Introduction

Breast cancer is the most common type of cancer affecting women worldwide, accounting for 23 % of the total new cancer cases. It is also the second leading cause of cancer death among women after the lung cancer [1]. Breast carcinoma is the most common type of cancer attacking women in Tunisia, with an age standardized rate of 16.7 per 100,000 per year. Average age at diagnosis is 50 [2]. Breast cancer is a sporadic disease in almost 90–95 % of the cases, while the familial form is observed in only 5–10 %. In this latter form, BRCA1 is the most commonly involved gene, along with BRCA2 gene [3]. Dysfunction or alteration within these two major genes predisposes to breast cancer with a poor prognosis, and it is often associated with high risk of distant metastases’ occurrence and even with lymph node metastases onset [4, 5]. BRCA1 gene is located on chromosome 17q21, containing 22 coding exons and two noncoding ones with a transcript of 7.8 kb which encodes a protein of 1,863 amino acids. BRCA1 protein is described as a tumor suppressor molecule which maintains genomic stability through its involvement in homologous recombination and in DNA repairs [6]. BRCA1 gene mutational spectrum is not entirely depicted. More than one thousand mutations are incorporated in breast cancer information core BIC database, which are classified as deleterious mutations, SNP and UV (unknown value) [7]. Half of the deleterious mutations are frame shift, nonsense or splice-site alterations that lead to premature truncation of the protein upon translation, and approximately 60 % of these deleterious mutations are unique to a family [8]. However, germ line deleterious mutations in BRCA1 are not always revealed in inherited breast cancer. Previous studies performed by direct sequencing of BRCA1 gene among Tunisian patients with familial breast cancer showed deleterious mutations in <20 % of the cases [9]. The same study showed that such mutations in BRCA1 gene are observed in 50 % of the families with breast and ovarian cancer and in only 4 % of the families with breast cancer without any history of ovarian cancer [10]. Since this last situation corresponds to the majority (70 %) of familial breast cancer cases in Tunisian population, the present research is only concerned with the involvement of BRCA1 gene across families with breast cancer only. Indeed, it could be hypothesized that mutations in other suppressor genes could be associated with this disease. However, other mechanisms involving BRCA1 gene could be suggested. More recently, several studies have showed that the lack of gene function may be due to epigenetic factors such as micro-RNAs [11]. Many micro-RNAs were described to be involved in cancer and were called “onco-miR” that down-regulate tumor suppressor genes. Others were described to act as “tumor suppressor miR” that down-regulate oncogenes [12]. Micro-RNAs, miRs or miRNAs are small noncoding RNAs ranging from 20 to 24 nucleotides in length that are highly conserved, encoded by genomes of plants and animals [13]. They are posttranscriptional regulators of gene expression by targeting mRNAs, and they also play an essential role in cell signaling pathways. Micro-RNAs precursors are transcribed either from nonprotein coding areas of DNA (RNA genes), or from introns of protein-coding genes as rather long (1,000 nucleotides) primary transcripts (pri-miRNA), which undergo a multistep maturation process to become biologically active miRs [13]. Some micro-RNAs were described in literature as being the target of BRCA1 gene such as miR182 and miR17 correlated with an increased risk of familial breast cancer [14]. This correlation needs to be more investigated, along with the absence of deleterious mutations in BRCA1 gene across families with breast cancer.

This research explores the hypothesis that the combination between genotypes of BRCA1 SNPs and onco-miR expression levels could generate a high risk of familial breast cancer along with the occurrence of distant metastases. In a previous study, we described at least 18 neutral SNPs in BRCA1 gene leading to different haplotypes based on 14 SNPs. Nineteen haplotypes were observed in Tunisian population among which only three were in common with American population [10].

In the present research, we conducted a case–control study including patients and healthy controls based on nine SNPs within the BRCA1 gene among the 14 ones already studied in our previous work, which are as follows: c.442.58delT, c.2082C>T, c.2311T>C, c.2612C>T, c.3113A>G, c.3119G>A, c.3548A>G c.4308T>C and 4837A>G in order to investigate in depth the involvement of BRCA1 SNPs in familial breast cancer. Moreover, to explain some aspects of this association, we investigated interactions between SNPs associated with familial breast cancer and their targeting of micro-RNAs.

Patients and methods

Patients

A cohort of total 34 subjects were recruited as healthy controls, from Charles Nicolle hospital in Tunis (capital of Tunisia), and the average age was 48.08 ranging from 24 to 72. We ensured that these individuals were unpaired and have no cancer history in their families. Patients with breast cancer were selected from chemotherapy service at Salah Azaiz Institute of Tunis. They were diagnosed between 2002 and 2012. Forty subjects were selected among those who have familial breast cancer based on personal and familial cancer history, and 46 ones were selected among those who have sporadic breast cancer with no cancer history in their families. Familial breast cancer is defined as the index case of family with at least three female relatives affected with breast and/or ovarian cancer. In the cases of families with fewer than three affected relatives, we took into account the presence of one relative diagnosed with breast or ovarian cancer before the age of 45, the presence of one member of the family diagnosed with ovarian cancer and at least one family member on the same side diagnosed with breast cancer (at any age) and the presence of one relative diagnosed with multiple primary breast cancers. Sporadic and familial breast cancer patients’ average age was, respectively, 48.24 raging from 22 to 74 and 41.01 raging from 18 to 64 years. Essential information including stage, grade, lymph nodes metastasis and distant metastasis was collected from the Salah Azaiz Institute data base. The peripheral blood was collected into tubes and centrifuged at 3,000 rpm for 15 min; the puffy coat was isolated and stored at −80 °C until use. Return to the archive of anatomopathology department of Salah Azaiz Institute in order to retrieve formalin-fixed, paraffin-embedded (FFPE) tissues of the same patients was a very heavy operation. Finally, we managed to retrieve thirty FFPE tissues for each group of patients with their corresponding normal tissues. Three 10-mm cores were obtained from both the tumor and the normal tissues. A great effort was made to avoid any adjacent normal tissue and to isolate areas of tissue containing more than 70 % of tumor cells. All personal data were concealed to guarantee patients’ protection. All patients signed and approved an informed consent. The procedures were in agreement with the regulations for use of human material in research issued by the Medical Ethics Committee of Pasteur Institute of Tunis. An ethical approval was signed by Dr M. Samir BOUBAKER president of the Medical Ethics Committee.

Immunohistochemistry

Immunohistochemistry (IHC) was performed on FFPE tissues. Routine sections of 2 μm thick were cut using DAKO Capillary Gap slides (S2024; DAKO Corp, Carpinteria, CA, USA), fixed in 10 % buffered formalin and dried at 60 °C overnight. Slides were dewaxed in xylene transferred to absolute alcohol and incubated in 3 % hydrogen peroxide in methanol for 10 min to block endogenous peroxidase.

Slides were then transferred to running tap water before being transferred to 3 l of boiling citrate buffer pH 6.0 in a 15-lb pressure cooker, rinsed in Tris-buffered saline (TBS) pH 7.4 and incubated in normal goat serum (1:10) for 10 min.

Serum was tipped off, and the sections were incubated in primary antibody for 60 min at the appropriate dilution. We used monoclonal antibodies anti-BRCA1 (GLK-2) with dilution 1/50. After incubation, slides were rinsed in TBS and incubated in DAKO Duet (K0492) biotinylated goat antimouse/rabbit secondary reagent (1:100) for 35 min. Slides were incubated in DAKO Duet (K0492) streptavidin–biotin–horseradish peroxidase complex for 35 min, rinsed in TBS and treated with DAB (3,3′ diaminobenzidine chromogen 896102, Kem-En-Tec, Copenhagen, Denmark) for 10 min. Finally, slides were rinsed in tap water, counterstained in Mayer’s hematoxylin and mounted.

Sequencing

Genomic DNA was first extracted by proteinase K digestion from peripheral blood mononuclear cells isolated from each sample and then purified using the phenol/Chloroform method. Exons of BRCA1 containing the studied SNPs were PCR amplified in a total reaction volume of 50 µl containing 10 mM Tris–HCl, pH 8.3, 50 mM KCl, 1.5–4.5 mM MgCl, 50 mM dNTPs, 10 µl of each primer (designed by Centre Jean Perrin; sequences available on demand), 100 ng of genomic DNA and 1 U of either Taq polymerase (Applied Biosystems, Roche). PCR cycling program comprised an initial denaturation at 94 °C for 5 min, followed by 30 cycles including 20 s at of 94 °C, following the annealing step at specific temperatures for each primer pair and the extension at 72 °C for 20 s. Amplicons were purified by solid-phase extraction using QIAquick column gel (QIAGEN). The product was sequenced in forward and reverse reactions using Applied Biosystems Taq DyeDeoxy terminator cycle sequencing kit according to the manufacturer’s instructions. Cycle sequencing consisted of 25 cycles at 96 °C for 30 s and 60 °C for 30 s. Sequence analysis was performed using SEQMAN (DNAstar, Madison, WI, USA) and SEQSCAPE V2.5 (Applied Biosystems) software.

Bioinformatics and statistical analysis

Case/control studies were conducted using the Epi-info seven program. P value was considered after the Bonferroni correction. Sequences identified relying on the Ensembl browser’s tools http://www.ensembl.org/index.html have served to determine mature micro-RNAs belonging to the class of Homo sapiens which target regions containing the studied SNPs across their wild and mutated alleles. This work was carried out with miRBase, available at http://www.mirbase.org/ [15], which is the central online repository for micro-RNAs nomenclature, sequence data, annotation and target prediction.

Selected micro-RNAs are associated with the lowest E value which is the most significant to target the analyzed regions. It is a method for comparing pair-wise alignments with different similarities and different lengths. It corresponds to the number of times the database match may have occurred just by chance [15]. Normality of data and micro-RNA fold expression distribution was checked using the Kolmogorov–Smirnov test. The comparison of micro-RNA fold expression among patients according to the clinical outcome, results of BRCA1 protein expression and genotyping were analyzed using the student T test. Statistical significance was set at the 95 % level (P < 0.05). These statistical analyses were performed relying on the SPSS 20 (IBM) software.

Quantitative real-time PCR (Q-PCR)

Total micro-RNAs were extracted from FFPE tissues using the miRNeasy FFPE kit GIAGEN and then converted to c.DNA using the miScript II RT Kit QIAGEN. Both miR-4668 5P and miR-1179 expressions were tested by Q-PCR SYBR Green using the miScript SYBR Green PCR Kit QIAGEN and miScript Primer Assays QIAGEN corresponding to these two selected micro-RNAs. Protocol details and cycling conditions were performed according to the manufacturer’s instructions. Total RNA concentrations were measured using an ND-1000 spectrophotometer. RNA integrity was confirmed on an Agilent 2100 Bioanalyzer. Micro-RNA expression levels (2−ΔCt) were normalized to these of Hs_RNU6B_2 as a reference and to the mean expression of the normal samples using the ΔΔCt method.

Results

Clinical outcome

Clinical data were compared between both sporadic and familial breast cancer patients (Table 1). No significant differences were obtained for the median age under and upper 35 years, along with stage (T0–T4), grade (G1–G3) and lymph node metastasis. However, a significant difference was observed for the distant metastases’ occurrence among patients: 45 % of them were categorized as FBC patients, and only 19.5 % of them as SBC ones having distant metastasis, P value = 0.03 and OR = 2.9 (Table 1).

Table 1 Clinical characteristics among familial breast cancer and sporadic breast cancer patients
Full size table

Immunohistochemistry study of BRCA1 protein

BRCA1 protein expression among sporadic and familial breast cancer patients was assessed according to the presence or the absence of BRCA1 protein within mammary tissues. Scores ranged from 0 to 3+. Negative staining was scored 0 and 1+, and positive staining was scored 2+ and 3+. We found a significant difference of the BRCA1 protein staining among the two groups of patients where BRCA1 protein was absent in (46.6 %) of the FBC cases compared to (16.6 %) of the SBC ones with P value = 0.02 and OR = 5 (Table 2).

Table 2 BRCA1 protein expression in mammary tumors revealed by Immunohistochemistry according to both familial and sporadic breast cancer patients
Full size table

Case–control study

We performed a case–control study to compare genotypes and alleles frequencies between the three analyzed groups based on the nine selected SNPs (Table 4). We found that c.2082C>T, c.3113A>G, c.3119G>A, c.3548A>G and c.4837A>G SNPs were not associated with breast cancer disease (FBC or SBC) with P value >0.05. However, the comparison of the three aforementioned groups revealed a significant difference for c.442.58 delT, c.2311T>C, c.2612C>T and c.4308T>C SNPs (P value <0.05; Table 3). Comparison of allelic and genotypic frequencies between sporadic and familial breast cancer groups considering c.442.58 del T SNP, showed that TT genotype which is the wild type, seems to be strongly associated with familial breast cancer relative to SBC patients. P value = 0.000001 and OR = 9.67. However, TT genotype appears to be protective against SBC form comparing with HC. P value = 0.02 and OR = 0.32. Considering allele frequency, we showed that T allele which is the wild type increases the risk of familial breast cancer occurrence with P value = 0.000001 and OR = 8.68 (Table 3). The comparison across genotypes’ frequencies of c.2311T>C BRCA1 SNP between sporadic and familial breast cancer patients showed a significant difference in TT wild-type genotype which is associated with the FBC group with P value = 0.02 and OR = 3.16, while CC genotype which is the mutated type was almost entirely absent in the same group. T wild-type allele is associated with familial breast cancer but not to the sporadic form with P value = 0.01 and OR = 2.76 (Table 3). Concerning c.2612C>T BRCA1 SNP, the association of CC genotype frequency among FBC patients compared to HC is significant with P value = 0.03 and OR = 3.14. C allele frequency is also associated with familial breast cancer cases compared to HC. P value = 0.01 and OR = 2.49 (Table 3). Regarding c.4308 T>C SNP, the TT wild-type genotype frequency is associated with FBC patients relative to both HC and SBC cases, respectively ((P value = 0.04, OR = 3.3) and (P value = 0.01, OR = 4.66)). Similarly, the T wild-type allele is also associated with FBC cases compared to HC and SBC patients, respectively ((P = 0.001, OR = 4.6) and (P value = 0.002, OR = 4.11; Table 3)).

Table 3 Case–control study of BRCA1 variants: c.442.58delT, c.2082C>T, c.2311T>C, c.2612C>T, c.3113A>G, c.3119G>A, c.3548A>G c.4308T>C and 4837A>G across patients groups and healthy controls
Full size table

Functional role of the associated BRCA1 SNPs with familial breast cancer patients

We focused on the eventual functional role that could play the four selected BRCA1 SNPs c.442.58 delT, c.2311T>C, c.2612C>T and c.4308T>C which are observed to be associated with FBC patients through their wild genotypes and alleles. At first, we used miRbase tools to predict potential micro-RNAs that may target regions within BRCA1 mRNA containing these SNPs. We selected 2 micro-RNAs: miR-4668-5p and miR-1179, respectively, targeting c.442-58delT and c.2311T>C regions at their wild-type allele based on a low E value, 2.3 and 1.5, respectively (Table 4). For c.2612 C>T, miRbase predicts a targeting miR-511 of the wild allele with a high E value = 6.8. Consequently, we did not take into account this predictable micro-RNA. Concerning c.4308 T>C SNP, miRbase did not find any predictable micro-RNA that could target its region with the wild-type allele (Table 4). In order to confirm these computing results, we assessed the expression levels of both miR-4668-5p and miR-1179 by Q-PCR SYBER GREEN. This was realized through analyzing 30 mammary tumors of each familial and sporadic breast cancer patients along with their corresponding normal tissues. MiScript Primer of miR-4668-5P did not show any result because of its high concentration in Guanine and Cytosine. We managed to get results only for miR-1179. The experiment was carried out three times for each sample. After normalizing the expression levels with the corresponding mean value of the gene reference RNU6B by the ΔCt method, the samples were checked for outliers to be excluded. The final expression levels of miR-1179 were considered after normalizing tumors mean expression levels with those of their corresponding normal tissues according to ΔΔCt method. The expression levels of miR-1179 across FBC and SBC patients were standardized (Log N transformed) and analyzed by descriptive analyses method. Thus, the expression levels were approximately normally distributed among FBC patients with skewness of 0.01 and Kurtosis of −0.32, the mean fold expression =5.83 with a standard deviation (SD) = 3.69. However, in SBC patients the mean fold expression =3.63 with SD = 2.5, the expression levels across those patients were also approximately normally distributed with skewness of −1.09 and Kurtosis of 3.11 (Table 5). The difference of the means was assessed among the two groups of patients using the one-way ANOVA method and was revealed statistically significant with F ratio = 7.28 and P value = 0.009 (Table 5).

Table 4 Predicted micro-RNAs targeting sequences that contain BRCA1 SNPs; c.442-58 delT, c.2311T>C, c.2612C>T and c.4308T>C according to their wild- or mutated-type alleles using miRbase tools
Full size table
Table 5 Expression level of miR-1179 in both familial and sporadic breast cancer patients
Full size table

Second, results were further explored by comparing the different means of miR-1179 fold expression according to our previous results, BRCA1 protein expression (negative or positive staining) and genotypes of c.2311T>C BRCA1 SNP (TT, TC and CC), having or not distant metastasis and according to the presence or absence of BRCA1 deleterious mutations. This last criterion concerns only FBC patients. Results of the deleterious mutations were already established on the same FBC patients by Troudi et al. [7]; our concern is to compare the mean fold expression on FBC patients having or not deleterious mutations to explain the role that could be assigned to miR-1179 in the absence of BRCA1 protein independently of deleterious mutations. miR-1179 fold expression among SBC patients according to BRCA1 protein status (negative and positive staining) did not show any significant difference, T test = −0.27, df = 28 and P value = 0.78 (Fig. 1). However, miR-1179 fold expression across FBC patients considering the negative staining of BRCA1 protein is clearly higher than the one having BRCA1 positive staining, T test = −6.87, df = 28 and P value = 0.00001 (Fig. 2). Then, we compared the miR-1179 fold expression between the two groups of patients (FBC and SBC) according to the different genotypes (TT, TC and CC) of c.2311T>C BRCA1 variant. We found that miR-1179 fold expression among FBC patients with TT genotype is higher than this among SBC patients with the same genotype, T test = 3.36, df = 40 and P value = 0.002; however, the expression between patients according to the other genotypes TC and CC did not show any significant difference (Fig. 3). In addition, we compared the mean fold expression of the analyzed micro-RNA between FBC and SBC patients according to both BRCA1 protein expression and c.2311 T>C genotypes, and combined together, we found that the mean fold expression of miR-1179 according to both TT genotypes and negative staining of BRCA1 protein is much higher among FBC patients than those with SBC form, T test = 4.41, df = 15 and P value = 0.01. Nevertheless, we did not record any significant difference of the fold expression within FBC and SBC patients according to TC and CC genotypes combined with positive or negative staining of the protein (Fig. 4). To further confirm the involvement of miR-1179 in FBC with BRCA1 negative staining, we considered previous results about deleterious mutations of BRCA1 gene that were already established by Troudi et al. [7] on the same FBC patients. We found that the mean fold expression of miR-1179 among FBC patients without BRCA1 deleterious mutation is absolutely higher than the fold expression in those with deleterious mutations, T test = 2.7, df = 22 and P value = 0.01 (Fig. 5). Finally, we assessed the mean fold expression levels of the analyzed micro-RNA to the occurrence of distant metastasis.

Fig. 1
figure 1

miR-1179 fold expression among sporadic breast cancer patients according to BRCA1 protein expression. Student T test was calculated considering independent samples method dealing with equal variances assumed. Arrows indicate the two analyzed groups by student T test. P value <0.05 is significant, df degree of freedom, SD standard deviation, SBC sporadic breast cancer

Full size image
Fig. 2
figure 2

miR-1179 fold expression among familial breast cancer patients according to BRCA1 protein expression. Student T test was calculated considering independent samples method dealing with equal variances assumed. Arrows indicate the two analyzed groups by student T test. P value <0.05 is significant, df degree of freedom. SD standard deviation, FBC familial breast cancer

Full size image
Fig. 3
figure 3

miR-1179 fold expression among familial and sporadic breast cancer patients according to TT, TC and CC genotypes of c.2311T>C BRCA1 variant student T test was calculated considering independent samples method dealing with equal variances assumed. Arrows indicate the two analyzed groups by student T test. P value <0.05 is significant, df degree of freedom, SD standard deviation, FBC familial breast cancer, SBC sporadic breast cancer

Full size image
Fig. 4
figure 4

miR-1179 fold expression according to TT, TC and CC genotypes of c.2311T>C BRCA1 variant among familial and sporadic breast cancer patients considering BRCA1 protein expression. Student T test was calculated considering independent samples method dealing with equal variances assumed. Arrows indicate the two analyzed groups by student T test. P value <0.05 is significant, df degree of freedom, SD standard deviation, FBC familial breast cancer, SBC sporadic breast cancer

Full size image
Fig. 5
figure 5

miR-1179 fold expression among familial breast cancer patients with TT genotype of c.2311T>C BRCA1 variant depending on whether or not to have deleterious mutation in BRCA1 gene. Student T test was calculated considering independent samples method dealing with equal variances assumed. Arrows indicate the two analyzed groups by student T test. P value <0.05 is significant, df degree of freedom, standard deviation, FBC familial breast cancer

Full size image

Our results show well that miR-1179 expression levels according to distant metastases’ occurrence are higher within FBC patients compared to those with SBC form, T test = 3.43, df = 15 and P value = 0.04 (Fig. 6).

Fig. 6
figure 6

miR-1179 fold expression according to TT, TC and CC genotypes of c.2311T>C BRCA1 variant among familial and sporadic breast cancer patients considering the occurrence or not of distant metastasis. Student T test was calculated considering independent samples method dealing with equal variances assumed. Arrows indicate the two analyzed groups by student T test. P value <0.05 is significant, df degree of freedom, SD standard deviation, FBC familial breast cancer, SBC sporadic breast cancer

Full size image

General discussion

BRCA1 gene is commonly the most characterized gene in familial breast cancer. Previous studies failed to demonstrate in most hereditary breast cancer cases, the role assigned to BRCA1 germ line deleterious mutations in the dysfunction of BRCA1 protein. The involvement of BRCA1 gene in mammary tumors through epigenetic mechanisms was proven by several studies. Methylation of promoters as well as over-expression of oncogenic micro-RNAs targeting tumor suppressor genes is well documented [16, 17]. Recently, it has been suggested that deregulation of BRCA1 transcription could contribute to a sporadic form of breast cancer, in relationship with microenvironment effect [18]. Several research works demonstrated that BRCA1 gene dysfunction plays a crucial role in distant metastases’ occurrence especially in brain and pulmonary metastases [19]. This could explain our clinical outcome where we observed high frequency of patients having distant metastases’ occurrence within FBC group compared to those with SBC form. This observation suggests that distant metastases’ occurrence among FBC patients may be due to the dysfunction of BRCA1 protein. Our investigation was performed on nine SNPs within BRCA1 gene. We analyzed their frequencies in three groups of unrelated women from Northern Tunisia: healthy subjects, patients with sporadic and familial breast cancer. Sample sizes were adapted to the number of familial breast cancer diagnosed in this region where 1,000 cases of breast cancer are observed per year among which only 5 % took the family form. According to our case–control study, four among nine SNPs revealed significant differences in genotypes and alleles frequencies among the studied groups. Wild-type alleles and genotypes of c.442-58 delT, c.2311T>C, c.2612 C>T and c.4308T>C SNPs are clearly associated with familial breast cancer with an odds ratio ranging from 2.49 to 4.66. Two questions are raised in the light of these results: The first is the unexpected feature related to the risk associated with wild-type genotypes of BRCA1 SNPs, and the second is related to the functional role that these SNPs could play across their wild-type genotypes in familial breast cancer onset.

Considering BRCA1 wild-type allele as an ancestral form that was selected during evolution and then adapted to the different ways of life for thousands of years, we notice that in the last 50 years, women lifestyle has changed dramatically.

Wild allelic form of BRCA1 gene adapted to the ancient lifestyle became at risk of breast cancer in the new environmental conditions. This observation is not unique for breast cancer, and it was also observed for other diseases such as obesity and high blood pressure (HBP) [20]. Indeed, genes selected in conditions of lack of food and salt became at risk, respectively, for obesity and HBP due to over nutrition and excessive salt consumption [21]. During the last decades, the Tunisian population just like others was subject to epidemiological transition wrapping an increasing number of breast cancer cases, including familial cases [22]. This could be explained by changes in lifestyle which became part of modern life societies including social and environmental factors such as stress, fatty food, hormonal contraception, lack of exercise, pregnancy history and breast-feeding and by the interaction between genetic and epigenetic features.

Among the four associated SNPs to familial breast cancer, the c.2612 C>T variant could have an effect on the protein sequence with an amino acid change (Proline to Leucine) at position 871, suggesting an alteration on the protein function and an ambivalent role of wild allele to familial breast cancer susceptibility. However, the c.4304T>C variant was not associated with any amino acid change. Its association with breast cancer might be indirect, due to linkage disequilibrium with other SNPs. However, its direct involvement in breast cancer could not be excluded. For c.442-58delT BRCA1 SNP, we found that it is complementary to mature miR-4668-5p. This micro-RNA was identified in normal and tumor breast tissues, suggesting its role as an onco-miR [23]. We tried to check the validity extent of this result by evaluating its expression levels in mammary tumors using Q-PCR, but we did not find any amplification because of the high G and C concentration in miR-4668-5P miScript primer assay. Concerning c.2311T>C BRCA1 SNP, our computing findings showed that the wild-type allele is targeted by miR-1179. The over-expression of this micro-RNA was already demonstrated in breast cancer and human sarcoma, and it was up-regulated in colorectal cancers with distant metastases’ occurrence [24]. In this research, using Q-PCR analysis, we intend to show that expression levels of miR-1179 within FBC tumors are significantly higher than those in SBC ones and this according to both TT genotype of c.2311T>C SNP and the absence of germ line BRCA1 deleterious mutations. In addition, the mean fold expression of the studied micro-RNA was associated with distant metastases’ occurrence with TT genotype among FBC patients compared to those with SBC form. This observation supports the functional role that could be assigned to BRCA1 SNPs with its wild-type alleles on FBC onset and therefore on distant metastasis.

In light of these results, we can propose the following model: Combination between the wild-type TT genotype of c.2311T>C and miR-1179 over-expression generates a lack of BRCA1 protein expression leading to a higher risk of developing a familial breast cancer with distant metastases’ occurrence.

We suggest that micro-RNAs are mostly directed against wild-type allele sequences. In a way of life, endogenous factors such as hormones and inflammatory cytokines could be influenced and therefore could have an effect on oncogenic micro-RNAs such as miR-4668-5p and miR-1179 by up-regulation mechanism. Therefore, a subject carrying wild-homozygous-type genotypes within the BRCA1 gene could be threatened by breast cancer with the onset of distant metastasis.

Conclusion

This study is in agreement with genetic end epigenetic interactions in breast cancer across micro-RNAs expression which could be considered as a mechanism rather than deleterious mutations, responsible for familial breast cancer onset with distant metastases’ occurrence. Onco-miRs targeting BRCA1 gene and other suppressor genes have to be better studied. Transduction ways leading to their expression in relation to environmental and endogenous factors remain to be further explained. This research gives the hope to prevent familial breast cancer inspecting BRCA1 germ line deleterious mutations along with the high risk of distant metastases’ occurrence.