Human cancers contain numerous genetic abnormalities, a subset of which drive tumour initiation or progression. Alternatively, cancer cells can develop dependence on genes that are not genetically altered but contribute to maintaining tumour cell viability. Identifying these genetic vulnerabilities is critical for designing targeted therapeutics. The emergence of RNA interference (RNAi) as a tool to study cancer cell biology has rapidly expanded to high-throughput screening formats that enable the systematic interrogation of potential genes involved in tumourigenesis. This article discusses the application of genome-wide, high-throughput
RNAi-based screens for the identification of essential genes in cancer cell proliferation and survival.
by Dr M. A. Medrano, Dr F. S. Saiz and Dr R. Rottapel
Over the last decade, the advent of high-throughput technologies including DNA microarrays, comparative genomic hybridisation (CGH) and single-nucleotide-polymorphism (SNP) genotyping, has enhanced our understanding of complex biological systems. These technical advances, coupled with the sequencing of the human genome, have enabled researchers to progress significantly in characterising the structure of the cancer genome. Large-scale efforts by the Cancer Genome Atlas (TCGA) have uncovered hundreds of mutations, gene copy number variations and chromosomal alterations in human cancers. Sequencing projects like those of the TCGA are being used to develop improved therapy and provide insight into some of the genetic disturbances cancer cells have adopted. The use of cancer-specific genetic mutations for therapeutic benefit includes the success of Gleevec, a tyrosine kinase inhibitor that blocks the BCR-Abl fusion product, for the treatment of leukaemia. Understanding the molecular networks that contribute to cancer pathogenesis in this way can be exploited to provide substantial clinical benefits. Despite the success of this type of approach, it has to be remembered that human cancers are hosts to hundreds of genetic mutations, but only a subset of these genes have the ability to influence tumour progression and maintenance.
Genetic complexity of cancer
Cancer is a genetically complex disease that is not only affected by the individual genetic lesions it carries but also by the interaction of all these lesions with genes that themselves may not be altered at the genomic level – creating a network of disrupted processes [1]. As a consequence of these modified genetic interactions, a cancer cell can acquire unique dependencies required for proliferation and survival that are distinct from non-cancerous cells. These critical bottlenecks that control information flow in the cancer cell genetic network may represent selective therapeutic targets. How have cancer-associated mutations rewired the normal cell signalling network? How do we identify points of vulnerability within the cancer genetic network? And how can we exploit this information for therapeutic benefit? These questions can be addressed through the use of RNAi screening technologies, which employs a genetic approach to identify cancer cell vulnerabilities.
RNA Interference (RNAi) : overview
RNAi is a highly conserved, genetic process responsible for the post-transcriptional, sequence-specific down-regulation of genes. Double-stranded RNA (dsRNA) is cleaved into short interfering RNA (siRNA), approximately 21-23 nucleotides in length, by the RNase-III enzyme, Dicer. One strand of the siRNA, known as the guide strand, is incorporated into a group of effector proteins called the RNA-induced silencing complex (RISC). Argonaute, the catalytic component of RISC, cleaves mRNAs complementary to the guide strand, thereby modulating the expression of the protein encoded by the specific targeted gene [Figure 1]. Since siRNAs are short-lived in a cell, they provide only transient knockdown of a target gene. Moreover, transfection efficiency of RNAi oligonucleotides is highly cell-type dependent. Therefore, the application of siRNAs in long-term culture systems using high-throughput formats is limited. To overcome these limitations, several groups have developed retro- and lentiviral-based short hairpin RNAs (shRNAs) that stably integrate into the host-cell genome – allowing for long-term, stable knockdown of the target gene [2 - 4]. The shRNA is transcribed and processed by the cell’s RNAi machinery to function like endogenous siRNA. Lentiviral shRNA libraries targeting most of the known genes within the genome have been developed. Typically, each vector is marked with a unique nucleotide barcode that can be used to deconvolute the library using microarray technology.
shRNA Libraries for genome-wide screening
The current library we are using was developed at the Broad Institute of MIT and Harvard through the RNAi Consortium (TRC) [5]. The library is composed of over 54, 000 individual shRNA-expressing lentivirus targeting approximately 11, 000 human genes. A lentivirus that infects a cell integrates its genetic material containing the shRNA sequence into the host-cell genome. The expression of the shRNA is driven by the human U6 promoter and processed into an active siRNA. By using pooled RNAi libraries, we are able to question the importance of 11, 000 genes in a particular cell line for proliferation and survival in one single experiment. Cultured cells are infected at a low multiplicity of infection such that each cell will harbour on average, a single lentivirus. Non-infected cells are removed from the population by puromycin antibiotic selection since each infected cell will express the puromycin resistant gene. Each cell containing a unique gene-silencing signature is then allowed to compete with other cells in culture for survival. Cells that harbour an shRNA, which targets a gene required for proliferation or survival, will be at a competitive disadvantage and will become under-represented in the population over time. The shRNA harboured by each cell marks that cell with a unique nucleotide identifier characteristic of each hairpin. Genomic DNA is collected over several time points to retrieve the hairpin markers remaining in the population. After a restriction enzyme digest and PCR amplification, the products are labelled and hybridised onto a custom-designed microarray. The microarray probes are used to detect the presence or relative loss of each of the hairpin markers within the library. Those shRNA markers that have depleted from the library over time represent cells that have died or proliferated at slower rates relative to the bulk population of cells [Figure 2]. This method can be used to identify genes, gene classes, pathways, processes and chromosomal regions that are essential for tumour cell growth and survival under in vitro growth conditions.
Ovarian cancer
We are using RNAi screening as a tool to investigate novel biology in serous ovarian cancer (SOC), a histological subtype of epithelial ovarian cancer (EOC). Epithelial ovarian cancer represents about 90% of all ovarian cancer cases. The most lethal and common subtype is serous ovarian cancer, occurring in more than 50% of EOCs yet responsible for approximately 80% of mortalities [6]. Ovarian cancer is a particularly difficult disease to diagnose early considering the organ is deep inside the body and there are no widely available biological screening markers for early detection. As a result, 70% of women are diagnosed with advanced stage disease (Stage III and Stage IV), meaning that the cancer has already metastasised from the pelvic region through the lymphatic vessels to distant organs. The five-year survival rate at this advanced stage is only 25%. Current therapy using surgical debulking and cytoreductive surgery combined with post-operative taxol- or platinum-based chemotherapy is initially highly effective but 70-90% of patients will have recurrence and die of their disease [7]. With such a poor prognosis, there is a pressing need to shape new strategies for developing therapies.
Future prospects
The detection of genetic interactions essential for cancer cells to survive will provide a new outlook for treatment opportunities, and this can be made possible with the use of shRNA screening. Libraries of RNAi are allowing researchers to interrogate the functional importance of nearly every gene in the human genome in high-throughput formats. Using this technique, laboratories will collect a compendium of functional genetic data that can be contextualised with ongoing sequencing efforts – leading to the rapid and efficient identification of potential pharmaceutical targets with greater tumour cytotoxic effects while sparing normal cell types. As RNAi tools continue to develop, high-throughput screens will be used in a wide-range of experimental contexts, providing insights into a multitude of biological phenotypes.
Acknowledgments
This publication was made possible by funding from the Terry Fox Research Institute (TFRI), the Ontario Institute for Cancer Research (OICR), the Ontario Ministry of Research and Innovation, and the collaborative effort with Dr. Jason Moffat, Dr. Troy Ketela and Dr. Benjamin Neel.
References
1. Luo J, Solimini NL, Elledge SJ. Principles of Cancer Therapy: Oncogene and Non-oncogene Addiction. Cell 2009; 136; 823-837.
2. Root DE et al. Genome-scale loss-of-function screening with a lentiviral RNAi library. Nature Methods 2006; 3; 715-719.
3. Silvia JM et al. Second-generation shRNA libraries covering the mouse and human genomes. Nature Genetics 2005; 37; 1281-1288.
4. Bernards R, Brummelkamp TR, Beijersbergen RL. shRNA libraries and their use in cancer genetics. Nature Methods 2006; 3; 701-706.
5. Moffat J et al. A Lentiviral RNAi Library for Human and Mouse Genes Applied to an Arrayed Viral High-Content Screen. Cell 2006; 124; 1283-1298.
6. Karst AM, Drapkin R. Ovarian Cancer Pathogenesis: A Model in Evolution. Journal of Oncology 2009; 2010; 932371.
7. Cannistra SA. Cancer of the Ovary. The New England Journal of Medicine 2004; 351; 2519-2529.
The authors
Dr Mauricio A. Medrano1,2, Dr Fernando S. Saiz1, Dr Robert Rottapel1,2,3,4,5
1Ontario Cancer Institute/Toronto Medical Discovery Tower, Toronto, Canada, M5G1L7
2Department of Medical Biophysics, University of Toronto, Toronto, Canada
3Department of Immunology, University of Toronto, Toronto, Canada
4Department of Medicine, University of Toronto, Toronto, Canada
5St. Michael’s Hospital, 30 Bond St., Toronto, Canada, M5B 1W8
Corresponding author:
Dr R. Rottapel
Ontario Cancer Institute/Toronto Medical,
101 College St
Toronto, ON, Canada M5G 1L7
email: rottapel@oci.utoronto.ca
