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 Google Scholar
Google Scholar
Right arrow Articles by Schuppert, F.
Right arrow Articles by Kohn, L. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schuppert, F.
Right arrow Articles by Kohn, L. D.
The Journal of Clinical Endocrinology & Metabolism Vol. 85, No. 2 858-867
Copyright © 2000 by The Endocrine Society

Original Studies

Increased Major Histocompatibility Complex (MHC) Expression in Nontoxic Goiters Is Associated with Iodide Depletion, Enhanced Ability of the Follicular Thyroglobulin to Increase MHC Gene Expression, and Thyroid Autoantibodies

Frank Schuppert, Dieter Ehrenthal, Andrea Frilling, Koichi Suzuki, Giorgio Napolitano and Leonard D. Kohn

Department of Clinical Endocrinology, Hannover Medical School (F.S., D.E.), D-30625 Hannover, Germany; Department of Surgery, Essen University (A.F.), D-45122 Essen, Germany; Chair of Endocrinology, University G. D’Annunzio (G.N.), I-66100 Chieti, Italy; and Metabolic Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health (K.S., L.D.K.), Bethesda, Maryland 20892

Address all correspondence and requests for reprints to: Frank Schuppert, M.D., Department of Clinical Endocrinology, Hannover Medical School, Carl-Neuberg-Strasse 1, D-30625 Hannover, Germany.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Recent studies suggest that thyroglobulin (TG) accumulated in the follicular lumen of colloid nodular goiters can increase major histocompatibility complex (MHC) class I gene expression in FRTL-5 thyrocytes. Iodide deficiency, also present in these patients, was separately suggested to enhance thyroidal MHC class I and class II gene expression in vivo and in vitro. To test the clinical relevance of these observations, we examined 41 nontoxic goiters surgically removed from patients who had compression problems. Northern analysis revealed that there was a mean 3.9-fold increase in MHC class I expression and a 8.3-fold increase in class II expression by comparison to 9 normal glands. In situ hybridization showed that thyrocytes were the main source of class I and class II transcripts; histological examination revealed that lymphocytic infiltration was minimal to nonexistent. The iodine content of the 41 nontoxic goiters was significantly lower than in normal glands, consistent with increased MHC class I and class II. There is also a profound accumulation of TG in the follicles of the nontoxic goiters, and TG purified from the follicles of these glands increased MHC class I gene expression in FRTL-5 thyroid cells significantly more than TG from normal glands per mg protein. Nearly all patients with nontoxic goiter had low, but significantly elevated, levels of antibodies against thyroid peroxidase and/or against TG in their sera compared with those in normal individuals. Moreover, there was a positive correlation between the titer of the serum antibodies against thyroid peroxidase and against TG and MHC class I and class II expression in the thyroid. The data support the possibility that the TG accumulated in the follicular lumen of nontoxic goiters together with relative iodine deficiency contributes to increased MHC expression in thyroid cells in vivo and that increased MHC gene expression contributes to the ability of thyroid antigens to trigger an autoimmune reaction.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
WE AND OTHERS have recently shown that serum samples and thyroid tissue obtained from patients suffering from nontoxic goiters display a number of autoimmune phenomena. Thus, there is an increase in autoantibodies against thyroid peroxidase (TPO) and/or thyroglobulin (TG) (1, 2) and increased levels of immunologically competent cells, such as CD8+CD57+ lymphocytes (3). There is also an increase in thyroidal interferon-{gamma} transcripts (4), other cytokine transcripts (5), major histocompatibility complex (MHC) class I transcripts and protein (6, 7, 8, 9), intercellular adhesion molecule 1 transcripts (10), and invariant chain transcripts (10). Despite this, we do not really consider nontoxic goiters a form of autoimmune thyroid disease because patients are usually euthyroid, because stimulating antibodies against the TSH receptor are absent (2), and because putative thyroid growth-stimulating autoantibodies are generally not considered to play an important role in the majority of patients suffering from nontoxic goiter (11).

Using a cultured thyroid cell model, we have shown that iodide down-regulates MHC class I and MHC class II messenger ribonucleic acid (mRNA) levels in human thyrocytes in vivo and in vitro, and conversely, low iodide levels increase MHC gene expression (9). In addition, we have recently shown that TG that accumulates in the follicular lumen, particularly in iodide-deficient goiters and colloid nodular goiters, increases MHC class I expression (12). Thus, TG has a potent negative regulatory effect on thyroid-restricted transcription factors, thyroid transcription factor-1 (TTF-1), TTF-2, and Pax-8 (12, 13). This decreases TG, TPO, sodium iodide symporter (NIS), and TSH receptor (TSHR) gene expression (12, 13, 14); however, MHC class I expression is increased because TTF-1 and TTF-1/Pax-8 elements surround a cAMP response element (CRE) on the MHC class I promoter. Decreasing TTF-1 and Pax-8 allows increased binding of the CRE-binding protein, CREB, to the CRE, thereby increasing class I expression (15, 16).

In this study we examined whether there was a higher level of thyroidal MHC transcript levels in nontoxic goiters, whether there was an associated increase in serum TG and TPO autoantibody levels in the patients, and whether increased thyroidal MHC transcript levels were correlated with increased serum TG and TPO autoantibody levels. We further tested whether the glands would be iodide depleted and if the follicular TG from these thyroids would increase MHC class I gene expression and suppress thyroid-restricted gene expression, consistent with the possibility that the increase in thyroidal MHC gene transcripts was associated with these phenomena and with the level of the TPO and/or TG autoantibodies. This report answers these questions affirmatively and, therefore, raises the possibility that the increase in MHC genes is clinically relevant, i.e. they decrease self-tolerance and result in an autoimmune phenomenon. We speculate, however, that full-blown autoimmune disease does not develop because the TG from nontoxic goiters simultaneously suppresses thyroid autoantigens, TG, TPO, NIS, and TSHR, more efficiently than TG from normal glands. This should decrease the ability of the autoantigens to be processed into peptides that can be presented to immune cells.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Patients: clinical parameters

Written consent was obtained from all participating patients after approval of the local ethics committee.

The following serum parameters were assessed in all participants (Table 1Go): total T3, total tetraiodothyronine, and T4-binding globulin (RIAs by Ciba Corning, Inc. Germany). The ratio of total T4 to T4-binding globulin was used to calculate the free T4 index. Basal TSH levels were measured with an immunoluminescence assay (Ciba Corning, Inc., Frankfurt A.M., Germany). Serum levels of TPO antibodies (TPOAb) and TG antibodies (TGAb) were measured with an enzyme-linked immunosorbent assay by Pharmacia & Upjohn, Inc. (Freiburg, Germany). TSH binding-inhibiting Ig (TBII) was determined with a radioreceptor ligand assay (TRAK, B.R.A.H.M.S., Berlin, Germany).


View this table:
[in this window]
[in a new window]
  		Table 1. Clinical parameters of all 50 patients

 
MHC expression was assessed in the thyroid tissues from a total of 50 patients: 41 with nontoxic goiters and 9 healthy controls. Patients with nontoxic goiters were euthyroid and had increased thyroid weight (>25 g). Patients with nontoxic goiter generally underwent surgery because of compression problems; patients with a healthy thyroid usually underwent surgery because of primary hyperparathyroidism. The histological examination of the thyroids confirmed the nontoxic goiter diagnosis in each case as well as the presence of normal thyroid tissue in the hyperparathyroid patients.

Lymphocyte infiltration was histologically assessed and semiquantitatively graded without knowledge of the clinical and experimental data. Twenty-two of 41 patients with nontoxic goiters received L-T4 (50 ± 15 µg; range, 0–100 µg daily). L-T4 medication was typically discontinued by the surgeons on admittance to the ward. The clinical parameters of all 50 patients are summarized in Table 1Go.

Northern blot analysis of patient thyroids

Total RNA isolation from thyroid fragments, RNA extraction, and Northern blot analysis were performed as described previously (8, 10). MHC class I [human leukocyte antigen I (HLA I)], measured using a 1500-bp PstI fragment of HLA-B8 complementary DNA (cDNA) in pUC9 (17), was provided by Dr. E. H. Weiss (Munich, Germany). MHC class II (HLA DR) was measured using a 1300-bp BamHI fragment of HLA-DR{alpha} cDNA (DR1) in pcDV1-pL2 (18), which was purchased from the American Type Culture Collection (Manassas, VA). In each case, 100 ng cDNA were labeled by random priming using a kit from Roche (Mannheim, Germany) and [a[lpha]-32P]deoxy-CTP (Amersham Pharmacia Biotech, Braunschweig, Germany). RNA integrity and the amount of RNA loaded per lane were assayed by measuring 18S and 28S ribosomal RNA (rRNA) in ethidium bromide stains. As a calibration standard, 1 µg of an oligonucleotide 28S rRNA probe was labeled using T4 polynucleotide kinase (Life Technologies, Inc., Gaithersburg, MD) and [{gamma}-32P]ATP (Amersham Pharmacia Biotech) as previously described (8, 10). Probes were typically labeled to a specific activity of approximately 110 cpm/µg DNA.

Autoradiographic signals were quantified by laser scanning densitometry (Pharmacia-LKB, Freiburg, Germany). Signals were compared to signals of oligonucleotide 28S rRNA to compensate for potential RNA loading differences (e.g. MHC class I mRNA/28S rRNA ratio). Expression levels of 28S rRNA showed only minimal, insignificant loading differences in all Northern blot analyses performed (data not shown). Therefore, differences in the expression levels of the parameters investigated represent true variations in mRNA content in the human thyroid gland.

In situ hybridization

Eight-micron thick cryostat sections mounted on sterilized slides pretreated with 3-aminopropyl-triethoxysilane (Sigma, St. Louis, MO) were fixed in 4% (wt/vol) paraformaldehyde in phosphate-buffered saline, pH 7.4, and acetylated to reduce nonspecific hybridization. In situ hybridization proceeded as previously described (8). A MHC class I or class II (HLA DR) complementary RNA probe (0.1 ng) was generated from a fragment of the original probe and cloned into a Bluescript vector.

Measurement of intrathyroidal total iodide content

Thyroid tissue was homogenized following the protocol of Tiran and co-workers (19). Digestion and photometric measurement proceeded as previously described (9).

TG preparations

Purified 19S follicular TG was prepared by salt extraction of sliced frozen thyroid glands, ammonium sulfate precipitation, and gel filtration chromatography on Sephacryl S-300 (Pharmacia Biotech, Uppsala, Sweden) as previously described (20, 21, 22). The purified TG preparation contained a single 330-kDa component by electrophoresis in SDS reducing gels (20, 21, 22).

Cell culture

The F1 subclone of FRTL-5 rat thyroid cells (Interthyr Research Foundation, Baltimore, MD) (23, 24) was grown in Coon’s modified F-12 medium containing 5% calf serum and 1 mmol/L nonessential amino acids supplemented with a mixture of six hormones (6H) including bovine TSH (1 x 10-10 mol/L), insulin (10 µg/mL), cortisol (0.4 ng/mL), transferrin (5 µg/mL), glycyl-L-histidyl-L-lysine acetate (10 ng/mL), and somatostatin (10 ng/mL). The FRTL-5 cells were diploid, between the 5th and 25th passage; their properties were previously described (12, 13, 14, 15, 16, 23, 24). Fresh medium was added every 2 or 3 days, and cells were passaged every 7–10 days.

RNA isolation from FRTL-5 cells and Northern analyses

RNA was prepared using a Total RNA Isolation Kit (5 Prime to 3 Prime, Inc., Boulder, CO) with minor modifications of the manufacturer’s protocol (12). Quantitation used a BAS-1500 Bioimaging Analyzer (Fuji Photo Film Co., Ltd., Stamford, CT). The probes for TG, TPO, MHC class I, TSHR, TTF-1, Pax-8, NIS, TTF-2, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were the same as those used previously (12).

To determine statistical significance in experiments measuring the effect of TG on class I RNA levels in Table 2Go and Fig. 6Go, the experiments were repeated at least three times using different batches of cells. The mean ± SD of these experiments were calculated. Significance between experimental values was determined by two-way ANOVA and are significant at P < 0.05.


View this table:
[in this window]
[in a new window]
  		Table 2. Relative amount of TG from a nontoxic goiter to increase MHC class I or decrease thyroid-restricted gene RNA levels in FRTL-5 thyroid cells, relative to TG from normal human thyroids

 


View larger version (16K):
[in this window]
[in a new window]
  		Figure 6. Ability of exogenous follicular TG isolated from nontoxic goiters to modulate MHC class I, TPO, TG, TSHR, NIS, TTF-2, TTF-1, and Pax-8 RNA levels in cultured thyrocytes relative to GAPDH. FRTL-5 cells in 6H medium were washed with fresh medium, and the incubation was continued in medium containing 5 mg/mL TG isolated from nontoxic goiters or 5 mg/mL human IgG (control). After 48 h, Northern analyses were performed using 20 µg total RNA. Blots were sequentially hybridized with probes for TG, TSHR, TPO, NIS, TTF-1, TTF-2, Pax-8, MHC class I, and GAPDH. The ratio of the binding of each probe to GAPDH was calculated, as GAPDH was not changed by TG. The ratio of probe to GAPDH signal intensity in control cells was set at 1 in each case. Experimental values were compared to their corresponding control values; data are the mean ± SD of results from five nontoxic goiters. The increase in MHC class I and the decreases in TG, TSHR, TPO, NIS, TTF-1, TTF-2, and Pax-8 to GAPDH ratios were significant at P < 0.01. The significance between control vs. experimental values was determined by two-way ANOVA; P < 0.05 was considered significant.

 

   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
MHC class I and MHC class II expressions in nontoxic goiters are increased, and expression of each is directly correlated

MHC class I and class II (HLA DR) mRNA expression levels were investigated in a total of 50 human thyroid glands. Representative Northern analyses for class I expression in 15 nontoxic goiters and a healthy normal control are presented (Fig. 1Go, upper panel) as are class II RNA levels (Fig. 1Go, lower panel). All thyroids with class I and/or class II transcripts showed the expected 1.4-kb band (8, 9).



View larger version (30K):
[in this window]
[in a new window]
  		Figure 1. MHC class I mRNA (upper panel) and MHC class II (HLA-DR) mRNA (lower panel) expression evaluated by Northern blot analysis in representative human thyroid glands. For Northern analysis, 20 µg total RNA were loaded per lane. Transcript size was assessed using a 9.5-kb RNA ladder (Life Technologies, Inc.), of which the 2.4- and 1.5-kb fragments are depicted. No significant differences in the amount of RNA loaded per lane were evident as assessed by measuring 18S and 28S rRNA in ethidium bromide stains and by measuring the signals of oligonucleotide 28S ribosomal RNA, which was used as a calibration standard (data not shown). After evaluating class I expression levels (upper panel), the same filters were rescreened with a class II (HLA-DR) cDNA probe (lower panel).

 
MHC class I mRNA expression levels were variable. Some nontoxic goiters displayed class I levels that were not different from those in the healthy controls, e.g. nontoxic goiter 10 (Fig. 1Go, upper panel). Other nontoxic goiters showed significant increased levels, i.e. nontoxic goiters 1, 5, 6, 12, and 13 (Fig. 1Go, upper panel). Nevertheless, by comparison to nine healthy thyroids, whose mean class I expression level was arbitrarily defined as 1, the nontoxic goiters had a mean level of 3.9 relative units (Fig. 2aGo). This was statistically different from the normal value (P < 0.001).



View larger version (16K):
[in this window]
[in a new window]
  		Figure 2. MHC class I (a) and MHC class II (HLA-DR; b) mRNA levels in all thyroids studied; correlation of MHC class I and class II (HLA-DR) mRNA levels in all 41 nontoxic goiters that were examined using the same blots (c). Class I and class II levels were quantitated and normalized to 28S rRNA levels to determine relative units (see Materials and Methods). Mean values of levels in 9 healthy normal thyroid glands were arbitrarily set at unity. All other values are plotted relative to these. Mean MHC class I and class II (HLA-DR) mRNA levels of the healthy thyroid glands were arbitrarily assigned a value of 1. The indicated P values were generated with the Mann-Whitney U test in relation to the healthy thyroids. P < 0.05 was considered significant. n.s., Not significantly different from healthy thyroids; *, P < 0.05; **, P < 0.01; ***, P < 0.001. Thick lines indicate the arithmetic mean. The Spearman rank correlation coefficient for nonnormal distributed samples in c is r = 0.64 (P < 0.001). Correlation was considered significant at P < 0.05.

 
Analysis of MHC class II expression yielded essentially the same picture. A wide variation was found, ranging from normal (e.g. nontoxic goiters 2, 3, 9, 10, and 11) to increased expression levels (e.g. nontoxic goiters 1, 5, 6, and 13; Fig. 1Go, lower panel). Again, however, by comparison to nine healthy thyroids, whose mean class II expression level was also arbitrarily defined as 1, the nontoxic goiters had a mean level of 8.3 relative units (Fig. 2bGo). This was again statistically different from the normal value (P < 0.001).

The signal densities of class I and class II RNA expressions were closely correlated (r = 0.64; P < 0.001; Fig. 2cGo). Measurements using multiple tissue fragments from the same glands yielded a correlation coefficient of more than 0.88; thus, these data appeared to be representative of all thyroids.

We considered the possibility that lymphocytic infiltration present in some glands might influence the results. We found that lymphocytic infiltration was minimal or nonexistent in our patients. Further, measurements from the multiple fragments yielded similar results. This suggested that lymphocytic infiltration was not influencing the results, as lymphocytic infiltrates are not uniform. Finally, we compared data from glands with and without lymphocytic infiltrates, as histologically assessed and semiquantitatively graded without knowledge of the clinical and experimental data. Results were not significantly different between glands with infiltrates and those without reexpression of class I or class II RNA levels, consistent with observations that the infiltrates were minimal relative to those in other autoimmune thyroid diseases.

In situ hybridization revealed that there was an increase in MHC class I as well as class II expression in all nontoxic goiters. An example of the increased MHC class I expression of thyrocytes in nontoxic goiters is shown in Fig. 3aGo. In contrast, a normal control shows lower autoradiographic grain densities in in situ positive thyrocytes (Fig. 3bGo). Lymphocytic infiltration was minimal to nonexistent. Histologically, there was profound accumulation of colloid (Fig. 3cGo). Follicles were heterogeneous in size and typically lined by a flat epithelium.



View larger version (90K):
[in this window]
[in a new window]
  		Figure 3. In situ detection of MHC class I mRNA in human thyroid tissue. Cryostat sections (8 µm) of two thyroid samples were hybridized with the 35S-labeled class I antisense complementary RNA probe and exposed for 4 days. As a control, no labeling was seen when thyroid tissue sections were hybridized with 35S-labeled sense probe (data not shown). Sections were counterstained with hematoxylin. Thyrocytes in nontoxic goiters show an increased MHC class I expression (a; 60-fold original magnification), whereas relatively low autoradiographic grain densities can be seen in thyrocytes of the healthy control tissue (b; 60-fold original magnification). Histologically, there was profound accumulation of colloid (c; 60-fold original magnification). Follicles were heterogeneous in size and typically lined by a flat epithelium. Lymphocytic infiltration in between follicles was minimal to absent.

 
Iodide content in the nontoxic goiters is decreased

Intrathyroidal total iodide content in the 41 nontoxic goiters investigated was significantly lower than that in the healthy controls, which displayed a well recognized wide variation (Fig. 4Go) (9, 25).



View larger version (13K):
[in this window]
[in a new window]
  		Figure 4. Intrathyroidal total iodide content in micrograms per 100 mg wet human thyroid tissue. Iodide was measured as denoted (see Materials and Methods). In each of the thyroids, values are the mean of at least two replicate tissue samples from different areas of the gland; in no case did values differ by more than 7%. The thick lines represent the arithmetic mean of all patients in one group, and an asterisk indicates a significant mean difference (P < 0.05) from the normal value, as determined by unpaired, two-tail t test for normally distributed samples.

 
TG levels in serum and the effect of follicular TG from the nontoxic goiters on expression of thyroid-restricted genes and on MHC class I

TG levels in serum were significantly elevated in patients with nontoxic goiters (Table 1Go and Fig. 5Go). The mean increase was 48-fold higher than the normal value.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 5. Serum TG levels were significantly elevated in patients with nontoxic goiters compared to serum levels in healthy controls. Only TG values with a recovery rate between 70–130% were considered (n = 40). The thick lines represent the arithmetic mean of all patients in one group. ***, Significant mean difference (P < 0.001) from the normal value, as determined by the Mann-Whitney U test.

 
Salt-extracted and Sephacryl-purified 19S TG from the nontoxic goiters had no effect on GAPDH RNA levels when incubated with FRTL-5 cells, but significantly increased MHC class I RNA levels, while decreasing RNA levels of TPO, TG, TSHR, NIS, TTF-2, TTF-1, and Pax-8 genes (Fig. 6Go). The effect on the thyroid-restricted transcription factors is specific because TG does not suppress ubiquitous transcription factors that also regulate TPO, TG, TSHR, and/or NIS (12). The mean ± SD for the effects of 5 mg/mL TG on MHC class I, TTF-1, TTF-2, Pax-8, TG, NIS, TPO, and TSHR gene, relative to GAPDH, by comparison to controls containing 5 mg/mL human IgG were significant at P < 0.01 for TG from five different nontoxic goiters (Fig. 6Go). The TG concentration tested in these experiments, 5 mg/mL, is in the range of TG measured in the colloid of individual follicles by fine needle aspiration biopsy (0.1–3 mg/mL in normal human thyroids and up to 14 mg/mL in colloid nodules) (26). Increased class I establishes that TG is not simply a general suppressor of all genes.

The effects of TG on TPO, NIS, TSHR, TG, TTF-2, TTF-1, Pax-8, and class I mRNA levels were not duplicated by albumin at 10 mg/mL or mannitol at 1 mg/mL, as previously described (data not shown) (12, 14). The effect of mannitol was tested to insure that the TG action was not an osmotic pressure effect. The osmotic pressure of 1 mg/mL mannitol is 5.5 mosmol; that of 10 mg/mL TG is 0.14 mosmol. Neither 10-7 mol/L T3, 10-7 mol/L T4, nor 10-3 mol/L iodide duplicated the TG effects (data not shown).

Salt-extracted/Sephacryl-purified human 19S follicular TG obtained from the nontoxic goiters was 2- to 3-fold more efficient than salt-extracted/agarose-purified normal human 19S follicular TG at increasing MHC class I transcripts and decreasing those of the thyroid-restricted genes (Table 2Go).

TPOAb and TGAb serum levels in patients with nontoxic goiters and their relationship to class I and class II MHC expression in the corresponding thyroid gland

TPOAb and TGAb serum levels in all 41 patients with nontoxic goiter were investigated (Fig. 7Go, a and b, respectively). The mean serum TPOAb level was 17 U/mL (Fig. 7aGo and Table 1Go); this was statistically different from the normal value (P < 0.05). The range was from less than 1 to 91 U/mL. The mean TGAb serum level was 35 U/mL (Fig. 7bGo and Table 1Go); again, this was statistically different from the normal value (P < 0.01). The range was from 4–99 U/mL. The correlation between both parameters was highly significant (r = 0.47; P = 0.0035; Fig. 7cGo) and in both cases was higher than that in normal controls, albeit lower than the commonly used cut-off for autoimmune disease (<100 U/mL). In no case were TBIIs detectable (Table 1Go). TPOAb and TGAb serum levels also significantly correlated with MHC class I and MHC class II (HLA DR) expression levels, as assessed in the corresponding goiters. In all cases P < 0.05 (Table 3Go).



View larger version (15K):
[in this window]
[in a new window]
  		Figure 7. TPOAb (a) and TGAb (b) serum levels in all 41 nontoxic goiters and all 9 healthy controls; correlation of TPOAb and TGAb serum levels in patients with nontoxic goiters (c). To facilitate statistical evaluation, a TPOAb serum level less than 1 U/mL was equated to 1 U/mL. The indicated P values were generated with the Mann-Whitney U test in relation to healthy thyroids. P < 0.05 was considered significant. *, P < 0.05; **, P < 0.01. Thick lines indicate the arithmetic mean. The Spearman rank correlation coefficient for nonnormally distributed samples in c is r = 0.47 (P < 0.01). Correlations were considered significant at P < 0.05.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Correlation of TPOAb and TGAb serum levels from those patients whose corresponding nontoxic goiters were investigated for MHC class I and class II mRNA expression levels

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Iodide has been reported to decrease MHC class I and class II gene expression in rat FRTL-5 thyrocytes, human thyrocytes, and Graves’ thyroids in vivo (9, 27, 28). The iodide action involves transcriptional regulation of nuclear factor-{kappa}B (28). More recently, TG stored in the follicular lumen was shown to significantly increase MHC class I gene expression coincident with its ability to act as a feedback suppressor of thyroid function and its ability to suppress TSH activation of thyroid-restricted genes, TPO, TG, and NIS (12, 13, 14). The basis for the TG action is its ability to suppress thyroid-restricted transcription factors that control these genes: TTF-1 (12, 13), TTF-2 (12), and Pax-8 (12). It is suggested that TG suppression of TTF-1 and Pax-8 allows CREB and activating transcription factor-1 to bind to a CRE site that is within -127 bp of the MHC class I 5'-flanking region and is associated with a downstream silencer (15, 16). This eliminates silencing activity and increases MHC class I levels (15, 16).

The mechanism by which TG suppression occurs involves an apical membrane receptor that binds TG; the receptor is a homolog of the asialoglycoprotein receptor on the liver (29, 30). This receptor will preferentially bind the low iodinated, poorly sialylated TG that is found in low iodine goiters and colloid adenomas (12, 20, 21, 31, 32).

The above observations raised the possibility that the TG in goiters associated with iodide deficiency might increase MHC class I expression by its effect on thyroid-restricted transcription factors, that low iodide would contribute to the increase in class I and class II by an independent action regulating nuclear factor-{kappa}B, a ubiquitous transcription factor, and that the increase in MHC expression might induce a clinically detectable autoimmune response. This possibility might explain, for example, why nontoxic goiters are associated with a number of autoimmune phenomena; for example, the increase in autoantibodies against TPO and/or TG (1, 2) in serum.

In this report we show that nontoxic goiters do indeed have significantly increased MHC class I and class II RNA levels and that the levels are directly correlated. We show that the goiters have a low iodide content relative to normal thyroids. They also have an accumulation of TG in their follicular lumen that is both greater in amount than that in normal glands by histological appearance and has a higher specific activity with respect to its ability to increase class I RNA levels compared to TG from normal glands. It is, therefore, plausible that the TG accumulation and the low iodide level increase class I transcripts, and that the low iodide level increases class II transcripts.

In the present report we show that the patients have significantly higher TPOAb and TGAb levels in their sera than those in the sera of normal humans, but below the cut-off (<100 U/mL) recommended by the manufacturer of the assays for the measurement of TPOAB and TGAb (Pharmacia & Upjohn, Inc.) to distinguish patients with classic thyroid autoimmune diseases such as Graves’ or Hashimoto’s from patients with other types of thyroid diseases. We specifically investigated only nontoxic goiters whose TGAb and TPOAb serum levels were less than100 U/mL to ensure that solely nontoxic goiters, but no samples from classic thyroid autoimmune disease such as Graves’ or Hashimoto’s, were included in this study. Moreover, we show the titers directly correlate with levels of the MHC gene transcripts expressed in the thyroid. In sum, there is circumstantial evidence that associates the increases in MHC gene transcripts with enhanced autoantibody levels and a plausible basis for the MHC increase to occur. Nontoxic goiter can, therefore, be reasonably construed as a mild form of autoimmune thyroid disease induced by relative iodide deficiency and the abnormal accumulation of TG within the thyroid follicles.

A major question emerging from these data is why the autoimmune phenomenon is so mild? We speculate that one reason relates to the ability of the nontoxic goiter TG to suppress thyroid autoantigens as reflected in the RNA decreases (Ref. 12 and this report). This would mean there is lower synthesis and lesser levels of TG, TPO, NIS, or TSHR peptides available to be presented to the immune cells in the context of the increased MHC class I and class II genes. For example, TG suppresses TTF-1 (12, 13), and TTF-1 is the sole thyroid-restricted transcription factor that regulates TSHR expression (33, 34, 35). The TSHR decrease would preclude the development of anti-TSHR autoantibodies, i.e. no TBIIs would be seen (Table 1Go).

TG in serum is increased. We presume this occurs because of the primary and unknown defect that causes its high accumulation in follicles and because of disordered architecture or secretion. TG in serum would not expected to be presented to immune cells in the context of abnormally expressed class I or class II on the target tissue. It will be interesting to epitope map the TG autoantibodies in these patients, as recent evidence suggests the TSHR and TPO yield different antibody populations when expressed in the context of abnormally expressed MHC genes (36, 37).

In some patients another inducing factor, genetic, viral, or traumatic, for example, might set off a more full-blown autoimmune disease, as is in the case of iodide-deficient goiters associated with growth autoantibodies (38, 39, 40, 41).


   Acknowledgments
 
We express gratitude to Mrs. S. Deiters and Mrs. S. Horter for their excellent technical assistance. We thank (in alphabetical order) H. Dralle, M.D., Chief of the Department of Surgery, Halle-Wittenberg University, Halle (Saale), Germany; C. Hoang Vu, Ph.D., Department of Surgery, Halle-Wittenberg University, Halle (Saale), Germany; G. F. W. Scheumann, M.D., Department of Abdominal and Transplant Surgery, Hannover Medical School, Hannover, Germany; S. Schröder, M.D., Institute for Immunology, Pathology, and Molecular Biology, Hamburg, Germany; J. Simanowski, M.D., Department of Heart, Thorax, and Vessel Surgery, Hannover Medical School, Hannover, Germany; and G. Weinland, M.D., Israelitic Hospital, Hamburg, Germany, for the assistance with collecting some of the thyroid tissue used in this study.

Received June 28, 1999.

Revised October 21, 1999.

Accepted November 8, 1999.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Zonenberg A, Kinalska I, Zarzycki W, Telejko B. 1994 Incidence of thyroid autoantibodies in the endemic goiter. Horm Metab Res. 26:238–242.[Medline]
  2. Schuppert F, Deiters S, Rambusch E, Sierralta W, Dralle H, Von zur Mühlen A. 1996 TSH-receptor expression and human thyroid disease: relation to clinical, endocrine and molecular thyroid parameters. Thyroid. 6:575–587.[Medline]
  3. Corrales JJ, Orfao A, Miralles JM, et al. 1993 Immunological features of sporadic multinodular goiter. Clin Invest. 71:552–558.[Medline]
  4. Schuppert F, Hiller WFA, Weinland G, Hoang-Vu C, Von zur Mühlen A, Kohn LD. 1996 Iodide, cytokines and TSH receptor expression in Graves’ disease. Exp Clin Endocrinol Diabetes. 104(Suppl 4):68–74.
  5. Watson PF, Pickerill AP, Davies R, Weetman AP. 1994 Analysis of cytokine gene expression in Graves’ disease and multinodular goiter. J Clin Endocrinol Metab. 79:355–360.[Abstract]
  6. Jansson R, Karlsson A, Forsum U. 1984 Intrathyroidal HLA-DR expression and T lymphocyte phenotypes in Graves’ thyrotoxicosis, Hashimoto’s thyroiditis and nodular colloid goitre. Clin Exp Immunol. 58:264–272.[Medline]
  7. Lucas-Martin A, Foz-Sala M, Todd I, Bottazzo GF, Pujol-Borrell R. 1988 Occurrence of thyrocyte HLA class II expression in a wide variety of thyroid diseases: relationship with lymphocytic infiltration and thyroid autoantibodies. J Clin Endocrinol Metab. 66:367–375.[Abstract/Free Full Text]
  8. Schuppert F, Reiser M, Scheumann GFW, et al. 1993 Expression levels of the thyrotropin receptor gene in autoimmune thyroid disease: coregulation with parameters of thyroid function and inverse relation to major histocompatibility complex classes I and II. J Clin Endocrinol Metab. 76:1349–1356.[Abstract]
  9. Schuppert F, Taniguchi SI, Schröder S, Dralle H, Von zur Mühlen A, Kohn LD. 1996 In vivo and in vitro evidence for iodide regulation of MHC class I and class II expression in Graves disease. J Clin Endocrinol Metab. 81:3622–3628.[Abstract]
  10. Schuppert F, Reiser M, Heldin N-E, et al. 1994 TSH-receptor and leukocyte adhesion molecules in autoimmune thyroid disease: a study of their gene expression by Northern blot analysis and in situ hybridization. Acta Endocrinol (Copenh). 131:480–488.[Abstract/Free Full Text]
  11. Peter, HJ, Bürgi U, Gerber H. 1996 Pathogenesis of nontoxic diffuse and nodular goiter. In: The thyroid, Braverman LE, Utiger RD, eds. Philadelphia, New York: Lippincott-Raven; 890–895.
  12. Suzuki K, Lavaroni S, Mori A, et al. 1998 Autoregulation of thyroid-specific gene transcription by thyroglobulin. Proc Natl Acad Sci USA. 95:8251–8256.[Abstract/Free Full Text]
  13. Suzuki K, Mori A, Lavaroni S, et al . 1999 In vivo expression of thyroid transcription factor-1 (TTF-1) RNA and its relation to thyroid function and follicular heterogeneity: identification of follicular thyroglobulin (TG) as a feedback suppressor of TTF-1 RNA levels and TG synthesis. Thyroid. 9:319–331.[Medline]
  14. Suzuki K, Mori A, Saito J, Ullianich L, Kohn LD. Follicular thyroglobulin suppresses iodide uptake by suppressing expression of the sodium/iodide symporter gene. Endocrinology. In press.
  15. Shong M, Taniguchi SI, Napolitano G, et al . 1995 Major histocompatibility class I gene expression in thyroid cells is regulated by tissue-specific transcription factors: thyroid transcription factor-1 and Pax-8. Thyroid. 5(Suppl 1):S18.
  16. Saji M, Shong M, Napolitano G, et al. 1997 Regulation of major histocompatibility complex class I gene expression in thyroid cells. Role of the cAMP response element-like sequence. J Biol Chem. 272:20096–20107.[Abstract/Free Full Text]
  17. Pohla H, Kuon W, Tabaczewski P, Doerner C, Weiss EH. 1989 Allelic variation in HLA-B and HLA-C sequences and the evolution of the HLA-B alleles. Immunogenetics. 29:297–307.[CrossRef][Medline]
  18. Sekaly RP, Tonnelle C, Strubin M, Mach B, Long EO. 1986 Cell surface expression of class II histocompatibility antigens occurs in the absence of the invariant chain. J Exp Med. 164:1490–1504.[Abstract/Free Full Text]
  19. Tiran B, Wawschinek O, Eber O, Beham A, Lax S, Dermelj M. 1991 Simple determination of iodine in small specimens of thyroid tissues. Exp Clin Endocrinol. 98:32–36.[Medline]
  20. Shifrin S, Kohn LD. 1981 Binding of thyroglobulin to bovine thyroid membranes. Role of specific amino acids in receptor recognition. J Biol Chem. 256:10600–10605.[Abstract/Free Full Text]
  21. Consiglio E, Shifrin S, Yavin Z, et al. 1981 Thyroglobulin interactions with thyroid membranes. Relationship between receptor recognition of N-acetylglucosamine residues and the iodine content of thyroglobulin preparations. J Biol Chem. 256:10592–10599.[Abstract/Free Full Text]
  22. Consiglio E, Acquaviva AM, Formisano S, et al. 1987 Characterization of phosphate residues on thyroglobulin. J Biol Chem. 262:10304–10314.[Abstract/Free Full Text]
  23. Ambesi-Impiombato FS. 1986 Fast-growing thyroid cell strain. US Patent 4:608:341.
  24. Kohn LD, Valente WA, Grollman EF, Aloj SM, Vitti P. 1986 Clinical determination and/or quantification of thyrotropin and a variety of stimulatory or inhibitory factors performed in vitro with an improved cell line FRTL-5. US Patent 4:609:622.
  25. Aeschimann S, Buergi U, Wagner HE, Kaempf J, Lauber K, Studer H. 1994 Low intrathyroidal iodine concentration in non-endemic human goitres: a consequence rather than a cause of autonomous goitre growth. J Endocrinol. 140:155–164.[Abstract/Free Full Text]
  26. Salabe GB, Corvo L, Lotz H. 1996 Thyroglobulin determined in thyroid fine needle aspiration biopsies by radial immunodiffusion and electroimmuno-diffusion. Eur J Clin Chem Clin Biochem. 34:43–47.[Medline]
  27. Saji M, Moriarty J, Ban T, Singer DS, Kohn LD. 1992 Major histocompatibility complex class I gene expression in rat thyroid cells is regulated by hormones, methimazole, and iodide as well as interferon. J Clin Endocrinol Metab. 75:871–878.[Abstract]
  28. Taniguchi SI, Shong M, Giuliani C, et al. 1998 Iodide suppression of major histocompatibility class I gene expression in thyroid cells involves enhancer A and the transcription factor NF-{kappa}B. Mol Endocrinol. 12:19–33.[Abstract/Free Full Text]
  29. Pacifico F, Aquaviva R, Liguoro D, et al. 1999 TG binding and TSH regulation of RHL-1 subunit of the asialoglycoprotein receptor in rat thyroid. Biochimie. 81:493–496.[Medline]
  30. Ulianich L, Suzuki K, Mori A, et al. 1999 Follicular thyroglobulin (TG) suppression of thyroid-restricted genes involves the apical membrane asialoglycoprotein receptor and TG phosphorylation. J Biol Chem. 274:25099–25107.[Abstract/Free Full Text]
  31. Consiglio E, Salvatore G, Rall JE, Kohn LD. 1979 Thyroglobulin interactions with thyroid plasma membranes: the existence of specific receptors and their potential rule. J Biol Chem. 254:5065–5076.[Abstract/Free Full Text]
  32. Kohn LD, De Luca M, Santisteban P, et al. 1985 Thyroglobulin interactions with thyroid membranes: implications for the regulation of thyroid hormone formation in thyroglobulin. In: Eggo M, Burrow G, eds. Thyroglobulin: the prothyroid hormone, progress in endocrine research and therapy. New York: Raven Press; 171–214.
  33. Saji M, Akamizu T, Sanchez M, et al. 1992 Regulation of thyrotropin receptor gene expression in rat FRTL-5 thyroid cells. Endocrinology. 130:520–533.[Abstract/Free Full Text]
  34. Shimura H, Okajima F, Ikuyama S, et al. 1994 Thyroid-specific expression and cyclic adenosine 3',5'-monophosphate autoregulation of the thyrotropin receptor gene involves thyroid transcription factor-1. Mol Endocrinol. 8:1049–1069.[Abstract/Free Full Text]
  35. Civitareale D, Castelli MP, Falasca P, Saiardi A. 1993 Thyroid transcription factor 1 activates the promoter of the thyrotropin receptor gene. Mol Endocrinol. 7:1589–1595.[Abstract/Free Full Text]
  36. Shimojo N, Kohno Y, Yamaguchi K, et al. 1996 Induction of Graves-like disease in mice by immunization with fibroblasts transfected with the thyrotropin receptor and a class II molecule. Proc Natl Acad Sci USA. 93:11074–11079.[Abstract/Free Full Text]
  37. Jaume JC, Guo J, Wang Y, Rapoport B, McLachlan SM. 1999 Cellular thyroid peroxidase (TPO), unlike purified TPO and adjuvant, induces antibodies in mice that resemble autoantibodies in human autoimmune thyroid disease. J Clin Endocrinol Metab. 84:1651–1657.[Abstract/Free Full Text]
  38. Suzuki K, Mori A, Ishii KJ, et al. 1999 Activation of target tissue immune recognition molecules by double strand polynucleotides. Proc Natl Acad Sci USA. 96:2285–2290.[Abstract/Free Full Text]
  39. Drexhage HA, Bottazzo GF, Doniach D, Bitensky L, Chayen J. 1980 Evidence for thyroid-growth-stimulating immunoglobulins in some goitrous thyroid diseases. Lancet. 2:287–292.
  40. Schatz H, Pschierer-Berg K, Nickel JA, et al. 1985 Assay for thyroid-growth stimulating immunoglobulins: stimulation of 3H-thymidine incorporation into isolated thyroid follicles by TSH together with thyroglobulin, by EGF, and by immunoglobulins of goitrous patients from an iodine-deficient region. In: Medeiros-Neto G, Gaitan E, eds. Frontiers in thyroidology. New York: Plenum Press; 885–891.
  41. Medeiros-Neto G, Halpern A, Cozzi ZS, Lima N, Kohn LD. 1986 Thyroid growth immunoglobulins in large multinodular endemic goiters: effect of iodized oil. J Clin Endocrinol Metab. 63:644–650.[Abstract/Free Full Text]




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 Google Scholar
Google Scholar
Right arrow Articles by Schuppert, F.
Right arrow Articles by Kohn, L. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schuppert, F.
Right arrow Articles by Kohn, L. D.

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