Inactivation of a diol-epoxide and a K-region epoxide with high efficiency by glutathione transferase X.

Four glutathione transferases (EC 2.5.1.18), glutathione transferases A, B, and C and a hitherto unknown form, termed X, were purified to apparent homogeneity from rat liver cytosol. They were investigated for their abilities to inactivate two mutagenic epoxides derived from the polycyclic aromatic hydrocarbon benz(a)anthracene, the K-region epoxide benz(a)anthracene 5,6-oxide and the diol-epoxide r-8,t-9-dihydroxy-t-10,11-oxy-8,9,10, 11-tetrahydrobenz(a)anthracene. Mutagenic activity was determined using Salmonella typhimurium his- strain TA100. Glutathione alone had little if any influence on the mutagenicity of the diol-epoxide but significantly decreased the mutagenic effect of the K-region epoxide. This inactivation was enhanced by the addition of glutathione transferases. Both epoxides were inactivated by glutathione in the presence of each of the four enzymes, but with varying efficiencies. Inactivation of the K-region epoxide (in terms of its mutagenicity in the presence of glutathione) required extremely little enzyme, about 1000 times less than for the diol-epoxide. On a molar basis, glutathione transferase X (followed by C greater than A greater than or equal to B) was clearly the most efficient enzyme in inactivating both substrates and also more efficient than were three other purified enzymes (microsomal epoxide hydrolase, cytosolic epoxide hydrolase, and dihydrodiol dehydrogenase) previously investigated in this test system. Taking into account the amounts of enzyme present in rat liver, the glutathione transferases C and X were most effective in inactivating the epoxides examined. Thus, the newly discovered glutathione transferase X appears to be of substantial significance in the inactivation of two structural prototypes of epoxides derived from polycyclic aromatic hydrocarbons, a K-region epoxide and a non-bay-region vicinal diol-epoxide.

Polymorphisms in P450 CYP1B1 affect the conversion of estradiol to the potentially carcinogenic metabolite 4-hydroxyestradiol.

Most drug metabolizing cytochrome P450s (P450) are predominantly expressed in the liver. In contrast, human CYP1B1 is an extrahepatic P450 which is overexpressed in many tumours and has been strongly implicated in the activation of carcinogens. Rare allelic variants of the CYP1B1 gene which encode an inactive protein have been identified. However, four polymorphisms which most likely do not abolish functionality have been described. In this report, we have characterized the functional consequences of these. A CYP1B1 cDNA, identical to a cDNA published previously, served as a template to introduce allelic changes either separately or in combination. The resulting effects on CYP1B1 activity were determined in membranes isolated from Escherichia coli which coexpressed CYP1B1 together with P450 reductase. None of the allelic changes affected the CYP1B1 expression level. The allelic changes Arg48 to Gly, Ala19 to Ser and Asn453 to Ser had little influence on the Vmax and the Km of the CYP1B1 mediated 2- and 4-hydroxylation of estradiol. In contrast, the Km of these metabolic pathways was increased at least three-fold by the allelic change Va432 to Leu or by simultaneously changing Val432 to Leu and Asn453 to Ser. However, these alterations had little effect on the kinetic parameters of other CYP1B1 mediated reactions such as the epoxidation of (-)-trans-(7R,8R)-benzo[a]pyrene 7,8-dihydrodiol as determined by (r-7,t-8,t-9,c-10)-benzo[a]pyrene tetraol formation, or such as the O-dealkylation of ethoxyresorufin and the 1'-hydroxylation of bufuralol. Molecular modelling suggests that amino acid residue 432 of CYP1B1 may be involved in the interaction between CYP1B1 and P450 reductase. Since 4-hydroxyestradiol has been implicated in hormonal carcinogenesis and CYP1B1 is expressed in target tissues, the data presented demonstrate that polymorphisms in CYP1B1 have the potential to affect disease susceptibility.

Functional co-expression of CYP2D6 and human NADPH-cytochrome P450 reductase in Escherichia coli.

The polymorphic human CYP2D6 has been co-expressed with human NADPH-cytochrome P450 oxidoreductase in Escherichia coli in order to generate a functional recombinant monooxygenase system for the study of xenobiotic metabolism. The two cDNAs were co-expressed from separate, compatible plasmids with different antibiotic selection markers. The CYP2D6 could be detected in bacterial cells at levels up to 700 nmol I-1 culture by Fe(2+)-CO versus Fe2+ difference spectroscopy, exhibiting the characteristic absorbance peak at 450 nm. Immunoblotting demonstrated the presence of both proteins in bacterial membranes, where they were expressed at levels significantly higher than those found in human liver microsomes. Membrane content was 150-200 pmol CYP2D6 (determined spectrally) and 100-230 pmol CYP-reductase (determined enzymatically) per mg protein. Critically, the two co-expressed proteins were able to couple to form a NADPH-dependent monooxygenase which metabolized the CYP2D6 substrate bufuralol (Vmax 3.30 nmol min-1 mg-1 protein; K(m) 11.1 microM) in isolated membrane fractions. This K(m) value was similar to the K(m) determined in human liver microsomes. Activity could be inhibited by the specific inhibitor quinidine. Of greater significance however, was the finding that intact E. coli cells, even in the absence of exogenous NADPH, were able to metabolize bufuralol at rates almost as high as those measured in membranes (4.6 +/- 0.4 min-1 versus 5.7 +/- 0.2 min-1 at 50 microM substrate). Such recombinant strains will greatly facilitate the molecular characterization of allelic variants of cytochrome P450 isoenzymes.

Sequence similarity of mammalian epoxide hydrolases to the bacterial haloalkane dehalogenase and other related proteins. Implication for the potential catalytic mechanism of enzymatic epoxide hydrolysis.

Direct comparison of the amino acid sequences of microsomal and soluble epoxide hydrolase superficially indicates that these enzymes are unrelated. Both proteins, however, share significant sequence similarity to a bacterial haloalkane dehalogenase that has earlier been shown to belong to the alpha/beta hydrolase fold family of enzymes. The catalytic mechanism for the dehalogenase has been elucidated in detail [Verschueren et al. (1993) Nature 363, 693-698] and proceeds via an ester intermediate where the substrate is covalently bound to the enzyme. From these observations we conclude (i) that microsomal and soluble epoxide hydrolase are distantly related enzymes that have evolved from a common ancestral protein together with the haloalkane dehalogenase and a variety of other proteins specified in the present paper, (ii) that these enzymes most likely belong to the alpha/beta hydrolase fold family of enzymes and (iii) that the enzymatic epoxide hydrolysis proceeds via a hydroxy ester intermediate, in contrast to the presently favoured base-catalyzed direct attack of the epoxide by an activated water.

Coexpression of a human P450 (CYP3A4) and P450 reductase generates a highly functional monooxygenase system in Escherichia coli.

The catalytic activities of recombinant cytochrome P450s expressed in E. coli have been impeded by the absence of endogenous P450 reductase. To solve this problem, we coexpressed P450 reductase with CYP3A4. Membranes from this strain contained 215 pmol P450/mg protein and a reductase activity of 1315 nmol cytochrome c reduced/min per mg. We detected 6beta-hydroxylation of testosterone and oxidation of nifedipine in vivo with turnover numbers of 15.2 and 17.3 min(-1), respectively. These values compare favourably with those obtained using an optimally reconstituted system. Our data demonstrate that a catalytically efficient human P450 system can be generated in E. coli.

Cytochrome P450 polymorphisms as risk factors for steroid hormone-related cancers.

The development of cancers of the breast, endometrium, ovaries and possibly prostate is modulated by steroid hormones. Many steroids and environmental carcinogens are subject to cytochrome P450 (P450)-mediated metabolism that generates reactive metabolites and modulates steroid potency, thereby influencing tumor initiation and promotion respectively. These pathways, which are modulated by polymorphisms in P450 genes, are therefore likely to play an important role in the etiology of hormone-related cancers. Several groups have evaluated genotypes of xenobiotic- and steroid-metabolizing P450 enzymes as risk factors for hormone-related cancers. Polymorphisms in P450s that are specifically involved in the metabolism of steroids appear to be single risk factors. The situation is less clear for xenobiotic-metabolizing P450s. For these genes, only combined genotypes of several P450s or combined genotypes of P450s together with other enzymes have been clearly correlated with disease frequency. Success in identifying the appropriate combination of candidate genes requires a thorough knowledge of the metabolic pathways and enzyme systems that control the initial stages of carcinogenesis, as will be illustrated in this review.

Studies of the expression of the cytochrome P450IA, P450IIB, and P450IIC gene family in extrahepatic and hepatic tissues.

We have studied the expression of three P-450 gene subfamilies in hepatic and extrahepatic tissues using the sensitive RNAse A protection assay. Members of the P450IA subfamily, which encodes the major methylcholanthrene-inducible cytochromes P-450, were found to be not expressed in extrahepatic tissues of untreated animals, raising the question whether these P-450 play a role in the metabolism of carcinogens in unexposed individuals. In contrast, members of the P450IIB family, some of which encode the major phenobarbital-inducible cytochromes P-450, were found to be expressed in some extrahepatic tissues of untreated rats and here most notably in the lung and in sebaceous glands. Members of the P450IIC family, which encode some constitutively expressed cytochromes P-450, were found to be expressed exclusively in the liver.

Mechanism-based predictions of interactions.

Exposure to more than one toxic compound is common in real life. The resulting toxic effects are often more than the simple sum of the effects of the individual compounds. It is unlikely that it will ever be possible to test all combinations. It is therefore highly desirable to improve or develop means for reasonably approximating predictions of interactions. In order to be valid and extrapolatable, these predictions are most promising if they are mechanism-based. Examples will be given for possibilities of mechanism-based predictions of interactions which exceed trivialities of simple increases by enzyme induction of enzymatic rates of a given biotransformation pathway leading to a toxic metabolite. Instead, examples will be provided where competition between various enzymes for shunting the same substrate into divergent pathways can lead to predictable dramatic changes in toxicity by shifting the metabolic routes under conditions of no significant changes of overall metabolism. Further examples are given on predictable interactions between chemicals which need bioactivation for exerting their toxicity and chemicals which effect hormonal status and other endogenous factors which in turn modify enzymes involved in the control of toxic metabolites.

Perfluorodecanoic acid decreases the enzyme activity and the amount of glutathione S-transferases proteins and mRNAs in vivo.

The effects of the anti-wetting agent perfluoro-n-decanoic acid (PFDA) on various glutathione S-transferase (GST) enzyme activities were studied in vitro and in vivo. In addition the effects of PFDA treatment on the amount of some glutathione S-transferase subunits and their corresponding translatable mRNA were studied in vivo. PFDA like some other peroxisome proliferators was a non-competitive inhibitor of several GST enzyme activities in vitro. In vivo PFDA reduced the enzyme activity towards substrates which are indicative for the Ya, Yb1 and Yb2 subunits of GSTs to a larger extent than the enzyme activity towards the substrate indicative for the Yc subunit. Whereas the reduction of GST enzyme activities by other peroxisome proliferators was shown to be caused by an inhibition of the relevant enzymes in vivo, PFDA was found to decrease the GST enzyme activities at least in part by lowering the amount of the various GST subunits in vivo due to a lowered concentration of translatable mRNA coding for these enzymes. In addition PFDA abolished the inducibility of GST mRNAs by phenobarbital. Thus PFDA might be an interesting tool for mechanistic studies of the control of GST expression in the liver.

Effects of lindane treatment on drug metabolizing enzymes and liver weight of CF1 mice in which it evoked hepatomas and in non-susceptible rodents.

In CF1 mice lindane treatment led to a significant increase in liver tumor incidence whilst in Osborne-Mendel rats it was not carcinogenic. Although somewhat less clear, the test in B6C3F1 mice led to the conclusion that under the conditions of the bioassay lindane was not carcinogenic for this strain. In this study, the specific activities of some enzymes which are thought to be involved in the metabolism of lindane were studied in these different strains in order to investigate whether differences exist in the activities of these enzymes. Because the enzyme pattern may change after lindane treatment during the carcinogenicity studies, we also investigated the enzyme activities in animals treated for 3 days or 3 months with various doses of lindane. The influence of lindane treatment on the relative liver weight was also determined. B6C3F1 mice showed no increase in absolute or relative liver weight even after 3 months of treatment with the highest tolerated dose of lindane. However, in the susceptible CF1 strain lindane led to a large increase of the absolute and relative liver weight in both sexes, whilst a smaller increase was observed in Osborne-Mendel rats. Basal glutathione-S-transferase activity was higher in males than in females in all three strains, bearing no apparent relationship to susceptibility for tumor formation. However, after treatment with the highest dose of lindane a 5-6-fold induction of this enzyme activity was observed in female CF1 mice, which then together with the male CF1 mice had a higher glutathione-S-transferase activity than untreated and treated B6C3F1 mice and Osborne-Mendel rats. Whether lindane or one of its metabolites is activated by conjugation with glutathione remains to be established. After treatment of the animals with high doses of lindane detergent-treated rat liver microsomes showed a higher UDP-glucuronosyltransferase activity than mouse liver microsomes. This high activity could lead to a rapid conjugation of phenols derived from lindane. The most striking difference observed in this study was the fact that together with the larger increase in absolute and relative liver weight, untreated and treated CF1 mice showed higher monooxygenase activity and, after treatment with lindane, lower epoxide hydrolase activity than rats. Whether the high monooxygenase and rather low epoxide hydrolase activity will lead to an accumulation of reactive epoxides derived from lindane remains to be clarified.

Molecular and cellular basis for adequate metabolic design of genotoxicity studies.

Genotoxic species and metabolites are usually under the control of a complex set of activating, inactivating and precursor sequestering enzymes. These enzymes differ greatly between test systems, animal species and man. An adequate metabolic design of genotoxicity studies requires careful attention to factors such as: Dilution of cofactors in in vitro tests which are present in much higher concentrations in the intact cell; Induction in high dose carcinogenicity bioassays of enzymes, which are constitutively not expressed and not induced at such doses of the compound, which occur in the situations of the practical use of the compound; Modifications of control enzymes, which are effected by hormones or other endogenous factors, which are differently influenced by high dose (bioassay) versus moderate dose (real exposure) or by in vivo (endocrine regulation) versus in vitro (no endocrine regulation) conditions.

In vitro investigation of cytochrome P450-mediated metabolism of dietary flavonoids.

Human and mouse liver microsomes and membranes isolated from Escherichia coli, which expressed cytochrome P450 (CYP) 1A2, 3A4, 2C9 or 2D6, were used to investigate CYP-mediated metabolism of five selected dietary flavonoids. In human and mouse liver microsomes kaempferol, apigenin and naringenin were hydroxylated at the 3'-position to yield their corresponding analogs quercetin, luteolin and eriodictyol, whereas hesperetin and tamarixetin were demethylated at the 4'-position to yield eriodictyol and quercetin, respectively. Microsomal flavonoid metabolism was potently inhibited by the CYP1A2 inhibitors, fluvoxamine and -naphthoflavone. Recombinant CYP1A2 was capable of metabolizing all five investigated flavonoids. CYP3A4 recombinant protein did not catalyze hesperetin demethylation, but showed similar metabolic profiles for the remaining compounds, as did human microsomes and recombinant CYP1A2, although the reaction rates in general were lower as compared to CYP1A2. CYP2C9 catalyzed the 4'-demethylation of tamarixetin, whereas CYP2D6 did not seem to play any role in the metabolism of the selected flavonoids. The major involvement in flavonoid metabolism of human CYP1A2, which mediates the formation of metabolites with different biochemical properties as compared to the parent compound and furthermore is known to be expressed very differently among individuals, raises the important question of whether individual differences in the CYP enzyme activity might affect the beneficial outcome of dietary flavonoids, rendering some individuals more or less refractory to the health-promoting potential of dietary flavonoids.

Toxicological implications of enzymatic control of reactive metabolites.

Many foreign compounds are transformed into reactive metabolites, which may produce genotoxic effects by chemically altering critical biomolecules. Reactive metabolites are under the control of activating, inactivating and precursor sequestering enzymes. Such enzymes are under the long-term control of induction and repression, as well as the short-term control of post-translational modification and low molecular weight activators or inhibitors. In addition, the efficiency of these enzyme systems in preventing reactive metabolite-mediated toxicity is directed by their subcellular compartmentalization and isoenzymic multiplicity. Extrapolation from toxicological test systems to the human requires information of these variables in the system in question and in man. Differences in susceptibility to toxic challenges between species and individuals are often causally linked to differences in these control factors.

Refinement of the Region for Split Hand/Foot Malformation 5 on 2q31.1.

Background: Deletions that encompass 2q31.1 have been proposed as a microdeletion syndrome with common clinical features, including intellectual disability/developmental delay, microcephaly, cleft palate, growth delay, and hand/foot anomalies. In addition, several genes within this region have been proposed as candidates for split hand-foot malformation 5 (SHFM5). Methods: To delineate the genotype-phenotype correlation between deletions of this region, we identified 14 individuals with deletions at 2q31.1 detected by microarray analysis for physical and developmental disabilities. Results: All subjects for whom detailed clinical records were available had neurological deficits of varying degree. Seven subjects with deletions encompassing the HOXD cluster had hand/foot anomalies of varying severity, including syndactyly, brachydactyly, and ectrodactyly. Of 7 subjects with deletions proximal to the HOXD cluster, 5 of which encompassed DLX1/DLX2, none had clinically significant hand/foot anomalies. In contrast to previous reports, the individuals in our study did not display a characteristic gestalt of dysmorphic facial features. Conclusion: The absence of hand/foot anomalies in any of the individuals with deletions of DLX1/DLX2 but not the HOXD cluster supports the hypothesis that haploinsufficiency of the HOXD cluster, rather than DLX1/DLX2, accounts for the skeletal abnormalities in subjects with 2q31.1 microdeletions.

Molecular biological methods for characterizing drug-metabolizing enzymes in hepatic and extrahepatic tissues.

Drug metabolism influences the pharmacotoxicological properties of a vast array of compounds and is controlled by a complex system of drug-metabolizing enzymes. A thorough understanding of this system allows the more effective development of therapeutic drugs as well as a significant improvement of risk assessment. The early identification of optimal therapeutic problems relating to drug metabolism could reduce the development costs for pharmaceuticals. Recently, techniques using transgenes have become available for this purpose. In these approaches the genetic information for the enzyme under investigation is expressed in vitro or in vivo, following gene transfer. This approach is called 'heterologous expression'. This article illustrates some examples in which molecular biological methods have been used to analyze those enzymes which control the pharmacotoxicological properties of drugs. Particular emphasis has been placed on the use of these methods to characterize extrahepatic drug-metabolizing enzymes such as those in the skin.

Colorimetric quantitation of trace amounts of sodium lauryl sulfate in the presence of nucleic acids and proteins.

A fast and sensitive procedure for the colorimetric detection of sodium lauryl sulfate (SDS) is presented. The assay is based upon the formation of a chloroform-extractable ion pair between lauryl sulfate and methylene blue that is quantified spectrophotometrically with an estimated detection limit of 150 ng of SDS. The method is suitable for the monitoring of contaminating traces of SDS in protein or nucleic acid samples that have the potential to interfere with enzymatic manipulations such as proteolytic digest, restriction analysis, or reverse transcription. Since the procedure is extremely simple and no special equipment is required it is accessible to every researcher concerned with SDS contamination.

The role of the hydrophobic domain in orienting natural signal sequences within the ER membrane.

The orientation of signal sequences during insertion into the endoplasmic reticulum membrane is largely determined by the charged residues flanking the apolar domain. Using recombinant and mutant proteins, also length and hydrophobicity of the apolar segment were shown to affect the orientation: translocation of the N-terminus was found to be favored by long hydrophobic sequences, and translocation of the C-terminus, by short ones. Here, we tested the physiological significance of this phenomenon by mutagenesis of the hydrophobic portion of two natural signals with unusual flanking charges. Extending the hydrophobic domain of the short, cleaved Ncyt/Cexo signal of pre-provasopressin-neurophysin II and shortening that of the Nexo/Ccyt signal anchor of microsomal epoxide hydrolase resulted in a significant fraction of polypeptides inserting in the opposite orientation to that of the wild-type proteins. The topogenic contribution of the hydrophobic domain is thus important for the correct and uniform orientation of natural proteins and can explain the behavior of some of the signals with unusual flanking charges.