Fetuin-A protects against atherosclerotic calcification in CKD

Abstract

Reduced serum levels of the calcification inhibitor fetuin-A associate with increased cardiovascular mortality in dialysis patients. Fetuin-A-deficient mice display calcification of various tissues but notably not of the vasculature. This absence of vascular calcification may result from the protection of an intact endothelium, which becomes severely compromised in the setting of atherosclerosis. To test this hypothesis, we generated fetuin-A/apolipoprotein E (ApoE)-deficient mice and compared them with ApoE-deficient and wild-type mice with regard to atheroma formation and extraosseous calcification. We assigned mice to three treatment groups for 9 wk: (1) Standard diet, (2) high-phosphate diet, or (3) unilateral nephrectomy (causing chronic kidney disease [CKD]) plus high-phosphate diet. Serum urea, phosphate, and parathyroid hormone levels were similar in all genotypes after the interventions. Fetuin-A deficiency did not affect the extent of aortic lipid deposition, neointima formation, and coronary sclerosis observed with ApoE deficiency, but the combination of fetuin-A deficiency, hyperphosphatemia, and CKD led to a 15-fold increase in vascular calcification in this model of atherosclerosis. Fetuin-A deficiency almost exclusively promoted intimal rather than medial calcification of atheromatous lesions. High-phosphate diet and CKD also led to an increase in valvular calcification and aorta-associated apoptosis, with wild-type mice having the least, ApoE-deficient mice intermediate, and fetuin-A/ApoE-deficient mice the most. In addition, the combination of fetuin-A deficiency, high-phosphate diet, and CKD in ApoE-deficient mice greatly enhanced myocardial calcification, whereas the absence of fetuin-A did not affect the incidence of renal calcification. In conclusion, fetuin-A inhibits pathologic calcification in both the soft tissue and vasculature, even in the setting of atherosclerosis.

Figure 1.

Serum chemistry, BUN, total calcium, phosphate, intact parathyroid hormone (PTH), total cholesterol, body weight, and mean arterial pressure (MAP) in various treatment groups of wt (▪), ApoE−/− ( ), and Ahsg−/−/ApoE−/− (□) mice. The columns represent means ± SD. *P < 0.05 versus group 1; ^aP < 0.05 between low phosphorus (LP) and matched HP group (1 versus 2; 4 versus 5; 7 versus 8); ^bP < 0.05 between CKD and matched control group (1 versus 3; 4 versus 6; 7 versus 9). Numbers of mice reaching the end of the study as well as total animal numbers (in parentheses) of every group are given.

Figure 2.

Atherosclerotic lesion histomorphometry at different vascular sites in wt (▪), ApoE−/− ( ), and Ahsg−/−/ApoE−/− (□) mice and representative photomicrographs of en face–mounted Oil Red O–stained aortas or stained tissue sections. (A) Atherosclerotic lesion size estimated from Oil Red O–stained whole-mount aortas as depicted on the right for experimental groups 3, 6, and 9 (D, E, and F). (B) Intima thickness (int) determined from serial sections cut through the aortic root displaying the aortic sinus and the aortic valve leaflets (lft) after Masson’s trichrome staining as depicted on the right for experimental groups 3, 6, and 9 (G, H, and I). (C) Incidence of coronary sclerosis. Representative sections of the experimental groups 3, 6, and 9 (J, K, and L) after Masson’s trichrome staining depict proximal coronary arteries (circles). All values in groups 4 through 9 differed significantly from the matching values in groups 1 through 3 (P < 0.05). Numbers of mice reaching the end of the study as well as total animal numbers (in parentheses) of every group are given.

Figure 3.

(A through C) Vascular calcification histomorphometry of von Kossa–stained tissue sections of (A) aorta calcification, (B) coronary artery calcification, and (C) ulcerating plaque incidence in the various treatment groups of wt (▪), ApoE−/− ( ), and Ahsg−/−/ApoE−/− (□) mice. (D through I) Representative photomicrographs are depicted on the right for the treatment groups 3 (D and G), 6 (E and H), and 9 (F and I). (J) A type 5b calcified plaque stained with von Kossa. The tissue was not decalcified before sectioning and thus partially ruptured. (K) A similarly large calcified atherosclerotic lesion stained strongly positive with hematoxylin, suggesting proteinaceous calcified debris, and also AnnexinV (brown stain), indicating apoptosis close to the calcified border region of the plaque. (L) Apoptosis in the aorta wall was further analyzed by fluorescence terminal deoxynucleotidyl transferase–mediated digoxigenin-deoxyuridine nick-end labeling staining. Note that the elastic layer of the tunica media (white arrows) is autofluorescent. White broken lines encircle strongly positive nuclei of apoptotic cells in an extended atherosclerotic plaque (group 9). The columns in A through C represent means ± SD. *P < 0.05 versus group 1; ^aP < 0.05 between LP and matched HP group (1 versus 2; 4 versus 5; 7 versus 8); ^bP < 0.05 between CKD and matched control group (1 versus 3; 4 versus 6; 7 versus 9); ^cP < 0.05 between Ahsg−/−/ApoE−/− and matched ApoE−/− group (4 versus 7; 5 versus 8; 6 versus 9). Numbers of mice reaching the end of the study as well as total animal numbers (in parentheses) of every group are given.

Figure 4.

(A and B) Phenotypic characterization of calcifying aorta tissue. (C) Apoptosis was assessed histochemically and expressed as percentage of vascular wall area. (D through F) Differential expression of OPN (marker for reactive macrophage activity; D), MGP (tissue restricted inhibitor of media calcification; E), and SM22-α (vascular smooth muscle cell marker; F). mRNA was analyzed by quantitative reverse transcriptase–PCR as detailed in the Concise Methods section. Shown is the fold change compared with untreated wt mice. The columns represent means ± SD. *P < 0.05 versus group 1; ^aP < 0.05 between LP and matched HP group (1 versus 2; 4 versus 5; 7 versus 8); ^bP < 0.05 between CKD and matched control group (1 versus 3; 4 versus 6; 7 versus 9); ^cP < 0.05 between Ahsg−/−/ApoE−/− and matched ApoE−/− group (4 versus 7; 5 versus 8; 6 versus 9). Histologic sections and mRNA expression levels were scored in 10 mice for each treatment group.

Figure 5.

Soft tissue calcification quantified by calcium analysis of tissue extracts and illustrated by von Kossa–stained tissue sections in various treatment groups of wt (▪), ApoE−/− ( ), and Ahsg−/−/ApoE−/− (□) mice. (A and B) Myocardial calcification incidence and extent, respectively. (C) Aortic valve calcification. (D) Kidney calcification. (E through M) Representative photomicrographs are depicted on the right for the treatment groups 3 (E, H, and K), 6 (F, I, and L) and 9 (G, J, and M). The columns represent means ± SD. *P < 0.05 versus group 1; ^aP < 0.05 between LP and matched HP group (1 versus 2; 4 versus 5; 7 versus 8); ^bP < 0.05 between CKD and matched control group (1 versus 3; 4 versus 6; 7 versus 9); ^cP < 0.05 between Ahsg−/−/ApoE−/− and matched ApoE−/− group (4 versus 7; 5 versus 8; 6 versus 9). Numbers of mice reaching the end of the study as well as total animal numbers (in parentheses) of every group are given.