In utero gene transfer to the glucocerebrosidase knockout mouse line, partial characterization of a potentially novel beta glucosidase, and construction and functionality of a mutant glucocerebrosidase transgene in ES cells and transgenic mice

We sought to examine the potential of in utero bone marrow transplantation to deliver GC activity to the most severely affected GC-knockout mouse model. We chose to use a retroviral vector (MFG GC) to deliver the GC cDNA to mouse, fetal liver-derived, hematopoeitic cells (FLD-HSC). We then transplanted 2--5 x 105 transduced cells to fetuses of the GC knockout mouse line by transplacental injection. Aliquots of these transduced cells were also grown in vitro to produce colony-forming units---granulocyte/macrophage (CFU-GM). These CFU-GM colonies were examined by PCR and GC enzyme activity assay to assess the transduction efficiency and increase in GC activity after transduction. Transplanted mice were allowed to come to term, and were sacrificed at one day, one week, and 2 months after birth. Of 80 animals collected, only four were homozygous for the GC gene disruption, indicating that homozygous knockouts may have been adversely affected by the transplant procedure itself. Spleen, liver, brain, and skin from the transplanted mice were analyzed by PCR for the presence of transduced, engrafted cells. A significant percentage (38%) showed engraftment of transduced cells in at least one of the tissues examined. GC enzyme activity assays were then performed to establish "normal" GC activities for these tissues in non-transplanted mice. These comparisons utilized a two-tailed Student's T-test (assume equal variances, P ≤ 0.05). Using this statistical analysis, 7 engrafted tissue samples were shown to exhibit significant increases in GC activity over control values. Of these, only one exhibited a GC activity that was well above the range of GC activity in the control pool. Western blot analysis failed to detect human GC in any of the tissues analyzed. A potentially novel beta-glucosidase activity was observed in the tissues of affected GC knockout homozygotes. This activity differed from GC primarily in that it was more tolerant of neutral pH, was resistant to inhibition by conduritol beta-epoxide (CBE) up to 250 muM, and was not activated or inhibited by sodium taurocholate. Finally, data is presented concerning the construction of a mutant (N370S) GC minigene and its use in transgenic ES cells and mice. It is concluded that in utero transplantation of genetically modified FLD-HSC is not an efficient means of delivering GC activity to the tissues of mice. Future directions for the work are discussed. (Abstract shortened by UMI.).