Camps Carme, Saini Harpreet K, Mole David R, Choudhry Hani, Reczko Martin, Guerra-Assunção José Afonso, Tian Ya-Min, Buffa Francesca M, Harris Adrian L, Hatzigeorgiou Artemis G, Enright Anton J, Ragoussis Jiannis
The Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Oxford, United Kingdom.
Mol Cancer. 2014 Feb 11;13:28. doi: 10.1186/1476-4598-13-28.
In mammalians, HIF is a master regulator of hypoxia gene expression through direct binding to DNA, while its role in microRNA expression regulation, critical in the hypoxia response, is not elucidated genome wide. Our aim is to investigate in depth the regulation of microRNA expression by hypoxia in the breast cancer cell line MCF-7, establish the relationship between microRNA expression and HIF binding sites, pri-miRNA transcription and microRNA processing gene expression.
MCF-7 cells were incubated at 1% Oxygen for 16, 32 and 48 h. SiRNA against HIF-1α and HIF-2α were performed as previously published. MicroRNA and mRNA expression were assessed using microRNA microarrays, small RNA sequencing, gene expression microarrays and Real time PCR. The Kraken pipeline was applied for microRNA-seq analysis along with Bioconductor packages. Microarray data was analysed using Limma (Bioconductor), ChIP-seq data were analysed using Gene Set Enrichment Analysis and multiple testing correction applied in all analyses.
Hypoxia time course microRNA sequencing data analysis identified 41 microRNAs significantly up- and 28 down-regulated, including hsa-miR-4521, hsa-miR-145-3p and hsa-miR-222-5p reported in conjunction with hypoxia for the first time. Integration of HIF-1α and HIF-2α ChIP-seq data with expression data showed overall association between binding sites and microRNA up-regulation, with hsa-miR-210-3p and microRNAs of miR-27a/23a/24-2 and miR-30b/30d clusters as predominant examples. Moreover the expression of hsa-miR-27a-3p and hsa-miR-24-3p was found positively associated to a hypoxia gene signature in breast cancer. Gene expression analysis showed no full coordination between pri-miRNA and microRNA expression, pointing towards additional levels of regulation. Several transcripts involved in microRNA processing were found regulated by hypoxia, of which DICER (down-regulated) and AGO4 (up-regulated) were HIF dependent. DICER expression was found inversely correlated to hypoxia in breast cancer.
Integrated analysis of microRNA, mRNA and ChIP-seq data in a model cell line supports the hypothesis that microRNA expression under hypoxia is regulated at transcriptional and post-transcriptional level, with the presence of HIF binding sites at microRNA genomic loci associated with up-regulation. The identification of hypoxia and HIF regulated microRNAs relevant for breast cancer is important for our understanding of disease development and design of therapeutic interventions.
在哺乳动物中，低氧诱导因子（HIF）是通过直接结合DNA来调控低氧基因表达的主要调节因子，然而其在低氧反应中对微小RNA（miRNA）表达调控的作用尚未在全基因组范围内阐明。我们的目的是深入研究乳腺癌细胞系MCF-7中低氧对miRNA表达的调控，建立miRNA表达与HIF结合位点、初级miRNA（pri-miRNA）转录以及miRNA加工基因表达之间的关系。
将MCF-7细胞在1%氧气条件下培养16、32和48小时。按照先前发表的方法使用针对HIF-1α和HIF-2α的小干扰RNA（siRNA）。使用miRNA微阵列、小RNA测序、基因表达微阵列和实时定量聚合酶链反应（Real time PCR）评估miRNA和mRNA表达。将Kraken流程与Bioconductor软件包一起应用于miRNA测序分析。使用Limma（Bioconductor）分析微阵列数据，使用基因集富集分析（Gene Set Enrichment Analysis）分析染色质免疫沉淀测序（ChIP-seq）数据，并在所有分析中应用多重检验校正。
低氧时间进程miRNA测序数据分析鉴定出41种显著上调和28种显著下调的miRNA，包括首次与低氧相关报道的hsa-miR-4521、hsa-miR-145-3p和hsa-miR-222-5p。将HIF-1α和HIF-2α的ChIP-seq数据与表达数据整合显示，结合位点与miRNA上调之间总体存在关联，以hsa-miR-210-3p以及miR-27a/23a/24-2和miR-30b/30d簇的miRNA为主要实例。此外，发现hsa-miR-27a-3p和hsa-miR-24-3p的表达与乳腺癌中的低氧基因特征呈正相关。基因表达分析表明pri-miRNA和miRNA表达之间没有完全协调，表明存在其他调控水平。发现几种参与miRNA加工的转录本受低氧调控，其中Dicer（下调）和AGO4（上调）是HIF依赖性的。在乳腺癌中发现Dicer表达与低氧呈负相关。
在一个模型细胞系中对miRNA、mRNA和ChIP-seq数据的综合分析支持以下假设：低氧条件下miRNA表达在转录和转录后水平受到调控，miRNA基因组位点上HIF结合位点的存在与上调相关。鉴定与乳腺癌相关的低氧和HIF调控的miRNA对于我们理解疾病发展和设计治疗干预措施具有重要意义。
