| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Canada Research Chair in Pharmacogenomics (C.G.); Centre de Recherche du Centre Hospitalier de l’Université Laval, Department of Molecular Endocrinology and Oncology (J.L., O.B., G.P., F.L., A.B., C.G.), G1V 4G2 Québec, Canada; Laval University, Faculty of Pharmacy (J.L., O.B., C.G.), Faculty of Medicine, Gynecologic Oncology Service, Hôtel-Dieu de Québec (M.P.), Department of Pathology Hôtel-Dieu de Québec (B.T.), G1K 7P4 Québec, Canada
Address all correspondence and requests for reprints to: Chantal Guillemette, Canada Research Chair in Pharmacogenomics, Pharmacogenomics Laboratory, Centre Hospitalier de l’Université Laval Research Center, T3-67, 2705 Boulevard Laurier, Québec G1V 4G2, Canada. E-mail: chantal.guillemette{at}crchul.ulaval.ca.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recent findings highlight the importance of CE metabolites as chemical mediators and their link to cancer development and progression (7, 8, 9). For instance, 4-OHE2 undergoes metabolic redox cycling to generate reactive oxygen species, which may damage lipids and DNA and initiate estrogen-related carcinogenesis (10). By contrast, biotransformation of CE to 2-MeOCE may result in the formation of a protective metabolite having very potent inhibitory effect on cell proliferation, tubulin activity, and angiogenesis (11, 12, 13). The formation of CEs and MeOCEs may proceed through reactions carried out by hepatic metabolism but also through local metabolic conversion in extrahepatic hormone responsive organs or through a combination of both. Because of their unique chemical and biological properties, it is critical to establish the contribution of extrahepatic formation and elimination of estrogen and of their metabolites.
Conjugation transforms estrogenic hormones into less active, more polar and water-soluble metabolites, thus facilitating their excretion in the bile and urine. Conjugative estrogen metabolism includes sulfonation (14) and glucuronidation (15, 16, 17, 18), two reactions catalyzed by sulfotransferases and uridine diphospho-glucuronosyltransferases (UGTs), respectively. In fact, sulfates and glucuronides are the most abundant circulating estrogen conjugates (19), whereas estrogen sulfates represent a form of estrogen storage that acts as a precursor of E2 (14). On the other hand, glucuronidation by UGTs leads to complete inactivation of estrogen (17, 19). To the best of our knowledge, human ß-glucuronidase activity has never been reported in the literature for the endometrium. Glucuronidation of steroids is not limited to the liver, and this metabolic pathway is recognized as contributing significantly to the modulation of specific physiological effects of steroids in target tissues such as the prostate (15, 20, 21, 22). Furthermore, the detection of significant amounts of estrogen glucuronides in the breast cystic fluid supports the role of the glucuronidation metabolism in the elimination of estrogens (23, 24). The presence of UGTs was reported in several estrogen target organs such as the breast and uterus (15, 25, 26), even though a comprehensive analysis of the expression of every UGT enzyme reactive toward estrogens in target tissues has never been completed. Several human UGTs are known to exert catalytic activities toward E2, E1, and their hydroxylated metabolites, whereas the glucuronidation of MeOCE remains to be clarified (17, 18, 27, 28, 29, 30). To date, 16 functional UGT proteins have been reported (31, 32). In vitro studies using human UGT cDNAs cloned and expressed in mammalian cells have been used to determine the specificity of UGT isoenzymes for some estrogens in a small number of studies (15, 17, 18, 28, 29, 30), whereas a systematic study using the same experimental conditions remains to be conducted. The detailed regioselectivity of UGTs has never been systematically established, possibly because of the lack of estrogen glucuronide standards and suitable analytical methods.
In the present study, we hypothesized that E2, E1, and their oxidatized metabolites undergo glucuronidation at various positions and that these site-specific reactions are catalyzed by different UGTs. Given that the enzymatic machinery for the formation of catechol and methoxy metabolites of estrogen was demonstrated in the uterus (5, 6), it was expected that UGT involved in the glucuronidation of estrogen would also be expressed. We have first developed a highly sensitive and specific mass spectrometry-based analytical method to thoroughly establish the substrate specificity and stereoselectivity of all known functional human UGT proteins toward E2 and E1 and their hydroxyl as well as their MeOCE derivatives under the same experimental conditions. In the course of this study, we accumulated data on the enzymatic kinetics [maximal velocity (Vmax), substrate concentration at half-maximal velocity (Km), and catalytic efficiencies] of individual UGT isoforms to determine which UGTs are the most efficient in catalyzing glucuronidation at the 2-, 3-, 4-, and 17-hydroxyl positions of estrogens and define their stereoselectivity. Subsequently we explored the expression of UGT2B7 that metabolizes with the highest catalytic efficiency the genotoxic 4-OHCEs in the uterus by RT-PCR and immunohistochemistry using a specific antibody. Altogether, the data indicate that glucuronidation of estrogen may be one of the mechanisms of major biological significance in the inactivation and elimination of estrogens, their reactive CEs, and MeOCE metabolites in the uterus.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
E1, E2, 2-OHE1, 4-OHE1, 2-OHE2, 4-OHE2, 2-MeOE1, 4-MeOE1, 2-MeOE2, 4-MeOE2, E1-3G, E2-3G, and E2-17G were purchased from Steraloids (Newport, RI). The glucuronide derivatives of 2-OHE1, 4-OHE1, 2-OHE2, 4-OHE2, 2-MeOE1, 4-MeOE1, 2-MeOE2, and 4-MeOE2 were synthesized by Endorecherche (Québec, Canada), and confirmation of the structure of each product was determined by nuclear magnetic resonance. Methanol (HPLC grade), acetonitrile (HPLC grade), 1-chlorobutane (HPLC grade), and iso-amyl-alcohol were obtained from VWR Canlab (Montréal, Canada). Ammonium hydroxide (trace metal grade) was purchased from Fisher Scientific (Nepean, Canada).
Characterization of human embryonic kidney (HEK)293 cell systems expressing UGT1A and UGT2B
Microsomal proteins used for UGT expression were prepared as previously described (33, 34, 35). To ascertain the level of UGT protein expression in the stable UGT1A- and UGT2B-HEK293 clones, a semiquantitative immunoblot analysis method was used. The microsomal fractions from UGT-HEK293 cells were used in all enzymatic assays, whereas in the case of UGT1A10, commercial microsomal fractions of Sf-9 insect cells infected with a baculovirus strain containing human UGT1A10 cDNA were used (Panvera, Madison, WI). Microsomal proteins (10 µg protein) were separated on SDS-PAGE and transferred onto nitrocellulose. For quantification of the UGT1A proteins, we used the antihuman UGT1A common carboxyl terminus region (amino acids 312–531) antiserum RC-71, as previously reported (15, 34, 35). UGT2B protein levels were quantified using the antihuman UGT2B antibody (EL-93), as previously described (36). To normalize sample loading, blots were stripped and reprobed with anticalnexin antibody (Stressgen Biotechnologies, Victoria, Canada) to detect a second endoplasmic reticulum-resident protein, calnexin. Bands were visualized using enhanced chemiluminescence (Amersham, Piscataway, NJ) and quantified by Bioimage Visage 110s from Genomic Solutions Inc. (Ann Arbor, MI). Ratios between UGT and calnexin signals were calculated for each UGT except for UGT1A10 baculosome for which the UGT signal alone was considered for the expression level.
Assay of UGT activity
Microsomal fractions from UGT-HEK293 cells and commercial microsomal fraction of UGT1A10 were used in enzymatic assays. Reactions (100 µl volume) contained 50 mM Tris-HCl (pH.7.5), 10 mM MgCl2, 100 µg/ml phosphatidylcholine, 8.5 mM saccharolactone, 2 mM uridine diphospho-glucuronic acid, and 40–60 µg membrane protein and steroid substrate (25 µM). Reactions were initiated by adding varying concentrations of each individual substrate. Blanks and controls contained all compounds except steroid substrates and microsomal preparation, respectively. The assays were terminated by adding 100 µl methanol and were centrifuged at 14,000 x g for 10 min before analysis. Time-course experiments were designed to determine the linearity of the glucuronidation reaction. All reaction rates were shown to be linear under the selected time interval, and inclusion of detergent was found to be unnecessary for assessment of the full glucuronidating potential of UGT-expressing HEK cell membranes. For determination of Vmax and Km values, HEK293 cells stably expressing UGT enzymes were incubated in the presence of varying substrate concentrations from 0.5 to 200 µM for 1 h. Absolute glucuronidation activities were divided by the level of UGT protein and expressed as relative glucuronidation activities in picomoles per minute per milligram.
Mass spectrometry analysis
Incubation medium was diluted with 0.1 ml methanol/water (50:50; vol/vol), vortexed, and then transferred into a conical vial for injection into the mass spectrometer. The HPLC and tandem mass spectrometry (MS/MS) system consisted of a mass spectrometer (model API 3000, Perkin-Elmer Sciex, Thornhill, Canada) equipped with an elecrospray ionization source in the negative ion mode and a HPLC pump plus autosampler (model 2690, Waters, Milford, MA). A chromatographic separation was achieved with a 100 x 3.9 mm Xterra MS C8 column 3.5 µ m packing material (Waters), using a three-solvent isocratic system (A: water + 0.1% ammonium hydroxide; B: methanol + 0.1% ammonium hydroxide; C: acetonitrile + 0.1% ammonium hydroxide) at a constant flow rate (0.8 ml/min) for 8 min with 90% A, 8.5% B, and 1.5% C. Afterward, the column was washed with 93.5% B and then reequilibrated to initial conditions over 4 min. The mass spectrometer was operated in the multiple reaction monitoring (MRM) mode, and the quantification of the products formed was performed using a calibration curve in the range of 200–10,000 pg/ml of incubation medium for each substrate.
Tissue procurement, RNA purification, reverse transcription, and expression analyses
Nonmalignant uterine tissue was obtained from postmenopausal women who had not received hormone replacement therapy for at least 3 wk and who had no menstrual bleeding for at least 1 yr. All subjects provided written consent for use of their specimens, and the present study was reviewed and approved by Institutional Review Boards (Hôtel-Dieu de Québec, Centre Hospitalier de l’Université Laval Research Center, and Laval University). Fresh material from patients were all collected by the pathologist and immediately deposited in liquid nitrogen within 30 min of surgery pending transfer to a freezer at –80 C. RNA was isolated by the acid guanidium thiocyanate/chloroform method (37). Total RNA was collected with Trizol (Molecular Research Center, Cincinnati, OH), and random hexamers (pDN6) were used to synthesize cDNA from total RNA (1 µg) using a SuperScript II cDNA synthesis kit (Invitrogen, Carlsbad, CA) according to the manufacturer’s specifications. Semiquantitative RT-PCR was used to study the expression of the UGT2B7 gene using primers 882 forward: 5'-ctacgtacctgttgttatgtcagaattaac and 883 reverse: 5'-actgatcccacttcttcatgtcaa. For amplification of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), these primers were used: 41 forward: 5'-tgggtgtgaaccatgag and 42 reverse: 5'-cccagcgtcaaaggtgg. PCR products were sequenced. All PCR amplifications were performed using Taq DNA polymerase at 95 C, 30 sec for denaturation, 59 C, 40 sec for annealing, and 72 C, 45 sec for extension in a thermal cycler, with a fixed number of 35 cycles for each pair of primers. GAPDH was used as internal standard.
Characterization of the polyclonal anti-UGT2B7 antibody
Anti-UGT2B7 antibody was raised against amino acids 55–164 (110 amino acids length) of the protein. This fragment corresponds to the portion of the UGT2B7 protein that is the least homologous to other human UGT2B. Several rabbits were injected with a total of 100 µg purified fusion proteins in the presence of incomplete Freund’s adjuvant. The production of antibodies was determined 12 d after the injections. To gain information concerning the novel anti-UGT2B7 antibody, Western blot experiments using the recombinant UGT2B7 protein fragment and microsomal proteins from UGT2B7-HK293 cells showed that the immune serum 1809 is immunoreactive for UGT2B7. The specificity of the anti-UGT2B7 1809 antibody was studied by Western blot experiments using microsomal proteins (10 µg) of HK293 cells stably expressing all known UGT2B subfamily members, namely UGT2B4, 2B7, 2B10, 2B11, 2B15, 2B17, and 2B28. Microsomal proteins of human liver were used as positive control. The gel was transferred onto nitrocellulose membrane and incubated with anti-UGT2B7 1809 antibody (dilution 1:2000). Bands were visualized using enhanced chemiluminescence (Amersham) and quantified by Bioimage Visage 110s (Genomic Solutions).
Immunohistochemistry
Six nonmalignant human uterine tissues were also fixed in formol and embedded in paraffin for immunohistochemical studies. The tissues were serially cut at 5-µm and sections mounted on glass slides. The sections were deparaffinized, hydrated, and incubated overnight at 4 C with the human UGT2B7 antiserum diluted 1:250 in Tris-saline (pH 7.6). The sections were then washed and incubated at room temperature for 4 h with peroxidase-conjugated goat antirabbit immunoglobulins (DakoCytomation, Glostrup, Denmark) diluted 1:200 as previously described (38). Endogenous peroxidase activity was eliminated by preincubation with 3% H2O2 for 30 min, and peroxidase was then revealed during incubation with 15 mg of 3,3'-diaminobenzidine in 100 ml Tris-saline buffer containing 0.03% H2O2. The intensity of the staining was controlled under the microscope. The sections were then counterstained with hematoxylin. Control experiments were performed on adjacent sections by substituting preimmune rabbit serum (1:250).
Enzyme kinetic analysis
Visual inspection of fitted functions [velocity as a function of substrate concentration] and Eadie-Hofstee plots [velocity as a function of (velocity/substrate concentration)] was used to select the best fit enzyme kinetic model (39). These include the Michaelis-Menten model [V = (Vmax x S)/(Km + S)], the substrate activation model [Hill equation, V = (Vmax x Sn)/(Kmn + Sn)] and the uncompetitive substrate inhibition model [V = (Vmax x S)/(Km + S x (1 + S/Ks))], where n is the Hill coefficient or the degree of curve sigmoidicity, and Ks is an inhibition constant. Analysis of data showing sigmoid kinetic was performed using SigmaPlot 8.0 with Enzyme Kinetics 1.1 (SPSS, Chicago, IL), and analysis of data showing substrate inhibition or Michaelis-Menten kinetics was performed using Enzfitter (Biosoft, Ferguson, MO). For Michaelis-Menten and substrate inhibition models, the intrinsic clearance (CLint) estimation values were calculated based on the following equation: CLint = Vmax/Km, whereas for sigmoid model the maximal clearance (CLmax) estimation values were calculated using the following equation: CLmax = (Vmax/Km) x [(n – 1)/(n x (n – 1)1/n)] (40).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The identity of all conjugated products formed by expressed UGTs was confirmed by comparison with the retention time of HPLC and mass fragmentation of authentic standards using HPLC and tandem MS (MS/MS) methods developed to specifically detect and quantify the E glucuronides. The analyses of those compounds by MS/MS were assessed by MRM mode (parent and daughter ion) to provide higher specificity than single ion monitoring (Table 1
). The high accuracy of this analytical method is also achieved by the quality of the chromatographic separation achieved between the two glucuronide products potentially formed from each individual 2-OHCEs and 4-OHCEs. As illustrated in Fig. 1A, despite the same parent and daughter ions for 2-OHE2-2G and 2-OHE2-3G, it was possible to chromatographically separate the two estrogen glucuronides. Good chromatographic resolution was also obtained for 4-OHE2-4G and 4-OHE2-3G, which also have the same ions (Fig. 1B).
|
|
Assays of human UGTs toward a series of estrogens, namely E1, E2, 2-OHE1/E2, 4-OHE1/E2, and their methoxy derivatives 2-MeOE1/E2 and 4-MeOE1/E2, were performed using microsomal preparations of HK293 cells engineered to stably express all known human UGTs, with the exception of UGT1A10, which was overexpressed in baculosomes.
The initial screening of the conjugating activity was performed by incubation with the selected microsomal preparation for 16 h containing 25 µM of the estrogenic substrate and 1 mM of glucuronic acid. Most of the UGT2B isoenzymes did not catalyze the glucuronidation of estrogens, except UGT2B7, whereas UGT2B4’s activity was minimal (Table 2). On the other hand, several UGT1As were capable of conjugating estrogens with high efficiency, namely UGT1A1, UGT1A3, UGT1A8, UGT1A9, and UGT1A10 (Table 2). The formation of glucuronidated products at positions 2, 3, and 4 of hydroxylated E2/E1 was achieved by several UGT isoforms, whereas conjugation at position 17 of 2- and 4-hydroxylated E2 could not be detected (data not shown).
|
). The values of apparent Km and relative Vmax presented are from one experiment performed in duplicate. Kinetics were repeated for most of the substrates and similar results were obtained in independent experiments. On visual inspection of fitted functions and Eadie-Hofstee plots to select the best fit enzyme kinetic model (39) among Michaelis-Menten, substrate activation, and uncompetitive substrate inhibition, kinetic parameters were calculated using equations described in Materials and Methods. The detailed experimentation has allowed us to delineate kinetic properties of UGTs toward estrogens. Various enzymatic kinetic profiles were observed depending on the substrate, especially for E1, E2, and 2-OHCEs, and also depending on the products formed as well as the position to which the glucuronic acid is transferred. Similar reports of atypical kinetic properties of human UGTs have recently been published (41, 42).
|
UGT1A1, UGT1A3, UGT1A8, and UGT1A10 conjugate both E2 and E1, whereas UGT2B7 and UGT1A9 glucuronidate specifically E2 and E1, respectively (Table 3
). Conjugation of E2 was exclusively at position 3 for all UGTs tested, except for UGT2B7, which glucuronidated E2 at position 17 and to a lesser extent by UGT1A3. Km values for the formation of E2-3G and E2-17G were almost in the same range varying from 10 to 47 µM, whereas higher Km values were observed for E1-3G (38–90 µM). In general, lower velocity was observed for E1-3G, compared with E2-3G and E2-17G conjugation, with the exception of UGT1A8, which demonstrated the highest Vmax for both E1-3G (142 pmol/min·mg) and E2-3G (195 pmol/min·mg).
Depending on which UGT was involved, glucuronidation of E1 and E2 followed sigmoid and hyperbolic kinetic profiles (Fig. 2). Data for UGT1A1 and UGT1A3 best fitted a sigmoid profile for E1 and E2, and UGT1A9 and UGT2B7 exhibited hyperbolic profiles for the conjugation of E1 and E2. In turns, UGT1A8 and UGT1A10 demonstrated a kinetic profile dependent of the estrogenic substrate with a hyperbolic profile for E2 and a sigmoid profile for E1.
|
). The highest value of catalytic efficiency was observed for E1-3G with UGT1A8 (1.3 µl/min·mg). Glucuronidation of 2-OHCE
Five UGTs, namely UGT1A1, UGT1A3, UGT1A8, UGT1A9, and UGT2B7, are responsible for the conjugation of 2-OHE2 and 2-OHE1 with the exception of UGT2B7, which shows activity for only 2-OHE2. Although position 17 of 2-OHE2 is not glucuronidated by UGTs (data not shown), every reactive UGT toward 2-OHCEs conjugates hydroxyls at both positions 2 and 3, except UGT1A1, which demonstrates a regioselective activity for the glucuronidation of 2-OHE2 at position 2 and not for 2-OHE1 at the same position (Table 3).
Glucuronidation of 2-OHCEs by UGT1A1, UGT1A3, and UGT1A8 display hyperbolic, sigmoid, or substrate inhibition profiles, whereas glucuronidation of 2-OHCEs by UGT1A9 and UGT2B7 follows a hyperbolic profile for both positions of 2-OHCEs (Fig. 3, A and B). The values of apparent Km for conjugation at positions 2 or 3 of 2-OHE2 vary widely, from 15 to 479 µM, whereas the Km values are slightly higher for 2-OHE1, especially at position 3, with UGT1A3 having a Km of 587 µM (Table 3). In fact, the catalytic efficiencies of 2-OHE2 conjugation by UGT1A8 and UGT2B7 are markedly higher than those of UGT1A1, UGT1A3, and UGT1A9. In fact, the catalytic efficiencies are as high as 11.3 µl/min·mg protein for glucuronidation at position 3-hydroxy of 2-OHE2 by UGT2B7. Values are approximately 10-fold lower for conjugation at position 2-hydroxy, with the highest being at 1.3 µl/min·mg protein. On the other hand, the values of the catalytic efficiencies are similar for all enzymes in the conjugation of 2-OHE1, with the exception of UGT1A1, which is much more efficient at conjugating 2-OHE1 at position 3 (Table 3).
|
Glucuronidation of 4-OHCE
As observed for 2-OHE1/E2, 4-OHE1/E2 can be conjugated in the two hydroxylated positions of ring A of the estrogen, whereas position 17 remains not conjugated (data not shown). For both 4-OHE2 and 4-OHE1, UGT1A1, UGT1A8, UGT1A9, and UGT2B7 have significant conjugating activities. In contrast to what was observed for 2-OHCEs, glucuronidation of 4-OHCEs by UGTs exhibits an hyperbolic profile, except for UGT1A1, which demonstrates substrate inhibition and sigmoid profiles for conjugation of both hydroxylated positions of 4-OHE2 and 4-OHE1, respectively (Fig. 3, C and D).
The Km values for the 3 and 4 positions of 4-OHE2 ranged from 38 to 118 µM, except UGT2B7, which has much higher affinity for 4-OHE2 with Km values of 10 and 22 µM at positions 4 and 3, respectively (Table 3). The catalytic efficiency of 4-OHE1 glucuronidation at position 3 by the UGT isoforms appears relatively low, compared with conjugation of the same molecule at position 4. Two isoforms are major in the conjugation of position 4 of 4-OHE2, namely UGT1A8 and UGT2B7, which demonstrate two of the highest Vmax values observed, 4723 and 5235 pmol/min·mg, respectively. Although the conjugation of 4-OHE1 was lower, compared with 4-OHE2, UGT2B7 demonstrates the highest Vmax at 8724 pmol/min·mg. The catalytic efficiency of UGT2B7 was about 5-fold higher for 4-OHE2 (523.5 µl/min·mg protein), compared with 4-OHE1 (140.7 µl/min·mg protein). Conjugation at the 3 hydroxyl position of 4-OHCE is preferentially catalyzed by UGT1A8, whereas the predominant substrate is 4-OHE1. The relative Vmax of 1975 pmol/min·mg and the catalytic efficiency of 10.7 µl/min·mg protein for UGT1A8 are at least 10-fold higher than the other UGTs catalyzing 4-OHE2-3G and 4-OHE1-3G. It is concluded that 4-OHE1 is predominantly conjugated at position 4 by UGT2B7, whereas for 4-OHE2, UGT1A8 and UGT2B7 are the predominant isoforms involved in the conjugation at position 4 (Table 3).
Glucuronidation of 2-MeOCE and 4-MeOCE
Our data demonstrate that five UGTs, namely UGT1A1, UGT1A3, UGT1A8, UGT1A9, and UGT1A10, conjugate 2-MeOCE. Of them, only UGT1A1 and UGT1A8 conjugate 4-MeOCE. As observed for other hydroxylated E, position 17 was not conjugated. All UGTs displayed a hyperbolic kinetic profile (curves not shown).
Glucuronidation of 2- and 4-MeOE2 occurs at higher rates than for 2- and 4-MeOE1, these reactions being mostly catalyzed by UGT1A8. The Km values of all substrates tested are almost in the same range, between 49 and 126 µM, with the exception for 2-MeOE2 by UGT1A10, 4-MeOE2 by UGT1A1, and UGT1A8 with Km values of 12, 14, and 8 µM, respectively (Table 3). The catalytic efficiencies of 4-MeOE2 and 2-MeOE2/E1 by UGT1A8 were the highest at 27.3, 15.9, and 10.9 µl/min·mg protein, respectively. UGT1A1 and UGT1A3 were also efficient at conjugating 2-MeOE2 with catalytic efficiencies of 4.5 and 3.1 µl/min·mg protein, respectively.
Specificity of the polyclonal anti-UGT2B7 antibody
Western blot analysis using microsomal proteins (10 µg) of HK293 cells and HK293 cells stably expressing UGT2B4, 2B7, 2B10, 2B11, 2B15, 2B17, and 2B28 revealed that anti-UGT2B7 1809 antibody is specific to UGT2B7 protein. Microsomal proteins of human liver were used as positive control (Fig. 4A).
|
The expression of the UGT2B7 was first assessed in endometrial tissues collected from six postmenopausal women by RT-PCR. Significant variation in UGT2B7 transcripts between endometrial samples was evident (Fig. 5). Immunostaining using specific antibodies to human UGT2B7 showed the presence of immunoreactive material in nonmalignant endometrium and myometrium. Both epithelial cells lining the glands and those covering the surface were strongly reactive (Fig. 4B). Endometrial stromal cells were also immunoreactive (Fig. 4B). In the myometrium, the vast majority of smooth muscle cells exhibited moderate immunolabeling (Fig. 4C). When the nonimmune serum was used, no staining could be detected (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Detailed kinetic analyses point out that only six UGT enzymes of all 16 tested are involved in this process, namely UGT1A1, UGT1A3, UGT1A8, UGT1A9, UGT1A10, and UGT2B7, whereas a limited number of UGTs display regioselectivity for the estrogenic molecules. The data further indicate that UGT2B7, the most reactive UGT toward the genotoxic 4-OHCEs, is expressed at variable levels in endometrial tissue of postmenopausal women, even though the sample size was limited. Our observations support the concept that E1 and E2 and their metabolites formed locally in estrogen sensitive tissues (5, 6) are conjugated by specific UGT enzymes in situ to achieve complete inactivation before their release in the circulation. The data further suggest a potential biological role of this metabolic pathway in the inactivation of genotoxic estrogens and that UGT2B7 is a key player in this process.
|
, 17ß-diol, and androsterone (22). Although less attention was given to estrogen glucuronidation, few studies indicated that E1 and E2 may also be conjugated by specific UGTs in estrogen-sensitive tissues (15, 25, 33, 43, 44). Besides, E1 and E2 are converted into multiple hydroxylated metabolites in not only the liver but also several peripheral tissues (3). Recently Lee et al. (45) conducted a study aimed at characterizing the major metabolites of E1 and E2 by selected human CYPs. The authors concluded that the main oxidative pathway of E1 and E2 is likely by CYP1A1, CYP3A4, and CYP1B1, which predominantly leads to catechol estrogen production. In fact, CYP1A1 and CYP3A4 are responsible for the production of 2-OHCE and are expressed in estrogen-sensitive tissues, including the endometrium. CYP1B1 is exclusively found in extrahepatic tissues, namely the breast and endometrium, and produces 4-OHCE (5, 46, 47). These observations are concurrent with the recent findings of Rogan et al. (48), who demonstrated substantial amounts of CEs and their methoxyestrogen derivatives in breast tissue. However, in that study, the measurements of estrogen metabolites in the tissue were performed after treatment of the homogenate extracts with ß-glucuronidase, thus indicating the potential for local formation of polar glucuronide derivatives. In fact, earlier studies demonstrated high concentrations of E1 and E2 glucuronides in the breast cyst fluid (23, 24), which is supported by the expression of several UGT isoenzymes (25, 33). The present findings suggest the formation of estrogen glucuronide derivatives in a second estrogen-sensitive tissue, namely the endometrium.
As demonstrated recently for CYPs (45), human UGTs catalyze the transfer of glucuronic acid to estrogenic substrates with a large range of catalytic efficiency and a unique stereoselectivity (Fig. 6). Even though more than one UGT is frequently involved in the conjugation of various estrogens, data indicate that several UGTs possess a distinctive regioselectivity for specific estrogenic molecules. The first general observation derived from our metabolic studies is that UGT1A family members conjugate estrogens at positions 2, 3, and 4 but not at position 17, with the exception of UGT1A3, which, however, shows a very low catalytic efficiency for this position. Previous observations have already indicated that UGT2B7 exclusively conjugates E2 at position 17, whereas UGT1A1 was the enzyme responsible for the glucuronidation at position 3 (18, 28, 29). However, the present data clearly show that UGT1A8 is the predominant enzyme for conjugation of E2 at position 3 followed by UGT1A1. The data further reveal that E2 is the preferred estrogenic substrate for glucuronidation by human UGTs, compared with its keto metabolite E1. This affirmation is also true for their corresponding 2-OHE1/E2, 4-OHE1/E2, and 4-MeOE1/E2 derivatives.
It is believed that the polarity of the estrogenic substrate has an important role in the subsequent conjugation via the glucuronidation pathway. Additional examination of data indicates that the glucuronidation of 2-OHE2 and 2-OHE1 is in large part at position 3 and that this conjugation is performed by UGT1A1, UGT1A8, and UGT2B7, whereas a modest activity is also observed for UGT1A3 and UGT1A9. In contrast, 4-OHE2 and 4-OHE1 are conjugated predominantly at position 4 and UGT2B7 is the major enzyme involved. In fact, it is the first time that the glucuronidation at position 4 of 4-OHE2/E1 is observed. Based on the in vitro enzyme kinetic data, the inactivation of 4-OHCEs is catalyzed by several enzymes but largely by UGT2B7. As previously observed for other estrogen metabolites, the change of the hydroxy group (E2) to the keto group (E1) at position 17 reduces by more than 50% the efficiency of UGT2B7. Recently Gestl et al. (25) have shown that the expression of UGT2B7 is particularly decreased in cancerous mammary gland tissue, compared with normal tissue. These authors suggested that UGT2B7 would protect the mammary gland from the genotoxic estrogen 4-OHE1. Our data support the involvement of UGT2B7 in the inactivation of 4-OHCE and demonstrate its expression in a second estrogen-responsive tissue, the endometrium, which also expresses CYP1B1 involved in the local formation of 4-OHCE (5). In the uterus of postmenopausal women, we further showed using the UGT2B7 antibody immunoreactivity in the cells that also express CYP1B1, involved in the formation of 4-OHCEs. As for CYP1B1, UGT2B7 staining was predominantly seen in the cytoplasm in epithelial cells lining the uterine glands and those covering the surface in addition to smooth muscle cells of the myometrium (49, 50). These observations suggest that the formation of genotoxic 4-OHCE by CYP1B1 and subsequent conjugation by UGT2B7 may occur within the same cells of the uterus, supporting the protective role of UGT2B7.
The present study further highlights the previously unrecognized complexity of estrogen metabolism and suggests that the glucuronidation pathway may be significant in estrogen metabolism in extrahepatic tissue such as the mammary gland and uterus. Because E2 and its 2-OH, 4-OH, and 2-MeO metabolites exert various biological effects, the present findings indicate that the glucuronidation pathway may represent a critical process in uterine cells, thus influencing estrogen exposure in this tissue. In agreement with this hypothesis, we recently demonstrated that a lower expression of UGT1A1, a UGT involved in the conjugation 2-OHCE and 2-MeOCE that mediate proliferative inhibitory effects (11, 12, 13), significantly decreases the risk of endometrial cancer (26). These data suggest that lower expression of UGT1A1 decreases the risk of endometrial cancer by reducing the excretion of the antiproliferative metabolite of E2, 2-MeOE2, in the endometrium. It is thus predicted from our kinetic data that additional UGTs, such as UGT2B7, play a significant role in protecting locally against 4-OHCEs by reducing their mutagenic potential in estrogen-responsive tissues such as the breast and uterus. In fact, the evaluation of the relative level of expression of the most reactive UGTs toward estrogens in nonmalignant endometrium tissues support the high level of expression of the UGT2B7 isoenzyme in this tissue (data not shown). Altogether, our data highlight the major importance of determining the presence of estrogen glucuronide derivatives in estrogen-sensitive tissues and into circulation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: CE, Catecholestrogen; COMT, catechol-O-methyltransferase; CYP, cytochrome P450; E1, estrone; E2, estradiol; -G, glucuronide; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HEK, human embryonic kidney; Km, substrate concentration at half-maximal velocity; 2-MeOCE, 2-methoxycatecholestrogen; 4-MeOCE, 4-methoxycatecholestrogen; MRM, multiple reaction monitoring; MS, mass spectrometry; MS/MS, tandem mass spectrometry; 2-OHCE, 2-hydroxycatecholestrogen; 4-OHCE, 4-hydroxycatecholestrogen; UGT, uridine diphospho-glucuronosyltransferase; Vmax, maximal velocity.
Received February 19, 2004.
Accepted June 8, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. Nguyen, J. A. Bonzo, S. Chen, S. Chouinard, M. J. Kelner, G. Hardiman, A. Belanger, and R. H. Tukey Disruption of the Ugt1 Locus in Mice Resembles Human Crigler-Najjar Type I Disease J. Biol. Chem., March 21, 2008; 283(12): 7901 - 7911. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Dassen, C. Punyadeera, R. Kamps, B. Delvoux, A. Van Langendonckt, J. Donnez, B. Husen, H. Thole, G. Dunselman, and P. Groothuis Estrogen metabolizing enzymes in endometrium and endometriosis Hum. Reprod., December 1, 2007; 22(12): 3148 - 3158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Sun, A. K. Sharma, R. W. Dellinger, A. S. Blevins-Primeau, R. M. Balliet, G. Chen, T. Boyiri, S. Amin, and P. Lazarus Glucuronidation of Active Tamoxifen Metabolites by the Human UDP Glucuronosyltransferases Drug Metab. Dispos., November 1, 2007; 35(11): 2006 - 2014. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Yamanaka, M. Nakajima, M. Katoh, and T. Yokoi Glucuronidation of Thyroxine in Human Liver, Jejunum, and Kidney Microsomes Drug Metab. Dispos., September 1, 2007; 35(9): 1642 - 1648. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Swanson, M. Lorentzon, L. Vandenput, F. Labrie, A. Rane, J. Jakobsson, S. Chouinard, A. Belanger, and C. Ohlsson Sex Steroid Levels and Cortical Bone Size in Young Men Are Associated with a Uridine Diphosphate Glucuronosyltransferase 2B7 Polymorphism (H268Y) J. Clin. Endocrinol. Metab., September 1, 2007; 92(9): 3697 - 3704. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Matsumoto, H. Iwano, H. Inoue, N. Iwano, N. Yamashiki, and H. Yokota Metabolic Barrier against Bisphenol A in Rat Uterine Endometrium Toxicol. Sci., September 1, 2007; 99(1): 118 - 125. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Fujiwara, M. Nakajima, H. Yamanaka, A. Nakamura, M. Katoh, S.-i. Ikushiro, T. Sakaki, and T. Yokoi Effects of Coexpression of UGT1A9 on Enzymatic Activities of Human UGT1A Isoforms Drug Metab. Dispos., May 1, 2007; 35(5 |
