Arterial gene transfer of acidic fibroblast growth factor for therapeutic angiogenesis in vivo: critical role of secretion signal in use of naked DNA

Objective: Previous studies have demonstrated that arterial gene transfer of naked DNA encoding for a secreted protein may permit modulation of the host phenotype despite a low transfection efficiency. Acidic fibroblast growth factor (aFGF) is an angiogenic growth factor, but is not secreted by intact cells. In the current study, we investigated the hypothesis that addition of a hydrophobic leader sequence to achieve active secretion of the gene product would permit therapeutic angiogenesis following arterial gene transfer of naked DNA encoding for aFGF. Methods: Ten days following surgical induction of unilateral hindlimb ischemia, New Zealand white rabbits were randomized to intra-arterial gene transfer with one of three plasmids: p267 (encoding non-secreted aFGF, n = 10), pMJ35 (encoding secreted aFGF) (n = 10), or 500 mu g of pGSVLacZ (control, n = 10) (500 mu g each). All animals were studied at 30 days post-gene transfer for evidence of therapeutic angiogenesis. Results: pMJ35 transfectants had more angiographically visible collaterals (angiographic score = 0.76 +/- 0.02) than either p267 (0.55 +/- 0.02, p < 0.01) or LacZ (0.47 +/- 0.02, p < 0.001). Limb blood pressure ratio for pMJ35 was 0.88 +/- 0.02 vs. 0.68 +/- 0.04 for p267 (p < 0.01) and 0.57 +/- 0.04 for LacZ(p < 0.001). Vascular resistance was significantly lower in the pMJ35 group, compared with that in pGSVLacZ group, both in resting state (3.2 +/- 0.4 vs. 7.4 +/- 1.4 respectively, p < 0.05) and after the administration of nitroprusside. Capillary density (per mm(2)) was also superior in pMJ35 group (274 +/- 10) vs. p267 (204 +/- 9, p < 0.01) and LacZ (177 +/- 6, p < 0.001). Conclusion: The paracrine effects of a secreted gem product may obviate the need for adjunctive vectors in strategies of arterial gene therapy. (C) 1997 Elsevier Science B.V.