Antisense to transforming growth factor-beta1 messenger RNA reduces vein graft intimal hyperplasia and monocyte...

Autogenous vein grafts are commonly used for arterial reconstructive procedures. Their success is limited by the development of intimal hyperplasia (IH), a fibroproliferative disease that predisposes the grafts to occlusive stenosis. Mesenchymal cell proliferation and the deposition of an extracellular matrix characterize neointimal development. Increasing evidence suggests that, regardless of blood vessel type, IH results from complex interactions among vessel wall cells, infiltrating leukocytes, and cytokines. Transforming growth factor-beta1 (TGF-beta1) is a pleiotropic cytokine with powerful effects on inflammatory cell chemotaxis; smooth muscle cell, fibroblast, and endothelial cell proliferation; and extracellular matrix synthesis.

Epigastric vein to common femoral artery interposition grafts were placed in male Lewis rats and harvested at 1, 2, 4, and 12 weeks after surgery. We used replication-defective adenoviruses to deliver a control reporter gene for the enzyme beta-galactosidase (Ad-GAL), empty virus (Ad-CMVpLpA), or the sequence encoding the antisense strand of TGF-beta1 (Ad-AST). The vein graft was transduced passively in medium containing 10 7 plaque-forming units per milliliter of Ad-GAL, Ad-CMVpLpA, or Ad-AST for 20 minutes at room temperature. The adenovirus-treated grafts were compared with grafts treated with medium without virus (sham).

The Ad-GAL control grafts showed beta-galactosidase activity from 3 days to 4 weeks. Twenty percent of cells were positive out to 2 weeks, at which time the number of cells positive for beta-galactosidase activity began to decline. Treatment with Ad-AST resulted in a significant reduction vs sham, Ad-CMVpLpA, and Ad-GAL in TGF-beta1 messenger RNA, total TGF-beta1 protein, and bioactive TGF-beta1 protein. Neointimal area was significantly reduced in the Ad-AST group vs Ad-GAL at 4 weeks, vs Ad-CMVpLpA at 4 and 12 weeks, and vs sham at 2 and 4 weeks. The medial/adventitial layer was significantly thicker in the Ad-AST group than the Ad-GAL group at 12 weeks. In addition, we studied the effect of Ad-AST on monocyte chemotactic protein 1 (MCP-1). Although the reduction in TGF-beta1 resulted in a reduction of MCP-1 messenger RNA in whole-graft homogenates and MCP-1 protein-positive staining in histologic sections from the perianastomotic region, no reduction in the number of ED1-positive cells (monocytes and macrophages) was observed.

Perioperative antisense TGF-beta1 treatment of the vein to be used in arterial reconstructions resulted in a prolonged diminution of IH; this emphasizes the importance of TGF-beta1 in neointimal thickening and indicates that ex vivo gene therapy can reduce the vessel's predisposition to IH.

The main cause of occlusion and graft failure after peripheral and cardiac arterial reconstruction is IH. The study of the mechanisms and mediators of IH, including TGF-beta1, should lead to future gene therapies to prevent or limit IH. The clinical effect of such treatments would be enormous, because they would increase graft longevity, thereby enhancing quality of life and enabling patients to live without the threat of limb loss or recurrent heart attack.