| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Molecular and Cellular Pathology, Ninewells Hospital and Medical School, University of Dundee, Dundee, Scotland DD1 9SY; and Department of Internal Medicine III, Erasmus University Medical School, Rotterdam 3015 GE, The Netherlands
Address all correspondence and requests for reprints to: Karen A. B. Findlay, Department of Molecular and Cellular Pathology, Ninewells Hospital and Medical School, University of Dundee, Dundee, Scotland DD1 9SY. E-mail: k.a.b.findlay{at}dundee.ac.uk
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The UGTs represent a family of membrane-bound isoenzymes located in the endoplasmic reticulum of hepatic (1) and extrahepatic tissues (2). These enzymes are encoded by two different gene clusters (3). The first subfamily are the UGT1A isoenzymes, which are encoded by a 350-kb gene complex (4, 5) located on chromosome 2 (6, 7). Transcription of the UGT1A gene using multiple start sites followed by differential splicing generates mRNAs, each of which combines 1 (of 13) variable exon and 4 constant exons. These UGT1A enzymes are often referred to as the bilirubin and phenol UGT enzymes. The second family, the UGT2B subfamily, of genes are located on chromosome 4 (8). The human UGT2B gene consists of 6 exons/5 introns. These UGT2 enzymes are responsible for steroid, bile acid, and xenobiotic glucuronidation.
Increased activity of UGTs in rat liver caused by exposure to xenobiotics such as polychlorobiphenyls can result in thyroid hyperplasia (9). The xenobiotics cause prolonged UGT induction, thereby increasing thyroid hormone glucuronidation and depleting the levels of circulating T4. Thus, the TSH concentration is elevated to release more T4 and T3, which, in turn, initiates thyroid hypertrophy, and eventually prolonged exposure to xenobiotics results in thyroid hyperplasia in rats. Initial work in rats has established that different UGTs are responsible for the glucuronidation of iodothyronines. T4 and rT3 are glucuronidated to a large extent by isoenzymes of the UGT1A subfamily in rats, as shown by a 50% decrease in glucuronide production in livers from Gunn rats, which have a mutation in the bilirubin/phenol UGT1A gene, compared to control rats. T3, however, is glucuronidated in rats predominantly by UGT2B2, as livers from Wistar-LA, WAG, and Fisher rat show a 70% reduction in T3 glucuronidation compared with normal Wistar rats due to a mutation in the UGT2B2 gene (10).
The possibility of xenobiotic-induced thyroid hyperplasia in rats has raised concern that similar problems might be experienced by man. Thus, it is important to study this problem in man using human tissues and cloned expressed cell lines. Earlier studies of cell lines expressing human UGTs suggested that the bilirubin UGT1A1 and the phenol UGT1A9 glucuronidate T4 and rT3 (11). T3 is not glucuronidated by either of these cell lines (12).
The work here used human tissues that have different UGT isoform compositions to further investigate the specificity of thyroid hormone glucuronidation. Normal human liver microsomal samples containing all liver UGTs have been compared to Crigler-Najjar (CN) liver samples. CN patients have a pathological condition such that no functional UGT1A1 is present due to a mutation in the gene sequence. This mutations results in an inability to glucuronidate bilirubin, and patients suffer severe unconjugated hyperbilirubinanemia. All of these samples are also compared to human kidney samples that do not express the UGT1A1 enzyme, but highly express UGT1A9 (13).
| Materials and Methods |
|---|
|
|
|---|
Genomic DNA preparation was carried out from frozen samples as a tissue source using a Nucleon DNA preparation kit (Scotlab Biosciences, Coatbridge, UK). PCR products for individual exons were amplified for automated sequencing using standard intronic/exonic PCR primers (MWG Biotech, Ebersberg, Germany) available in the laboratory on a Biometra Trio Thermoblock and sequenced by automated sequencing using an PE Applied Biosystems 377 DNA sequencer (Perkin-Elmer Corp., Palo Alto, CA). The CN liver sample was a donation from Prof. R. Verbeeck, University of Catholique (Louvain, Belgium).
Preparation of tissue microsomal fractions
Microsomal samples for enzyme assay and Western blotting were prepared from human tissue (samples were obtained from healthy parts of tissue removed with tumors or postmortem tissue at less than 5 h after death and stored at -70 C after rapid freezing in liquid nitrogen) by differential centrifugation as follows. The tissue was homogenized in 5 x tissue weight vol 0.25 mol/L sucrose/0.005 mol/L HEPES, pH 7.4, and then centrifuged for 10 min at 10,000 x g. The supernatant was filtered through glass wool and then centrifuged for 45 min at 100,000 x g. The cytosolic supernatant was removed, and the remaining microsomal pellet was resuspended in sucrose/HEPES at 1 mL/g original tissue. Protein content was determined by the method of Lowry et al. (14) before aliquots were snap-frozen in liquid nitrogen and stored at -70 C.
Western blotting
Western blotting of all samples was carried out essentially as described by Towbin et al. (15). Proteins were separated using 10% bis-acrylamide gels, transferred to nitrocellulose, and detected by RAK antibody [antibody raised against rat kidney UGTs, previously described by Coughtrie et al. (16)]. Dilutions were as follows; RAK primary, 1:1000 for 2 h; and secondary, 1:20,000 for 1 h. Horseradish peroxidase-linked secondary antibodies were used for enhanced chemiluminescence and elucidation of UGT proteins transferred to nitrocellulose.
UGT activity assays
Reverse phase high performance liquid chromatography analysis and kinetic characterization of all samples were carried out essentially as described by Ethell et al. (17) using the probe substrates described in Results. In short, 0.2 mg microsomal protein was incubated in 100 mmol/L Tris/maleate buffer, pH 7.4, containing 5 mmol/L MgCl2, with increasing concentrations of probe substrates. Reactions were prewarmed to 37 C before the addition of 2 mmol/L uridine diphosphate glucuronic acid cofactor (0.1 µCi [14C] uridine diphosphate glucuronic acid/assay) and incubated for 30 min. The only exception to the protocol described was that 0.5 mg protein was used in morphine UGT assays, and incubation was for 60 min at pH 9. Termination was achieved by the addition of 100 µL methanol, and proteins were removed by centrifugation at 1000 x g for 5 min. The resulting supernatant was removed to a high performance liquid chromatography vial, and 150 µL of the volume were injected onto a ThermoSeparation gradient liquid chromatograph with an AS300 autosampler and UV100 detector attached (ThermoSeparation Products, Staffordshire, UK) for elucidation of glucuronides. Radiochemical detection was through a model 9701 radioactivity monitor (Reeve Analytical, Glasgow, Scotland), and data were collected on JCL6000 for Windows software (Jones Chromatography, Hengoed, Wales, UK).
Thyroid hormone assays were carried as described by Visser et al. (10) with minor modifications. Organic extraction assays were performed as described by Van Roy and Heirwegh. (18), Otani et al. (19) and Coughtrie and Sharp (20) were used to determine bilirubin, 1-naphthol, and imipramine UGT activities, respectively.
1-Naphthol extraction assays were carried out on all samples to obtain an optimal detergent/protein ratio (D/P) for maximum microsomal activation using the detergent Lubrol-PX. This involves incubations of 00.5 D/P over the standard assay conditions and provides maximal detergent activation conditions for bilirubin and imipramine assays.
In each extraction assay the substrate concentration was varied to allow a kinetic profile of UGT activity to be determined. Bilirubin and imipramine assays were at optimal D/P over a concentration range of 0200 µmol/L bilirubin, 02 mmol/L imipramine (this is based on a modification of the original concentrations of 122 µmol/L bilirubin and 150 µmol/L imipramine to allow kinetic analysis to be carried out). In contrast to the normal liver samples, the CN and human kidney samples showed no UGT activity toward bilirubin. In all experiments positive control incubations with normal rat liver microsomes indicated optimal assay performance.
| Results |
|---|
|
|
|---|
Individual exons of the human UGT1 were amplified by PCR using
genomic DNA extracted from the livers of two CN patients. PCR products
were sequenced to reveal the same mutation present in both tissue
samples. The mutation was found in the variable region of the gene,
exon 1. Deletion of a TTC base pair sequence (bases 510512 inclusive)
results in an amino acid deletion at position 170 and disruption of a
diphenylalanine repeat, as shown in Fig. 1
. This mutation causes a total loss of
bilirubin UGT activity, but has no effect on the expression of any
other member of the UGT1 family. Therefore, the CN sample can be used
in comparison with normal liver microsomal samples to determine the
contribution of UGT1A1 to thyroid hormone glucuronidation. This
mutation does not lead to loss of the UGT1A1 protein.
|
CN liver, normal human liver, and kidney microsomal samples were
examined by Western blotting using RAK antibody (see Materials
and Methods). UGTs were detected by enhanced chemiluminescence
(shown in Fig. 2
) and alkaline
phosphatase methods. RAK detected only one band in the kidney samples
(Fig. 2
, lanes 14), but detected two bands in the control human liver
and CN liver sample lanes (control lanes 5 and 6, and CN lanes 7 and
8). The single band detected in the kidney corresponded to the lower
band observed in the liver samples, indicating that the lower band is
that of the phenol UGTs, and the upper band is bilirubin UGT (UGT1A1),
which is not present in human kidney. Both bands in the CN sample lanes
appear enhanced compared with those in the control liver sample. The
explanation of this is likely to be an observed induction of the UGTs
in the tissue resulting from standard barbiturate therapy for CN
patients. As the mutation is an amino acid deletion and is located in
the variable exon 1 of the CN sample the protein will be present
although nonfunctional and therefore still detectable under SDS-PAGE
separation and Western blotting conditions.
|
All liver and kidney samples to be assayed with the thyroid
hormones were screened for UGT activities using the following probe
substrates: UGT1A1, bilirubin; UGT1A4, imipramine; UGT1A6, 1-naphthol;
UGT1A9, propofol; UGT2B4, hydrodeoxycholate; UGT2B7, morphine; and
UGT2B15, androstanediol. Incubations of varying concentrations of probe
substrates were used to obtain kinetic data for each sample and
substrate. Duplicate data points for each substrate and tissue sample
were used to obtain a kinetic profile, and maximum velocity
(Vmax) values are shown in Table 1
. Examination of the UGTs in all three
tissue types revealed normal UGT activities toward all probe
substrates.
|
CN liver microsomes, normal human liver microsomes, and normal
human kidney microsomes were assayed for UGT activity toward
T4, T3, and
rT3. rT3 was glucuronidated
faster by all microsomal samples, although substantial levels of
T4 glucuronidation were observed (Table 1
). The
two CN liver samples lacking a functional UGT1A1 enzyme due to an amino
acid deletion (Fig. 1
and Table 1
) have a substantial loss of activity
toward T4 and rT3. Overall,
these data suggest that in human liver, UGT1A1 is the main contributor
to the glucuronidation of T4 and
rT3, with rT3 being more
extensively glucuronidated. Human kidney microsomes do not express the
UGT1A1 isoform and thus have no activity toward bilirubin (Table 1
).
Table 1
also shows that human kidney glucuronidates
rT3 more extensively than
T4, but, again, minimal T3
glucuronidation was observed.
Examination of the data obtained using normal human kidney microsomes or CN liver microsomes, neither of which expresses functional UGT1A1, suggests that UGT1A9 has a more important role in thyroid hormone glucuronidation in the kidney than in the liver. This can be explained by the higher expression of UGT1A9 in the kidney than in the liver.
| Discussion |
|---|
|
|
|---|
This work looks again at the role of UGTs in thyroid hormone glucuronidation using normal and pathological tissues with varying expression of different isoenzymes. The availability of CN liver samples provides us with the opportunity to determine the role of UGT1A1 in substrate glucuronidation, particularly in this case when only one isoform is affected. Comparison of the UGT activity in normal liver, CN liver, and normal kidney microsomes toward iodothyronines and a range of probe substrates can be used to investigate which individual isoform plays the key role in glucuronidation of the thyroid hormones.
We have shown using microsomes from these pathological tissues that
UGT1A1 and UGT1A9 are responsible for the glucuronidation of
T4 and rT3, but not
T3. In all samples tested
rT3 was glucuronidated more extensively than
T4, suggesting a preference for
rT3 as a substrate. This concurs with the earlier
work using rats and cells expressing human UGTs (10, 21) as summarized
in Table 2
.
|
Under normal physiological conditions it appears that glucuronidation is not a significant route by which the thyroid hormones are metabolized. However, when thyroid hormone levels are elevated, glucuronidation may become more important as the UGT enzymes would be recruited in an attempt to return the circulating thyroid levels to normal physiological conditions through elimination of glucuronides as well as hormone recirculation through sulfation. Alternatively, when the activity of the UGTs responsible for thyroid hormone glucuronidation becomes elevated, for instance as a result of xenobiotic exposure, thyroid hormone metabolism may be altered from normal as the glucuronidation becomes a more active pathway. Investigation of the regulation of the glucuronidation pathways should reveal the implication of glucuronidation under altered thyroid states and/or after xenobiotic exposure and aid in the understanding of new therapies.
In summary, we have shown that human UGT1A1 and UGT1A9, responsible for bilirubin and bulky phenol glucuronidation, respectively, also glucuronidate T4 and rT3, but not T3. Whether T3 is also a substrate for an as yet unidentified UGT in humans remains to be elucidated.
| Footnotes |
|---|
Received December 30, 1999.
Revised March 22, 2000.
Accepted April 6, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Y. Kato, S.-i. Ikushiro, Y. Emi, S. Tamaki, H. Suzuki, T. Sakaki, S. Yamada, and M. Degawa Hepatic UDP-Glucuronosyltransferases Responsible for Glucuronidation of Thyroxine in Humans Drug Metab. Dispos., January 1, 2008; 36(1): 51 - 55. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. Tong, H. Li, I. Goljer, O. McConnell, and A. Chandrasekaran In Vitro Glucuronidation of Thyroxine and Triiodothyronine by Liver Microsomes and Recombinant Human UDP-Glucuronosyltransferases Drug Metab. Dispos., December 1, 2007; 35(12): 2203 - 2210. [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] |
||||
![]() |
S. Luci, H. Kluge, F. Hirche, and K. Eder Clofibrate Increases Hepatic Triiodothyronine (T3)- and Thyroxine (T4)-Glucuronosyltransferase Activities and Lowers Plasma T3 and T4 Concentrations in Pigs Drug Metab. Dispos., November 1, 2006; 34(11): 1887 - 1892. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. French, M. R. Wilkinson, W. Yang, L. de Chaisemartin, E. H. Cook, S. Das, M. J. Ratain, W. E. Evans, J. R. Downing, C.-H. Pui, et al. Global gene expression as a function of germline genetic variation Hum. Mol. Genet., June 15, 2005; 14(12): 1621 - 1629. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. M. Zavacki and P. R. Larsen CARs and Drugs: A Risky Combination Endocrinology, March 1, 2005; 146(3): 992 - 994. [Full Text] [PDF] |
||||
![]() |
G. D. Anderson Pharmacogenetics and enzyme induction/inhibition properties of antiepileptic drugs Neurology, November 23, 2004; 63(10_suppl_4): S3 - S8. [Abstract] [Full Text] |
||||
![]() |
J. M. Maglich, J. Watson, P. J. McMillen, B. Goodwin, T. M. Willson, and J. T. Moore The Nuclear Receptor CAR Is a Regulator of Thyroid Hormone Metabolism during Caloric Restriction J. Biol. Chem., May 7, 2004; 279(19): 19832 - 19838. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. T. Ethell, J. Riedel, H. Englert, H. Jantz, R. Oekonomopulos, and B. Burchell GLUCURONIDATION OF HMR1098 IN HUMAN MICROSOMES: EVIDENCE FOR THE INVOLVEMENT OF UGT1A1 IN THE FORMATION OF S-GLUCURONIDES Drug Metab. Dispos., August 1, 2003; 31(8): 1027 - 1034. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Chen, J. L. Staudinger, and C. D. Klaassen NUCLEAR RECEPTOR, PREGNANE X RECEPTOR, IS REQUIRED FOR INDUCTION OF UDP-GLUCURONOSYLTRANSFERASES IN MOUSE LIVER BY PREGNENOLONE-16{alpha}-CARBONITRILE Drug Metab. Dispos., July 1, 2003; 31(7): 908 - 915. [Abstract] [Full Text] [PDF] |
||||
![]() |
O. Barbier, L. Villeneuve, V. Bocher, C. Fontaine, I. P. Torra, C. Duhem, V. Kosykh, J.-C. Fruchart, C. Guillemette, and B. Staels The UDP-glucuronosyltransferase 1A9 Enzyme Is a Peroxisome Proliferator-activated Receptor alpha and gamma Target Gene J. Biol. Chem., April 11, 2003; 278(16): 13975 - 13983. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Xie, M.-F. Yeuh, A. Radominska-Pandya, S. P. S. Saini, Y. Negishi, B. S. Bottroff, G. Y. Cabrera, R. H. Tukey, and R. M. Evans Control of steroid, heme, and carcinogen metabolism by nuclear pregnane X receptor and constitutive androstane receptor PNAS, April 1, 2003; 100(7): 4150 - 4155. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |