bind to the brain cannabinoid receptor. This study assessed anandamide as a substrate for a brain lipoxygenase and characterised the brain metabolite 12-hydroxyanandamide. Anandamide was also compared with arachidonic acid as a lipoxygenase substrate by examining enzyme kinetics in the presence of either of the two compounds. In addition, a non-mammalian enzyme was used to generate 11- and 15-hydroxy-anandamide in order to compare the cannabinomimetic properties of a range of anandamide derivatives. A ligand displacement assay indicated a large variation in the affinity of anandamide metabolites for the brain cannabinoid receptor. The brain metabolite, 12-hydroxyanandamide had an affinity twice that of anandamide, although the 11- and 15- hydroxy-metabolites were considerably poorer ligands of this receptor. Consistent with the receptor binding data, 12-hydroxyanandamide (unlike 15-hydroxyanandamide) inhibited forskolin-stimulated cAMP synthesis, indicating it to be a functional agonist at the brain cannabinoid receptor. Pharmacological studies of the capacity of anandamide and its metabolites to inhibit the murine vas deferens twitch response indicated the 12-hydroxy-metabolite to be less active than the parent compound, but a better cannabinomimetic than 15-hydroxyanandamide.

A universal approach to the expression of human and rabbit cytochrome P450s of the 2C subfamily in Escherichia coli.

Human cytochrome P450s 2C8, 2C9, 2C18, and 2C19 and rabbit cytochrome P450s 2C1, 2C2, 2C4, 2C5, and 2C16 were expressed from their respective cDNAs in Escherichia coli as chimeric enzymes in which a portion of the N-terminal membrane anchor sequence was replaced with a modified sequence derived from P450 17A. For 2C1 and 2C2 removal of the extraneous 3'-untranslated sequence allowed the successful expression of constructs that were unproductive in its presence. The levels of expression varied from 180 to 1500 nmol/liter of culture and the addition of delta-aminolevulinic acid to the culture media increased the amount of spectrally detectable P450 for several of these enzymes 2- to 10-fold. The catalytic properties of the modified human 2C P450s expressed in E. coli were concordant with previously published data for several marker substrates including (S)-mephenytoin for P450 2C19, tolbutamide and tetrahydrocannabinol (THC) for P450 2C9, and taxol for P450 2C8. Interestingly, P450 2C19 catalyzed the 21-hydroxylation of progesterone and, to a lesser extent, catalyzed the formation of 16 alpha-hydroxyprogesterone. The rabbit enzyme P450 2C16 catalyzed the formation of 17 alpha- and 16 alpha-hydroxyprogesterone in addition to 21-hydroxylation. P450 2C19 also catalyzed the methylhydroxylation of tolbutamide and the 7-hydroxylation of THC at rates that were similar to or greater than that of P450 2C9. This work has identified important factors required for the high-level expression of 2C subfamily P450s in E. coli. The availability of these enzymes will facilitate detailed kinetic measurements for known and yet to be identified substrates.

Effect of cannabidiol pretreatment on the kinetics of tetrahydrocannabinol metabolites in mouse brain.

Tetrahydrocannabinol (THC), a major constituent of marijuana, and several of its metabolites are psychoactive in humans. Cannabidiol (CBD), a nonpsychoactive cannabinoid, inhibits hepatic microsomal THC metabolism and also modulates subjective psychological responses to THC in humans. Treatment of mice with CBD markedly decreased the hepatic microsomal in vitro formation of the major THC metabolites, 6 alpha-OH-THC and 7-OH-THC and increased formation of the minor metabolite 6 beta-OH-THC. THC blood levels were modestly elevated after CBD pretreatment and THC administration, compared with untreated controls, and area under the curve (AUC) of THC increased 50% as a function of decreased clearance. CBD pretreatment modestly increased the Cmax, AUC, or t1/2 of the major THC metabolites in the blood, whereas those kinetic parameters for 6 beta-OH-THC were dramatically increased. Changes in brain concentrations and kinetic parameters of the major THC metabolites did not reflect the relatively modest changes found in blood levels after CBD pretreatment, but exhibited large increases in AUC (7- to 15-fold) and t1/2 (2- to 4-fold), as well as in tmax. Changes in brain concentrations and kinetic parameters for 6 beta-OH-THC reflected the marked changes observed in blood levels after CBD pretreatment. Thus, CBD pretreatment resulted in large increases in AUC and t1/2 of all THC metabolites in brain, with a modest increase in AUC of THC. These changes in THC metabolite brain pharmacokinetics may contribute to the modulation of psychological responses to THC observed after CBD treatment.

Microsomal cytochrome P450-mediated liver and brain anandamide metabolism.

Anandamide (AN) is an arachidonic acid congener, found in the brain, that binds to the cannabinoid receptor and elicits cannabinoid-like pharmacological activity. Cytochromes P450 (P450s) are known to oxidize arachidonic acid to a wide variety of metabolites, yielding many physiologically potent compounds. To determine if AN could be similarly oxidized by P450s, its metabolism by mouse liver and brain microsomes was examined. Mouse hepatic microsomal incubation of AN with NADPH resulted in the generation of at least 20 metabolites, determined after HPLC separation by increased UV-absorbance at 205 nm. Pretreatment of mice with various P450 inducers resulted in increased hepatic microsomal formation of several AN metabolites, with dexamethasone being the most effective inducer. Phenobarbital pretreatment resulted in a metabolic profile similar to that observed after dexamethasone pretreatment, whereas 3-methylcholanthrene pretreatment selectively increased the formation of several other metabolites. Clofibrate pretreatment had no effect on hepatic AN metabolism. Polyclonal antibodies prepared against mouse hepatic P450 3A inhibited the formation of several AN metabolites by hepatic microsomes from untreated mice as well as the formation of those metabolites found to be increased after dexamethasone pretreatment. AN metabolism by brain microsomes resulted in the formation of two NADPH- and protein-dependent metabolites. Hepatic P450 3A antibody partially inhibited the formation of only one of these metabolites. Thus, P450 3A is a major contributor to AN metabolism in the liver but not in the brain. The physiological consequences of P450-mediated AN metabolism remain to be determined.

Induction and genetic regulation of mouse hepatic cytochrome P450 by cannabidiol.

Cannabidiol (CBD) has been shown to be a selective inactivator of cytochromes P450 (P450s) 2C and 3A in the mouse and, like many P450 inactivators, it can also induce P450s after repeated administration. The inductive effects of CBD on mouse hepatic P450s 2B, 3A, and 2C were determined using cDNA probes, polyclonal antibodies, and specific functional markers. P450 2B10 mRNA was increased markedly after repeated CBD administration and correlated well with increased P450 2B immunoquantified content and functional activity. On the other hand, although the 2-fold increase in P450 3A mRNA detected after repeated CBD administration was consistent with the increased immunoquantified P450 3A protein content, the lack of an observable increase in P450 3A-specific functional activity suggested subsequent inactivation of the induced P450 3A. Repeated CBD treatment increased P450 2C mRNA content 2-fold, but did not increase either the P450 2C immunoquantified content or its functional activity. The effect of CBD treatment on the ability of tetrahydrocannabinol (THC) to induce P450 2B was also determined. A THC dose that did not induce P450 2B significantly was administered alone or in combination with a CBD dose that markedly inactivated P450s 2C- and 3A but submaximally increased P450 2B functional activity. The combination of THC and CBD did not increase P450 2B-catalyzed activity significantly over that observed after CBD treatment alone. Thus, prior CBD-mediated P450 inactivation does not appear to increase the ability of THC to induce P450 2B. To further characterize the relationship between P450 inactivation and induction, several structurally diverse CBD analogs with varying P450 inactivating potentials were tested for their ability to induce P450 2B. At least one CBD analog that is an effective P450 inactivator failed to induce P450 2B, while at least one CBD analog that is incapable of inactivating P450 was found to be a very good P450 2B inducer. It therefore appears that inherent structural features of the CBD molecule rather than its ability to inactivate P450 determine P450 2B inducibility. The complex effects of CBD treatment on P450 inactivation and induction have the potential to influence the pharmacological action of many clinically important drugs known to be metabolized by these various P450s. The mechanism of CBD-mediated P450 induction remains to be elucidated but does not appear to be related to CBD-mediated P450 inactivation.

The effect of cannabidiol on mouse hepatic microsomal cytochrome P450-dependent anandamide metabolism.

Anandamide (arachidonylethanolamide) has been identified as a brain constituent that selectively binds to the cannabinoid receptor and possesses cannabimimetic activity. Cytochromes P450 catalyze the oxidation of arachidonic acid to several metabolites possessing very potent pharmacological activity. We examined whether P450 would also metabolize anandamide, and whether cannabidiol (a cannabinoid which inactivates several P450s) would affect this metabolism. Mouse hepatic P450s were found to metabolize anandamide to at least 10 different metabolites, four of which were characterized by mass spectrometry. Cannabidiol selectively inhibited the formation of two of these four anandamide metabolites. The significance of anandamide metabolism remains to be explored.

Characterization of cannabidiol-mediated cytochrome P450 inactivation.

Cannibidiol (CBD) has been shown to impair hepatic drug metabolism in several animal species and to markedly inhibit mouse hepatic microsomal delta 1-tetrahydrocannabinol (THC) metabolism by inactivating specific cytochrome P450s (P450) belonging to the 2C and 3A subfamilies. Elucidation of the mechanism of CBD-mediated P450 inhibition would be clinically very important for predicting its effect on metabolism of THC and the many other clinically important drugs known to be metabolized by P450s 2C and 3A. CBD-mediated inactivation of mouse hepatic microsomal P450s did not decrease hepatic microsomal heme content. However, [14C]CBD was found covalently bound to microsomal protein in an approximately equimolar ratio to P450 loss. Immunoprecipitation of microsomal protein with antibodies raised against either P450 2C or 3A revealed that approximately equal amounts of [14C]-CBD were bound to each of these P450s after CBD-mediated inactivation. Furthermore, this specific P450 binding was equivalent to P450 loss and accounted for nearly all of the microsomal [14C]CBD irreversible binding. Although > 80% of the enzyme activities attributed to P450s 2C and 3A were inactivated by CBD at the anticonvulsant dose of 120 mg/kg, P450 2C was approximately 3-fold more sensitive than P450 3A at the lower CBD doses tested. CBD analogs were synthesized in order to elucidate the chemical pathways for CBD-mediated P450 inactivation in vivo. Oxidations at allylic carbon positions or saturation of either the exocyclic double bond or both double bonds of the terpene moiety did not markedly affect the inhibitory properties of the analogs. Methylation of both phenolic groups of the resorcinol moiety completely blocked the P450-inhibitory properties of this analog, revealing the involvement of a free hydroxyl group in the inactivation process. Rotation of the resorcinol moiety in abnormal-CBD did not impair the inhibitory properties of the analog, suggesting that the position of the hydroxyl group relative to the terpene ring is unimportant. Further studies are required to fully understand the chemical basis of CBD-mediated P450 inactivation.

Human hepatic microsomal metabolism of delta 1-tetrahydrocannabinol.

Hepatic microsomal metabolism of delta 1-tetrahydrocannabinol (THC) has been extensively studied in many rodent species, but there have been few reports describing such metabolism in humans. Because several THC metabolites are known to be pharmacologically active, identifying the P-450 subfamilies responsible for their formation is of clinical importance. We have found that, in addition to catalyzing the formation of significant amounts of 7-hydroxy-THC, hepatic microsomes from nine human livers also formed 6 beta-hydroxy-THC at approximately the same rate. In addition, 1 alpha,2 alpha-epoxyhexahydrocannabinol (EHHC) was formed at approximately one-third the rate of 7-hydroxy- and 6 beta-hydroxy-THC, and small amounts of 6 alpha-hydroxy- and 6-keto-THC were also found. Immunoinhibition studies with antibodies raised against human hepatic P-450 2C9, or a mouse hepatic P-450 isozyme belonging to the P-450 3A subfamily, revealed that P-450 2C9 catalyzed the formation of 7-hydroxy-THC, whereas P-450 3A catalyzed the formation of 6 beta-hydroxy-THC, EHHC, and the relatively minor metabolites. In contrast, antibodies raised against human P-450 2C8 had no affect on human microsomal THC hydroxylation. Excellent correlations were found between hepatic microsomal P-450 2C9 and 3A content and 7-hydroxy- and 6 beta-hydroxy-THC formation, respectively. In addition, purified P-450 2C9 catalyzed the formation of 7-hydroxy-THC at a 7-fold higher rate than that observed with microsomes. Microsomal 7-hydroxy-THC formation varied less than 5-fold between the livers, suggesting that this activity is normally expressed and probably not subject to environmental influences.(ABSTRACT TRUNCATED AT 250 WORDS)

Topological analysis of the active sites of cytochromes P450IIB4 (rabbit), P450IIB10 (mouse), and P450IIB11 (dog) by in situ rearrangement of phenyl-iron complexes.

The reaction of phenyldiazene with purified, phenobarbital-inducible rabbit cytochrome P450IIB4, mouse cytochrome P450IIB10, and dog cytochrome P450IIB11 yields complexes with absorbance maxima at 480 nm. Treatment of the cytochrome P450 complexes with K3Fe(CN)6 results in disappearance of the 480-nm absorption. Extraction of the prosthetic group from the proteins after these reactions yields the two isomers of N-phenylprotoporphyrin IX with the N-phenyl group on pyrrole rings A and D as the major products and the regioisomer with the N-phenyl on pyrrole ring C as a minor product. The A:C:D arylated pyrrole ring ratio is 3:2:3 for rabbit P450IIB4, 3:1:3 for mouse P450IIB10, and 4:1:2 for dog P450IIB11. Formation of the A and D regioisomers is consistent with the results obtained previously for rat isozymes IA1, IIB1, IIB2, and IIE1, but the rabbit, mouse, and dog P450IIB enzymes differ from the four rat enzymes in that a substantial amount of the isomer with the N-phenyl on pyrrole ring C is also formed. The results indicate that the region over pyrrole ring B is masked by protein residues in all the active sites and suggest that the region over pyrrole ring C is more hindered by protein residues in the rat than in the rabbit, mouse, or dog enzymes so far examined.