Screening of Chimeric GAA Variants in a Preclinical Study of Pompe Disease Results in Candidate Vector for Hematopoietic Stem Cell Gene Therapy

Abstract

Pompe disease is a rare genetic neuromuscular disorder caused by acid alpha-glucosidase (GAA) deficiency resulting in lysosomal glycogen accumulation and progressive myopathy. Enzyme replacement therapy (ERT) is the current standard of care, which prolongs the quality of life for Pompe patients. However, ERT has limitations due to lack of enzyme penetration into the central nervous system (CNS) and skeletal muscles, immunogenicity against the recombinant enzyme, and requires life-long biweekly infusions. In a preclinical mouse model, a clinically relevant promoter to drive lentiviral vector-mediated expression of engineered GAA in autologous hematopoietic stem and progenitor cells (HSPC) was tested with nine unique human chimeric GAA coding sequences incorporating distinct peptide tags and codon-optimization iterations. Vectors including glycosylation independent lysosomal targeting (GILT) tags resulted in effective GAA enzyme delivery into key disease tissues with enhanced reduction of glycogen, myofiber and CNS vacuolation, compared to non-tagged GAA in Gaa knockout mice, a model of Pompe disease. Genetically modified microglial cells in brains were detected at low levels, but provided robust correction. Furthermore, an aminoacid substitution in the tag added to reduced capacity to induce insulin signaling and there was no evidence of off-target effects. This study demonstrated the therapeutic potential of lentiviral HSPC gene therapy exploiting optimized GAA tagged coding sequences to reverse Pompe disease pathology in a preclinical mouse model providing a promising vector candidate for further investigation.

Isogenic GAA-KO Murine Muscle Cell Lines Mimicking Severe Pompe Mutations as Preclinical Models for the Screening of Potential Gene Therapy Strategies

“…Our data of the murine IGF2-GAA (2G) chimera contrast with the results obtained by several groups using human IGF2-GAA proteins, showing clear improvements in their therapeutic activity in comparison with WT GAA [ 48 ]. The differences in IGF2-GAA processing in human versus murine myeloid cells could be a consequence of several factors.…”

“…Mice were sacrificed one month after the injection, and the vector genome copy number ( Figure 7 C), as well as the GAA activity in the liver ( Figure 7 D) and serum ( Figure 7 E, left graph), were analysed to determine in vivo GAA expression and secretion activity for each of the constructs. As already reported [ 48 ], the unmodified-codon optimised GAA appears as the best performer in terms of expression in the liver ( Figure 7 D, right graph). However, modified GAAs were secreted with similar efficacy, as evidenced by normalising the GAA activity in serum versus GAA activity in the liver ( Figure 7 E, right).…”

Gene Therapy Developments for Pompe Disease

“…To that effect, a lentiviral vector using the MND promoter driving expression of a IGF2-tagged GAA was used in LV HSPC gene therapy. In this study, nine GAA chimeric variants driven by the MND promoter were investigated [ 133 ], including tags described to enhance delivery to the CNS, such as an apolipoprotein E (ApoE) tag [ 127 ] and a modified IGF2 tag containing a R37A mutation for GILT delivery [ 134 ]. The MND-GILT-R37A-GAAco vector was more effective in reducing glycogen in skeletal muscles and CNS than the non-tagged GAAco vector 16 weeks after transplantation of genetically modified HSPCs.…”

“…Since microglia account for ~5–12% of cells in the rodent brain [ 136 ], the therapeutic enzyme levels depend on the percentage of microglial engraftment in the brain, the transgene promoter activity in microglial cells, and the efficiency of cross-correction. The GAA enzyme activity levels in the CNS of treated Pompe mice were lower than the activity level observed in wild-type animals [ 133 ]. However, the low enzyme activity levels were similar to those seen in other studies using HSPC gene therapy, such as Arsa −/− mice reaching close to 10% of wild-type arylsulfatase A enzyme activities [ 121 ], and Ids −/− treated mice which did not exceed 4% of wild-type levels in the brain.…”