In vivo LV expression of human GAA. (A) Gaa^−/− mice received a transplant with 5 × 10⁵ Lin⁻ cells transduced with the therapeutic vector (LV-SF-GAA) at a multiplicity of infection of 9 to 10. The GAA activity in peripheral leukocytes of a total of 32 LV-SF-GAA–treated mice was measured, starting from day 29 after transplantation to the last data point at day 549 (n = 4). Fifteen Gaa^−/− mice that received a transplant of LV-SF-GFP served as negative controls. The leukocyte α-glucosidase levels (3.4 ± 1.3 nmol/h/mg) of congenic FVB mice (n = 6) are indicated by the dotted line. (B) Illustration of GAA expression and lysosomal localization in BM cells of a WT, a Gaa^−/− mouse, and a Gaa^−/− mouse that received a transplant of LV-SF-GAA–transduced HSCs harvested 8 months after transplantation [blue nuclear DAPI (4′-6′-diamidine-2-phenylindole) stain and green fluorescent Alexa Fluor 488 immunostaining for GAA protein; original magnification ×1000]. (C) At 8 months after transplantation, Western blot analysis of GAA expression in BM and spleen (SPL) shows a known pattern of 95-kDa (intermediate species) and 76-kDa (mature species) forms of human GAA, reflecting GAA synthesis, posttranslational modification, maturation, and lysosomal localization.²²  Mice that received a transplant of LV-SF-GFP do not express human GAA, whereas the level of GAA expression in WT animals is too low to detect. A dilution of recombinant human GAA (rhGAA) of 110 kDa serves as a marker. (D) Plasma GAA enzyme levels. Gaa^−/− mice that received a transplant of LV-SF-GAA showed variable but detectable GAA enzyme levels in plasma compared with background activity (BL) detected in Gaa^−/− mice. Low activity levels were also detected in WT mice. Numbers in the figure that accompany the plasma data represent GAA activity (nmol/h/mg) in leukocytes. (E) Reconstitution of GAA activity in Gaa^−/− mice. GAA activity in tissues of mice that received a transplant of LV-SF-GFP and LV-SF-GAA and age-matched WT mice at 10 months of age (n = 3-6 mice for all tissues, except for LV-SF-GFP brain). *Significant difference (P < .05) of mice that received a transplant of LV-SF-GAA compared with both WT and LV-SF-GFP mice. †Significant difference (P < .05) between mice that received a transplant of LV-SF-GAA and LV-SF-GFP but not between LV-SF-GAA and WT mice. (F) Tolerance to rhGAA in Gaa^−/− mice that received a transplant of LV-SF-GAA. Both WT (n = 2) and LV-SF-GFP–treated Gaa^−/− (n = 2) mice responded by the generation of anti–human GAA antibodies after a first injection of rhGAA (Myozyme, 20 mg/kg) with Freund adjuvant (rhGAA + A) followed by a second injection of rhGAA without adjuvant 2 weeks later. LV-SF-GAA–treated Gaa^−/− mice (n = 13) did not respond to the challenge with rhGAA. (G) The percentage of reconstituted male cells in BM of LV-SF-GFP and LV-SF-GAA female recipients (chimerism) was determined by Y chromosome Sry q-PCR and corrected for overall genomic DNA by Gapdh q-PCR (n = 6). (H) Copy number of LV vector integrations. Q-PCR was used to determine the average copy number in male BM cells (n = 6). Horizontal lines in panels G and H represent mean value.