PAX8 and peroxisome proliferator-activated receptor gamma 1 gene expression status in benign and malignant thyroid tissues.

OBJECTIVE Genetic alterations involving the thyroid transcription factor PAX8 and the peroxisome proliferator-activated receptor gamma 1 (PPARgamma1) genes have been described in thyroid neoplasms. We investigated in a series of thyroid samples, including 14 normal, 13 hyperfunctioning tissues, 26 follicular adenomas, 21 follicular and 41 papillary carcinomas, both the frequency of the PAX8-PPARgamma1 rearrangement and the expression of the PAX8 and PPARgamma transcripts. METHODS Using RT-PCR followed by sequencing PCR products, PAX8-PPARgamma1 translocation was not detected in benign tissues nor in papillary carcinomas and was detected in 4 (19%) of 21 follicular carcinomas and in one (4%) of 26 follicular adenomas. RESULTS Specific real-time quantitative RT-PCR (Q RT-PCR) methods detected high levels of PPARgamma transcripts in follicular carcinomas presenting the rearrangement. Interestingly, the level of PPARgamma transcripts was significantly decreased in papillary carcinomas in comparison with those found in benign adenomas and follicular carcinomas. Finally, PAX8 gene expression was decreased in both papillary and follicular thyroid carcinomas, and in these tumors to the same extent in the presence or absence of the rearrangement. These alterations in both PPARgamma and PAX8 gene expression may explain the poorly differentiated histotype of follicular carcinomas harboring the translocation. Immunohistochemistry showed that nuclear PPARgamma staining was weak in normal tissues, adenomas, papillary carcinomas and in some follicular carcinomas, and strong in the follicular carcinomas positive for the PAX8-PPARgamma1 translocation, but also in some follicular tumors in which no translocation could be evidenced. CONCLUSION These observations confirm that the PAX8-PPARgamma1 translocation characterizes a subset of thyroid follicular carcinomas but is not a specific marker of carcinoma and that its frequency is lower than that initially reported. Finally, immunohistochemistry is not a reliable method for the specific detection of the translocation, that can be specifically evidenced by Q RT-PCR.