CRISPR library screening to develop HEK293-derived cell lines with improved lentiviral vector titers

Figures

• 

  FIGURE 1. Workflow of CRISPR knockout library screening to identify genomic targets affecting LV vector production. The CRISPR knockout (Brunello) plasmid library (top panel left) was first subcloned into a LV vector plasmid backbone containing an intact (wild type) 3′ LTR (top panel middle). LV vectors were produced by transfecting this vector plasmid library plus plasmids encoding LV packaging functions and VSV-G into HEK293T cells (top right panel) to generate LV vector particles. The resulting LV vector library was then used to transduce HEK293 cells at an MOI of less than 0.4. LV vector particles were rescued from transduced cells by rescue transfection using LV packaging and VSV-G-encoding plasmids. This process of transduction followed by rescue transfection was iterated. The gRNAs encoded by the LV vectors were sequenced by performing RT-PCR using vector RNA from concentrated vector stocks or by using plasmid DNA as a template. PCR products were then sequenced.

• 

  FIGURE 2. Characterization of the modified LV vector CRISPR knockout library. The fold-change in gRNAs targeting essential genes was determined. The y-axis shows the fold-change in gRNA representation of essential genes, where a value of 1 indicates that the number of reads in the library mapping to essential genes is proportional to their abundance in the library. The x-axis shows the progress of the workflow, where each name represents a pool of PCR products that was sequenced, starting from the initial knockout plasmid library from Addgene (referred to as Addgene Brunello), the plasmid library involving a LV vector plasmid with a wild type LTR (referred to as “Cloned Brunello”), lentiviral particles produced by using the LV vector plasmid library (referred to as “third Gen Lenti”) and ending with iterative rounds of viral rescue (referred to as “Rescue 1-8”).

• 

  FIGURE 3. Abundance of enriched gRNAs over time. The line chart shows the representation of the top 255 gRNAs throughout the screen, where the y-axis shows the percentage of the mapped reads attributed to an individual gRNA among all the mapped reads. The x-axis shows the progress of the workflow as described in Figure 1.

• 

  FIGURE 4. Workflow of knockout cell population generation. HEK293 cells were first transduced with LV vectors targeting selected exons or a control locus. Transduced cells were then selected by puromycin selection. Cas9 expression was induced by introducing doxycycline. After a recovery period, cells were co-transfected with an LV packaging plasmid and a VSV-G-encoding plasmid to assess the titers of LV vectors produced.

• 

  FIGURE 5. Characterization of bulk knockout populations. (A) 30 gRNAs enriched in the library screen that were selected for further characterization. The Top Hits category are the genes with individual gRNA that were most enriched. Multiple hits show genes that had multiple gRNAs within the top 500 of library hits. The HIV interactions category shows genes that had enriched gRNA within the top 1,000 of library members and may increase HIV titers when perturbed based on previously described HIV interactions (Ptak et al., 2008; Fu et al., 2009; Pinney et al., 2009). Genes downregulated in HEK293T vs. HEK293 lists genes downregulated by a factor of four or more within the top 1,000 library members (Lin et al., 2014). (B) LV vector titers of bulk knockout populations from panel (A) relative to cells transduced with gRNAs targeting a safe harbor locus (Chaikind et al., 2016). Genes from panel (A) listed in gray were not tested in panel (B). Bars represent standard error of the mean of three biological replicates.

• 

  FIGURE 6. Gene enrichment of Brunello and Calabrese gRNA after 8 rounds of rescue transfection. Read counts were analyzed by MAGeCK (Li et al., 2014), where reads generated from cloned library plasmids were classed as controls and reads generated from viral supernatant were classed as treatment. Outputs of the most enriched genes are shown in plots: (A) Brunello library enrichment, (B) Calabrese Set A library enrichment and (C) Calabrese Set B library enrichment, where the y-axis indicates p-value and the x-axis represents genes ordered from most to least enriched. (D) Brunello library enrichment, showing top 50 most enriched genes on x-axis. (E) Combined Calabrese library enrichment, showing top 50 most enriched genes on x-axis and average p-value on the y-axis.