BACKGROUND: Complexing recombinant DNA with cationic liposomes is a convenient means of introducing foreign genes into cells (lipofection) and could potentially form the basis for genetically modifying diseased blood vessels in patients. The mechanism of lipofection is incompletely understood, but it is recognized that the degree of successful gene transfer is highly dependent on cell type. To date, there has been no reported experience with lipofection of human vascular smooth muscle cells. METHODS AND RESULTS: Primary cultures of human vascular smooth muscle cells were transfected under optimized conditions with a plasmid expressing either firefly luciferase (Luc) or nuclear-localized beta-galactosidase (NL-beta-gal). Cells were derived from either normal human internal mammary arteries (n = 6), fragments of primary atherosclerotic plaque (n = 4), or fragments of restenotic lesions (n = 5). Concurrent lipofection of rabbit vascular smooth muscle cells and NIH 3T3 cells was performed as well. Cultures derived from 15 patients all demonstrated positive expression of the reporter gene. Compared with NIH 3T3 cells, however, expression in human vascular smooth muscle cells was markedly reduced: in cells derived from internal mammary artery, Luc expression, normalized for protein content, was 123-fold lower than in NIH 3T3 cells, whereas the proportion of cells expressing NL-beta-gal was 30-fold lower. Luc expression in cells derived from restenotic tissue was significantly greater than from cells derived from primary plaque (P < .03). Within a given population of cells, the mitotic index of cells expressing the recombinant gene was significantly higher than the mitotic index for the total population of cells (P < .05). Finally, cotransfection experiments, in which lipofection of smooth muscle cells was performed using genes for NL-beta-gal and for human growth hormone, showed that among positive transfects, a high proportion of cells (23% to 36%) coexpressed both genes. CONCLUSIONS: The efficiency of successful lipofection in human vascular smooth muscle cells in vitro is low. Transfection appears to be preferentially facilitated in cells derived from restenotic tissue, and specific properties of smooth muscle cells, including growth rates, appear to be critical for successful transfection. Further elucidation of cell properties that promote transfection is required to augment the efficiency of liposome-mediated gene transfer in human vascular cells.
BACKGROUND: Complexing recombinant DNA with cationic liposomes is a convenient means of introducing foreign genes into cells (lipofection) and could potentially form the basis for genetically modifying diseased blood vessels in patients. The mechanism of lipofection is incompletely understood, but it is recognized that the degree of successful gene transfer is highly dependent on cell type. To date, there has been no reported experience with lipofection of human vascular smooth muscle cells. METHODS AND RESULTS: Primary cultures of human vascular smooth muscle cells were transfected under optimized conditions with a plasmid expressing either firefly luciferase (Luc) or nuclear-localized beta-galactosidase (NL-beta-gal). Cells were derived from either normal human internal mammary arteries (n = 6), fragments of primary atherosclerotic plaque (n = 4), or fragments of restenotic lesions (n = 5). Concurrent lipofection of rabbit vascular smooth muscle cells and NIH 3T3 cells was performed as well. Cultures derived from 15 patients all demonstrated positive expression of the reporter gene. Compared with NIH 3T3 cells, however, expression in human vascular smooth muscle cells was markedly reduced: in cells derived from internal mammary artery, Luc expression, normalized for protein content, was 123-fold lower than in NIH 3T3 cells, whereas the proportion of cells expressing NL-beta-gal was 30-fold lower. Luc expression in cells derived from restenotic tissue was significantly greater than from cells derived from primary plaque (P < .03). Within a given population of cells, the mitotic index of cells expressing the recombinant gene was significantly higher than the mitotic index for the total population of cells (P < .05). Finally, cotransfection experiments, in which lipofection of smooth muscle cells was performed using genes for NL-beta-gal and for humangrowth hormone, showed that among positive transfects, a high proportion of cells (23% to 36%) coexpressed both genes. CONCLUSIONS: The efficiency of successful lipofection in human vascular smooth muscle cells in vitro is low. Transfection appears to be preferentially facilitated in cells derived from restenotic tissue, and specific properties of smooth muscle cells, including growth rates, appear to be critical for successful transfection. Further elucidation of cell properties that promote transfection is required to augment the efficiency of liposome-mediated gene transfer in human vascular cells.