Crispr-Cas9 Induced MLL-Rearrangements Cause Clonal Outgrowth in CD34+ Hematopoietic Stem Cells

Reciprocal chromosomal translocations are the causative genetic aberration in almost 60% of the pediatric acute myeloid leukemia cases. Amongst these, rearrangements of the MLL1/KMT2A gene are most frequent. Retroviral overexpression of MLL fusion genes has been shown to be sufficient to transform human hematopoietic stem and progenitor cells (HSPCs). Whether endogenous MLL-rearrangements have a similarly potent transformation capacity remains an open question. As an emerging technology, the clustered regularly interspaced short palindromic repeats (CRISPR) - CRISPR-associated-9 (Cas9) system now offers the opportunity to engineer chromosomal rearrangements, allowing the investigation of fusion oncogenes in the endogenous context. The successful transfer to the target cell type represents the only limitation.

With the aim to elucidate the transformative nature of endogenous MLL-rearrangements in primary human HSPCs, we developed and advanced an all-in-one lentiviral CRISPR-Cas9 system with two sgRNA expression cassettes (LentiCRISPR-CT2.0). The improved lentiviral architecture with additional viral enhancer elements yielded a vector capable of producing higher-titer virus (2.5-fold; p=0<0.0001), compared to our previously published vectors. Utilizing established reporter-based sgRNA testing, we selected highly efficient sgRNAs targeting MLL1 and ENL intronic sequences (cleavage rates >80%) to generate the t(11;19)/MLL-ENL translocation. T7 endonuclease assays for the top five off-target sites and the on-target sites of our pre-selected sgRNAs verified high on-target and no detectable off-target activity at the endogenous loci. Dual sgRNA expression from a H1 promoter in combination with a U6 promoter was incorporated thereby establishing an efficient, recombination- and off-target-free all-in-one lentiviral CRISPR-Cas9 system for induction of chromosomal rearrangements.

Based on these results, we tested generation of chromosomal rearrangements in hematopoietic cell lines. MLL-ENL transcript and the genomic breakpoint were robustly detectable in the transduced bulk population (K562 cells).

To determine the impact of endogenous MLL-ENL on HSPCs, we transduced cord blood derived CD34+ HSPCs. In three independent experiments using methylcellulose-based colony-forming assays, MLL-ENL expression was detectable, resulting in a rearrangement efficacy of at least 1.58x10^-3 (detection/total colony number). MLL-ENL containing cells, verified on DNA and RNA level, had an extended -but not unlimited- replating capacity. Our experiments thus provide strong evidence that endogenous MLL-ENL translocations provide a growth advantage and limited self-renewal to human HSPCs. These findings were further supported by clonal outgrowth in one out of two experiments performed in liquid culture.

Transformation by MLL-rearrangements is guided by up-regulation of HOXA genes and their co-factors MEIS1 and PBX3. In line with these findings, MLL-ENL harboring cells showed robust up-regulation of HOXA9, HOXA10, MEIS1, and PBX3. Interestingly, genes associated with leukemic stem cell activity (CBX5, HMGB3, MYBL2) after retroviral MLL fusion gene expression in mice, were found down-regulated in our study. This finding highlights crucial differences to the previous, retrovirus-based studies in mice and the need to study chromosomal rearrangements at their endogenous locus in the primary human cell context.

With the results of our in vitro studies, we next aimed to interrogate the transforming capacity of endogenous MLL-rearrangements in vivo. CD34+ HSPCs, freshly transduced with the LentiCRISPR-CT2.0, were transplanted into immunodeficient mice. Detection of MLL-ENL genomic breakpoints in the mice (8 weeks post transplant) strongly supports our in vitro findings of successful HSPC modification and underlines the power of our approach. Further follow up of our in vivo studies will yield new insights on clonal behavior and downstream events of endogenous MLL-rearrangements in human HSPCs.

In aggregate, our study uncovers the oncogenic potency and limitations of endogenous MLL translocations in human HSPCs and highlights the power of the CRISPR-Cas9 system to generate precise cancer models, which will allow us to test the efficacy of targeted therapies, and to investigate the mechanisms of drug resistance in vitro and in vivo.