Genome-Wide RNAi Screens with Human and Mouse Pooled shRNA Libraries

Overview

The past few years has witnessed the development of new technological platforms enabling the creation of genome-scale RNAi libraries and methodologies for their use in functional genomic research. The ultimate goal is to produce genome-wide mammalian RNAi libraries that can achieve stable and specific gene knockdown in a wide variety of cell types, and to develop the requisite tools for their implementation in high-throughput screening assays.

The study of normal or disease processes in the context of the whole organism produces an enormous amount of information that can be complex to analyze. With cell culture systems, many stages in the progression of a disease, e.g. cancer, can be modeled at the cell-based level, thus enabling elucidation of the genetic mechanisms in a controlled environment. Researchers are now moving toward larger, cell culture-based screening projects (often genome-wide) to identify candidate genes involved in their biological question of interest. The flexibility of its applications has made RNAi a valuable tool for researchers who are interested in gene function characterization, signaling pathway analysis, and drug target validation. RNAi is particularly useful for characterizing molecular targets, screening genes, and looking at gene interactions with specific drugs, making it indispensable to the field of pharmacogenomics.

Lentiviral-based shRNA is an RNAi approach that facilitates delivery and screening in a wide variety of different cell types, including primary and non-dividing cells. While individual genes can be efficiently targeted using arrayed lentiviral libraries, large numbers of genes that effect phenotypes of interest can be screened by the introduction of a complex pool of shRNAs. These pooled screens are designed so that the majority of cells are transduced with a single shRNA, to aide in the process of hit deconvolution. In positive selection screens, cells are typically treated after transduction with some form of selective pressure. Survivors are identified by either colony selection or FACS analysis. The shRNA sequences, and thereby the corresponding gene targets, are determined from this surviving population by standard and deep sequencing or microarray analysis. Finally, the positive identification of hits is often used to develop hypotheses regarding the biological role of the corresponding gene(s). However, validation of leads by independent methods such as siRNA, small molecule inhibition, or complete gene knockout, is required.

The MISSION® TRC1 shRNA libraries from Sigma Aldrich, www.sigmaaldrich.com,consist of over 150,000 plasmid-based shRNA constructs, targeting 15,000+ human and 15,000+ mouse genes. The shRNA sequences are designed using an algorithm developed by the Broad Institute of MIT and Harvard University. On average, there are five shRNA designs for each gene. The MISSION LentiPlex Human Pooled shRNA Library and the MISSION LentiPlex Mouse Pooled shRNA Library are high-titer lentiviral shRNA pools. Both are generated from the RNAi Consortium's TRC1 collection of shRNA clones and are intended for rapid, whole-genome pooled RNAi screening projects.

Representation of individual shRNAs from the library is tested before product release to ensure adequate library coverage. Each library is provided as 2x ~25 µL aliquots of ready-to-use lentivirus, with titers of at least 5x10⁸ TU/mL (as determined via p24 assays), and is pre-divided into ten sub-pools containing approximately 8,000 shRNA constructs each. Amplification and sequencing primers are also provided for identification of hits.

Overview of Pooled Library Screen

The LentiPlex pooled library enables researchers to screen for novel phenotypes in relevant cell types. The study of gene function and disease in cells recalcitrant to RNAi delivery (primary, non-dividing, or terminally differentiated cells) is typically difficult to implement using standard transfection methodologies. However, lentiviral-based delivery mechanisms allows for screens to be done in these often more relevant cell systems. The five primary steps required for utilizing the LentiPlex-pooled library in a screen are outlined below.

The following is a brief description of each step.

1. Optimization of puromycin selection conditions.

To effectively eliminate colonies with no shRNA insert, optimization of puromycin selection conditions is suggested for each newly tested cell type. This is accomplished by performing an antibiotic kill curve to determine the optimal concentration of puromycin needed to eliminate untransduced cells. The lowest concentration of puromycin that kills all untransduced cells should be utilized for subsequent screening experiments. Utilizing higher concentrations of puromycin may lead to unacceptably high cytotoxicity and potential off-target effects.

2.Optimization of transduction efficiency.

To successfully identify shRNA sequences of interest it is critical to design the screen to ensure that each cell receives only one shRNA construct. By using a low multiplicity of infection (MOI of 1 or lower) the probability of multiple integrants per cell is greatly decreased. However, transduction efficiencies, and consequently the desired MOIs, depend strongly on the target cell type. Therefore, it is imperative to determine the optimal MOI for each new cell type prior to starting the screen. The optimal MOI can be determined by testing a range of MOIs using either MISSION pLKO.1-Puro or MISSION TurboGFP Control Transduction Particles on a fixed cell density.

3.Transduction and selection.

Transduced cells may be selected using puromycin before the positive selection pressure is applied to the screen. Whether or not puromycin selection is used depends on the type of screen being performed. For screens utilizing a reporter enzyme, the puromycin selection step will eliminate untransduced cells.This minimizes the number of cells that will need to be sorted in the end, thereby maximizing the dynamic range of the assay. For screens assessing changes in viability, the puromycin selection step may not be required if the selective pressure is sufficient to eliminate untransduced cells.

Another variable is the length of time between transduction and application of selective pressure. Shorter incubation times will result in fewer cells to maintain and to sort while longer times may result in a broader, more representative screen. A longer incubation provides sufficient time for the degradation of proteins with long half-lives and increases the number of targets that can be identified in the screen. Finally, a longer incubation time enables the detection of genes that influence pathways downstream of their immediate targets. This balance will be screen-specific and will depend on the type of targets that are being screened for. It is always important to include the most appropriate positive and negative controls for your particular screen.

4.PCR amplification.

Total genomic DNA (gDNA) is isolated from the selected and expanded cell populations and the shRNA inserts are amplified using the primers provided. The provided MISSION shRNA Human Positive Control Vector should be used as a template in a separate reaction in order to ensure that the PCR reactions are working optimally.

5.Identification of positive hits or deconvolution.

Sequence analysis of the PCR amplicons recovered from cells that show the phenotype of interest can be used to identify hits. There are three deconvolution strategies available to generate sequencing data from clonal or polyclonal population of cells:

1. From clonal cell population, standard sequencing.
2. From polyclonal cell population, standard sequencing and
3. Solexa deep sequencing.

For clonal gDNA screening, hits are identified, and colonies are isolated and expanded. Genomic DNA is then harvested from the clonal isolates, and PCR amplification is performed using the provided primers. LentiPlex PCR primers are designed to amplify the region of the viral insert containing the shRNA sequence. The adjacent figure shows the PCR amplification of gDNA from A549 cells transduced with individual shRNA clones. The expected size of the amplicon is 310 bp, with an amplification of pLKO.1 empty vector control showing a smaller size (Lane 4) due to absence of the shRNA insert in the vector. Following PCR, the products are sequenced.

PCR amplification of A549 colonies transduced with a single MISSION shRNA clone

Lane 1 – PCR Ladder
Lane 2 –PCR amplicon from clonal shRNA   construct A
Lane 3 –PCR amplicon from clonal shRNA construct B
Lane 4 –PCR amplicon from clonal pLKO.1 empty vector control

For polyclonal DNA, PCR amplification is performed on the gDNA harvested from the pool of positive clones, and all of the shRNAs are amplified. The resulting PCR fragments are subcloned and sequenced to identify individual shRNA. Standard sequencing of polyclonal gDNA may require additional processing time. Following isolation of the gDNA, PCR is performed to amplify all DNA containing shRNA inserts. While the PCR product is uniform in size (see picture below) it is actually a pool of different sequences because it originates from multiple clones.

The resulting shRNA sequences are identified using the LentiPlex shRNA Sequence Search database that provides the corresponding TRC number for each shRNA sequence. Additional information about the identified TRC shRNA sequence(s) and the corresponding gene targets can be readily found on Sigma-Aldrich's "Your Favorite Gene" site at http://www.sigma.com/yfg

Another approach for identification of hits from a pooled screen is to perform deep sequencing analysis on the gDNA isolated from the polyclonal cell population. Deep sequencing technology enables the rapid sequencing of complex, heterogeneous gDNA samples. Additionally, pooled samples may be multiplexed within the same run when tagging unique sequences to the PCR oligo sequences. Deep sequencing can deliver more than three million bases from a single sample preparation.

The raw sequencing data is queried against the TRC library database to match each individual sequence to its corresponding TRC clone. Additionally, the number of reads for each TRC clone is identified and the TRC clones are then grouped according to the gene target. The example of data that is shown here is the number of shRNA clones identified for the top 25 hits from one sub-pool of 10 sub-pools screened. From this screen, multiple shRNA clones targeting individual genes were identified. This demonstrates the reliability of pooled screen results, and focusing on hits with this type of redundancy can reduce the chance of pursuing false positives caused by possible off-target effects.

Analysis of deep sequencing data will require bioinformatics support, a service which is offered free of charge through Sigma-Aldrich.

MISSION LentiPlex pooled shRNA libraries combine the RNAi Consortium's shRNA collection with Sigma's lentiviral manufacturing expertise to enable easier, more affordable genome-wide RNAi screening options. The LentiPlex pooled screening system provides enhanced delivery and long-term gene silencing in non-dividing and primary cell types. Rapid whole-genome RNAi screens are now accessible to any researcher with minimal reagent, time, or capital equipment investment.