Abstract
BACKGROUND:
Impaired miRNAs processing pathway is one interesting scenario for global downregulation of the miRNAome in various types of malignancy. We previously reported that DGCR8 and Dicer genes dysregulated in patients with breast cancer.
OBJECTIVE:
To evaluate the expression pattern of Drosha in patients with breast cancer.
METHODS:
We evaluated the mRNA expression level of Drosha in 70 fresh breast carcinomas and adjacent non-neoplastic tissue using quantitative real-time PCR and assessed the possible correlation between its expression and clinicopathological parameters.
RESULTS:
Our results revealed that mRNA expression level of Drosha was decreased in tumors when compared to adjacent non-neoplastic tissue. However, this difference is not statistically significant (P > 0.05). Downregulation of Drosha is related to older age at diagnosis, higher histological grade, higher tumor size and metastasis. However, there was no significant correlation between Drosha expression level and clinicopathological parameters (P > 0.05). We found that Drosha expression negatively correlated with DGCR8 (P = 0.043), whereas dysregulated expression levels of Drosha and Dicer are positively correlated with to each other (P < 0.0001).
CONCLUSION:
This study provides evidence that the expression of Drosha is impaired in breast cancer. However, the molecular basis of observed expression pattern have remained inexplicable and should be further investigated.
Introduction
MicroRNAs (miRNAs) is a class of endogenous non-coding short RNAs which delicately regulate the expression of protein-coding genes at the post-transcriptional level through targeting mRNAs in a sequence-specific manner by triggering translational repression or, more commonly, degradation of the mRNAs [1]. Transcripts targeted by miRNAs are involved in a wide array of biological processes such as chromatin structure remodeling, genome rearrangement, DNA repair, cell cycle progression and cell programmed death [2,3]. Not surprisingly, dysregulation of miRNAs expression contributes to development and progression of various human malignancies [3]. The underlying mechanisms of dysregulated miRNAs expression have not been completely elucidated. However, it seems that dysregulation of miRNAs expression could be caused by DNA copy number aberrations, mutations, aberrant DNA methylation, improper histone modification, dysregulation of transcription factors targeting miRNAs and alterations in the miRNAs processing pathway [4]. The biogenesis of miRNAs is a stepwise process, which is characterized by two cleavage reactions mediated by RNase III enzymes, Drosha and Dicer [5]. The miRNAs genes initially are transcripted into the pri-miRNAs by RNA polymerase II [5]. Processing of pri-miRNAs takes place concurrently with or shortly after transcription by microprocessor complex in the nucleus to liberate ∼70 nucleotide stem–loop structures, termed pre-miRNAs [5]. Initial processing of the vast majority of miRNAs is performed by this complex, although a small subset of miRNAs, mirtrons, can bypass this step [4,5]. The main components of microprocessor complex are RNase III enzyme Drosha and DGCR8 proteins [6]. The N-terminus of Drosha is required for nuclear localization and, the middle and C-terminal part of Drosha, containing the two RNase III domains (RIIIDa and RIIIDb) and one double-stranded RNA binding domain (dsRBD) are essential for pri-miRNAs processing [5,6]. DGCR8, as a “molecular ruler”, recognizes the unique features of the pri-miRNAs and determines the cleavage site within the pri-miRNAs [6]. Following the cleavage of pri-miRNAs by the Drosha microprocessor complex, pre-miRNAs transport to the cytosol by exportin 5/RanGTP complex and are processed by another RNase III endonuclease, Dicer, to generate the ∼ 22 nt mature miRNAs [4,5].
Over the past decade, a gradual accumulation of evidence demonstrated that the aberrant expression or function of miRNA processing component(s) could potentially alter the miRNAome and contribute to transforming healthy cells into cancer cells [7–11]. Therefore, the precise regulation of miRNA processing component(s) expression may be crucial in guarding the cells against malignancy. Accordingly, dysregulation of miRNA processing component(s) such as Drosha, DGCR8, XPO5, Dicer, TRBP and Ago2 have been reported in various cancers [7–19]. Along the same lines, several studies have been recently conducted to discover the link between dysregulation of Drosha and breast cancer, but these studies have shown conflicting results [12–17]. Because of these contradictory data, in the current study, we posed a question as to whether mRNA expression pattern of Drosha is altered in patients with breast cancer, and whether clinicopathological parameters have any effect on the Drosha expression. Moreover, we previously demonstrated that the expressions of DGCR8 and Dicer dysregulated in breast cancer [10,11]. Hence, the possible correlation of Drosha expression with DGCR8 and Dicer expression levels was also assessed in this study.
Materials and methods
Sample preparation
Breast cancerous and matched adjacent non-neoplastic tissues were obtained from seventy patients diagnosed with breast cancer undergoing radical surgery at the Nour-nejat hospital, Tabriz, Iran during 2013–2015. The mean age of patients was 47.27 ± 8.09 years (ranging between 33 and 71 years). The histological grade, tumor size and lymph node metastasis of all resected samples were determined by histopathological analysis. Patients who received any preoperative neoadjuvant chemotherapy or radiotherapy were excluded from the study. The present study was approved by the Ethics and Human Rights Committee of Hormozgan University of medical sciences, Hormozgan, Iran and the informed consents were filled out by all participants.
RNA extraction and cDNA synthesis
All tissue samples were stored in RNAlater solution (Qiagen, Germany) at − 80∘ C until RNA extraction. Total RNA was extracted from tissue samples (approximately 100 mg) using Tri-Pure® Isolation Reagent (Roche, Germany) and RNA yield was treated with RNase-free DNase I (Thermo Scientific, USA), according to the manufacturer’s instructions. The quality and quantity of extracted RNA were confirmed by agarose gel electrophoresis and NanoDrop® ND-1000 Spectrophotometer (Thermo Scientific, USA), respectively. Then, total RNA (1 μ g) was reverse transcribed to cDNA using RevertAidTM First Strand cDNA Synthesis Kit (Fermentas, Canada) with random hexamer primer, following the manufacturer’s protocol.
Quantitative Real-time PCR
Real-time PCR was conducted in triplicate using gene-specific primers including F 5′ - AGCCCTGGTGCCTGAGGAGGAGAT -3′ and R 5′ -TGCAGGGCGTATCCCAAAGTGGAC-3′ for Drosha, F 5′ -CAGCCATGTACGTTGCTATCCAGG-3′ and R 5′ -AGGTCCAGACGCAGGATGGCATG-3′ for β -actin and Syber Green-I dye in AccuPower® 2X GreenStarTM qPCR Master Mix (Bioneer, Korea) by the Rotor-GeneTM 6000 system (Corbett Research, Australia) according to the manufacturer’s instructions. Each reaction mixture contained 100 ng cDNA, master mix 2X, ROX dye 50X, and 10 pmol of each primer pairs for Drosha and β -actin in a final volume of 25 μ l. Initial denaturation at 94∘ C for 5 min was followed by 50 cycles (for Drosha) and 35 cycles (for β -actin) of denaturation at 94∘ C for 10 s, annealing at temperatures 60.2∘ C and 59.8∘ C for Drosha and β -actin, respectively for 15 s, extension at 72∘ C for 20 s. The mean expression of the housekeeping gene β -actin was used as a control to normalize the variability in expression level of Drosha. We also used a no template sample as a negative control. A standard curve was included in each run for assay validation.
Statistical analysis
Statistical analysis was performed using SPSS software version 16.0 (SPSS Inc., Chicago, IL, USA). Histogram on a log2 scale of the expression ratio was created to determine the distribution of Drosha expression level around cutoff points, and tested for normality with Kolmogorov-Smirnov test. We used student’s t -test to determine the statistically significant difference in the expression of Drosha between tumors and matched adjacent non-neoplastic tissues. The association between clinicopathological parameters and the expression level of Drosha was examined using Fisher’s exact test. Pearson’s correlation test was used to analysis of the correlation between Drosha expression and clinicopathological features. The correlation of Drosha expression with DGCR8 and Dicer expression levels was analyzed using Spearman’s correlation test. P values less than 0.05 were statistically considered as significant.
Results
Expression of Drosha in breast cancerous and matched adjacent non-neoplastic tissues
The mRNA expression level of Drosha in seventy fresh breast cancerous and matched adjacent non-neoplastic tissues were evaluated using quantitative real-time PCR method and the ratio of the Drosha mRNA expression analyzed by the 2−ΔΔct method. Initial analysis of the distribution of Drosha mRNA expression level using histogram and Kolmogorov-Smirnov test showed that the data were normally distributed (P = 0.003). Therefore, the median expression level (i.e. 1.30 of the log2 value) was chosen as the cut-off value in subsequent analysis. Accordingly, patients divided into two distinct groups with low and high expression of Drosha (Table 1). Our findings revealed that the mRNA expression level of Drosha varied among patients and its expression is downregulated in slightly more than half (52.86%) of the tumors when compared to matched adjacent non-neoplastic tissue (Table 1). As shown in Table 1, the median ratio of Drosha expression in patients with low and high Drosha mRNA levels were 0.64 (range, 0.08–1.29) and 2.39 (range, 1.37–4.16) respectively. When, the Drosha expression level were compared between tumors and matched adjacent non-neoplastic tissue using two-tailed t-test, we found that there is no statistical significant differences between two groups (P = 0.710).
Relation between Drosha expression level and clinicopathological parameters
To determine the influence of the clinical features on expression level of Drosha, patients were subdivided according to each clinicopathological parameters (Table 1). With regard to age of patients, Drosha down regulated in 45.83% (≤50 years) and 68.18% (>50 years) of patients (Table 1). The mean of Drosha mRNA expression level in ≤50 years patients was higher than >50 years; however, this difference is not statistically significant (P = 0.303) (Table 1). Regarding histopathological grade, downregulation of Drosha expression was observed in 61.54% (grade I), 48.72% (grade II) and 55.56% (grade III) (Table 1). Comparison of Drosha expression between different histological grades using one way ANOVA test showed that the mean of Drosha mRNA expression level was lower in grade III compared to both grade I and grade II, but did not show a statistically significant difference (P = 0.608) (Table 1). Among the patients with tumor size ≤2, 2–5 and >5 cm, Drosha mRNA expression level decreased in 53.33%, 51.11% and 60.00%, respectively (Table 1). The mean of Drosha expression in patients with tumor >5 cm was lower than ≤5, but with no significant difference (P = 0.431) (Table 1). Based on the data in Table 1 we observed downregulation of Drosha expression in 50.00% of patient with lymph node metastasis. The mean of Drosha expression in patients with lymph node metastasis was lower than patients without lymph node metastasis; however, the mean of Drosha expression was not statistically different between two groups (P = 0.309) (Table 1).
Correlation of Drosha expression level with clinicopathological parameters
The correlation between relative expression of Drosha in specimens and clinicopathological parameters were evaluated using Pearson’s correlation test (Table 2). Our results showed that Drosha mRNA expression level was not statistically correlated with age, histopathological grade, tumour size and lymph node metastasis. However, we observed that the expression of Drosha mRNA was negatively correlated with age and tumor size (Table 2).
Correlation of Drosha expression with DGCR8 and Dicer expression levels
As shown in Fig. 1, the mRNA expression level of Drosha was negatively correlated with DGCR8 expression (Fig. 1a), whereas strong positive correlation between Drosha and Dicer expressions was observed (Fig. 1b).
Discussion
Due to the critical role of Drosha in canonical miRNA processing, we investigated the mRNA expression of Drosha in breast cancer, with the aim of evaluating the alteration of Drosha expression at mRNA level and its correlation with clinicopathological parameters. Our results showed that level of Drosha mRNA varied among the cancerous samples from different cases, and slightly more than half of the tumor samples (52.86%) had low Drosha mRNA expression compared with adjacent non-neoplastic tissue. We did not observe any significant differences in the mRNA level of Drosha between tumor and adjacent non-neoplastic tissue (P = 0.710). Our results are in accordance with the findings of several previous studies that have been published in this field [12–15]. Kwon et al. reported that the mRNA expression of Drosha downregulated in 78% of patients with breast cancer [12]. However, the differences of Drosha expression between cancerous tissues and adjacent non-neoplastic breast tissues were not statistically significant [12]. In the study conducted by Yan et al. decreased levels of Drosha mRNA have reported in 51.9% of the breast cancerous tissues [13]. Moreover, a gradual loss of Drosha cytoplasmic expression was observed along tumour progression from ductal carcinoma in situ, to invasive and to metastatic cancer cells [14]. Dedes et al. investigated Drosha expression in 245 patients with breast cancer receiving adjuvant anthracycline-based chemotherapy and reported that Drosha downregulated in 18% of cases [15]. In contrast, Avery-Kiejda et al. and Passon et al. observed increased Drosha mRNA expression level in triple-negative breast cancers samples [16,17]. In light of these studies and our data, dysregulation of Drosha may be involved in pathogenesis of breast cancer. In line with our observation in breast cancer, previous investigations in invasive epithelial ovarian cancer [18], endometrial cancer [9], gall bladder adenocarcinoma [20], nasopharyngeal carcinoma [19], colon cancer [21], neuroblastoma [22] and cutaneous melanoma [7] revealed that level of Drosha mRNA decreased in cancerous specimens compared with normal tissue (Table 3). Conversely, in serous ovarian cancer [23], bladder urothelial carcinoma [24,25], adrenocortical carcinoma [26], pleomorphic adenomas of the salivary gland [27], colorectal cancer [28], non-small cell lung cancer [29], epithelial skin cancer [8], esophageal cancer [39] and gastric cancer, the level of Drosha mRNA were increased (Table 3). With respect to the finding of studies mentioned above, it seems that Drosha plays a dual role in cancer development, either through upregulation or downregulation, depending on the tumor type.
Although, the causes of the downregulated expression of Drosha observed in our study and other reports are not understood very well, there are several plausible explanations for reduced expression of Drosha among various types of malignancy. It has been shown that Drosha expression regulates at transcriptional and post-transcriptional levels [30–33]. A chromatin immunoprecipitation (ChIP) experiment showed that c-Myc transcription factor binds directly to the E-box of the Drosha promoter, and transactivates Drosha mRNA expression [30]. Using ChIP assays and promoter reporter studies, Rupaimoole et al. demonstrated that Drosha is downregulated by ETS1 and ELK1 transcription factors through recruitment of HDAC1 and ARID4B complexes to methylate Drosha promoter region [31]. Kim et al. reported that EWS, an RNA/DNA-binding protein, regulates the expression of Drosha and EWS deficiency resulted in increased expression of Drosha [32]. At the post-transcriptional level, ARF, a multifunctional tumor suppressor, regulates Drosha mRNA translation to prevent aberrant cell proliferation and Ras-dependent transformation [33]. On the other hand, it has been recently demonstrated that various factors such as BRCA1, p53, ATM and SMAD are involved in regulation of miRNA biogenesis through direct or indirect interactions with Drosha [34–37]. It should be noted that the role of dysregulated expression and/or mutation of c-Myc, ETS1, ELK1, ARF, BRCA1, p53, ATM and SMAD proteins have been frequently reported in pathogenesis of breast cancer. It is probably no exaggeration to say that observed downregulation of Drosha in our study resulting from the dysregulated expression and/or mutation of above mentioned factors. Considering all these data, the expression and function of Drosha precisely regulated with various molecular mechanisms. However, further studies may be required to disclose other possible mechanisms involving in control of Drosha expression and function.
The correlation between Drosha mRNA expression level and clinicopathological parameters was also evaluated. Unfortunately, we failed to find significant correlation between altered expression of the Drosha and clinical features including Age, histological grade, tumor size and lymph node metastasis. These findings are in agreement with several previous studies [12,13,16]. These studies clarified that there is no evidence to support a correlation between Drosha mRNA expression level and clinicopathological parameters in breast cancer. However, in spite of the non-correlation reported in this investigation and some previous studies, possible correlation between altered expression of the Drosha and clinicopathological parameters could not be completely ruled out. As previously, Dedes et al. revealed that decreased Drosha mRNA expression was correlated with high histological grade and high Ki-67 index in triple negative breast cancer cases receiving adjuvant anthracycline-based chemotherapy [15]. Another study revealed that the downregulated of Drosha protein expression was associated with shorter disease free survival [14]. On the other hand, we previously reported that DGCR8 mRNA expression is upregulated in patients with breast cancer, whereas Dicer is downregulated [10,11]. We found that Drosha expression negatively correlated with DGCR8 (Fig. 1a). This negative correlation may be a consequence of posttranscriptional crossregulation between Drosha and DGCR8. It has been reported that DGCR8 expression is negatively regulated by the Drosha-DGCR8 complex at the posttranscriptional level through mRNA cleavage [38]. In other words, if Drosha and DGCR8 levels are increased in the cell, Drosha-DGCR8 complex will be cleaved and destabilize the DGCR8 mRNA, resulting in the reduction of DGCR8. Additionally, DGCR8 stabilizes the Drosha protein via protein-protein interaction. This auto regulatory feedback loop may help maintain the homeostatic control of miRNA biogenesis [38]. However, further studies may be required to disclose other possible mechanisms. Our finding showed a strong positive correlation between Drosha and Dicer expressions (Fig. 1b). However, no reasons were given for this positive correlation.
In conclusion, our study provides additional evidence that Drosha expression downregulated in patients with breast cancer, which may implicate its involvement in genesis and progression of breast cancer. However, upregulated expression of Drosha has been previously reported in triple-negative breast cancer [16,17]. The reasons for this difference have remained inexplicable yet. A small sample analysis in our study may account for the non-statistical significance results. Therefore, further large scale and functional studies are required to determine the precise role of Drosha in breast cancer pathology.
Footnotes
Acknowledgements
The authors would like to express our sincerest appreciation to all the subjects for participating in this study.
Conflict of interest
There are no potential conflicts of interest for each author, concerning the submitted manuscript.
