Celastrol (Tripterygium wilfordii) has been reported to play an important role in anti-cancer tumorigenesis such as colon cancer. However, the mechanism of its role in thyroid cancer remains unclear. The aim of this study was to investigate the killing effect and underlying mechanisms of celastrol in different types of thyroid cancer cell lines. After the treatment of celastrol at different concentrations (1, 5 and 10 μM) for 24 h, the cell shape of papillary carcinoma cell BCPAP and anaplastic carcinoma cell 8505C cells turned round at a low concentration of 1 μM, while follicular carcinoma cell FTC-133 showed morphological changes at 5 and 10 μM of celastrol treatment for 24 h. Then IC50 values of BCPAP, FTC-133 and 8505C cells were determined by MTT assays to further determine the growth-inhibition effect of celastrol. The IC50 values of BCPAP, FTC-133 and 8505C cells were 4.7, 8.7 and 5.9 μM, respectively, indicating that celastrol had a good killing effect on different kinds of thyroid cancer cells. We selected BCPAP cells to further explore the mechanism of celastrol in thyroid carcinoma. The proportion of JC-1 monomer, a sensitive marker of mitochondrial membrane potential, increased from 10.3% to 27.1% and 36.2% after celastrol (2.5 and 5 M) treatment for 24 h in BCPAP cells. PDHA2 is a member of the Pyruvate dehydrogenase complex (PDHc), an important rate-limiting enzyme family involved in mitochondrial metabolic glycolysis to the tricarboxylic acid (TCA) cycle. After 24 h of treatment with celastrol (1.25, 2.5 and 5 M) on BCPAP cells, the protein expression of PDHA2 decreased significantly, suggesting that celastrol may kill thyroid cancer cells through targeting PDHA2. Altogether, these findings elucidated the potential anti-tumor mechanisms of celastrol and shed a light on a prospective therapeutic target for thyroid cancer treatment.