| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Department of Endocrinology and Metabolism, Academic Medical Center, University of Amsterdam, 1100 DD Amsterdam, The Netherlands
Address all correspondence and requests for reprints to: Dr. Leon J. S. Brokken, Department of Endocrinology, F5-171, Academic Medical Center, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. E-mail: l.j.brokken{at}amc.uva.nl
Abstract
Antithyroid treatment for Graves’ hyperthyroidism restores euthyroidism clinically within 1–2 months, but it is well known that TSH levels can remain suppressed for many months despite normal free T4 and T3 levels. This has been attributed to a delayed recovery of the pituitary-thyroid axis. However, we recently showed that the pituitary contains a TSH receptor through which TSH secretion may be down-regulated via a paracrine feedback loop. In Graves’ disease, TSH receptor autoantibodies may also bind this pituitary receptor, thus causing continued TSH suppression. This hypothesis was tested in a rat model. Rat thyroids were blocked by methimazole, and the animals were supplemented with L-T4. They were then injected with purified human IgG from Graves’ disease patients at two different titers or with IgG from a healthy control (thyroid hormone binding inhibitory Ig, 591, 127, and < 5 U/liter). Despite similar T4 and T3 levels, TSH levels were indeed lower in the animals treated with high TSH receptor autoantibodies containing IgGs; the 48-h mean TSH concentration (mean ± SEM; n = 8) was 11.6 ± 1.3 ng/ml compared with 16.2 ± 0.9 ng/ml in the controls (P < 0.01). The intermediate strength TSH receptor autoantibody-treated animals had levels in between the other two groups (13.5 ± 2.0 ng/ml). We conclude that TSH receptor autoantibodies can directly suppress TSH levels independently of circulating thyroid hormone levels, suggesting a functioning pituitary TSH receptor.
GRAVES’ DISEASE IS an autoimmune thyroid disorder characterized by circulating Ig directed against the TSH receptor (TSH-R) (1, 2). The majority of these TSH-R autoantibodies (TRAb) act as agonists by mimicking TSH binding leading to Graves’ hyperthyroidism and goiter. Antithyroid drug treatment usually restores euthyroidism within 4–6 wk in patients with hyperthyroidism (3). However, it may take much longer for TSH values to normalize. Many treated Graves’ disease patients who are clinically euthyroid and have normal T4 and T3 serum levels continue to show decreased TSH levels (4, 5).
The explanation for this continued suppression of TSH is unknown, but it is usually attributed to a delayed recovery of the pituitary-thyroid axis (6). We offer an alternative explanation, involving a direct effect of TRAb on TSH secretion by the pituitary. We have recently postulated that in addition to a negative feedback control by T4 levels, TSH secretion is influenced through a negative ultra-short-loop feedback mechanism within the pituitary. We indeed demonstrated that the TSH-R is expressed in the human anterior pituitary on folliculo-stellate (FS) cells (7). When TSH is secreted by the thyrotrophs, it can bind to this receptor on FS cells, which then signal the thyrotrophs to decrease their TSH secretion. That the FS cells are involved in this feedback control is likely, because they are well known for their paracrine regulatory capabilities (8, 9). Apart from this physiological control, the TSH-R on FS cells may also bind circulating TRAb, which, by mimicking TSH, subsequently can cause a decrease in TSH secretion independently of thyroid hormone levels. Such a mechanism may very well be responsible for the low TSH levels observed in otherwise euthyroid Graves’ patients receiving antithyroid drug treatment. TRAb often remain present in patients treated for Graves’ disease (10, 11) and can be responsible for the long-term suppression of TSH.
To test this hypothesis, we used a modified long-acting thyroid stimulator bioassay in which we measured the plasma TSH response to the administration of TRAb in rats that were unable to mount a thyroid response to TRAb because of prior antithyroid drug treatment.
Materials and Methods
Animals
Adult female Wistar rats (Harlan Sprague Dawley, Inc., Zeist, The Netherlands), weighing approximately 325 g, were housed in cages at 21 C under a 12-h light, 12-h dark cycle (lights on at 0700 h and off at 1900 h). The animals received food and water ad libitum. The experiments described here were approved by the animal welfare committee.
Experimental design
To suppress thyroidal T4 production, 24 rats were treated with the antithyroid drug methimazole (1-methylimidazole-2-thiol; Sigma, St. Louis, MO) at a concentration of 0.05% (wt/vol) in the drinking water in combination with L-T4 (Sigma) dissolved in 0.9% NaCl (wt/vol) at 0.3 mg/ml and administered daily via a gastric tube in a dose of 1 ml/100 g BW. These dosages were determined in pilot experiments and resulted in slightly elevated basal TSH levels between 5–10 ng/ml. After 1 wk, the animals received a control IgG preparation (n = 8), a TRAb-containing IgG preparation of intermediate strength (n = 8), or a preparation with a high TRAb titer (n = 8) at 0900 h. Blood was collected in heparinized tubes immediately before and 1, 2, 4, 8, 24, and 48 h after the administration of 1 ml of the appropriate IgG preparation and was centrifuged at 3000 x g for 10 min at 4 C. Plasma was stored at -20 C for later analysis. Administration of IgG and subsequent withdrawal of blood were performed via the tail vein under mild fentanyl fluanisone/midazolam anesthesia (0.25 ml/100 g BW). Hematocrit was determined before and 8 h after the onset of the experiment.
IgG purification
TRAb-containing serum was obtained from 32 patients with Graves’ disease who had thyroid hormone binding inhibitory Ig (TBII) titers greater than 100 U/liter. Control serum was obtained from a healthy subject. The pooled TRAb-containing serum and the control serum were filtered through a 0.22-µm pore size, low protein binding filter (Millipore Corp., Bedford, MA), and the IgGs were isolated by affinity chromatography as described by Harlow and Lane (12). In short, the samples were passed over a 5-ml protein G-Sepharose column (HiTrap Protein G, Pharmacia Biotech, Piscataway, NJ) that was equilibrated with 0.1% BSA in run buffer (20 mM sodium phosphate buffer, pH 7.0). After washing with run buffer, the IgGs were eluted with 0.1 M glycine/HCl, pH 2.7, and 1-ml fractions were collected in tubes containing 44 µl 1 M Tris, pH 9.0, to neutralize the acid-labile IgGs. The protein-containing fractions were pooled and concentrated by ammonium sulfate precipitation (50%, wt/vol). The IgG preparations were dissolved in a minimal volume of PBS (pH 7.4) and finally dialyzed for 16 h at 4 C against several changes of PBS. Both preparations were diluted in PBS to 30 mg/ml protein. The TBII titer in the control IgG was less than 5 U/liter. The TRAb-containing IgG had a TBII titer of 591 U/liter. To include an intermediate strength preparation, this high TRAb pool was partly diluted with control IgG, yielding a TBII titer of 127 U/liter. The purity of the IgG preparation and the yield of the different IgG isotypes were assessed by immune electrophoresis and ELISA.
Hormone assays
TBII titers were measured by TRAK assay (Brahms Diagnostica, Berlin, Germany). TSH plasma levels were determined in a highly sensitive chemiluminescent enzyme immunoassay (Immulite Third Generation TSH kit, rat TSH application, Diagnostic Products, Los Angeles, CA). Total T4 (TT4) and total T3 plasma levels were determined by in-house RIAs (13) using rat null plasma as diluent. As an estimate of free T4 levels, the free T4 index (FT4I) was calculated as the product of T4 and T3 resin uptake. The latter was determined with a T3 Uptake Kit (Ortho-Clinical Diagnostics, Amersham Pharmacia Biotech, Little Chalfont, UK). Mean plasma levels of TSH, TT4, FT4I, and T3 levels over the 48-h period were calculated as the area under the curve divided by 48 h. All samples were measured within one run. Data are expressed as the mean ± SEM.
Statistical analysis
The data were analyzed using SPSS, Inc. software (version 7.5.2, SPSS, Inc., Chicago, IL). Time series were analyzed by ANOVA with repeated measurements and two grouping factors (time and treatment). A t test was used to compare the 48-h mean plasma levels. Differences between groups were considered significant at P < 0.05.
Results
The pooled serum samples yielded IgG preparations that were more than 99% pure with respect to total protein. The recovery of IgG1, IgG2, and IgG3 was more than 99% and 85% for IgG4.
At baseline, there were no differences in TSH,
TT4, T3, and
FT4I among the three groups (Fig. 1
). After injection of IgG, thyroid
function remained unaffected, as documented by similar
T3 levels in all three groups.
TT4 levels as well as FT4I
decreased transiently in all groups (Fig. 1, B and C, D). There
were no statistically significant differences in
TT4, FT4I, and
T3 values among the three groups.
|
). However, TSH levels in rats treated with high
TRAb-containing IgG were lower during the entire observation period
than those in rats that received control IgG (P <
0.01, by ANOVA). Rats treated with intermediate strength TRAb showed
TSH levels in between those in the other two groups.
The 48-h mean plasma hormone levels were calculated and showed no
differences among the groups with respect to T3,
TT4, or FT4I (Fig. 2). However, 48-h mean TSH plasma levels
were significantly reduced in the rats treated with the highest
concentration of TRAb (P < 0.01). Hematocrit did not
change during the experiment (data not shown).
|
In this rat model we showed that TRAb are capable of suppressing TSH levels through an extrathyroidal pathway. Intravenous administration of TRAb containing human IgG, in contrast to normal control IgG, to methimazole-treated rats induced a decrease in TSH levels without affecting T4 or T3 levels. This extrathyroidal effect of TRAb is most likely caused by the binding of these IgGs to the TSH-R in the pituitary. In a recent study we have shown that the TSH-R is expressed in the human anterior pituitary on the so-called FS cells (7). Other researchers not only confirmed this finding (14), but they also showed activation of adenylate cyclase by TSH in a mouse FS cell line. These cells make up approximately 10% of the pituitary cell population and are known for their regulatory effects on pituitary hormone secretion (8, 9). We hypothesized that these TSH-R-bearing FS cells play a role in fine-tuning of TSH secretion by the thyrotrophs through an ultra-short-loop negative feedback mechanism, possibly mediated via cytokines. In Graves’ disease, this may have a further consequence. The anterior pituitary resides outside the blood-brain barrier and is thus accessible to circulating IgGs. Thus, TRAb may very well bind to the TSH-R on FS cells, which may then send a paracrine signal to the thyrotrophs to diminish their TSH secretion. The present rat study strongly supports this postulate.
We do not think that the observed suppression of TSH levels upon administration of TRAb-containing IgGs can be explained otherwise. First, we included a TBII-negative, control IgG preparation that was administered in the same concentration as the two TBII-positive preparations, thus correcting for aspecific general effects of IgGs on the pituitary-thyroidal axis. Next, we used two concentrations of TRAb-containing IgGs and found an indication for a dose-response effect. Thirdly, the thyroid hormone levels were similar in all three groups over a 48-h period, and T4 and FT4I levels actually decreased slightly in all groups. This not only shows the effectiveness of the methimazole-induced block of thyroid hormone synthesis, but it also makes it highly unlikely that TRAb suppressed TSH levels via stimulation of the thyroid gland. In view of these considerations, we strongly believe that the TRAb sera indeed suppressed TSH levels via an extrathyroidal pathway.
We found that TSH levels increased rather sharply shortly after the administration of IgGs in all groups. We suggest that this was due to a stress response in the animals. Similar increases in TSH levels were found upon skin incision in patients undergoing cholecystectomy (14A ). In addition, part of the TSH increase may be explained by the naturally occurring morning surge in rats (15). We made our rats slightly hypothyroid in view of the mildly elevated TSH levels. This was done to ascertain that changes in TSH levels would indeed be detectable by the TSH assay.
We suggest that these data can be extrapolated to the human situation. When patients with Graves’ hyperthyroidism are rendered euthyroid, it is frequently seen that their TSH levels remain suppressed for a long time despite normal thyroid hormone levels (4, 5). It is also known that TRAb can remain present for a variable period of months to even years, and our data now support the hypothesis that TRAb may be responsible for extrathyroidal TSH suppression. We believe that this is a better explanation for continued TSH suppression than a delayed recovery of the pituitary-thyroid axis. The hypothesis can also explain another poorly understood phenomenon encountered in clinical practice. After 1 yr of antithyroid drug treatment, approximately 50% of Graves’ patients relapse. This occurs in patients with large goiters, in patients with high TBII titers (16, 17), and also in those who continue to have suppressed TSH levels in the absence of detectable TBII titers (18). We suggest that these suppressed TSH values result from biologically active TRAb below the detection limit of routine TBII assays.
Acknowledgments
We thank Adrie Maas (Department of Experimental Internal Medicine, Academic Medical Center) for skillful technical assistance.
Footnotes
Abbreviations: FS, Folliculo-stellate; FT4I, free T4 index; TBII, thyroid hormone binding inhibitory Ig; TRAb, TSH receptor autoantibodies; TSH-R, TSH receptor; TT4, total T4.
Received April 2, 2001.
Accepted June 12, 2001.
References
and cortisol in
the pathogenesis of enthyroid sick syndrome? Proceedings of the 81st
Annual Meeting of The Endocrine Society, San Diego, CA, 1999; Abstract
P3-422, p 529
This article has been cited by other articles:
 |
|
 |
|
  G. Gelwane, N. de Roux, D. Chevenne, J. C. Carel, and J. Leger Pituitary-Thyroid Feedback in a Patient with a Sporadic Activating Thyrotropin (TSH) Receptor Mutation: Implication That Thyroid-Secreted Factors Other Than Thyroid Hormones Contribute to Serum TSH Levels J. Clin. Endocrinol. Metab., August 1, 2009; 94(8): 2787 - 2791. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. V. H. Grommen, S. Taniuchi, T. Janssen, L. Schoofs, S. Takahashi, S. Takeuchi, V. M. Darras, and B. De Groef Molecular Cloning, Tissue Distribution, and Ontogenic Thyroidal Expression of the Chicken Thyrotropin Receptor Endocrinology, August 1, 2006; 147(8): 3943 - 3951. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. F. Prummel, C. B. Terwee, M. N. Gerding, L. Baldeschi, M. P. Mourits, L. Blank, F. W. Dekker, and W. M. Wiersinga A Randomized Controlled Trial of Orbital Radiotherapy Versus Sham Irradiation in Patients with Mild Graves' Ophthalmopathy J. Clin. Endocrinol. Metab., January 1, 2004; 89(1): 15 - 20. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. E. Kempers, D. A. van Tijn, A. S. P. van Trotsenburg, J. J. M. de Vijlder, B. M. Wiedijk, and T. Vulsma Central Congenital Hypothyroidism due to Gestational Hyperthyroidism: Detection Where Prevention Failed J. Clin. Endocrinol. Metab., December 1, 2003; 88(12): 5851 - 5857. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. J. S. Brokken, W. M. Wiersinga, and M. F. Prummel Thyrotropin Receptor Autoantibodies Are Associated with Continued Thyrotropin Suppression in Treated Euthyroid Graves' Disease Patients J. Clin. Endocrinol. Metab., September 1, 2003; 88(9): 4135 - 4138. [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 |
and 
Data
after administration of TRAb containing IgG obtained from patients with
Graves’ disease at intermediate and high concentrations, respectively.
•, Data obtained after administration of control IgG. Curves of
TRAb-treated animals are statistically compared with control curves by
ANOVA with repeated measurements and two grouping factors (time and
treatment). P = 0.009, difference between high
serum TRAb vs. control.