Mouse model of venous bypass graft arteriosclerosis

Abstract

Saphenous vein grafts are widely used for treatment of severe atherosclerosis via aortocoronary bypass surgery, a procedure often complicated by later occlusion of the graft vessel. Because the molecular mechanisms of this process remain largely unknown, quantitative models of venous bypass graft arteriosclerosis in transgenic mice could be useful to study this process at the genetic level. We describe herein a new model of vein grafts in the mouse that allows us to take advantage of transgenic, knockout, or mutant animals. Autologous or isogeneic vessels of the external jugular or vena cava veins were end-to-end grafted into carotid arteries of C57BL/6J mice. Vessel wall thickening was observed as early as 1 week after surgery and progressed to 4-, 10-, 15-, and 18-fold original thickness in grafted veins at age 2, 4, 8, and 16 weeks, respectively. The lumen of grafted veins was significantly narrowed because of neointima hyperplasia. Histological and immunohistochemical analyses revealed three lesion processes: marked loss of smooth muscle cells in vein segments 1 and 2 weeks after grafting, massive infiltration of mononuclear cells (CD11b/18+) in the vessel wall between 2 and 4 weeks, and a significant proliferation of vascular smooth muscle cells (alpha-actin+) to constitute neointimal lesions between 4 and 16 weeks. Similar vein graft lesions were obtained when external jugular veins or vena cava were isografted into carotid arteries of C57BL/6J mice. Moreover, no significant intima hyperplasia in vein-to-vein isografts was found, although there was leukocyte infiltration in the vessel wall. Thus, this model, which reproduces many of the features of human vein graft arteriosclerosis, should prove useful for our understanding of the mechanism of vein graft disease and to evaluate the effects of drugs and gene therapy on vascular diseases.

Figure 1.

Schematic representation of vein bypass graft. The right common carotid artery was ligated with an 8-0 silk suture (a) and dissected between the middle ties and passed through the cuffs, respectively (b). The vessel, together with the cuff handle, was fixed with microhemostat clamps; the suture at the end of the artery was removed; and a segment of the artery was turned inside out with a stent and fine tweezers to cover the cuff body (c), which was fixed to the cuff with an 8-0 silk suture (d). The right external jugular or vena cava vein segment (1 cm) was harvested and grafted between the two ends of the carotid artery by sleeving the ends of the vein over the artery cuff and suturing them together with an 8-0 suture ligation (e). The cuff handle was cut off, and the vascular clamps were removed; pulsations were seen in the grafted vein.

Figure 2.

H&E-stained sections of mouse control vein and vein grafts. Mice underwent anesthesia, and the external jugular vein was grafted into the carotid artery. Animals were sacrificed at 0 (A), 1 (B), 2 (C), 4 (D), 8 (E), and 16 (F) weeks after surgery, and the grafted tissue fragments were fixed in 4% phosphate-buffered (pH 7.2) formaldehyde, embedded in paraffin, sectioned, and stained with H&E. Arrowheads indicate the control vessel wall (A) and neointima (B through D). A portion of neointima was shown in E and F; bar = 50 μm.

Figure 3.

H&E-stained section of mouse vein graft 16 weeks after operation. Arrows indicate neointima; bar = 50 μm.

Figure 4.

Neointima thickness in vein autografts. The procedure for animal models and the preparation of H&E-stained sections are the same as described in the legend to Figure 2 ▶ . Thickness was measured microscopically. Four regions of each section along a cross were measured, and five sections per animal were selected. A shows a graph of intima and media or neointima thickness (mean ± SD) obtained from six to eight animals per time point. B shows data (mean) in fold converted from A, where the thickness of the control vein intima and media is taken as one. *Significant difference from the control; P < 0.05.

Figure 5.

Neointima hyperplasia in vein autografts. The procedure for animal models and the preparation of H&E-stained sections are the same as those described in the legend to Figure 2 ▶ . Total H&E-stained nuclei in 100-μm lengths of (neo)intima and media of veins were counted manually. Two opposite areas from each section were counted, and five sections per animal were selected. The graph shows data (mean ± SD) obtained from six to eight animals per time point. *Significant difference from the control; P < 0.05.

Figure 6.

H&E-stained sections of control vein, autografts, and isografts. Mice underwent anesthesia, and the external jugular vein (A) was autografted (D) or isografted (E), or the vena cava (B) was isografted (F) into the carotid artery. C represents the result of vein-to-vein isograft. Animals were sacrificed 4 weeks postoperatively, and the grafted tissue fragments were harvested, embedded in paraffin, sectioned, and stained with H&E. Arrows indicate the vessel wall or neointima; bar = 50 μm.

Figure 7.

Comparison of neointima thickness between venous bypass auto- and isografts. The procedure for animal models and the preparation of H&E-stained sections are the same as those described in the legend to Figure 6 ▶ . The thickness was measured microscopically. The graph shows data of neointima thickness (mean ± SD) obtained from six to eight animals per group. There are no significant differences among groups.

Figure 8.

Immunohistochemical staining demonstrates the presence of MAC-1⁺ mononuclear cell infiltration in vein grafts. Sections derived from vein grafts at 1 (A and B), 2 (C), 4 (D), 8 (E), and 16 (F) weeks were labeled with normal rat serum (A) or a rat monoclonal antibody (CD11b/18; B through F) against MAC-1⁺ leukocytes and developed with alkaline phosphatase-anti-alkaline phosphatase techniques. A counterstaining (blue) with hematoxylin was performed. Filled arrows indicate neointima, and open arrows indicate examples of typical MAC-1-positive cells (red). D through F show a portion of neointima; bar = 20 μm.

Figure 9.

Immunohistochemical staining demonstrates the presence of α-actin⁺ smooth muscle cells in vein grafts. Sections derived from vein grafts at 0 (A), 4 (B), 8 (C), and 16 (D) weeks were labeled with a mouse monoclonal antibody against α-actin conjugated with an alkaline phosphatase and developed with the substrate with counterstaining. Filled arrows indicate normal intima/media (A) or neointima (B through D), and open arrows indicate examples of typical α-actin-positive cells (red). B through D show a portion of neointima; bar = 50 μm.