MicroRNAExpressionProfilingbyBeadArray1

MicroRNA Expression Profiling by Bead Array Technology in Human Tumor Cell Lines Treated with Interferon-Alpha-2a

MicroRNAs are positive and negative regulators of eukaryotic gene expression that modulate transcript abundance by specific binding to sequence motifs located prevalently in the 3′ untranslated regions of target messenger RNAs (mRNA). Interferon-alpha-2a (IFNα) induces a large set of protein coding genes mediating antiproliferative and antiviral responses. Here we use a global microarray-based microRNA detection platform to identify genes that are induced by IFNα in hepatoma- or melanoma-derived human tumor cell lines. Despite the enormous differences in expression levels between these models, we were able to identify microRNAs that are upregulated by IFNα in both lines suggesting the possibility that interferon-regulated microRNAs are involved in the transcriptional repression of mRNA relevant to cytokine responses.

Key Words: MicroRNAs - Oligonucleotide Array Sequence Analysis - Interferons - Melanoma - Hepatoma - Reverse Transcriptase Polymerase Chain Reaction - Suppressor of Cytokine Signaling Proteins

Introduction

The gene expression patterns of tumor-derived cell lines differ greatly, as do their responses to antiproliferative effects of interferons (IFNs). The cause of this variation has been under investigation for more than 40 years, but only basic regulatory mechanisms of interferon signaling are understood today. Small regulatory genome encoded RNAs, such as microRNAs, have recently attracted attention in genomic research. New methods to analyze the levels of these regulatory elements are now commercially available, but the power of these techniques is still discussed extensively. Our study was designed to compare two methods for microRNA detection with respect to usefulness in defined cell culture assays. The experimental design assesses variation between the two cell lines and the treatment effects of IFNα.

A hallmark of the therapeutic activity of type I interferons is the induction of antiproliferative activity mediated by the upregulation of several hundred response genes with pleiotropic functions (1). These genes can be divided into two major classes based on the kinetic properties of induction (2). Primary response genes (PRGs) are upregulated within 24 h after the cytokine signal and the secondary response genes (SRGs) are induced following day 1 when the activity of the PRGs decays. In contrast to SRGs, all PRGs studied to date contain bona fide interferon response elements in the promoter region, which are required for binding of the interferon-stimulated gene factor 3 (ISGF3) complex and for janus kinase/signal transducer of transcription (JAK/STAT)-pathway-mediated signaling.

Expression of PRGs is turned off by proteins termed suppressors of cytokine signaling (SOCS) (3). As the nomenclature indicates, this class of polypeptides has the capacity to interfere and silence other cytokine-induced signaling cascades (for review see (4)). SOCS1 for instance is part of the early inducible PRG cluster and down modulation occurs together with the other genes before onset of SRG expression. It is believed that feedback inhibition of JAK/STAT signaling by SOCS1 represses transcriptome modulation of IFNα signaling (5). Regulation of SOCS protein translation by interferon-regulated microRNAs (IRmiRs) would enhance the potential of cytokine fine regulation. It has been reported that miR-19 antagonists lead to higher SOCS1 levels and miR-19 mimics can repress SOCS1 reporter constructs, thus obviously supporting the bioinformatic predictions that SOCS1 is a direct target of miR-19 (6). Inhibition of SOCS activity could for instance prolong the duration of cytokine activity, which has obvious clinical implications.

Following the discovery of microRNAs in virtually all higher eukaryotic organisms significant research efforts were initiated to address the function of these catalytic oligonucleotides which are the natural counterparts of synthetic small inhibitory RNAs (siRNAs) used for experimental gene silencing (for review see (7)). MicroRNAs are positive and negative modulators of the expression of entire gene clusters that contain complementary microRNA recognition sequence motifs in the 3′-UTR. Today, prediction of microRNA target genes by homology-based algorithms is still ambiguous (8). The activity of one or several microRNAs could explain suppression of the entire PRG cluster provided that microRNA abundance is regulated by IFNα. Alternatively, microRNA-mediated degradation of transcripts encoding negative regulatory proteins would also abolish PRG expression and restore IFNα responsiveness.

Some recent reports showed that interferon beta (IFNβ) stimulation can boost microRNA levels in cell culture together with inhibition of viral replication (9). At this point it is an open question whether this induction is IFNβ specific or a shared feature of all type I interferons. To investigate whether microRNA are also involved in regulation of IFNα response, we used two human-tumor-derived cell lines: the melanoma line ME-15 (10) and the hepatoma line HuH7 (11). We have chosen these cell lines as models, because we have a good understanding of the IFNα responses at the mRNA and the protein levels in these cell lines. Further we chose to use a melanoma cell line because IFN is also used for treatment of this cancer type. HuH7 is commonly used as a model for testing antiviral effects of IFN in the HCV replicon system. In both models efficient responses to IFNα have been shown at the functional and transcriptional level. IFNα response genes carry response elements in their promoter region and these motifs are responsible for gene expression with similar efficiency in many cell types. Therefore we expected to find a similar regulated set of genes in both lines given that IRmiR genes are regulated by the same mechanism, whereas some constitutively expressed microRNA genes were expected to be cell type specific for functional reasons. We have chosen a DNA-microarray-based technology (Illumina) for the multi-parallel expression analysis of all known human microRNAs (http://microrna.sanger.ac.uk/; Release 10.0: August 2007). This method allowed us to process total RNA as template, allowing the possibility of mRNA gene expression profiling in further experiments. Briefly, annealing of microRNA specific primers combined with enzymatic polyadenylation allows multi-parallel polymerase chain reaction (PCR)-mediated amplification of individual microRNAs. The output of this step is a DNA amplicon library that reflects to a large extent the original stoichiometry of mature microRNAs in a cell or tissue (12). PCR amplification is performed with fluorescently labeled primers, which allows quantitative signal detection by conventional confocal laser scanning.