Plasma carboxyl ester lipase activity modulates apolipoprotein B-containing lipoprotein metabolism in a transgenic mouse model

Abstract

Pancreatic carboxyl ester lipase (CEL) is in the plasma of many mammals, including humans and rats, but not mice. In vitro, CEL hydrolyzes cholesterol esters of apolipoprotein B-containing lipoproteins (apo B-Lp). To study the effect of CEL on metabolism of apo B-Lp and atherosclerosis in vivo, apo E-knockout (EKO) mice, which have high plasma levels of apo B-Lp and are prone to atherosclerosis, were made to secrete CEL into plasma by introducing a transgene containing a liver-specific promoter and rat CEL complementary DNA. Plasma CEL activity in EKO-CEL mice was comparable with that found in rats. Evidence of modification of apo B-Lp by plasma CEL in vivo was an increase in the free cholesterol to cholesterol ester ratio of apo B-Lp from mice on chow or a Western-type diet. In addition, plasma total cholesterol levels were elevated in EKO-CEL mice, with the elevation found exclusively in the apo B-Lp fraction. Associated with the increase in steady-state apo B-Lp levels was an increase in the plasma half-life of very low-density lipoproteins (VLDL) in EKO-CEL mice, measured by the clearance rate of injected VLDL. Interestingly, despite the increase of apo B-Lp, the atherosclerotic lesion did not differ between EKO and EKO-CEL mice on a Western-type diet. In summary, our results demonstrate that plasma CEL modulates apo B-Lp metabolism in vivo, resulting in reduced VLDL clearance and elevated plasma cholesterol levels.

Fig. 1. Construction of rat CEL transgene

A Bluescript plasmid was previously modified to contain β-globin exons 2 and 3 (E2, E3) and the intervening intron (I). The 324 bp hepatic-specific promoter for apolipoprotein A-I (AI promoter) was inserted into the plasmid multiple cloning site and the complete rat CEL cDNA (∼ 2 kb) inserted in the EcoRI site of exon 3 of β-globin. The plasmid was linearized by restriction digestion with NotI and purified for microinjection.

Fig. 2. Plasma lipid concentrations

Plasma samples were collected from 2 months old EKO mice (n=10) and EKO-CEL mice (n=10) on chow (A) or Western diet (B). Plasma total cholesterol (TC), HDL-cholesterol (HDL-C), and TG concentrations were determined enzymatically by using commercial kits. Non-HDL-C was calculated as the difference between TC and HDL-C. EKO-CEL mice had significantly higher plasma TC than EKO mice on either chow or Western diet. The difference was found exclusively in the non-HDL-C fraction. No significant difference between two groups was found in the plasma HDL and TG concentrations. * P < 0.05; ** P < 0.01.

Fig. 3. Plasma lipoprotein cholesterol profiles

Plasma samples were pooled from EKO (n=10) and EKO-CEL mice (n=10), respectively, on chow (A) or Western diet (B). FPLC was used to separate plasma lipoproteins, and the total cholesterol profiles were determined. VLDL-cholesterol from EKO-CEL mice was significantly higher than that from EKO mice either on chow or Western diet. IDL-and LDL-cholesterol also tended to be higher in EKO-CEL mice. No difference between two groups was observed in HDL fraction.

Fig. 4. Free cholesterol (FC) to cholesterol ester (CE) ratios in apoB-Lp

VLDL and LDL fractions of the plasma were collected from the FPLC elution. FC and TC concentrations were determined enzymatically. Concentrations of CE were obtained by subtracting FC from TC. On chow diet (A), the FC/CE ratios were significantly higher in both VLDL and LDL from EKO-CEL mice compared to those from EKO mice. On Western diet (B), the FC/CE ratio in LDL was significantly higher in EKO-CEL mice and in VLDL it tended to be higher (P = 0.09) compared to that in EKO mice. * P < 0.05; ** P < 0.01.

Fig. 5. Clearance of plasma VLDL on chow diet

The turnover rate of VLDL was measured by intravenous injection of a bolus radiolabeled VLDL. Percent recovery of radioactivity at different time points after injection was determined. The rate of disappearance of radiolabeled VLDL in EKO-CEL (n=10) mice was significantly slower than that in EKO (n=9) mice 30 min after injection (P = 0.045 by repeated measures ANOVA). * P < 0.05.

Fig. 6. Atherosclerotic lesion area in the aortic root of mice on Western diet for 10 weeks

No significant difference was observed between EKO (n=10) and EKO-CEL (n=10) mice.