Acute perioperative-stress-induced increase of atherosclerotic plaque volume and vulnerability to rupture in apolipoprotein-E-deficient mice is amenable to statin treatment and IL-6 inhibition

Abstract

Myocardial infarction and stroke are frequent after surgical procedures and consume a considerable amount of benefit of surgical therapy. Perioperative stress, induced by surgery, is composed of hemodynamic and inflammatory reactions. The effects of perioperative stress on atherosclerotic plaques are ill-defined. Murine models to investigate the influence of perioperative stress on plaque stability and rupture are not available. We developed a model to investigate the influence of perioperative stress on plaque growth and stability by exposing apolipoprotein-E-deficient mice, fed a high cholesterol diet for 7 weeks, to a double hit consisting of 30 min of laparotomy combined with a substantial blood loss (approximately 20% of total blood volume; 400 µl). The innominate artery was harvested 72 h after the intervention. Control groups were sham and baseline controls. Interleukin-6 (IL-6) and serum amyloid A (SAA) plasma levels were determined. Plaque load, vascular smooth muscle cell (VSMC) and macrophage content were quantified. Plaque stability was assessed using the Stary score and frequency of signs of plaque rupture were assessed. High-dose atorvastatin (80 mg/kg body weight/day) was administered for 6 days starting 3 days prior to the double hit. A single dose of an IL-6-neutralizing antibody or the fusion protein gp130-Fc selectively targeting IL-6 trans-signaling was subcutaneously injected. IL-6 plasma levels increased, peaking at 6 h after the intervention. SAA levels peaked at 24 h (n=4, P<0.01). Plaque volume increased significantly with the double hit compared to sham (n=8, P<0.01). More plaques were scored as complex or bearing signs of rupture after the double hit compared to sham (n=5-8, P<0.05). Relative VSMC and macrophage content remained unchanged. IL-6-inhibition or atorvastatin, but not blocking of IL-6 trans-signaling, significantly decreased plaque volume and complexity (n=8, P<0.01). Using this model, researchers will be able to further investigate the pathophysiology of perioperative plaque stability, which can result in myocardial infarction, and, additionally, to test potential protective strategies.

Keywords: Atherosclerosis; Mouse model; Perioperative stress.

Fig. 1.

Lipoprotein profiles and plasma IL-6 levels in C57BL6/J animals and in ApoE-deficient animals on a Western diet subjected to the combination of surgery and blood loss (double hit). (A) For baseline (BSL) measurements, blood was drawn from the retrobulbar plexus without any intervention and lipoprotein profiles were measured. Measurements for the other groups took place after the procedure. There were no differences in any of the components except for a decrease in VLDL cholesterol levels in the surgery group. Kruskal–Wallis and Dunn's post hoc test, n=5 each, P<0.01. (B) ELISA for IL-6 detected a significant increase in wild-type mice subjected to the double-hit protocol. Separate animals had to be used to avoid interference with the plaque growth protocol due to additional blood draws. The IL-6 peak occurred at 6 h after procedure. ApoE-deficient mice on a Western diet had slightly higher IL-6 levels compared to the wild types and the IL-6 increase was significantly higher compared to the baseline levels as well as compared to the levels of the wild-type mice. One-way ANOVA and t-test, n=4, for each time point, *P<0.05; **P<0.01; ^§P<0.05; vs BSL.

Fig. 2.

Plaque volume in atherosclerosis-prone mice increased with perioperative stress. Matched ApoE-deficient mice of mixed gender, 8 weeks of age, were placed on a Western diet for 7 weeks and subjected to blood loss, laparotomy or the combination thereof to exert perioperative stress (double hit). (A) Plaque area was determined on H&E staining every 42 µm throughout the innominate artery and plaque volume was calculated. Only blood loss or the combination of surgery and blood loss caused a significant increase of plaque volume in the innominate artery. Kruskal–Wallis test. BSL, baseline. (B) Representative micrographs of H&E staining of plaques at the branching of the innominate artery. The overall plaque load was rather small, but eccentric plaques in the double-hit group had complex lesions, whereas animals in the surgery or blood-loss groups had lesions resembling foam cell lesions.

Fig. 3.

Composition of plaques. (A-D) Relative macrophage and smooth muscle cell content did not change in plaques from mice subjected to the combination of surgery and blood loss (double hit). (A,C) Macrophages (CD68) and (B,D) vascular smooth muscle cells (VSMCs; α-smooth muscle actin) were stained by immunofluorescence, and relative areas staining positive for the respective antigens were morphometrically determined. No significant change of the relative contribution of VSMC or macrophages to the larger lesions was detected when double hit was inflicted. Kruskal–Wallis test, sham n=3; surgery n=4; bleed n=7; double hit n=8; n varies due to differential presence of measureable lesions in groups, P=n.s. (E) Collagen content was very low in the plaques and distribution signatures did not overtly differ between groups. (F) Polymorphonuclear (PMN) cell recruitment to the plaques was low. Mainly, PMNs were located in the subendothelial areas likely owing to the acute recruitment of myeloid cells as outlined in the text. (G) T-cell recruitment was mainly observed in the adventitia. Within the plaques, T cells were extremely rare. (H) To assess whether differences in leukocyte proliferation could be responsible for differences in leukocyte content, proliferating cell nuclear antigen (PCNA) was stained. No differences between the groups were detected. Proliferation in the plaques was extremely rare. Arrows indicate typical areas with antigen positive staining.

Fig. 4.

Signs preceding rupture of plaques were detected in animals exposed to the combination of surgery and blood loss (double hit). (A) A small proportion of lesions in control, but a large proportion of mice exposed to the double hit, demonstrated necrotic areas in the core of the plaque. Such necrosis had to be detectable on at least 50% of the sections to score 1 point. (B) Hemorrhage in the plaque was detected on H&E staining by the presence of red blood cells in the center of the lesion (arrowheads). Signs for intraplaque bleeding had to be present on at least 10% of the analyzed sections that span the whole innominate artery. (C) Buried fibrous caps were defined as αSMA-positive streaks (white arrowheads) that had been overgrown by new plaque material and were interpreted as buried fibrous caps secondary to plaque rupture, reorganization and overgrowth by plaque material. (D) Scoring of the features depicted in A-C with one point each revealed the presence of complex lesions in 60% of the double-hit mice, whereas control mice only had one feature in 32% of the cases, which mostly represented necrotic areas as shown in A. Fisher's exact test, sham, double hit n=8; surgery n=7; bleed n=5; **P<0.01 vs SHAM.

Fig. 5.

High-dose statin therapy 3 days prior to and 3 days after surgery in mice exposed to the double hit, and IL-6 signaling pathway interception. Mice were treated with 80 mg/kg body weight/day atorvastatin, blocking antibodies against IL-6 or the selective inhibition of IL-6 trans-signaling by gp130-Fc by a single subcutaneous injection. (A) Total cholesterol (n=5; P<0.05), (B) IL-6 plasma levels (n=4; P<0.05) and (C) plaque volume (double hit n=8; atorvastatin n=9; P<0.05) decreased significantly under atorvastatin. Mann–Whitney U-test. (D) Plaque complexity as assessed by the score was likewise significantly reduced (Fisher's exact test, double hit n=8; atorvastatin n=9; P<0.05). (E) The blocking antibody against IL-6 reduced plaque volume, whereas the selective inhibition of IL-6 trans-signaling by gp130-Fc did not yield any reduction in plaque volume. Mann–Whitney U-test; double hit, gp130-Fc n=8; isotype n=5; IL-6 ab n=8; P<0.05. (F) Plaque complexity was reduced by blocking antibody (Fisher's exact test; double hit, gp130-Fc n=8; isotype n=5; IL-6 ab n=8; **P<0.05 vs double hit).