Construction of NK7
The full length bovine α1–3 GT cDNA was cut at Apa- and BamH-sites of N3GA  and ligated to an pEGFP-1 vector (CLONTECH, Palo Alto, CA) to form an E1GA vector. The hTERT promoter region was amplified by PCR and subcloned, as previously described . Briefly, genomic DNA was extracted from human pancreatic carcinoma cell line, MIA PaCa-2 . Then, 100 ng of genomic DNA were amplified by PCR using Gene Taq (Nippon gene, Tokyo, Japan). The following primer sequences were used for PCR: 5-TCTGGATTCCTGGGAAGTCCTCA-3, 5-ACGCAGCGCTGCCTGAAACTCG-3. A 1082-bp PCR product was subcloned into the pGEM-T vector (Promega, Madison, WI). The fragment was cut at Sal- and Sph-sites and ligated to a pSP72 vector (Promega), then cut at Xho- and Kpn-sites and finally ligated to the E1GA vector (NK7, Figure 1).
Dual luciferase assay
The luciferase reporter plasmid was constructed by subcloning of the hTERT promoter region to Kpn I and Bgl II sites in the pGL3-Basic vector (Promega). MIA cells were seeded at 3 × 104 per 16-mm well and transfected 24 h later with complexes containing 1.2–1.9 μl of LipofectAMINE, 4 μl Plus reagent (GIBCO, Rockville, MD) which contained the Renilla luciferase gene as a transfection efficiency control, and 420 ng of firefly luciferase reporter plasmid per well. Lysates were prepared 24–48 h after transfection by adding 100 μl of reporter lysis buffer (dual luciferase reporter system, Promega), and their luciferase activities were measured with an analytical luminometer (model TD-20/20, Turner Designs, Sunnyvale, CA) and expressed as relative luciferase units (RLU), calculated by determining the ratio of the intensity of the light produced by the Renilla luciferase pRL-TK plasmid. All luciferase assays were performed in triplicate.
Cell lines and transfection
MIA cells and SUSM-1 cells  were used. SUSM-1 cells were negative for telomerase-activity, such that the hTERT gene-promoter was expected to not be active in SUSM-1 cells. Cells were cultured in Dulbecco's modified eagle medium (GIBCO), containing 10 % fetal calf serum in a 5% CO2 incubator. Twenty micrograms of NK7 were electroporetically transfected into 2 × 106 cells with a GENE PULSER (BIO RAD, Hercules, CA). Approximately 1 × 103 cells per well were cultured in a medium containing neomycin (Geneticin, GIBCO), in a 96-well plate, for 10–14 days. The neomycin concentrations for the selection of MIA cells and SUSM-1 cells were 1 mg/ ml and 0.4 mg/ ml, respectively. Individual neomycin resistant clones were picked up and transferred to a 24 well tissue culture plate, and DNA extracted from these cells was subjected to PCR to confirm integration of the α1–3 GT cDNA. PCR was performed using the following primers. Sense strand primer: 5'-AGCTCAGTAGAACTTGGTACTTTT-3', and anti-sense strand primer: 5'-CATCTGATTACCACAGGTTCATT-3'. After an initial denaturation for 3 minutes at 94°C, amplification was performed for 25 cycles in a final volume of 50 μl using 2.5 u of Taq DNA polymerase. Each cycle consisted of 30 seconds at 94°C, 30 seconds at 60°C and 60 seconds at 72°C. The PCR products were separated by 2% agarose gel electrophoresis and the products were visualized by ethidium bromide. Ten neomycin-resistant clones from MIA cells or SUSM-1 cells, which were positive for the 1355 bp-PCR product, were transferred to 10-cm tissue culture plates to obtain a sufficient number of cells for flow cytometry and complement dependent cross-match test (CDC).
The promoter activity of the hTERT gene was expected to operate in the endogenously telomerase-positive MIA cells, but not in the telomerase-negative SUSM-1 cells. Promoter activity was reflected by detection of the αGal epitope on the MIA cell surface, in cells which had been efficiently transfected with NK7. The αGal epitope was detectable by flow cytometry using FITC-conjugated BS-I ISOLECTIN B4 (IB4 lectin, Sigma, St. Louis, MO). IB4 lectin recognizes the terminal galactosyl epitope in the α linkage . Flow cytometry was performed as follows. Approximately, 1 × 106 cells were incubated in 10 μg of IB4 lectin. After a 60-min incubation, the cells were washed twice with PBS, and analyzed on a FACscan (Becton-Dickinson, Mountain View, CA).
Complement dependent cytotoxic cross-match test (CDC)
As previously reported, we used a CDC assay to examine the susceptibility of transfected cells to antibody-mediated cell lysis . Approximately 2 × 106 cells were cultured with 50 μl of human serum at 37°C. To exclude the possible involvement of anti-blood type antibodies, pooled human sera from blood types A, B, O and AB donors were used as the source of natural antibodies, at each experiment. After a 50-min incubation, 5 μl of rabbit complement were added to the plates and a further 70-min incubation was carried out. Then, a 5 % eosin solution was added to the plates, and dead cells which appeared darker than the living cells were counted under a microscope. In each experiment, the three wells containing the highest number of dead cells were selected, and the percentage of dead cells was calculated. The ANOVA was used for the statistical analysis. CDC assays were performed on the same day of the flowcytometry.
Evaluation of apoptosis by the annexin V binding assay
The annexin V binding assay was performed using the Annexin V-FITC Apoptosis Detection kit (PharMingen) according to the supplier's instruction. At least 1 × 106 cells prepared by human serum and rabbit complement described in CDC were incubated with FITC-conjugated annexin V at room temperature for 15 min. The cells were then analyzed by flow cytometry.