E2F1 copy number variations in germline and breast cancer: a retrospective study of 222 Italian women

Abstract

Background

Breast cancer is the most common neoplasia among women in developed countries. The risk factors of breast cancer can be distinguished in modifiable and unmodifiable factors and, among the latter, genetic factors play a key role. Copy number variations (CNVs) are genetic variants that are classified as rare when present in less than 1% of the healthy population. Since rare CNVs are often cause of diseases, over the last years, their contribution in carcinogenesis has become a relevant matter of study. E2F1 is a transcriptional factor that plays an important role in regulating cell cycle and apoptosis. Its double and conflicting role is the reason why it acts both as oncogene and as tumour suppressor, depending on cell context. Since anomalies in expression or in number of copies of E2F1 have been related to several cancers, we aimed to study number of germline copies of E2F1 in women with breast cancer in order to better elucidate their contribution as predisposing factor to this tumour.

Methods

We performed, hence, a retrospective study on 222 Italian women with breast cancer recruited from October 2002 to December 2007. TaqMan CNV assay and Real-Time PCR were carried out to analyse, respectively, E2F1 CNV and E2F1 expression in the subjects of the study. Chi square test or Fisher’s exact test and Student's t‐test were used to calculate the frequency of CNVs and differences in continuous variables between groups, respectively.

Results

Intriguingly, we found that 10/222 (4.5%) women with breast cancer had more copies than controls (0/200, 0%), furthermore, the number of copies positively correlated with E2F1 gene expression in breast cancer tissue, suggesting that the constitutive gain of the gene could translate into an increased risk of genomic instability. Additionally, we found that altered E2F1 copies were present prevalently in the patients with contralateral breast cancer (20%) and all of them had a positive family history, both typically associated with hereditary cancer.

Conclusions

Our findings suggest that copy number variations of E2F1 might be a susceptibility factor for breast cancer, however, further studies on large cohorts are to be performed in order to better delineate the phenotype linked to the gain of E2F1 copies.

Background

Breast cancer (BC) is the most diagnosed neoplastic disease and represents the second cause of cancer-related death, after lung cancer, in women (DeSantis et al. 2019). It is a complex and heterogeneous neoplasia whose aetiology involves several risk factors of genetic, environmental and behavioural origin (Orlandella et al. 2020). Risk factors of BC are generally distinguished in two groups: extrinsic and intrinsic factors. Extrinsic risk factors include obesity, physical activity, alcohol consumption and ionizing radiation exposure, while intrinsic risk factors are unmodifiable parameters such as race, ethnicity, sex, age, early menarche, late menopause, late age at first birth, nulliparity, hormonal factor, family history and genetic mutations (Kaminska et al. 2015). The latter include rare high-risk mutations (BRCA1 and BRCA2 genes), more moderate susceptibility variants (CHEK2 and ATM genes) and several still unidentified common susceptibility variants associated with low to moderate increased risk. These known risk factors, however, fail to fully elucidate the high incidence of BC.

Based on the status of estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor (HER2), different intrinsic subtypes of BC have been classified as: luminal A, luminal B, HER2-enriched and triple negative breast cancer (TNBC).

ER signalling plays a key role in BC development, in fact, up to 75% of all breast cancers are ERα positive (Allred et al. 2004; Osborne and Shiff 2011; Dai et al. 2016); therefore, the deregulation of downstream target proteins of ER could partially explain the underlying mechanisms of initiation and development of most of BC.

Transcriptional factor E2F1, member of E2F family including both transcriptional activators and repressors, is a downstream target of ER pathway resulting overexpressed in breast cancer tissue (Li et al. 2018).

E2F1 is a transcriptional activator promoting proliferation, following mitogenic stimulation, or apoptosis, as a response to DNA damage. Therefore, E2F1 can act both as an oncogene and as a tumour suppressor, depending upon the cellular context (Engelmann and Pützer 2012). Abnormalities in E2F1 gene expression or E2F1 gene amplification have been reported in many types of human cancer (Nelson et al. 2006; Ma et al. 2013; Liang et al. 2016; Kent al. 2017).

Beside gene mutations and hormone receptor (HR) status, copy number variations (CNVs), covering about 12% of whole human genome (Schaschl et al. 2009), have been largely investigated in BC, because of their role as risk factors for several diseases, including tumours (Petrij-Bosch et al.1997; Montagna et al. 2003; Casilli et al. 2006; Lesueur et al. 2008; Cybulski et al. 2006; Cybulski et al. 2007; Shlien and Malkin 2009; Kumaran et al. 2017).

In our previous studies, we found that the frequency of germinal CNV of E2F1 gene in patients with testicular cancer and melanoma was higher compared to healthy controls and, furthermore, the increased number of copies of E2F1 correlated with an increased gene expression, especially under stress conditions, suggesting germline E2F1 CNVs as risk factor of these two tumours (Rocca et al., 2017; Rocca et al. 2018).

Based on these recent evidence, the aim of this study was to investigate the frequency of CNV of E2F1 in 222 Italian women with breast cancer in order to better elucidate the contribution of this structural variant as a potential predisposing factor to breast carcinogenesis.

Materials and methods

This study was approved by the Ethics Committee of Padova University Hospital (identifier: prot#448).

A total of 222 women of Italian origin were retrospectively selected among patients referred to the First Surgical Clinic, University Hospital of Padua—Veneto Institute of Oncology, Italy for breast cancer. We extracted the clinic-pathological data of treated patients between October 2002 and December 2007, using a prospectively maintained database linked to the biobank of the First Surgical Clinic—University Hospital of Padua, Italy. To be included in the study, each case had to meet the following requirements: (1) histologically confirmed diagnosis of breast cancer or metastasis from breast cancer; (2) pathology-based information on TNM stage; (3) follow-up data (minimum follow up: 6 months); (4) availability of DNA for genotyping purposes (Table 1). Tissue specimens were available for 35 out of 222 patients.

Table 1 Description of phenotypic characteristics of patients

200 women, referred to Unit of Andrology and Reproduction Medicine of University Hospital of Padua, with no history of any malignancy were used as controls. All subjects provided written informed consent.

Copy number variation analysis

Genomic DNA of patients was provided by the above-mentioned biobank. DNA was isolated from peripheral blood leucocytes using QIAamp DNA Blood Mini Kit, according to the manufacturer’s protocol (Qiagen Inc., Hilden, Germany). Copy number variation was evaluated on 20 ng of genomic DNA. Quantitative real-time polymerase chain reaction (PCR) TaqMan Copy Number Assays were performed using three probes targeting different regions of the E2F1 gene (Hs00576444_cn, Hs01758822_cn and Hs00919582_cn)(Applied Biosystems, Foster City, CA, USA). TaqMan CNV reactions were performed in triplicate using the FAM-dye-labeled assay for E2F1 and VIC-dye labeled RNase P assay. RNase P assay was used to normalize the genomic DNA input.

An internal DNA resulted with two copies of E2F1 both by TaqMan Copy Number Assay and array CGH was used as calibrator. Real-time data were collected by the StepOne Plus 2.1 software, and ABI CopyCaller 2.0 software (Thermo Fisher Scientific Inc, Waltham, MA, USA) was used for data analysis. Copy Number ranging from 1.5 to 2.5 were predicted as CNV = 2. Two independent assays were performed for each sample to confirm results.

RNA isolation, cDNA synthesis and real-time PCR

Total RNA was extracted from 35 breast cancer tissues of patients included in the study using the RNeasy Mini Kit (Qiagen, Hilden, Germany). cDNA was synthesized from 250 ng of total RNA using SuperScript III (Invitrogen, Carlsbad, CA, USA) and random hexamers. Real Time PCR were performed in a 20 µl final volume containing 20 ng of cDNA, 1X Power SYBR Green PCR Master Mix (Applied Biosystem, Foster City, CA, USA), and a mix of forward and reverse primers (1 mmol/l each). The following primers were used: E2F1: forward 5′-CATCAGTACCTGGCCGAGAG-3′ and reverse 5′-CCCGGGGATTTCACACCTTT-3′. Human GAPDH was used as a housekeeping gene: forward 5′-TCGACAGTCAGCCGCATCTT-3′ and reverse 5′-AGGCGCCCAATACGACCAAA-3′. Real Time PCR was performed on thermocycler StepOne plus (Applied Biosystems, Foster City, CA, USA) and relative quantification was performed using Delta Delta Ct (ΔΔCt) method.

Statistical analysis

Statistical analysis of the data was conducted with SPSS 21.0 for Windows (SPSS, Chicago, IL). Statistical power was calculated comparing two proportions: the frequency of E2F1 altered copies in healthy controls and the frequency of E2F1 altered copies in patients with BC. The α level of significance was set at 0.05.

Differences in the frequency of CNVs between groups were compared using the Chi square test, or Fisher’s exact test when expected frequency was < 5. Differences in continuous variables between groups were analysed by Student's t‐test. P value ≤ 0.05 was considered statistically significant.

Results

On Table 1 are reported main characteristics of patients. Age did not differ between cases and controls (Age = 60 ± 14.5 and 59 ± 10.4, respectively). Pre-, peri- and post-menopausal women were also comparable between cases and controls (data not shown). In the group of cancer patients, we found a significantly higher portion of women (4.5%, 10/222) with more than two copies of E2F1, compared with controls where none of subjects harboured E2F1 CNVs > 2 (0%, 0/200; p = 0.002). The statistical power of this study was 80%.

At this point, we compared patient’s phenotypic characteristics between subjects with CNV > 2 and those with CNV = 2. Table 2 summarises this analysis. In particular, patients with more than 2 copies of E2F1 had less lymph nodes-positive (p = 0.04) and prevalently a bilateral tumour (p = 0.04), and, additionally, they had positive BC family history (p = 0.001). No difference has emerged for other traits.

Table 2 Pathological characteristics in two group of patients. Data are presented as proportion of individuals within each category

Gene expression analysis on breast cancer tissue of 35 out 222 patients (N = 4 with CNV > 2 and N = 31 with CNV = 2 in blood) (Table 3) has revealed a positive and statistically significant correlation between E2F1 expression and number of copies (p = 0.004) (Fig. 1a). Specifically, three out four individuals with germline CNV > 2 showed more copies also in breast tissue.

Table 3 Description of phenotypic characteristics of 35 out of 222 patients on which E2F1 expression and copy number assay were performed

Fig. 1

Scatter plots and box plot illustrating respectively the associations between exact CNVs and E2F1 expression (a) and the different E2F1 expression between patients with CNV = 2 (N = 31) and CNV > 2 (N = 4) in somatic tissue (b). *p value < 0.001

Furthermore, expression levels were significantly different between subjects with CNV = 2 and subjects with CNV > 2 (p < 0.001) (Fig. 1b).

Discussion

BC is the most common malignant tumour among women in the majority of the developed countries and it is normally distinguished into four main molecular subtypes. Although established risk factors are associated with each subtypes (Yang et al. 2007; Sisti et al. 2016; Holm et al. 2017), however, the constant research of novel biomarkers is of fundamental importance in clinical practice so that the patients may have an early diagnosis and specific treatment.

Approximately 5–10% of breast cancers are hereditary and up to 25% are due to mutations in the following high-penetrance genes: BRCA1, BRCA2, PTEN, TP53, CDH1, and STK11 (Oosterwijk et al. 2014; Shiovitz and Korde 2015). Therefore, molecular screening programmes are of huge relevance in terms of prevention, especially in women with BC family history. Indeed, individuals having first degree relatives with genetic mutations, mainly in BRCA1 and BRCA2 genes, have an increased risk of developing BC (Ripperger et al. 2009).

Over last years, several groups have studied germinal CNVs as predisposing factors to cancer. CNVs are structural variations ranging from 1 kb to several Mb in length, referred to as polymorphisms if present in more than 1% of population. However, rare CNVs have been described in some tumours, including breast cancer (Kumaran et al. 2017), suggesting a role of these variants as susceptibility factors (Lee and Scherer 2010).

Recently, germinal CNVs of E2F1 have been proposed as genetic risk factor for testicular cancer and melanoma (Rocca et al. 2017; Rocca et al. 2019). In both studies, patients showed a higher frequency of altered E2F1 copies compared to controls.

E2F1 is a member of E2F family and acts as a transcriptional factor of genes encoding proteins implicated in cell-cycle progression and apoptosis (Attwool et al. 2004). E2F1 expression has been found to be increased in breast cancer tissue, pointing out the important contribution of this transcriptional factor in breast carcinogenesis (Li et al. 2018).

Based on large body evidence regarding the role of CNV in disease susceptibility and the altered expression of E2F1 in breast cancer, we have analysed CNVs of E2F1 in a cohort of 222 women with breast cancer.

Interestingly, we found that the frequency of CNVs of E2F1 was higher in the patients compared to controls (4.5%), suggesting therefore that altered CNVs of E2F1 may play a key role as predisposing factor of BC as well as it was reported in patients with testicular cancer and melanoma (Rocca et al. 2017; Rocca et al. 2019). This finding, intriguingly, is corroborated by gene expression analysis that has pointed out a higher expression of E2F1 in breast tissue from three out of four patients with more germline copies of the gene, included in somatic tissue analysis. Therefore, it is likely that inherited germinal CNVs, in combination with other extrinsic or intrinsic factors, may contribute to increase genomic instability in breast tissue, resulting in the deregulation of E2F1 target proteins.

Patients with more copies of E2F1 were almost similar to patients with normal copies, except for three clinical characteristics. Indeed, we found a statistically significant difference among two groups for following parameters: lymph nodes status, laterality and family history.

Contralateral breast cancer (CBC) is less common than unilateral breast cancer (UBC) and it has been supposed to have mainly a genetic cause (Mack et al. 2002). Indeed, women carrying mutations in BRCA1, BRCA2 and CHECK genes have a higher risk of developing CBC and the estimation of the risk is higher in carriers of germinal mutations, confirming the strong genetic contribution (Robson et al. 2017; Kramer et al. 2020). Based on these evidence, the finding of altered E2F1 CNVs in patients with CBC suggests that this structural variant, likely inherited, as well as it occurs in inherited germline mutations of BRCA genes, may contribute to an increased risk of CBC. This hypothesis, hence, could explain the early age at diagnosis and BC family history of one of the patients with more copies of E2F1 and CBC.

It is well known, indeed, that hereditary cancers are generally characterized from an earlier age of onset of BC (Brandt et al. 2008); therefore, the combination of early age of onset and positive BC family history represents a very strong risk factor and is generally associated with germline mutation in BRCA1 gene (Anders et al. 2009).

Interestingly, the patient harbouring more copies of E2F1 with CBC and positive family history was also carrier of a mutation in BRCA1 gene. This finding could suggest that carriers of inherited rare germline mutations and CNVs of E2F1 may have an increased risk of BC compared to the individuals having only one of these two genetic variants. It is known that BRCA1 plays a crucial role in the preserving genome integrity, hence, the loss of function of BRCA1 could promote cell cycle progression also of those cells with more copies of E2F1, translating into the deregulation of E2F1 target genes due its overexpression.

In this study, additionally, we have found a significant difference between the two groups of the patients in relation to the presence of lymph nodes positive. Although we have not enough data to clarify this result, we could suppose that this parameter may contribute to better delineate the phenotype linked to the patients having more copies of E2F1.

In conclusion, although this study adds information to the research of new biomarkers of BC, the small size of samples and the missing data about the inheritance pattern of the identified CNVs is a limitation, therefore, these results should be confirmed from study on a large cohort of women selected for BC family history.

However, based on our results, we suppose that the high frequency of E2F1 CNV might predispose to BC, therefore, molecular screening of this gene, mainly in women with positive family history, would be of fundamental importance to estimate the risk of recurrence into the family. Furthermore, the confirmation of germinal CNVs of E2F1 as novel predictive biomarkers of BC could have impressive implications in clinical practice regarding the choice of targeted therapies against this malignant cancer.

Conclusions

In conclusion, although this study adds information to the research of new biomarkers of BC, the small size of samples and the missing data about the inheritance pattern of the identified CNVs is a limitation, therefore, these results should be confirmed from study on a large cohort of women selected for BC family history.

However, based on our results, we suppose that the high frequency of E2F1 CNV might predispose to BC, therefore, molecular screening of this gene, mainly in women with positive family history, would be of fundamental importance to estimate the risk of recurrence into the family. Furthermore, the confirmation of germinal CNVs of E2F1 as novel predictive biomarkers of BC could have impressive implications in clinical practice regarding the choice of targeted therapies against this malignant cancer.