High overexpression of the human alpha-galactosidase A gene driven by its promoter in transgenic mice: implications for the treatment of Fabry disease.

BACKGROUND: Human alpha-galactosidase A (alpha-Gal A) is the lysosomal enzyme that cleaves alpha-galactosyl residues from glycoconjugates and is the deficient enzyme in Fabry disease. To date, there have been no studies on the regulation of this "housekeeping" gene. METHODS: Transgenic mice were established with either 1) a 13.3-kilobase (kb) human genomic fragment that contained 246 bp of 5'- and approximately 2.8 kb of 3'- untranslated sequences, or 2) an "intronless" construct derived from the genomic sequence with the 5' and 3' flanking regions intact. Tissues that expressed high levels of alpha-Gal A activity were examined by light and electron microscopy. RESULTS: Transgenic mice were generated with 2 and 12 copies of the genomic sequence (Lines 1 and 2) or about 60 copies of the intronless construct (Lines 3 and 4). In mice hemizygous for the genomic transgene (Lines 1 and 2), tissue alpha-Gal A activities were 12 to 155 times higher than those in the respective wild-type tissue, depending on tissue and transgene copy number. Of note, the high overexpression did not alter the cellular or subcellular cytoarchitecture. In contrast, alpha-Gal A activities expressed by mice that carried the intronless construct were only two- to sixfold more than in wild-type tissues in which the genomic transgene was highly expressed. CONCLUSIONS: The remarkably high levels of alpha-Gal A expression in transgenic mice carrying the intact genomic sequence versus the intronless construct suggested that the genomic sequence contained a strong intronic enhancer element. Identification of this regulatory element or elements may be useful in efforts to overexpress human alpha-Gal A for gene therapy endeavors. In addition, overexpression of human alpha-Gal A did not affect cellular morphology, which indicates that its overexpression in gene therapy endeavors should be safe.