Background

Breast cancer (BC) is the most commonly diagnosed type of cancer in women, accounting for 25% of all cancer cases in the world; with much more cases recorded in developing countries than developed ones. In 2012, 1.67 million cases of BC resulted in 522,000 deaths [1,2,3]. In Africa, 324,000 deaths were reported to be caused by BC [1, 2]. The predisposition to BC appears to be affected by several factors, one of them is the high-risk BC gene mutation in BRCA2 (OMIM: 600185) (Gene ID: 675) (RefSeqGene: NG_012772) [4]. Although the incidence rate of gene mutation in BRCA2 is low, it is associated with a high lifetime risk of BC [5, 6]. This lifetime risk is variable among different population [7,8,9,10]. BRCA2 is believed to be the primary cause of 5 to 10% of all cases of BC [11]. About 45% of women, who inherited a defective BRCA2 allele, will develop BC when they reach the age of 70 [12, 13]. Mean age at onset of BC for BRCA2 mutation carriers is reported to be 42.8 years [14].

The human BRCA2 gene contains 27 exons, among which exon 11 is the largest one. The coding sequence (RefSeq transcript mRNA: NM_000059) was determined to be 11,385 bp, which codes for a protein of 3418 amino acids (Uniprot: P51587) (RefSeq protein: NP_000050) [15]. A study conducted in Central Sudan from 2001 to 2002 concluded that this gene plays a role in the etiology of BC [16]. In addition, in a genetic analysis performed on secondary school female students in Northern Sudan, some variants were detected in two groups free of BC, one with a family history of BC and the other without familial risks. Two BRCA2 mutations were reported in the group without a family history [17].

It is known that the majority of deleterious mutations in BRCA2 are either a frameshift or nonsense mutations [14, 18, 19]. The nonsense mutations have been reported more within exon 11 of early onset BC cases with high pathogenicity [14, 18]. It is found that about 90% of reported BRCA2 mutations are protein truncating [20]. In addition, the formation of nonsense-mediated RNA decay -as premature terminating inactivation codon- could lead to the production of a toxic partial protein [14]

Heterozygosity of BRCA2 mutations was found to be associated with a distinctive phenotype, which could lead to BRCA2 tumorigenesis, as altered heterozygous BRCA2 does not function well and the wild allele alone is not enough to maintain genomic stability. In other cases, it was suggested to be haploinsufficient. Furthermore, BRCA2 monoallelic carrier mutations were detected in patients with pancreas and breast cancer [21, 22].

Etiologically, scientific literature from African countries showed that reproductive factors more commonly associated with the development of BC are early menarche, pregnancy, and multiparity [23]. The situation is globally similar; as early menarche, late menopause, carriers of BRCA2 damaging variants, and early pregnancy before age of 30 years confer high-risk conditions for BC [24].

Unfortunately, the scientific articles from African countries lacked data about the risk conferred by familial cases as it has not been well investigated, although some studies suggested its etiological companion [16, 23]. This study aimed to screen BRCA2 mutations, taking into consideration the biggest region in the gene, exon 11, to find out and investigate variants or single nucleotide polymorphisms (SNPs) among known BC patients.

Methods

Study area

This study was carried out in Khartoum state at the Radiation and Isotope Center in Khartoum (RICK), which is one of the only two oncology centers in Sudan, and it provides oncological services for people from all parts of Sudan.

Sampling

Out of all Sudanese female patients diagnosed with BC (45 patients) attending RICK during March 2015, 10 patients were selected randomly for genetic sequencing and analysis. Four healthy subjects with no family history of BC and another one diagnosed with essential thrombocythemia who are free of BC have been added as controls. Blood specimens were collected using EDTA-vacutainer tubes from the selected patients and controls. The specimens were preserved at −20 °C.

Ethical considerations

All patients were informed and consented to participate in the study before collecting the samples. All patients were consented to publish the results of the study. Ethical approval was obtained from the ethical committee of Sudan Ministry of Health-Khartoum state.

DNA extraction

For both patients and controls, DNA was extracted by Salting out technique according to the published protocol [25]. In addition, we added proteinase K at 56 °C to enhance white cells membrane breakdown. After 1 h, the DNA was extracted with concentration of 30 ng/ul, dissolved in 100 ul Tris-EDTA (TE) Buffer, and kept for overnight at 4 °C, then preserved at −20 °C until use.

PCR amplification

Forty-five patients and five control samples were subjected to amplification using three primers sets (A, B and C) targeting three regions within BRCA2 gene exon 11 as described in (Table 1). This study focused only on the product of the second primer set (primer B) based upon stability and quality of this primer [26]. Primers were synthesized and purchased from Macrogen Incorporation (Seoul, South Korea). Annealing temperature was adjusted using Maxime PCR PreMix Kit i-Taq 20 μl (INTRON Biotechnology, South Korea) on several runs of PCR. The adjusted temperatures are described in (Table 1). Amplification for the targeted regions was done after addition of 15 ul Distilled water, 3 ul sample DNA and 1 ul of each forward and reverse to the ready-to-use master mix volume. PCR mixture was subjected to an initial denaturation step at 96 °C for 5 min, followed by 35 cycles of denaturation at 96 °C for 30 s, primer annealing at 50 °C for 30 s, followed by a step of elongation at 72 °C for 60 s, the final elongation was at 72 °C for 10 min [26]. The PCR products were checked and analyzed by 2% agarose gel electrophoresis at 100 V for 30 to 45 min and then bands were visualized by automated gel photo documentation system (Fig. 1). Only 10 patients and five controls yielded sufficient quality bands, and were subsequently selected for sequencing by the Sanger sequencing technique.

Table 1 List of primers used to amplify BRCA2 gene selected regions
Full size table
Fig. 1
figure 1

Illustrates PCR amplification results of the three tested regions (A, B & C) on 2% gel electrophoresis. MW: DNA ladder’s molecular weight where 100 bp was used. C7 to C1 lanes indicate primer C bands. A1 indicates primer A band. B7 down to B1 Lanes indicate primer B bands

Full size image

Sequencing of BRCA2 gene

Sanger sequencing was performed for the PCR products. Both DNA strands were sequenced by Macrogen Company (Seoul, South Korea).

Bioinformatics analysis

For each sample, the two purified chromatogram (forward and reverse) nucleotide sequences were viewed and checked for quality by FinchTV program version 1.4.0 [27]. The NCBI Nucleotide database was searched for reference sequences. BRCA2 nucleotide sequence (NM_000059.3) was obtained and all regions were analyzed accordingly [10]. Additional high similarity sequences (AY436640.1) and (X95161.1) were obtained from NCBI database and were added as control sequences using nucleotide Basic Local Alignment Search Tool (BLAST) [28]. Any apparent changes within the tested sequences were noticed through multiple sequence alignment using BioEdit software [29]. All sequences were translated into amino acid sequences using online Expasy translate tool [30]. The resulted amino acid sequences were compared all together using BioEdit software.

SNP prediction

SIFT-software was used to check for the effect of SNPs on the protein; whether they are damaging or not [31]. Also, SNPs structural and functional impact on resultant protein was predicted by PolyPhen-2; which performs searches in several protein structure databases for 3D protein structures, multiple alignments of homologous sequences and amino acid contact information. [32]

Project hope was used to analyze the structural and conformational variations that have resulted from single amino acid substitutions corresponding to the single nucleotide substitutions [33], then the protein stability was assessed by I-Mutant [34], In addition to web-based applications for rapid evaluation of the disease-causing potential of DNA sequence alterations called MutationTaster2 [35]

Results

Study population characteristics

Patient characteristics, clinical and histological parameters

Forty-five women with BC, who attended RICK-center for treatment and follow-up, were selected for the study, their age ranged between 27 to 80 years (mean age was 45.9 years). Out of 45 patients, 25 (55.6%) were premenopausal women (Early onset cases) with a mean age of 36.6 years. On the other hand, late onset cases - who were 46 years or more - had a mean age of 57.4 years. The majority of women in the study were multiparous 30/45 (66.6%), with an average number of 4.1 parities. Patients were from 17 tribes, Ja’alya, Shaygeya, and Dnagla were the most frequent tribes (Table 2). Familial history of any type of cancer was found in 11 cases; of which six cases had BC in the family. Abortion was detected in 10 cases (22.2%), with an estimated frequency of 1–5 times. Among the married cases (88.8%), three cases were married at less than 20 years of age.

Table 2 Patients demographic and characteristics
Full size table

Available histotype data showed that ductal tumors were the predominant type (detected in 22 cases (48.8%)). Lobular and mucinous were reported in 5 and 2 cases respectively. Papillary adenocarcinoma was detected in only one patient, as a secondary deposit in bone. The right side was affected by the disease in 20 patients (44.4%). Four patients had bilateral disease (Table 2).

Mean age at diagnosis in the group selected for DNA sequencing was 39 years (27 to 57 years). Nine patients were multiparous (mean of parity was 3.5). In this group, while the right-side was predominantly affected, one patient had bilateral breast involvement. Cancer grades were between II to III. Clinical staging showed lymph nodes involvement in five cases. Distal metastasis was noted in the liver in one patient; while bone and lung involvement were documented in another case. Control individuals were free of BC and free of family history involvement. The youngest patient within the study was 27 years old and was the only case free of lymphatic involvement (Table 3).

Table 3 The highly purified Breast Cancer Patients demographic, clinical, histological parameters with the nonsense mutation
Full size table

Bioinformatics result analysis

The sequencing data was checked for consistency and quality, and one patient’s sequence has been excluded for inconsistency.

By using the multiple sequence alignment tool BioEdit, the analysis of nine tested patients and five controls of the modified sequencing results -compared to NCBI RefSeq transcript mRNA (NM_000059.3) - revealed a single nucleotide change (substitution) within region B at position 3385 yielding a stop codon (TGA) in four patients as (TTA/TGA). The corresponding amino acid sequences appeared as gaps in (Fig. 2); in which the normal amino acid Leucine no longer existed as a result of premature termination (L1053X).

Fig. 2
figure 2

a: I. patient. Illustrates the sequencing result of the chromatogram of one of the tested patients with the substitution mutation marked by a small square. The monoallelic change is more apparent. II. Control. Illustrates the sequencing result of the chromatogram of the control with the normal sequence. A - Adenine, G - guanine, T - thymine, C - cytosine. III. Illustrates Bioedit multiple sequences alignment with substitution of thymine by guanine. b: I. This frame illustrates the nucleotide sequence (in small letters) and their corresponding amino-acids sequence (in capital letters) of a selected frame (5' to 3' frame 1) of the tested region (region B) of BRCA2. The dash (−) represents absence of amino-acid (stop codon). This figure was taken from Expasy online translate tool. II. this frame illustrates the amino-acids sequence in a compacted form. The dash (−) represents absence of amino acid. This figure was taken from Expasy online translate tool

Full size image

Another two single nucleotide changes had been noticed. The first one occurred in two patients with the previously noted L1053X and resulted in Adenine being replaced by Guanine at position 3474 (haplotype), and the corresponding amino acid change was N1083D. This variant was predicted to alter normal protein features in both function and structure -as shown by SIFT sequence and Project Hope. Also it was predicted to decrease protein stability -by I-Mutant. However, it was expected to probably harmless by MutationTaster2 and benign by polyphen-2. The other detected mutation -rs1801406- was silent (K1132 K) and noted in six cases, two of them had both L1053X and N1083D changes, (Table 4).

Table 4 Detected patients among the refined group to carry the following variants within BRCA2 exon 11 primer B region
Full size table

Nonsense mutations

Patients carrying this mutation were premenopausal, with a mean parity of 3.0. The mean age of patients with and without the nonsense mutation was 36.5 and 40.5 years respectively, with a mean difference of four years as illustrated in (Fig. 3). Two patients bearing this SNP were from Ja’alya tribe and one of them had a history of secondary liver deposits (Table 3).

Fig. 3
figure 3

The mean age in breast cancer patients with and without the detected nonsense mutation

Full size image

Discussion

The significant change noted in this study was a monoallelic T3385G stop codon. A variant found with different nomenclatures, c.3158 T > G and n.3386 T > G (Table 5). This SNP was previously identified by Lubiniski in a study aimed to screen familial cases presented with seven different phenotypes including BC and Ovarian Cancer. He studied Ovarian Cancer Cluster Region (OCCR) within the BRCA2 coding sequence. This region was noted more consistently to determine hereditary familial cancer cases. He found termination sequence at position T3386G [18, 36, 37]. The change was similar in both studies (T converted to G) but appears in different positions. However, the resulted-corresponding amino acid sequence provided the same change in both studies (L1053X). Also the mutation has been found as a germline-type but in prostatic cancer cases [38, 39] and one study found this variant within a control subject [40]. The geographic distribution of the variant within detected population has been covered (Table 6).

Table 5 highlights the stop codon L1053X with different nomenclatures described by ClinVar NCBI database
Full size table
Table 6 The geographic provenience of the samples previously detected with the mutation L1053X
Full size table

The patients carrying the mutation had a mean age of 36 years; similar to what was previously reported in Sudan by Awadelkarim et al. who analyzed 35 patients with breast cancer. In terms of parity and menopausal status of the subjects, both studies showed the same trend as the majority of BC cases were premenopausal and multiparous. Furthermore, patients from Ja’alya tribe were found to have truncating mutations in both studies [16].

Our mutation is located within the central region, which possesses eight functional BRC repeats to bind RAD51 -that is essential for Homologous Recombination (HR)- to facilitate its loading onto single strand DNA, where a repair process is needed [41,42,43,44]. Accordingly, any defect of this loading will result in failure of Homologous recombination and the DNA double strand breaks remain altered [45].

From the NCBI database; BRCA2 human has a total of about 10,736 known SNPs, and more than 466 reported truncating mutations. One of these mutations is the K3326X (rs11571833). This mutation has been associated with a 26% increase in the risk of developing breast cancer in European, Latin Americans, and Indian populations. K3326X mutation has been associated with a 2.5 fold increase in risk of squamous lung cancer [46]. Another example of stop codon mutation in BRCA2 is Y3308X (rs4987049) which has been found in Asian, European, Sub-Saharan and African American populations. Other stop mutations in BRCA2 coding region lack frequency data [47]. Seventy Nigerian breast cancer patients with ages younger than 40 years were studied, and one BRCA2 truncating mutation 3034del4 within exon 11 has been reported [48]. The same mutation has been reported in a study of 39 early onset breast cancer (< 40 years) patients in Nigeria. Although 30 variants of BRCA2 were detected, there was only one (3034del4) truncating mutation, located in exon 11 [49].

The N1083D mutation was not previously reported and such a companion is shown in this study by this variant regarding the position to be in continuation -sitting- few steps later after the monoallelic nonsense variant L1053X, so this position proves to be of no significance because it is situated after the nonsense mutation. The other variant, A3623G, was silently expressed as K1132 K, was detected with high frequency among earlier cases, and was involved with three cases detected with the nonsense L1053X including the two N1083D variants. The silent mutation K1132 K was reported among familial cases as the benign non-virulent bearing-characteristic and was found frequently within early onset <50 with mean age 37.5 and more frequently among Asian population and was noticed its high occurrence among a Chinese population [50, 51]. This variant has been recorded with other 13 variants as a recurrent situation among a Belgian population [52].

A technical facility to establish the outcome/resulting truncation inactivation is not available and it is very difficult to handle such a technical assessment. Though all 45 patients’ DNA had been extracted, only 10 patient’s extracts were sequenced owing to financial constraints. Also, due to these financial constraints only the product of one primer with the highest stability was subjected to further analysis in this study. Moreover, the sample size limits the generalizability of this study, but for this variant to be generalized to the Sudanese population, further studies using larger sample size will be needed in the future. In a general context, BRCA genes have not got wide assessment within our geographic region, thus in such scarce way of expression of BC genetic characteristics regarding some countries including Sudan, data presented in our study could be more raised. Most of BRCA2 mutations variants detected within African literature have been gathered in (Table 7) with their corresponding country of origin.

Table 7 Most of the BRCA2 mutations variants detected within African literature
Full size table

Conclusion

This study detected monoallelic L1053X mutation causing the same stop codon in BRCA2 protein sequence at the same position in four Sudanese female BC patients out of nine from different families. This nonsense mutation should be evaluated in further studies in a larger number of BC patients in both hetero-homozygosity re-evaluation and to check the reliability of using this stop codon as a screening tool for early detection of BC.