Uremia Accelerates both Atherosclerosis and Arterial Calcification in Apolipoprotein E Knockout Mice

Animals

All experiments were performed in apoE^−/− mice, which were obtained from IFFA-Credo (Lyon, France) and bred at the Necker Medical Faculty (Paris, France). The mice were housed in polycarbonate cages in a pathogen-free facility set on a 12-h light-dark cycle and given free access to water and regular laboratory chow (Harlan Teklad Global Diets). The components of the diet as listed by the manufacturer were 5.7% fat, 18.9% protein, 72.3% carbohydrates, 1.01% calcium, 0.65% phosphorus, 0.2% magnesium, and 1540 IU/kg vitamin D₃. All procedures were in accordance with National Institutes of Health guidelines for the care and use of experimental animals (NIH publication No. 85-23). After sexual maturation at 4 wk of age, the mice were separated by sex and housed in groups of up to five.

Creation of CRF

At 8 wk of age, the animals were randomly assigned into a uremic group (14 female, 8 male mice) and a nonuremic control group (19 female, 6 male mice). We used a two-step procedure to create uremia. Briefly, we applied cortical electrocautery to the right kidney through a 2-cm flank incision and performed left total nephrectomy through a similar incision 2 wk later as described by Gagnon et al. (17). In an initial series aimed at inducing severe reductions in renal function, we observed a 100% mortality of apoE^−/− mice at 24 h after contralateral kidney ablation, although 12 of 16 C57BL/6J mice survived for several weeks after the same degree of renal mass reduction (unpublished personal observations). Because the intensity of electrocoagulation determines the degree of CRF, we then aimed to produce a more moderate reduction in renal function and obtained serum urea concentrations between 25 and 30 mM. Control animals received sham operation that included decapsulation of both kidneys. Special care was taken to avoid damage to the adrenals. At 6 wk after surgery, mice were killed by exsanguination under ketamine/xylazine anesthesia (100 mg/kg and 20 mg/kg) in the nonfasting state.

Biochemical Serum Determinations

Blood was collected from the abdominal aorta into chilled dry tubes and spun in a refrigerated centrifuge, and serum was stored at −80°C. Serum levels of urea, total cholesterol, and triglycerides were assessed using a Hitachi 917 autoanalyzer (Roche, Meylan, France). Serum levels of advanced oxidation protein products (AOPP) were measured by spectrophotometry with a microplate reader as described previously (18). In separate groups of male and female apoE^−/−- mice (18 uremic mice and 24 control mice), we measured serum calcium and phosphorus with a Hitachi 917 autoanalyzer and mouse intact parathyroid hormone (iPTH) using a two-site ELISA (Immutopics, Quidel, France).

Quantification of Atheromatous Lesions in Aorta

For the aortic root sections, the vasculature was perfused with sterile PBS, and the ventricular edge and approximately 1 mm of the aortic root were immediately dissected under a microscope and cryomounted in optimal cutting temperature embedding medium. Cryosections, cut in 10-μm-thick slices, were collected and numbered from the appearance of the first cusp (point 0). Sections at 200-, 400-, 600-, and 800-μm distances from the cusps were stained with oil red-O, counterstained with hematoxylin, and mounted under coverslips. For the thoracic aorta sections, the vasculature was perfused and aortic sections from the left subclavian artery to the last intercostal artery branch were dissected and cryomounted. Cryosections, cut in 10-μm thicknesses, were numbered beginning at 4 mm distal from the left subclavian artery (point 0). Sections at 100-, 200-, 300-, and 400-μm distances from point 0 were stained as described above. The stained cryosections were magnified at ×5, and the images were captured on a microcomputer equipped with a program developed in Quips language that permitted quantification of the lesions (Leica, Cambridge, UK). The cross-section surface area of the vessel and cross-section surface area of the lesion were assessed.

Quantitative and Qualitative Evaluation of Aorta Calcification

Vascular calcifications in aortic root were evaluated by von Kossa staining in cryosections of aortic tissue. Briefly, the cryosections were placed in 5% silver nitrate solution (Sigma Aldrich, St. Louis, MO) for 30 min in darkness. Then, they were put in revelator solution (Kodak) for 5 min and fixed in 5% sodium-thiosulfate solution for another 5 min. Finally, they were stained with 2% eosin. Calcium deposits appeared in black on a bright red-colored surrounding tissue. We developed morphologic image processing algorithms for computer-assisted automated quantitative measurement of calcification from aortic sections (19), revealed from the von Kossa’s silver nitrate staining. Calcium deposits were measured inside and outside atherosclerotic lesions, reflecting atherosclerotic and medial calcium deposits, respectively. Data were expressed as the relative proportion of calcified area to total surface area of either the inside or the outside of atherosclerotic lesions and as the size of calcification granules on either surface, as described previously (19).

To perform additional qualitative characterization of vascular calcification type in CRF apoE^−/− mice, we performed Fourier transform infrared (FTIR) spectroscopy of the vascular tissue (Bruker vector 22, Bruker A590 microscope) (20).

Quantification of Monocyte-Macrophage Infiltration and Collagen Content in Aortic Lesions

Monocyte-macrophages (MOMA) were detected in atherosclerotic lesion by immunostaining as described previously (21). Briefly, aortic sections were incubated with 10% normal goat serum for 30 min at room temperature, washed in PBS, and incubated with a primary rat monoclonal antibody against mouse macrophages (clone MOMA-2; BioSource International, Camarillo, CA) for 1 h at room temperature. The secondary antibody was a biotin-horseradish peroxidase-conjugated goat anti-rat IgG (Vector Laboratories, Biovalley, Marne la Vallée, France). Immunostainings were visualized after incubation with a peroxidase detection system (Vectastain ABC kit, Vector Laboratories) using 3-amino 9-ethyl carbazole (Sigma Aldrich) as substrate. At least four sections per animal were analyzed for each immunostaining.

The lesion collagen content was determined by staining with Sirius red (21). The image was captured on a microcomputer equipped with Histolab software (Microvision Instruments, Envy, France) and analyzed by computerized image analysis measuring the relative area/density in 12 continuous fields in each Sirius red–stained section. Because collagen content may depend on the size of the lesions, only large lesions were examined for determination of content.

BP Measurements

In a separate set of CRF apoE^−/− mice (3 female, 3 male mice) and age-matched nonuremic apoE^−/− control mice (3 female, 3 male mice), we measured mean arterial BP at the day of killing using direct intra-arterial recording as described previously (22). Briefly, the mice were anesthetized with ketamine/xylazine (100 mg/kg and 20 mg/kg, intraperitoneally), and a PE50 catheter, stretched to reduce the tip diameter, was inserted into the abdominal aorta. The catheter was filled with heparinized saline (20 U/ml), and the distal end was attached to a BP transducer (Gould pressure processor). After a 5-min stabilization period, three BP tracings were obtained, and BP was averaged from the three measurements.

Statistical Analyses

Data were analyzed by ANOVA, linear regression analyses, and unpaired t test, as appropriate. Results were expressed as means ± SEM. Differences between groups were considered significant at P < 0.05.

Body Weight and Serum Biochemistry

Body weight did not differ between the uremic and nonuremic apoE^−/− mice throughout the study (24.4 ± 1.3 versus 23.5 ± 1.2 g, respectively). Serum urea levels were used to assess renal function. At 6 wk after surgery, serum urea concentrations in CRF mice increased by 225% above baseline and tended to be stable, with minimal variations during the 6-wk period (Figure 1A). Serum total cholesterol and triglyceride concentrations were significantly higher in CRF mice compared with non-CRF controls (Figure 1, B and C). Serum AOPP levels were significantly higher in CRF animals than in controls (Figure 1D). By linear regression analysis, AOPP values were closely associated with serum urea values (r² = 0.71, P < 0.001; Figure 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Serum concentrations of urea (A), total cholesterol (B), triglycerides (C), and advanced oxidation protein products (AOPP; D). Apolipoprotein E-deficient (apoE^−/−) mice were subjected to either sham operation (control mice) or a two-step creation of chronic renal failure (CRF). CRF mice, n = 22; control mice, n = 25. EC, electrocoagulation of right renal cortex; Nx, left total nephrectomy. P < 0.001 ANOVA between groups. *P < 0.05 Fischer post hoc least significant difference test between control and CRF.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Correlation between serum AOPP and serum urea in the CRF and control apoE^−/− groups collapsed together. CRF mice, n = 22; control mice, n = 25.

Uremic animals had higher serum calcium and phosphate concentrations (P < 0001, P < 0.05, respectively; Table 1) and higher serum iPTH values (P < 0.05; Table 1). Mean arterial BP was comparable between the groups (Table 1).

Atheromatous Lesions of Aorta

Six weeks of CRF significantly increased aortic plaque area fraction in thoracic aorta in CRF mice compared with controls (P < 0.04; Figure 3). The impact of uremia on atherosclerosis in the aortic root seemed to be less pronounced because lesion sizes at this site were comparable in the two groups (Table 2). Also in the aortic root, lesions were larger and more advanced in female than in male apoE^−/− mice (P = 0.03; data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Atherosclerotic lesion cross-section area in thoracic aorta of control apoE^−/− and CRF apoE^−/−mice. CRF mice, n = 22; control mice, n = 25. Data are means ± SEM.

Aortic Calcification, Total Collagen Content, and MOMA Infiltration

CRF apoE^−/− mice had significantly larger areas of atheromatous and medial calcification than nonuremic apoE^−/− mice (Table 2, Figure 4). Moreover, intimal and medial calcification sizes were increased in CRF mice compared with controls, although this difference reached statistical significance only in the intima. There were no correlations between intimal or medial calcification area or size and serum calcium, phosphorus, calcium × phosphorus product, cholesterol, triglycerides, AOPP, or atherosclerotic lesion area (data not shown). However, a significant correlation was observed in uremic mice between intimal and medial calcification area (r² = 0.94, P < 0.001) and between intimal and medial calcification sizes (r² = 0.66, P < 0.001), suggesting a link between atherosclerotic and medial vascular calcification.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

Representative images of aorta root sections of control apoE^−/− and CRF apoE^−/− mice. (A) Von Kossa silver nitrate staining (calcification in black). (B) Morphologic image processing (calcification in red) (C) Representative images of aortic root of CRF apoE^−/− mouse showing a higher magnification (×10) of the object after 2% eosin (1) as well as after 2% eosin and von Kossa staining (2). M, media; I, intima; L, lumen.

FTIR spectroscopy confirmed a significant overall increase of mineral density in the plaques of CRF apoE^−/− mice compared with control mice. Moreover, it showed that the calcium deposits were made of hydroxyapatite (carbapatite) and calcite (calcium carbonate anhydrous) crystals in both the intimal and medial lesion sites. The calcium deposits in the intimal lesions were frequently co-localized with the presence of cholesterol crystals (Figure 5).

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

Fourier transformed infrared spectroscopy of aortic tissue of an apoE^−/− mouse with CRF and a control apoE^−/− mouse, respectively, showing an overall increase in mineral density (calcite and hydroxyapatite peaks) in the plaques of CRF apoE^−/− mice. Typical peaks of calcite are seen at 1410, 876, and 719 cm⁻¹ and those of hydroxyapatite at 1035 and 957 cm⁻¹, respectively.

Finally, uremia was associated with a 38% increase in plaque collagen content (Table 2). However, the percentage of lesion cross-section area occupied by macrophages, as revealed by MOMA-2 staining, was comparable between the two groups (Table 2).