Efficient gene knockout in primary human and murine myeloid cells by non-viral delivery of CRISPR-Cas9

Abstract

Myeloid cells play critical and diverse roles in mammalian physiology, including tissue development and repair, innate defense against pathogens, and generation of adaptive immunity. As cells that show prolonged recruitment to sites of injury or pathology, myeloid cells represent therapeutic targets for a broad range of diseases. However, few approaches have been developed for gene editing of these cell types, likely owing to their sensitivity to foreign genetic material or virus-based manipulation. Here we describe optimized strategies for gene disruption in primary myeloid cells of human and murine origin. Using nucleofection-based delivery of Cas9-ribonuclear proteins (RNPs), we achieved near population-level genetic knockout of single and multiple targets in a range of cell types without selection or enrichment. Importantly, we show that cellular fitness and response to immunological stimuli is not significantly impacted by the gene editing process. This provides a significant advance in the study of myeloid cell biology, thus enabling pathway discovery and drug target validation across species in the field of innate immunity.

Graphical abstract

Figure S1.

Screening of optimal Cas9-RNP nucleofection protocols for KO in murine monocytes and BMDMs. (A) Workflow for generation of murine BMDMs and screening of crRNA/Cas9-mediated CD11b KO. (B) Heatmaps depicting relative impact of nucleofection conditions on cell viability and CD11b MFI. Red box indicates the best condition (Buffer P3, program CM-137). (C) Representative histograms depicting MFI of CD11b for indicated nucleofection conditions following 5 d of culture in BMDM medium. (D) Ranking of nucleofection conditions for crRNA/Cas9-mediated eGFP KO and viability in monocyte-derived macrophages following 5 d of culture. Green bar: Buffer P3, Program CM-137. Blue bar: Buffer P5, Program CM-150. Red bar: Buffer P5, Program CA-137. Black bars: Nucleofection controls without crRNA-Cas9. All data are representative of one screening experiment.

Figure 1.

Efficient gene editing in murine monocytes, macrophages, and DCs obtained from the BM. (A) Workflow for screening crRNA-Cas9-RNP–mediated KO of eGFP in mouse monocytes. Representative flow cytometry plots show gating strategy for identifying eGFP KO, F4/80⁺ macrophages following 5 d of culture in M-CSF (the workflow was previously published in Lim et al., 2019). (B) Heatmaps depicting relative impact of nucleofection conditions on cell viability and eGFP KO. Red box indicates the best condition (Buffer P3, Program CM-137). (C) eGFP KO efficiency following nucleofection with single or pooled crRNAs. NTC, non-targeting control crRNA-Cas9-RNP. (D) Workflow for CD45 crRNAXT-Cas9-RNP–mediated KO in mouse BMDC cultures. Top flow-cytometry plots show gating strategy for identifying macrophage, pDC, CD24⁺ DC, and Sirpα⁺ DC cells. Bottom flow-cytometry plots depict representative gating strategy using NTC crRNAXT-Cas9-RNP as a control for determining CD45-negative cells in each cell population. (E) CD45 KO efficiency as measured by flow cytometry (experimental workflow and gating strategy shown in D) in BM-derived CD24⁺, Sirpα⁺, pDC, and macrophage cells nucleofected with Cas9-RNPs (IDT V3) loaded with NTC or CD45 targeting crRNAXT using either P3/CM-137 or P3/EN138 combinations. Data in B–D are from single screening experiments. Data in E are presented as mean ± SEM (n = 4–5) and collected from two independent experiments.

Figure S2.

Screening of optimal Cas9-RNP nucleofection protocols for KO in murine BMDCs. (A) Data from the initial optimization screen are shown in four heatmaps reporting the cell abundances and CD45 KO efficiency in CD24⁺ (left panels) and Sirpα⁺ (right panels) DCs. Red boxes indicate the five conditions that showed the highest KO efficiency while maintaining acceptable cell abundance. Data are from one experiment. (B) Confirmation of deletion efficiency of the top five conditions from the initial optimization using CD45 crRNAXT-Cas9-RNPs compared with NTC crRNAXT-Cas9-RNPs and no-nucleofection (NN) controls. Data are averaged from technical replicates. (C) Relative CD80 levels as measured by flow cytometry in BM-derived CD24⁺ DC, Sirpα⁺ DC, pDC, and macrophage cells not nucleofected (No Nuc) or nucleofected with NTC crRNAXT-Cas9-RNPs using either P3/CM-137 or P3/EN138 combinations. Data are presented as mean ± SEM (n = 5) and collected from two independent experiments.

Figure 2.

Population-level gene editing in human monocyte-derived DCs and macrophages. (A) Workflow for B2M-specific KO in human monocyte-derived dendritic and macrophage cultures. Representative flow cytometry plots show gating strategy for using Cas9-RNPs loaded with NTC gRNA to determine B2M-negative cells in each cell population. (B) B2M KO efficiency in monocyte-derived DC (pink, left bar graph) and macrophage (right, blue bars) cells nucleofected with distinct B2M targeting sequences in either crRNA (B2M cr1, B2M cr2) or crRNAXT (B2M crXT1, B2M crXT2) format, or NTC complexed with IDT V3 Cas9 (dark bars) or Thermo Fisher TruCut V2 Cas9 (light bars). Data are from one experiment. (C) B2M KO efficiency in monocyte-derived macrophages nucleofected with IDT V3 Cas9-RNPs loaded with two different crRNAXTs (crXT1 and crXT2) targeting B2M or NTC (NTC crXT). Cas9-RNPs were added individually or in combination. Each Cas9-RNP is labeled with 1× or 2× to indicate the relative molar quantity nucleofected into the cells. Data are presented as mean ± SD (n = 3). (D) Same as B, but gRNA-Cas9-RNPs were loaded with sgRNAs (B2M sg1 and sg2) instead of crRNAs or crRNAXTs. (E and F) Dose–response curve of B2M KO efficiency in monocyte-derived macrophages. IDT V3 Cas9-RNPs loaded with two different sgRNAs targeting B2M (sg2 or sg4) were nucleofected into cells at the indicated quantities. Cas9-RNPs were complexed and delivered with and without 4 µM of IDT electroporation enhancer. Data in B–D are representative of two independent experiments (n = 1 donor per experiment). Data in E and F are from one experiment (n = 1 donor). (B–E) No nucleofection (No Nuc) cells or cells nucleofected with NTC crRNA, crRNAXT, or sgRNA-Cas9-RNPs were used as controls. Buffer P3, CM-137 condition was used for all Cas9-RNP delivery.

Figure S3.

Supporting data for population-level gene disruption in human monocyte-derived DCs and macrophages. (A) B2M KO efficiency in monocyte-derived macrophages nucleofected with four different crRNAXTs (crXT1, crXT2, crXT3, and crXT4) targeted against B2M and complexed with Cas9. Data are from one experiment. (B) B2M KO efficiency in monocyte-derived macrophages nucleofected with two unique crRNAXTs (crXT1 and crXT2) targeted against B2M and complexed with Cas9. Data from three different donors are displayed, each performed in triplicate (n = 3). (C) B2M KO efficiency in monocyte-derived macrophages nucleofected with indicated crRNAs targeted against B2M or NTC and complexed with Cas9. (D) B2M KO efficiency in monocyte-derived macrophages nucleofected with a single sgRNA (sg2) and complexed with Cas9. Data from three different donors are displayed, each performed in triplicate (n = 3), and are displayed as mean ± SD. (E) B2M KO efficiency in monocyte-derived macrophages nucleofected with IDT V3 containing Cas9-RNPs loaded with two different sgRNAs (sg1 and sg2) targeting B2M or a NTC sgRNA (NTC sg). Cas9-RNPs were added individually or in combination. Each Cas9-RNP is labeled with 1× or 2× to indicate the relative molar quantity nucleofected into the cells. Data in C and E are mean ± SD (n = 3) and representative of two independent experiments. (F) B2M KO efficiency in monocyte-derived macrophages for IDT V3 Cas9-RNPs loaded with NTC sgRNA or B2M sgRNA2 and complexed with an sgRNA:Cas9 molar ratio of 2:1 or 3:1. (G and H) Cell viability in monocyte-derived macrophages from samples in Fig. 2, E and F. Data in F–H are from single donors (n = 1) and one individual experiment.

Figure S4.

Analysis of activation markers, cytokine release, and phagocytosis in human monocyte-derived macrophages following RNP nucleofection. (A) Cell surface levels of indicated phenotypic markers measured by flow cytometry 5 d after nucleofection. (B) Efficiency of B2M-KO following nucleofection with two unique sgRNA:Cas9 RNPs. (C) Comparison of cell surface levels of CD80 and CD86 measured by flow cytometry 5 d after nucleofection. (D) ELISA measurements of TNF levels in cell culture medium of monocyte-derived macrophages following nucleofection. (E) Quantification of live monocyte-derived macrophages using imaging of live cell nuclei. Micrographs depict representative images of cultured cells. (F and G) Quantification of kinetics of particulate phagocytosis (F, myelin-pHrodo; G, beads-pHrodo). Micrographs depict representative images of phagocytosis following nucleofection, taken at the 5 h time point. Graphs in A–C depict MFI. Graphs in F and G depict intensity of pHrodo signal measured hourly following incubation with depicted particulate cargo. (H) Quantification of phagocytic index measured as AUC of data in G and H over 5 h of imaging. Data in A and B represent single measurements from two individual donors. Data in C–H are mean ± SD (n = 3) from three individual donors and a single experiment. Scale bars = 150 µm.

Figure 3.

Disruption of single or multiple genes in murine BMDCs and BMDMs to study TLR signaling. (A) Percentage of cells that were TLR7 negative from BM-derived CD24⁺, Sirpα⁺ DCs, pDCs, and macrophages nucleofected with Cas9-RNPs loaded with an NTC or two different Tlr7-specific sgRNAs (sg1 and sg2). Left: Buffer P3, Program CM-137; Right: Buffer P3, Program EN-138. TLR7-negative cells were assayed by intracellular flow cytometry. Data are presented as mean ± SEM (n = 3, biological triplicates) and representative of two independent experiments. (B) Cytokine levels measured by Luminex in supernatant from BMDC culture (combined cell types) in A after stimulation with mock or 800 ng/ml of the TLR7 agonist R848 for 17 h. Data are presented as mean ± SD (n = 3, technical triplicates) and are representative of two independent experiments. (C) Representative Western blots depicting MyD88 or TRIF knockdown by sgRNA-Cas9-RNP in murine BMDMs. (D) Assessment of gene editing efficiency by Sanger sequencing 5 d after electroporation. Data are presented as mean ± SEM (n = 3) and representative of one experiment. (E) ELISA measurement of IFNβ levels in cell culture medium of BMDMs 24 h after electroporation and 5 d after electroporation treated as described. (F) ELISA measurements of TNF in cell culture supernatant following stimulation with the indicated ligand for 18 h. Data in E and F are presented as mean ± SD (n = 3) and representative of three independent experiments.

Figure S5.

Analysis of Tlr7 editing efficiency and impact of nucleofection on BMDC phenotypes. (A) Representative histograms of TLR7 flow cytometry following nucleofection with indicated RNP complexes using Buffer P3, Program CM-137. Quantification of TLR7-KO shown in Fig. 3 A. (B) Frequencies of indicated myeloid cell subsets 12 d following nucleofection and Flt3 ligand-mediated BMDC differentiation. (C) Relative abundance of BMDCs cultured in B. (D) Assessment of cell surface levels of indicated phenotypic and activation markers on BMDCs cultured as in B. (E and F) Quantification of CD8⁺ T cell/OT-I (E) or CD4⁺ T cell/OT-II (F) proliferation following 3 d of coculture with BMDCs nucleofected and pulsed with indicated concentrations of OVA. Histograms on right depict proliferation measured by CFSE dilution of OT-I or OT-II cells. Data in B–D are mean ± SEM (n = 3, biological triplicates) and are representative of two independent experiments. Data in E and F are mean ± SEM (n = 3, biological triplicates). *, P < 0.05; **, P < 0.01 derived using pairwise statistical analyses using an unpaired Student’s two-sided t test.

Figure 4.

Disruption of single and multiple genes in human monocyte-derived macrophages. (A) Histograms depicting cell surface B2M, CD14, and CD81 protein levels in monocyte-derived macrophages as measured by flow cytometry. (B) Quantification of gene deletion as measured by flow cytometry. Data are presented as mean ± SD (n = 3) and representative of three independent donors. (C) Assessment of gene editing efficiency by Sanger sequencing 7 d after nucleofection. Data are mean ± SEM (n = 3) and representative of three independent donors.