OBJECTIVES: To find whether germline and somatic gain-of-function mutations of the thyrotropin receptor (TSHR) differ in location and/or mutational mechanisms, as well as to explore the degree to which these mutations are specific to TSHR compared with pituitary glycoprotein hormone receptors. METHODS: We examined the data on the TSHR website (www.unvi-leipzeig approximately innerre/TSH) supplemented with recent literature. Comparisons were also made with gain-of-function mutations of lutropin/choriogonadotropin (LH/CGR) and follicle-stimulating hormone receptors (FSHR). RESULTS: Some mutations (at residues 183, 505, 509 and 597) are exclusively germline, whereas mutations at 630 and 633 are characteristic of somatic mutations. Several residues located mainly in a mutation cluster region (619-639) are shared by both. Germline mutations are more likely to be transitions than transversions compared with somatic mutations. The lack of mutations involving deamination of CpG dinucleotides, a common mechanism for C-->T transitions, reflects the low CG prevalence in the mutable regions of TSHR. Comparison of the mutation sites with the equivalent positions in LH/CGR showed a significant difference (P<0.0001), whereas those in the mutation cluster region comprising the sixth transmembrane helix (TM6) and the adjoining third intracellular loop were concordant (P>0.90). We suggest that there is specific clustering of mutations in the juxtacytoplasmic end of TM6 in LH/CGR, a hydrophobic patch that is tightly packed with a face on TM5 whose sequences diverge from those of TSHR. CONCLUSIONS: TSHR exhibited frequent mutations outside the mutation cluster region. A role for a mutagenic environment created by the thyroid for other TSHR-specific codons cannot be discounted, nor can genetic factors, when accounting for the variation in the prevalence of TSHR-activating mutations worldwide.