Characterization of metabolites of the chemopreventive agent curcumin in human and rat hepatocytes and in the rat in vivo, and evaluation of their ability to inhibit phorbol ester-induced prostaglandin E2 production.

Curcumin, the yellow pigment in turmeric, has been shown to prevent malignancies in a variety of tissues in rodents, especially in the intestinal tract. Pharmacological activities of curcumin in cells in situ germane to chemoprevention, such as inhibition of expression of cyclooxygenase-2 (COX-2), require drug concentrations in the 10(-5) - 10(-4) M range. The systemic bioavailability of curcumin is low, so that its pharmacological activity may be mediated, in part, by curcumin metabolites. To investigate this possibility, we compared curcumin metabolism in human and rat hepatocytes in suspension with that in rats in vivo. Analysis by high-performance liquid chromatography with detection at 420 and 280 nm permitted characterization of metabolites with both intact diferoylmethane structure and increased saturation of the heptatrienone chain. Chromatographic inferences were corroborated by mass spectrometry. The major metabolites in suspensions of human or rat hepatocytes were identified as hexahydrocurcumin and hexahydrocurcuminol. In rats, in vivo, curcumin administered i.v. (40 mg/kg) disappeared from the plasma within 1 h of dosing. After p.o. administration (500 mg/kg), parent drug was present in plasma at levels near the detection limit. The major products of curcumin biotransformation identified in rat plasma were curcumin glucuronide and curcumin sulfate whereas hexahydrocurcumin, hexahydrocurcuminol, and hexahydrocurcumin glucuronide were present in small amounts. To test the hypothesis that curcumin metabolites resemble their progenitor in that they can inhibit COX-2 expression, curcumin and four of its metabolites at a concentration of 20 microM were compared in terms of their ability to inhibit phorbol ester-induced prostaglandin E2 (PGE2) production in human colonic epithelial cells. Curcumin reduced PGE2 levels to preinduction levels, whereas tetrahydrocurcumin, previously shown to be a murine metabolite of curcumin, hexahydrocurcumin, and curcumin sulfate, had only weak PGE2 inhibitory activity, and hexahydrocurcuminol was inactive. The results suggest that (a) the major products of curcumin biotransformation by hepatocytes occur only at low abundance in rat plasma after curcumin administration; and (b) metabolism of curcumin by reduction or conjugation generates species with reduced ability to inhibit COX-2 expression. Because the gastrointestinal tract seems to be exposed more prominently to unmetabolized curcumin than any other tissue, the results support the clinical evaluation of curcumin as a colorectal cancer chemopreventive agent.

Effects of dietary curcumin on glutathione S-transferase and malondialdehyde-DNA adducts in rat liver and colon mucosa: relationship with drug levels.

Curcumin prevents colon cancer in rodent models. It inhibits lipid peroxidation and cyclooxygenase-2 (COX-2) expression and induces glutathione S-transferase (GST) enzymes. We tested the hypothesis that 14 days of dietary curcumin (2%) affects biomarkers relevant to cancer chemoprevention in the rat. Levels of inducible COX-2, as reflected by prostaglandin E(2) production by blood leukocytes, were measured ex vivo. Total GST activity and adducts of malondialdehyde with DNA (M(1)G), which reflect endogenous lipid peroxidation, were measured in colon mucosa, liver, and blood leukocytes. Curcumin and its metabolites were analyzed by high-performance liquid chromatography in plasma, and its pharmacokinetics were compared following a diet containing 2% curcumin versus intragastric (i.g.) administration of curcumin suspended in an amphiphilic solvent. The curcumin diet did not alter any of the markers in the blood but increased hepatic GST by 16% and decreased colon M(1)G levels by 36% when compared with controls. Administration of carbon tetrachloride during the treatment period increased colon M(1)G levels, and this increase was prevented by dietary curcumin. Dietary curcumin yielded low drug levels in the plasma, between 0 and 12 nM, whereas tissue concentrations of curcumin in liver and colon mucosa were 0.1--0.9 nmol/g and 0.2--1.8 micromol/g, respectively. In comparison with dietary administration, suspended curcumin given i.g. resulted in more curcumin in the plasma but much less in the colon mucosa. The results show that curcumin mixed with the diet achieves drug levels in the colon and liver sufficient to explain the pharmacological activities observed and suggest that this mode of administration may be preferable for the chemoprevention of colon cancer.

The comparative toxicity of chlorambucil and chlorambucil-spermidine conjugate to BALB/c mice.

The acute intraperitoneal toxicities of chlorambucil and chlorambucil-spermidine conjugate have been compared, in mice. Both compounds were neurotoxic and also caused a prolonged fall in bodyweight and a depletion of lymphocyte numbers associated with a fall in the total leukocyte count and loss of spleen and thymus weight. Alanine aminotransferase and aspartate aminotransferase activities and blood urea nitrogen concentration were increased at 24 h after conjugate administration, but had returned to normal at 72 h. Chlorambucil significantly decreased blood urea nitrogen concentration for 72 h, but did not affect transferase activity. Tissue concentrations of conjugate were measurable in liver and kidney for 12 days and lung for 5 days after dosing. The toxicity of both compounds was cumulative. In mol/kg, the chlorambucil-spermidine conjugate was 10-fold more toxic than chlorambucil, on the basis of their neurotoxicity, but only 2- to 3-fold more toxic on the basis of their effects on lymphocyte depression. The increased toxicity of the conjugate does not improve its therapeutic index relative to chlorambucil.

Cell-specific loss of cytochrome P450 2B1 in rat lung following treatment with pneumotoxic and non-pneumotoxic trialkylphosphorothioates.

This study was designed to test the hypothesis that the reduction in cytochrome P450 (CYP) 2B1 content and activity of rat lung microsomes, following dosing with pneumotoxic trimethylphosphorothioates, results from damage to specific cell types. Of the lung cells exhibiting immunolabelling for CYP2B1, only type I cells showed signs of susceptibility to the pneumotoxins O,O.S-trimethylphosphorothioate and O,S,S-trimethylphosphorodithioate. While most type I cells became necrotic, type II and Clara cells showed no signs of injury, despite their gradual loss of CYP2B1, as detected by immunogold labelling. This loss of labelling was accompanied by a 75% reduction in the immunoreactive CYP2B1 content and an 85% reduction in pentoxyresorufin O-dealkylase activity in lung microsomes. In contrast, the non-pneumotoxic analogue O,O,S-trimethylphosphorodithioate, differing from O,O,S-trimethylphosphorothioate by only the presence of a P = S rather than a P = O moiety, caused an even more rapid fall in pulmonary pentoxyresorufin O-dealkylase activity, but only a slight reduction in the microsomal content of CYP2B1. The recovery of this activity began within 12 hr of dosing. O,O,S-Trimethylphosphorodithioate, which acts as a suicidal inhibitor of pulmonary CYP2B1, did not cause any detectable lung injury or increase in cell division. These results are consistent with the initial reduction in both enzyme content and activity caused by the P = O - containing pneumotoxins resulting, almost entirely, from death of type I cells. Subsequent reductions that occur long after clearance of the toxin may be exacerbated by the onset of mitosis in Clara and type II cells.

Association of tamoxifen biliary excretion rate with prior tamoxifen exposure and increased mdr1b expression.

ATPase transporter proteins are commonly found in the hepatocyte canalicular membrane. Some of these, in particular the multidrug resistance (mdr1b) gene, have been previously demonstrated to be inducible genes. In this study, we found that tamoxifen induced expression of the mdr1b gene in the liver up to 40-fold after 14 days' exposure to tamoxifen in the diet at a concentration of 420 ppm. As tamoxifen and its metabolites are primarily excreted into the bile, we investigated if the increased expression of mdr1b in the liver following tamoxifen exposure had any effect on its excretion in rats. We found that the excretion of tamoxifen and its metabolites into bile was increased from 8 +/- 1% to 51 +/- 18% (mean +/- SD) of an administered dose of 180 nmol/kg over a collection period of 3 hr in rats that had received tamoxifen (35 mg/kg) orally for 12 days (plus a 3-day rest) prior to the experiment. These data suggest that prolonged treatment with tamoxifen may result in lower serum and tumour concentrations, due to a self-mediated enhancement of excretion via mdr1b gene-encoded P-glycoprotein. This may have implications for other drugs sharing the same route of excretion and co-administered with tamoxifen.

Protection against chemical-induced lung injury by inhibition of pulmonary cytochrome P-450.

Protection afforded by trialkyl phosphorothionates against the lung injury caused by trialkyl phosphorothiolates probably results from the inhibition by the P = S moiety of the thionates, of one or more pulmonary cytochrome P-450 isozymes. The aromatic hydrocarbons p-xylene and pseudocumene also protect against this injury and inhibit some P-450 isozymes, but by a different mechanism. OOS-Trimethylphosphorothionate and p-xylene were compared as protective agents against the effect of OOS-trimethylphosphorothiolate and two other lung toxins ipomeanol and 1-nitronaphthalene that are known to be activated by cytochrome P-450. The effects of these protective compounds, in vivo, on pulmonary cytochrome P-450 activity were also determined. Both compounds inhibited pentoxyresorufin O-deethylase activity, but not ethoxyresorufin O-deethylase. The phosphorothionate was most effective against lung injury caused by the phosphorothiolates and 1-nitronaphthalene, whereas p-xylene was much more effective against ipomeanol. beta-Naphthoflavone, which induces pulmonary ethoxyresorufin O-deethylase activity, did not protect against phosphorothiolate or 1-nitronaphthalene injury, and it was only marginally effective in decreasing the toxicity of ipomeanol.

A comparison of the acute pathology induced by 3-phenylamino-1,2-propanediol (PAP) and its mono-oleoyl ester in rodents with the toxic oil syndrome in man.

Phenylamino-1,2 propanediol (PAP) and its mono-oleoyl ester have been identified in samples of the cooking oil thought to be responsible for the Toxic Oil Syndrome (TOS) which occurred in Spain in 1981. The acute toxicity of PAP and its mono-oleoyl ester have been examined in rats and mice, after daily administration for periods of up to 14 days to determine whether these compounds could produce any of the pathologies of TOS. Even at the highest dose, the 1-mono-oleoyl ester of 3-phenylamino-1,2 propanediol did not cause any toxicity in rats or mice when given intraperitoneally. 3-Phenylamino-1,2 propanediol, however, was toxic when administered to rats by this route. After 6-10 consecutive daily doses of PAP, at the highest dose administered (350 mg kg-1), all of the rats became unwell. Postmortem examination showed that the major pathology present was massive pulmonary thromboembolism. Further investigations of the toxicity of PAP after intravenous administration showed that it was not directly vasotoxic. The pulmonary thromboembolism seen with intraperitoneally administered PAP was due to the toxic effect of PAP on the mesenteric tissue and blood vessels, causing thrombosis which subsequently embolised the blood vessels in the lung. Intra-gastric administration of PAP caused no toxicity in rats. Comparatively, the pathology seen after intraperitoneal administration of PAP was not thought to be representative of the pathology of the toxic oil syndrome in man.

Evaluation of the cancer chemopreventive efficacy of rice bran in genetic mouse models of breast, prostate and intestinal carcinogenesis.

Brown rice is a staple dietary constituent in Asia, whereas rice consumed in the Western world is generally white, obtained from brown rice by removal of the bran. We tested the hypothesis that rice bran interferes with development of tumours in TAg, TRansgenic Adenocarcinoma of the Mouse Prostate (TRAMP) or Apc(Min) mice, genetic models of mammary, prostate and intestinal carcinogenesis, respectively. Mice received rice bran (30%) in AIN-93G diet throughout their post-weaning lifespan. In TAg and TRAMP mice, rice bran did not affect carcinoma development. In TRAMP or wild-type C57Bl6/J mice, dietary rice bran increased kidney weight by 18 and 20%, respectively. Consumption of rice bran reduced numbers of intestinal adenomas in Apc(Min) mice by 51% (P<0.01), compared to mice on control diet. In parallel, dietary rice bran decreased intestinal haemorrhage in these mice, as reflected by increased haematocrit. At 10% in the diet, rice bran did not significantly retard Apc(Min) adenoma development. Likewise, low-fibre rice bran (30% in the diet) did not affect intestinal carcinogenesis, suggesting that the fibrous constituents of the bran mediate chemopreventive efficacy. The results suggest that rice bran might be beneficially evaluated as a putative chemopreventive intervention in humans with intestinal polyps.

The interaction between phosphorothionate insecticides, pneumotoxic trialkyl phosphorothiolates and effects on lung 7-ethoxycoumarin O-deethylase activity.

A number of phosphorothionate (P = S) insecticides, including bromophos and fenitrothion, prevent trialkyl phosphorothiolate (P = O)-induced lung toxicity and the resulting increase in lung weight normally observed at 3 days in the rat. Measurement of 7-ethoxycoumarin O-deethylase (7-EC) activity after both phosphorothionate and phosphorothiolate dosing revealed differing patterns of loss of enzyme activity. Depletion of 7-EC activity by phosphorothionates was maximal between 2 and 10 h after dosing, with recovery between 24 and 72 h. Phosphorothiolates, however, appear to cause two phases of loss of 7-EC activity, an initial fall of approximately 30% observed at 2 h and a secondary fall, maximal on day 3, with loss of 97% of activity, apparently associated with the pathological changes in the lung. It is suggested that oxidative metabolism of phosphorothionates known to occur at the P = S moiety, with suicidal loss of P-450, may then prevent oxidative activation of an S-methyl on the phosphorothiolates, the most likely site for production of a reactive intermediate capable of damaging the lung. Lung 7-EC in rat is sensitive to concentrations of the phosphorothionates bromophos and fenitrothion at 5-25 times less than those causing loss of liver 7-EC activity and at doses 125-600 times less than their LD50s. If repeated in man this may have implications for personnel occupationally exposed to these compounds.

Structure-activity relationships of some pyrethroids in rats.

The intravenous toxicity to the rat of 36 pyrethroids has been examined. With two exceptions they cause either (1) T-syndrome, consisting of aggressive sparring, sensitivity to external stimuli, fine progressing to gross whole body tremor and prostration or (2) CS-syndrome, consisting of pawing and burrowing behaviour, salivation, coarse tremor, progressing to sinuous writhing (choreoathetosis) and clonic seizures. The two exceptions presented a TS-syndrome with salivation associated with the T-syndrome. No clearcut relationship between chemical structure and symptoms of poisoning has emerged through some generalisations are discussed.

The acute toxicity of some organolead and organotin compounds in the rat, with particular reference to a gastric lesion.

The acute toxicity of triphenyl, tributyl and dibutyllead in the rat appears to result from their irritating properties to membranes regardless of their route of administration. A more detailed examination of their oral toxicity has shown that a single dose inhibits gastric emptying with resulting fluid distension. A similar effect was observed with a number of other organolead and organotin compounds. Analysis of gastric fluid from pyloric ligated rats following an oral organolead dose showed increased concentrations of Na+ and glucose and decreased concentrations of H+ and K+. Dilation of the gastric mucosal microcirculation was also observed along with desquamation of surface mucous cells which resulted in shallow erosions of the upper portions of the gastric glands. It is suggested that oral dosing with a number of organoleads and, possibly, organotins has an effect on mucosal membranes that results in erosion of the gastric mucosa and increases permeability of the mucosal microcirculation. This would allow leakage of plasma constituents into interstitial fluid and ultimately the gastric lumen with back-diffusion of H+ into the gastric mucosa. Such a lesion would impair gastric function and reduce normal food and water consumption. The starvation and dehydration may then be the important factors for death that follows oral dosing with organolead and organotin compounds.

Ultrastructural effects of phosphorothionates and other inhibitors of lung monooxygenases: protection against trialkylphosphorothiolate-induced lung injury.

An oral dose (25 mg/kg) of O,O,S-triethylphosphorothiolate (OOSEtO) to rats results in selective injury of type I pneumocytes, degranulation of Clara cells, and pronounced increase in lung weight. A dose (12.5 mg/kg) of the related compound O,O,S-trimethylphosphorothionate (OOSMeS) causes neither injury nor degranulation but, when administered 2 h before OOSEtO (25 mg/kg), protects against all the signs of lung injury that would otherwise result from this dose of the compound. The administration of OOSMeS also results in the formation of large, electron-lucent granules within the apical cytoplasm of the Clara cells. The granules are not birefringent, and histochemical procedures indicate that they do not contain carbohydrate but may consist of lipid accumulated around a proteinaceous core. Similar granules are also observed after administration of p-xylene, pseudocumene, and the pesticide bromophos. These compounds, like OOSMeS, inhibit 7-ethoxycoumarin O-deethylase activity in the lung and are capable of protecting against trialkylphosphorothiolate toxicity. This inhibition of 7-ethoxycoumarin O-deethylase activity suggests loss of pulmonary cytochrome P-450. This loss may account for both the protective action of these compounds and the formation of abnormal granules within Clara cells.

Selective inhibition and induction of CYP activity discriminates between the isoforms responsible for the activation of butylated hydroxytoluene and naphthalene in mouse lung.

1. Selective induction and inhibition experiments have been used to identify the cytochrome P450 (CYP) isoforms responsible for butylated hydroxytoluene (BHT) bioactivation in mouse lung. 2. Pre-treatment of BALB/c mice with O,O,O-trimethylphosphorothioate (OOOMeP(S)), which prevented all the signs of toxicity observed following BHT treatment, inhibited the pulmonary activity of pentoxyresorufin O-dealkylase (PROD) and coumarin hydroxylase but not 4-nitrophenol hydroxylase. 3. Pulmonary coumarin hydroxylase activity was greater in DBA than in BALB/c mice but the severity of BHT-induced lung injury was similar. 4. Pre-treatment with pyrazole, which exacerbated BHT-induced lung injury, did not affect pulmonary coumarin hydroxylase or 4-nitrophenol hydroxylase activity but increased that of PROD. 5. Pre-treatment with OOOMeP(S) prevented the lethargy and weight-loss associated with naphthalene poisoning but not the pulmonary injury. Pre-treatment with pyrazole did not exacerbate naphthalene-induced injury. 6. Members of both CYP2F and 2B sub-families have been shown to exhibit PROD activity and 2F2 activates naphthalene in mouse lung. The current studies, however, indicate that 2F2 is unlikely to be a significant component of PROD activity in mouse lung. 2F2, like coumarin hydroxylase (2A5) and 4-nitrophenol hydroxylase (2E1), is not responsible for the pulmonary activation of BHT, which is largely attributable to an isoform of 2B, probably 2B10.

The toxicity and neuropathology of 2,4,5-tribromoimidazole and its derivatives in rats.

2,4,5-tribromoimidazole and its 1-n-butylcarboxylate and 1-dimethylcarbamoyl derivatives, when administered to rats, induced poisoning typical of uncouplers of oxidative phosphorylation. At 48 h rats surviving a single toxic dose of 20-60 mg/kg developed permanent incoordination of the hindlimbs in the absence of brain oedema. Neuropathologic examination of brain and spinal cord from perfused fixed rats at 24 h revealed neuronal necrosis and chromatolysis in the vestibular nucleus, the outer parietal neocortex and red nucleus. Chromatolysis and necrosis in these areas had increased at 72-96 h and were also observed in the deeper layers of the neocortex, the medial entorhinal cortex, the reticular formation, the grey matter of the spinal cord extending into the ventral horns, the dorsal, and ventral cochlear nuclei and the deep cerebellar nuclei, in decreasing order of severity. Neuronal necrosis was accompanied by an increased glial response, including neuronophagia and at 16 days with astroglial hypertrophy and hyperplasia.

The determination of O,S,S-trimethylphosphorodithioate in the plasma and various tissues of rats using high-resolution gas chromatography with nitrogen-phosphorus detection.

A method has been developed for the determination of O,S,S-trimethylphosphorodithioate in the plasma, lung, liver, brain and thymus of rats using high-resolution gas chromatography. The organophosphorus compound was extracted from the biological sample with ethyl acetate and analysed on a carbowax 20M fused-silica capillary column with a nitrogen-phosphorus specific detector. O,S,S-Triethylphosphorodithioate was used as an internal standard added to the sample before extraction. The sensitivity of the method allowed the compound to be measured in 0.1-ml aliquots of plasma or in 20-mg wet weight of tissue down to a level of 5 ng/sample. The method has been applied to a pharmacokinetic study in the rat after an oral or intravenous dosage with 25 mg/kg of O,S,S-trimethylphosphorodithioate.

Cytochrome P450 2B isoenzymes are responsible for the pulmonary bioactivation and toxicity of butylated hydroxytoluene, O,O,S-trimethylphosphorothioate and methylcyclopentadienyl manganese tricarbonyl.

O,O,S-Trimethylphosphorothioate and methylcyclopentadienyl manganese tricarbonyl damage the type I pneumocytes of the alveolar epithelium in rats. Butylated hydroxytoluene causes similar damage in mice. The toxicity of these compounds is dependent on their bioactivation by the cytochrome P-450 (CYP) system. A range of compounds, that modifies the activity of specific CYP isoenzymes, has been used to establish those particular isoenzymes involved in bioactivation. Pulmonary toxicity was assessed by measurement of lung weight and changes in the activity of gamma-glutamyltranspeptidase and alkaline phosphatase in bronchoalveolar lavage fluid. O,O,S-Trimethylphosphorodithioate, bromophos, p-xylene and 2,4-dichloro-(6-phenyl-phenoxy)ethylamine all inhibited the dealkylation of pentoxyresorufin, an indicator of CYP2B1 activity, and also prevented pulmonary toxicity. There was a significant negative correlation between the level of pulmonary pentoxyresorufin dealkylation after pretreatment with O,O,S-trimethylphosphorodithioate and the severity of lung injury. This pretreatment also reduced the toxicity of butylated hydroxytoluene by a factor of 20 and methylcyclopentadienyl manganese tricarbonyl by a factor of 10. Modification of the activity of CYP1A1, CYP2E1 and CYP4B1 did not alter the toxicity of these compounds. These results indicate that pulmonary CYP2B1 is responsible for the bioactivation and toxicity of O,O,S-trimethylphosphorothioate and methylcyclopentadienyl manganese tricarbonyl in rats and the orthologous 2B isoenzyme in mice activates butylated hydroxytoluene.

Pulmonary cytochrome P450 in rats exposed to formaldehyde vapor.

The lungs of rats exposed to formaldehyde vapor, for 6 hr/day over 4 consecutive days, were examined for signs of injury and for changes in the level, or activity, of cytochrome P450. The animals were supplied with 10 ppm formaldehyde vapor generated, in two separate experiments, either from an aqueous solution of formaldehyde or from heated paraformaldehyde. All rats were exposed for 6 hr, on each of 4 consecutive days, and killed 1 day after the onset of the fourth period of exposure. The lung weights and gains in body weight of exposed animals were indistinguishable from those of their controls. Lungs from the formaldehyde-exposed animals did not show any signs of injury, even at the ultrastructural level. Bronchoalveolar lavage samples from exposed animals showed no increase in alkaline phosphatase or gamma-glutamyl transpeptidase activity. The total concentration of cytochrome P450 in the lungs of exposed animals was similar to that found in their controls. The P450 activity of pulmonary microsomes from exposed animals was not significantly different from that obtained with samples from the control animals. These results indicate that repeated exposure to 10 ppm formaldehyde vapor does not injure the deep lung of rats and has no effect on the level of lung P450 or on its activity against substrates for the most common pulmonary forms of this enzyme.

Inhibition and induction of cytochrome P450 isoenzymes in rat lung.

The O-dealkylation of pentoxyresorufin, a substrate for P450 2B1, was decreased in lung microsomes from rats dosed with O,O,S-trimethylphosphorodithioate, O,O,O-trimethylphosphorothioate, bromophos, fenitrothion, p-xylene and 2,4-dichloro-(6-phenylphonoxy)ethylamine. This activity was decreased by antibodies to P450 2B1 but unaffected by antibodies to P450 1A1 or 4B1. This reduction reflected both inactivation and destruction of P450 2B1; destruction of this protein was particularly marked after bromophos and fenitrothion. Pyrazole was the only compound in this study to induce the O-dealkylation of pentoxyresorufin. None of these compounds altered the rate of ethoxyresorufin O-dealkylation, an indicator of P450 1A1 activity, but this activity was induced greatly by both Aroclor and beta-naphthoflavone, p-Xylene was the only compound to decrease P450 4B1 activity, as determined by the N-hydroxylation of 2-aminofluorene. In the liver, bromophos, fenitrothion, p-xylene and 2,4-dichloro-(6-phenylphonoxy)ethylamine all had marked effects on the O-dealkylation of ethoxyresorufin and pentoxyresorufin but, at the dose used, O,O,O-trimethylphosphorothioate and O,O,S-trimethylphosphorodithioate had minimal effects in this tissue. Thus, both O,O,O-trimethylphosphorothioate and O,O,S-trimethylphosphorodithioate are exquisitely selective inhibitors of pulmonary P450 2B1 activity. Their use, together with pyrazole, will facilitate future studies of the pulmonary activation of toxins by P450 2B1.