Breast cancer (BC) is one of the most widespread forms of malignant neoplasms, next only to lung cancer and colorectal cancer. BC incidence has been growing in many parts of the world [1–4]. Early detection of the pathology and screening for BC is therefore a key task.
Suppressor genes BRCA1 and BRCA2 are important actors in regulating the signaling pathways associated with the functioning of DNA repair systems. Mutations in these genes entail an elevated risk of developing BC and some other forms of malignant tumors.
A substantial proportion of the mutations in tumors are somatic mutations, playing an important role both in the pathogenesis of sporadic BC and in the development of de novo resistance to anticancer drugs. Sporadic forms of cancer constitute, on average, 70–80% of BC cases, whereas only 10% of all the patients carry inherited mutations in the BRCA1 and BRCA2 genes [5].
The actual task of oncogenetics today is the development and improvement of approaches to the effective selection of anticancer drugs, taking into account the molecular-genetic features of tumor development.
The aim of this study was to identify the spectrum of mutations in the BRCA1 and BRCA2 genes in patients with BC by Illumina next-generation sequencing.

METHODS

Material for the study. Clinical characteristics of the patients

The collection of tumor samples for the study was taken from 66 patients with malignant breast neoplasms in hospital care at NN Blokhin National Medical Research Centre of Oncology of the Russian Health Ministry and Russian Scientific Center for X-ray Radiology of the Ministry of Health of the Russian Federation, (Moscow). The average age of the patients was 52.5 ± 9.7 years. The criteria for being included in the study were: age of 18 to 70, female, clinically verified BC diagnosis. Exclusion criteria: history of other forms of neoplasms, pregnancy. BC was staged according to TNM classification [6]. The study involved patients with stages T1–3N0–3M0–1. The study adhered to the principles of voluntariness and confidentiality. All patients provided informed consent to the study. The principal clinical characteristics of the patients are given in tab. 1.

DNA isolation and quality control. Oncogene panel sequencing

Genomic DNA was isolated from tumor tissue samples by using DNeasy Blood and Tissue Kit (Qiagen; USA) as instructed by the manufacturer. The concentration of the extracted DNA specimens was measured with a Qubit 3.0 fluorometer (Thermo Fisher Scientific; USA). The quality of the DNA samples was additionally tested by electorphoresis in 1% agarose gel with ethidium bromide.
DNA fragment libraries were prepared using NEBNext Ultra DNA Library Prep Kit for Illumina (New England Biolabs; USA). The libraries were barcoded by PCR using two reagent kits: NEBNext Ultra DNA Library Prep Kit for Illumina and NEBNext Multiplex Oligos for Illumina (Dual Index Primers Set 1, New England Biolabs; USA). DNA library quality control was done by measurements with Agilent Bioanalyzer 2100 (Agilent Technologies; USA) using High Sensitivity Kit as instructed by the manufacturer.
Coding regions of the tumor genome were enriched using MYbaits Onconome KL v1.5 Panel (Mycroarray; USA). The analysis was performed with a high-throughput genome sequencing system HiSeq 2500 (Illumina; USA) using paired 100-nucleotide reads. The samples were prepared and initiated according to Illumina protocols.

Bioinformatic processing of NGS data

Bioinformatic processing of the resultant NGS data was carried out using a previously developed algorithm [7, 8]. At first, the quality of the reads from DNA sequencing was assessed by Cutadapt software, and they were mapped to the reference genome hg19 (GRCh37.p13) by using the BWA tool (Burrows-Wheeler Aligner). Paired reads were removed by running the specialized rmdup command in the SAMtools software package. Mutations in the NGS dataset were detected by MuTect, and DNA sequences covered by at least 12 readswere considered the most significant.
The mutation abundance was defined as the proportion (%) of mutation-supporing reads at a position. The functional effect of the mutations was assessed relying on ActiveDriverDB database [9]. The mutations affecting the coded protein were visualized using the ProteinPaint application [10].

RESULTS

We analyzed DNA samples from breast tumors (n = 66) for the presence of mutations in the BRCA1 and BRCA2 genes by Illumina next-generation sequencing. Bioinformatic processing of the NGS data revealed mutations in the BRCA1 and BRCA2 genes in 39 (59.1%) out of the 66 BC patients. Altogether 78 unique genetic variants were detected in the study, including 30 mutations in BRCA1 and 48 mutations in BRCA2. Among all these mutations, 70 of the detected variants were identified as new mutations (89.7%). All the detected genetic variants are listed in the tab. 2.

The highest frequency in the analysis was demonstrated by the mutations 17:41246746:T>C in BRCA1 gene (52%) and 13:32914688:G>T in BRCA2 gene (47%). The mutation 13:32910800:A>C in BRCA2 gene occurred the most frequently among all samples, being identified in 10.7% (n = 3/28) of tumors with BRCA2 mutations. Mutations in both BRCA1 and BRCA2 were found in 11 patients with BC (16.7%; n = 66).
Annotation against databases revealed 33 mutations (42.3%) influencing the sequence of the coded protein, including 16 in BRCA1 gene and 17 in BRCA2 gene. The mutations affecting the sites of post-translational modification in proteins (PMT mutations) are shown in the figure.

DISCUSSION

Personalized targeted therapy is gaining ground in modern oncology. The development of a highly sensitive and cost-efficient approach to affordable routine diagnosis of tumors is therefore a priority task.
The “gold standard” for mutation detection today is Sanger sequencing, but its diagnostic capabilities are limited compared to next-generation genetic analysis systems. Tumor cells are histologically and genetically heterogeneous, contributing to the advantage of NGS-based techniques, which allow developing efficient bioinformatics pipelines for detecting genetic variants both in pairs of tumor and normal tissues samples and within individual biopsies containing a fraction of normal cell DNA.

Mutations in the key BC oncogenes BRCA1 and BRCA2 are among the most frequent and significant molecular aberrations, whose analysis can help in assessing the risk of tumor development, clinical prediction for BC patients, and in predicting the effectivenesss of anticancer drug therapy.
The BRCA1 gene was identified by Y. Miki et al. in 1994 by positional cloning on the long arm of chromosome 17. The second gene — BRCA2, was mapped and isolated on chromosome 13q. BRCA1 and BRCA2 are suppressor genes, characterized by the autosomal dominant inheritance pattern and high penetrance. Recent molecular studies of BRCA1 and BRCA2 have demonstrated a wide spectrum of mutations present in these genes [5].

The international COSMIC database [11] contains over 900 somatic coding mutations of the BRCA1 gene and over 1400 coding mutations of the BRCA2 gene. A substantial part of these mutations result in structural transformations modifying the function of protein products, thus undermining the capacity of repair systems to effectively fix DNA lesions. Many of the mutations in BRCA1/BRCA2 are missence mutations, where the coding sequence is altered and one functional codon is changed to another.

Having analyzed the NGS data for the BC tumors in our study by bioinformatics techniques, we identified 78 unique mutations in the genes BRCA1 and BRCA2. A majority of the mutations were found in BRCA2. According to the literature, the frequency of mutations differs notably between the genes BRCA1 and BRCA2 [5].
Further analysis using ActiveDriverDB showed that a large part of the genetic variants produce a functional effect on post-translational modification sites of the coded proteins (figure). Our study revealed 33 PMT-mutations, many of them previously unannotated. To confirm the pathogenic variants detected in the study and the status of the mutations, the research results need to be verified by Sanger sequencing using normal tissue samples.

CONCLUSIONS

Targeted next-generation sequencing appears to be the most promising approach for molecular profiling of tumors for clinical application. An integrated NGS-based analysis of mutations in the genes BRCA1 and BRCA2 in BC patients enables the identification of a greater number of mutations, including low mutant allele frequency variants, as well as genetic variants in biopsy samples with low tumor cell content. NGS-based approaches revealing mutations in the entire BRCA1 and BRCA2 coding sequence will enable a more effective identification of the patients to whom an adequate therapy with targeted anticancer drugs can be administered.