Miscoding properties of 6alpha- and 6beta-diastereoisomers of the N(2)-(estradiol-6-yl)-2'-deoxyguanosine DNA adduct by Y-family human DNA polymerases.

Treatment with estrogen increases the risk of breast, ovary, and endometrial cancers in women. DNA damage induced by estrogen is thought to be involved in estrogen carcinogenesis. In fact, Y-family human DNA polymerases (pol) eta and kappa, which are highly expressed in the reproductive organs, miscode model estrogen-derived DNA adducts during DNA synthesis. Since the estrogen-DNA adducts are a mixture of 6alpha- and 6beta-diastereoisomers of dG-N(2)-6-estrogen or dA-N(6)-6-estrogen, the stereochemistry of each isomeric adduct on translesion synthesis catalyzed by DNA pols has not been investigated. We have recently established a phosphoramidite chemical procedure to insert 6alpha- or 6beta-isomeric N(2)-(estradiol-6-yl)-2'-deoxyguanosine (dG-N(2)-6-E(2)) into oligodeoxynucleotides. Using such site-specific modified oligomer as a template, the specificity and frequency of miscoding by dG-N(2)-6alpha-E(2) or dG-N(2)-6beta-E(2) were explored using pol eta and a truncated form of pol kappa (pol kappaDeltaC). Translesion synthesis catalyzed by pol eta bypassed both the 6alpha- and 6beta-isomers of dG-N(2)-6-E(2), with a weak blockage at the adduct site, while translesion synthesis catalyzed by pol kappaDeltaC readily bypassed both isomeric adducts. Quantitative analysis of base substitutions and deletions occurring at the adduct site showed that pol kappaDeltaC was more efficient than pol eta by incorporating dCMP opposite both 6alpha- and 6beta-isomeric dG-N(2)-6-E(2) adducts. The miscoding events occurred more frequently with pol eta, but not with pol kappaDeltaC. Pol eta promoted incorporation of dAMP and dTMP at both the 6alpha- and 6beta-isomeric adducts, generating G --> T transversions and G --> A transitions. One- and two-base deletions were also formed. The 6alpha-isomeric adduct promoted slightly lower frequency of dCMP incorporation and higher frequency of dTMP incorporation and one-base deletions, compared with the 6beta-isomeric adduct. These observations were supported by steady-state kinetic studies. Taken together, the miscoding property of the 6alpha-isomeric dG-N(2)-6-E(2) is likely to be similar to that of the 6beta-isomeric adduct.

Synthesis of oligodeoxynucleotides containing a single 6alpha- or 6beta-diastereoisomer of N2-(estradiol-6-yl)-2'-deoxyguanosine.

DNA damage induced by estrogens is associated with developing breast, ovary, and endometrial cancers. The quinone of 2-hydroxyestrogen (2-OHE), a major estrogen metabolite, produces 2-OHE-derived dG and dA adducts in DNA. N(2)-[Estradiol-6(alpha or beta)-yl]-2'-deoxyguanosine [dG-N(2)-6(alpha or beta)-E(2)] lacking a 2-OH moiety may also be formed through sulfonation of 6-hydroxyestrogen. To explore the biological properties of such estrogen-DNA adducts, oligodeoxynucleotides modified by estrogen-derived DNA adduct were prepared by chemical synthesis. Initially, 6alpha- and 6beta-aminoestradiol 17-acetate (6alpha- and 6beta-NH(2)-E(2) 17Ac) were prepared by reductive amination of 6-oxo-estradiol 3,17-diacetate. The DMT-phosphoramidite derivative of N(2)-(3,17-diacetoxyestradiol-6alpha-yl)-2'-deoxyguanosine and its 6beta-isomer were prepared by coupling 5'-O-(4,4'-dimethoxytrityl)-2-fluoro-O(6)-[2-(4-nitrophenyl)ethyl]-2'-deoxyinosine separately with 6alpha- and 6beta-forms of NH(2)-E(2) 17Ac, respectively, followed by selective acetylation of the steroidal 3-hydroxyl group. The desired oligodeoxynucleotide containing a single dG-N(2)-6alpha-E(2) or dG-N(2)-6beta-E(2) was prepared efficiently by an automated DNA synthesizer. Synthesis of these site-specifically modified oligodeoxynucleotides will benefit further research into the biological properties and three-dimensional structure of 6alpha- and 6beta-diastereoisomers of estrogen-DNA adducts.

Studies on the interactions between drugs and estrogen. III. Inhibitory effects of 29 drugs reported to induce gynecomastia on the glucuronidation of estradiol.

To determine the inhibition effects of drugs on the glucuronidation of estradiol (E2), 29 drugs that have been reported to induce gynecomastia were examined in the presence of UDP-glucuronic acid using human hepatic microsomes (pooled) as the enzyme source. The percentage inhibition of the E2 glucuronidation was determined at drug concentrations of 1 microM (approximate therapeutic concentration) and 100 microM (non-clinical overdose concentration) based on the rate constants for the 3- and 17-glucuronidation of E2 (11.2 and 2.52 pmol/min/mg protein, respectively). The only drug that exhibited 50% or higher inhibition of the 3-glucuronidation at a concentration of 1 microM was manidipine (54.4%). When the concentration was 100 microM, manidipine exhibited 100% inhibition of the 3-glucuronidation, and other drugs that exhibited 50% or higher inhibition of the 3-glucuronidation were nicardipine (92%), nisoldipine (90%), nifedipine (84%), domperidone (81%), tacrolimus (80%), nitrendipine (77%) and ketoconazole (69%). Conversely, ipriflavone accelerated the formation of estradiol 3-glucuronide in the activity of 165% at the concentration of 100 microM. On the 17-glucuronidation, all of the drugs showed less than 50% inhibition at the concentration of 1 microM, but at the concentration of 100 microM, drugs that exhibited 50% or higher inhibition consisted of manidipine (79%), chlormadinone acetate (74%), nisoldipine (66%), nitrendipine (60%) and ketoconazole (55%). Although IC(50) values of these drugs were all lower than the K(m) value (285 microM) for the 3-glucuronidation of E2, they were higher than the K(m) value for the 17-glucuronidation (18.8 microM). Thus, the effect of the drugs on the E2 glucuronidation should be greater for hydroxy group at the C-3 than that at the C-17 of E2 molecule. On the other hand, metabolic clearances (V(max)/K(m)) of the 3- and 17-glucuronidation were about 1/14th and 1/18th of that of the 2-hydroxylation of E2, respectively. The result implies that, when the contribution of the glucuronidation to enterohepatic circulation is taken into consideration, the effect of this metabolic inhibition in the estrogen pool cannot be ignored.

Translesion synthesis past estrogen-derived DNA adducts by human DNA polymerases eta and kappa.

Newly discovered human DNA polymerase (pol) eta and kappa are highly expressed in the reproductive organs, such as testis, ovary, and uterus, where steroid hormones are produced. Because treatment with estrogen increases the risk of developing breast, ovary, and endometrial cancers, miscoding events occurring at model estrogen-derived DNA adducts were explored using pol eta and a truncated form of human pol kappa (pol kappaDeltaC). These enzymes bypassed N(2)-[3-methoxyestra-1,3,5(10)-trien-6-yl]-2'-deoxyguanosine (dG-N(2)-3MeE) and N(6)-[3-methoxyestra-1,3,5(10)-trien-6-yl]-2'-deoxyadenosine (dA-N(6)-3MeE), which were embedded in site-specifically modified oligodeoxynucleotide templates. Quantitative analysis of base substitutions and deletions occurring at the lesion site showed that pol kappaDeltaC was more efficient at incorporating dCMP opposite the dG-N(2)-3MeE lesion than pol eta. Surprisingly, the frequency of translesion synthesis beyond the dC*dG-N(2)-3MeE pair was 13% of the normal dC*dG pair and was 4 and 6 orders of magnitude higher than that of dC*(+)-trans-dG-N(2)-benzo[a]pyrene and dC*dG-C8-acetylaminofluorene pairs, respectively, suggesting that dG-N(2)-3MeE is a natural substrate for pol kappa. In contrast, the bypass frequency beyond the dT*dA-N(6)-3MeE pair was 7 orders of magnitude less than that for the normal dT*dA pair. dA-N(6)-3MeE is a more miscoding lesion than dG-N(2)-3MeE. Pol eta promoted incorporation of dAMP and dCMP at the dA-N(6)-3MeE lesion, while with pol kappaDeltaC, deletions were more frequently observed, along with incorporation of dAMP and dCMP opposite the lesion. These observations were also supported by steady-state kinetic studies. When taken together, the properties of pol eta and kappa are consistent with the mutagenic events attributed to estrogen-derived DNA adducts.

Comparison of ex vivo inhibitory effect between 2-hydroxyestradiol and its 17-sulfate on rat hepatic microsomal lipid peroxidation.

Two endogenous antioxidants that are speculated to be defense substances against preeclampsia, 2-hydroxyestradiol (2-OH-E2) and its 17-sulfate, 2-hydroxyestradiol 17-sulfate (2-OH-E2-17-S), were administered to rats to compare their inhibitory effects on hepatic microsomal lipid peroxidation, and the lipid peroxides were determined in NADPH- and ascorbic acid-dependent systems. The two catechols showed a strong inhibitory effect on lipid peroxidation in both systems, and the effect was dose dependent. However, a large difference was observed in their inhibition patterns. After administration of 2-OH-E2, the effect appeared immediately and decreased gradually with time. In contrast, the effect of 2-OH-E2-17-S appeared some time after administration and persisted for a longer time. Both catechols also showed a striking difference in their dynamics. After administration, 2-OH-E2 was detected in the blood together with its metabolites, 2-methoxyestradiol and 2-methoxyestrone, and they disappeared immediately. In contrast, 2-OH-E2-17-S was present in the blood for a longer time together with its O-methylated product, 2-methoxyestradiol 17-sulfate, but disappeared from liver microsomes within 2 h after administration. The results imply no occurrence of a direct inhibition effect of 2-OH-E2-17-S.

Metabolism of [6,7-3H, 35S] estradiol 17-sulfate in rats.

To confirm whether or not the sulfo group of estradiol 17-sulfate (ES) is removed during in vivo metabolism in rats, the doubly labeled conjugate [6,7-3H, 35S] ES was injected into rats, and its biliary and urinary metabolites were determined by reverse isotope dilution method (RIDM). In male rats, the major radioactivity was detected in biliary disulfate fraction, which was composed of mainly ES and its two minor metabolites, 2-hydroxyestradiol 17-sulfate (2-OH-ES) and 2-methoxyestradiol 17-sulfate (2-MeO-ES). In female rats, in contrast, the radioactivity was dispersed into three fractions:biliary monosulfate, biliary disulfate, and urinary monosulfate fractions (Frs.) In both monosulfate Frs., 7beta-hydroxyestradiol 17-sulfate was detected as the major metabolite followed by 6alpha-, 6beta-, and 15beta-hydroxyestradiol 17-sulfates. Like male rats, 2-OH-ES and 2-Meo-ES as the minor products were detected in biliary disulfate fraction. The isotope ratios of ES and its metabolites in both sexes were essentially the same as that of the dose except that of 6alpha-hydroxylated metabolite, which may be derived from the loss of the tritium labeled at C6. These results confirm the occurrence of the direct metabolism of ES in rats.

Studies on the interactions between drug and estrogen. II. On the inhibitory effect of 29 drugs reported to induce gynecomastia on the oxidation of estradiol at C-2 or C-17.

A study was investigated on the inhibitory effect of 29 drugs that have been reported to induce gynecomastia on the 2-hydroxylation of estradiol (E2) by recombinant P450 CYP3A4 and on the 17-oxidation of E2 by hepatic microsomal type II 17beta-hydroxysteroid dehydrogenase (17beta-HSD) of human male. The IC(50) values were determined for each drug relative to the 2-hydroxylation of E2 (catalytic activity: 1.54 nmol/nmol P450/min), and the inhibition constants (K(i)) were determined for 13 drugs of which IC(50) values were 100 microM or less. Ketoconazole exhibited the lowest inhibitory concentration, and IC(50) and K(i) values of 0.007 and 0.01 microM, respectively, were obtained. The IC(50) and K(i) values for each of the 12 remaining drugs were as follows: cyclosporin A (IC(50): 0.064, K(i): 0.30), nicardipine hydrochloride (0.55, 0.29), tacrolimus (0.64, 0.88), mandipine hydrochloride (3.9, 2.6), nisoldipine (10, 3.3), verapamil hydrochloride (10, 20), domperidone (13, 7.2), haloperidol (14, 55), nitrendipine (14, 2.5), chlormadinone acetate (16, 10), flutamide (30, 39) and omeprazole (49, 47). With the exception of cyclosporin A that exhibited a competitive inhibition, the inhibition mechanisms of these drugs were all non-competitive. Next, the percentage inhibition of the above 29 drugs relative to the 17-oxidation of E2 (catalytic activity: 0.47 nmol/mg protein/min) was investigated at the approximate therapeutic concentration (1 microM) and at the non-clinical overdose concentration (100 microM). Although none of the drugs investigated exhibited inhibitory effects at a concentration of 1 microM, spironolactone and ketoconazole at 100 microM demonstrated percentage inhibitions of 96% and 77%, respectively. When the K(i) values were determined for these two drugs, the former had a K(i) value of 2.4 microM and the latter, 41 microM, and both of their inhibition mechanisms were non-competitive. On the basis of the above results, a total of 14 drugs consisting of the above 13 drugs plus spironolactone were found to inhibit the 2-hydroxylation or 17-oxidation of E2 in the liver, and this is presumed to act as a trigger that causes as increase in the estradiol pool, followed by induction of gynecomastia.

On the inhibitory action of 29 drugs having side effect gynecomastia on estrogen production.

To examine the influence on aromatase and sulfatase pathways in estrogen pool by drugs reported to cause gynecomastia as the side effect, 29 ethical drugs were incubated with human placental microsomes as an enzyme source. The percent inhibition of drugs on aromatase pathway was obtained by sum of the velocity constants of two products, estrone (E1) and estradiol (E2) from testosterone (T) as the substrate, and that on sulfatase pathway was obtained as the velocity constant of production of E1 from estrone sulfate (E1S). Although several drugs including ketoconazole showed a significant inhibition effect on aromatase pathway at their non-clinical over-dose concentration (100 microM), no influence on the inhibition was observed in any drugs at their approximately therapeutic concentration (1 microM). However, several drugs including spironolactone gave the product ratio (E2/E1) having higher value than that of the control, the result means spironolactone inhibits the conversion of E2 to E1. No inhibitory effect of the drugs tested on estrogen production from E1S (sulfatase pathway) was confirmed. The results suggest the possibility that the tested drugs known to cause gynecomastia have no inhibitory effect essentially on aromatase and sulfatase pathways.

On the structures of hydroxylated metabolites of estradiol 17-sulfate by rat liver microsomes.

The metabolism of estradiol 17-sulfate (ES) by hepatic microsomes of female rats produced four new metabolites in addition to 2- and 4-hydroxyestradiol 17-sulfates (2- and 4-OH-ES), which were detected on an HPLC chromatogram. By comparison with synthetic specimens, three of these compounds were identified as 6alpha-, 6beta-, and 7beta-hydroxyestradiol 17-sulfates. To elucidate the structure of the remaining metabolite, a large-scale incubation of ES was carried out, followed by isolation using preparative HPLC to give the single material, which was assigned as 15beta-hydroxyestradiol 17-sulfate by instrumental analyses. On the other hand, when ES was incubated with the microsomes of male rats, 2-OH-ES was produced accompanied by two minor products: 4-OH-ES and a metabolite of unknown structure. The results show clearly that the metabolism of ES by rat hepatic microsomes is remarkably different between the sexes.