One of the hypotheses that explains the pathogenesis of thyroid-associated ophthalmopathy (TAO) is that thyroglobulin (Tg) is transported through a retrograde lymphatic route to orbital tissues (OT), where it elicits autoimmune damage. In a previous study we demonstrated the presence of intact Tg of thyroid origin in OT from three patients with TAO. The present study was undertaken to investigate this issue further, by increasing the number of patients, by analyzing the distribution of Tg in OT, and by investigating possible relations between the presence of Tg in OT and the clinical features of patients. OT was obtained from seven patients with TAO who underwent decompressive orbitotomy via a transpalpebral approach. Patients were designated P10 to P16. Inflamed palpebral skin, retrobulbar fibroadipose tissue and extraocular muscle surgical samples were collected separately. Tissue extracts were prepared by homogenization and analyzed for the presence of Tg using two different techniques. We first performed immunoprecipitation experiments, in which a rabbit polyclonal anti-Tg antibody was used to capture Tg on protein A and a mouse monoclonal anti-Tg antibody was used to re-veal captured Tg by Western blotting. Intact 330-kd Tg was detected in retrobulbar fibroadipose tissue extracts from three patients (P10, P11, and P16), whereas no Tg was detected in retrobulbar fibroadipose tissue extracts from the remaining four patients. Tg was not detected in the extraocular muscle extracts from all patients studied. In addition, intact 330-kd Tg was found in the inflamed palpebral skin extract from one patient (P10). No Tg was detected in OT extracts from two patients without thyroid or eye disease and in abdominal adipose tissue extracts from two obese patients without thyroid or eye disease. We then searched for Tg by enzyme-linked immunosorbent assay (ELISA), using the same antibodies used for immunoprecipitation. Tg was detected in retrobulbar fibroadipose tissue extracts from four patients (P10, P11, P12, and P16) and in the inflamed palpebral skin extract from patient P10, in amounts ranging from approximately 125 to approximately 400 pg/microg of tissue protein. Again, Tg was not detected in extraocular muscle extracts. A positive gradient between Tg in OT and Tg in the serum was observed in patient P12. Using an ELISA approach, we found that Tg in OT from three TAO patients (P10, P11, and P12) contained thyroxine (T4) residues (mean T(4)
Content: 2.42 molecules per molecule of Tg), indicating that Tg had originated in the thyroid. Combining the results obtained in our previous and present study, we found a possible relation between the presence of Tg in OT and the previous treatment with radioiodine. Thus, of the seven patients (3 in the previous and 4 in the present study) in whose OT Tg was found, six had been treated with radioiodine, whereas of the three patients with no Tg in their OT none had been treated with radioiodine. In conclusion, Tg was found in OT extracts from patients with TAO by immunoprecipitation in three of seven cases and by ELISA in four of seven cases. Tg was found in retrobulbar fibroadipose tissue, but not in extraocular muscles. There was a relation between the presence of Tg in OT and the previous treatment with radioiodine. Our results support the hypothesis that Tg may play a role as a coantigen in the pathogenesis of TAO. Further studies are needed to investigate this possibility.