MicroRNA-based molecular classification of non-BRCA1/2 hereditary breast tumours

Topology preserving stratification of tissue neoplasticity using Deep Neural Maps and microRNA signatures

Background

Accurate cancer classification is essential for correct treatment selection and better prognostication. microRNAs (miRNAs) are small RNA molecules that negatively regulate gene expression, and their dyresgulation is a common disease mechanism in many cancers. Through a clearer understanding of miRNA dysregulation in cancer, improved mechanistic knowledge and better treatments can be sought.

Results

We present a topology-preserving deep learning framework to study miRNA dysregulation in cancer. Our study comprises miRNA expression profiles from 3685 cancer and non-cancer tissue samples and hierarchical annotations on organ and neoplasticity status. Using unsupervised learning, a two-dimensional topological map is trained to cluster similar tissue samples. Labelled samples are used after training to identify clustering accuracy in terms of tissue-of-origin and neoplasticity status. In addition, an approach using activation gradients is developed to determine the attention of the networks to miRNAs that drive the clustering. Using this deep learning framework, we classify the neoplasticity status of held-out test samples with an accuracy of 91.07%, the tissue-of-origin with 86.36%, and combined neoplasticity status and tissue-of-origin with an accuracy of 84.28%. The topological maps display the ability of miRNAs to recognize tissue types and neoplasticity status. Importantly, when our approach identifies samples that do not cluster well with their respective classes, activation gradients provide further insight in cancer subtypes or grades.

Conclusions

An unsupervised deep learning approach is developed for cancer classification and interpretation. This work provides an intuitive approach for understanding molecular properties of cancer and has significant potential for cancer classification and treatment selection.

MicroRNA-548m Suppresses Cell Migration and Invasion by Targeting Aryl Hydrocarbon Receptor in Breast Cancer Cells

Breast cancer is the most commonly diagnosed cancer among women and one of the leading causes of cancer mortality worldwide, in which the most severe form happens when it metastasizes to other regions of the body. Metastasis is responsible for most treatment failures in advanced breast cancer. Epithelialmesenchymal transition (EMT) plays a significant role in promoting metastatic processes in breast cancer. MicroRNAs (miRNAs) are highly conserved endogenous short noncoding RNAs that play a role in regulating a broad range of biological processes, including cancer initiation and development, by functioning as tumor promoters or tumor suppressors. Expression of miR-548m has been found in various types of cancers, but the biological function and molecular mechanisms of miR-548m in cancers have not been fully studied. Here we demonstrated the role of miR-548m in modulating EMT in the breast cancer cell lines MDA-MB-231 and MCF-7. Expression data for primary breast cancer obtained from NCBI GEO data sets showed that miR-548m expression was downregulated in breast cancer patients compared with healthy group. We hypothesize that miR-548m acts as a tumor suppressor in breast cancer. Overexpression of miR-548m in both cell lines increased E-cadherin expression and decreased the EMT-associated transcription factors SNAI1, SNAI2, ZEB1, and ZEB2, as well as MMP9 expression. Consequently, migration and invasion capabilities of both MDA-MB-231 and MCF-7 cells were significantly inhibited in miR-548m-overexpressing cells. Analysis of 1,059 putative target genes of miR-548m revealed common pathways involving both tight junction and the mTOR signaling pathway, which has potential impacts on cell migration and invasion. Furthermore, this study identified aryl hydrocarbon receptor (AHR) as a direct target of miR-548m in breast cancer cells. Taken together, our findings suggest a novel function of miR-548m in reversing the EMT of breast cancer by reducing their migratory and invasive potentials, at least in part via targeting AHR expression.

The Implications of ncRNAs in the Development of Human Diseases

The mammalian genome comprehends a small minority of genes that encode for proteins (barely 2% of the total genome in humans) and an immense majority of genes that are transcribed into RNA but not encoded for proteins (ncRNAs). These non-coding genes are intimately related to the expression regulation of protein-coding genes. The ncRNAs subtypes differ in their size, so there are long non-coding genes (lncRNAs) and other smaller ones, like microRNAs (miRNAs) and piwi-interacting RNAs (piRNAs). Due to their important role in the maintenance of cellular functioning, any deregulation of the expression profiles of these ncRNAs can dissemble in the development of different types of diseases. Among them, we can highlight some of high incidence in the population, such as cancer, neurodegenerative, or cardiovascular disorders. In addition, thanks to the enormous advances in the field of medical genomics, these same ncRNAs are starting to be used as possible drugs, approved by the FDA, as an effective treatment for diseases.

The emerging role of miRNA clusters in breast cancer progression

Micro RNAs (miRNAs) are small non-coding RNAs that are essential for regulation of gene expression of the target genes. Large number of miRNAs are organized into defined units known as miRNA clusters (MCs). The MCs consist of two or more than two miRNA encoding genes driven by a single promoter, transcribed together in the same orientation, that are not separated from each other by a transcription unit. Aberrant miRNA clusters expression is reported in breast cancer (BC), exhibiting both pro-tumorogenic and anti-tumorigenic role. Altered MCs expression facilitates to breast carcinogenesis by promoting the breast cells to acquire the various hallmarks of the cancer. Since miRNA clusters contain multiple miRNA encoding genes, targeting cluster may be more attractive than targeting individual miRNAs. Besides targeting dysregulated miRNA clusters in BC, studies have focused on the mechanism of action, and its contribution to the progression of the BC. The present review provides a comprehensive overview of dysregulated miRNA clusters and its role in the acquisition of cancer hallmarks in BC. More specifically, we have presented the regulation, differential expression, classification, targets, mechanism of action, and signaling pathways of miRNA clusters in BC. Additionally, we have also discussed the potential utility of the miRNA cluster as a diagnostic and prognostic indicator in BC.

MiR-29b-1-5p is altered in BRCA1 mutant tumours and is a biomarker in basal-like breast cancer

Depletion of BRCA1 protein in mouse mammary glands results in defects in lactational development and increased susceptibility to mammary cancer. Extensive work has focussed on the role of BRCA1 in the normal breast and in the development of breast cancer, the cell of origin for BRCA1 tumours and the protein-coding genes altered in BRCA1 deficient cells. However, the role of non-coding RNAs in BRCA1-deficient cells is poorly understood. To evaluate miRNA expression in BRCA1 deficient mammary cells, RNA sequencing was performed on the mammary glands of Brca1 knockout mice. We identified 140 differentially expressed miRNAs, 9 of which were also differentially expressed in human BRCA1 breast tumours or familial non-BRCA1 patients and during normal gland development. We show that BRCA1 binds to putative cis-elements in promoter regions of the miRNAs with the potential to regulate their expression, and that four miRNAs (miR-29b-1-5p, miR-664, miR-16-2 and miR-744) significantly stratified the overall survival of basal-like tumours. Importantly the prognostic value of miR-29b-1-5p was higher in significance than several commonly used clinical biomarkers. These results emphasise the role of Brca1 in modulating expression of miRNAs and highlights the potential for BRCA1 regulated miRNAs to be informative biomarkers associated with BRCA1 loss and survival in breast cancer.

Transcriptome mining of non-BRCA1/A2 and BRCA1/A2 familial breast cancer

About 10% of all breast cancer cases are the familial type. Mutations in two highly penetrance breast cancer susceptibility genes, BRCA1 and BRCA2, can only explain 20% to 25% of genetic susceptibility to breast cancer, and most familial breast cancer cases have intact BRCA1 and BRCA2 genes that refer to non-BRCA1/A2 or BRCAX familial breast cancer. Despite extensive studies, more than 50% of genetic susceptibility to breast cancer remained to be disclosed. Finding the differences between these two types of breast cancer (non-BRCA1/A2 and BRCA1/A2) at genomic, transcriptomic, and proteomic levels can help us to elucidate fundamental molecular processes and develope more promising therapeutic targets. Here, we used expression data of 391 patients with familial breast cancer including 195 non-BRCA1/A2 and 196 BRCA1 and/or BRCA2 cases from four independent studies by means of meta-analysis to find differences in gene expression signature between these two types of familial breast cancer. As well as, we applied comprehensive network analysis to find crucial protein complexes and regulators for each condition. Our results revealed significant overexpression of cell cycle processes in BRCA1/A2 patients and significant overexpression of estrogen axis in non-BRCA1/A2 patients. Moreover, we found FOXM1 as the central regulator of cell cycle processes and GATA3, FOXA1, and ESR1 as the main regulators of estrogen axis.

Genetic alterations detected by comparative genomic hybridization in BRCAX breast and ovarian cancers of Brazilian population

Background

About 5–10% of breast/ovarian cancers are hereditary. However, for a large proportion of cases (around 50%), the genetic cause remains unknown. These cases are grouped in a separated BRCAX category. The aim of this study was to identify genomic alterations in BRCA1/BRCA2 wild-type tumor samples from women with family history strongly suggestive of hereditary breast/ovarian cancer.

Results

A cohort of 31 Brazilian women was included in the study. Using the GISTIC algorithm, we identified 20 regions with genomic gains and 31 with losses. The most frequent altered regions were 1q21.2, 6p22.1 and 8p23.3 in breast tumors and Xq26 and Xp22.32-22.31 among the ovarian cancer cases. An interesting association identified was the loss of 22q13.31-13.32 and the presence of ovarian cancer cases. Among the genes present in the frequently altered regions, we found FGFR1, NSMCE2, CTTN, CRLF2, ERBB2, STARD3, MIR3201 and several genes of RAET and ULBP family.

Conclusions

In conclusion, our results suggest that alterations on chromosomes 1, 6, 8 and X are common on BRCAX tumors and that the loss on 22q can be associated with the presence of ovarian cancer.

Methods

DNA copy number alterations were analyzed by 60K array comparative genomic hybridization in breast and ovarian FFPE tumors.

MiR-578 and miR-573 as potential players in BRCA-related breast cancer angiogenesis

The involvement of microRNA (miRNAs), a new class of small RNA molecules, in governing angiogenesis has been well described. Our aim was to investigate miRNA-mediated regulation of angiogenesis in a series of familial breast cancers stratified by BRCA1/2 mutational status in BRCA carriers and BRCA non-carriers (BRCAX). Affymetrix GeneChip miRNA Arrays were used to perform miRNA expression analysis on 43 formalin-fixed paraffin-embedded (FFPE) tumour tissue familial breast cancers (22 BRCA 1/2-related and 21 BRCAX). Pathway enrichment analysis was carried out with the DIANA miRPath v2.0 web-based computational tool, and the miRWalk database was used to identify target genes of deregulated miRNAs. An independent set of 8 BRCA 1/2-related and 11 BRCAX breast tumors was used for validation by Real-Time PCR. In vitro analysis on HEK293, MCF-7 and SUM149PT cells were performed to best-clarify miR-573 and miR-578 role. A set of 16 miRNAs differentially expressed between BRCA 1/2-related and BRCAX breast tumors emerged from the profile analysis. Among these, miR-578 and miR-573 were found to be down-regulated in BRCA 1/2-related breast cancer and associated to the Focal adhesion, Vascular Endothelial Growth Factor (VEGF) and Hypoxia Inducible Factor-1 (HIF-1) signaling pathways. Our data highlight the role of miR-578 and miR-573 in controlling BRCA 1/2-related angiogenesis by targeting key regulators of Focal adhesion, VEGF and HIF-1 signaling pathways.

miRGate: a curated database of human, mouse and rat miRNA-mRNA targets

MicroRNAs (miRNAs) are small non-coding elements involved in the post-transcriptional down-regulation of gene expression through base pairing with messenger RNAs (mRNAs). Through this mechanism, several miRNA-mRNA pairs have been described as critical in the regulation of multiple cellular processes, including early embryonic development and pathological conditions. Many of these pairs (such as miR-15 b/BCL2 in apoptosis or BART-6/BCL6 in diffuse large B-cell lymphomas) were experimentally discovered and/or computationally predicted. Available tools for target prediction are usually based on sequence matching, thermodynamics and conservation, among other approaches. Nevertheless, the main issue on miRNA-mRNA pair prediction is the little overlapping results among different prediction methods, or even with experimentally validated pairs lists, despite the fact that all rely on similar principles. To circumvent this problem, we have developed miRGate, a database containing novel computational predicted miRNA-mRNA pairs that are calculated using well-established algorithms. In addition, it includes an updated and complete dataset of sequences for both miRNA and mRNAs 3'-Untranslated region from human (including human viruses), mouse and rat, as well as experimentally validated data from four well-known databases. The underlying methodology of miRGate has been successfully applied to independent datasets providing predictions that were convincingly validated by functional assays. miRGate is an open resource available at http://mirgate.bioinfo.cnio.es. For programmatic access, we have provided a representational state transfer web service application programming interface that allows accessing the database at http://mirgate.bioinfo.cnio.es/API/ Database URL: http://mirgate.bioinfo.cnio.es

Bias in microRNA functional enrichment analysis

MOTIVATION

Many studies have investigated the differential expression of microRNAs (miRNAs) in disease states and between different treatments, tissues and developmental stages. Given a list of perturbed miRNAs, it is common to predict the shared pathways on which they act. The standard test for functional enrichment typically yields dozens of significantly enriched functional categories, many of which appear frequently in the analysis of apparently unrelated diseases and conditions.

RESULTS

We show that the most commonly used functional enrichment test is inappropriate for the analysis of sets of genes targeted by miRNAs. The hypergeometric distribution used by the standard method consistently results in significant P-values for functional enrichment for targets of randomly selected miRNAs, reflecting an underlying bias in the predicted gene targets of miRNAs as a whole. We developed an algorithm to measure enrichment using an empirical sampling approach, and applied this in a reanalysis of the gene ontology classes of targets of miRNA lists from 44 published studies. The vast majority of the miRNA target sets were not significantly enriched in any functional category after correction for bias. We therefore argue against continued use of the standard functional enrichment method for miRNA targets.

miRNA expression profiling of formalin-fixed paraffin-embedded (FFPE) hereditary breast tumors

Hereditary breast cancer constitutes only 5-10% of all breast cancer cases and is characterized by strong family history of breast and/or other associated cancer types. Only ~ 25% of hereditary breast cancer cases carry a mutation in BRCA1 or BRCA2 gene, while mutations in other rare high and moderate-risk genes and common low penetrance variants may account for additional 20% of the cases. Thus the majority of cases are still unaccounted for and designated as BRCAX tumors. MicroRNAs are small non-coding RNAs that play important roles as regulators of gene expression and are deregulated in cancer. To characterize hereditary breast tumors based on their miRNA expression profiles we performed global microarray miRNA expression profiling on a retrospective cohort of 80 FFPE breast tissues, including 66 hereditary breast tumors (13 BRCA1, 10 BRCA2 and 43 BRCAX), 10 sporadic breast carcinomas and 4 normal breast tissues, using Exiqon miRCURY LNA™ microRNA Array v.11.0. Here we describe in detail the miRNA microarray expression data and tumor samples used for the study of BRCAX tumor heterogeneity (Tanic et al., 2013) and biomarkers associated with positive BRCA1/2 mutation status (Tanic et al., 2014). Additionally, we provide the R code for data preprocessing and quality control.

Combined microRNA and ER expression: a new classifier for familial and sporadic breast cancer patients

BACKGROUND

The role of miRNAs in familial breast cancer (fBC) is poorly investigated as also in the BRCA-like tumors. To identify a specific miRNA expression pattern which could allow a better fBC classification not only based on clinico-pathological and immunophenotypical parameters we analyzed miRNA profile in familial and sporadic samples. Moreover since BRCA1 tumors and sporadic triple negative (TN) breast tumors share similarities regarding clinical outcomes and some histological characteristics, we focused on TN and not TN cases.

METHODS

The sample set included fresh frozen tissue samples, including 39 female fBCs (19 BRCA-related and 20 BRCAX) and 12 male fBC (BRCAX). Moreover, we considered TN and non TN (NTN), 21 BRCA-related and 27 sporadic BCs. MiRNA profiling was performed through GeneChip miRNA v.1.0 Array (Affymetrix). ANOVA, hierarchical and consensus clustering analyses allowed identification of pattern of expression of miRNAs and pathway enrichment analysis, considering validated target genes, was carried out to achieve a deeper biological understanding.

RESULTS

ANOVA test led to the identification of 53 deregulated miRNAs; hierarchical and consensus clustering of female fBCs (fFBCs) and male fBCs (fMBCs) highlighted the presence of 3 sample clusters named FBC1, FBC2 and FBC3. We found a correlation between ER-status and the three sample clusters. The three clusters are distinct by a different expression of two clusters of miRNAs (CLU1 and CLU2), which resulted to be different in targeted pathways. In particular, CLU1 targets cellular pathways and CLU2 is involved in epigenetic activities. Considering TN and NTN BRCA-related and sporadic tumors, a hierarchical clustering identified two clusters of miRNAs, which were not so different from CLU1 and CLU2, both in miRNA content and targeted pathways.

CONCLUSIONS

Our results highlighted the importance of miRNA regulation to better clarify similarities and differences between familial and sporadic BC groups.

MicroRNA expression profiling in male and female familial breast cancer

Background:

Gender-associated epigenetic alterations are poorly investigated in male and female familial breast cancer (fBC). MicroRNAs may contribute to the different biology in men and women particularly related to RASSF1A pathways.

Methods:

Microarray technology was used to evaluate miRNA profile in 24 male and 43 female fBC. Key results were validated using RT–qPCR in an external samples set. In vitro studies were carried out to verify microRNA–target gene interaction.

Results:

Pathway enrichment analysis with the 287 differentially expressed microRNAs revealed several signalling pathways differently regulated in male and female cases. Because we previously hypothesised a peculiar involvement of RASSF1A in male fBC pathogenesis, we focussed on the MAPK and the Hippo signalling pathways that are regulated by RASSF1A. Male miR-152 and miR-497 upregulation and RASSF1A and NORE1A interacting gene downregulation were observed, confirming a possible indirect interaction between miRNAs and the two genes.

Conclusions:

For the first time, a different microRNA expression pattern in male and female fBC has been shown. Moreover, the importance of RASSF1A pathway in male fBC carcinogenesis has been confirmed, highlighting a possible role for miR-152 and miR-497 in controlling MAPK and Hippo signalling pathways, regulated by RASSF1A.

MicroRNAs discriminate familial from sporadic non-BRCA1/2 breast carcinoma arising in patients ≤35 years

The influence of genetic factors may contribute to the poor prognosis of breast cancer (BC) at a very young age. However BRCA1/2 mutations could not explain the majority of cases arising in these patients. MicroRNAs (miRs) have been implicated in biological processes associated with BC. Therefore, we investigated differences in miRs expression between tumors from young patients (≤35 years) with sporadic or familial history and non-carriers of BRCA1/2 mutations. Thirty-six young Brazilian patients were divided into 2 groups: sporadic (NF-BC) or familial breast cancer (F-BC). Most of the samples were classified as luminal A and B and the frequency of subtypes did not differ between familial or sporadic cases. Using real time qPCR and discriminant function analysis, we identified 9 miRs whose expression levels rather than miR identity can discriminate between both patient groups. Candidate predicted targets were determined by combining results from miRWalk algorithms with mRNA expression profiles (n = 91 differently expressed genes). MiR/mRNA integrated analysis identified 91 candidate genes showing positive or negative correlation to at least 1 of the 9 miRs. Co-expression analysis of these genes with 9 miRs indicated that 49 differentially co-expressed miR-gene interactions changes in F-BC tumors as compared to those of NF-BC tumors. Out of 49, 17 (34.6%) of predicted miR-gene interactions showed an inverse correlation suggesting that miRs act as post-transcriptional regulators, whereas 14 (28.6%) miR-gene pairs tended to be co-expressed in the same direction indicating that the effects exerted by these miRs pointed to a complex level of target regulation. The remaining 18 pairs were not predicted by our criteria suggesting involvement of other regulators. MiR-mRNA co-expression analysis allowed us to identify changes in the miR-mRNA regulation that were able to distinguish tumors from familial and sporadic young BC patients non-carriers of BRCA mutations.

MicroRNA expression signatures for the prediction of BRCA1/2 mutation-associated hereditary breast cancer in paraffin-embedded formalin-fixed breast tumors

Screening for germline mutations in breast cancer-associated genes BRCA1 and BRCA2 is indicated for patients with breast cancer from high-risk breast cancer families and influences both treatment options and clinical management. However, only 25% of selected patients test positive for BRCA1/2 mutation, indicating that additional diagnostic biomarkers are necessary. We analyzed 124 formalin-fixed paraffin-embedded (FFPE) tumor samples from patients with hereditary (104) and sporadic (20) invasive breast cancer, divided into two series (A and B). Microarray expression profiling of 829 human miRNAs was performed on 76 samples (Series A), and bioinformatics tool Prophet was used to develop and test a microarray classifier. Samples were stratified into a training set (n = 38) for microarray classifier generation and a test set (n = 38) for signature validation. A 35-miRNA microarray classifier was generated for the prediction of BRCA1/2 mutation status with a reported 95% (95% CI = 0.88-1.0) and 92% (95% CI: 0.84-1.0) accuracy in the training and the test set, respectively. Differential expression of 12 miRNAs between BRCA1/2 mutation carriers versus noncarriers was validated by qPCR in an independent tumor series B (n = 48). Logistic regression model based on the expression of six miRNAs (miR-142-3p, miR-505*, miR-1248, miR-181a-2*, miR-25* and miR-340*) discriminated between tumors from BRCA1/2 mutation carriers and noncarriers with 92% (95% CI: 0.84-0.99) accuracy. In conclusion, we identified miRNA expression signatures predictive of BRCA1/2 mutation status in routinely available FFPE breast tumor samples, which may be useful to complement current patient selection criteria for gene testing by identifying individuals with high likelihood of being BRCA1/2 mutation carriers.

Sparse representation for tumor classification based on feature extraction using latent low-rank representation

Accurate tumor classification is crucial to the proper treatment of cancer. To now, sparse representation (SR) has shown its great performance for tumor classification. This paper conceives a new SR-based method for tumor classification by using gene expression data. In the proposed method, we firstly use latent low-rank representation for extracting salient features and removing noise from the original samples data. Then we use sparse representation classifier (SRC) to build tumor classification model. The experimental results on several real-world data sets show that our method is more efficient and more effective than the previous classification methods including SVM, SRC, and LASSO.