Characterization of multiple Chinese hamster carbonyl reductases.

Carbonyl reductase (CR) is an enzyme which can catalyze the oxidoreduction of various carbonyl compounds in the presence of NAD(P)H. With the PCR method, using primers carrying the conserved nucleotide sequence among mammalian CRs, we isolated three different cDNAs (CHCR1, CHCR2 and CHCR3) which encode a unique carbonyl reductase from the Chinese hamster. The PCR products of CHCR1 and CHCR2 were clearly isolated with Bpu1102I, BspEI and XmaI restriction enzymes. The nucleotide-sequence of CHCR3 was completely different from those of CHCR1 and CHCR2. The predicted double-wound betaalphabetaalpha-structures of the CHCRs suggests the presence of a typical NADP(+)-binding motif and is similar to the corresponding region of 3alpha,20beta-hydroxysteroid dehydrogenase and mouse lung tetrameric carbonyl reductase. The deduced amino acid sequence of CHCR1 showed a high homology to CHCR2 (>96%) and the other mammalian CRs (>81%). However, CHCR3 showed a high homology to human CBR3 (>86%) and a relatively lower homology to the other CHCRs (<76%). Bacterial recombinant CHCRs showed typical carbonyl reductase activities towards 4-benzoylpyridine, 4-nitrobenzaldehyde and pyridine 4-carboxyaldehyde. These three CRs showed not only 3-keto reductase of steroids, but also 20-keto reductase. However, these CRs did not show any activity of 17-keto reductase activity. Both CHCR1 and CHCR2 have prostaglandin 9-keto reductase and 15-hydroxyprostaglandin dehydrogenase activities towards PGE(2) and PGF(2alpha) from the analyses of enzymatic reaction products. The results of Western blotting and RT-PCR suggest these CHCRs have a tissue-dependent-distribution in the Chinese hamster.

Cloning and bacterial expression of monomeric short-chain dehydrogenase/reductase (carbonyl reductase) from CHO-K1 cells.

Mammalian carbonyl reductase (EC 1.1.1.184) is an enzyme that can catalyze the reduction of many carbonyl compounds, using NAD(P)H. We isolated a cDNA of carbonyl reductase (CHO-CR) from CHO-K1 cells which was 1208 bp long, including a poly(A) tail, and contained an 831-bp ORF. The deduced amino-acid sequence of 277 residues contained a typical motif for NADP+-binding (TGxxxGxG) and an SDR active site motif (S-Y-K). CHO-CR closely resembles mammalian carbonyl reductases with 71-73% identity. CHO-CR cDNA had the highest similarity to human CBR3 with 86% identity. Using the pET-28a expression vector, recombinant CHO-CR (rCHO-CR) was expressed in Escherichia coli BL21 (DE3) cells and purified with a Ni2+-affinity resin to homogeneity with a 35% yield. rCHO-CR had broad substrate specificity towards xenobiotic carbonyl compounds. RT-PCR of Chinese hamster tissues suggest that CHO-CR is highly expressed in kidney, testis, brain, heart, liver, uterus and ovary. Southern blotting analysis indicated the complexity of the Chinese hamster carbonyl reductase gene.

[Identification of environmental estrogens with a three-dimensional quantitative structure-activity relationship(3D-QSAR) analysis].

The quantitative structure-activity relationship(QSAR) technique has been used by computing the structural data from many chemical compounds. In order to predict the estrogenic actions of chemical compounds, various steric(van der Waals) and electrostatic(Coulombic) interaction energy from those compounds are calculated at the standard with a comparative molecular field analysis(CoMFA) and the comparison of these results with template estradiol is drawn as a three-dimensional image. The three-dimensional structural features of various chemical compounds are compared with those of estradiol. The binding affinities of various chemical compounds with estrogen receptor could be predicted. Steric bulk of a chemical compound molecule and electric potential of substituents bound to the molecule can be also drawn on the basis of the three-dimensional data obtained. The results from CoMFA/QSAR analysis correlates well with ones from the presently widely-used bioassay, indicating compatibility of two methods. The 3D-QSAR technique with CoMFA/QSAR analysis seems to be useful for selection of the target disruptors prior to the time-consuming and tedious bioassays.

hCG-stimulated ovarian carbonyl reductase in relation to preovulatory ovarian follicular growth.

The relationship between carbonyl reductase, which is localized in the theca and interstitial cells, and ovarian follicular development was evaluated by measuring the enzyme activity and concentration of carbonyl reductase, and examining its immunohistochemical localization in immature rats treated with equine chorionic gonadotrophin (eCG), ovine(o) FSH and hCG. eCG and oFSH were administered s.c. to 26-day-old rats, and the ovaries were isolated 48 and 72 h after treatment. hCG was administered s.c. 48 h (at 28 days of age) after eCG or oFSH, and the ovaries were isolated 6 or 9 h after treatment. Both eCG and oFSH significantly increased the ovarian carbonyl reductase concentration 48 h after treatment compared with the saline-treated group, but the increase with eCG was approximately twice that with oFSH. Seventy-two hours after treatment, eCG further increased both carbonyl reductase activity and its concentration. Ovulation was also induced. oFSH stimulated only ovarian follicular growth without further increase in either activity or concentration. In addition, eCG increased the immunoreactivity to anti-carbonyl reductase antibody in the theca and interstitial cells, but oFSH did not. Treatment with hCG in saline-pretreated rats increased carbonyl reductase activity by 2.8-fold and carbonyl reductase concentration by 4.1-fold after 9 h compared with the 0 h concentration, without producing the large follicles observed with eCG and oFSH. However, the stimulatory effect of hCG on ovarian carbonyl reductase in eCG- or oFSH-pretreated rats was weaker than that in saline-pretreated rats. These results suggest that ovarian carbonyl reductase is induced by LH/hCG but not by FSH and that the enzyme induced by hCG is unrelated to preovulatory ovarian follicular growth.

Pre- and post-ovulatory changes in carbonyl reductase in ovarian follicles and corpora lutea in rats.

We investigated the relationship between the carbonyl reductase (CR) and the ovarian function in normal cycling and gonadotropin treated rats. The preovulatory mature large follicles, the neo corpora lutea (CL) after ovulation, and the old CL were isolated from the other ovarian component. The CR activity (4-benzoylpyridine reducing activity) in the preovulatory mature large follicles did not change between 1500 h and 1900 h on the day of proestrus. In the neo CL after ovulation, the activity and concentration of the enzyme were significantly increased to 2.2 and 1.7 times, respectively, of the levels in the mature large follicles. The enzyme activity and its concentration in the old CL were the lowest. The enzyme activity in the neo CL after human chorionic gonadotropin (hCG) treatment in equine chorionic gonadotropin (eCG)-pretreated rats was also increased to 1.9 times of that in the mature large follicles formed by treatment with eCG alone. Positive immunohistochemical localization of ovarian CR was observed in the neo CL formed on the day of estrus and after hCG treatment. The results suggest that the rat ovarian CR is involved in the formation of neo CL after ovulation rather than the ovulatory process.

Effects of psychotropic drugs on aldo-keto reductase activity in rat ovary and adrenal gland.

We investigated the effects of minor and major tranquilizers on ovarian and adrenal aldo-keto reductase activity towards five substrates in relation to ovulation in mature cycling rats. Nitrazepam (NZP) did not alter ovarian and adrenal weights or body weight, although ovulation was inhibited at 5 and 10 mg/kg. NZP decreased ovarian 13,14-dihydro-15-ketoprostaglandin F2 alpha (15KD-PGF2 alpha) and 4-benzoylpyridine (4BP) reducing activities. None of the doses of zopiclone (ZPC) influenced uterine and adrenal weights or body weight, but it increased ovarian weight at 10 mg/kg. No significant effects of ZPC on ovarian aldo-keto reductase activity were observed. NZP had inhibitory effects on adrenal aldo-keto reductase activity, whereas ZPC had a stimulatory effect. Chlorpromazine (CPZ) did not alter ovarian or adrenal weight, whereas the estrous cycles were abolished at 5 and 10 mg/kg. Reserpine (RSP) decreased ovarian weight and completely inhibited ovulation at 5 and 10 mg/kg, but it increased adrenal weight. Both CPZ and RSP decreased, dose dependently, ovarian aldo-keto reductase activity towards five substrates in agreement with the inhibition of ovulation. On the other hand, differences were found between the effects of CPZ and RSP on adrenal aldo-keto reductase activity. CPZ significantly increased 4BP reducing activity at 5 and 10 mg/kg, although no significant changes were observed in the other four reducing activities. RSP decreased 15KD-PGF2 alpha reducing activity in a dose-dependent manner, whereas the other four activities did not change.

Ovarian carbonyl reductase activity in rats at persistent oestrus.

Age-related changes and effects of gonadotrophins on ovarian carbonyl reductase (CR) of virgin female Wistar-Imamichi rats were investigated. In mature rats ovarian CR is an LH-dependent oxidoreductase and is closely involved in the periovulatory process. After 8 months of age, all of the rats exhibited persistent oestrus and the ovarian mass was significantly decreased, whereas 4- and 5-month-old rats showed the regular 4 day oestrous cycles and ovulation occurred. The ovarian CR activity towards four substrates declined steeply consistent with persistent oestrus and the enzyme activity for each substrate at 15 months of age was 1.4-53.7% lower than at 4 months of age. The ovarian CR concentration measured by western blot analysis was also lower at 8 months of age. Equine chorionic gonadotrophin (eCG) and hCG were administered to 10- and 15-month-old rats at persistent oestrus to examine whether the ovarian CR in rats at persistent oestrus is influenced by gonadotrophins. Treatment with hCG not only increased ovarian CR activity and its concentration but also caused ovulation at both ages. The immunoreactivity to anti-ovarian CR antibody in the theca cells of the hCG-treated ovary was greater than that of the control ovary. Equine(e) CG did not stimulate ovarian CR activity in rats at persistent oestrus, although ovarian mass of 10-month-old rats was increased after administration of 50 iu of eCG. These results indicate that changes in ovarian CR with ageing are related to the transition of regular oestrous cyclicity to acyclicity and that the expression of ovarian CR markedly decreased in aged rats is stimulated by hCG.

Ovarian carbonyl reductase delayed onset of persistent estrus in the offspring of parathyroidectomized mother rats.

We investigated the effects of maternal parathyroidectomy on day 5 of pregnancy on the ovarian carbonyl reductase (CR) in the offspring of rats. Changes in the ovarian CR of the offspring were examined at 4 and 8 months of age. In 8-month-old female offspring, the ovarian weight and the ovarian CR activity were significantly higher in rats from parathyroidectomized mothers than in rats from sham-operated mothers, and the offspring of the parathyroidectomized mothers showed the regular 4-day estrous cycles at 8 months of age, while the offspring of the sham-operated mothers were in the state of persistent estrus. Furthermore, intense immunostaining was found in the theca interna cells and the interstitial gland cells in the ovary of rats from the parathyroidectomized mothers, whereas in the ovary of the age-matched normal rats, immunostaining was faint. These results suggest that the maternal parathyroidectomy affect the activity and localization of CR in the ovary of female offspring.

Changes in adrenal carbonyl reductase activity in pregnant rats.

Changes in adrenal carbonyl reductase (CR) activity during pregnancy and postparturition in rats were investigated using two substrates, 13,14-dihydro-15-keto-prostaglandin F2 alpha (15KD-PGF2 alpha) and p-nitroacetophenone (PNAP). 15KD-PGF2 alpha reductase activity showed the lowest level on day 16 of pregnancy and steeply increased on the morning of postpartum day 1. PNAP reductase activity did not change up to day 18 of pregnancy, then was gradually increased and maintained the highest levels between day 19 and the morning of day 21 of pregnancy. Thereafter, the enzyme activity was rapidly decreased after parturition, on the contrary of changes in the 15KD-PGF2 alpha reductase activity. On the other hand, exogenous glucocorticoid treatment increased the 15KD-PGF2 alpha activity and decreased the PNAP reductase activity in adrenal glands of nonpregnant animals. Thus, it is found that the adrenal CRs in the above two classes are regulated by different mechanisms. As maternal corticosterone level is increased rapidly during late pregnancy by maternal transfer of fetal corticosterone, the present results suggest that adrenal CR activity during pregnancy, in particular at periparturition, may be influenced by fetal corticosteroids.

Localization and activity of adrenal carbonyl reductase during the estrous cycle.

Changes in carbonyl reductase (CR) activity towards seven substrates in adrenal glands of female rats were investigated during estrous cycle and the circadian patterns of adrenal CR activity towards seven substrates in male rats were examined. The localization of the enzyme protein in the adrenal gland was carried out by immunohistochemistry. In female rats, adrenal CR activity was lowest between 1200 and 2000 h on the day of proestrus, the phase in which a large, significant increase in ovarian steroids, gonadotropins and prolactin is observed. Immunoreactivity in the zona reticularis of the adrenal cortex to anti-adrenal CR, CR-K2, antibody on the day of proestrus was faint or negative compared with that on other days of the estrous cycle. In male rats, no major changes in adrenal CR activity were observed compared with females, although these were slight changes in all the enzyme activities. These results suggest that adrenal CR may be negatively regulated by the pituitary-adrenal axis in female rats and, in particular, on the day of proestrus.

Purification and properties of six aldo-keto reductases from rat adrenal gland.

Six aldo-keto reductases from rat adrenal gland have been highly purified to apparent homogeneity. These enzymes were identified as carbonyl reductases (CR-K1, CR-K2, CR-A, and CR-B), aldehyde reductase (AR-H), and aldose reductase (AR-L) in terms of substrate specificity, molecular weight (33,000-39,000), inhibitor susceptibility, cofactor requirement, and immunochemical properties. Both CR-K1 and CR-K2 were characterized as possessing high affinity towards 13,14-dihydro-15-ketoprostaglandin F2 alpha and were localized immunohistochemically in the zona glomerulosa and the zona reticularis of adrenal cortex, and in the ganglion cell of adrenal medulla. Immunoreactive proteins to anti-CR-K2 antibody were observed in male and female reproductive tissues of rats. Positive immunoreactive protein to anti-CR-A antibody was found in mouse, hamster, and rabbit adrenal gland, whereas that to anti-CR-K2 antibody was present only in rat adrenal gland. AR-H and AR-L mainly reduced aromatic and aliphatic aldehydes. All the aldo-keto reductases from rat adrenal gland were completely inhibited by p-chloromercuribenzoate. Barbiturate and 3,3'-tetramethylene glutarate potently inhibited AR-H, and quercitrin significantly decreased the activity of CR-K1, CR-K2, and AR-L. We propose that these aldo-keto reductases may play important roles in the rat adrenal functions.

Activation by human chorionic gonadotropin of ovarian carbonyl reductase in mature rats exposed in vivo to estrogens.

We investigated the effects of exogenous estrogens and human chorionic gonadotropin (hCG) on the activity, content, and immunohistochemical localization of ovarian carbonyl reductase (CR) in mature cycling rats. Estrogens, estradiol, hexestrol (HEX) and diethylstilbestrol (DES) were given s.c. to rats daily for 3 days from the first day of diestrus, and hCG was given s.c. at 3:00 p.m. on the day of expected proestrus. The ovaries were isolated on the day of expected estrus. Ovarian CR activity was measured by using two substrates that reflect the activity of the enzyme in rats, and the enzyme content was determined by western blot analysis. Ovarian CR activity and content were decreased by estrogens as well as by inhibition of ovulation; hCG restored both the activity and the content decreased by estrogens to levels produced by hCG alone. Nevertheless, the number of ova in the oviduct when ovulation was decreased or blocked by estrogens was not restored completely by hCG treatment. Faint immunostaining in the interstitial gland cells of HEX-treated rat ovaries was observed. These results suggest that (i) although hCG activates ovarian CR in estrogen-treated rats, this increase in both enzyme activity and content may not be an obligatory event in the ovulatory process, and (ii) exogenous estrogens may predominantly influence the ovarian CR in the interstitial gland cells in mature rats by inhibiting luteinizing hormone release from the pituitary.

Effects of age and calcium ion on testis carbonyl reductase in rats.

To examine the role of carbonyl reductase (CR) in the development and function of testis, age-related changes in CR and the effect of exogenous calcium on CR in rat testis were studied. Testicular CR activity was the highest at 3 weeks of age when the enzyme activity was measured at 2, 3, 4, 8 and 20 weeks of age. The intensity of positive protein bands in testicular cytosol was similar among these age groups in Western blot analysis with anti-rat ovarian CR antibody. In 3-week-old rats, intravenous administration of 4 mg/kg calcium markedly suppressed the activity to about 50% of the control level 1 min after the treatment. Pretreatment with nicardipine prevented the inhibitory effect of exogenous calcium on the testicular CR activity in 3-week-old rats, indicating that the suppressing action of calcium on the enzyme activity was mediated by the entry of extracellular calcium via calcium channels. There were no significant changes in the CR activity following the treatment with calcium and nicardipine in the other 3 age groups examined. These findings suggest that testicular CR plays an important role in the functional development of the testis during the infantile period in rats.

Interindividual variability of carbonyl reductase levels in human livers.

Interindividual variability of carbonyl reductase levels in human livers (N = 11) was examined by measuring reductase activity toward various substrates and by western blot analysis using anti-rat ovarian carbonyl reductase CR2 antibody. The carbonyl reductase activity toward p-nitrobenzaldehyde (PNBA) (58.1 +/- 5.4 nmol/mg protein/min, mean +/- SE) was highest among the substrates examined, followed by 4-benzoylpyridine (4BP) (14.4 +/- 2.0 nmol/mg protein/min) and p-nitroacetophenone (PNAP) (2.00 +/- 0.37 nmol/mg protein/min). The reductase activity (6.33 +/- 0.56 pmol/mg protein/min) toward 13,14-dihydro-15-keto-prostaglandin F2 alpha (15KD-PGF2 alpha), which is a diagnostic substrate for rat ovarian carbonyl reductases, was relatively high compared to that in other species. Western blot analysis revealed that each human liver contained several immunoreactive proteins to anti-CR2 antibody. The activities toward 15KD-PGF2 alpha (r = 0.85, P < 0.01) and 4BP (r = 0.84, P < 0.01), but not PNBA (r = 0.53, not significant) or PNAP (r = 0.52, not significant), were closely correlated with the relative amounts of the high molecular weight immunoreactive proteins determined with a densitometer. Thus, the major carbonyl reductases in human liver are similar to those of rat ovarian enzymes.

Gonadotropin-induced ovarian carbonyl reductase in mice and hamsters: comparison with carbonyl reductase in rats.

We investigated the effects of pregnant mare serum gonadotropin (PMSG) and human chorionic gonadotropin (hCG) on ovarian carbonyl reductase activities towards 13,14-dihydro-15-ketoprostaglandin F2 alpha (15KD-PGF2 alpha), p-nitroacetophenone (PNAP) and p-nitrobenzaldehyde (PNBA) in mice and hamsters, and compared with their effects on those we observed previously in rats. The treatment with PMSG and hCG caused a significant increase in ovarian weights and superovulation in both mice and hamsters. Hamster ovary possessed appreciable carbonyl reductase activities towards all three substrates, whereas the activities were lower than those in rat ovary. The reductase activities were not increased by the treatment with gonadotropins, differing from rat ovarian carbonyl reductase. In untreated mice, carbonyl reductase activity towards 15KD-PGF2 alpha was not detected, whereas the activities towards PNAP and PNBA were detected, which activities were lower than those in rats and hamsters. The PNAP and PNBA reductase activities in mouse ovary were significantly increased up to 7.1- and 1.7-fold, respectively, by the treatment with gonadotropins. These results show that there are species differences in ovarian carbonyl reductase and response of the enzyme to gonadotropins.

Carbonyl reductase from human testis: purification and comparison with carbonyl reductase from human brain and rat testis.

Carbonyl reductase (EC 1.1.1.184) is a cytosolic, monomeric, NADPH-dependent oxidoreductase with broad specificity for carbonyl compounds and a general distribution in human tissues. A carbonyl reductase closely resembling the human enzyme is exclusively expressed in rat reproductive tissues and adrenals (Iwata, N., Inazu, N. and Satoh, T. (1989) J. Biochem. 105, 556-564). In order to investigate the relationship between the human and rat enzyme, carbonyl reductase from human testis was purified to homogeneity. The enzyme was indistinguishable from carbonyl reductase from other human tissues on the basis of physicochemical properties, substrate specificity, inhibitor sensitivity and immunological reactivity. Likewise, the human and rat testis enzymes exhibited greatly overlapping substrate specificities for prostaglandins, steroids as well as many xenobiotic carbonyl compounds, and showed the same susceptibility to inhibition by flavonoids and sulfhydryl-blocking agents. Structural homology between the two enzymes was indicated by the mutual cross-reactivity of antibodies against carbonyl reductase from one species and the enzyme protein from the other species. Unlike the rat enzyme, which is confined to Leydig cells, the human enzyme was detectable in Leydig cells as well as Sertoli and spermatogenic cells.

Consecutive treatment of disulfiram inhibits ovarian carbonyl reductase activity in rats.

We investigated the effects of disulfiram (DS) on ovarian carbonyl reductase activity in rats to determine the influence of DS on female reproductive function. Three consecutive treatments with DS significantly inhibited ovarian carbonyl reductase activity as well as ovulation, dose-dependently. Single treatment with DS had no effect on ovarian carbonyl reductase activity. Our observations indicate that consecutive treatment with DS has an inhibitory action on female reproductive function, although DS is well-known to inhibit aldehyde dehydrogenase and dopamine beta-hydroxylase.

Human chorionic gonadotropin causes an estrogen-mediated induction of rat ovarian carbonyl reductase.

We earlier reported that human chorionic gonadotropin (hCG) stimulates rat ovarian carbonyl reductase (CR) activity and content, and that estrogen enhances the stimulatory effect. The present study was performed to determine the mode of action of the gonadotropin. Cycloheximide (CHX) and actinomycin D (AD) were given to estradiol-pretreated immature rats 6 h before hCG treatment. The enzyme activity was measured with three substrates, and the enzyme content was determined by the method of Western-blot analysis using anti-rat ovarian CR anti-serum as the first antibody. Both protein inhibitors significantly prevented hCG from increasing the enzyme activity and content in estradiol-pretreated ovary. These results indicate that rat ovarian CR is induced by LH via the action of estrogen.

Effects of antiestrogens on ovarian aldo-keto reductase in relation to ovulation in rats.

The pharmacological effects of antiestrogens on ovarian aldo-keto reductase and ovulation were investigated in rats. The activities of reduction of 13,14-dihydro-15-keto-PGF2 alpha, 4-benzoylpyridine and menadione in ovarian cytosol were significantly decreased by antiestrogen treatments, and the ovulation was completely inhibited. However, administration of luteinizing hormone-releasing hormone at 3:00 p.m. on the day of proestrus restored both enzyme activities and ovulation, which were inhibited by antiestrogens, to control levels. These results indicate that nonsteroidal antiestrogen inhibits the luteinizing hormone surge on the day of proestrus in 4-day cycling rats and that ovarian aldo-keto reductase may be closely involved in the ovulatory process in rats.

Regulation of ovarian carbonyl reductase mediated by estrogen receptor in immature rats.

In the present study, the enhancing effect of synthetic estrogen on ovarian carbonyl reductase, a new prostaglandin (PG)-metabolizing enzyme, was investigated, and the antagonistic effect of antiestrogen on this enhancement was examined in immature rats. Ovarian carbonyl reductase activity towards 13,14-dihydro-15-keto-PGF2 alpha (15KD-PGF2 alpha), 4-benzoylpyridine (4BP) and menadione was determined photometrically and radiochemically, and quantitation of ovarian carbonyl reductase content was performed by Western blot analysis. Diethylstilbestrol (DES) and hexestrol (HX) enhanced the increasing effect of human chorionic gonadotropin (hCG) on ovarian carbonyl reductase activity and content when these synthetic estrogens (0.2 mg/kg) were administered for 3 days from 26 days of age, before hCG treatment. On the other hand, tamoxifen, which inhibits the binding of estradiol to the estrogen receptor, significantly prevented estradiol (E2) from enhancing the effect of hCG on ovarian carbonyl reductase. Furthermore, the ovarian carbonyl reductase activities towards the three substrates correlated well with the ovarian carbonyl reductase content. These results indicate that ovarian carbonyl reductase in immature rats may be regulated by a specific increase in the ovarian response to luteinizing hormone mediated by estrogen receptor.