(Micrographs) Representative micrographs of MT-stained sections (original magnification 40x). Myocardiocytes are stained red, elastin fibers and nuclei appear black, extracellular matrix appears green. Fibrofatty atherosclerotic lesions are present in the 3 aortic cups in PBS and CD31Rg 1-2 groups, whereas they are limited to 2 of the 3 cusps by CD31Rg treatment. Adventitial inflammation was reduced in the CD31Rg group (white arrows). (Bottom right panel) Morphometric analysis was performed at 200, 400, 600, and 800 µm from the appearance of the first aortic cusp, starting from the ventricular side of the aortic root, by oil-red O staining and computer-assisted image analysis. The CD31Rg-treated group showed significantly smaller lesions as compared to the CD31Rg 1-2 and PBS control groups. *p < 0.01 versus CD31Rg 1-2 and PBS, n = 14 mice/group. CD31Rg = CD31 receptor globulin; CD31Rg 1-2= truncated CD31 receptor globulin lacking domain 1-2; MT = Masson's trichrome staining; PBS = phosphate-buffered saline.

Masson's trichrome staining and double immunohistochemistry for CD31 (MEC 13.3) and endothelia (lectin BSI) on comparable brachiocephalic plaques. Representative micrographs showing blunted neovascularization and intraplaque hemorrhage in CD31Rg-treated mice (3 of 14 mice, 21%, p < 0.05, chi-square test) as compared to CD31Rg 1-2 (6 of 14 mice 42%) and PBS (9 of 14 mice, 64%) control mice. Intraplaque fibrin deposition (red, MT column, black arrows) localizes at areas of rich neovascularization as detected by FITC-lectin BSI staining of endothelial cells within the plaques (green, BSI column, white arrows). Of note, most of the intraplaque neovessels were negative for CD31 (red, MEC13.3 column) as showed by the merge of fluorescent images (MERGE + DAPI column). Original magnification 100x. Abbreviations as in Figure 1.

Immunohistochemistry of CD31Rg in aortic root crysosections. CD31Rg (A), but not CD31Rg 1-2 (B), binds a consistent number of plaque-infiltrating cells. CD31Rg binding was detected on CD31^bright cells (white arrows) as well as on CD31^low cells (white arrow heads). CD31Rg staining was revealed by HRP/DAB (brown, A to B; green, C and F), nuclei were counterstained by hematoxylin (light blue, A to B) or DAPI (blue, E to F). CD31⁺ cells were visualized by a monoclonal rat antimouse CD31 antibody (MEC13.3) and an Alexafluor 555-conjugated secondary antibody (red, D and F). Original magnification 200x. A = adventitia; L = lumen; M = media; other abbreviations as in Figure 1.

The percentage of circulating activated (CD69⁺) helper (CD4⁺) cells within T lymphocytes (double gate FSC/SSC x CD3⁺) was analyzed by flow cytometry. At the end of the study, CD31Rg-treated mice showed significantly reduced numbers of circulating activated T-helper lymphocytes as compared to PBS and CD31Rg 1-2 controls. Similar data were obtained 1 week after the first electroporation round. *p < 0.05 versus PBS and CD31Rg 1-2, n = 14 mice/group. Abbreviations as in Figure 1.

Spleen cells were stimulated with concanavalin A for 3 days. ³[H]thymidine was added to the cultures for the last 18 h. Cell proliferation was measured in a beta counter and is expressed in counts per minute (cpm). T-cell proliferation of CD31Rg-treated mice was significantly reduced as compared control mice, even in response to a powerful stimulus such as the polyclonal mitogen concanavalin A. *p < 0.05 versus CD31Rg 1-2 and PBS, n = 14 mice/group. Abbreviations as in Figure 1.

(Top) The percentage of CD25⁺/FoxP3⁺ regulatory T cells (Tregs) within total CD4⁺ T lymphocytes was significantly increased in the peripheral blood of CD31Rg-treated mice as compared with the controls. *p < 0.01, n = 14 mice/group. (Bottom) The suppressive capacity of Tregs was not altered by CD31Rg treatment (n = 14 mice/group, NS). CD4⁺CD25 negative T-cells were incubated with Tregs at the indicated ratios, and proliferation was measured as before. Other abbreviations as in Figure 1.