Successful CRISPR genome editing in myeloid cells

Freund EC, Lock JY, et al. (2020) Efficient gene knockout in primary human and murine myeloid cells by non-viral delivery of CRISPR-Cas9. J Exp Med 217(7):e20191692.2.

Citation summary: Myeloid cells, which are immune cells such as dendritic cells, monocytes, macrophages, and granulocytes, are important in research on immunological therapies. These cells are naturally sensitive to viruses and external nucleic acids. For this reason, myeloid cells have presented a challenge to CRISPR genome editing involving viral vectors or stable expression of Cas9. Freund et al. have now demonstrated that ribonucleoprotein (RNP)-based delivery of CRISPR components gives reliably high levels of genome editing in both mouse and human myeloid cells, and they discuss future applications of these results.

Background

Myeloid cells are immune cells that are important in defense against microbes and tumors. Many varieties of myeloid cells exist, such as monocytes, dendritic cells, macrophages, and granulocytes. Methods for inducing modifications of these cells are important areas of research in the development of therapeutics for modulation of the immune system. Myeloid cells demonstrate natural challenges to genome editing via CRISPR, because their immune responses against viruses and external nucleic acids prevent reliable genome editing using plasmids and viral vectors. In this paper, Freund et al. describe how they improved the approach to CRISPR genome editing of myeloid cells.

Experiment

Bone marrow derived macrophages (BMDMs) can be obtained by culturing mouse bone marrow. BMDMs serve as an important model for the study of macrophage involvement in inflammation, infection, tissue regeneration, fibrosis, and similar events. Freund et al. first decided to determine optimal conditions for electroporation of BMDMs with Alt-R® S.p. Cas9 Nuclease V3 as an RNP targeting CD11b. They tested a matrix of parameters (buffers and instrument settings) using a Lonza 4D-Nucleofector® and then observed the expression level of this protein by flow cytometry. As a second step, the optimal parameters (Lonza buffer P3 and program CM-137) were verified in another myeloid cell type (monocytes) expressing eGFP by electroporation of these cells with Cas9 RNP targeting eGFP. Knockouts of eGFP and CD11b were both found to occur at the highest level when 2 gRNA sequences were used simultaneously. Based on these findings, Freund et al. proceeded with their studies of other myeloid cells.

The researchers reasoned that dendritic cells might respond differently to electroporation than BMDMs. To investigate optimal conditions for introducing CRISPR reagents into dendritic cells, 3 types of dendritic cells were first induced by differentiation from bone marrow cells. Freund et al. again tested a large number of parameters for electroporation with the 4D-Nucleofector, using RNP targeting CD45. They found that slightly different parameters were optimal for the 3 different subsets of dendritic cells, but each optimum set of parameters resulted in high editing efficiency. They also investigated how the expression level of a marker of normal myeloid activation, CD80, was affected by the electroporation treatment, and found that this was largely unaffected for 2 of the 3 types of dendritic cells.

In further studies, Freund et al. moved from mouse cells to human myeloid cells obtained from healthy blood donors. To determine whether there is a difference in editing efficiency between Alt-R CRISPR-Cas9 crRNA XT and the standard Alt-R CRISPR-Cas9 crRNA , they targeted the B2M gene in human myeloid cells obtained from healthy blood donors. They found that standard crRNA and crRNA XT resulted in similar, very high, editing efficiencies. As with the initial experiments in murine cells, they also found that the human cells showed highest knockout when 2 crRNA sequences were used in the same experiment. Replacing 1 of the targeting crRNA sequences with a non-targeting control did not result in a similarly high level of genome editing, indicating that the effect of using 2 crRNAs was specific.

Freund et al. also tested Alt-R CRISPR-Cas9 single guide RNA (sgRNA) targeting B2M and compared it to the use of 2-part guide RNA directed to the same target site. They found that in primary myeloid cells, sgRNA resulted in the highest levels of gene knockout. They also tested Alt-R Cas9 Electroporation Enhancer in combination with the sgRNA-containing RNPs and found that this allowed them to decrease the amount of RNP significantly while still obtaining the highest levels of genome editing efficiency.

Using their optimized conditions, Freund et al. knocked out TLR7 in mouse bone marrow cells, and checked the levels of downstream signaling cytokines. This condition resulted in decreases of TNF, IFNa, IL-12p40, and IL-6, demonstrating that the knockout not only knocked out the gene but also its function. They also showed that they could knock out B2M, CD14, and CD81 simultaneously in human monocyte-derived macrophages. Flow cytometry confirmed the triple knockout status.

Conclusion

Freund et al. pointed out that the parameters they optimized have the advantage of allowing simultaneous, rather than step-wise, knockout of multiple targets in human and mouse myeloid cells. They observed no chronic immune cell activation when they followed their protocols and suggested that cells which have undergone genome editing in this way may be suitable for adoptive cell therapy.

Published May 15, 2020