Correction of the murine Wiskott-Aldrich syndrome phenotype by hematopoietic stem cell transplantation

Wiskott-Aldrich syndrome (WAS) is an X-linked immunodeficiency of variable penetrance, presenting classically with the triad of immunodeficiency, thrombocytopenia, and eczema.1,2 Severely affected males who receive symptomatic care alone rarely survive beyond the second decade of life. In contrast, hematopoietic stem cell transplantation (HSCT) from matched sibling donors is curative, showing excellent long-term survival.3-5 However, many patients lack such a donor, and the use of alternate donors has shown less promising results. Hence, the identification of an animal WAS model that recapitulates the human phenotype would provide an opportunity to test novel therapeutic strategies, such as transplantation with nonmyeloablative conditioning or gene therapy. However, exacerbation of a pre-existingHelicobacter-associated colitis in WAS-deficient 129/SvEv mice after transplant conditioning results in mortality in the majority of animals studied.6 Our studies were designed to determine whether transfer of the WAS genotype to the C57/BL6J strain would provide mice to test novel therapies.

129/SvEv-Wasp^tm1Sbs, C57/BL6J, and C57/BL6J Ly5.1 (B6.SJL-Ptprca Pep3b/BoyJ) mice were obtained from the Jackson Laboratory (Bar Harbor, ME). 129/SvEv-Wasp^tm1Sbs mice were bred with the C57/BL6J mice for 2 to 4 generations (see figure legends). Genetic screening of progeny was performed by polymerase chain reaction amplification of the neo insert and/or by Southern blotting.6 Donor marrow cells were obtained from the hind limbs of C57/BL6J Ly5.1 mice (Figure 2B-D) or from littermates of the recipients. Recipients were lethally irradiated (1100 cGy) and injected with 1.0 to 7.5 × 10⁶cells into the tail vein. Flow cytometry of peripheral leukocytes, gating lymphocytes on forward versus side scatter and subsequently on fluorescently tagged cell-bound anti-Ly5.1 (CD45.1), CD90.2 (Thy-1.2), or CD45R/B220 (BD Pharmingen, San Diego, CA) was used to monitor engraftment.

Whole spleens were crushed and the cells were filtered through 70-μm mesh. Following red cell lysis and removal of B lymphocytes by adsorption to antibody-coated flasks (goat anti-mouse IgG + IgM; Jackson Immunoresearch, West Grove, PA), T cells were treated with anti-CD3ε antibody (BD Pharmingen) that had been bound overnight (4°C) to 96-well dishes, or with human interleukin-2 (IL-2; Chiron, Emeryville, CA). A total of 1.0 μCi/well (0.037 MBq/well)³H-thymidine was added at 48 hours, and incorporation was measured at 72 hours.

Hematologic values were measured on a HemaVet modified Coulter counter (CDC Technologies, Seymour, CT). Statistical analysis was performed using the exact Wilcoxon rank test. Significance values noted in Figure 2 were assessed atP < .05/3 using the exact Wilcoxon rank test with Bonferroni correction.

The utility of the C57/BL6J mouse strain for study of HSCT is well established.7-9 We crossed 129/SvEv WAS-null mice (129−WAS−) for 2 to 4 generations onto the C57/BL6J background. Necropsy studies of C57/BL6J WAS− (C57−WAS−) males revealed normal colons without the inflammatory infiltrates and crypt abscesses previously reported on the 129/SvEv background.6A slightly increased number of mononuclear cells in the lamina propria was found in WAS-null animals compared with wild-type (WT) littermates, an abnormality slightly exacerbated by irradiation (Figure1). Necropsy studies of 129−WAS− males revealed only acute and chronic inflammatory cells in the colonic lamina propria in 1 of 3 animals.

More than 60 mice have received lethal irradiation and bone marrow rescue, with a survival rate (longer than 6 weeks) of greater than 80% (Figure 2A). Donor marrow was derived from WT littermates or from C57/BL6J Ly5.1 animals (see below). No consistent differences in morbidity or mortality were observed between these donor sources, nor was there a correlation between posttransplantation mortality and the number of back-crosses (Figure2A).

Of 29 transplant recipients for which fecal specimens were tested prior to lethal irradiation, 24 were positive for Helicobacter hepaticus and 5 for other Helicobacter species. All but 2 mice in the first group, and all of the second group, have survived for more than 6 weeks after transplantation. Necropsy of the nonsurviving mouse positive for H hepaticusrevealed no significant colitis (Figure 1).

Hematologic and immunologic parameters were measured in several cohorts of animals and compared with WT and C57−WAS− littermate controls. We found that C57−WAS− males (age range, 8-34 weeks) showed a significant thrombocytopenia and lymphocytopenia in comparison with age-matched WT littermates (Figure 2B). Age did not significantly discriminate between the genotypes in this study. As is the case for human WAS, T cells from C57−WAS− proliferated poorly in response to T-cell receptor stimulation, but responded to IL-2 stimulation (Figure2C). HSCT corrected the thrombocytopenia and lymphocytopenia of C57−WAS− mice by 12 to 14 weeks after transplantation (Figure 2B). The inability of WAS-derived splenic T cells to proliferate normally following stimulation with immobilized anti-CD3 antibody was also corrected (Figure 2C), demonstrating that T cells derived from donor marrow had repopulated the spleen normally.

To follow the rate and extent of donor engraftment accurately, we subjected a cohort of 6 C57−WAS− males (which express the Ly5.2 epitope of CD45 on their leukocytes) to transplantation with marrow derived from Ly 5.1 (B6.SJL-Ptprca Pep3b/BoyJ) mice. At 6 weeks after transplantation, all lineages except peripheral blood T cells were derived entirely from the donor (Figure 2D and data not shown). Consistent with past reports of radiation-resistant T-cell subsets,10 we found that a fraction (12%) of peripheral T cells were host derived at 7 weeks after transplantation, but no host T cells were found at 24 weeks (data not shown). Spleen-derived T cells from the transplant recipients were also 100% donor derived at late times after transplantation (data not shown).

The lack of colitis in our WAS knockout mice at less than 4 months contrasts with prior results in older mice, and we found that the presence of Helicobacter species in WAS− mice is not sufficient to induce colitis. Given the variable penetrance of WAS mutations clinically, the phenotypic difference we have seen between these strains of WAS-null mice could reflect a similarly variable penetrance in this species.

Our results demonstrate that C57−WAS− mice can survive HSCT, which corrects several key aspects of the murine WAS− phenotype. These observations strengthen the similarity between these mice and their human counterparts and will allow us to study whether murine WAS can be phenotypically corrected by alternative transplantation methods such as gene therapy.