Research Grants Awarded
Genome-wide mapping of MicroRNA binding sites in breast cancer cells and associated fibroblasts
Tumor Cell Biology VI
Background: miRNAs are small genome-encoded non-coding RNAs (ncRNAs) that regulate gene expression through sequence-specific interactions with target genes. More than 470 miRNAs have been identified in humans, and many of the miRNA genes are located at chromosomal fragile sites associated with cancer. Because experimental identification of miRNA targets is difficult, there has been an explosion of computational target predictions. Each of the ~470 miRNAs is predicted to bind ~200-1000 target mRNAs. On a small throughput, candidate targets were confirmed via reporter experiments, and sometime even through RNA immunoprecipitation (RNA-IP) of miRNA complexes, by pulling down tagged DICER and AGO (protein components of the miRNA complexes). More recently, by combining chromatin immunoprecipitation (ChIP) with tiling arrays, chromatin regions were identified, both proximal and distal to promoters, which are directly bound to transcription factors, or modified chromatin proteins, such as (methyl)3-K4H3 Histones and (methyl)3-K27H3. Objective: Genome-wide detection of miRNA-bound genomic sites in both mRNA and chromatin of breast cancer cells, using RNA-IP or ChIP, on-chip approach. Specific Aims: 1) To identify genome wide distribution of miRNA complexes on breast cancer RNA. 2) To identify miRNAs associated with nuclear genomic chromatin. 3) To identify genome wide distribution of miRNA complexes on breast cancer genome-DNA. Study design: To identify mRNA-targets of miRNA complexes, we will RNA-IP breast cancer extracts (as above), and derived cDNA loaded on commercial Human microarrays, which interrogates sequences located closer to the 3' untranslated region of transcripts (usual location of miRNA-binding elements). The RNA-IP signal would be detected if the binding element is within ~500 bps from the microarray probe set. To assess incorporation of miRNAs in euchromatin and heterochromatin, we will utilize ChIP of (methyl)3-K4H3 Histones and (methyl)3-K27H3 Histones, correspondingly. Pooled down material will be proteinase K and DNAse treated, and enriched RNA will be assessed on commercial miRNA arrays. If we detect miRNA-enrichment with heterochromatin we would identify target genes for this modification in breast cancer cells, by performing ChIP-on-chip experiments, with antibodies against DICER and AGO. We will load the product DNA on commercial genome tilling microarrays that interrogates 25,500 promoters (generally 7.5 Kb upstream and 2.5Kb downstream from transcription start site), and ~59% of CpG islands. If miRNA-association is observed on either the transcript-3' untranslated regions or Promoter platforms, we will repeat the experiment on tilling arrays that interrogate the full human genome. Potential outcome and benefits of the research: The data generated in this study is aimed for publicly available web-interface databases, and would assist the growing interest in this new potential targets for breast cancer treatment.
The central dogma of molecular biology alludes to a flow of sequence information through three macro-molecules; Eukaryotic DNA-dependent RNA polymerases produce RNA from the genomic DNA that ultimately code for proteins. The sequence of proteins dictates the phenotype of cells. A new paradigm has emerged in biology (2006 Nobel award) in which RNA molecules, rather than coding for proteins (noncoding RNAs-ncRNAs), actively participate in regulation and catalysis of biological processes. These ncRNAs associate with a number of proteins to produce these effects, proteins that can be pooled out of a complex extract via affinity reagents, such as antibodies. Recently, a genomic site of aberration associated with breast cancer phenotype, was found to code for MicroRNAs (miRNAs), a class of ncRNAs that base-pair with target genes, and attenuate protein synthesis from target-RNA. The targets of such miRNAs are commonly identified via computational analyses. Genomics has the potential to revolutionize the diagnosis and management of cancer by offering an unprecedented comprehensive view of the molecular underpinnings of pathology. Computational analysis is essential to transform the masses of generated data into a mechanistic understanding of disease. Occasionally, however, the genetic element responsible for the disease involves an unexplored biological process, the genomics data is obscured to us. Such is also the case with mutations that promote breast cancer, such as BRCA1 or BRCA2, that map to non-coding regions of the gene. We first propose to use genomics to define ncRNA distribution within active and inactive genomic areas. We further propose to determine a large fraction of the miRNAs binding elements in breast cancer cell genome, using a genomic, currently unprecedented approach. Our approach is a modification of an existing method in the genomics field that explores protein-nucleic acid interactions directly, using antibodies as described above. We propose to use this approach RNA Immunoprecipitation, to enrich for the complexes formed with miRNAs, and explore the enriched genomic sequences, using whole genome arrays. After optimizing the method on breast cancer cell lines, we wish to explore the corresponding patterns of miRNA binding sites in DCIS and invasive breast cancer-derived epithelial and fibroblast cells. We propose to make this data available through web-interface, for future genetic screens to compare their loci with this data, in case a locus that codes for a breast cancer associated phenotype is dictated by altering the miRNA target recognition. This data would be used to custom microarrays that are directly aimed at profiling the miRNA binding distributions in different biological conditions, and would provide essential experimental information to improve the currently existing algorithms that identify miRNA binding sites by computational analysis.