RNAi-Based Functional Genomics Screening of the Human Kinome Identifies Major Growth Regulating Kinases in Acute Myeloid Leukemia Cells

In order to improve treatment strategies for Acute Myeloid Leukemia (AML), we adapted a functional genomics approach using RNAi screening to identify molecular targets that are vital to the growth of AML. Herein we report the first large-scale kinome gene silencing screen in AML. A high throughput RNAi screen was developed for the efficient siRNA transfection of AML cell lines. Eight commercially available cationic lipid-based transfection reagents were tested for their ability to transfect several AML cell lines with siRNA. These extensive transfection optimization experiments identified two AML cells lines TF-1 and ML4 with up to 95–100 and 70–75% transfection efficiency respectively. Two independent replicate kinome screens were performed on both cell lines using a siRNA library targeting 572 kinase genes with 2 siRNA/gene. At 96 hours post transfection, cell proliferation was assessed and the B-score method was used to background correct and analyze the screening data. Several siRNA to specific kinases were identified that significantly inhibit cell proliferation of up to ~40–88%. Hits were defined at two thresholds: siRNA having a B-score of <−2 providing a statistically significance of p<0.05 (confidence of > 95%) and a cutoff B-score of <−1.5 providing greater than 87% confidence for each siRNA hit. Two different kinases (2 siRNA/gene/screen) were identified as major growth regulating kinases in TF1 cells with all 4 siRNA/gene having a B-score <−2. For two additional kinases, 3/4 siRNA for each gene had a Bscore <−2. Expanding the cutoff to a B-score <−1.5 three further kinases were targeted by at least 3/4 siRNA/gene. Similar analysis using the same criteria for ML4 cells identified one kinase targeted by 3/4 siRNA at a B-score <−2, seven kinases with 2/4 siRNA <−2 and two kinases with 3/4 siRNA/gene at a B-score of <−1.5. Common hits for both cell lines with at least 6/8 siRNA per gene from 4 screens performing at a B-score <−2 identified two kinases, one of them PLK1. Applying a B-score threshold of <−1.5, we identified five kinases for which at least 5/8 siRNA/gene from 4 screens met these criteria. Kinases/genes will be presented at the meeting.Confirmation of gene silencing and validation of growth response is currently underway for a subset of genes. Among the strongest hits are siRNA targeting PLK1, as well as siRNA targeting three other kinase-genes involved in regulating cell cycle progression and checkpoints and gene ontology (GO) analysis showed enrichment in cell cycle and cell cycle-checkpoint processes. Inhibitors against PLK1 and other kinase hits identified in the screen are in (pre)-clinical development and if confirmed, our experiments provide a strong rational to test these in AML. The application of RNAi based screening is useful in the identification of genes important in AML proliferation, which could serve as targets for therapeutic intervention and guide AML drug development. Furthermore, results from these types of functional genomics approaches hold promise to be rapidly translated into clinical application.