Serum amyloid A augments the atherogenic effects of cholesteryl ester transfer protein

Abstract

Serum amyloid A (SAA) is predictive of CVD in humans and causes atherosclerosis in mice. SAA has many proatherogenic effects in vitro. However, HDL, the major carrier of SAA in the circulation, masks these effects. The remodeling of HDL by cholesteryl ester transfer protein (CETP) liberates SAA restoring its proinflammatory activity. Here, we investigated whether deficiency of SAA suppresses the previously described proatherogenic effect of CETP. ApoE^-/- mice and apoE^-/- mice deficient in the three acute-phase isoforms of SAA (SAA1.1, SAA2.1, and SAA3; "apoE^-/- SAA-TKO") with and without adeno-associated virus-mediated expression of CETP were studied. There was no effect of CETP expression or SAA genotype on plasma lipids or inflammatory markers. Atherosclerotic lesion area in the aortic arch of apoE^-/- mice was 5.9 ± 1.2%; CETP expression significantly increased atherosclerosis in apoE^-/- mice (13.1 ± 2.2%). However, atherosclerotic lesion area in the aortic arch of apoE^-/- SAA-TKO mice (5.1 ± 1.1%) was not significantly increased by CETP expression (6.2 ± 0.9%). The increased atherosclerosis in apoE^-/- mice expressing CETP was associated with markedly increased SAA immunostaining in aortic root sections. Thus, SAA augments the atherogenic effects of CETP, which suggests that inhibiting CETP may be of particular benefit in patients with high SAA.

Keywords: HDL; apolipoprotein; atherosclerosis; cholesteryl ester transfer protein; inflammation; lipid metabolism; serum amyloid A.

Fig. 1

CETP remodeling of SAA-enriched HDL triggers IL-1β secretion in macrophages. A: J774 cells were incubated for 24 h with serum-free media lacking HDL (“control”) or serum-free media containing TKO-HDL (1.18 mg devoid of SAA) or SAA-HDL (1.18 mg HDL protein containing 0.19 mg of SAA) with or without CETP remodeling as indicated. IL-1β levels in the conditioned media were quantified by ELISA. B: TKO-HDL remodeled with CETP (lane 1), SAA-HDL treated with remodeling mixture minus CETP (lane 2), and SAA-HDL incubated with remodeling mixture containing CETP (lane 3) were separated by NDGGE followed by Western blotting to detect SAA. About 5 μl of the reaction mix was loaded per well.

Fig. 2

Expression of CETP or lack of SAA does not change lipoprotein profiles. Mice deficient in apoE (apoE^−/−) and apoE^−/− mice deficient in all three inducible SAA isoforms (apoE^−/− SAATKO) were administered an AAV-expressing CETP or a null AAV and fed a normal chow diet for 14 weeks. Plasma was collected at the end of the study and fractionated by fast performance liquid chromatography; cholesterol content of 0.5 ml fractions was determined enzymatically (Wako Chemicals). apoE^−/− null (n = 5), apoE^−/− SAATKO null (n = 5), apoE^−/− CETP (n = 7), and apoE^−/− SAATKO CETP (n = 7).

Fig. 3

CETP expression increases atherosclerosis only in apoE^−/− mice and not in apoE^−/− SAATKO mice. ApoE^−/− and apoE^−/− SAATKO mice were injected with an AAV-expressing CETP or a null AAV and fed a normal chow diet for 14 weeks. Atherosclerosis in the aortic arch region was quantified by en face analysis. Each point represents an individual mouse with mean ± SEM shown for the group; expressed as the percent of the total aortic arch surface area. Statistical analysis was performed using two-way ANOVA followed by Tukey’s multiple comparisons test. apoE^−/− null (n = 15), apoE^−/− SAATKO null (n = 15), apoE^−/− CETP (n = 15), and apoE^−/− SAATKO CETP (n = 14).

Fig. 4

Expression of CETP in mice decreases plasma SAA concentrations. Mice deficient in apoE (apoE^−/−) and apoE^−/− in all three inducible SAA isoforms (apoE^−/− SAATKO) were administered an AAV-expressing CETP or a null AAV and fed a normal chow diet for 14 weeks. A: CETP activity levels in plasma after 2, 9, and 14 weeks after AAV injection, determined by the detection of CETP-mediated transfer of neutral lipids from a synthetic substrate to a physiological acceptor. apoE^−/− null (n = 11), apoE^−/− SAATKO null (n = 11), apoE^−/− CETP (n = 15), and apoE^−/− SAATKO CETP (n = 16). B: SAA levels in the plasma collected at −2, 2, 9, and 14 weeks after AAV injection, determined by ELISA. Statistical analysis was performed using two-way ANOVA followed by Tukey’s multiple comparisons test: ∗∗P < 0.01, apoE^−/− null (n = 5), apoE^−/− CETP (n = 7).

Fig. 5

SAA is enhanced in the lesions of aortic roots of apoE^−/− mice expressing CETP. ApoE^−/− and apoE^−/− SAATKO mice were injected with an AAV-expressing CETP or a null AAV and fed a normal chow diet for 14 weeks. Aortic root sections from (A) apoE^−/− null mice, (B) apoE^−/− CETP mice, and (C) apoE^−/− TKO CETP mice were processed to detect macrophages (red fluorescence) and SAA (green fluorescence) by immunostaining, nuclei were identified using 4′,6-diamidino-2-phenylindole (blue fluorescence). Sections were photographed under 4× objective magnification, scale bar represents 200 μm and 20× objective magnification, scale bar represents 50 μm. D: SAA immunostaining in aortic roots of apoE^−/− null mice and apoE^−/− CETP mice was quantified using NIS-Elements software. Each point represents the area of SAA expression in the aortic root section of an individual mouse. Statistical analysis was performed using two-tailed t-tests, ∗P < 0.05.