Expression analysis of long non-coding A T B and its putative target in breast cancer

Abstract

BACKGROUND:

A long noncoding RNA (lncRNA) activated by transforming growth factor (TGF)-β (lncRNA-ATB) has been recently shown to promote the invasion-metastasis cascade in various types of cancers via upregulation of some targets including ZEB1.

OBJECTIVES:

The aim of the present study was to elucidate the expression of lncRNA-ATB and ZEB in breast cancer patients.

METHODS:

The expression of these genes was evaluated by real-time reverse transcription polymerase chain reaction in tumor samples form 50 newly diagnosed breast cancer patients as well as their corresponding adjacent non-cancerous tissues (ANCTs). Patients were divided into subsequent groups according to the median lncRNA-ATB expression.

RESULTS:

LncRNA-ATB has been shown to be downregulated in about two third of tumor samples compared with their ANCTs.

A significant association has been found between ZEB1 expression and Ki-67 status. In addition, we demonstrated a correlation between expression of lncRNA-ATB and ZEB1 in tumor samples and not in ANCTs.

CONCLUSION:

Collectively, out data show downregulation of lncRNA-ATB in a significant number of breast tumor tissues compared with ANCTs and imply that lncRNA-ATB might have distinct roles in the pathogenesis of different cancers or even different subtypes of a certain cancer which should be evaluated in future studies.

Introduction

Although more than 98% of the human genome encompasses non-protein coding sequences, recent studies have demonstrated that majority of them are transcribed into non-protein coding RNA transcripts. These transcripts have variable sizes ranging from very small RNAs such as the microRNAs (miRNAs) with 20–25 base pairs sizes to long non coding RNAs (lncRNAs) that can be up to 100 kb or more. They have been shown to take part in chromatin rearrangement, histone modification, modification of alternative splicing genes, control of gene expression, dosage compensation, genomic imprinting, cell differentiation and organogenesis as well as tumorigenesis [1]. Notably, their significant role in breast cancer evolution has been highlighted in several studies [2,3]. Based on their functions and patterns of expression they can be categorized to tumor suppressors and oncogene classes. The lncRNA-activated by transforming growth factor β (lncRNA-ATB) is a non-polyadenylated lncRNA which localizes mainly in the cytoplasm, and has three close homologs in the human genome [4]. It has been firstly shown to be upregulated in hepatocellular carcinoma metastases and linked with poor prognosis. It has been shown to be a competing endogenous RNA (ceRNA) which binds to miR-200 family and consequently upregulates ZEB1 and ZEB2 and promotes epithelial-mesenchymal transition (EMT) and invasion. Another independent role of lncRNA-ATB is its function in organ colonization of disseminated tumor cells which is mediated by its binding with IL-11 mRNA, autocrine induction of IL-11, and prompting STAT3 signaling [5]. Afterwards, its high expression in colorectal cancer patients has been shown to be associated with larger tumor size, penetration of tumor invasion, lymphatic invasion, vascular invasion, lymph node metastasis as well as poorer patients’ survival [6]. Subsequently, its expression has been shown to be upregulated in the renal cell carcinoma tissues and renal cancer cells compared with the adjacent non-tumor tissues and normal human proximal tubule epithelial cells. Besides, its expression has been demonstrated to be dramatically higher in the renal cell carcinoma patients with metastasis. This study has shown the association between level of lncRNA-ATB expression and tumor stages, histological grade, vascular invasion, lymph node metastasis as well as distant metastasis [7]. Its role in breast cancer has been highlighted in a recent study which has shown that it has been the most significantly upregulated lncRNA in trastuzumab resistance SKBR-3 cells and the tissues of trastuzumab resistant breast cancer patients. In addition, in vitro analyses showed that lncRNA-ATB could promote trastuzumab resistance and invasion-metastasis cascade in breast cancer by competitively binding miR-200c, increasing ZEB1 expression and subsequent induction of EMT [8]. EMT has been recognized as a process which is strictly linked with cancer progression and mediated by many of components of tumor microenvironment such as TGFβ  [9]. Considering the close crosstalk between lncRNAs and proteins involved in different stages of cancer initiation and progression, expression analysis of lncRNAs and their protein coding targets in cancer cells might pave the way to understand their biological roles. Consequently, in the present study, we aimed at expression analysis of lncRNA-ATB and its main target ZEB1 in breast cancer samples versus adjacent non cancer tissues (ANCTs) in an attempt to find its significance in breast cancer pathogenesis.

Material and methods

Subjects

This study was approved by the Research Ethics Committee of SBMU. Informed consent was acquired from all of the participants. Fifty cases of breast cancer patients based on accepted clinicopathological were enrolled in this study. The study group consisted of 44 invasive ductal carcinomas (IDC), 3 in situ ductal carcinomas, 2 lobular carcinomas and one mammary Paget’s disease. Tumor samples as well as their ANCT samples were taken from all patients. None of patients had received any treatment before the surgery. ANCT was delineated as the normal breast tissue diagnosed by the pathologists through H.E. staining. Complete and detailed clinical examination including staging based on the International Union Against Cancer TNM (tumor size, nodes, metastases) classification [10], ultrasound examination and bilateral mammography were performed. All samples were immediately snap-frozen in liquid nitrogen and then stored in − 70^∘ C, until being used for RNA extraction.

RNA extraction and quantitative real-time reverse transcription polymerase chain reaction (RT-PCR)

Total RNA was extracted from tissue samples using the AccuZol^TM total RNA extraction solution (Bioneer, Korea) according to the manufacturer’s instructions, except for an extended 1 h treatment with DNase I. RNA purity and concentration was assessed by Thermo Scientific NanoDrop^TM 1000 Spectrophotometer. One μ g of RNA was used for cDNA synthesis by using PrimeScript RT reagent kit (Takara Bio, Ohtsu, Japan). Quantitative real-time reverse transcription polymerase chain reaction (RT-PCR) reaction was performed on a rotor gene 6000 corbett detection system using SYBR Premix Ex Taq (Takara Bio, Ohtsu, Japan). Thermal cycling conditions were a preliminary activation step for 5 minutes at 95^∘ C followed by 40 cycles at 95^∘ C for 15 seconds, 60^∘ C for 10 seconds and 72^∘ C for 30 seconds followed by melting curve acquisition. No template control (NTC) consisting of H₂ O was included in each run. B2M gene was used as normalizer. Forward and reverse primers sequences are as follows, respectively: ZEB Forward: 5′ -TGCCCAAACTGCAAGAAACG-3′ , ZEB Reverse: 5′ -TGAGTCCTGTTCTTGGTCGC-3′ , lncRNA-ATB Forward: 5′ -GGGATTCGATCAACAGAGAG-3′ , lncRNA-ATB Reverse: 5′ -GTCCAAAGTCATACTGCCCC-3′ , B2M Forward: 5′ -AGATGAGTATGCCTGCCGTG-3′ and B2M Reverse: 5′ -GCGGCATCTTCAAACCTCCA-3′ .

Estrogen receptor (ER)/progesterone receptor (PR), HER2/neu, Ki-67 and E-cadherin status

These results were obtained from patients’ medical records which were performed by immunohistochemical (IHC) staining as previously described [11]. For HER2/neu a test result of 0 to 2+ was regarded as negative and 3+ as positive. Ki-67 values were reported as the fraction of positively stained malignant cells among the whole number of malignant cells evaluated by the means of the anti-human Ki-67 monoclonal antibody MIB1. E-cadherin immunoreactivity was reported based on a 4-point scale E-cadherin scoring system as previously explained [11].

Statistical analysis

Fold changes in gene expression were calculated by LinRegPCR (2) and Relative Expression Software Tool-RG©-version 3 (QIAGEN, Korea) considering the amplification efficiencies and cycle thresholds. The amounts of mRNAs in the tissues were standardized to the B2M mRNA and compared between tumor and noncancerous tissues. For evaluation of significance, the pairwise fixed reallocation randomization test with 2000 iterations in the REST 2009 software was used. The level of statistical significance was set at P < 0.05. Moreover, for defining the association between genes expressions in a certain type of samples (tumor or ANCTs), the results were stated as the ratio of target genes over B2M expression (relative expression, REx). PCR assays were carried out in duplicate for each sample, and the results were averaged. LncRNA-ATB relative expression (REx) was determined and normalized using the ΔCt (Ct _{lncRNA- ATB} -Ct_B2M ) method relative to B2M.

SPSSv.18.0.1 (SPSS Inc., Chicago, IL) was applied for statistical analyses of demographic and clinical data. The McNemar test was used when comparing paired tumor and ANCTs. Chi-square and independent t tests were used for testing the association between categorical variables. Statistical significance was based on a P -value of 0.05.

Results

General statistical information

Data have been analyzed based on the information taken from questionnaires, interviews and clinical and laboratory tests. Table 1 summarizes the demographic and clinical data of patients.

Expression of lncRNA-ATB and ZEB in patients’ samples

LncRNA-ATB has been shown to be downregulated in IDC tumor tissues compared with their paired ANCTs in 61% (27 out of 44) of paired samples, while in the remaining a significant upregulation of lncRNA-ATB has been detected in tumor samples compared with ANCTs. ZEB has been dowregulated in 52% (23 out of 44) of IDC tumor tissues compared with their paired ANCTs and upregulated in the remaining samples. Comparison of lncRNA-ATB expression levels between total tumor and ANCT tissues showed a significant dowregulation in tumor tissues (P = 0.02) (Fig. 1a), while such analysis showed no statistically significant difference between these two tissue types in ZEB expression (P > 0.05) (Fig. 1b).

Among the total number of malignant tissues, two were negative for ER, PR and HER2/neu expression (triple negative). In one of them lncRNA-ATB was dramatically upregulated in the tumor tissue compared with the ANCT, while in the other it was significantly downregulated in tumor tissue. In the single Paget’s disease sample, significant downregulation of lncRNA-ATB and upregulation of ZEB has been detected. In lobular samples both genes were downregulated in tumor tissues compared with the paired ANCTs.

Correlations between genes expressions and clinical characteristics

The relationship between expression of genes and clinicopathological variables are shown in tables 2 and 3. A significant association has been found between ZEB expression and Ki-67 status. However, no statistically significant association has been shown between the level of lncRNA-ATB expression in tumor samples (up- versus downregulation in tumor samples compared with their ANCT) and clinicipathological and demographic characteristics such as age, tumor size, ER status, PR status, HER2/neu expression level, lymph node status, histologic grade and stage. We further divided patients’ samples in which lncRNA-ATB expression has been up- or down-regulated in tumor tissues into high (>= 10 folds change) versus low up- or down-regulation (<10 folds change) groups and repeated all statistical analyses. The frequency and cumulative percentage of samples in each subgroup based on relative expression levels are shown in Fig. 2. No statistically significant associations have been detected between the levels of gene expression and clinicopathological data. In addition, using both the proliferation rate divisions of >10% and = <10% and positive vs. negative Ki-67 expression classifications, lncRNA-ATB expression did not significantly correlate with proliferation index.

All samples diagnosed as IDC showed expression of E-cadherin as anticipated, but no correlation has been detected between E-cadherin expression level and genes expression or other prognostic variables.

Relative expression of lncRNA-ATB and ZEB in individual samples

In order to discover any correlation between the expressions of these two genes, the relative expression of these genes has been compared in individual set of samples. In tumor tissues, the analysis showed that there is a moderate correlation between the expression of these genes (R² = 0.35581, P < 0.05) (Fig. 3a). However, such correlation has not been found in ANCTs (R² = 0.0002, P > 0.05) (Fig. 3b). In addition, we further evaluated the association between fold changes in the mentioned genes in tumor tissues versus ANCTs and failed to find a significant association (R² = 0.003, P > 0.05) (Fig. 3c).

Discussion

Previous studies have demonstrated that several transcripts have differential expression pattern between tumor tissues and ANCTs which facilitates their application as cancer biomarkers [2,12,13]. LncRNA-ATB is a mediator of TGF-β signaling which induces invasion-metastasis cascade in various types of cancer [5,7,8]. It has been shown to be upregulated by TGF-β in MCF7 breast cancer cell line and SMAD4-deficient SW480 colorectal cancer cell line, implying that it may be stimulated via the SMAD-independent, noncanonical TGF-β pathway [5]. Furthermore, lncRNA-ATB knockdown has been shown to prevent cell proliferation, induce apoptosis, decrease EMT program and inhibit cell migration and invasion in renal cell carcinoma [7]. In hepatocellular carcinoma, lncRNA-ATB has been demonstrated to exert an oncogenic activity with pleiotropic effects on cell invasion, colonization, and metastasis [5].

Besides, previous studies has revealed its upregulation in several cancer tissues including hepatocellular carcinoma [5], colon cancer [14] and renal carcinoma [7]. However, it has been shown to be downregulated in pancreatic cancer tissues and pancreatic cancer cell lines. Low lncRNA-ATB expression levels have been significantly correlated with lymph node metastases, neural invasion, and clinical stage in pancreatic cancer patients [15]. In breast cancer its upregulation has been shown to be associated with trastuzumab resistance in both cell line and patient samples [8], but there has been no published data regarding its transcript levels in breast cancer tissues compared with normal tissues. In the present study, we demonstrated a total dowregulation of lncRNA-ATB in tumor tissues compared with ANCTs regardless of the patients’ response to any kind of treatment as we evaluated its transcript levels in newly diagnosed and previously untreated patients which exclude any possible effect of therapies on the gene expression. Transcript analysis in matched pairs of primary breast cancer biopsies before and after chemotherapy showed enhanced expression of genes associated with cancer stem cells and TGF-β 1 signaling after chemotherapy [16] which indicates that in order to find the role of these transcripts in tumorigenesis process, expression analysis of such genes should be conducted in previously untreated patients. In addition, in the single Paget’s disease sample as well as both lobular samples, significant downregulation of lncRNA-ATB has been detected. This is in line with the previous hypothesis that the fundamental reacting component in almost all dysplastic, metaplastic, hyperplastic, anaplastic, and neoplastic diseases of the human breast is the terminal duct lobular unit (TDLU) [17] and both ductal and lobular carcinoma originate from a common TDLU [18].

The results of the mentioned studies indicate that lncRNA-ATB might have distinct roles in the pathogenesis of different cancers or even different subtypes of a certain cancer which should be evaluated in future studies. This is in accordance with the pleiotropic and highly context-dependent function of TGFβ on tumor induction or inhibition. As revealed by various studies, during early stages of tumorigenesis, increased TGFβ levels have tumor suppressive effects, while at later stages TGFβ contribute in malignant conversion and progression [19]. LncRNA-ATB as a downstream effector of TGFβ cascade might follow such mechanism of regulation. Besides, in the present study we demonstrated a correlation between expression of lncRNA-ATB and ZEB1 just in tumor samples and not in ANCTs. A previous study has shown that lncRNA-ATB significantly upregulates ZEB1 in an in vitro model of hepatocellular cancer cells [5]. Future studies should evaluate the correlation between the expression of these genes in different models (patient samples versus cell line) as well as various tissue types to find if the mechanism of regulation of ZEB expression by lncRNA-ATB is similar in different physiologic or pathologic conditions.

Taken together, although we demonstrated lncRNA-ATB downregulation in a significant number of breast tissues compared with ANCTs, considering the distinct function of TGFβ in different stages of breast tumorigenesis, evaluation of lncRNA-ATB expression as a target of TGFβ in distinct breast cancer subtypes as well as different disease stages are valuable and facilitate administration of personalized treatment strategies.