New somatic mutations and WNK1-B4GALNT3 gene fusion in papillary thyroid carcinoma

Abstract

Papillary thyroid carcinoma (PTC) is the most frequent thyroid malignant neoplasia. Oncogene activation occurs in more than 70% of the cases. Indeed, about 40% of PTCs harbor mutations in BRAF gene, whereas RET rearrangements (RET/PTC oncogenes) are present in about 20% of cases. Finally, RAS mutations and TRK rearrangements account for about 5% each of these malignancies. We used RNA-Sequencing to identify fusion transcripts and mutations in cancer driver genes in a cohort of 18 PTC patients. Furthermore, we used targeted DNA sequencing to validate identified mutations. We extended the screening to 50 PTC patients and 30 healthy individuals. Using this approach we identified new missense mutations in CBL, NOTCH1, PIK3R4 and SMARCA4 genes. We found somatic mutations in DICER1, MET and VHL genes, previously found mutated in other tumors, but not described in PTC. We identified a new chimeric transcript generated by the fusion of WNK1 and B4GALNT3 genes, correlated with B4GALNT3 overexpression. Our data confirmed PTC genetic heterogeneity, revealing that gene expression correlates more with the mutation pattern than with tumor staging. Overall, this study provides new data about mutational landscape of this neoplasia, suggesting potential pharmacological adjuvant therapies against Notch signaling and chromatin remodeling enzymes.

Keywords: RNA-sequencing; gene fusions; mutations; papillary carcinomas; thyroid.

Figure 1. Point mutations, gene fusions and gene expression signatures in papillary thyroid carcinoma

A, Schematic representation of protein-altering mutations and gene fusions identified in PTC samples. Each vertical column represents a PTC patient. In the upper panel, known missense mutations and fusion transcripts associated with papillary thyroid carcinoma are shown. In the lower panel are depicted newly identified somatic mutations and other somatic alterations in cancer driver genes reported in other tumors but described for the first time in PTC (indicated by asterisks). Red boxes indicate HRAS-mutated patients or those with a RAS-like transcriptional profile. Green boxes indicate BRAF-mutated or RET/PTC patients with a BRAF-like transcriptional profile. B, Heatmap of the hierarchical clustering of differentially expressed genes between BRAF-like and RAS-like PTC samples. Black bars indicate samples with point mutations in HRAS and BRAF genes. ^# indicate samples with gene fusions.

Figure 2. New mutations identified in papillary thyroid carcinoma

A-D, the genomic localization and the exon/intron structure of each mutated gene are schematized (A, CBL; B, NOTCH1; C, PIK3R4; D, SMARCA4). In each panel, the ectropherogram shows the nucleotide variation identified by RNA-Seq, and the protein graphic representation shows the functional domains affected. In panel D, a detail of the three-dimensional structure of SMARCA4 bromodomain highlights the salt interactions among wild-type residue (colored in green) and the surrounding amino acids (colored in orange). These interactions are lost in the mutated protein (the mutated residue colored in red).

Figure 3. New WNK1-B4GALNT3 gene fusion in papillary thyroid carcinoma

A, Schematic representation of the localization of the fusion partners, WNK1 and B4GALNT3, on chromosome 12. The exons of WNK1 and B4GALNT3 genes that are involved in the fusion are indicated in red and grey, respectively. The RNA-Sequencing reads that map across the fusion breakpoint are shown in the black box. The red arrow indicates the exact fusion breakpoint. B, RT-PCR validation of the WNK1-B4GALNT3 fusion performed on the RNA of 18 PTC samples of the discovery cohort. C, Qualitative RT-PCR assay on the mRNAs of WNK1 and B4GALNT3 parent genes. Agarose gel picture shows the PTC sample with the fusion and a negative sample. D, Schematic mRNA structure of the two isoforms of WNK1-B4GALNT3 fusion gene. The electropherograms show the nucleotide sequences of the breakpoint (indicated by red arrows).

Figure 4. Genomic landscape of PTC mutations and related pathways

Co-occurrence of protein-altering nonsense and missense mutations identified in PTC patients (n=18) by RNA-Sequencing. Most relevant shared mutations in biological pathways associated to tumorigenesis are shown. Each vertical column represents a PTC patient. HRAS_Q61R, BRAF_V600E and RET/PTC patients are indicated by ^†, ^* and ^#, respectively. The severity of the amino acid change is proportional to the intensity of red and green boxes (according to MA, “Mutation Assessor”, and Sift scores).