An activating mutation in the BRAF gene is the most common genetic alteration in papillary thyroid carcinomas (PTCs). The mutation in PTCs is almost a c.1799T>A transversion, resulting in a p.V600E amino acid substitution (BRAFV600E). Here, we report a novel complex BRAF mutation identified in 4/492 Japanese PTC cases (0.81%). The mutation was comprised of one nucleotide substitution at position 1798, followed by an in-frame insertion of three nucleotides, c.1798delinsTACA in Exon 15, resulting in p.V600delinsYM. In silico three-dimensional protein structure prediction implied altered kinase activity of this mutant. In vitro kinase assay and western blotting revealed that this mutation conferred high kinase activity on the BRAF protein, leading to constitutive activation of the MAPK signaling pathway. The mutation also showed high transforming ability in focus formation assay using NIH3T3 cells. The degree of all the functional characteristics was comparable to that of BRAFV600E, and treatment with a BRAF inhibitor Sorafenib was also equally effective in this mutant. These findings suggest that the novel BRAF mutation, BRAFV600delinsYM, is a gain-of-function mutation and plays an important role in PTC development.
Papillary thyroid carcinoma (PTC) represents the most common malignant tumor in thyroid. PTCs have particular genetic alterations leading to the activation of the mitogen-activated protein kinase (MAPK) signaling pathway. Those include RET/PTC rearrangement and point mutations in BRAF and RAS genes. They are detected in ∼60–70% of all PTCs and rarely overlap in the same tumor.1–3 The lack of coexistence of these mutations provides strong genetic evidence for the importance of constitutive activation of the MAPK cascade for PTC carcinogenesis.
The BRAF mutation is the most prevalent genetic alteration in PTCs, ranging 29–83%, 44% on average.4 The mutation in PTCs is almost a thymine-to-adenine transversion at nucleotide 1799 (c.1799T>A) in Exon 15, resulting in a valine-to-glutamic acid substitution at amino acid 600 (p.V600E). However, several other mutations affecting the BRAF gene have been found in rare cases (Table 1).
Table 1. Summary of rare BRAF mutations identified in PTCs
Most of oncogenic mutations in the BRAF protein are located in two regions: activation loop (A-loop) and phosphate-binding loop (P-loop). Activating mutations in these areas are believed to disrupt their hydrophobic interactions that maintain the inactive conformation of the BRAF protein, resulting in its conversion into constitutively active form.19 Not only activating mutations (e.g., p.V600E) but also mutants with impaired activity (e.g., p.G466E/V and p.G596R) are also associated with cancers. Intriguingly, the latter do activate MEK indirectly by activating CRAF, possibly via dimerization and an allosteric or transphosphorylation mechanism, leading to the constitutive activation of the MAPK signaling.19 This mechanism might have clinical implications because promoted dimerization of CRAF and BRAF could lead to resistance to RAF inhibitors.20, 21
In this study, we report a novel complex BRAF mutation, BRAFV600delinsYM, identified in 4 out of 492 Japanese PTC cases (0.81%). We also performed its functional characterization, and it showed constitutively active kinase function and transforming ability, suggesting that it is a new PTC oncogene that activates the MAPK signaling pathway as does BRAFV600E.
Material and Methods
A total of 509 sporadic PTC samples were collected at Kuma Hospital (Kobe, Japan). Histological diagnosis was performed by a thyroid pathologist (MH). All patients had no history of radiation exposure. The protocol was approved by the ethics committees of Nagasaki University and Kuma Hospital.
DNA extraction and BRAF status screening
DNA was extracted from formalin-fixed paraffin-embedded PTC tissues using QIAamp DNA mini kit (QIAGEN) according to the manufacturer's protocol. BRAF mutations were analyzed by direct genomic DNA sequencing. First, PCR amplification was done using KOD FX (TOYOBO). PCR products were then treated with ExoSAP-IT PCR clean-up reagent (GE Healthcare), and Sanger-sequencing was performed with Big Dye Terminator sequencing kit version 3.1 (Applied Biosystems) on an ABI3100 automated sequencer (Applied Biosystems). To distinguish each allele, amplified PCR products were cloned into pGEM-T Easy vector (Promega), and sequencing was then performed. Primer sequences used for both PCR amplification and direct sequencing were: BRAFi14F, 5′-ACATACTTATTGACTCTAAGAGGAAAGATGAA-3′ and BRAFi15R, 5′-GATTTTTGTGAATACTGGGAACTATGA-3′. For sequencing cloned DNA, standard SP6 and T7 primers were used.
Cell lines and reagents
Murine fibroblast NIH3T3 cell line was maintained in Dulbecco's modified Eagle medium (DMEM) (Wako Pure Chemicals) supplemented with 5% calf serum and 1% penicillin/streptomycin (Wako Pure Chemicals). The 293FT packaging cell line (Invitrogen) was maintained according to the manufacturer's protocol. COS7 cell line was maintained in DMEM supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. Sorafenib tosylate (Selleck Chemicals) was dissolved in dimethylsulfoxide (DMSO) at a stock concentration of 5 or 10 mM.
BRAF protein expression
Full length coding portion of V600E mutant BRAF (BRAFV600E) cDNA was subcloned using pLenti6/V5 directional TOPO cloning kit (Invitrogen) according to the manufacturer's instruction. Primers used for subcloning were: BRAFV600EF, 5′-CACCATGGCGGCGCTGAGCGGTGGC-3′ and BRAFV600ER, 5′-GTGGACAGGAAACGCACCATATCCCCCTGC-3′. Wild-type (BRAFwt) and V600delinsYM mutant BRAF (BRAFV600delinsYM) were generated using QuikChange XL site-directed mutagenesis kit (Agilent Technologies). Primers used for generating wild-type and V600delinsYM mutants were: BRAFwtF, 5′-GGTCTAGCTACAGTGAAATCTCGATGGAG-3′ and BRAFwtR, 5′-CTCCATCGAGATTTCACTGTAGCTAGACC-3′; BRAFinsF, 5′-GGTCTAGCTACATACATGAAATCTCGATGGAG-3′ and BRAFinsR, 5′-CTCCATCGAGATTTCATGTATGTAGCTAGACC-3′, respectively. The integrity of coding regions was confirmed by sequencing. For transient transfection, we used Lipofectamine 2000 (Invitrogen) following the manufacturer's instruction. For stable transduction, recombinant lentiviruses were produced by introduction of the plasmid together with ViraPower lentiviral packaging mix (Invitrogen) into the packaging cell line 293FT using Lipofectamine 2000 and concentrated using PEG-it virus precipitation solution (System Biosciences). Viral titers were determined using NIH3T3 cells.
Protein samples were separated with SDS-PAGE and transferred onto PVDF membrane (Millipore). After incubation with an appropriate primary antibody, the antigen–antibody complexes were visualized using HRP-conjugated secondary antibody (Cell Signaling Technology) and a chemiluminescence system (Nacalai Tesque) Detection was performed using a LAS3000 imaging system (Fujifilm). Primary antibodies were obtained from the following sources: anti-phospho-ERK (Thr202/Tyr204), anti-ERK, anti-phospho-MEK (Ser217/221), and anti-MEK from Cell Signaling Technology Japan; anti-BRAF and anti-β-actin from Santa Cruz Biotechnology; anti-V5 from Invitrogen.
In vitro kinase assay
Cells were lysed in cell lysis buffer (20 mM Tris-HCl pH 7.5, 1 mM EDTA, 150 mM NaCl, 0.5% Triton-X, 50 mM NaF, 10 mM Na pyrophosphate, and 1 mM Na3VO4) supplemented with Complete protease inhibitor cocktail (Roche). V5-tagged BRAF protein was immunoprecipitated using agarose-conjugated anti-V5-tag polyclonal antibody (MBL, Nagoya, Japan), and the precipitates were washed twice with washing buffer (20 mM Tris-HCl pH 7.5, 1 mM EDTA, 150 mM NaCl and 0.1% Triton-X). BRAF kinase activity was measured using B-RAF kinase assay kit (Millipore) followed by manufacturer's instruction with minor modification. Briefly, kinase reaction was carried out in the presence of ATP and recombinant MEK substrate at 16°C for 10 min. Phosphorylation level was measured by Western blotting using anti-phospho-MEK antibody (supplied in the kit). Precipitated BRAF protein level was also measured using anti-V5-tag HRP-DirecT antibody (MBL).
Focus formation assay
NIH3T3 cells were infected with the appropriate lentivirus in the presence of 4 μg ml−1 polybrene (Sigma–Aldrich). The infected cells were cultured for 2 weeks in the presence of 5 μg ml−1 Blasticidin (InvivoGen), and the number of transformed foci was counted.
For the quantitative estimation of cell morphology, image analysis was done with Image J software (NIH). The spindle index was calculated according to the formula: a/b, where a is the longest cell diameter and b is the longest diameter perpendicular to a.
In silico protein structure prediction
The three-dimensional (3D) structure was predicted by the Protein Homology/analogY Recognition Engine V 2.0 (Phyre2)22 and visualized using PyMOL software.
Differences between groups were examined for statistical significance with ANOVA followed by Tukey's post test. A p value not exceeding 0.05 was considered statistically significant.
Results and Discussion
A novel BRAF mutation detected in PTC
In the course of our previous study analyzing the BRAF mutation and FOXE1/NKX2-1 polymorphisms in the Japanese PTC cases,23 we identified a novel complex mutation in the BRAF gene. By direct genomic DNA sequencing, the chromatogram showed an overlapping pattern starting at base position 1798 (guanine in wild-type) (Fig. 1), which most likely indicates heterozygous deletion or insertion. To separate each allele, PCR product was cloned into pGEM-T Easy vector and sequenced. As illustrated in Figure 1, in the mutated allele, the guanine at 1798 was substituted by a thymine followed by an in-frame insertion of three nucleotides, adenine–cytosine–adenine (ACA), resulting in a valine-to-tyrosine substitution at codon 600, followed by an insertion of one methionine (p.V600delinsYM). The other allele was wild-type. To the best of our knowledge, this complex mutation has not been reported in any type of human disease.
The rare BRAF mutations reported in PTCs are summarized in Table 1. We screened a total of 509 Japanese PTC cases, and the results of direct genomic sequencing were available in 492 cases. Of the 492 cases, we found 4 cases harboring the same heterozygous BRAFV600delinsYM mutation (0.81%). The number of screened cases was the largest among all the previous studies, and the BRAFV600delinsYM mutation may be relatively frequent among the rare mutations (Table 1). Note that the BRAFV600E mutation was identified in 388 cases and we did not find any other type of mutation in Exon 15 of the BRAF gene such as p.K601E. All four cases with BRAFV600delinsYM were classic papillary subtype of PTC, and two had lymph node metastasis. No specific clinicopathological characteristics were found in these cases.
Functional characterization of BRAFV600delinsYM
First, we estimated the effect of the mutation on the 3D structure of the BRAF protein, especially in the A-loop using the protein structure prediction software Phyre2, because the hydrophobic interaction between the P-loop and the A-loop is believed to maintain inactive conformation and its disruption may lead to constitutive activation of BRAF. The Phyre2 server uses a library of known protein structures and enables highly accurate prediction. A detailed protocol was described elsewhere.22 As shown in Figure 2a, the A-loop was predicted to be apparently dissociated from the P-loop in BRAFV600delinsYM as well as in BRAFV600E. This result suggests that the BRAFV600delinsYM mutation may have an effect on kinase function by altering the protein structure, especially the association between P-loop and A-loop.
Next, we sought to determine the effects of BRAFV600delinsYM on the MAPK signaling pathway. BRAFwt, BRAFV600E and BRAFV600delinsYM were stably expressed in NIH3T3 cells using a lentivirus expression system, and the activation status of the MAPK pathway was evaluated by immunoblotting using phospho-specific antibodies. Compared with negative control (LacZ expression) and BRAFwt, the phosphorylation levels of MEK and ERK were slightly higher in the BRAFV600E and in the BRAFV600delinsYM-expressing cells (Fig. 2b), indicating that the MAPK pathway was constitutively activated in these stable transductants.
As described above, some BRAF mutants (e.g., p.G466E/V and p.G596R) have impaired kinase activity but can induce activation of MEK via CRAF, resulting in activation of the MAPK pathway.19 To evaluate BRAFV600delinsYM kinase activity, we performed in vitro kinase assay. V5-tagged each BRAF or LacZ protein was transiently overexpressed in COS7 cells, and immune complexes against V5 were subjected to the kinase assay. BRAFV600delinsYM robustly induced MEK phosphorylation in vitro, which was comparable to that by BRAFV600E (Fig. 2c), suggesting that BRAFV600delinsYM is a high activity mutant.
We also examined the effect of a BRAF inhibitor on this mutant. Sorafenib is the most commonly and widely used nonspecific BRAF inhibitor. It also crossreacts with CRAF, VEGFR, PDGFRβ and FGFR1, and has already been approved for the treatment of renal cell carcinoma and hepatocellular carcinoma.24 As shown in Figure 2d, both MEK and ERK phosphorylation levels were reduced by the treatment with Sorafenib in a dose-dependent manner. These results indicate that Sorafenib is equally effective in BRAFV600delinsYM and in BRAFV600E. Note that this does not ensure that all types of BRAF inhibitors would be effective against this mutant.
To determine the transforming potential of BRAFV600delinsYM, we stably transduced NIH3T3 cells and performed focus formation assay. BRAFV600delinsYM dramatically changed cell morphology into spindle-like shape as did BRAFV600E, whereas LacZ and BRAFwt did not (Fig. 2e). As shown in Figure 2f, a number of foci were formed in cells expressing BRAFV600E and BRAFV600delinsYM. In contrast, the expression of BRAFwt or LacZ did not develop transformed foci. Thus, like BRAFV600E, BRAFV600delinsYM most likely represents a driver of tumorigenesis with high kinase activity.
In conclusion, the present study reports for the first time a novel complex BRAF mutation, BRAFV600delinsYM. The functional characterization demonstrates its high kinase activity and transforming potential, suggesting that this mutation plays an important role in PTC tumorigenesis. Analysis of large series of PTCs in patients of different ethnicities would be interesting to clarify the incidence of this particular mutation.