Antibody targeting KIT as pretransplantation conditioning in immunocompetent mice

Chronic granulomatous disease (CGD) is an inherited hematological disease resulting from absence or dysfunction of the leukocyte superoxide-generating nicotinamide adenine dinucleotide phosphate (NADPH) oxidase that plays an important role in microbial killing.^1-3  Gene defects in any one of 5 NADPH oxidase subunits can cause CGD, with two-thirds of cases due to mutations in an X-linked gene encoding the gp91^phox (NOX2) subunit of the oxidase flavocytochrome b.^1,4  Current management includes antibiotic prophylaxis and aggressive treatment of acute infections, and HSC transplantation has been used for patients failing standard therapies.^1,2,5  CGD is also a candidate for gene therapy using autologous hematopoietic stem cells (HSCs).^1,5  Because CGD affects neutrophil function and no selective advantage exists for gene correction, cytoreduction of uncorrected endogenous marrow cells is required in order to achieve a sufficient level of long-term engraftment of gene-modified HSCs, ideally with regimens that minimize exposure to genotoxic agents.

KIT and its ligand stem cell factor (SCF) contribute to the survival and proliferation of hematopoietic stem and progenitor cells (HSPCs).⁶  The monoclonal antibody ACK2 blocks binding and activation of murine KIT by SCF.⁷  Administration of ACK2 transiently reduces murine HSPCs, permitting substantial repopulation with donor HSCs in immunodeficient Rag2^−/−γc^−/− mice, but not immunocompetent mice.⁸  Here, we used congenic models to examine the effectiveness of combining ACK2 with low-dose irradiation (LD-IR) as conditioning for transplantation of immunocompetent wild-type mice and for HSC gene therapy of X-linked CGD (X-CGD) mice.

Mice used in this study include wild-type C57Bl/6J (C57; CD45.2⁺), B6.SJL-PtrcaPep3b/BoyJ (BoyJ; CD45.1⁺), and F₁ progeny of C57 and BoyJ mice. X-CGD mice⁹  in C57, BoyJ, and CD45.1 × CD45.2 F₁ backgrounds were also used. Mice were treated with 2 mg of the KIT-blocking monoclonal antibody ACK2⁷  (injected intraperitoneally), 300 cGy, or both, followed by analysis of marrow HSPCs or transplantation of congenic marrow, either freshly isolated or transduced with the lentiviral vector CL20-gp91-OPT¹⁰  for expression of gp91phox, performed as described in supplemental Methods (available on the Blood Web site; see the Supplemental Materials link at the top of the online article). All animal protocols were approved by the Institutional Animal Care and Use Committee of the Indiana University School of Medicine.

To determine whether ACK2 + LD-IR at 300 cGy could deplete HSCs in immunocompetent mice, marrow was harvested from wild-type mice treated with ACK2, LD-IR, or both (Figure 1A), and the frequency of HSCs (lin⁻ Sca-1⁺ KIT⁺ CD135⁻CD150⁺) and HPCs (lin⁻ Sca-1⁻KIT⁺ CD34^lowFcγR^low or colony forming unit-granulocyte, monocyte) was determined. Mice in all 3 treatment arms showed HSC and HPC depletion on day 7; whereas recovery occurred in mice treated with either agent alone, these remained profoundly depressed in mice treated with ACK2 + LD-IR through day 14 (Figure 1B). Bone marrow (BM) cellularity followed a similar pattern (Figure 1B). Peripheral blood (PB) counts obtained on day 10 also showed more substantial effects of the combination treatment compared with ACK2 or LD-IR alone (supplemental Table 1).

HSC function of these 3 cohorts of mice was assessed with a competitive long-term repopulating assay.^11-13  7 days after ACK2 administration, marrow long-term repopulating activity (LTRA) was normal (Figure 1C), indicating that HSC functional capacity was intact, despite the marked reduction in marrow lin⁻ Sca-1⁺ KIT⁺ CD135⁻CD150⁺ cells (Figure 1B). This is consistent with the full recovery in the number of marrow lin⁻ Sca-1⁺ KIT⁺ CD135⁻CD150⁺ cells 10 days after ACK2 (Figure 1B). As reported,^12,13  LTRA was reduced to approximately 30% of normal 3 days following LD-IR alone. However, the combination of ACK2 + LD-IR almost entirely eliminated marrow LTRA (2.5% ± 0.8% of normal; Figure 1C).

We examined the clearance of ACK2 antibody in serum using a serial dilution assay (supplemental Methods). ACK2 concentration was estimated at 2-5 μg/mL serum at day 7 and was undetectable by day 10 (supplemental Figure 1). These results suggest that serum ACK2 levels were reduced by at least 200-fold 7 days after administration compared with the predicted level of 1 mg/mL on the day of ACK2 administration.

To determine whether marrow depletion of HSCs using ACK2 + LD-IR could allow engraftment of donor HSC in immunocompetent mice, 1 × 10⁶ marrow cells from CD45.2+ C57 mice were transplanted into CD45.1+CD45.2+ (F1) recipients on day 7 after ACK2 ± LD-IR (see Figure 1A). In mice preconditioned with ACK2 + LD-IR, donor chimerism in PB leukocytes reached 79.3% ± 8.3%, while LD-IR alone permitted only 15.3% ± 4.6% engraftment 24 weeks posttransplantation (Figure 2A). In a second experiment, we transplanted 20 × 10⁶ marrow cells into mice treated with ACK2 alone to exclude the possibility that a low number of donor HSCs may be insufficient to engraft in the host niche in this setting, while 1 × 10⁶ cells were transplanted into mice conditioned with LD-IR or ACK2 + LD-IR. Donor chimerism in ACK2 + LD-IR–treated mice (76.5% ± 5.5%) was significantly higher than in the single-treatment groups: LD-IR group, 7.3% ± 2.6%; ACK2 group, 5.0% ± 2.5% (Figure 2B). The enhancement of PB donor chimerism in mice conditioned with ACK2 + LD-IR was multilineage (supplemental Figure 2A). Long-term donor chimerism in marrow lin⁻ Sca-1⁺ KIT⁺ CD135⁻CD150⁺ HSC from primary recipients was similar to that in PB (supplemental Figure 2B), and donor chimerism following secondary transplantation into lethally irradiated recipients was similar to that seen in primary recipients (Figure 2C), further establishing that HSC depletion with ACK2 + LD-IR permits engraftment of multilineage, long-term repopulating HSCs.

We then sought to implement ACK2 + LD-IR as conditioning for transplantation of X-CGD mice with HSC transduced with a lentiviral vector for gp91^phox expression. In mice transplanted with of 5 × 10⁵ CL20-gp91-OPT¹⁰ –transduced lin⁻ cells, donor chimerism at 20 weeks posttransplantation was only approximately 1.0% in recipients conditioned with LD-IR but ranged from 8.0% to 76.0% (45.5% ± 32.1% mean ± standard deviation) in mice conditioned with ACK2 + LD-IR (Figure 2D). The percentage of DHR-positive neutrophils in this latter cohort ranged from 9.1% to 42.0% at 20 weeks (Figure 2D), indicating engraftment of CL20-gp91-OPT–transduced HSC and expression of gp91^phox in neutrophils to reconstitute NADPH oxidase activity. This fraction of functional phagocytes is expected to provide significant activity against pathogenic microbes.^1,5,14 

In summary, we show for the first time that combination treatment with the KIT-blocking antibody ACK2 and LD-IR nearly ablates marrow HSCs in immunocompetent mice, as measured by a HSC competitive repopulation assay. This combination enables substantial and durable engraftment after transplantation of modest numbers of congenic donor BM. ACK2 given as a single agent transiently reduced HSPCs in immunocompetent mice, as in a previous report⁸ ; however, HSC recovery was rapid and LTRA was unaffected on day 7 after administration. These findings likely explain why there is little engraftment of congenic donor marrow in immunocompetent mice conditioned with ACK2 alone, even with transplantation of high cell doses. Our data further demonstrate that conditioning with ACK2 and LD-IR in X-CGD mice enables efficient engraftment of HSC cultured ex vivo for lentivirus-mediated gene transfer in a murine model of autologous transplantation for gene therapy. We conclude that ACK2 combined with LD-IR is a promising approach for reducing exposure to genotoxic agents while still achieving substantial cytoreduction of endogenous marrow to facilitate engraftment of transplanted HSCs. This conditioning strategy may be useful for gene therapy of CGD and other inherited blood cell disorders for which no selective advantage exists for gene correction, and may also be applicable for depletion of recipient HSCs in allogeneic transplantation.

We thank Catherine Matthews for help with preparing the manuscript and Karl Staser and Wade Clapp for bone marrow–derived mast cells as controls for ACK2 staining. We also thank Harry Malech, Uimook Choi, Suk See De Ravin, and Elizabeth Kang (Laboratory for Host Defenses, Genetic Immunotherapy Section, National Institute of Allergy and Infectious Diseases) for the CL20-gp91-OPT lentiviral vector.

This work was supported by National Institutes of Health grants P01 HL53586 (M.C.D., M.C.Y., and E.F.S.) and the Riley Children's Foundation (M.C.D. and M.C.Y.). The Indiana University Simon Cancer Center (P30 CA082709) provided partial support for flow cytometry and mouse cores.

National Institutes of Health

Contribution: X.X., N.K.P., W.C.S., M.C.Y., and M.C.D. designed, performed, and analyzed experiments; X.X., M.C.Y., and M.C.D. drafted the manuscript and figures; E.F.S. helped with experimental design, analysis, and editing of the manuscript; and M.C.D. oversaw the project including experimental design, analysis, interpretation of the data, and preparation of the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

The current affiliation for M.C.D. is Department of Pediatrics, Washington University School of Medicine, St Louis, MO.

Correspondence: Mary Dinauer, Washington University School of Medicine, 660 South Euclid Ave, Box 8208, St Louis, MO 63100; e-mail: dinauer_m@kids.wustl.edu.