Recombinants plasmids encoding wild-type and V600E mutant BRAF were constructed as described by Board et al (44). Briefly, corresponding outer and mutant primers were used to yield half fragments with complimentary ends using wild type tissue DNA as a template (first half primer sequences: forward, 5′-CCAGGAGTGCCAAGAGAATA-3′, and reverse, 5′-CCATCGAGATTTCTCTGTAGCTAGACCA-3′; second half primer sequences, forward, 5′-TGGTCTAGCTACAGAGAAATCTCGATGG-3′, and reverse, 5′-TTTCAACAGGGTACACAGAACA-3′), with each half fragment containing a mutant base. PCR was performed in a 50 µl mixture containing 5 µl 10X PCR Buffer (Takara Bio Inc., Otsu, Japan), 1.25 U Takara TaqTM polymerase (Takara Bio Inc.), 4 µl dNTP mixture (Takara Bio Inc.), 0.5 µM primers and 5 µl DNA (Sigma-Aldrich, Merck KGaA, Darmstadt, Germany). The reaction procedure was as follows: Initial denaturation at 95°C for 5 min, followed by 35 cycles at 95°C for 15 sec, primer annealing at 53°C for 20 sec, and 72°C for 50 sec, final extension took place at 72°C for 5 min. The two half fragment products were mixed equally as a template for the second round PCR. The second round of PCR used inner nested primers (forward, 5′-AGCATCTTCATTCCAATGAAGAGCC-3′, and reverse, 5′-CATCCACAAAATGGATCCAGACAAC-3′. The second round was performed in a 50 µl mixture containing 5 µl 10X PCR Buffer (Takara Bio Inc.), 1.25 U Takara TaqTM polymerase (Takara Bio Inc.), 4 µl dNTP mixture (each 2.5 mM; Takara Bio Inc.,), 0.5 µM primers and 5 µl template. The thermocycling conditions included initial denaturation at 95°C for 5 min, followed by 40 cycles of 95°C for 15 sec, 50°C for 30 and 72°C for 50 sec, the final extension took place at 72°C for 5 min. Self-priming of the complementary half fragments and the subsequent amplification created a final product harboring the mutant base. The products were ligated into the pMD19 plasmid (Takara Bio Inc., Otsu, Japan), and recombinants containing mutant alleles were produced and confirmed by sequencing performed by Sangon Biotech Co., Ltd., (Shanghai, China). The sequencing machine used was ABl-PRISM 3,730 (Applied Biosystems; Thermo Fisher Scientific, Inc., Waltham, MA, USA) and the analysis software was DNASTAR 5.0 (DNASTAR, Inc., Madison, WI, USA). Recombinant plasmid DNA was extracted using a Tiangen Plasmid DNA kit (Tiangen Biotech Co., Ltd., Beijing, China). As a positive control, the recombinants were mixed with an equal amount of human genomic DNA (Sigma-Aldrich, Merck KGaA, Darmstadt, Germany).

In total, 53 patients with melanoma treated at the Department of Dermatology, The Third Affiliated Hospital of Sun Yat-sen University (Guangzhou, China), were enrolled in the present study from March 2011 to December 2012. Unrelated patients diagnosed with melanoma were included. These samples were all biopsies. The median age of patients was 68 years (age range 39-84 years). A total of 33 cases (62%) were females, and 20 cases (38%) were males. A total of 10 patients (19%) exhibited Clark level II disease (penetration of melanoma into the second layer of the skin, the dermis) (45), and 18 patients (34%), 24 patients (45%) and 1 patient (2%) exhibited III, IV and V stage disease, respectively. The study protocol was approved by the Ethics Committee of The Third Affiliated Hospital of Sun Yat-sen University, and written informed consent was obtained from all patients. DNA was extracted from 10% formalin-fixed that was fixed at room temperature for 24 h, paraffin-embedded melanoma samples using a QIAamp DNA FFPE Tissue kit (Qiagen GmbH, Hilden, Germany). Briefly, formalin-fixed, paraffin-embedded blocks containing the maximum number of tumor-rich areas from the patients were selected and sliced into 3 5-µm thick sections. Each section was processed by proteinase K digestion at 56°C for >16 h, and the obtained lysate was loaded onto a QIAamp column. Following 2 washes, the DNA was eluted with 100 µl ddH₂O. The extracted DNA was kept at −20°C until it was used for PCR analysis.

The RFLP-ARMS TaqMan PCR assay described in the present study was a one-step PCR that used a mutation-enriching reaction and ARMS primer genotyping process to selectively eliminate wild-type genes and detect the mutant alleles. For the BRAF V600E mutation, a restriction enzyme was used to digest the wild-type genomic DNA, thereby enriching the mutant allele, and then a pair of RFLP primers: Forward, 5′-AGCATCTTCATTCCAATGAAGAGCC-3′; and reverse, 5′-CATCCACAAAATGGATCCAGACAAC-3′, designed to amplify a 400–500 bp fragment containing the mutant allele [melting temperature (Tm), 65°C], were used. The ARMS primers were designed to selectively amplify a mutant allele with a lower Tm (60°C) with the following respective sequences: Forward, 5′-TAGGTGATTTTGGTCTAGCTACACA-3′ (mismatched base is underlined); and reverse, 5′-CCACAAAATGGATCCAGACAAC-3′. To improve the specificity of the ARMS primers, an additional mismatch base was introduced at the penultimate nucleotide of the mutation site (G was replaced by C), based on the principles described by Newton et al (19). Mutant allele enrichment and genotyping were performed in a single tube via a 3-phase reaction: i) 65°C for 30 min, allowing the restriction enzyme, TspRI, to cut the wild-type DNA; ii) enrichment of the mutant allele with the RFLP primers and thermocycling conditions of 95°C for 10 min, followed by 5 cycles of 95°C for 15 sec, 65°C for 20 sec and 72°C for 60 sec; and iii) selective amplification of the mutant allele with the ARMS primers, and 40 cycles of 95°C for 15 sec and 60°C for 35 sec (fluorescence collection). During the second phase, the ARMS primers were unable to bind at the higher temperature (65°C), while in the third phase, the RFLP primers did not function as the 400–500 bp product was incompletely synthesized during the 35 sec elongation phase.

The PCR reaction mixtures contained 12.5 µl TaqMan universal PCR MasterMix (Applied Biosystems; Thermo Fisher Scientific, Inc., Waltham, MA, USA), 0.2 µM probe, 0.25 µM primer (each), 3 µl DNA, 5 IU TspRI (New England Biolabs, Inc., Ipswich, MA, USA), and ddH₂O to 5 µl. qPCR was performed using an ABI7300 (Applied Biosystems; Thermo Fisher Scientific, Inc., Waltham, MA, USA). DNA samples were extracted from all clinical tissue samples using a QIAamp DNA FFPE Tissue kit (Qiagen GmbH, Hilden, Germany), and tested using our RFLP-ARMS TaqMan PCR detection system.

To determine the specificity of the proposed RFPL-ARMS TaqMan PCR method, reactions were performed with 2–200 ng of wild-type genomic DNA (Sigma-Aldrich, Saint Louis, Missouri, USA) per reaction and assessed the inefficiency caused by extension from wild-type DNA. An internal control assay was used to assess the total DNA concentration from 2–200 ng in each sample. The forward primer was designed to begin at c.1798G; it amplified wild-type and V600E mutant BRAF using the aforementioned reverse primer and probe. The change in the threshold cycle (ΔCq) [ΔCq=(mutation Cq)-(control Cq)] was defined for each sample. The reactions were performed five times for each DNA concentration, and each reaction was repeated in triplicate to define a cut-off ΔCq value (46).

The wild-type and mutant plasmid DNA samples were diluted 10-fold. The effective copy number of plasmids was obtained by comparing the Cq of the diluted samples with that obtained from Human Random Control DNA Panels (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) with a known concentration of 100 ng/µl. In a high-quality DNA sample, there was an average of 1 set genomic DNA per 3 pg.

To assess the sensitivity of our assay for the BRAF V600E mutation, it was compared with the ARMS TaqMan PCR protocol without a removed enzyme digestion step. The mutant-encoding plasmid (50,000, 5,000, 500, 50 or 5 copies) was mixed with wild-type genomic DNA (30 ng/µl), corresponding to 80, 8, 0.8, 0.08 and 0.008% mutation rate. For quantification, a standard curve was generated by plotting the Cq cycle numbers against the log of each corresponding DNA copy numbers for the known standards. The linear correlation coefficients (R²) and slopes were calculated using ABI7300 software (v.1.3; Applied Biosystems; Thermo Fisher Scientific, Inc.).

All samples were re-analyzed by PCR sequencing, described as following. Kimura et al (47) previously demonstrated that direct sequencing failed to yield satisfactory results from samples containing mixtures of wild-type and mutant DNA. In the present study, the existing mutation-enriched PCR sequencing method was improved by restriction enzyme selectively cutting wild type alleles and leaving the mutant alleles enriched, increasing the mutation rate. This method was adapted for the detection of the BRAF V600E mutation. This method involved a first PCR amplification step, enzymatic digestion (to remove the wild-type DNA), a second PCR step for mutant enrichment and a final sequencing step. PCR was performed in a total volume of 25 µl containing 12.5 µl 2X Gold Fast PCR mix (Tiangen Biotech Co., Ltd., Beijing, China), 0.4 µM each primer (round 1: forward, 5′-AGCATCTTCATTCCAATGAAGAGCC-3′, and reverse, 5′-CATCCACAAAATGGATCCAGACAAC-3′; round 2: forward, 5′-CATAATGCTTGCTCTGATAGGA-3′, and reverse, 5′-CCACAAAATGGATCCAGACAAC-3′) and 3 µl DNA (for the first round of PCR) or 3 µl digested product (for the second round of PCR). The cycling conditions consisted of 95°C for 10 min, followed by 20 (round 1) or 35 (round 2) cycles of 95°C for 15 sec, 58°C for 20 sec and 72°C for 30 sec, with a final extension at 72°C for 7 min and a final hold at 4°C. Following the first round of PCR, digestion of the wild-type product was performed in a 50 µl volume containing 10 µl first-round PCR product, 5 µl 10X CutSmart Buffer (New England Biolabs, Inc., Ipswich, MA, USA) and 5 IU TspRI (New England Biolabs, Inc.) at 65°C for 30 min. All of the obtained second-round PCR products were sequenced by Sangon Biotech Co., Ltd. (Shanghai, China). The sequencing machine used was ABl-PRISM 3730 (Applied Biosystems; Thermo Fisher Scientific, Inc.) and the analysis software was DNASTAR (v.5.0; DNASTAR, Inc.).

Fisher's exact test was used to compare the sensitivity difference of the two methods (RFLP-ARMS TaqMan PCR and PCR sequencing) in the detection of clinical samples with V600E mutation rates <5%. P<0.05 was considered to indicate a statistically significant difference. SPSS for Windows was used (version 23.0; IBM Corp., Armonk, NY, USA).

To assess the specificity of the RFPL-ARMS TaqMan PCR method, RFLP-ARMS TaqMan PCR was performed using wild-type genomic DNA (2–200 ng/µl). By testing all the wild-type genomic DNA (2–200 ng) cases, the cut-off ΔCq value was determined to be 3 Cq below the lowest ΔCq value observed; the final cut-off ΔCt value was determined to be 14. The ΔCq values were then calculated as the difference between the mutant and control Cq values. If the difference was smaller than the cut-off ΔCt value, the sample was classified as positive (a larger ΔCq reflected the presence of fewer mutant alleles). If the difference was larger than the cut-off point, the sample was classified as mutation-negative or beyond the limits of detection.

To determine the minimal detection limit of the RFPL-ARMS TaqMan PCR method, a mimic human genomic DNA panel containing the mutant plasmid and normal wild-type human genomic DNA was used. The results revealed that the mutation-enriched PCR sequencing method exhibited increased sensitivity compared with direct PCR sequencing (Fig. 1), and was able to identify mutations making up ~1% of the total genomic DNA content. Furthermore, the RFLP-ARMS TaqMan PCR assay was demonstrated to be a sensitive and practical method to screen for the BRAF V600E mutation. As indicated in Fig. 2, this RFLP-ARMS TaqMan PCR method allowed the detection of mutants within mixed samples containing <0.01% of the V600E mutation (corresponding to <10 copies), while 0.8% V600E mutations were detected by ARMS TaqMan PCR, but 0.08 and 0.008% gave no amplification signal (Fig. 3).

Mutation analysis of clinical samples. RFLP-ARMS TaqMan PCR and PCR sequencing was performed on 53 clinical samples. Among them, 21 samples were identified to be positive for the BRAF V600E mutation by RFLP-ARMS TaqMan PCR, while only 18 positive samples were identified by PCR sequencing (Table I). The three discordant samples were then analyzed by Droplet Digital PCR (ddPCR; QX200, Bio-Rad Laboratories, Inc., Hercules, CA, USA). The results indicate that the mutation frequency of the three samples was 3.8, 1.2 and 0.6%, respectively, and the results (3.8, 1.2 and 0.6%) indicate that the samples were positive for the BRAF V600E mutation, which verified the sensitivity of RFLP-ARMS TaqMan PCR. The result of Fisher's exact test revealed a significant difference between the two methods in detecting low-frequency mutations (<5%). Future experiments will use a larger sample size, which may result in increased statistical significance.