Effect of BRAF mutational status on expression profiles in conventional papillary thyroid carcinomas.

ABSTRACT

BACKGROUND: Whereas 40 % to 70 % of papillary thyroid carcinomas (PTCs) are characterized by a BRAF mutation (BRAFmut), unified biomarkers for the genetically heterogeneous group of BRAF wild type (BRAFwt) PTCs are not established yet. Using state-of-the-art technology we compared RNA expression profiles between conventional BRAFwt and BRAFmut PTCs. METHODS: Microarrays covering 36,079 reference sequences were used to generate whole transcript expression profiles in 11 BRAFwt PTCs including five micro PTCs, 14 BRAFmut PTCs, and 7 normal thyroid specimens. A p-value with a false discovery rate (FDR) < 0.05 and a fold change > 2 were used as a threshold of significance for differential expression. Network and pathway utilities were employed to interpret significance of expression data. BRAF mutational status was established by direct sequencing the hotspot region of exon 15. RESULTS: We identified 237 annotated genes that were significantly differentially expressed between BRAFwt and BRAFmut PTCs. Of these, 110 genes were down- and 127 were upregulated in BRAFwt compared to BRAFmut PTCs. A number of molecules involved in thyroid hormone metabolism including thyroid peroxidase (TPO) were differentially expressed between both groups. Among cancer-associated molecules were ERBB3 that was downregulated and ERBB4 that was upregulated in BRAFwt PTCs. Two microRNAs were significantly differentially expressed of which miR492 bears predicted functions relevant to thyroid-specific molecules. The protein kinase A (PKA) and the G protein-coupled receptor pathways were identified as significantly related signaling cascades to the gene set of 237 genes. Furthermore, a network of interacting molecules was predicted on basis of the differentially expressed gene set. CONCLUSIONS: The expression study focusing on affected genes that are differentially expressed between BRAFwt and BRAFmut conventional PTCs identified a number of molecules which are connected in a network and affect important canonical pathways. The identified gene set adds to our understanding of the tumor biology of BRAFwt and BRAFmut PTCs and contains genes/biomarkers of interest.