help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ebmeier, C. C.
Right arrow Articles by Anderson, R. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ebmeier, C. C.
Right arrow Articles by Anderson, R. J.
The Journal of Clinical Endocrinology & Metabolism Vol. 89, No. 11 5597-5605
Copyright © 2004 by The Endocrine Society

Human Thyroid Phenol Sulfotransferase Enzymes 1A1 and 1A3: Activities in Normal and Diseased Thyroid Glands, and Inhibition by Thyroid Hormones and Phytoestrogens

Christopher C. Ebmeier and Robert J. Anderson

Section of Endocrinology, Veterans Affairs Medical Center, Creighton University Medical Center, Omaha, Nebraska 68105

Address all correspondence and requests for reprints to: Dr. Robert J. Anderson, Veterans Affairs Medical Center, 4101 Woolworth Avenue, Omaha, Nebraska 68105. E-mail: robert.anderson4{at}med.va.gov.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Sulfation by sulfotransferase enzymes (SULTs) is an important pathway for the metabolism of thyroid hormones and phytoestrogens. Intrathyroidal SULTs may contribute to the processing of thyroid hormones for the reutilization of iodide. SULT1A1 and SULT1A3 activities were identified in normal and diseased human thyroid glands. Biochemical properties that included apparent Km values, thermal stabilities, and responses to inhibitors were characterized in a normal human thyroid high speed supernatant pool. Apparent Km values for SULT1A1 and SULT1A3 activities with the model substrates p-nitrophenol and dopamine were 0.58 ± 0.04 and 11.3 ± 1.3 µM, respectively. Activities of SULT1A1 and SULT1A3 determined in individual normal thyroid (n = 35), nodular goiter (n = 26), and autoimmune thyroid disease (n = 25) glands were 0.34 ± 0.06, 0.52 ± 0.09, and 0.82 ± 0.19 U/mg protein for SULT1A1, respectively, and 0.22 ± 0.04, 0.21 ± 0.04, and 0.48 ± 0.11 U/mg protein for SULT1A3, respectively. Both SULT activities in autoimmune thyroid disease glands were significantly higher than those in normal thyroids. Only 3,3'-diiodothyronine (3,3'-T2) and the phytoestrogen daidzein served as substrates for the normal thyroid SULT activities, yet each thyroid hormone and phytoestrogen tested were found to inhibit thyroid SULT1A1 and SULT1A3 activities. The preference of thyroid gland SULT activities for 3,3'-T2 suggests that sulfation may enhance degradation of intrathyroidal 3,3'-T2 for iodide reutilization. Inhibition of these SULT activities by the exogenous phytoestrogens daidzein and genistein, with a potential decrease in iodide reutilization, presents another mechanism through which these compounds may adversely affect human thyroid function.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
METABOLISM OF THYROID hormones through sulfate conjugation (sulfation) provides a unique pathway that accelerates deiodination of the hormones (1). The enhanced deiodination leads to inactivation of thyroid hormones and efficient reutilization of iodide (2). Alterations in either one of these enzymatic steps could affect the availability of thyroid hormone to the receptor or decrease iodide reutilization, and thus adversely affect normal thyroid hormone homeostasis (3). For example, excess type 3 monodeiodinase in hepatic hemangiomas causes profound hypothyroidism (4, 5). In addition, increased sulfated thyroid hormone in hyperthyroidism may serve a protective role by providing substrate for deiodination and subsequent inactivation (6, 7). Recently, mRNA message for one sulfotransferase isozyme (SULT1C2) has been identified in the human thyroid gland, previously reported as SULT1C1, now reassigned as SULT1C2 (8, 9). The possibility that intrathyroidal sulfation of thyroid hormones might contribute to thyroid hormone metabolism has not been addressed. Because of the lack of information about the number or potential role of sulfotransferase activities within the human thyroid, we have investigated human thyroid sulfotransferase activities in this report.

Sulfation affects a broad spectrum of substrates that include simple phenolic drugs (acetaminophen), neurotransmitters such as dopamine, and estrogens, dehydroepiandrosterone (DHEA), phytoestrogens, and thyroid hormones (10, 11). At least 10 members of the human cytosolic SULT superfamily have been described (9, 10, 11). Sulfation contributes to the inactivation of T3 and other iodothyronines by the addition of a sulfuryl moiety to the 4'-hydroxyl group. 3'-Phosphoadenosine-5'-phosphosulfate (PAPS) donates the sulfuryl group in the reaction. T3 sulfate is a major product of human T3 metabolism (6, 12). T3 sulfate is a better substrate than T3 for type I 5'-deiodinase (D1), and this results in enhancement of deiodination (2). Several human SULTs are capable of sulfating thyroid hormones. These include SULT1A1 (TS PST1, P-PST) (13, 14, 15), SULT1A3 (TL PST, M-PST) (13, 15), SULT1B1 (16, 17), SULT1E1 (EST) (18, 19), and SULT2A1 (DHEA ST) (18). Most studies have been performed with transiently expressed SULT enzymes. Endogenous human liver SULT1A1 has been shown to sulfate T3 and other thyroid hormones (13, 14, 15). Detailed characterizations of other human tissue (including thyroid gland) SULT1A1 activities with iodothyronines as the substrates are lacking.

By contrast, deiodinase activities are well characterized (20). Deiodinases are responsible for bioactivation and degradation of thyroid hormones. Type I 5'-deiodinase (D1) has been found in many human tissues, with the highest levels present in the thyroid. The preferred substrates for type I 5'-deiodinase are rT3 and T4, which yield 3,3'-diiodothyronine (3,3'-T2) and T3, respectively, in the reaction. Sulfation of the latter two substrates accelerates and subsequently facilitates their deiodination and reutilization of iodide (20). Animal and human cell culture studies of other tissues (2, 21) suggest an important cooperation between these intrathyroidal enzymes, but this has not been investigated in the human thyroid.

In response to the rising popularity of natural supplements and the intense marketing of them as alternative treatments for the potential prevention of cancer, cardiovascular disease, osteoporosis, and relief from menopausal symptoms, phytoestrogen supplements are readily available. As noted above, some phytoestrogens are metabolized through sulfation (22, 23). In addition, the phytoestrogens daidzein and genistein inhibit the thyroid peroxidase-catalyzed iodination of tyrosine for the synthesis of T4 within the thyroid (24). Individuals may ingest large enough quantities of these phytoestrogens to cause a goiter, as was noted in infants given soy formula (25). Potential effects of genistein and daidzein on intrathyroidal SULT activities, and subsequently thyroid hormone metabolism, have not been described.

We performed these studies to identify several SULT enzyme activities within the human thyroid, to establish radioenzymatic assays for their measurement, and to test the levels of SULT activities in a range of human normal thyroid samples, nodular goiters, and goiters affected by autoimmune thyroid disease. We also investigated the potential effects of inhibition by the exogenous phytoestrogens daidzein and genistein on these SULT activities to determine other mechanisms through which these compounds may adversely affect human thyroid function.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials

Dopamine, pargyline, p-nitrophenol, BSA, 2,6-dichloro-4-nitrophenol (DCNP), sodium chloride, ecteola cellulose (fine mesh), ammonium hydroxide, T3, rT3, T4, daidzein, genistein, trifluoroacetic acid, chlorosulfonic acid, 2 M ammonia in ethanol, and dimethylformamide were obtained from Sigma-Aldrich Corp. (St. Louis, MO). Acetonitrile was obtained from Fisher Scientific (Pittsburgh, PA). 3,3'-Diiodothyronine (3,3'-T2) was a gift from Dr. S.-Y. Wu (Long Beach, CA). Dithiothreitol was purchased from Calbiochem (La Jolla, CA). [35S]PAPS (2.3–3.0 Ci/mmol) was purchased from PerkinElmer LAS (Shelton, CT). TRIzol reagent, Superscript first-strand synthesis system for RT-PCR, and Platinum PCR SuperMix were obtained from Invitrogen Life Technologies, Inc. (Carlsbad, CA).

Thyroid cytosol preparations

Frozen thyroid tissue samples were obtained at surgery and autopsy from the Cooperative Human Tissue Network, which is funded by the NCI, and the National Disease Research Interchange and were placed at –80 C immediately after acquisition. Surgical samples were processed to provide normal adjacent tissue determined by gross inspection to be uninvolved with tumor or nodular processes. Autopsy samples were handled in a similar fashion. Clinical information was limited in the majority of cases to demographic data and histology. Concomitant drugs and treatments were not available. Each sample was homogenized in a buffer that included 5 mM potassium phosphate (pH 7.5), 0.25 M sucrose, 1.5 mM dithiothreitol, and 1.25 mM disodium EDTA at 4 C for 15–30 sec with a tissue homogenizer (Kinematica, Lucerne, Switzerland). Homogenates were centrifuged at 13,000 x g for 10 min at 4 C. The resultant supernatants were transferred and centrifuged again at 100,000 x g for 1 h at 4 C. Aliquots of the high speed supernatant (HSS) were removed for protein determination. The remaining HSS was mixed with an equal volume of 5 mM potassium phosphate, pH 7.5, that contained 5 mg/ml BSA and was stored at –80 C until assay. These studies were approved by the Veterans Affairs Nebraska Western Iowa Health Care System institutional review board and research and development committee.

RNA isolation and RT-PCR

Total RNA was extracted from thyroid tissue using TRIzol reagent as recommended by the manufacturer (Invitrogen Life Technologies, Inc.). First strand cDNA synthesis was performed using Superscript II reverse transcriptase with oligo(deoxythymidine) priming for mRNA only, as recommended by the manufacturer (Invitrogen Life Technologies, Inc.). The first strand cDNA was used as the template for PCR with the primers described in Table 1Go (26, 27). PCR was performed with Platinum PCR Supermix as recommended by the manufacturer (Invitrogen Life Technologies, Inc.) with an Eppendorf thermal cycler. An initial denaturation at 94 C for 2 min was followed by 35 cycles of 94 C for 30 sec, 55 or 65 C for 1 min and 15 sec, and 72 C for 2 min to amplify PCR amplicons. The PCR products were separated by 2% agarose gel electrophoresis and visualized with ethidium bromide and UV light.


View this table:
[in this window]
[in a new window]
  		TABLE 1. RT-PCR primers for SULT genes tested in normal human thyroid

 
Sulfotransferase and protein assays

SULT activities were measured in vitro with 4 µM p-nitrophenol for SULT1A1, 60 µM dopamine for SULT1A3, and 0.4 µM [35S]PAPS as the cosubstrate by the method of Foldes and Meek (28), as modified by Anderson and Liebentritt (29). Assay conditions were optimized with respect to protein concentration, reaction time, and buffer pH. Samples were incubated at 37 C for 30 min. These standard conditions were used for assays with daidzein and genistein and for all inhibitor studies with thyroid hormones. Thyroid hormones were dissolved in alkaline double-distilled water. Daidzein and genistein were dissolved in dimethylsulfoxide for a final concentration of 4.6% (vol/vol) dimethylsulfoxide in double-distilled water. SULT activities toward thyroid hormones were measured in vitro with T4, T3, rT3, and 3,3'-T2 as substrates and 0.4 µM PAPS as the cosubstrate by the method of Young et al. (13), as modified by Li and Anderson (18). Samples were incubated at 37 C for 15 min. Blank samples without substrate capable of accepting a sulfuryl group were used as controls. One unit of enzyme activity represented the formation of 1 nmol sulfated product/h incubation at 37 C. Total protein for each sample HSS was determined by the dye-binding method of Bradford (30) with BSA as a standard.

Thermal stability

Thermal stability was tested by the methods described by Reiter et al. (31) as modified by Anderson et al. (32). Aliquots of pooled normal thyroid HSS samples were heated before assay for 15 min at temperatures of 35–47 C, with 2 C increment increases. Unheated aliquots were kept at 4 C as controls. Activities at each temperature were expressed as a percentage of the unheated activity.

Identification of reaction products by HPLC

Identification of T2 sulfate and daidzein sulfate was performed via HPLC analyses on a Waters system consisting of two Waters 501 pumps, an automated gradient controller and a Waters model 440 absorbance detector at 280 nm (Waters Corp., Milford, MA). For separation of T2 and daidzein derivatives, a 3.9 x 20-mm Waters Symmetry Shield rpC18 column with 3.5 µm resin was operated at room temperature. Solvents were 0.1% (vol/vol) trifluoroacetic acid in water (A) and 0.1% (vol/vol) trifluoroacetic acid in acetonitrile (B). The gradient profile was: 0–1 min, 15% B; 1–11 min, 15–60% B; and 11–12 min, 60–15% B. Blank and active samples were eluted at 2 ml/min and collected in 0.5-ml fractions before counting. Daidzein and T2 sulfate standards for HPLC were prepared with chlorosulfonic acid and dimethylformamide (33).

Data analysis

Apparent Km values were calculated by the Wilkinson method and the direct linear plot method with Enzpack for Windows version 1.4 (Elsevier-Biosoft, Cambridge, UK). Kinetic data were presented as double reciprocal plots. The median inactivation temperature (T50) and IC50 with DCNP and NaCl were determined using a curve-fitting program. Statistical significance was analyzed by the Mann-Whitney nonparametric t test and by nonparametric one-way ANOVA with the Kruskal-Wallis test. Dunn’s multiple comparison posttest was used (GraphPad, Inc., San Diego, CA). Significance was defined as P < 0.05.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The experiments were performed in three parts. First, the RT-PCR results were reviewed, human thyroid SULT1A1 and SULT1A3 assays were developed, and biochemical properties of the enzymes were established. Second, normal and diseased human thyroid samples were assayed to establish the means and ranges of SULT activities. Third, thyroid hormones and the phytoestrogens daidzein and genistein were evaluated as substrates and inhibitors of intrathyroidal SULT enzymes.

RT-PCR detection of SULTs in thyroid tissue

Amplification products of the expected size were obtained from the RT-PCR performed on the thyroid tissue mRNA with the listed SULT primers (Fig. 1Go). The results indicated the presence of SULT1A1, SULT1A3, SULT1B1, and SULT1C2 mRNA in the human thyroid tissue. Because specific assays for SULT1A1 and SULT1A3 activities would be most feasible with low concentrations of the model substrates p-nitrophenol and dopamine, respectively, we chose to establish assays for thyroid cytosol SULT1A1 and SULT1A3 activities in this study.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 1. RT-PCR of SULTs in normal human thyroid. Using 1 µg mRNA from the pool of normal thyroids (n = 3) as the template, RT-PCR was performed with SULT primers as described in Materials and Methods. Lane 1, 100-bp standard. Each labeled lane indicates RT-PCR with the respective SULT primers. PCR products were visualized with ethidium bromide during 2% agarose gel electrophoresis.

 
Biochemical properties of thyroid SULTs

Effects of varying concentrations of prototypical substrates. The effects of varying concentrations of p-nitrophenol and dopamine on the levels of SULT1A1 and SULT1A3 activities in a normal thyroid HSS pool (n = 3) were tested (Fig. 2Go). Apparent Km values with the substrates p-nitrophenol and dopamine are listed in Table 2Go. Varying concentrations of [35S]PAPS (0.15–4 µM) were tested with stable concentrations of each substrate to estimate the apparent Km values listed in Table 2Go. These results were similar to values obtained with cytosol preparations of human liver (34), pituitary (35), small intestine (36), and platelets (37).



View larger version (17K):
[in this window]
[in a new window]
  		FIG. 2. Effects of substrate concentration on the activities of SULT1A1 (A) and SULT1A3 (B) in a human thyroid HSS pool (5.8 µg protein/assay tube). The insets show the corresponding double-reciprocal plots. Results are expressed as the mean ± SEM of three experiments with triplicate determinations in each experiment. Reaction substrates varied from 0.03–8 µM p-nitrophenol (A) and 0.23–120 µM dopamine (B), respectively.

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Biochemical properties of human thyroid SULT

 
Effect of temperature. Thermal stability is an important indicator of differences in protein structure and is one of the well defined properties that distinguish between SULT1A1 and SULT1A3. Enzyme thermal stabilities were tested in an HSS pool of normal thyroid glands (n = 3). Temperatures at which 50% of the unheated control activity remained (T50 values) are listed in Table 2Go. Again, the values were similar to those obtained with other human tissue preparations (34, 35, 36, 37, 38).

Effects of inhibition by DCNP and NaCl

The responses of SULT1A1 and SULT1A3 activities to DCNP and NaCl provide another distinction between the enzymes. The thyroid HSS pool was tested with varying concentrations of DCNP and NaCl. Control samples contained no inhibitor. The IC50 values are listed in Table 2Go. The greater sensitivity of SULT1A1 to inhibition by both DCNP and NaCl compared with SULT1A3 was typical of the human SULT enzymes reported previously for other tissues (38).

SULT activities in normal and diseased thyroids

Once uniform assay conditions and biochemical properties were established for the human thyroid sulfotransferases, SULT1A1 and SULT1A3 activities were measured in individual normal and diseased thyroid samples to determine mean levels and ranges of activities. The mean normal thyroid, nodular goiter and autoimmune thyroid disease gland (AITD) sulfotransferase enzyme activities are presented in Table 3Go. The activities of a single normal sample for SULT1A1 and SULT1A3 were 17- and 25-fold greater, respectively, than the mean normal activities. Calculations and statistical analyses were performed with and without this high activity normal thyroid outlier to evaluate how the single outlier sample affected the significance of the comparisons. The normal autopsy tissue as a group was lower in both activities than were the normal surgery samples. The normal autopsy group was larger (n = 23 without the outlier) than the surgery group (n = 12), and few lower activities were found in the surgery samples. Because it is not clear whether the normal adjacent tissue obtained at surgery is completely unaffected by the disease process, because the surgical samples are underrepresented by low activities, and the data reported below did not support a significant correlation of activity level with length of time from death to autopsy, we have chosen to include all normal samples in the statistical analyses (Table 3Go and Fig. 3Go). There were no significant differences between mean SULT1A1 activities in normal thyroid and nodular goiter, with and without the high activity outlier, or between SULT1A3 activities in the same tissues. There were significant differences between mean normal thyroid activities (n = 35) and the combined AITD activities (n = 25) when the high activity normal thyroid outlier was excluded (SULT1A1, P = 0.02; SULT1A3, P = 0.01). With the outlier included, the significant differences were the same in both tissue groups (Table 3Go).


View this table:
[in this window]
[in a new window]
  		TABLE 3. SULT1A1 and SULT1A3 enzyme activities measured with p-nitrophenol and dopamine, respectively, in normal and diseased thyroid tissue samples

 


View larger version (19K):
[in this window]
[in a new window]
  		FIG. 3. SULT1A1 and SULT1A3 activities (mean ± SEM) in samples of normal thyroid glands (without the high activity outlier), nodular goiters, Graves’ disease, and Hashimoto’s thyroiditis. These groups were used with and without the high activity outlier for comparison by nonparametric ANOVA. Significant differences between tissue SULT activities were detected for both SULT1A1 and SULT1A3 without the outlier included. The significance was lost only for SULT1A1 with inclusion of the outlier sample.

 
After subclassification of the AITD group into Graves’ disease (n = 10) and Hashimoto’s thyroiditis (n = 15), comparison of all groups that included the normal thyroid and nodular goiter by nonparametric ANOVA indicated significant differences between tissues for SULT1A1 activity (P = 0.04) and for SULT1A3 activity (P = 0.03) with the high activity normal thyroid outlier excluded (Fig. 3Go). With the outlier included, there was no longer a significant difference in SULT1A1 activity (P = 0.06), but the significance for SULT1A3 remained (P = 0.04). Dunn’s multiple comparison posttest indicated that the only significant comparison was between the SULT1A1 activities in normal thyroid and Graves’ disease samples. The posttest did not identify the elevated SULT1A3 activity in Hashimoto’s thyroiditis glands as significantly different, although SULT1A3 in this group is obviously increased. It remains to be determined whether this significant difference in SULT1A1 activity in glands with Graves’ disease was a causative factor or whether it was the result of possible induction of the enzyme activity by high levels of thyroid hormone (7).

Normal thyroid SULT1A1 activity was 2- to 3-fold higher than activity reported for the skin and pituitary enzymes (35, 38), but was 78- to 105-fold lower than SULT1A1 activity in human small intestine and liver, respectively (34, 36). Normal thyroid SULT1A3 activity was 5.5-fold higher than pituitary activity, but was 2.7-, 375-, and 7.2-fold lower than skin, small intestine, and liver SULT1A3 activities, respectively (34, 35, 36, 38).

Correlation of SULT activities and collection time of normal thyroids obtained at autopsy

SULT activities in tissue samples obtained at autopsy may be altered in part due to varying temperature conditions from the time of death to actual autopsy and tissue storage (35). Examination of the potential relationship of normal thyroid SULT activities with the length of time before autopsy revealed no significant correlations between SULT activities and sample collection time for SULT1A1 (r = 0.2; n = 19; P = 0.4) and SULT1A3 (r = 0.1; n = 19; P = 0.6; Fig. 4Go). It was interesting that the high activity outlier thyroid sample was a normal autopsy sample collected 12 h after death. The data provided evidence that the enzyme activities did not decline significantly postmortem before sample collection.



View larger version (11K):
[in this window]
[in a new window]
  		FIG. 4. Correlations of SULT1A1 (A) and SULT1A3 (B) activities with collection time in hours for normal thyroids obtained after autopsy. The mean collection time was 11.1 ± 1.1 h (mean ± SEM). The pattern supports the conclusion that minimal loss of SULT activity occurred in autopsy tissue.

 
Correlation of thyroid SULT activities

To test whether the human thyroid SULT1A1 and SULT1A3 enzymes were regulated in a similar fashion, possible correlations of the SULT1A1 and SULT1A3 activities were examined in each thyroid tissue group. There were significant positive correlations of SULT1A1 and SULT1A3 activities in the normal thyroid, with the high activity outlier (r = 0.7; n = 36; P < 0.0001) and without the outlier (r = 0.7; n = 35; P < 0.0001), in nodular goiter (r = 0.5; n = 26; P = 0.005) and in AITD (r = 0.7; n = 25; P = 0.0002; Fig. 5Go). Positive correlations between tissue SULT1A1 and SULT1A3 activities also were found with human skin and human small intestinal mucosa (36, 38). These findings may reflect the equal importance of the two enzymes for local tissue substrate metabolism. By contrast, human platelet, pituitary; and liver SULT1A1 and SULT1A3 activities were not correlated (34, 35, 37), findings that may indicate separate tissue regulation of these two enzymes for specific local metabolism of their respective substrates.



View larger version (12K):
[in this window]
[in a new window]
  		FIG. 5. Correlations of SULT1A1 and SULT1A3 activities in normal thyroid (A), nodular goiter (B), and AITD (C) groups. Each point represents the mean of three determinations. The high activity outlier was excluded from the normal thyroid group (A), but the significance of the correlation did not change.

 
Finally, there was no significant correlation between normal thyroid SULT activities and patient age when men and women were compared together or separately, with or without the high activity normal outlier. Nor were there significant correlations between either SULT activities in nodular goiter and AITD samples and patient age or gender.

Thyroid SULT activities with endogenous and exogenous compounds

SULT activities with thyroid hormones as substrates. The normal thyroid pool was assayed under standard conditions used in previous studies with thyroid hormones as substrates (39). Adequate activity was detectable only with 3,3'-T2. The apparent Km for thyroid SULT activity with 3,3'-T2 was 4.0 ± 0.2 µM by the Wilkinson method and 13.1 ± 1.8 µM by the direct linear method (mean ± SEM for three experiments with three determinations for each experiment; Fig. 6Go). The results indicated that 3,3'-T2 was the preferred substrate for thyroid SULT activity and raised the possibility that endogenous thyroid hormones interfered with the SULT assays when T4, T3, and rT3 were tested as substrates.



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 6. Effects of substrate concentration on the sulfation of 3,3'-T2 by SULT activities in a human thyroid HSS pool (57.8 µg protein/assay tube). The concentration of 3,3'-T2 varied from 0.5–50 µM. The inset shows the corresponding double-reciprocal plot. Results are expressed as the mean ± SEM of three experiments, with triplicate determinations in each experiment.

 
Because the normal human thyroid HSS pool contains at least SULT1A1 and SULT1A3 activities, it was not possible to determine precisely which SULT used 3,3'-T2 as a substrate. However, the Km for bacterial expressed SULT1A1 activity was 260-fold more specific for 3,3'-T2 than was a Km for the bacterially expressed SULT1A3 activity (15). This suggested that SULT1A1 may be the predominant SULT activity for the sulfation of 3,3'-T2 in the human thyroid. Because sulfation of 3,3'-T2 enhances deiodination, the human thyroid SULT1A1 activity may be of crucial importance for the reutilization of iodide and the continued production of active thyroid hormone.

SULT activities with phytoestrogens as substrates. To evaluate potential SULT activity toward the phytoestrogens, the normal thyroid pool was assayed with daidzein and genistein (1 µM) as substrates in the standard SULT assay. Measurable activity was detectable only with daidzein. Under the standard conditions, the apparent Km for thyroid SULT activity with daidzein was 0.62 ± 0.05 µM by the Wilkinson method and 0.68 ± 0.11 µM by the direct linear method (mean ± SEM for three experiments, with three determinations for each experiment; Fig. 7Go). Because this phytoestrogen serves as a substrate for thyroid SULT activities, it might also inhibit SULT activities through competition with thyroid hormone substrates.



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 7. Effects of substrate concentration on the sulfation of daidzein by SULT activities in a human thyroid HSS pool (60.4 µg protein/assay tube). The concentration of daidzein varied from 0.1–5.0 µM. The inset shows the double-reciprocal plot. Results are expressed as the mean ± SEM of three experiments, with triplicate determinations in each experiment.

 
Inhibition of thyroid SULT activities

To test thyroid hormones and phytoestrogens as potential inhibitors, the normal thyroid pool was assayed with the prototypical substrates p-nitrophenol (0.25–4 µM) for SULT1A1, and dopamine (3.75–60 µM) for SULT1A3 as controls, and with the addition of 10–200 µM T4, T3, and rT3; 10–100 µM 3,3'-T2 (Fig. 8Go); and 1–10 µM daidzein and genistein (Table 4Go). The concentrations of thyroid hormones and phytoestrogens used in the inhibition experiments were chosen based on previous reports of concentrations in other tissues that led to measurable activities (13, 14, 18, 39). Ranges of concentrations were then tested with the thyroid tissue cytosol, and those concentrations that provided measurable activities were chosen to avoid amounts that might contribute to low activity due to substrate inhibition. The effects of the addition of thyroid hormones as inhibitors of SULT1A1 (p-nitrophenol) and SULT1A3 (dopamine) are shown in Fig. 8Go. The type of inhibition was uncompetitive or mixed type for all thyroid hormones depending on the concentration of the inhibitor (Enzpack software). The results support the conclusion that the presence of endogenous thyroid hormones in the thyroid preparation inhibited the enzyme activities when measurements were attempted with thyroid hormones as substrates.



View larger version (15K):
[in this window]
[in a new window]
  		FIG. 8. Inhibition of human thyroid pool SULT1A1 (4 µM p-nitrophenol; A) and SULT1A3 (60 µM dopamine; B) activities by 0–200 µM T4, T3, rT3, and 0–100 µM 3,3'-T2 concentrations, respectively. The mean absolute activity in counts per minute for the SULT1A1 control sample was 2973, and the control activity for SULT1A3 was 1249 cpm. Results are expressed as the mean ± SEM of six determinations (two experiments with triplicate determinations in each experiment).

 

View this table:
[in this window]
[in a new window]
  		TABLE 4. Inhibition of human thyroid SULT activities by phytoestrogens

 
Daidzein and genistein were found to inhibit both human thyroid SULT1A1 and SULT1A3 activities in a concentration-dependent fashion (Table 4Go). Human thyroid SULT1A1 was susceptible to a somewhat greater inhibition by both daidzein and genistein. If SULT1A1 activity contributes to the metabolism of 3,3'-T2 for efficient recycling of iodide, these phytoestrogens potentially could interfere with the reutilization process. Concentrations of daidzein and genistein up to approximately 0.5–1 µM have been reported in human serum after the consumption of soy supplements containing daidzein and genistein (40). Although intrathyroidal concentrations with a diet supplemented with large doses of phytoestrogens are unknown, the possibility of intrathyroidal SULT inhibition by these compounds exists. Inhibition of SULT activities and the known inhibition of thyroid peroxidase activities by daidzein and genistein may combine for a more potent goitrogenic effect (24).

Identification of reaction products by HPLC

Reverse phase HPLC was used to identify products of the thyroid sulfotransferase reactions using T2 and daidzein as substrates. Eighty-seven percent of the net counts coeluted with the T2 sulfate standard, and 84% of the net counts coeluted with the standard single-sulfated daidzein derivative. The results supported the conclusion that the products of the reactions of thyroid sulfotransferases with T2 and daidzein were T2 sulfate and daidzein sulfate, respectively.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Sulfation of endogenous and exogenous compounds and hormones is a widespread process that can affect receptor availability and biological activity of the hormones (41). The expanding number of human SULT enzymes, many with specific tissue localizations and substrate specificities, emphasizes the importance of sulfation (9, 10). SULT1A1 and SULT1A3, the most thoroughly studied human SULTs, are ubiquitous in human tissue and are unique in their abilities to catalyze the sulfoconjugation of a broad range of substrates that includes thyroid hormones (13, 14, 15, 39). Both catalyze the sulfation of iodothyronines, but SULT1A1 displays the greatest specificity for these hormones (15). The availability of cDNA-expressed SULT enzymes has allowed more detailed characterization of the activities toward thyroid hormones (15, 16, 17, 18, 19). Our characterization of SULT1A1 and SULT1A3 activities within the human thyroid expands the possibilities of enzymatic regulation of thyroid homeostasis. For example, intrathyroidal sulfation may be an essential step in the metabolism of excess thyroid hormone that is not secreted, but degraded for reutilization of iodide (3). Sulfation of T3 and T2 leads to more rapid deiodination by D1 and subsequent degradation (2). Alteration of intrathyroidal SULT activities may affect the efficiency of the proposed enhanced iodide reutilization. That is, a decrease in SULT activity might inhibit iodide reutilization, whereas an increase in SULT activity might enhance the breakdown of excess thyroid hormone within a hyperthyroid gland. Intrathyroidal SULT activities may metabolize exogenous compounds such as phytoestrogens to prevent their adverse effects on thyroid hormone homeostasis. A final possibility is that specific thyroid diseases may be associated with changes in the thyroid SULT1A1 and SULT1A3 activities. Although our studies were not designed to answer these questions, the results do provide intriguing insights into possible physiological roles of the SULT activities.

This study documented the presence of SULT1A1 and SULT1A3 activities within the human thyroid gland and confirmed the presence of two of the intrathyroidal SULT activities predicted by RT-PCR. Biochemical properties were essentially the same as properties reported for other human tissues (34, 35, 36), and they were similar to values obtained with recombinant SULT1A1 and SULT1A3 enzyme activities (42, 43). Thyroid SULT1A1 and SULT1A3 activities were inhibited in a concentration-dependent fashion by several iodothyronines. The type of inhibition was variable, no doubt because the cytosol preparation contained competing SULT enzyme activities. However, SULT activity was readily measured with 3,3'-T2 as the substrate (18). This indicated 3,3'-T2 may be the preferred intrathyroidal SULT substrate for facilitation of subsequent deiodination and iodide reutilization. The thyroid SULT activity represented the combination of at least SULT1A1 and SULT1A3, with the greater contribution by SULT1A1 activity. In addition, other SULT activities, if present, would be expected to contribute because 3,3'-T2 is known to be sulfoconjugated by at least five of the 10 human enzymes (SULT1A1, SULT1A3, SULT1B1, SULT1E1, and SULT2A1) (13, 14, 15, 16, 17, 18, 19).

Examination of the thyroid pool preparation showed no measurable activity for SULT1E1 and SULT2A1 when tested with the prototypical substrates estrone (SULT1E1) and DHEA (SULT2A1), respectively. One report indicated the presence of SULT1E1 mRNA by RT-PCR in one sample from an elderly Japanese male (44). We did confirm the presence of SULT1C2 and noted SULT1B1 mRNA transcripts by RT-PCR. The results for SULT1C2 were consistent with previous reports of intrathyroidal human SULT1C2 (8). Confirmation of activities of both enzymes within the thyroid will require careful development of assays for each enzyme with the human thyroid cytosol preparation.

Mean levels of SULT1A1 and SULT1A3 activities for all normal thyroids and nodular goiters were not significantly different even when the markedly elevated activity outlier sample was included. It was not possible to discern whether SULT activities within nodules were different compared with normal adjacent tissue activities. We are currently investigating such potential differences. Of interest was the significantly higher level of SULT1A1 activity in the entire AITD group and the Graves’ disease group in particular compared with the normal thyroids. One explanation for this difference would be the induction of intrathyroidal SULT1A1, perhaps by excessive levels of thyroid hormone. Such an increase in SULT1A1 activity might protect the gland by enhancing deiodination and inactivation of the excess hormone. High levels of T3 sulfate have been reported in hyperthyroid patient serum, but no measurement of tissue SULT levels in hyperthyroid patients has been described (6).

Phytoestrogen contributions to altered thyroid function are known. The phytoestrogens daidzein and genistein are substrates for thyroid peroxidase, and genistein can inhibit T4 synthesis (24). Because sulfation of these phytoestrogens occurs, the intrathyroidal SULT activities may serve as a protective mechanism. Greater sulfation of daidzein may decrease the availability of the compound for inhibition of thyroid peroxidase (24). Our pool of normal thyroid SULT activities catalyzed the sulfation of daidzein with an apparent Km value within the range reported for plasma daidzein after oral ingestion (24). In our study both human thyroid SULT1A1 and SULT1A3 activities were inhibited by daidzein and genistein in a dose-dependent manner. A potential consequence of higher plasma levels of these phytoestrogens might be inhibition of intrathyroidal SULT activities and a subsequent decline in deiodination and iodide reutilization for an enhanced goitrogenic effect. In a recent study evaluating phytoestrogen supplements for the treatment of postmenopausal hot flashes, subjects consumed 50–80 mg isoflavones, which included daidzein and genistein, daily for 12 wk (45). Ingestion of these quantities of phytoestrogens has led to plasma levels associated with thyroid peroxidase inhibition and to levels in the range required to serve as substrates for thyroid SULT activities. Potential dangers to the thyroid of excessive, concentrated soy and phytoestrogen preparations may involve adverse effects on both sulfotransferases and thyroid peroxidase activities.

In conclusion, these studies provide the basis for examining SULT1A1 and SULT1A3 activities in normal and diseased thyroids. The identification, biochemical characterization, and description of levels of SULT activities are essential initial steps toward explaining the role of intrathyroidal sulfation in thyroid hormone homeostasis. More detailed characterization of SULT1A1 and SULT1A3 activities in autopsy and surgical samples and in different types of thyroid nodules and thyroid cancers, and investigation of other thyroid SULTs will be performed. Finally, study of the potential deleterious effects of daidzein, genistein, and multiple other phytoestrogens in over-the-counter preparations on intrathyroidal sulfation will enhance the safety information with respect to these alternative medicines.


   Acknowledgments
 
We thank Dr. Dahn Clemens for his helpful support during these studies.


   Footnotes
 
This work was supported by the Veterans Affairs Medical Research Service.

Abbreviations: AITD, Autoimmune thyroid disease; D1, type I 5'-deiodinase; DCNP, 2,6-dichloro-4-nitrophenol; DHEA, dehydroepiandrosterone; HSS, high speed supernatant; PAPS, 3'-phosphoadenosine-5'-phosphosulfate; SULT, sulfotransferase enzyme; 3,3'-T2, 3,3'-diiodothyronine; T50, median inactivation temperature.

Received November 10, 2003.

Accepted July 27, 2004.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Mol JA, Visser TJ 1985 Rapid and selective inner ring deiodination of thyroxine sulfate by rat liver deiodinase. Endocrinology 117:8–12[Abstract/Free Full Text]
  2. Visser TJ, van Buuren JCJ, Rutgers M, Rooda SJE, de Herder WW 1990 The role of sulfation in thyroid hormone metabolism. Trends Endocrinol Metab 1:211–218
  3. Li X, Clemens DL, Anderson RJ 2000 Sulfation of iodothyronines by human sulfotransferase 1C1 (SULT1C1). Biochem Pharmacol 60:1713–1716[CrossRef][Medline]
  4. Huang SA, Tu HM, Harney JW, Venihaki M, Butte AJ, Kozakewich HPW, Fishman SJ, Larsen PR 2000 Severe hypothyroidism caused by type 3 iodothyronine deiodinase in infantile hemangiomas. N Engl J Med 343:185–189[Free Full Text]
  5. Huang SA, Fish SA, Dorfman DM, Salvatore D, Kozakewich HPW, Mandel SJ, Larsen PR 2002 A 21-year-old woman with consumptive hypothyroidism due to a vascular tumor expressing type 3 iodothyronine deiodinase. J Clin Endocrinol Metab 87:4457–4461[Abstract/Free Full Text]
  6. Chopra IJ, Wu S-Y, Chua Teco GN, Santini F 1992 A radioimmunoassay for measurement of 3,5,3'-triiodothyronine sulfate: studies in thyroidal and nonthyroidal diseases, pregnancy, and neonatal life. J Clin Endocrinol Metab 75:189–194[Abstract]
  7. Huang W, Kuo S, Chen W, Fuh M, Wu S 1996 Increased urinary excretion of sulfated 3,3',5-triiodothyronine in patients with nodular goiters receiving suppressive thyroxine therapy. Thyroid 6:91–96[Medline]
  8. Her C, Kaur PG, Athwal RS, Weinshilboum RM 1997 Human sulfotransferase SULT1C1: cDNA cloning, tissue-specific expression, and chromosomal localization. Genomics 41:467–470[CrossRef][Medline]
  9. Coughtrie M 2002 Sulfation through the looking glass: recent advances in sulfotransferase research for the curious. Pharmacogenomics J 2:297–308[CrossRef][Medline]
  10. Weinshilboum R, Otterness D, Aksoy I, Wood T, Her C, Raftogianis R 1997 Sulfotransferase molecular biology: cDNAs and genes. FASEB J 11:3–14[Abstract]
  11. Glatt H 2002 Sulphotransferases. In: Ioannides C, ed. Enzyme systems that metabolize drugs and other xenobiotics. New York: Wiley & Sons; 353–439
  12. LoPresti JS, Nicoloff JT 1994 3,5,3'-Triiodothyronine (T3) sulfate: a major metabolite in T3 metabolism in man. J Clin Endocrinol Metab 78:688–692[Abstract]
  13. Young WF, Gorman CA, Weinshilboum RM 1988 Triiodothyronine: a substrate for the thermostable and thermolabile forms of human phenol sulfotransferase. Endocrinology 122:1816–1824[Abstract/Free Full Text]
  14. Anderson R, Babbit L, Liebentritt D 1995 Human liver triiodothyronine sulfotransferase: copurification with phenol sulfotransferases. Thyroid 5:61–66[Medline]
  15. Kester MHA, Kaptein E, Roest TJ, van Dijk CH, Tibboel D, Meinl W, Glatt H, Coughtrie MWH, Visser TJ 1999 Characterization of human iodothyronine sulfotransferases. J Clin Endocrinol Metab 84:1357–1364[Abstract/Free Full Text]
  16. Fujita K, Nagata K, Ozawa S, Sasano H, Yamazoe Y 1997 Molecular cloning and characterization of rat ST1B1 and human ST1B2 cDNAs, encoding thyroid hormone sulfotransferases. J Biochem (Tokyo) 122:1052–1061[Abstract/Free Full Text]
  17. Wang J, Falany JL, Falany C 1998 Expression and characterization of a novel thyroid hormone-sulfation form of cytosolic sulfotransferase from human liver. Mol Pharmacol 53:274–282[Abstract/Free Full Text]
  18. Li X-Y, Anderson RJ 1999 Sulfation of iodothyronines by recombinant human liver steroid sulfotransferases. Biochem Biophys Res Commun 263:632–639[CrossRef][Medline]
  19. Kester MHA, van Dijk CH, Tibboel D, Hood AM, Rose NJM, Meinl W, Pabel U, Glatt H, Falany CN, Coughtrie MWH, Visser TJ 1999 Sulfation of thyroid hormone by estrogen sulfotransferase. J Clin Endocrinol Metab 84:2577–2580[Abstract/Free Full Text]
  20. Kohrle J 1999 Local activation and inactivation of thyroid hormones: the deiodinase family. Mol Cell Endocrinol 151:103–119[CrossRef][Medline]
  21. Van Stralen PGJ, Van Der Hoek HJ, Docter R, De Jong M, Krenning EP, Lim CF, Hennemann G 1993 Reduced T3 deiodination by the human hepatoblastoma cell line HepG2 caused by deficient T3 sulfation. Biochim Biophys Acta 1157:114–118[Medline]
  22. Eaton EA, Walle UK, Lewis AJ, Hudson T, Wilson AA, Walle T 1996 Flavonoids, potent inhibitors of the human P-form phenolsulfotransferase. Drug Metab Dispos 24:232–238[Abstract]
  23. Falany CN 1997 Enzymology of human cytosolic sulfotransferases. FASEB J 11:206–216[Abstract]
  24. Divi RL, Chang HC, Doerge DR 1997 Anti-thyroid isoflavones from soybean isolation, characterization and mechanisms of action. Biochem Pharmacol 54:1087–1096[CrossRef][Medline]
  25. Shepard T, Pyne G, Kirschvink J, McLean C 1960 Soybean goiter. N Engl J Med 262:1099–1103
  26. Dooley TP, Haldeman-Cahill R, Joiner J, Wilborn TW 2000 Expression profiling of human sulfotransferase and sulfatase gene superfamilies in epithelial tissues and cultured cells. Biochem Biophys Res Commun 277:236–245[CrossRef][Medline]
  27. Dubin RL, Hall CM, Pileri CL, Kudlacek PE, Li X-Y, Yee JA, Johnson ML, Anderson RJ 2001 Thermostable (SULT1A1) and thermolabile (SULT1A3) phenol sulfotransferases in human osteosarcoma and osteoblast cells. Bone 28:617–624[Medline]
  28. Foldes A, Meek JL 1973 Rat brain phenolsulfotransferase: partial purification and some properties. Biochim Biophys Acta 327:365–374[Medline]
  29. Anderson RJ, Liebentritt DK 1990 Human platelet thermostable phenol sulfotransferase: assay of frozen samples and correlation between frozen and fresh activities. Clin Chim Acta 189:221–230[CrossRef][Medline]
  30. Bradford MM 1976 A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254[CrossRef][Medline]
  31. Reiter C, Mwaluko G, Dunnette J, Loon JV, Weinshilboum R 1983 Thermolabile and thermostable human platelet phenol sulfotransferase. Naunyn Schmiedebergs Arch Pharmacol 324:140–147[CrossRef][Medline]
  32. Anderson RJ, Jackson BL, Liebentritt DK 1988 Human platelet thermostable phenol sulfotransferase from black and whites: biochemical properties and variations in thermal stability. J Lab Clin Med 112:773–783[Medline]
  33. Mol JA, Visser TJ 1985 Synthesis and some properties of sulfate esters and sulfamates of iodothyronines. Endocrinology 117:1–7[Abstract/Free Full Text]
  34. Campbell NRC, Van Loon JA, Weinshilboum RM 1987 Human liver phenol sulfotransferase: assay conditions, biochemical properties and partial purification of isozymes of the thermostable form. Biochem Pharmacol 36:1435–1446[CrossRef][Medline]
  35. Anderson RJ, Yoon JK, Sinsheimer EG, Jackson BL 1986 Human pituitary phenol sulfotransferase: biochemical properties and activities of the thermostable and thermolabile forms. Neuroendocrinology 44:117–124[Medline]
  36. Sundaram RS, Szumlanski C, Otterness D, Van Loon JA, Weinshilboum RM 1989 Human intestinal phenol sulfotransferase: assay conditions, activity levels and partial purification of the thermolabile form. Drug Metab Dispos 17:255–264[Abstract]
  37. Anderson RJ, Weinshilboum RM, Phillips SF, Broughton DD 1981 Human platelet phenol sulfotransferase: assay procedure, substrate and tissue correlations. Clin Chim Acta 110:157–167[CrossRef][Medline]
  38. Kudlacek PE, Anderson RJ, Liebentritt DK, Johnson GA, Huerter CJ 1995 Human skin and platelet minoxidil sulfotransferase activities: biochemical properties, correlations and contribution of thermolabile phenol sulfotransferase. J Pharmacol Exp Ther 273:582–590[Abstract/Free Full Text]
  39. Li X, Clemens DL, Cole JR, Anderson RJ 2001 Characterization of human liver thermostable phenol sulfotransferase (SULT1A1) allozymes with 3,3',5-triiodothyronine as the substrate. J Endocrinol 171:525–532[Abstract]
  40. Doerge DR, Chang HC, Churchwell MI, Holder CL 1999 Analysis of soy isoflavone conjugation in vitro and in human blood using liquid chromatography-mass spectrometry. Drug Metab Dispos 28:298–307
  41. Strott CA 2002 Sulfonation and molecular action. Endocr Rev 23:703–732[Abstract/Free Full Text]
  42. Veronese ME, Burgess W, Zhu X, McManus ME 1994 Functional characterization of two human sulphotransferase cDNAs that encode monoamine- and phenol-sulphating forms of phenol sulphotransferase: substrate kinetics, thermal-stability and inhibitor-sensitivity studies. Biochem J 302:497–502
  43. Wood T, Aksoy I, Aksoy S, Weinshilboum R 1994 Human liver TL PST: cDNA cloning, expression and characterization. Biochem Biophys Res Commun 198:1119–1127[CrossRef][Medline]
  44. Miki Y, Nakata T, Suzuki T, Darnel AD, Moriya T, Kaneko C, Hidaka K, Shiotsu Y, Kusaka H, Sasano H 2002 Systemic distribution of steroid sulfatase and estrogen sulfotransferase in human adult and fetal tissues. J Clin Endocrinol Metab 87:5760–5768[Abstract/Free Full Text]
  45. Tice J, Ettinger B, Ensrud K, Wallace R, Blackwell T, Cummings SR 2003 Phytoestrogen supplements for the treatment of hot flashes: the isoflavone clover extract (ICE) study: a randomized controlled trial. JAMA 290:207–214[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 EndocrinologyHome page
C. A. Pietsch, T. S. Scanlan, and R. J. Anderson
Thyronamines Are Substrates for Human Liver Sulfotransferases
Endocrinology, April 1, 2007; 148(4): 1921 - 1927.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ebmeier, C. C.
Right arrow Articles by Anderson, R. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ebmeier, C. C.
Right arrow Articles by Anderson, R. J.

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