Interrogation of Individual CLL Loss-of-Function Lesions By CRISPR In Vivo Editing Reveals Common and Unique Pathway Alterations

Mouse models represent invaluable tools for the systematic evaluation of cancer drivers, yet models that address the impact of putative genetic drivers of chronic lymphocytic leukemia (CLL) on B cell development and function are largely lacking.

To study recurrent loss-of-function (LOF) mutations observed in human CLL, we established a transplant model that can rapidly evaluate genetic lesions. First, we crossed mice carrying B-cell restricted Cre expression (Cd19-cre) with mice carrying conditional Cas9-GFP, to generate a strain expressing B cell-restricted Cas9 (Cd19-Cas9). Next, we optimized methods for in vitro engineering of early stem and progenitor cells (Lin^- Sca-1⁺ c-kit⁺ [LSK]) from Cd19-Cas9 mice using lentivirus expressing sgRNAs (mCherry⁺)targeting Atm, Trp53, Chd2, Birc3, Mga, or Samhd1. We chose LSKs because of their high transducibility and long-term repopulating potential. Last, we transplanted the single sgRNA-expressing LSKs into sub-lethally irradiated CD45.1 recipient mice, and then confirmed presence of ~45-85% gene-edited sequences (>70% carrying frameshift mutations) in edited B cells (GFP⁺mCherry⁺) at 2 months post-transplant, by PCR-based targeted deep sequencing and CRISPResso software analysis. We also verified presence of gene alterations (and putative off-target lesions) at the single cell DNA level (targeted sequencing by Tapestri, Mission Bio).

We first asked whether presence of the 6 LOFs could impact B cell developmental trajectories in marrow, spleen and peritoneum at 4 months post-transplant, a time point by which B cells are considered to achieve optimal host reconstitution (n=5/group, including a non-targeting control group). No marked changes were observed in mice with Atm^indel, Trp53^indel, Chd2^indel, Birc3^indel or Samhd1^indel, as analyzed by flow cytometry. Of interest, however, Mga^indel mice were detected to have increased germinal center (B220⁺CD95⁺CD38^-) and marginal zone (B220⁺CD21^highCD23^-) splenic B cells, and also showed increased B1a (CD5⁺ B220^low CD23^- CD43⁺) and decreased B1b (CD5^- B220^low CD23^- CD43⁺) cells in the peritoneum (p<0.05, ANOVA). These results indicate that the likely negative regulatory role that Mga exerts on MYC networks may directly impact germinal center formation and cell fate determination in B cells.

The overall abundance of edited B cells in spleen and blood of each group was higher (overall median: 17.0%; 90%CI 6.7-58.8%) than the non-targeting control (8.4%; 90%CI 1.6-14.2%) at 4 months post-transplant (n=8/group, p<0.05, ANOVA), and abundance of edited cells increased in peripheral bleeds at 4 vs. 2 months (n=8/group, p<0.05, Wilcoxon signed rank test). This suggests that presence of individual alterations can alter pro-survival pathways in mature B cells, through mechanisms that may, at least partly, be shared across LOFs.

To address this question, we analyzed the transcriptional profiles of edited B cell splenocytes (n=3/group), and compared them to their non-edited counterparts (GFP⁺mCherry^- splenocytes from the same animal), identifying a total of ~3900 differentially expressed genes among the 6 groups (p<0.05, paired Student's t test). Notably, changes in gene expression were highly concordant across 5 of the 6 groups (Spearman r >0.37 for each of the 10 pairs of 5 groups), with the exception of Mga^indel, consistent with its unique phenotype, observed in developmental studies. Gene ontology analyses using Enrichr confirmed commonalities in pathway dysregulations across the 5 similar groups of mice (p<0.05), such as modulation of Notch signaling in Chd2^indel, Samhd1^indel, and Birc3^indel, serine/glycine metabolism in Atm^indel, Trp53^indel, and Chd2^indel, and oxidative phosphorylation in Atm^indel and Samhd1^indel. Unique to Mga^indel, we saw enrichment of the GOs for transcriptional mis-regulation in cancer and cellular senescence, both relevant for tumorigenesis and B cell development.

In conclusion, we demonstrate that common LOFs typical of patients with CLL lead to increased cellular fitness in B-cell restricted mouse models, while dysregulating pro-survival pathways relevant to B cell development, CLL pathogenesis and more broadly to tumorigenesis. We are currently exploring phenotypic similarities and differences through tailored functional assays, while addressing the relative contribution of each alteration to CLL development in multiplexed edited mouse lines.