CRISPR/Cas9-mediated Cxcr4 disease allele inactivation for gene therapy in a mouse model of WHIM syndrome

Abstract

WHIM syndrome is an autosomal dominant immunodeficiency disorder caused by gain-of-function mutations in chemokine receptor CXCR4 that promote severe panleukopenia because of retention of mature leukocytes in the bone marrow (BM). We previously reported that Cxcr4-haploinsufficient (Cxcr4+/o) hematopoietic stem cells (HSCs) have a strong selective advantage for durable hematopoietic reconstitution over wild-type (Cxcr4+/+) and WHIM (Cxcr4+/w) HSCs and that a patient with WHIM was spontaneously cured by chromothriptic deletion of the disease allele in an HSC, suggesting that WHIM allele inactivation through gene editing may be a safe genetic cure strategy for the disease. We have developed a 2-step preclinical protocol of autologous hematopoietic stem and progenitor cell (HSPC) transplantation to achieve this goal. First, 1 copy of Cxcr4 in HSPCs was inactivated in vitro by CRISPR/Cas9 editing with a single guide RNA (sgRNA) that does not discriminate between Cxcr4+/w and Cxcr4+/+ alleles. Then, through in vivo natural selection, WHIM allele-inactivated cells were enriched over wild-type allele-inactivated cells. The WHIM allele-inactivated HSCs retained long-term pluripotency and selective hematopoietic reconstitution advantages. To our knowledge, this is the first example of gene therapy for an autosomal dominant gain-of-function disease using a disease allele inactivation strategy in place of the less efficient disease allele repair approach.

Graphical abstract

Figure 1.

Feasibility of an allele-nonselective Cxcr4 inactivation cure strategy for WHIM syndrome. (A) Possible gene-editing cure strategies for the gain-of-function mutations causing WHIM syndrome. ∗ denotes WHIM mutation location in Cxcr4. (B) The 4 possible genotypes produced by allele-nonselective Cxcr4 inactivation. (C) Absolute leukocyte subtype counts in mouse blood stratified by Cxcr4 genotype. N > 19 mice in each group. (D-E) Hematopoietic reconstituting activity of Cxcr4^o/w vs Cxcr4^+/w BM cells during competitive transplantation of lethally irradiated Cxcr4^+/w recipient mice (D) and after single genotype BM transplantation of unconditioned Cxcr4^+/w recipient mice (E). The experimental schemes are shown at the top of panels D and E. Donor and recipient cells were marked by different CD45 isoforms for detection and quantitation via flow cytometry. n = 5 mice in each group in panels D and E. ∗P < .05; ∗∗P < .01; ∗∗∗P < .001. Data in panels D and E are from a single experiment and are representative of 2 experiments for panel D.

Figure 2.

CRISPR/Cas9-mediated Cxcr4 inactivation. (A) Sequences and gene locations of 2 sgRNAs that nonspecifically target both the wild-type and WHIM Cxcr4 alleles. sgRNA sequences are in blue and the PAM sequence is in red. (B) Cell-free system. Cxcr4 amplicons generated via PCR were incubated with Cxcr4-sgRNA/Cas9 RNP complex in vitro and the cleavage products were detected by agarose gel electrophoresis. Experimental conditions are indicated at the top of each lane: M, 1 kb DNA ladder markers; no sgRNA, control Cxcr4-amplicon incubated with Cas9 but without sgRNA; Right guide RNA and left guide RNA refer to Cxcr4 amplicons incubated with RNPs containing the indicated sgRNA, each performed in triplicate. ∗ denotes expected fragment sizes based on the sgRNA location within the amplicon. (C-F) Cell-based transfection system using Cxcr4-sgRNA/Cas9 RNP complex. (C-D) Editing of the pre–B-cell lymphoma cell line L1.2. (C) FACS analysis of nonpermeabilized L1.2 cells with Cxcr4 antibody. The sgRNA is specified at the upper left of each panel. Red and orange histograms, Cxcr4-sgRNA/Cas9 RNP-transfected cells tested with isotype control antibody (red) or anti-Cxcr4 antibody (orange). Blue histograms, mock-transfected cells tested with anti-Cxcr4 antibody. (D) T7 endonuclease 1 (T7E1) editing assay of L1.2 cells. Cxcr4 amplicons from mock-transfected cells or cells transfected with either the left or right sgRNA/Cas9 RNP-transfected, as specified at the top of each lane, were treated with T7E1 and the products were revealed via agarose gel electrophoresis. M, 100 bp DNA ladder. Red stars indicate the DNA fragments cleaved by T7E1. (E-G) Editing of primary cKit⁺ HSPCs from BM of C57BL/6 mice. (E) T7E1 assay. (F) Identification of cleavage site indels. Sequences of 3 Cxcr4 amplicons from edited HSPCs are shown. The PAM sequence is located within the red boxes. Red arrows, deletions of 1 (left) or 2 nucleotides (middle), and insertion of 1 nucleotide (right). (G) Location of premature stop codons in Cxcr4 introduced by CRISPR/Cas9-generated indels. In the DNA sequence, the PAM site is highlighted in red and the sgRNA sequence is in blue. Protein sequence numbers at the top demarcate amino acid positions; red stars show positions of chain termination introduced by the 1 deletion (first star) and 2 deletions or 1 addition (second star).

Figure 3.

Cxcr4-sgRNA/Cas9 RNP-transfected WHIM HSPCs have a selective advantage for durable hematopoietic reconstitution over mock-transfected cells in lethally irradiated Cxcr4^+/w recipient mice. (A) Competitive gene therapy experimental design. The editing status and CD45 isoforms used for distinguishing donor and recipient cells are given at the tops and bottoms of the mouse icons, respectively. Left-Cxcr4-sgRNA/Cas9 RNP-transfected (edited) and mock-transfected (mock) HSPCs were mixed 1:1 and transplanted into lethally irradiated Cxcr4^+/w recipients. (B) Cxcr4-sgRNA/Cas9 RNP–transfected (edited) and mock-transfected (mock) donor HSPCs were transplanted in approximately equal numbers. Numbers indicate the percentage of total cells in each gate before transplantation. (C) Hematopoietic reconstitution of the blood. The leukocyte subsets are indicated above each panel. Each data point represents an individual mouse and the solid lines trace the means across time for the experimental conditions coded at the right of the panel. (D-E) Hematopoietic reconstitution of the BM. Mature BM cells (D) and HSPCs (E) were analyzed on day 381 after transplantation when the experiment was arbitrarily terminated. In panel D, data for the same leukocyte subsets in the blood are regraphed from the same day from panel C to facilitate comparison. Data are presented as a scatter plot (n = 3) and as the mean ± SEM of the percent of total donor-derived cells for each subset. Data are from a single experiment, representative of 3 independent experiments. LT-HSC, long-term hematopoietic stem cell (CD34^-Flt3⁻LSK); MPP, multipotential progenitor (CD34⁺Flt3⁺LSK); ST-HSC, short-term hematopoietic stem cell (CD34⁺Flt3⁻LSK).

Figure 4.

WHIM allele–inactivated HSPCs have a selective advantage for engraftment to reconstitute hematopoietic cells. (A) Flow cytometry plot of sorted CD11b⁺ cells in blood from a recipient mouse 112 days after transplantation. DNA was isolated for analysis of allele-specific edits from the sorted CD11b⁺ cells, which were derived from RNP-transfected and mock-transfected HSPCs post transplantation, as described in Figure 3. (B) PCR cloning design for sequencing Cxcr4 that covers both the CRISPR/Cas9 PAM site and the WHIM mutation site. (C) Edit analysis for the sorted CD11b⁺ cells shown in panel A. Edits were scored on the wild-type and WHIM alleles for CD11b⁺ cells derived from either the mock-transfected or Cxcr4-sgRNA/Cas9 RNP–transfected donor HSPCs. Eighteen and 25 clones were sequenced for the CD11b⁺ cells derived from mock-transfected and Cxcr4-sgRNA/Cas9 RNP–transfected HSPCs, respectively. (D) Flow cytometry plot of LSK cells sorted from BM of the same recipient mouse shown in panel A 381 days after transplantation. (E) Edit analysis for the sorted LSK cell populations shown in panel D. A total of 6 and 10 clones were sequenced for the cells derived from the mock-transfected and Cxcr4-sgRNA/Cas9 RNP–transfected HSPCs, respectively.