Search

In order to characterize the role of leukocyte-activating antigens and other immunological parameters in autoimmune thyroid disease, mRNA levels of intercellular adhesion molecule 1 (ICAM-1), endothelial leukocyte adhesion molecule 1 (ELAM-1, E-selectin), invariant chain (Ii) and the thymic hormone thymosin beta (T beta 4) were investigated in 18 human thyroid glands, including eight Graves' thyroids, two Hashimoto's thyroids, two endemic goiters and six healthy controls. Northern blot analysis showed that in autoimmune thyroid disease, expression of ICAM-1 and T beta 4 was correlated to transcript levels of Ii, whereas in the healthy controls, expression of T beta 4, ICAM-1 and ELAM-1 was low or nearly absent. ELAM-1 and TSH receptor (TSH-R) expression, the latter serving as a thyroid specific marker, was increased in some diseased gland but showed no relation to the immunological parameters mentioned above. Localization of the specific mRNAs by in situ hybridization demonstrated a cell-specific expression of TSH-R (thyrocytes), ELAM-1 (vascular endothelial cells) and T beta 4 (cells of hematopoietic origin). In contrast, transcripts of Ii and ICAM-1 were found in thyrocytes, leukocytes and endothelial cells. Our results implicate a coordinate expression of ICAM-1, T beta 4 and Ii in autoimmune thyroid disease, yielding distinct cellular expression patterns. Differential expression of ICAM-1, Ii and the TSH-R in thyroid epithelial cells indicates active regulatory events within the thyrocyte.

Using a human TSH receptor (TSH-R) cDNA probe, we investigated TSH-R transcript levels in 13 human thyroid fragments by Northern blot analysis; 7 Graves' disease, 2 Hashimoto's disease, 3 endemic goiter, and 1 healthy thyroid gland were studied. TSH-R expression levels were variable, but displayed a close correlation to the expression of thyroid peroxidase (r = 0.703; P < 0.05), thyroglobulin (r = 0.817; P < 0.01), and the nuclear oncogene c-fos (r = 0.935; P < 0.001), but not c-myc. Overall, TSH-R transcript levels were low or absent in those thyroids in which expression of the major histocompatibility complex class I or II (MHC I or II) was high, thus establishing an inverse relation (MHC I, r = -0.791; P < 0.01; MHC II, r = -0.784; P < 0.01). In situ hybridization showed that apart from lymphocytes, thyroid cells themselves were the source of MHC II transcripts. gamma-Interferon expression was only detectable in 1 Hashimoto's goiter. Our findings suggest that next to lymphocyte infiltration, active regulatory events in the thyrocyte are responsible for the inverse relation between functional parameters (TSH-R, thyroid peroxidase, thyroglobulin, and c-fos) and immunological markers (MHC I and II).

Nucleotide sequence analysis of PCR fragments of TSH receptor cDNA synthesized from thyroid RNA of a patient with autoimmune hyperthyroidism, revealed two different sequences in the first position of codon 36. In one of the sequences, there was a C for G substitution leading to the D36----H substitution in the predicted peptide. Both variants were also found in genomic DNA of thyroid tissue. However, only the germ line sequence was found in other tissues representing all three different germ layers. The novel sequence is therefore likely to represent a somatic mutation in the thyroid tissue, of possible relevance for the pathogenesis of the patient's thyroid disorder.

There has been much controversy about the presence of TNF-α within thyroid tissue. We therefore conducted a study to determine if TNF-α mRNA is present in thyroid tissue and thyroid-derived cells. Semiquantitative reverse transcriptase-polymerase chain reaction (RT-PCR) was employed with a heterologous competitor fragment. Significantly lower levels of TNF-α mRNA were found in the autonomous nodules from patients with thyroid autonomy (TA; n = 4; 5.7 ± 1.3 arbitrary units (AU) (mean ± s.e.m.); P < 0.03) and in normal thyroid tissue (n = 2, 7.0 ± 3.1 AU) compared with tissue from patients with Graves' disease (GD; n = 13; 27.9 ± 10.3 AU), non-toxic multinodular goitre (NTG; n = 5:20.9 ± 5.8 AU) and perinodular tissue from TA patients (20.3 ± 4.0 AU). Higher levels were detected in tissues from patients with Hashimoto's thyroiditis (HT; n = 2; 51.3 ± 10.3 AU). Cultures of pure thyroid-derived fibroblasts (46 ± 18AU), thyrocytes (33 ± 8AU), and the anaplastic thyroid carcinoma cell lines 8505 C (39 ± 11 AU), SW 1736 (214 ± 16 AU) and C 643 (3 ± 1 AU) showed significantly lower TNF-α mRNA levels than thyroid-derived lymphocytes (1650 ± 32AU). TNF-α was detected in the supernatants or unstimulated lymphocytes (22.1 ± 1.1 pgs/ml) and SW 1736 cells (3.5 ± 0.9 pg/ml), but not in unstimulated fibroblasts and thyrocytes. Using an intracellular labelling technique in flow cytometry, the immunophenotype of stimulated TNF-α-positive lymphocytes was determined as predominantly CD3+CD45RO+. Our results suggest that TNF-α is present in the thyroid tissue of different thyroid disorders. Thyroid-derived lymphocytes are potential TNF-α producers and may thus locally influence thyroid function. [ABSTRACT FROM AUTHOR]