Phenotype Interaction of apobec-1 and CETP, LDLR, and ApoE Gene Expression in Mice
Role of ApoB mRNA Editing in Lipoprotein Phenotype Expression
Abstract—Apolipoprotein (apo) B mRNA editing determines the amount of apoB-100 and apoB-48 produced. Surprisingly, apobec-1 knockout mice, which do not edit apoB, have an essentially normal lipoprotein phenotype. By selected cross-breeding of mice of different genotypes, we show in this report that inactivation of editing produces profound phenotypic effects in cholesteryl ester transfer protein (CETP) transgenic mice and in apoE and low density lipoprotein receptor (LDLR) knockout mice. Compared with mice with an apobec-1+/+ background, CETP expression in apobec-1−/− mice caused a doubling of the plasma apoB-100 concentration (from 3.5±0.6 to 8.8±1.9 mg/dL, P<.01) and a much greater shift of plasma cholesterol from HDL to IDL/LDL as assayed by fast protein liquid chromatography analysis; the ratio of non-HDL to HDL cholesterol was 0.47, 0.46, 0.76, and 1.43 in apobec-1+/+/CETP−/−, apobec-1−/−/CETP−/−, apobec-1+/+/CETP+/−, and apobec-1−/−/CETP+/−animals, respectively. Feeding of a Western-type diet further exaggerated the shift in this ratio. In LDLR−/− mice, inactivation of apobec-1 caused an ≈200% rise in plasma apoB-100 concentration, an ≈60% increase in apoE concentration, and a 70% increase in total plasma cholesterol, which resulted exclusively from an increase in non-HDL cholesterol. The exaggerated hypercholesterolemia involving the VLDL+LDL fractions was further enhanced by a Western-type diet. In contrast, in apoE−/− mice, inactivation of apobec-1 caused a massive increase (from <0.5 to 55.5±16.4 mg/dL) in plasma apoB-100 concentration but an ≈55% reduction in hypercholesterolemia due to partial amelioration of the marked VLDL+IDL elevation. However, the difference in lipid profiles between apobec-1+/+/apoE−/−andapobec-1−/−/apoE−/− mice was abolished in a time-dependent manner as further increases in total plasma cholesterol were induced by a Western-type diet. Whereas apobec-1 inactivation in wild-type mice produced little or no change in lipoprotein phenotype, giving rise to speculation that apoB mRNA editing does not have significant effect on lipoprotein dynamics, we show herein that there is important gene-gene interaction between apobec-1 and the CETP, LDLR, and apoE loci, which is subject to further substantial modulation by environmental factors such as a Western-type diet in mice.
- Received October 17, 1997.
- Accepted December 5, 1997.
Apolipoprotein B exists in two forms, apoB-100 and apoB-48 (for a review, see Reference 11 ). ApoB-100 is synthesized in the liver and is required for the production of VLDL, IDL, and LDL. ApoB-48 is produced in the small intestine in humans and is required for chylomicron production. A major difference between the two proteins is the presence of the LDLR-binding domain in the C-terminal half of apoB-100, which is absent in apoB-48. Thus, apoB-100 is a physiological ligand for the LDLR, whereas apoB-48 is incompetent in this respect.
The mouse is a popular model for lipoprotein metabolism and atherosclerosis.2 3 4 However, the value of the mouse as a human lipoprotein disease model is limited by some important differences in lipoprotein metabolism between the two species. One major difference is the preferential production of apoB-48 by the mouse liver. In this species, apoB-48 accounts for some 60% to 70% of hepatic apoB. Coupled with the essentially exclusive production of apoB-48 in the small intestine, apoB-48 accounts for a major proportion (10% to 50%, depending on the strain) of circulating plasma apoB. In comparison, apoB-48 is often undetectable in human postabsorptive plasma because of the exclusive production of apoB-100 in the liver.5
Another important difference between mouse and human lipoprotein metabolism is the presence of CETP in human but not mouse plasma. In humans, CETP redistributes CE and TGs between HDL and VLDL.6 7 The absence of CETP in mice partially accounts for the low VLDL, IDL, and LDL and the high HDL levels in these animals. Another reason for the different lipoprotein distribution is the very low plasma apoB-100 content in mice, which is caused by the (1) diversion of much of the hepatic apoB synthetic capacity to apoB-48, as discussed above, and (2) very low expression of the apoB gene. In fact, in mice, instead of apoB-100, apoE plays a dominant role in plasma lipoprotein distribution and metabolism. ApoE is the major physiological ligand for the LDLR as well as the remnant receptor. ApoE deficiency produced by gene targeting in mice is associated with a much more severe hyperlipidemic phenotype than is observed with apoE deficiency in humans.8 9 10
Thus, the apoE and LDLR genes and the absence of the CETP gene in mice account for many of the differences in lipoprotein metabolism between humans and mice. These genes can potentially display coupled interactions with apobec-1, the gene that controls the relative amounts of apoB-100 and apoB-48 production by mediating the hepatic and intestinal editing of apoB-100 mRNA.11 12 13 It is interesting that the genetic annulment of apoB mRNA editing by the inactivation of apobec-1 is associated with the disappearance of plasma apoB-48 and an approximately threefold increase in plasma apoB-100 concentration but only minimal changes in plasma lipoproteins.14 15 16 We reasoned that this relative lack of lipoprotein phenotypic expression is related, at least partly, to the importance of apoE and the absence of CETP expression, which minimize the phenotypic consequences of the absence of editing in apobec-1−/− mice. In this article, we have examined the interaction of the apobec-1 gene and apoE, LDLR, and CETP genes in mice by genetic manipulation and cross-breeding. The plasma lipoproteins of the various genotypes were analyzed when the animals were fed regular laboratory chow and when they were fed a high-fat, high-cholesterol (Western-type) diet. Indeed, compared with apobec-1−/− animals with a wild-type background,14 15 16 noteworthy phenotypic effects were brought out in the apobec-1 knockout mice when their genetic complements of apoE, LDLR, and CETP were altered. Our observations provide insight into the interactive role of the apobec-1, CETP, LDLR, and apoE genes in the regulation of lipoprotein metabolism and the effect of an environmental factor, a Western-type diet, on lipoprotein phenotype expression.
Apobec-1 knockout mice, apobec-1−/−, were created by targeted gene disruption as described.16 ApoE knockout mice, apoE−/−, were a generous gift of Dr N. Maeda10 ; the LDLR−/− mice, a gift from Drs J. Herz, M.S. Brown, and J.L. Goldstein17 ; and the CETP+/− transgenic mice, from Dr A. Tall.18 The double-knockout mice homozygous for both apobec-1 and apoE mutant alleles (apobec-1−/−/apoE−/−), single-knockout, and wild-type mice were all generated by mating male apobec-1−/− to female apoE−/− mice. The double-knockout mice for both apobec-1 and LDLR (apobec-1−/−/LDLR−/−) and apobec-1−/−/CETP+/− mice, were created in a similar manner. In all cases, littermates were used as control groups. The various genotypes were identified by Southern blot or polymerase chain reaction analysis as described previously.10 16 17 18
All mice were initially maintained on a normal chow diet (Teklad 4% mouse/rat diet 7001, Harlan Teklad Premier Laboratory Diets) that contained 4% (wt/vol) animal fat and <0.04% (wt/vol) cholesterol. For the atherogenic diet study, mice were maintained on a humanlike Western-type diet (Teklad Adjusted Calories Western-type diet),9 which contained 21% (wt/vol) fat by weight (0.15% by weight cholesterol and 19.5% by weight casein, without sodium cholate) for 2 or more weeks. All animal experiments were conducted in accordance with guidelines of the Animal Protocol Review Committee of Baylor College of Medicine.
Lipoprotein Fraction by FPLC
All mice were maintained on either the normal chow diet or the Western-type diet for 2 to 16 weeks. They were fasted 4 to 5 hours before blood was removed by retro-orbital puncture under anesthesia. Mice were anesthetized by exposure to methoxylflurane in a chamber; they were killed by cervical dislocation under anesthesia. Total plasma cholesterol and TG were determined enzymatically with commercial kits (Sigma Diagnostics). Lipoprotein fractions were isolated by gel filtration chromatography by using a Beckman system gold high-performance liquid chromatography/FPLC with two Superose 6 columns (Pharmacia Biotech Inc) connected in series.19 For each analysis, 200 μL of mouse plasma was applied to the FPLC column, and 0.5-mL fractions were eluted with 1 mmol/L EDTA, 154 mmol/L NaCl, and 0.02% NaN3 (pH 8.2). Lipid contents in individual fractions were determined by enzymatic assay kits (Sigma Diagnostics).
Quantification of Mouse Plasma ApoB’s, ApoA-I, and ApoE
ApoB’s were analyzed by quantitative scanning densitometry of Coomassie brilliant blue R-250 (Bio-Rad) –stained polyacrylamide gels as described before.16 ApoB-containing lipoproteins were isolated by density gradient ultracentrifugation (Beckman 42.2Ti rotor, 40 000 rpm, 10°C, 8 hours) after adjusting 25-μL plasma samples to a density of 1.063 g/mL with a KBr solution (d=1.35 g/mL) in tubes containing a KBr overlay solution (d=1.063 g/mL). The upper 50-μL lipoprotein fractions were dialyzed against salEN buffer [150 mmol/L NaCl, 1 mmol/L EDTA (pH 7.4), and 0.05% NaN3 (wt/vol)] and assayed enzymatically for cholesterol. Cholesterol recoveries were compared with VLDL plus LDL cholesterol measurements of identical plasma samples treated with PEG.20 For analysis, purified lipoproteins (2 to 10 μg cholesterol) were solubilized in SDS sample buffer at 60°C for 30 minutes. ApoB-48 and B-100 were resolved on a 2% to 20% linear polyacrylamide gel containing dilutions of purified mouse β-VLDL, previously quantified for apoB-48 and apoB-100 protein content by densitometry to purified human LDL apoB-100. Scanned peak areas in the range of 0.125 to 2.00 μg were linear (r=.989, n=12), and chromogenicities were similar for both proteins.
ApoA-I and apoE were determined by radial immunodiffusion essentially as previously described,21 using monospecific rabbit antisera raised against purified apolipoproteins. The sample diluent for the assay of apoE was adjusted to 10% normolipidemic human plasma (vol/vol) and 1% Triton X-100 (vol/vol) in salEN buffer.
Isolation of Lipoproteins by Sequential Flotation Ultracentrifugation and Fractionation by Nondenaturing Gradient Gels
Plasma samples were taken from six mice, and lipoproteins were isolated from 2 mL of plasma by sequentially adjusting the densities with KBr.22 We isolated four different density lipoprotein fractions (d<1.006, d=1.006 to 1.019, d=1.019 to 1.063, and d=1.063 to 1.21 g/mL) by ultracentrifugation at 40 000 rpm in a Beckman Ti-70.1 rotor at 10°C for 20, 24, 36, and 48 hours, respectively. A 20-μL aliquot from each fraction was assayed for total cholesterol and TG. The apolipoprotein profiles of the lipoprotein fractions were analyzed by SDS–polyacrylamide gel electrophoresis (4% to 20%) and Coomassie brilliant blue R-250 staining. Total protein content of each fraction was measured using a Bio-Rad protein assay. Relative particle sizes of VLDL (d<1.006 g/mL), LDL (d=1.006 to 1.063 g/mL), and HDL (d=1.063 to 1.21 g/mL) were determined by nondenaturing gel electrophoresis performed on 3% to 27% gradient polyacrylamide gels.23
Inactivation of the apobec-1 gene gave rise to only minimal phenotypic changes in plasma lipoproteins in mice with a wild-type genetic background.14 15 16 We examined the interactions of the CETP, LDLR, and apoE genes in apobec-1+/+ and apobec-1−/− mice. Contrary to observations in mice with a wild-type genetic background, we found that apobec-1 inactivation produces major phenotypic effects in CETP transgenic mice and in apoE and LDLR knockout mice.
Effect of apobec-1 Gene Inactivation in CETP+/− Transgenic Mice
Genetic crosses between apobec-1 knockout and human CETP transgenic mice18 were healthy and viable. They displayed little difference in their plasma lipids (Table 1⇓). All four genotypes, wild-type, apobec-1−/−, CETP+/−, and apobec-1−/−/CETP+/−, had similar basal plasma cholesterol levels. The apobec-1−/−/CETP+/− animals had a minimal elevation in plasma TG compared with wild-type littermates.
Plasma apolipoproteins were measured in the different genotypes (Table 1⇑). The most notable changes were the absence of apoB-48 and the increase in apoB-100 in mice with inactivated apobec-1. There was only a minor variation in apoA-I and no change in apoE concentrations.
The distribution of the plasma lipoproteins as analyzed by FPLC fractionation was greatly altered (Fig 1⇓). As reported previously,18 the presence of the CETP transgene resulted in a marked lowering of the HDL concentration, concomitant with an increase in the VLDL and IDL/LDL fractions. Although the absence of functional apobec-1 in animals with a wild-type background caused little change in any of the lipoprotein fractions,14 15 16 the absence of a functional apobec-1 in CETP+/− animals led to a further (26%) reduction in HDL cholesterol and a reciprocal (38%) increase in the IDL/LDL fractions, changes that were highly significant in both lipoprotein fractions (Fig 1⇓). By nondenaturing gradient gel electrophoresis, we did not detect any significant difference in size of the plasma lipoproteins (data not shown).
The effect of a 21% fat, 0.15% cholesterol (Western-type) diet on CETP+/− mice in the presence and absence of apobec-1 inactivation was examined. After 2 weeks of this diet, there was a doubling of plasma cholesterol in mice of all genotypes (Table 2⇓). When the special diet treatment was continued for an additional 2 weeks, CETP+/− transgenic mice with an inactivated apobec-1 gene developed significantly higher total plasma cholesterol levels than did littermate CETP+/− mice that carried the wild-type apobec-1 gene; in these CETP+/− transgenic animals, the diet-induced increases in cholesterol were 2.7-fold and 1.8-fold for animals with an apobec-1−/− and apobec-1+/+ background, respectively. We extended the feeding of the Western-type diet another 12 weeks to examine whether this trend was consistent. At 16 weeks the total cholesterol level in both groups of animals showed a gradual decline, suggesting some adaptation to the Western-type diet. However, the cholesterol level remained 46% higher in the apobec-1−/−/CETP+/− mice than in the apobec-1+/+/CETP+/− animals. There was no significant change in plasma TG throughout the dietary treatment.
The effect of 4 weeks of the Western-type diet on the lipoprotein profile was analyzed by FPLC. This dietary manipulation raised the plasma HDL level in all groups. As a result, the difference in HDL between apobec-1−/−/CETP+/− and apobec-1+/+/CETP+/− animals was no longer evident. However, the IDL/LDL fraction was also increased in both groups of animals, with the apobec-1−/−/CETP+/− animals persistently displaying an approximately twofold higher IDL/LDL cholesterol than the apobec-1+/+/CETP+/− animals (FPLC data on these early effects are not shown).
VLDL after 2 to 4 weeks of the Western-type diet was significantly higher in the apobec-1−/−/CETP+/− mice. An additional 6 weeks of Western-type diet treatment caused the IDL/LDL fraction to decline in the apobec-1+/+/CETP+/− animals; in apobec-1−/−/CETP+/− mice, however, it stayed elevated, further exaggerating the difference in IDL/LDL between these two groups of animals (Fig 2⇓). It is evident that the higher total plasma cholesterol of apobec-1−/−/CETP+/− mice was accounted for entirely by the higher IDL/LDL cholesterol in these animals compared with apobec-1+/+/CETP+/− animals.
Effect of apobec-1 Gene Inactivation in LDLR−/− Mice
We also examined the effect of apobec-1 knockout in another hyperlipidemic model, the LDLR−/− mouse.17 Again, the apobec-1−/−/LDLR−/− animals were viable, fertile, and grossly normal. As shown in Table 3⇓, when the animals were maintained on regular chow, the absence of apoB mRNA editing caused these mildly hypercholesterolemic mice to develop even more severe hypercholesterolemia. The plasma cholesterol level in the double-knockout mice was ≈300% to 400% of that of wild-type mice and 70% higher than that in LDLR single-knockout animals. Interestingly, the double-knockout animals also had a doubling of plasma TG compared with LDLR−/− animals, which had normal basal plasma TG concentrations.
Inactivation of the LDLR gene per se increased plasma apoB-100 by fivefold to sixfold and apoB-48 by about ninefold compared with wild-type animals (Table 3⇑). The absence of apoB mRNA editing in LDLR−/− mice further increased plasma apoB-100 about threefold. The plasma apoE level was much increased in the two groups of LDLR−/− mice. Inactivation of apobec-1 in the LDLR−/− animals further increased the plasma apoE concentration an additional 62%.
Fractionation of the plasma lipoproteins by FPLC (Fig 3⇓) revealed that the increase in plasma cholesterol in LDLR−/− animals was accounted for exclusively by an elevation in IDL/LDL cholesterol. Inactivation of apobec-1 exaggerated the cholesterol changes, mainly in the same lipoprotein species. These changes in cholesterol were paralleled by qualitatively similar changes in the TG contents of the various lipoprotein fractions (Fig 3⇓, right-hand panel). There was no difference in the apparent size of the various lipoprotein fractions as analyzed by nondenaturing gradient gel electrophoresis (data not shown).
A high-fat, high-cholesterol diet was shown to markedly increase plasma cholesterol in LDLR−/− mice.17 We confirmed this observation by feeding the experimental animals a 21% fat, 0.15% cholesterol Western-type diet (Table 4⇓). The total plasma cholesterol concentration increased about sevenfold in both the LDLR−/− and apobec-1−/−/LDLR−/− animals. Thus, irrespective of whether the animals were fed low-fat, low-cholesterol chow or the Western-type diet, the apobec-1−/−/LDLR−/− mice maintained a more severe hypercholesterolemia than did their apobec-1+/+/LDLR−/− littermates. This diet-induced hypercholesterolemia peaked early, at about 2 weeks of diet feeding, in both genotypes. However, the difference in plasma cholesterol concentration increased with the time of diet treatment, so that at 14 weeks, it was 65% higher in the apobec-1+/+/LDLR−/− mice. The diet-induced hypertriglyceridemia (an ≈300% rise), in contrast, was more severe in the LDLR−/− animals with intact apobec-1 than in those with an inactivated apobec-1 (which had only a ≈34% increase) at 2 weeks of diet feeding. The difference in plasma TG between the two groups of animals was maintained in the first 4 weeks of Western-type diet feeding. Interestingly, the TG concentration of the apobec-1+/+/LDLR−/− animals decreased at 14 weeks to a level similar to that in apobec-1−/−/LDLR−/− mice.
We studied the lipoprotein profiles in apobec-1−/−/LDLR−/− and apobec-1+/+/LDLR−/− mice after Western-type diet feeding. As indicated earlier in this section, the major plasma lipoprotein affected by LDLR gene inactivation in either the presence or absence of functional apobec-1 was IDL/LDL. After Western-type diet treatment, the IDL/LDL fractions were increased further, being much more so in LDLR−/− mice with coexisting apobec-1 gene inactivation (Fig 4⇓). The VLDL and IDL/LDL TG contents of these particles were also increased, significantly more so in apobec-1−/−/LDLR−/− than in apobec-1+/+/LDLR−/− animals. The relative amounts of HDL cholesterol stayed low in both groups of animals.
apobec-1 Gene Inactivation Partially Ameliorates the Hypercholesterolemia of ApoE−/− Mice
We then cross-bred apobec-1−/− mice16 with apoE−/− mice10 and obtained animals that were apobec-1−/−/apoE−/−, apobec-1−/−/apoE+/+, apobec-1+/+/apoE−/−, and apobec-1+/+/apoE+/+. All animals were viable and grew normally with no obvious ill effects. They were initially maintained on regular laboratory chow, and their plasma lipids, apoB, and apoE levels were determined. The laboratory values among littermates of the various genotypes are presented in Table 5⇓.
As reported previously, basal plasma lipids were similar in wild-type and apobec-1−/− animals,14 15 16 and apoE−/− animals had a marked (approximately fivefold) elevation in total plasma cholesterol but no change in TG.9 10 The absence of a functional apobec-1 gene in apoE−/− animals caused a significant (≈55%) reduction in total plasma cholesterol but an insignificant change in TG.
To examine the potential role of apoB and apoE proteins in the lipoprotein phenotypes, we measured the plasma concentrations of these apolipoproteins as described in “Methods.” As shown in Table 5⇑, plasma apoB-100 increased about twofold in apobec-1−/− animals compared with the wild type. It increased further to about sixfold the level in wild-type animals in apobec-1−/−/apoE−/− animals. Interestingly, in apoE−/− animals on a regular chow diet, the basal plasma apoB-100 was barely detectable; these animals had almost exclusively apoB-48 in their plasma at a concentration that was about 200-fold greater than in wild-type controls. As expected, animals with inactivation of apobec-1 had no detectable apoB-48 in the circulation. Plasma apoE levels were not different in apobec-1−/− and wild-type animals; they were undetectable in apoE−/− animals, whether apobec-1 was functional or not.
To analyze the lipoprotein species affected by apobec-1 inactivation in apoE−/− animals, we fractionated the plasma lipoproteins by FPLC. As shown in Fig 5⇓, the increase in plasma cholesterol in apoE−/− mice was caused by an increase in VLDL and IDL/LDL, with essentially no change in HDL. In apoE−/− animals the absence of a functional apobec-1 gene resulted in substantial amelioration of the VLDL and IDL/LDL elevation without affecting HDL. Thus, the diversion of apoB-48 to apoB-100 production in apoE−/− mice is associated with substantial improvement in the hyperlipidemia. The apparent size of the various lipoproteins was analyzed by nondenaturing gradient gels. Of the various lipoprotein fractions, only HDL showed a difference in size, in that it was larger in apobec-1−/−/apoE−/− than in apobec-1+/+/apoE−/− animals (10.8±0.22 nm, n=5 versus 10.1±0.21 nm, n=5, P<.01).
We studied the effect of a 21% fat, 0.15% cholesterol (Western-type) diet9 on plasma lipids and lipoproteins in these groups of mice. Within 2 to 3 weeks of Western-type diet feeding, the wild-type and apobec-1−/− mice with a normal apoE+/+ background developed moderate hypercholesterolemia (Table 6⇓). On FPLC analysis, the plasma lipoproteins in chow-fed apobec-1−/− animals were similar to those of apobec-1+/+ littermates except for a mildly reduced HDL (data not shown, previously published in Reference 1616 ; also see Reference 1515 ). On the Western-type diet treatment, apobec-1 knockout mice developed a significantly elevated VLDL and IDL/LDL, approximately doubling the levels in wild-type controls (data not shown). The HDL in apobec-1−/− animals was increased twofold over chow-fed animals. The difference in HDL between apobec-1−/− and apobec-1+/+ animals disappeared with this dietary manipulation.
In animals with an apoE−/− background, the Western-type diet produced a massively increased plasma cholesterol level (Table 6⇑). Interestingly, the difference between apobec-1+/+ and apobec-1−/− mice with an apoE−/− background was no longer evident. In other words, animals with the apobec-1 knockout developed an enhanced lipoprotein response, so that an equally high plasma cholesterol level was achieved in apobec-1−/−/apoE−/− and apobec-1+/+/apoE−/− mice. The apobec-1−/−/apoE−/− animals also developed more severe hypertriglyceridemia than did the apobec-1+/+/apoE−/− animals, although the difference between the two groups was not significant. After 10 weeks on the Western-type diet, we analyzed the lipoproteins by FPLC (Fig 6⇓). We observed that the massive hypercholesterolemic response involved mainly the VLDL and IDL/LDL fractions with a concomitant lowering of the HDL fraction. Again, the marked difference between apobec-/-/apoE−/− and apoE−/− animals while they were on regular chow was virtually abolished. The hypertriglyceridemia, which was relatively mild, involved the VLDL peak and was higher in apobec-1−/−/apoE−/− than apoE−/− animals.
Apobec-1 is the catalytic subunit of a multicomponent mammalian apoB mRNA editing complex (reviewed in Reference 1313 ). When the enzyme was inactivated by gene targeting in mice, there was a complete absence of editing, and the apobec-1 knockout mice produced only apoB-100 and no apoB-48. Surprisingly, the apobec-1−/− mice displayed essentially no abnormalities in their lipoprotein profile.14 15 16 In this study, we explored the interactions between the CETP, LDLR, and apoE loci and apoB mRNA editing. The effect of a Western-type diet, an environmental factor, on the gene-gene interaction was also examined. The phenotypic changes in the various genetic crosses provide information on the role of the various gene products in lipoprotein homeostasis.
The first genetic manipulation that we performed on apobec-1−/− mice was to provide them with a human CETP gene driven by its native promoter.18 The mouse has no CETP. The absence of functional CETP in mouse plasma partly accounts for the very high HDL and relatively low VLDL, IDL, and LDL contents in this species compared with other mammalian species with CETP. Introduction of the CETP gene to mice causes a marked lowering of plasma HDL cholesterol with a reciprocal increase in IDL/LDL cholesterol (Fig 1⇑ and Reference 1818 ). In this study, we showed that the effect of this CETP is greatly enhanced in animals with an inactivated apobec-1 gene (Fig 1⇑). In these animals, the ratio of (IDL+LDL) to HDL is 1.43 compared with a ratio of 0.76 in apobec-1+/+/CETP+/− and 0.47 in apobec-1+/+/CETP−/− (wild-type) animals (Table 1⇑). We had previously shown that apobec-1 gene inactivation per se leads to a mild reduction in plasma HDL.16 The observations on the CETP transgenic animals suggest that there may be a synergistic effect between apoB-100 and CETP expression on the reciprocal changes in plasma IDL/LDL and HDL concentrations.
The LDLR is a convergent pathway for the metabolism of circulating plasma apoB-100 and apoE. Genetic inactivation of the LDLR locus interferes with the catabolism of both apoB-100–and apoE-containing lipoproteins. The absence of apoB mRNA editing in apobec-1−/−/LDLR−/− mice causes plasma apoB-100 to accumulate to a even higher level, because the LDLR is the only known physiological receptor for apoB-100. As a consequence, plasma concentrations of all apoB-100–containing lipoproteins, VLDL, IDL, and LDL, increased further in apobec-1−/−/LDLR−/− animals, whether they were on laboratory chow that is low in fat and cholesterol content or on a Western-type diet containing 21% fat and 0.15% cholesterol. These lipoprotein changes are in sharp contrast to the absence of any significant phenotypic change in apobec-1−/− mice with normal LDLR function.14 15 16
As a physiological ligand for the LDLR and the remnant receptor, apoE is a major modulator of lipoprotein metabolism in mice. As reported previously,9 10 apoE−/− mice displayed severe spontaneous hypercholesterolemia when they were fed regular laboratory chow (Table 5⇑). By FPLC analysis, the hypercholesterolemia in apoE−/− animals was accounted for entirely by elevated VLDL and IDL/LDL levels (Fig 5⇑). Interestingly, inactivation of apobec-1 in apoE−/− animals resulted in amelioration of their hypercholesterolemia (Table 6⇑). This blunting of the hypercholesterolemia in apobec-1−/−/apoE−/− animals was the result of a reduction in VLDL and IDL/LDL cholesterol with no change in HDL cholesterol in these animals. Partial reversal of the hypercholesterolemia involving the apoB-containing lipoproteins is consistent with an important role for apoB-100 as a ligand for the LDLR in the absence of apoE. ApoB-48 is the predominant apoB species in apoE−/− mice, in which apoB-100 is barely detectable. Inactivation of apoB mRNA editing led to a marked increase in plasma apoB-100. In the absence of apoE, apoB-100 serves as the sole physiological ligand for the LDLR, enabling the animal to remove the apoB-100–containing LDL from the circulation and to lower the concentration of its precursors, VLDL and IDL, at the same time. In the presence of normal apoE expression, despite a threefold increase in plasma apoB-100 in apobec-1−/− mice with a wild-type genetic background, the ligand function of apoB-100 for the LDLR was overshadowed by that of apoE.14 15 16 It was only in the absence of apoE that apoB-100 emerged as an important ligand mediating the removal of LDL in apobec-1−/− mice.
When the mice were fed a Western-type diet, the difference in plasma cholesterol contents between the apobec-1−/−/apoE−/− and apobec-1+/+/apoE−/− mice disappeared as the hypercholesterolemia became even more severe. As shown in Fig 6⇑, the massive increase in plasma cholesterol was caused by a marked elevation in VLDL and IDL/LDL cholesterol in both groups of animals. Interestingly, in response to this diet, the apoE−/− animals developed a hypertriglyceridemia that was milder and more transient than in the apobec-1+/+/apoE−/− animals but more severe and persistent than in their apobec-1−/−/apoE−/− littermates. The basis for the difference between the two groups is not clear.
apobec-1 inactivation in mice with a wild-type genetic background produced only minimal changes in lipoprotein profile,14 15 16 leading to speculation that perhaps apoB mRNA editing is not that important a factor in lipoprotein metabolism in mice. In this study, we examined the gene-gene interaction between apobec-1 and the CETP, LDLR, and apoE loci in detail and showed that, indeed, the abolition of apoB mRNA editing produces substantial lipoprotein phenotypic effects in the various genotypes; furthermore, our study revealed how an environmental factor, a Western-type diet, modulates the effect of gene-gene interaction on lipoprotein phenotypic expression.
Selected Abbreviations and Acronyms
|CE(TP)||=||cholesteryl ester (transfer protein)|
|FPLC||=||fast protein liquid chromatography|
This work was supported by National Institutes of Health grants HL56668, HL16512, and HL51586 (to L.C.). K.K. was supported by a postdoctoral fellowship from the Juvenile Diabetes Foundation International. S.T. and K.K. also received fellowship support from the Children’s Nutrition Research Center in Houston. We thank Dr Michael E. DeBakey for his interest and support. We thank Dr N. Maeda of the University of North Carolina for providing apoE−/− mice; Dr A Tall of Columbia University for providing CETP+/− mice; Drs J. Herz, M.S. Brown, and J.L. Goldstein of the University of Texas Southwestern Medical School for providing LDLR−/− mice; Lan Li for her technical assistance; and Irene A. Harrison and Sylvia A. Ledesma for expert secretarial assistance.
Chan L. Apolipoprotein B, the major protein component of triglyceride-rich and low density lipoproteins. J Biol Chem. 1992;267:25621–25624.
Chien KR. Genes and physiology: molecular physiology in genetically engineered animals. J Clin Invest. 1996;98:S19–S26.
Breslow JL. Transgenic mouse models of lipoprotein metabolism and atherosclerosis. Proc Natl Acad Sci U S A. 1993;90:8314–8318.
Kane JP, Havel RJ. Disorders of the biogenesis and secretion of lipoproteins containing the B apolipoproteins. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The Metabolic and Molecular Bases of Inherited Disease. New York, NY: McGraw-Hill Inc; 1995:1853–1885.
Schaefer EJ, Gregg RE, Ghiselli G, Forte TM, Ordovas JM, Zech LA, Brewer HB Jr. Familial apolipoprotein E deficiency. J Clin Invest. 1986;78:1206–1219.
Zhang SH, Reddick RL, Piedrahita JA, Maeda N. Spontaneous hypercholesterolemia and arterial lesions in mice lacking apolipoprotein E. Science. 1992;258:468–471.
Teng B-B, Burant CF, Davidson NO. Molecular cloning of an apoB mRNA editing protein. Science. 1993;260:1816–1819.
Chan L, Seeburg PH. RNA editing. Sci Med. 1995;2:68–77.
Hirano K-I, Young SG, Farese RV Jr, Ng J, Sande E, Warburton C, Powell-Braxton LM, Davidson NO. Targeted disruption of the mouse apobec-1 gene abolishes apolipoprotein B mRNA editing and eliminates apolipoprotein B48. J Biol Chem. 1996;271:9887–9890.
Morrison JR, Paszty C, Stevens ME, Hughes SD, Forte T, Scott J, Rubin EM. Apolipoprotein B RNA editing enzyme-deficient mice are viable despite alterations in lipoprotein metabolism. Proc Natl Acad Sci U S A. 1996;93:7154–7159.
Nakamuta M, Chang BH-J, Zsigmond E, Kobayashi K, Lei H, Ishida B, Li R, Chan L. Complete phenotypic characterization of apobec-1 knockout mice with a wild-type genetic background and a human apolipoprotein B transgenic background, and restoration of apolipoprotein B mRNA editing by somatic gene transfer of apobec-1. J Biol Chem. 1996;271:25981–25988.
Ishibashi S, Brown MS, Goldstein JL, Gerard RD, Hammer RE, Herz J. Hypercholesterolemia in LDL receptor knockout mice and its reversal by adenovirus-mediated gene delivery. J Clin Invest. 1993;92:883–893.
Jiang XC, Agellon LB, Walsh A, Breslow JL, Tall AR. Dietary cholesterol increases transcription of the human cholesteryl ester transfer protein gene in transgenic mice. J Clin Invest. 1992;90:1290–1295.
Cole T, Kitchens RT, Daugherty A, Schonfeld G. An improved method for separation of triglyceride-rich lipoproteins by FPLC. FPLC BioCommunique. 1988;4:4–6.
Dias VC, Parsons HG, Boyd ND, Keane P. Dual-precipitation method evaluated for determination of a high-density lipoprotein (HDL), HDL2, and HDL3 cholesterol concentrations. Clin Chem. 1988;34:2322–2327.
Ishida BY, Paigen B. Silver-enhanced radial immunodiffusion assay of plasma apolipoproteins. J Lipid Res. 1992;33:1073–1078.
Lindgren FT, Jensen LC, Hatch FT. The isolation and quantitative analysis of serum lipoproteins. In: Nelson GJ, ed. Blood Lipids and Lipoproteins. New York, NY: John Wiley-Interscience; 1972:181–274.
Nichols AV, Krauss RM, Musliner TA. Nondenaturing polyacrylamide gradient gel electrophoresis. In: Segrest JP, Albers JJ, eds. Methods in Enzymology. New York, NY: Academic Press; 1986:417–431.