Gal alpha(1, 3) Gal (gal epitope) is a carbohydrate epitope and synthesized in large amount by alpha(1, 3) galactosyltransferase [alpha(1, 3) GT] enzyme on the cells of lower mammalian animals such as pigs and mice. Human has no gal epitope due to the inactivation of alpha(1, 3) GT gene but produces a large amount of antibodies (anti-Gal) which recognize Gal alpha(1, 3) Gal structures specifically. In this study, a replication-deficient recombinant adenoviral vector Ad5sGT containing pig alpha(1, 3) GT cDNA was constructed and characterized. Adenoviral vector-mediated transfer of pig alpha(1, 3) GT gene into human tumor cells such as malignant melanoma A375, stomach cancer SGC-7901, and lung cancer SPC-A-1 was reported for the first time. Results showed that Gal epitope did not increase the sensitivity of human tumor cells to human complement-mediated lysis, although human complement activation and the binding of human IgG and IgM natural antibodies to human tumor cells were enhanced significantly after Ad5sGT transduction. Appearance of gal epitope on the human tumor cells changed the expression of cell surface carbohydrates reacting with Ulex europaeus I (UEA I) lectins, Vicia villosa agglutinin (VVA), Arachis hypogaea agglutinin (PNA), and Glycine max agglutinin (SBA) to different degrees. In addition, no effect of gal epitope on the growth in vitro of human tumor cells was observed in MTT assay.
To test the potential of human secretor type alpha(1, 2) fucosyltransferase [Se alpha(1,2)FT] to downregulate the expression of Gal alpha(1,3)Gal epitope (gal epitope) in cultured cell lines. Expression of Se alpha(1,2) FT was mediated by human adenoviral vector. Flow cytometric analysis was used to compare the expression level of H blood group antigen or gal epitope. MTT was employed to assess the susceptibility of mouse NIH3T3 cells to human natural antibody and complement mediated lysis. A recombinant replication-deficient adenovirus (rAdv) containing human Se alpha(1,2)FT cDNA (Ad5hSeFT) was designed and successfully constructed. Flow cytometric analysis showed that after mock infection, Ad5null infection, and Ad5hSeFT infection, the mean fluorescence intensity (MFI) values for the binding of Ulex europaeus I (UEA-I) lectin to NIH3T3 cells were 2.3 +/- 0.6, 2.1 +/- 1.0, and 36.5 +/- 5.9, respectively; MFI values for the binding of Griffonia simplicifolia isolectin B4 (GS-IB4) lectin to NIH3T3 cells were 167 +/- 23, 170 +/- 19, and 100 +/- 14, respectively; MFI values for the binding of human natural IgG and IgM antibodies to NIH3T3 cells were 31 +/- 3, 32 +/- 4, and 22 +/- 4, respectively. H blood group antigen was detected on NIH3T3 cells after Ad5hSeFT infection and resulted in more than 40% reduction in the level of gal epitope on the cell surface. This reduction increased the resistance of NIH3T3 cells to lysis by normal human serum.
To examine the effects of the expression of antisense RNA transcripts complementary to the pig alpha(1,3) galactosyltransferase [alpha(1,3)GT]mRNA on the expression of Gal alpha(1,3) Gal structure (gal epitope) in cultured cell lines. Human adenoviral vectors were used to mediate the expression of antisense RNA. The expression levels of H blood group antigens and gal epitopes were analyzed by flow cytometry using FITC-UEA-I and FITC-GS-IB4 lectins, respectively. Recombinant adenoviruses, Ad5anti-sGT600 and Ad5-anti-sGT1100, which express antisense RNA complementary to different regions of the pig alpha(1,3) GT mRNA, were constructed and used to infect cell line of NIH3T3. The results showed about 30% reduction in the expression level of gal epitopes on the surface of NIH3T3 cells. In addition, co-expression of human secretor type alpha(1,2) fucosyltransferase [alpha(1,2)FT]cDNA and antisense RNA complementary to the pig alpha(1,3) GT mRNA led to a further reduction in the gal epitope level. Recombinant adenoviruses, Ad5anti-sGT600 and Ad5anti-sGT1100, are effective to down-regulate the gal epitope expression.