Cholesterol and apolipoprotein B metabolism in Tangier disease.

Tangier disease (TD), caused by mutations in the gene encoding ATP-binding cassette 1 (ABCA1), is a rare genetic disorder in which homozygotes have a marked deficiency of high density lipoproteins (HDL), as well as concentrations of low density lipoproteins (LDL) that are typically 40% of normal. Although it is well known that the reduced levels of HDL in TD are due to hypercatabolism, the mechanism responsible for the low LDL levels has not been defined. Recently, it has been reported that intestinal cholesterol absorption is altered in ABCA1 deficient mice, suggesting that aberrant cholesterol metabolism may contribute to the LDL reductions in TD. In order to explore this possibility, as well as to define the role that ABCA1 plays in the metabolism of apolipoprotein (apoB)-containing lipoproteins, we determined the kinetics of apoB-100 within lipoproteins, and cholesterol absorption, biosynthesis, and turnover, in a compound heterozygote for TD. The levels of HDL cholesterol, LDL cholesterol and LDL apoB-100 in this subject were 7, 27 and 69% of normal, respectively, the latter of which was due to a two-fold increase in LDL catabolism (0.54 vs. 0.26+/-0.07 poolsday(-1)) relative to controls (n=11). NMR analysis of plasma lipoproteins revealed that 91% of the LDL cholesterol in the TD subject was contained within small, dense LDL, as compared with only 20% for controls (n=70). Cholesterol absorption was 97% of the value for controls (n=15) in the TD subject, at 45%, with cholesterol synthesis and turnover increased modestly by 17 and 25%, respectively. Our data are consistent with the concept that the reductions of LDL observed in TD are due to enhanced catabolism, secondary to changes in LDL composition and size, with neither cholesterol absorption nor metabolism significantly influenced by mutations in ABCA1.

A study of the metabolism of apolipoprotein B100 in relation to insulin resistance in African American males.

The purpose of this study was to determine the relationship between insulin resistance and apoB100 metabolism in African American males. Fifteen subjects, 33 +/- 7.6 years old, were divided into two groups, insulin-resistant (IR) or insulin-sensitive (IS), based on the sum of the plasma insulin concentrations during an oral glucose tolerance test. The IR group (n = 8) differed significantly from the IS group (n = 7) with respect to body mass index (BMI) (30.1 vs 23.1 kg/m2; P = 0.0003), fasting triglycerides, (118 vs 54 mg/dl, P = 0. 013), and total plasma apolipoprotein B100 (80 vs 59 mg/dl, P = 0.014). Significantly elevated apoB100 levels in the IR group were seen in very low density lipoprotein (VLDL) (5.1 vs 3.4 mg/dl, P = 0.045) and intermediate density lipoprotein (IDL) (18 vs 12 mg/dl, P = 0.017) but not in low density lipoprotein (LDL) (57 vs 46 mg/dl, P = 0.19). Total cholesterol, high density lipoprotein cholesterol (HDL-C), low density lipoprotein cholesterol (LDL-C), apolipoprotein A-I, and blood pressure were not significantly different between the two groups. There was a high correlation between the sum of insulins during the oral glucose tolerance test and the BMI (rho = 0.88, P = 0.0001). In five IR and five IS subjects, apoB100 kinetics were determined in the fasting state using a bolus dose of deuteroleucine and multicompartmental modeling. IR subjects had significantly lower fractional catabolic rates (FCR) in the larger VLDL1 (-70%), the smaller VLDL2 (-71%), and the IDL (-53%) fractions. No significant differences in production rates were observed for any lipoprotein class. There was a significant correlation between the sum of insulins and the FCR of the apoB100 of VLDL1 (rho = -0.65, P = 0.05) and of IDL (rho = -0.85, P = 0.004). The correlation coefficient of the sum of insulins and the FCR of VLDL2 was -0.61 with P = 0.067. We conclude that in this population of African American males, IR is correlated with a decreased FCR of apoB100 in VLDL and IDL and elevated plasma levels of apoB and triglycerides (TG). These changes might be explained by decreased clearance of the TG-rich lipoproteins. We postulate that this may reflect decreased lipoprotein and/or hepatic lipase activity related to insulin resistance and its association with obesity.

Effect of experimental nephrosis on hepatic lipoprotein secretion and urinary lipoprotein excretion in rats expressing the human apolipoprotein A-I gene.

When human apolipoprotein A-I was expressed in transgenic rats, induction of the nephrotic syndrome resulted in plasma A-I levels exceeding 10 mg/ml. Plasma lipids were no higher than in non-transgenic nephrotic rats. To explain this, the livers from four groups of rats were perfused: wild-type controls (WC), high expressor human apoA-I transgenic controls (TrGC), wild-type nephrotics (WN), and high expressor transgenic nephrotics (TrGN). Compared to the WC group, TrGC rats secreted the same amount of d < 1.063 g/ml lipoproteins but 50% more high density lipoprotein (HDL), with a 5-fold increase in total apoA-I output due to human apoA-I. Compared to the WC group, nephrosis in the WN rats caused a 2-fold increase in both d < 1.063 g/ml lipoproteins and HDL secretion with a 4.6-fold increase in rat apoA-I output. Compared to the TrGC group, nephrosis in the TrGN rats did not increase d < 1.063 g/ml lipoprotein secretion, but caused a 50% increase in HDL secretion and a 6-fold increase in human apoA-I output. The hepatic levels of mRNA for apoB and for HMG-CoA reductase, as well as the degree of apoB mRNA editing, were unchanged. Examination of the perfusate HDL by electron microscopy revealed spherical particles averaging 30 nm in diameter in the WC and WN rats and 17 and 20 nm in the TrGC and TrGN rats. Urinary HDL particles from the TrGN rats did not contain rat apoA-I and averaged 8.2 nm versus 11 nm in the WN rats. We conclude that the size of the nascent HDL, and subsequently of the mature HDL, is determined by the primary structure of apoA-I. In the TrGN rats, the heterogeneous mature HDL has a population of smaller human HDL which is more readily lost in the urine, accounting for the failure of plasma HDL levels to rise above those in TrGC rats. The fact that plasma triglyceride levels in TrGN rats were also not increased may relate to the failure of hepatic apoB secretion to increase, which in turn may have been due to saturation of the protein synthetic capacity by human apoA-I production.

Overexpression of human apolipoprotein A-I in transgenic rats and the hyperlipoproteinemia associated with experimental nephrosis.

Hyperlipoproteinemia contributes both to kidney disease progression and the development of atherosclerosis. Elevated high density lipoprotein cholesterol and apolipoprotein A-I (apoA-I) serum levels are independent factors protective against the atherosclerotic process. We examined the effects in a transgenic rat model of human apoA-I expression on the hyperlipoproteinemia and edema after puromycin aminonucleoside-induced nephrosis in three groups of animals: low line (TgR[hAI]low, human plasma apoA-I = 16.0 mg/dl); high line (TgR[hAI]high, 284 mg/dl); and non-transgenic litter mates (TgR[hAI]non). Nephrosis increased total plasma apoA-I levels 2-fold in TgR[hAI]non rats (75 vs. 162 mg/dl) and 4-fold in the TgR[hAI]low (97 vs. 458 mg/dl) and TgR[hAI]high rats (356 vs. 1,346 mg/dl). In both transgenic lines, this increase was due mainly to elevations of serum human apoA-I. The hepatic steady-state levels of rat apoA-I mRNA increased 5- to 7-fold in all three groups, while human apoA-I mRNA levels increased 21- and 65-fold in the low and high expressing groups, respectively, indicating a different degree of responsiveness of the rat and human genes. While nephrotic TgR[hAI]non and TgR[hAI]low rats showed severe hyperlipoproteinemia and edema, much lower levels of edema and of serum triglycerides, phospholipids, and cholesterol were seen in the TgR[hAI]high group. Urinary excretion of apoA-I, phospholipids, and cholesterol was significantly increased in the TgR[hAI]high group, indicating this as one possible mechanism for the relatively lower serum levels of these lipids. We conclude that the human apoA-I gene is responsive to nephrosis and that human apoA-I-transgenic rats with this syndrome provide an animal model for the study of human high density lipoprotein and apoA-I metabolism.

Kinetic evidence for both a fast and a slow secretory pathway for apolipoprotein A-I in humans.

The kinetics of apolipoproteins A-I and A-II were examined in human subjects using leucine tracers administered intravenously. High density lipoproteins were separated and apoA-I and A-II were isolated. The specific activity or enrichment data for these apolipoprotein were analyzed by mathematical compartmental modeling. In 11 of 14 subjects studied with a bolus-injected [3H]leucine tracer, in 3 subjects studied similarly with [3H]leucine, and in one subject studied by primed dose, constant infusion of [3H]leucine, a rapidly turning-over apoA-I fraction was resolved. A similar component was observed in 7 of 10 studies of apoA-II. The apoA-I data were analyzed using a compartmental model (Zech, L.A. et al. 1983. J. Lipid Res. 24: 60-71) modified to incorporate plasma leucine as a precursor for apoprotein synthesis. The data permitted resolution of two apoA-I pools, one, C(2), turned-over with a residence time of less than 1 day, the other, C(1), a slowly turning-over pool, appeared in plasma after a delay of less than half a day. C(1) comprised the predominant mass of apoA-I and was also the primary determinant of the residence time of apoA-I. Although the mass of the fast pool, C(2), was considerably less than that of C(1), because of its rapid turnover, the quantities of apoA-I transported through this fast pathway were 2- to 4-fold greater. These kinetic studies indicate that apoA-I is secreted into both fast and slowly turning-over plasma pools. The latter is predominantly measured with radioiodinated apoA-I tracers. The data can be analyzed by postulating either separate input pathways to each of the pools or by assuming the fast pool is the precursor to the slow pool. Thus, apoA-I could be initially secreted as a family of particles that are rapidly cleared from plasma, and a portion of this apoprotein then reappears in a slowly turning-over pool that constitutes the major mass of apoA-I. The physiologic identity of these kinetically distinct apoA-I species is unknown; however, the fast pool of apoA-I demonstrated in these studies is strikingly similar to that seen in subjects with Tangier disease who lack the slow pool.

Use of [15N]glycine in the measurement of apolipoprotein B synthesis in perfused rat liver.

Rat livers were perfused with [15N]glycine and unlabeled sodium benzoate by the single-pass technique via the portal vein or in retrograde fashion via the inferior vena cava. Perfusate [15N]hippurate enrichment was significantly greater than that of hepatic free glycine from 15 to 90 min, regardless of the direction of the perfusion. This result implies that differential labeling by periportal versus perivenous hepatocytes is not likely. When fasted animals were compared to those fed a chow diet or a sucrose-enriched diet, the labeling ratio of medium hippurate/hepatic free glycine decreased by only 9% in spite of a 5-fold decrease in the concentration of intrahepatic free glycine. Administration of nembutal to the intact animal significantly increased the enrichment of medium hippurate by 24% but did not affect the enrichment of the hepatic free glycine. We conclude that the difference between hippurate and free glycine enrichment is related to intracellular compartmentation of glycine transport. We suggest that measurement of the enrichment of hippurate after the administration of [15N]glycine with benzoate in intact animals or human subjects can therefore be used to estimate the enrichment of the intracellular precursor pool of glycine with a correction factor that does not vary appreciably under fed or fasted conditions. When uniformly labeled deuteroglycine was used as the tracer, enrichment of hepatic free glycine was decreased fivefold compared with [15N]glycine. Isotopic enrichments of apoBH and apoBL from the d less than 1.063 g/ml lipoprotein fraction isolated from the perfusion medium between 30 and 90 min averaged 3.7 and 4.1% excess, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Isolation of nascent high-density lipoprotein from rat liver perfusates by immunoaffinity chromatography: effects of oleic acid infusion.

Immunoaffinity chromatography on a column of rabbit IgG anti-rat apolipoprotein (apo) A-I covalently bonded to agarose was used to isolate nascent high-density lipoprotein (nHDL) from recirculated perfusates of rat livers. After passage through the affinity column, the bound material was eluted with sodium thiocyanate and analyzed for apolipoproteins and lipids. The protein content was 52% and the lipid composition was 37% triglyceride, 40% phospholipid, and 23% cholesterol. Apolipoproteins E and A-I each comprised approximately one third of the total, and very little apo B was detectable as judged by SDS-PAGE analysis. The affinity-isolated particles were therefore similar in composition to the major apo A-I:apo E-rich subfraction of nHDL isolated by ultracentrifugation in earlier work. It is concluded that the apo E in this class of nHDL (containing both apo E and apo A-I) is present in the secreted particle and is not a consequence of a loss of apo E from very-low-density lipoprotein (VLDL) during ultracentrifugation. The high triglyceride content in the virtual absence of apo B confirms and extends previous analyses and reinforces the conclusion that nHDL particles are enriched in triglyceride compared to plasma HDL. The inclusion of 4% albumin in the perfusion medium did not significantly change the total triglyceride output of 115 micrograms/g liver/h, but it decreased the triglyceride output isolated by anti-apo A-I affinity chromatography from 3.2 to 0.48 micrograms/g liver/h. The addition of oleic acid complexed to albumin increased the total triglyceride output by 70% and that associated with the immunoaffinity column increased from 0.48 to 2.7 micrograms/g liver/h.(ABSTRACT TRUNCATED AT 250 WORDS)