Dana-Farber Cancer Institute (DFCI): Genome-wide shRNA Screens with DEMETER Inferred Gene Effects

In this study RNA interference (RNAi) screens were performed on 285 cell lines and combined with 216 lines previously screened, which were then analyzed together with DEMETER to discover genetic dependencies across the entire pool of cell lines.

Read the abstract

Experimental Approaches

DFCI extended their previous study of 216 cell lines1 by performing genome-wide pooled loss - of - function screening on additional 285 cancer cell lines across approximately 100k shRNAs (final files include 107,523 shRNA values in Achilles_v2.19.2 to produce 17,098 DEMETER gene solutions in Achilles_v2.20.2). Each cell line was infected with the shRNA pool by lentivirus, in quadruplicate and propagated for at least 16 population doublings or 40 days, whichever came first. To determine the viral volume needed to achieve the desired transduction rate of ~40%, each cell line was titrated with 6 volumes of virus (0-500 ul) in a 12 well plate at a concentration of 3E6 cells/well. Then cells were cultured in the presence or absence of puromycin in 6 well dishes before infection rates were determined. Cells were expanded for infection in quadruplicate with a target of 3.7E7 infected cells. Before infection, cells were filtered through a 40 um cell strainer to remove clumps, then resuspended in media containing 4 ug/ml polybrene, and the appropriate volume of 98K library lentivirus to achieve a cell concentration of 1.5E6 cells/ml. This cell suspension was seeded into 12 well plates at 2 ml/well and centrifuged for 2 hours at 930xg at 30 °C. After the spin infection, 2 ml of fresh media was added to each well. After 24 hours, the cells from each replicate infection were pooled into T225 flasks with 60ml medium containing puromycin. To provide an in-line assessment of transduction rate, 150k of infected and uninfected cells were cultured in 6 well dishes in the presence or absence of puromycin. After 96 hours, both the in-line assay wells and the screen replicates were trypsinized. The infection rate was determined by calculating the number of viable cells selected in puromycin divided by the number of viable cells without puromycin selection.

Screening was continued if the infection rates were within the range of 30–65% so that the selected cells were nearly all MOI = 1 and so that there was a sufficient number of cells to provide adequate representation of each shRNA. For each of the replicates, 6E7 cells were plated into new T225 flasks in 60ml of media with puromycin. For the remaining passages, only 3E7 cells per replicate were carried over, and the remaining cells were spun down and resuspended in PBS for genomic DNA isolation. Passaging for each cell line was continued for at least 16 population doublings or 28 days, whichever was longer. Puromycin selection was maintained until day 7. At the end of passaging, genomic DNA from the screen endpoints were used to measure the abundance of shRNAs in comparison to the initial DNA plasmid pool. Samples were sequenced using a custom sequencing primer using standard Illumina conditions. Deconvolution was performed similar to that described in Ashton et al.2 and all steps are described more completely in Cowley et al.1 with the following alterations. A total of 280 μg gDNA was used as template for PCR from each replicate. Thermal cycler PCR conditions consisted of heating samples to 95 °C for 5 min; 28 cycles of 94 °C for 30 s, 53 °C for 30 s, and 72 °C for 20 s; and 72 °C for 10 min. PCR reactions were then pooled per sample. After PCR and additional of sample barcodes, 20 replicates were multiplexed into a single Illumina sample, and run on multiple lanes to achieve a minimum of 27 reads per replicate.

Read the detailed Experimental Approaches

Data

Access the Raw/Analyzed Data (DCC)

 

For questions, please contact Joshua Dempster.  

References

  1. Cowley GS, et al. (2014). Parallel genome-scale loss of function screens in 216 cancer cell lines for the identification of context-specific genetic dependencies. Sci Data 1:140035. (PMID: 25984343)
  2. Ashton JM, et al. (2012). Gene sets identified with oncogene cooperativity analysis regulate in vivo growth and survival of leukemia stem cells. Cell Stem Cel.l 11,:359-372. (PMID: 22863534)
  3. Tsherniak A, et al. (2017). Defining a Cancer Dependency Map. Cell. 170 (3):564-576. (PMID: 28753430)
Last updated: September 14, 2018