Characterization of a Human Bone Marrow-Derived CRISPR/Cas9 Cell Model for Chronic Myeloid Leukemia

The Philadelphia chromosome (Ph) is known as the hallmark of chronic myeloid leukemia (CML) and is formed by the reciprocal translocation t(9;22) (q34;q11). The result is a fusion of the ABL proto-oncogene 1 gene (ABL1) on chromosome 9 and the breakpoint cluster region gene (BCR) on chromosome 22. The BCR/ABL1 gene encodes for an oncogenic fusion protein with constitutively enhanced kinase activity. Deregulation of different molecular pathways by BCR/ABL1 leads to increased cell proliferation, genomic instability, and resistance to cell death. In the majority of CML cases, the breakpoint occurs within the major breakpoint region of the BCR gene, leading to the expression of the BCR/ABL1 transcripts e13a2 or e14a2. These transcripts encode for the fusion protein p210. In contrast, chromosomal breakage in the minor breakpoint region of the BCR gene results in the expression of the BCR/ABL1 transcript e1a2, which encodes the fusion protein p185 and is associated with poor a prognosis. Here we report the characterization of our CRISPR/Cas9-generated cell model for early-phase Ph⁺ leukemia.

We established a human CRISPR/Cas9-induced cell model to analyze the impact of the e1a2 and e13a2 transcripts on leukemogenesis. We used the CRISPR/Cas9 system to generate cell lines from human bone marrow (BM)-derived CD34⁺ hematopoietic stem and progenitor cells (HSPCs) from healthy donors. By using sgRNAs that target either the minor or the major breakpoint region of BCR, we induced expression of the e1a2 or e13a2 fusion transcript. After inducing t(9;22) in BM-derived HSPCs, the cells were cultured in liquid culture using cytokines for myeloid differentiation. After confirming the translocation by genomic PCR, FISH and karyotyping and validating the expression of the e1a2 or the e13a2 fusion transcript, the cells were further analyzed. By Western blot, we evaluated the phosphorylation status of common BCR/ABL1 target proteins. We performed tyrosine kinase inhibitor treatments to investigate whether imatinib and nilotinib have inhibitory effects on BCR/ABL1 kinase activity. Furthermore, we performed qRT-PCR to analyze the expression of potentially dysregulated genes. Finally, sublethally irradiated NOD scid gamma (NSG) mice were intrafemorally injected with t(9;22) cells three days after translocation induction. Engraftment of human cells was continuously monitored by FACS analysis of peripheral blood.

Induction of t(9;22) by the CRISPR/Cas9 system was successful in almost 100% of our attempts. Cells expressing the e13a2 transcript reached a proportion of up to 60% within the culture by day 60 after translocation induction. Cells expressing the e1a2 transcript reached a proportion of only 20% within the culture in the same time period. Western blot confirmed increased phosphorylation of STAT5, CRKL, and MAPK (ERK1/2) by BCR/ABL1 compared to healthy control (wt) cells from the same donor. The effect was reversed after 24 h treatment with increasing concentrations of imatinib (0.05 µM to 1 µM) or nilotinib (0.05 µM to 2 µM). qRT-PCR revealed a trend for the upregulation of MYC, HDAC3, BCL2L1, and TRIAP1 in t(9;22) cells compared to wt cells. After injection of wt, e1a2, or e13a2 cells into NSG mice, cell engraftment was monitored on days 44, 56, and 83 after injection by FACS analysis of peripheral blood. The percentage of huCD45⁺ cells in peripheral blood at day 83 after transplantation was highest in mice transplanted with e13a2 cells (4%) compared to mice transplanted with e1a2 (1%) or wt (1%) cells.

In summary, our CRISPR/Cas9-generated human cell line model allows, for the first time, a direct comparison of the BCR/ABL1 fusion variants e1a2 and e13a2 expressed under their endogenous promoter. Despite the cells exhibiting a proliferative advantage over the wt cells in culture and showing elevated BCR/ABL1 kinase activity, they fail to reach 100% purity in in vitro culture systems. By single-cell RNA sequencing, we will gain new insights into the transcriptome of the BCR/ABL1-expressing cells and be able to compare the effects of the different isoforms on a single-cell level. The ongoing in vivo experiments will shed light on the leukemia-initiating potential of e1a2 and e13a2. In the future, our model may facilitate a more comprehensive understanding of the biology of CML and the differences between p185 and p210.