Transesterification of p-hydroxybenzoate esters (parabens) by human intestinal (Caco-2) cells.

p-Hydroxybenzoate ester (paraben) preservatives are used in numerous orally administered products. The recognized route of metabolism for parabens is hydrolysis to p-hydroxybenzoic acid followed by conjugation and excretion. However, in the presence of alcohols, a presystemic transesterification pathway not previously reported for the human intestine can occur. Using human intestinal (Caco-2) cells, it was observed that hydrolysis of parabens to p-hydroxybenzoic acid is reduced markedly by ethanol concentrations that can occur in the human intestine, 0.25-0.5% (v/v). Ethanol concentrations of 1.0-2.5% (v/v) were optimal for transesterification to ethylparaben in Caco-2 cell homogenates. The kinetics of the transesterification reaction with regard to ethanol concentration (0-20%), time, pH (3-9), protein concentration (1-5 mg ml-1) and substrate concentration (6.25-200 microM) as well as the effects of different alcohols were studied. The Km and Vmax values for transesterification with ethanol for methyl, propyl, butyl, heptyl and octyl parabens were 449.7, 165.7, 86.1, 24.2 and 45.9 microM and 114.4, 37.5, 19.5, 7.5 and 7.6 micromol h-1 mg-1 Caco-2 cell protein, respectively. The Vmax values for transesterification of methylparaben with ethanol, propan-1-ol, butan-1-ol were 114.4, 5.1 and 4.9 micromol h-1 mg-1, respectively. Collectively, the kinetic data demonstrate that the enzyme responsible for the transesterification reaction has a preference for short-chain esters and represents the first report of transesterification in human intestinal cells. An implication of this mechanism is that alcohol-containing in vitro biosystems or protocols for the study of parabens disposition could generate transesterified artefacts. The clinical or toxicological implication is that, following co-ingestion of ester compounds with ethanol, transesterification could provide the basis for a previously unrecognized drug-alcohol interaction.

Mechanisms of chromium toxicity, carcinogenicity and allergenicity: review of the literature from 1985 to 2000.

Laboratory and clinical reports about the pathogenesis of the carcinogenicity and allergenicity of chromium compounds published between 1985 and 2000 have been reviewed as a basis for consideration of the pathogenetic mechanisms involved. There is good evidence from the clinic and the laboratory that Cr[VI] is the ion responsible for most of the toxic actions, although much of the underlying molecular damage may be due to its intracellular reduction to the even more highly reactive and short-lived chemical species Cr[III] and Cr[V]. Exposure to Cr[VI] can result in various point mutations in DNA and to chromosomal damage, as well as to oxidative changes in proteins and to adduct formation. The relative importance of these effects of chromium ions and of the free oxidising radicals they may generate in the body in causing tumours and allergic sensitisation remain to be demonstrated. Biochemical studies of the DNA-damaging effects and of the pathogenesis of the allergic reactions to chromium ions have not kept up with advances in understanding of the molecular basis of the effects of other carcinogens and allergens.

Fibronectin-mediated hepatocyte shape change reprograms cytochrome P450 2C11 gene expression via an integrin-signaled induction of ribonuclease activity.

A major limitation to the use of rat hepatocytes in the study of drug metabolism and toxicity is the rapid loss of CYPs. We demonstrate that the culture of rat hepatocytes results in a rapid loss of liver-specific CYP2C11 mRNA and transcripts encoding the general housekeeping gene copper-zinc superoxide dismutase (CuZnSOD) as well as poly(A(+)) mRNA. These losses are accelerated by fibronectin, which has no effect on the transcription of CYP2C11 and CuZnSOD. However, fibronectin, an extracellular matrix protein involved in cell adhesion and spreading, induces ribonuclease (RNase) activity. Fibronectin also increases hepatocyte diameter and data are presented that cell spreading is involved in the loss of both CYP2C11 and CuZnSOD mRNAs. The use of functional blocking antibodies demonstrates that fibronectin is operating through its alpha(5)beta(1) integrin receptor and genistein, a tyrosine kinase inhibitor, prevents hepatocyte spreading, RNase induction, and CYP2C11 mRNA loss. Collectively, the data indicate that hepatocytes in vitro actively promote the extinction of their phenotype via the autocrine effects of fibronectin rather than the current consensus that they simply lose differentiated function, such as CYP2C11 expression, through the absence of extracellular matrix proteins. The substrate specificity of the ribonuclease induced is also considered.

Effect of the adrenal 11-beta-hydroxylase inhibitor metyrapone on human hepatic cytochrome P-450 expression: induction of cytochrome P-450 3A4.

The drug metyrapone in the presence of glucocorticoid has been shown to induce the expression of rat hepatic cytochrome P-450 (CYP) 1A1 mRNA in vivo and in vitro through disruption of endogenous CYP1A1 regulator homeostasis and without either compound's binding to the aryl hydrocarbon receptor. Addition of metyrapone to human liver cancer cell cultures, with or without dexamethasone, did not induce CYP1A1 mRNA, in contrast to the aryl hydrocarbon receptor ligand beta-naphthoflavone. Addition of metyrapone to primary cultures of human hepatocytes also failed to induce detectable levels of CYP1A1 mRNA or CYP1A protein in two separate preparations, whereas the treatment with 2,3,7,8-tetrachlorodibenzo-rho-dioxin or omeprazole induced detectable levels of CYP1A1 mRNA in one preparation and CYP1A protein in both preparations. Addition of metyrapone to human hepatocyte cultures resulted in the induction of CYP3A4 expression. The pregnane X receptor (PXR), which has recently been shown to mediate the transcriptional induction of CYP3A4 expression in response to rifampicin, was activated by metyrapone in CV-1 cells transiently cotransfected with an expression vector encoding the human PXR and a reporter construct containing the everted repeat sequence that confers CYP3A4 induction responsiveness to inducers within its promoter. Metyrapone activated the human PXR at concentrations that also resulted in the induction of CYP3A4 in human cultured hepatocytes. Metyrapone treatment is therefore unlikely to result in the induction of CYP1A1 but may induce the expression of CYP3A4 in humans.

Rapid induction of NF-kappaB binding during liver cell isolation and culture: inhibition by L-NAME indicates a role for nitric oxide synthase.

This study is the first to demonstrate activation of NF-kappaB binding just 10 minutes into the commonly employed hepatocyte isolation procedure. It is further reported that the anti-oxidant Trolox can prevent the induction of NF-kappaB during the well established hepatocyte isolation procedure but not during their subsequent culture. However both phases of NF-kappaB activation are inhibited by L-NAME intimating a role for NO production, via nitric oxide synthase. These findings demonstrate that at least 2 different signal transduction pathways are operative during hepatocyte isolation and culture. Thus further studies employing Trolox and L-NAME will help delineate how each pathway contributes to the generalised loss of liver function commonly observed in vitro.

Disruption of endogenous regulator homeostasis underlies the mechanism of rat CYP1A1 mRNA induction by metyrapone.

The transcriptional induction of the cytochrome P-450 1A1 (CYP1A1) gene by xenobiotics such as polyaromatic hydrocarbons is dependent on their interaction with the aryl hydrocarbon receptor. Administration of the structurally unrelated compounds metyrapone (a cytochrome P-450 inhibitor) or dexamethasone (a glucocorticoid) to male rats does not induce hepatic CYP1A1 mRNA. However, administration of both metyrapone and dexamethasone to male rats results in the induction of hepatic CYP1A1 mRNA expression. The induction response is mimicked in vitro in cultured rat hepatocytes by the addition of metyrapone and dexamethasone to a serum-free culture medium, suggesting that these compounds act directly on the liver in vivo to effect hepatic CYP1A1 mRNA induction. An examination of the characteristics of CYP1A1 induction by metyrapone and dexamethasone in combination in vitro indicate that at least 6 h of treatment is required for detectable levels of CYP1A1 mRNA to accumulate in hepatocytes. In contrast, beta-naphthoflavone, which is known to bind to the aryl hydrocarbon receptor to effect CYP1A1 gene expression, induces detectable levels of CYP1A1 mRNA within 2 h of treatment. CYP1A1 mRNA is also induced when hepatocytes are treated with metyrapone in combination with the protein synthesis inhibitor cycloheximide but not with dexamethasone in combination with cycloheximide, indicating that CYP1A1 mRNA induction is strictly dependent on the presence of metyrapone and suggesting that the metyrapone-associated induction of CYP1A1 mRNA is dependent on a loss of a constitutively expressed protein that functions to suppress CYP1A1 gene expression. The role of dexamethasone in metyrapone-associated induction of CYP1A1 is probably mediated through the glucocorticoid receptor since the glucocorticoid receptor antagonist RU486 reduces the levels of CYP1A1 mRNA induced by metyrapone and dexamethasone in combination. Increasing the levels of the photosensitizer riboflavin present in the culture medium 10-fold and exposure to light increases the levels of CYP1A1 mRNA induced by metyrapone and dexamethasone in combination in vitro, suggesting that photoactivation of inducing medium constituent(s) might be required for induction. Failure to induce CYP1A1 mRNA by co-administration of metyrapone and dexamethasone in hepatocytes cultured in a balanced salt solution with or without photoactivation indicates that induction is dependent on a photoactivated component of the culture medium and not on metyrapone or dexamethasone alone. The addition of tryptophan in the presence of riboflavin to the balanced salt solution restores CYP1A1 mRNA induction by metyrapone alone and induction is increased when medium is exposed to light, indicating that induction is dependent on tryptophan photoactivation in vitro. Metyrapone failed to compete with 2,3,7,8-tetrachlorodibenzo-p-dioxin for specific binding to the aryl hydrocarbon receptor in rat liver cytosolic fractions. These results suggest that CYP1A1 might be induced in rats by metyrapone through an indirect mechanism associated with an elevation in the level of an endogenously generated inducer such as photoactivated product(s) of tryptophan and not because of metyrapone's interacting with the aryl hydrocarbon receptor. The dependence of CYP1A1 induction on dexamethasone or cycloheximide suggests that derepression by a glucocorticoid receptor-modulated negative-acting factor of CYP1A1 gene expression might be critical to induction by metyrapone.

Developmental changes in the constitutive and inducible expression of cytochrome P450 3A2.

Using a CYP3A2-specific oligonucleotide and an antipeptide antibody raised against the C terminus of CYP3A2 (VINGA) it is demonstrated that metyrapone administration to adult (12 weeks old) but not immature (3 weeks old) male Sprague Dawley rats induces the hepatic expression of CYP3A2 mRNA and protein. The constitutively expressed level of CYP3A2 protein in adult male rats is markedly lower than the levels expressed in immature rats as determined using the anti-VINGA antibody, in contrast to previous reports using antibodies that do not discriminate between CYP3A forms. Hepatic microsomal CYP3A2 protein expression, examined between 3 and 15 weeks of age, is extinguished between 9 and 12 weeks of age in contrast to immunoreactive CYP3A protein (determined using a nonselective antibody) and CYP3A-dependent androstenedione 6beta-hydroxylase activity. These data suggest that the regulation of the induction of CYP3A2 is developmentally controlled and that the major expressed adult form(s) of constitutively expressed CYP3A is not CYP3A2.

Effect of vitamin A deficiency on the expression of low affinity glucocorticoid binding site activity and glucocorticoid-dependent induction of CYP3A2 in rat liver.

Maintenance of rats on a vitamin A-deficient diet resulting in undetectable levels of plasma retinol and significant reductions in relative testes weight compared to age-matched controls leads to the loss of liver membrane-bound low affinity glucocorticoid binding site (LAGS) activity without any effects on the levels of constitutively expressed CYP3A2 protein. Subsequent daily administration of retinol acetate to vitamin A-deficient rats results in the re-expression of LAGS activity to control levels by 7 days. To determine any role for the LAGS in the modulation of CYP3A2 expression by glucocorticoids, a single dose of dexamethasone 21-phosphate was administered to vitamin A-deficient rats and vitamin A-deficient rats induced to re-express LAGS by daily retinol acetate treatment. Retinol acetate administration alone induces CYP3A2 protein to apparent maximal levels since dexamethasone 21-phosphate does not further increase the induction response. However, CYP3A2 remains inducible to dexamethasone 21-phosphate in vitamin A-deficient rats. These data suggest that vitamin A status affects the expression of LAGS and CYP3A2 but that glucocorticoids regulate the induction of CYP3A2 by a mechanism(s) independent of their interaction with the LAGS.

Temperature-sensitive mRNA degradation is an early event in hepatocyte de-differentiation.

The isolation and culture of metabolically active hepatocytes by proteolytic digestion of the extracellular matrix of the liver results in the transcriptional silencing of liver-specific genes encoding cytochromes P-450 (CYP) and albumin together with an induction of cellular RNase activity. The levels of albumin mRNA are maintained in cultured hepatocytes at similar levels to that present in the intact liver for at least 24 h, whereas the major constitutively expressed CYP2C11 mRNA is rapidly degraded. Hepatocytes heat-shocked at 40 degrees C during the isolation procedure (which results in an induction of heat-shock protein mRNA species) blocks the increase in RNase activity and abrogates the loss of CYP2C11 mRNA for at least 4 h. Cycloheximide-dependent inhibition of protein synthesis blocks the temperature-dependent induction of heat-shock proteins without affecting the protection afforded to CYP2C11 mRNA, indicating that CYP2C11 mRNA levels are not directly dependent on heat-shock protein induction and suggesting that the induction of RNase activity might be responsible for the specific loss of CYP2C11 mRNA in hepatocytes isolated at 37 degrees C. Differential rates of degradation of CYP2C11 transcribed in vitro and of albumin mRNA are observed in the presence of cellular extracts from cultured hepatocytes isolated at 37 degrees C (which have maximally induced levels of cellular RNase activity) but not in comparable extracts from cultured hepatocytes isolated at 40 degrees C, supporting the hypothesis that an RNase activity is induced in culture that specifically degrades CYP2C11 mRNA but not albumin mRNA. These results suggest that an early event in hepatocyte de-differentiation involves the induction of RNase activity in addition to transcriptional silencing of liver-specific genes and that the induced RNase activity demonstrates specificity within liver-specific gene products.

Glucocorticoid receptor-independent transcriptional induction of cytochrome P450 3A1 by metyrapone and its potentiation by glucocorticoid.

Metyrapone administration to 21- and 90-day-old male rats causes a transcriptional induction of the hepatic glucocorticoid-inducible CYP3A1 gene within an hour as determined by nuclear run-on experiments. Analyses performed 24 hr after metyrapone administration in both ages of rat demonstrate that the transcriptional induction of CYP3A1 gene expression is followed by significant increases in CYP3A1 mRNA, CYP3A-immunoreactive microsomal protein and total microsomal cytochrome P450 (CYP). In 21-day-old rats, there is a significant increase in microsomal CYP3A dependent steroid 6 beta-hydroxylase activity but not in 90-day-old rats, possibly because of a slower clearance of this drug, which inhibits CYP activities. In hepatocytes cultured in serum- and glucocorticoid hormone-free medium, metyrapone alone induces CYP3A1 mRNA expression, which demonstrates that metyrapone transcriptionally induces hepatic CYP3A1 by a direct interaction with the liver. Metyrapone does not compete with the binding of the synthetic glucocorticoid and potent transcriptional CYP3A1 inducer dexamethasone to the glucocorticoid receptor (GR) in soluble fractions from liver. This suggests that metyrapone is not a ligand for the GR and induces CYP3A1 by a mechanism independent of the GR. Addition of glucocorticoid to cultured hepatocytes at levels that induce GR-dependent genes potentiate CYP3A1 mRNA induction by metyrapone without inducing CYP3A1 mRNA alone. A GR-dependent mechanism may therefore mediate the potentiation of CYP3A1 transcriptional induction by metyrapone. The CYP3A1 transcriptional inducer and glucocorticoid antagonist pregnenolone 16 alpha-carbonitrile at 100 microM blocks dexamethasone binding to the GR in 21-day-old rat liver soluble fractions but is less effective in 90-day-old rat liver soluble fractions in contrast with 10 microM glucocorticoid antagonist RU486, which is equally effective at blocking dexamethasone binding to the GR. The inability of pregnenolone 16 alpha-carbonitrile to fully compete with dexamethasone for cytosolic binding in adult animals suggests that there may exist variant receptors with different affinities for dexamethasone and pregnenolone 16 alpha-carbonitrile and may explain the mechanism by which low concentrations of dexamethasone potentiate the transcriptional induction of CYP3A1 mediated by high concentrations of pregnenolone 16 alpha-carbonitrile [J. Biol, Chem. 270:28917-28923 (1995)]. Examination of membrane-bound dexamethasone binding activity, with which other steroidal and nonsteroidal CYP3A inducers have been shown to compete, indicates that binding activity is detectable in 90- but not 21-day-old rat liver microsomes. The absence of membrane-bound glucocorticoid binding site activity and the presence of a functional CYP3A1 transcriptional response in 21-day-old rats suggest that membrane-bound glucocorticoid binding site activity is not involved in the transcriptional activation of CYP3A1 expression. These data suggest that both glucocorticoids and nonsteroidal compounds may trigger the transcriptional induction of CYP3A1 by a GR-independent mechanism that may be potentiated by a GR-dependent mechanism.

Validity and reliability of in vitro systems in safety evaluation.

In vitro data could make an important contribution to the application of the proposed scheme for the subdivision of the usual 10-fold safety factors (used in risk assessment for inter-species and inter-individual differences) into two separate aspects of toxicokinetics and toxicodynamics. Whereas toxicokinetics (or delivery of the chemical to its site of action via the general circulation) is amenable to direct in vivo measurement, toxicodynamics (or the assessment of the sensitivity of the target tissue to the presence of the chemical) is open to in vitro investigation. Human risk assessment requires human data to be able to replace any of the default safety (or uncertainty) factors (Renwick, 1993). Because human tissues are of limited availability, it is likely that the main quantitative contribution of in vitro data will be to allow chemical specific inter-species differences in toxicodynamics to replace the proposed default value. Although in vitro data from human tissues could be used to define human variability in target organ sensitivity (toxicodynamics) this would require a large number of specimens and the variability detected in vitro should be representative of that present in vivo.

Characteristics of a membrane-associated steroid binding site in rat liver.

The glucocorticoid dexamethasone binds a site in microsomes in a saturable manner which by competition studies also binds other classes of steroids. The characteristics of dexamethasone binding to microsomes is distinct from the cytosolic glucocorticoid receptor by virtue of a slower rate of association; a differential competition by the glucocorticoid receptor agonist triamcinolone acetonide and antagonist RU38486; and a lack of sensitivity to the reversible thiol reactive agent arsenite. However, both binding sites have a similar rate constant for complex dissociation; are sensitive to covalent thiol modification by N ethylmaleimide and iodoacetamide; and have a similar concentration-dependent sensitivity to the reversible thiol reactive agent methyl methanethiosulfonate. The binding of dexamethasone by microsomes therefore exhibits distinct properties from the soluble glucocorticoid receptor.

Induction of the cytochrome P450 3A subfamily in rat liver correlates with the binding of inducers to a microsomal protein.

The specific binding of the archetypal cytochrome P450 3A subfamily (CYP3A) inducer dexamethasone is examined in microsomal fractions since rat and human liver CYP3A is induced by glucocorticoids through a mechanism which is apparently independent of the cytosolic glucocorticoid receptor. Dexamethasone binds in a specific and saturable manner to microsomes with an affinity constant (Kd) of 59 +/- 12.9 nM which compares to a Kd of 2.3 +/- 0.17 nM in cytosol. The total receptor concentrations ([LR]emax) in microsomes and cytosol are 9.5 +/- 1.67 pmols/mg protein and 410 +/- 167 fmol/mg protein respectively. The microsomal binding of dexamethasone is antagonised by several transcriptional and/or post-transcriptional CYP3A inducers with decreasing potency pregnenolone 16 alpha carbonitrile > metyrapone > phenobarbitone. Troleandomycin, which indirectly induces CYP3A1 in vivo and by protein stabilisation, does not antagonise the binding of dexamethasone in microsomes. The transcriptional and/or post-transcriptional induction of CYP3A may therefore be associated with the interaction of inducers with a microsomal protein.

Induction of rat hepatic glucocorticoid-inducible cytochrome P450 3A by metyrapone.

The bipyridyl compound metyrapone is a potent inhibitor of cytochromes P450, a gene superfamily of haemoproteins involved in the metabolism of many xenobiotics as well as endogenous compounds such as steroid hormones. Administration of metyrapone to male rats induces the expression of the cytochrome P450 sub-family 3A (CYP3A). In order to determine whether metyrapone was causing the induction of CYP3A by blocking endogenous glucocorticoid metabolism, CYP3A levels were examined in rat hepatocytes cultured in serum-free medium supplemented with hydrocortisone 21-hemisuccinate plus or minus metyrapone. Western blotting indicated that metyrapone alone induces CYP3A and that hydrocortisone 21-hemisuccinate is ineffective. However, hydrocortisone 21-hemisuccinate enhanced the levels of CYP3A induced by metyrapone. In contrast, glucocorticoid-inducible tyrosine aminotransferase (TAT) activity was unaffected by metyrapone but metyrapone enhanced the levels induced by hydrocortisone 21-hemisuccinate. An examination of the metabolism of hydrocortisone by rat hepatocytes in vitro indicated that metyrapone perturbed the catabolism of hydrocortisone under conditions which give rise to an enhancement of hydrocortisone 21-hemisuccinate and hydrocortisone-dependent TAT induction. However, evidence is presented to suggest that such a perturbation of hydrocortisone metabolism could not account for the glucocorticoid potency amplifying property of metyrapone. Thus the induction of CYP3A and the enhancement of glucocorticoid-mediated TAT induction appears not to be associated with any perturbation in glucocorticoid metabolism but with some other as yet undefined mechanism(s).

Alteration of transcription factor mRNAs during the isolation and culture of rat hepatocytes suggests the activation of a proliferative mode underlies their de-differentiation.

The commonly observed loss of liver specific phenotype regularly described in rat hepatocyte culture is typified by the loss of total cytochrome P450 (CYP) content and the altered abundance of CYP mRNAs. The current work shows that these changes are preceded by the induction of the mRNA encoding the transcription factor c-jun during the hepatocyte isolation procedure. Then as the hepatocytes attach to the substratum the induced expression of c-jun subsides and two patterns of CYP mRNA loss are observed. The mRNAs encoding CYPs 2C11, 2C13, 2E1, 3A1, 3A2 and 4A1 continuously decline while CYP 1A2, 2A1/2 and 2B1/2 mRNAs are temporarily stabilised for 2 to 2.5 hours at a reduced level before declining further. The loss of CYP1A2 and 2B1/2 mRNAs parallels the loss of the mRNAs encoding the liver specific transcription factors C/EBP alpha and HNF-1. The early and rapid increase in c-jun mRNA followed by a decline in C/EBP alpha mRNA are characteristic of the changes in the expression of these transcription factor mRNAs following the stimulation of hepatocyte proliferation after partial hepatectomy. The finding that the rate of loss of total P450 following partial hepatectomy parallels that in rat hepatocyte culture suggests that the commonly employed hepatocyte isolation procedure "primes" the normally quiescent hepatocytes to enter the cell cycle and de-differentiate especially as both systems lose the major constitutively expressed CYP2C11 isozyme.

Altered expression of cytochrome P-450 mRNAs, and potentially of other transcripts encoding key hepatic functions, are triggered during the isolation of rat hepatocytes.

The abundance of 12 cytochrome P-450 (CYP) mRNAs was quantified in the caudate lobe of rat livers before dissociation of the organ into single cells by perfusion with 0.025% (w/v) collagenase. Comparison of the initial abundance of CYP-1A1, -1A2, -2A subfamily, -2B1/2, -2C7, -2C11, -2D subfamily, -2E1, -3A1/2 and -4A1 transcripts in the caudate lobe of the intact liver with the values found in freshly isolated hepatocytes demonstrated that the relatively brief (1 h) cell isolation and washing procedures routinely caused 2-3-fold increases in the mRNAs encoding CYP-1A2, -2B1/2, -3A1/2, and -4A1, concomitant with a 50% decline in CYP2C11 mRNA. Further changes in the expression of CYP mRNAs occurred when the hepatocytes were cultured. Thus CYP1A1 mRNA, which is not constitutively expressed in rat liver, became detectable in hepatocytes cultured for 1 h, and after 6 h CYP3A1/2 mRNA levels began to increase. In contrast, levels of all other CYP mRNAs studied had declined after 24 h of culture concomitant with the loss of total cytochrome P-450 content. Culture of hepatocytes with 0.5 mM metyrapone (which prevents the loss of total P-450 content) increased CYP1A1 and CYP3A1/2 mRNA levels still further, such that after 72 h of culture these transcripts were conservatively 10-18-fold higher than in hepatocytes prior to culture. This suggests that these two isoenzymes comprise the bulk of the total cytochrome P-450 content maintained by metyrapone. Collectively, these results demonstrate that the technique commonly used to isolate rat hepatocytes alters hepatic gene expression, as illustrated by the elevation of the mRNAs encoding CYP-1A2, -2B1/2, -3A1/2 and -4A1, and that such perturbations are exacerbated during culture under standard conditions by the loss of the constitutive CYP2C11 and the precocious induction of CYP1A1 and CYP3A1/2 mRNAs.

Maintenance of total cytochrome P-450 content in rat hepatocyte culture and the abundance of CYP1A2 and CYP2B1/2 mRNAs.

mRNAs encoding cytochrome P-450s CYP1A2 and CYP2B1/2 have been quantified in rat hepatocytes cultured for periods up to 72 h under several different culture conditions that maintain total cytochrome P-450 content. When hepatocytes were cultured at either 37 or 30 degrees C in Williams E media, both CYP1A2 and CYP2B1/2 mRNAs declined dramatically. However, when cultured at 30 degrees C for 24 h, the decline in these mRNAs was not as great as that observed in cells grown at 37 degrees C. The addition of dimethyl sulphoxide to cells grown at 37 degrees C did not affect the rate of disappearance of the CYP1A2 or CYP2B1/2 mRNAs. These mRNAs also declined rapidly in cells grown in 'P-450 medium' i.e. RPMI 1640 medium without cyst(e)ine but supplemented with 0.1 mM-delta-aminolaevulinic acid. However, the levels of CYP2B1/2 mRNAs were maintained when hepatocytes were cultured in Williams E medium supplemented with 0.5 mM-metyrapone. These conditions did not, however, maintain the levels of CYP1A2 mRNA.

Effects of 2-acetyl-4-tetrahydroxybutyl imidazole (THI) on the thymus of rats.

2-Acetyl-4-tetrahydroxybutyl imidazole (THI), a component of the food colouring ammonia caramel, has been shown to produce suppression of cell-mediated immunity and a reduction in circulating TH and TC/S lymphocytes in rats. Accordingly in this study the effects of THI on the thymus has been investigated. THI (1 mg/kg/day) was given for up to 7 days in the drinking water to Fischer 344 rats on a vitamin B6 deficient diet. No marked change in thymus weight was found but the cellularity was marginally decreased and flow cytometric analysis of the lymphocyte subsets revealed an increase in the number of CD4+CD8- and CD4-CD8+ single positive (SP) cells and a reduction in the number of CD4+CD8+ double positive (DP) thymocytes. This reduction was in agreement with histological findings of increased numbers of pyknotic cells in the cortex, mainly engulfed by macrophages. Mitogen-induced proliferation of thymocytes prepared from THI-treated animals was increased, concordant with the gradual increase in the percentage of mature SP cells. No change in normal proliferation of thymocytes cultured in vitro, or, in proliferation in vivo, detected as 5-bromo-2'-deoxyuridine (BrdU) incorporation, was found. It is concluded that THI produced an increase in death of immature DP cells. However, THI did not affect thymocyte proliferation or their differentiation into mature SP cells in the thymus, but rather impairs their migration into the circulation. The mechanism of action of THI appears to be indirect, but THI does not act through increasing the release of adrenal corticosteroids to supra-physiological levels as the same histopathological changes in the thymus were found in adrenalectomized rats.