CRISPR-Cas9 Screening Identifies Therapeutic Targets for AML1-ETO Positive Leukemia

Objectives: Targeted therapy has achieved tremendous success in leukemia therapy over the past few decades, especially in those with fusion proteins, such as t(15;17) which generates PML-RARa and t(9;22) which generates BCR-ABL. However, effective targeted therapy for AML1-ETO-positive leukemia, which arises from one of the most frequent chromosomal translocations t(8;21), remains largely elusive. In this study, we performed CRISPR-Cas9 screening on the AML1-ETO-positive Kasumi-1 cell line using an AML1-related sgRNA library to identify potential therapeutic targets that could reverse the differentiation blockade in AML1-ETO-positive AML.

Methods: To identify the critical genes capable of reversing differentiation blockage in AML1-ETO-driven leukemogenesis, we constructed an AML1-related sgRNA library which focused genes and signaling pathways regulated by AML1-ETO. This sgRNA library consist of (1) coding genes of proteins interacting with AML1, (2) genes activated by AML1-ETO, and (3) genes downregulated upon knockdown of AML1-ETO as observed in CHIP-seq results. This sgRNA library comprises 4130 sgRNAs targeting 929 human genes (~4 sgRNA/gene), along with 100 control sgRNAs. CD13^high and CD13^low populations were collected by flow cytometry at 10th day post-transduction of sgRNA library. The enrichment of gRNAs between CD13^high and CD13^low group was analyzed using the MAGeCK program, and the candidate genes were identified based on the log2 fold change of sgRNA abundance. To make the screening results more reliable, we used CD11b as another screening marker to repeat the aforementioned process. We further preformed single candidate gene knockout in Kasumi-1 and SKNO-1 cell line to validate the screening phenotype. The regulatory function of candidate gene in AML1-ETO positive leukemia was explored through various in vitro and in vivo assays.

Results: The gene set enriched in the CD13^high group is considered to be positive hit. Silence of these genes may reverse blockage of differentiation, making them potential therapeutic targets for AML1-ETO leukemia. Additionally, genes such as CBFB, RUNX1, DOT1L, PRMT1, MED12 and KDM1A, which had been previously reported to promote myeloid differentiation when down-regulated, were also significantly enriched in the CD13^high group, these findings demonstrated the credibility of our screening strategy and results. Cyclin-dependent kinase (CDK) 12 emerged as one of the top-ranking genes in our study. CDK12 engages in diverse biological functions, including transcription, post-transcriptional modification, cell cycle regulation and cellular proliferation. Based on the information provided by the GEPIA and Bloodspot databases, CDK12 is highly expressed in the AML1-ETO-positive leukemia subtype and is associated with poor prognosis. The expression of differentiation-related surface markers in Kasumi-1 cells is upregulated in CDK12 knockout cells. Moreover, morphological analysis revealed a more mature cell morphology characterized by a decrease in nucleus-to-cytoplasm ratio and indented nuclei, suggesting that repression of CDK12 promotes the terminal differentiation of AML blasts. Cell biology experiments indicated that the knockout of CDK12 significantly inhibited the proliferation of Kasumi-1 and SKNO-1 cells, while accelerating apoptosis in these cells. Additionally, in vivo assays demonstrated that CDK12 knockout efficiently reduced tumor burden and prolonged the survival of the AML-ETO-positive leukemia mouse model. These findings indicate that the downregulation of CDK12 has a significant tumor-suppressive effect in AML1-ETO leukemia.

Conclusions: Our findings delineate the landscape of genes potentially involved in the regulation of differentiation in AML1-ETO-positive leukemia and highlight CDK12 as a promising therapeutic target for AML1-ETO-driven leukemogenesis.