PubMed2002
In an attempt to develop an animal model for thyroid-associated ophthalmopathy (TAO) we have genetically immunized BALB/c and outbred (CD-1) mice with cDNAs encoding the thyroid and eye muscle shared protein G2s and full length human thyrotropin receptor (TSHr). Firstly, BALB/c mice were immunized with cDNAs for G2s and the TSHr, alone or in tandem with cDNAs for interleukin (IL)4 or IL12. Control mice were immunized with empty vehicle only. Sera from the great majority of experimental mice contained antibodies against a G2s fusion protein and the flavoprotein (Fp) subunit of mitochondrial succinate dehydrogenase, the "64 kDa protein", with the greatest levels being found at sacrifice (17 wk). Antibody levels in mice immunized with G2s + TSHr or G2s + IL12 were generally higher than those in mice immunized with G2s only. TSHr antibodies (TRAb), measured as TSH binding inhibition, were detected in only two mice. On histological examination of the orbits, mild edema, eye muscle fiber separation and mast cell infiltration in and around the eye muscles were found in the majority of experimental mice, but not in control mice. Splenocytes were transferred from selected G2s-immunized mice to normal syngeneic litter mates. None of the transfer mice had serum antibodies against G2s, Fp or TSHr but their orbital tissue showed the same degree of mast cell infiltration as primary mice. No major histological changes were observed in the thyroid or other skeletal muscle in either primary or transfer mice. Similar results were observed in CD-1 mice although, overall, the model was better expressed than in BALB/c mice. In these mice, serum anti-G2s antibody levels were not significantly different between the various experimental groups except at 16 wk, when they were slightly greater than in control animals. Anti-Fp antibodies were detected at 12, 14 and 16 wk, in all experimental groups, including those immunized with G2s only, and were greatest in mice immunized with TSHr alone. TRAb levels were greatest in mice immunized with both G2s and the TSHr in the presence of TL4, but not IL12. The finding of negative anti-G2s but positive anti-Fp antibodies in some CD-1 mice suggests that eye muscle damage and Fp release must have been mediated by T lymphocytes, rather than antibodies, targeting G2s or some other as yet unidentified cell membrane antigen. Histological changes in the orbit were similar to those observed in BALB/c mice although mast cell numbers were greater, in both primary and transfer mice. Overall, the greatest histological changes were observed in CD-1 mice immunized with both G2s + TSHr + IL4. None of the animals became overtly hyperthyroid or hypothyroid during the course of the study although several of the CD-1 mice had abnormal TSH or T4 levels. These results indicate that we have established a valid model for human ophthalmopathy using the novel thyroid and eye muscle expressed protein G2s, now recognized as a fragment of the winged-helix transcription factor Foxp1, and TSHr, and that G2s and the TSHr are both primary antigens in TAO. Reactivity against a TSHr-like protein may be the first event leading to ophthalmopathy in humans with TAO and experimental mice and eye muscle damage may result from autoimmunity against G2s and Fp as a result of "antigen spreading".