Search

Peroxisome proliferator-activated receptors (PPARs) are nuclear receptors that regulate lipid and glucose metabolism and cellular differentiation. PPAR-alpha and PPAR-gamma are both expressed in human macrophages where they exert anti-inflammatory effects. The activation of PPAR-alpha may promote foam-cell formation by inducing expression of the macrophage scavenger receptor CD36. This prompted us to investigate the influence of different PPAR-activators on cholesterol metabolism and foam-cell formation of human primary and THP-1 macrophages. Here we show that PPAR-alpha and PPAR-gamma activators do not influence acetylated low density lipoprotein-induced foam-cell formation of human macrophages. In contrast, PPAR-alpha and PPAR-gamma activators induce the expression of the gene encoding ABCA1, a transporter that controls apoAI-mediated cholesterol efflux from macrophages. These effects are likely due to enhanced expression of liver-x-receptor alpha, an oxysterol-activated nuclear receptor which induces ABCA1-promoter transcription. Moreover, PPAR-alpha and PPAR-gamma activators increase apoAI-induced cholesterol efflux from normal macrophages. In contrast, PPAR-alpha or PPAR-gamma activation does not influence cholesterol efflux from macrophages isolated from patients with Tangier disease, which is due to a genetic defect in ABCA1. Here we identify a regulatory role for PPAR-alpha and PPAR-gamma in the first steps of the reverse-cholesterol-transport pathway through the activation of ABCA1-mediated cholesterol efflux in human macrophages.

Abstract —To investigate the in vivo role that hepatic lipase (HL) plays in HDL metabolism independently of its lipolytic function, recombinant adenovirus (rAdV) expressing native HL, catalytically inactive HL (HL-145G), and luciferase control was injected in HL-deficient mice. At day 4 after infusion of 2×10 8 plaque-forming units of rHL-AdV and rHL-145G-AdV, similar plasma concentrations were detected in postheparin plasma (HL=8.4±0.8 μg/mL and HL-145G=8.3±0.8 μg/mL). Mice expressing HL had significant reductions of cholesterol (−76%), phospholipids (PL; −68%), HDL cholesterol (−79%), apolipoprotein (apo) A-I (−45%), and apoA-II (−59%; P P 125 I-labeled apoA-I HDL, 131 I-labeled apoA-II HDL, and [ 3 H]cholesteryl ester (CE) HDL revealed that compared with mice expressing luciferase control (fractional catabolic rate [FCR] in d −1 : apoA-I HDL=1.3±0.1; apoA-II HDL=2.1±0; CE HDL=4.1±0.7), both HL and HL-145G enhanced the plasma clearance of CEs and apoA-II present in HDL (apoA-II HDL=5.6±0.5 and 4.4±0.2; CE HDL=9.3±0.0 and 8.3±1.1, respectively), whereas the clearance of apoA-I HDL was enhanced in mice expressing HL (FCR=4.6±0.3) but not HL-145G (FCR=1.4±0.4). These combined findings demonstrate that both lipolytic and nonlipolytic functions of HL are important for HDL metabolism in vivo. Our study provides, for the first time, in vivo evidence for a role of HL in HDL metabolism independent of lipolysis and provides new insights into the role of HL in facilitating distinct metabolic pathways involved in the catabolism of apoA-I– versus apoA-II–containing HDL.

Epithelial cells contain two distinct membrane surfaces, the apical and basolateral plasma membranes, which have different lipid and protein compositions. In order to assess the effect of the compositional differences of the apical and basolateral membranes on their ability to undergo cholesterol efflux, MDCK cells were radiolabeled with [3H]cholesterol and grown as a polarized monolayer on filter inserts, that separate the upper apical compartment from the lower basolateral compartment. The rate of cholesterol efflux from the basolateral membrane into media containing HDL in the basolateral compartment was 6.3%/h ± 0.7, whereas HDL-mediated efflux from the apical membrane was approximately 3-fold slower (1.9%/h ± 0.3). In contrast, Fu5AH cells, which do not form distinct polarized membrane domains, had a similar rate of HDL-mediated cholesterol efflux into the apical and basolateral compartments. Similar to HDL, other cholesterol acceptors, namely LDL, bovine serum albumin, and a lipid emulsion, also showed a decreased rate of cholesterol efflux from the apical membrane surface versus the basolateral membrane. Compared to the basolateral membrane, the apical membrane was also found to be more resistant to cholesterol oxidase treatment, to bind less HDL, and to take up less cholesterol from the medium. In conclusion, cholesterol efflux occurred less readily from the apical membrane than from the basolateral membrane for all types of acceptors tested. These results suggest that differences in the composition of the apical and basolateral membrane lead to a relative decrease in cholesterol desorption from the apical membrane and hence a reduced rate of cholesterol efflux.—Remaley, A. T., B. D. Farsi, A. C. Shirali, J. M. Hoeg, and H. B. Brewer, Jr. Differential rate of cholesterol efflux from the apical and basolateral membranes of MDCK cells. J. Lipid Res. 1998. 39: 1231–1238.

We identified a 38-yr-old male patient with the clinical expression of homozygous familial hypercholesterolemia presenting as severe coronary artery disease, tendon and skin xanthomas, arcus lipoides, and joint pain. The genetic trait seems to be autosomal recessive. Interestingly, serum concentrations of cholesterol responded well to diet and statins. We had no evidence of an abnormal low density lipoprotein (LDL)-apolipoprotein B (apoB) particle, which was isolated from the patient using the U937 proliferation assay as a functional test of the LDL-binding capacity. The apoB 3500 and apoB 3531 defects were ruled out by PCR. In addition, we found no evidence for a defect within the LDL-receptor by skin fibroblast analysis, linkage analysis, single-strand conformational polymorphism and Southern blot screening across the entire LDL-receptor gene. The in vivo kinetics of radioiodinated LDL-apoB were evaluated in the proband and three normal controls, subsequently. The LDL-apoB isolated from the patient showed a normal catabolism, confirming an intact LDL particle. In contrast the fractional catabolic rate (d−1) of autologous LDL in the subject and the normal controls revealed a remarkable delayed catabolism of the patient’s LDL (0.15 vs. 0.33–0.43 d−1). In addition, the elevation of LDL-cholesterol in the patient resulted from an increased production rate with 22.8 mg/kg per day vs. 12.7–15.7 mg/kg per day. These data indicate that there is another catabolic defect beyond the apoB and LDL-receptor gene causing familial hypercholesterolemia.

Lecithin:cholesterol acyltransferase (LCAT) is an enzyme well known for its involvement in the intravascular metabolism of high density lipoproteins; however, its role in the regulation of apolipoprotein (apo) B-containing lipoproteins remains elusive. The present study was designed to investigate the metabolic mechanisms responsible for the differential lipoprotein response observed between cholesterol-fed hLCAT transgenic and control rabbits. 131I-labeled HDL apoA-I and 125I-labeled LDL kinetics were assessed in age- and sex-matched groups of rabbits with high (HE), low (LE), or no hLCAT expression after 6 weeks on a 0.3% cholesterol diet. In HE, the mean total cholesterol concentration on this diet, mg/dl (230 +/- 50), was not significantly different from that of either LE (313 +/- 46) or controls (332 +/- 52) due to the elevated level of HDL-C observed in HE (127 +/- 19), as compared with both LE (100 +/- 33) and controls (31 +/- 4). In contrast, the mean nonHDL-C concentration for HE (103 +/- 33) was much lower than that for either LE (213 +/- 39) or controls (301 +/- 55). FPLC analysis of plasma confirmed that HDL was the predominant lipoprotein class in HE on the cholesterol diet, whereas cholesteryl ester-rich, apoB-containing lipoproteins characterized the plasma of LE and, most notably, of controls. In vivo kinetic experiments demonstrated that the differences in HDL levels noted between the three groups were attributable to distinctive rates of apoA-I catabolism, with the mean fractional catabolic rate (FCR, d-1) of apoA-I slowest in HE (0.282 +/- 0.03), followed by LE (0.340 +/- 0.01) and controls (0.496 +/- 0.04). A similar, but opposite, pattern was observed for nonHDL-C levels and LDL metabolism (h-1), such that HE had the lowest nonHDL-C levels with the fastest rate of clearance (0.131 +/- 0.027), followed by LE (0.057 +/- 0.009) and controls (0.031 +/- 0.001). Strong correlations were noted between LCAT activity and both apoA-I (r= -0.868, P < 0.01) and LDL (r = 0.670, P = 0.06) FCR, indicating that LCAT activity played a major role in the mediation of lipoprotein metabolism. In summary, these data are the first to show that LCAT overexpression can regulate both LDL and HDL metabolism in cholesterol-fed rabbits and provide a potential explanation for the prevention of diet-induced atherosclerosis observed in our previous study.

Niemann-Pick C disease (NP-C) is a rare inborn error of metabolism with hepatic involvement and neurological sequelae that usually manifest in childhood. Although in vitro studies have shown that the lysosomal distribution of LDL-derived cholesterol is defective in cultured cells of NP-C subjects, no unusual characteristics mark the plasma lipoprotein profiles. We set out to determine whether anomalies exist in vivo in the cellular distribution of newly synthesized, HDL-derived or LDL-derived cholesterol under physiologic conditions in NP-C subjects. Three affected and three normal male subjects were administered [14C]mevalonate as a tracer of newly synthesized cholesterol and [3H]cholesteryl linoleate in either HDL or LDL to trace the distribution of lipoprotein-derived free cholesterol. The rate of appearance of free [14C]- and free [3H]cholesterol in the plasma membrane was detected indirectly by monitoring their appearance in plasma and bile. The plasma disappearance of [3H]cholesteryl linoleate was slightly faster in NP-C subjects regardless of its lipoprotein origin. Appearance of free [14C] cholesterol ill the plasma (and in bile) was essentially identical in normal and affected individuals as was the initial appearance of free [3H]cholesterol derived from HDL, observed before extensive exchange occurred of the [3H]cholesteryl linoleate among lipoproteins. In contrast, the rate of appearance of LDL-derived free [3H]cholesterol in the plasma membrane of NP-C subjects, as detected in plasma and bile, was retarded to a similar extent that LDL cholesterol metabolism was defective in cultured fibroblasts of these affected subjects. These findings show that intracellular distribution of both newly synthesized and HDL-derived cholesterol are essentially unperturbed by the NP-C mutation, and therefore occur by lysosomal-independent paths. In contrast, in NP-C there is defective trafficking of LDL-derived cholesterol to the plasma membrane in vivo as well as in vitro. The in vivo assay of intracellular cholesterol distribution developed herein should prove useful to quickly evaluate therapeutic interventions for NP-C.

In order to evaluate the coordinate role that he- patic lipase (HL) and 1ecithin:cholesterol acyltransferase (LCAT) play in modulating HDL particle heterogeneity and function in vivo we utilized recombinant adenovirus to ex- press HL in control and LCAT transgenic mice. Adenovirus- mediated expression of human HL in control (n = 4, LCAT activity = 42 2 1 nmol/ml per h) and LCAT-tg mice (n = 4, LCAT activity = 3566 I 93 nmol/ml per h) resulted in post heparin HL activities of 24358 5 6080 and 27266 2 7985 nmol/ml per min, respectively. Overexpression of HL led to significant reductions in total cholesterol, phospholipids, and HDL cholesterol in both LCAT-tg (62,62, and 63%, P< 0.05) and control mice (68,63, and 78%, P< 0.01) as well as to the formation of more homogenous HDL. However, compared to control animals, the reductions in the plasma concentrations of HDL-cholesterol and apoA-I were less in LCAT-tg mice (HDL-cholesterol: -62 t 15% vs. -78 +- 15%, P = 0.18; apoA-I: -36 2 7% vs. -76 2 8%, P < 0.0005). Gel filtration analysis revealed that in LCAT-tg mice the apoE-rich HDL, was preferentially reduced by expression of HL in vivo. Com- pared to control mice the reduction in the apoA-I/A-I1 HDL in transgenic mice was significantly less indicating that a sub- set of HDL in LCAT transgenic mice are resistant to the action of HL.M These combined data support a role for both HL and LCAT in modulating HDL heterogeneity and function, properties which may ultimately affect the ability of LCAT transgenic mouse HDL to function in the process of reverse cholesterol transport.-Dugi, K. A, B. L. Vaisman, N. Sakai, C. L. Knapper, S. M. Meyn, H. B. Brewer, Jr., and S. Santama- rina-Fojo. Adenovirus-mediated expression of hepatic lipase in LCAT transgenic mice. ,J Lipid Res. 1997. 38: 1822-1832.

We have established a mouse model for human LCAT deficiency by performing targeted disruption of the LCAT gene in mouse embryonic stem cells. Homozygous LCAT-deficient mice were healthy at birth and fertile. Compared with age-matched wild-type littermates, the LCAT activity in heterozygous and homozygous knockout mice was reduced by 30 and 99%, respectively. LCAT deficiency resulted in significant reductions in the plasma concentrations of total cholesterol, HDL cholesterol, and apoA-I in both LCAT -/- mice (25, 7, and 12%; p < 0. 001 of normal) and LCAT +/- mice (65 and 59%; p < 0.001 and 81%; not significant, p = 0.17 of normal). In addition, plasma triglycerides were significantly higher (212% of normal; p < 0.01) in male homozygous knockout mice compared with wild-type animals but remained normal in female knockout LCAT mice. Analyses of plasma lipoproteins by fast protein liquid chromatography and two-dimensional gel electrophoresis demonstrated the presence of heterogenous prebeta-migrating HDL, as well as triglyceride-enriched very low density lipoprotein. After 3 weeks on a high-fat high-cholesterol diet, LCAT -/- mice had significantly lower plasma concentrations of total cholesterol, reflecting reduced levels of both proatherogenic apoB-containing lipoproteins as well as HDL, compared with controls. Thus, we demonstrate for the first time that the absence of LCAT attenuates the rise of apoB-containing lipoproteins in response to dietary cholesterol. No evidence of corneal opacities or renal insufficiency was detected in 4-month-old homozygous knockout mice. The availability of a homozygous animal model for human LCAT deficiency states will permit further evaluation of the role that LCAT plays in atherosclerosis as well as the feasibility of performing gene transfer in human LCAT deficiency states.

Cholesteryl ester transfer protein (CETP) deficiency is the most common cause of hyperalphalipoproteinemia in Japan. However, the genetic basis of this disorder has not been fully characterized. We have studied a 49-year-old Japanese male presenting with total cholesterol, HDL-cholesterol, and apolipoprotein A-I levels of 300, 236, and 233 mg/dl, respectively, and total absence of CETP activity and mass in plasma. Sequence analysis of the patient's CETP gene revealed that the splice donor consensus GT was substituted by GG in intron 10 (intron 10 splice defect) and by AT in intron 14 (intron 14 splice defect). Restriction digestion of PCR-amplified DNA using NdeI and MaeIII established that the patient was a compound heterozygote for both gene defects. Sequencing of cDNA amplified by RT-PCR from the patient's monocyte-derived macrophage RNA demonstrated abnormal splicing with deletion of exon 10 as well as alternative splicing at a native AG site located 31 nucleotides 5' of the normal splice acceptor in intron 13. Thus, the intron 10 splice defect results in exon 10 skipping and the insertion of a 31 bp fragment between exon 13 and exon 14, which contains an in frame stop codon. The presence of abnormally spliced mRNA was further confirmed by amplification of patient cDNA using CETP specific primers. Abnormal splicing of exon 14 as a result of the intron 14 splice defect was not detected, indicating potential unstable CETP mRNA derived from that mutation. These findings demonstrate that a novel splice site mutation in intron 10 of the CETP gene results in the skipping of exon 10, as well as disruption of downstream splicing at intron 13 identifying a novel mechanism leading to CETP deficiency.

Lecithin cholesterol acyltransferase (LCAT) is an enzyme involved in the intravascular metabolism of high density lipoproteins (HDLs). Overexpression of human LCAT (hLCAT) in transgenic rabbits leads to gene dose-dependent increases of total and HDL cholesterol concentrations. To elucidate the mechanisms responsible for this effect, 131I-HDL apoA-I kinetics were assessed in age- and sex-matched groups of rabbits (n=3 each) with high, low, or no hLCAT expression. Mean total and HDL cholesterol concentrations (mg/dl), respectively, were 162+/-18 and 121+/-12 for high expressors (HE), 55+/-6 and 55+/-10 for low expressors (LE), and 29+/-2 and 28+/-4 for controls. Fast protein liquid chromatography analysis of plasma revealed that the HDL of both HE and LE were cholesteryl ester and phospholipid enriched, as compared with controls, with the greatest differences noted between HE and controls. These compositional changes resulted in an incremental shift in apparent HDL particle size which correlated directly with the level of hLCAT expression, such that HE had the largest HDL particles and controls the smallest. In vivo kinetic experiments demonstrated that the fractional catabolic rate(FCR, d(-1)) of apoA-I was slowest in HE (0.328+/-0.03) followed by LE (0.408+/-0.01) and, lastly, by controls (0.528+/-0.04). ApoA-I FCR was inversely associated with HDL cholesterol level (r=-0.851,P

ApoA-II is a major apolipoprotein constituent of high density lipoproteins (HDL) and may play an important role in lipoprotein metabolism and predisposition to atherosclerosis. Previous radiotracer kinetic studies have suggested that the metabolism of apoA-II in humans may be different than the metabolism of apoA-I, the major HDL apolipoprotein. In the present study, we have used an endogenous labeling technique using stable isotopically labeled amino acids to study apoA-II metabolism and compared the results to those obtained by a simultaneous exogenous radiotracer labeling method. Seven subjects with HDL cholesterol levels ranging from 9 to 93 mg/dl and apoA-II levels from 13 to 60 mg/dl were investigated in this study. [13C6]phenylalanine and 131I-labeled apoA-II were simultaneously administered as a primed-constant infusion and a bolus injection, respectively. In the endogenous labeling study, plateau tracer/tracee ratios of VLDL apoB-100 were used as estimates for the precursor pool tracer/tracee ratios for apoA-II synthesis. Residence times of apoA-II using these two independent methods were found to be highly correlated (r = 0.973, P < 0.0002). These results indicate that the endogenous labeling of apoA-II using stable isotopically labeled amino acids is a reasonable alternative to the conventional exogenous radiotracer labeling method for the investigation of apoA-II turnover. However, under the conditions of our experimental design and modeling strategy, the apoA-II residence times as determined by endogenous labeling were significantly longer (mean 5.33 days) than by exogenous radiotracer (mean 4.65 days). This suggests that apoA-II turnover may be even slower than believed based on radiotracer studies, and further supports the concept that HDL containing apoA-II are metabolized differently than HDL without apoA-II.

Lipoprotein (a) [Lp(a)] is an atherogenic lipoprotein which is similar in structure to, but metabolically distinct from, LDL. Factors modulating plasma Lp(a) concentrations are poorly understood. We hypothesized that patients with hyperlipidaemia have elevated Lp(a) levels and determined the phenotype, concentration and distribution of Lp(a) in a group of hyperlipidaemic patients (n = 107) compared with a control group (n = 128). Lp(a) concentrations were significantly increased in the hyperlipidaemic patients (mean, 34 +/- 4 mg dL-1; median, 19 mg dL-1) as compared with the controls (20 +/- 3 mg dL-1; 9 mg dL-1) (P0.01). Interestingly, after dividing the patients into one group with elevated cholesterol (200 mg dL-1) (n = 44) and another group with elevated triglycerides (200 mg dL-1) (n = 51) we found that Lp(a) concentrations were 2.3-fold higher in the high cholesterol patients (mean, 45 +/- 5; median, 41 mg dL-1) compared to the high triglyceride subjects (20 +/- 4; 8 mg dL-1) (P0.01). Furthermore, a negative correlation between triglyceride and Lp(a) plasma concentrations was found in patients exhibiting triglyceride levels300 mg dL-1 (r = -0.41, P = 0.04, n = 36) and with triglycerides400 mg dL-1 (r = -0.52, P = 0.03, n = 17). These data indicate that plasma Lp(a) concentrations are elevated in hyperlipidaemia if the patients have high cholesterol levels, whereas Lp(a) is normal to low in patients with elevated triglycerides.(ABSTRACT TRUNCATED AT 250 WORDS)

Type III hyperlipoproteinemia is characterized by delayed chylomicron and VLDL remnant catabolism and is associated with homozygosity for the apoE-2 allele. We have identified a kindred in which heterozygosity for an apoE mutant, apoE-1 (Lys146-->Glu), is dominantly associated with the expression of type III hyperlipoproteinemia. DNA sequence analysis of the mutant apoE gene revealed a single-point mutation that resulted in the substitution of glutamic acid (GAG) for lysine (AAG) at residue 146 in the proposed receptor-binding domain of apoE. The pathophysiological effect of this mutation was investigated in vivo by kinetic studies in the patient and six normal subjects, and in vitro by binding studies of apoE-1 (Lys146-->Glu) to LDL receptors on human fibroblasts and to heparin. The kinetic studies revealed that apoE-1 (Lys146-->Glu) was catabolized significantly slower than apoE-3 in normals (P < 0.005). In the proband, the plasma residence times of both apoEs were substantially longer and the production rate of total apoE was about two times higher than in the control subjects. ApoE-1 (Lys146-->Glu) was defective in interacting with LDL receptors, and its ability to displace LDL in an in vitro assay was reduced to 7.7% compared with apoE-3. The affinity of apoE-1 (Lys146-->Glu) to heparin was also markedly reduced compared with both apoE-2 (Arg158-->Cys) and apoE-3. These abnormal in vitro binding characteristics and the altered in vivo metabolism of apoE-1 (Lys146-->Glu) are proposed to result in the functional dominance of this mutation in the affected kindred.

Several key genes involved in cholesterol metabolism are known to be directly regulated by cholesterol. The possible indirect effect, however, of increased levels of cellular cholesterol on gene expression and its possible role in cholesterol metabolism and atherosclerosis has not been thoroughly explored. In order to determine the overall effect of cholesterol on gene expression, we isolated differentially expressed genes from a PCR-based subtraction library prepared from the liver of chow-fed and cholesterol-fed rabbits. A total of nine upregulated and four down-regulated cDNA fragments were isolated. As determined by Northern blot analysis, the expression of the isolated cDNAs began to change as early as the first week on the cholesterol-rich diet or as late as 4 weeks, which corresponded with hepatic cholesterol accumulation. Three of the cDNAs were identified by DNA sequence homology, whereas the remaining cDNAs had no significant homology match. CYP1A1, a cytochrome P450 isoenzyme, was found to be down-regulated in hepatocytes by cholesterol feeding. Osteopontin and Mac-2, which are produced by macrophages, were found to be up-regulated in Kupffer cells by cholesterol feeding. Overall these results demonstrate the usefulness of the subtraction library approach for identifying new candidate genes for exploring the pathogenesis of atherosclerosis.

The metabolism of high density lipoproteins (HDL) is tightly linked to the metabolism of apoB-containing lipoproteins through the exchange and transfer of lipids and apolipoproteins within the plasma compartment. Abetalipoproteinemia (ABL), a genetic disease in which apoB is absent from the plasma and HDL are the sole plasma lipoproteins, is a model for the investigation of HDL metabolism without modification by apoB-containing lipoproteins. Apolipoproteins A-I and E are two of the major apolipoproteins in HDL. Plasma apoA-I levels, but not apoE levels, have been reported to be decreased in patients with ABL. Furthermore, HDL from ABL patients is enriched in apoE compared with normal subjects. The purpose of the present study was: 1) to elucidate the metabolic basis of the low apoA-I levels in ABL; 2) to determine whether in vivo apoE production rates are normal in the absence of apoB-lipoprotein secretion; and 3) to test the hypothesis that apoE influences apoA-I and HDL catabolism in ABL. 131I-labeled apoA-I and 125I-labeled apoE were reassociated with autologous lipoproteins and injected into two unrelated ABL patients and control subjects. The mean residence time of apoA-I in ABL (2.4 days) was significantly decreased by nearly 50% compared with control subjects (4.7 +/- 0.6 days). ApoA-I production rates were also significantly decreased by 40% in ABL (7.1 mg/kg-d) compared with control subjects (11.8 +/- 1.7 mg/kg-d). The mean residence time of apoE in ABL (0.50 days) was somewhat shorter than that of control subjects (0.66 +/- 0.15 days), whereas the mean apoE production rate in ABL (2.14 mg/kg-d) was not substantially different from that of control subjects (1.55 +/- 0.62 mg/kg-d). HDL subfractions LpA-I and LpA-I:A-II were isolated using immunoaffinity chromatography. In contrast to the normal metabolism, apoA-I in LpA-I:A-II particles was catabolized at a faster rate than apoA-I in LpA-I, accounting for the greater decrease of plasma LpA-I:A-II relative to LpA-I in the ABL patients. HDL subfractions with and without apoE were also isolated using anti-apoE immunoaffinity chromatography. Labeled apoA-I in apoE-containing HDL was catabolized faster than that in HDL without apoE. Among the three different forms of apoE, the apoE monomer was catabolized at the fastest rate, the apoE homodimer at an intermediate rate, and the apoE-A-II heterodimer had the slowest rate of catabolism.(ABSTRACT TRUNCATED AT 400 WORDS)

Lipoprotein lipase (LPL) is a complex enzyme consisting of multiple functional domains essential for the initial hydrolysis of triglycerides present in plasma lipoproteins. Previous studies have localized the catalytic domain of LPL, responsible for the hydrolytic function of the enzyme, to the N-terminus whereas the C-terminal end may play a role in lipid and heparin binding. To date, most described missense mutations resulting in a nonfunctional LPL have been located in the N-terminal region of the enzyme. In this manuscript we describe the defect in the LPL gene of a patient with triglycerides ranging from normal to 12,000 mg/dl, low LPL mass, and no LPL activity in post-heparin plasma. Sequencing of patient PCR-amplified DNA identified two separate mutations in the C-terminal domain of LPL: an A-->T transversion at nucleotide 1484 resulting in a Glu410-->Val substitution and a C-->G mutation at position 1595 that introduces a premature stop codon at position 447. Digestion with MaeIII and MnII established that the patient is a true homozygote for both mutations. In order to investigate the functional significance of these defects, mutant enzymes containing either the Val410 or the Ter447 mutations as well as both Val410 and Ter447, were expressed in vitro. Compared to the wild-type enzyme, LPL447 demonstrated a moderate reduction of specific activity using triolein (70% of normal) and tributyrin (74% of normal) substrates, while LPL410 had a significant (11% and 23% of normal) reduction of the normal lipase and esterase specific activities, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Lipoprotein(a) (Lp[a]) is an atherogenic lipoprotein which is similar in structure to low density lipoproteins (LDL) but contains an additional protein called apolipoprotein(a) (apo[a]). Apo(a) is highly polymorphic in size, and there is a strong inverse association between the size of the apo(a) isoform and the plasma concentration of Lp(a). We directly compared the in vivo catabolism of Lp(a) particles containing different size apo(a) isoforms to establish whether there is an effect of apo(a) isoform size on the catabolic rate of Lp(a). In the first series of studies, four normal subjects were injected with radio-labeled S1-Lp(a) and S2-Lp(a) and another four subjects were injected with radiolabeled S2-Lp(a) and S4-Lp(a). No significant differences in fractional catabolic rate were found between Lp(a) particles containing different apo(a) isoforms. To confirm that apo(a) isoform size does not influence the rate of Lp(a) catabolism, three subjects heterozygous for apo(a) were selected for preparative isolation of both Lp(a) particles. The first was a B/S3-apo(a) subject, the second a S4/S6-apo(a) subject, and the third an F/S3-apo(a) subject. From each subject, both Lp(a) particles were preparatively isolated, radiolabeled, and injected into donor subjects and normal volunteers. In all cases, the catabolic rates of the two forms of Lp(a) were not significantly different. In contrast, the allele-specific apo(a) production rates were more than twice as great for the smaller apo(a) isoforms than for the larger apo(a) isoforms. In a total of 17 studies directly comparing Lp(a) particles of different apo(a) isoform size, the mean fractional catabolic rate of the Lp(a) with smaller size apo(a) was 0.329 +/- 0.090 day-1 and of the Lp(a) with the larger size apo(a) 0.306 +/- 0.079 day-1, not significantly different. In summary, the inverse association of plasma Lp(a) concentrations with apo(a) isoform size is not due to differences in the catabolic rates of Lp(a) but rather to differences in Lp(a) production rates.

Sandwich ELISAs with an anti-apo(a) trapping antibody and either an anti-apolipoprotein B or anti-apolipoprotein(a) reporting antibody, were used to measure the concentrations of Lp(a) in 230 plasma samples that were either freshly drawn or stored at −20°C for 4–6 weeks. The assays produced significantly different results for the fresh and frozen samples, however, the magnitudes of these differences were small, about 8% higher for the frozen samples, and independent of total cholesterol, HDL cholesterol, triglyceride, apolipoprotein B or Lp(a) concentration or assay configuration. A similar difference was seen for a freshly drawn plasma sample assayed at the same time as the fresh and frozen samples, indicating the differences were due to inherent differences in the assays at the times the assays were performed. The assay configuration was an important factor in determining the Lp(a) concentrations for identically treated samples. ELISAs using the apoB reporting antibody yielded concentrations that were significantly less than those determined by ELISAs using the anti-apo(a) reporting antibody. The assay differences did not correlate with total cholesterol, HDL cholesterol, triglyceride, or apoB concentration. However, the magnitude of the difference did correlate well with Lp(a) amount. Low Lp(a) concentrations produced greater assay differences than high Lp(a) concentrations.

Epidemiological studies have established that concentrations of plasma high-density lipoproteins (HDL) are inversely associated with premature atherosclerosis, but the physiological basis of this relationship remains unknown. We investigated 5 probands with very low plasma HDL. None had clinical or biochemical findings typical of the known genetic disorders with low HDL nor had evidence of premature coronary atherosclerosis by sensitive diagnostic methods. All 5 probands and the son of 1 of them had rapid catabolism of the HDL apolipoproteins A-I and A-II. These results indicate that not all people with low HDL are necessarily at risk of premature coronary heart disease and that further investigation is required before decisions can be made about their management.

Apolipoprotein A-I is the major apolipoprotein constituent of high density lipoproteins (HDL). Methods used to investigate in vivo kinetics of apoA-I include exogenous labeling with radioiodine and endogenous labeling with stable isotopically labeled amino acids. We report here a direct comparison of these methods to determine the in vivo kinetics of apoA-I in four normal subjects. Purified apoA-I was labeled with 125I, reassociated with autologous plasma, and injected into study subjects. At the same time, [13C6]phenylalanine was administered as a primed constant infusion for up to 14 hours. The kinetic parameters of apoA-I were determined from the 125I-labeled apoA-I plasma curves. For the analysis of data from stable isotope studies, very low density lipoprotein (VLDL) apoB-100, VLDL apoB-48, and total apoA-I were isolated by ultracentrifugation and subsequent preparative NaDodSO4-PAGE, hydrolyzed, and derivatized. The tracer/tracee ratio was determined by gas chromatography-mass spectrometry. Monoexponential function analysis was used to determine the tracer/tracee curves of VLDL apoB-100 and VLDL apoB-48, and total apoA-I. The mean plateau tracer/tracee ratio of VLDL apoB-100 (primarily liver-derived) was 5.19%, whereas that of VLDL apoB-48 (intestinally derived) was only 3.74%. Using the VLDL apoB-100 plateau tracer/tracee ratio as the estimate of the precursor pool enrichment for apoA-I, the mean apoA-I residence time (RT) was 5.14 +/- 0.41 days, compared with 4.80 +/- 0.30 days for the exogenous labeling method. The apoA-I RTs using these two methods were highly correlated (r = 0.874).(ABSTRACT TRUNCATED AT 250 WORDS)

SELECTED CASE A 30-year-old woman was referred to the National Institutes of Health (NIH) with malabsorption, ataxia, visual impairment, and low serum cholesterol level. As an infant, she had frequent colic and diarrhea, which improved when the amount of fat in her diet was decreased. She continued to have malabsorptive symptoms after ingesting a fatty meal and was diagnosed at 8 years of age as having celiac disease. A gluten-free diet did not, however, improve her symptoms. During her late teens and twenties, she experienced slowly progressive difficulty maintaining her balance, numbness in her legs below the knee, and difficulty in seeing at night with worsening color vision. At age 28 years she had a small-intestine biopsy; the findings were not consistent with celiac disease but rather revealed lipid-engorged enterocytes. Her family history was remarkable in that her parents were fourth cousins and her three younger sisters had similar, though

Lipoprotein lipase (LPL) and hepatic lipase (HL) mediate the hydrolysis of triglycerides and phospholipids present in circulating lipoprotein particles and are essential for normal lipid metabolism. Both enzymes have a similar primary amino acid structure and share requirements for intact catalytic, lipid binding, and heparin binding domains. However, LPL and HL exhibit different substrate specificities and cofactor requirements. In order to characterize the functional domains necessary for LPL activity, a chimeric lipase consisting of the amino-terminal 314 amino acids of human LPL and the carboxyl-terminal 146 amino acids of human HL was synthesized by joining the cDNA of both lipases at the 5'-end of exon 7. Northern blot hybridization and Western blot analyses revealed the size of the chimera mRNA and protein to be approximately 1.5 kb and 55 kDa, respectively. The chimeric enzyme hydrolyzed both long chain and short chain fatty acid triacylglycerols and had catalytic properties that were similar to lipoprotein lipase. Thus, apolipoprotein (apo)C-II was required for maximal lipase activity, and high salt concentration abolished the ability of the chimera to hydrolyze triolein even in the presence of apoC-II. A monospecific anti-HL polyclonal antibody interacting with the C-terminal HL-derived domain of the chimeric enzyme abolished the enzyme's ability to hydrolyze triglyceride emulsion but not tributyrin substrates. Analysis of the heparin binding properties of the chimeric enzyme using heparin-Sepharose affinity chromatography revealed an elution pattern which was intermediate between that of lipoprotein and hepatic lipase. In summary, we have characterized the functional properties of an LPL-HL chimeric enzyme.(ABSTRACT TRUNCATED AT 250 WORDS)

Apolipoprotein (apo) A-IV is a polymorphic, intestinally derived apolipoprotein that is genetically linked to and similar in structure to apoA-I, the major apolipoprotein in high density lipoproteins (HDL). ApoA-IV plays a potentially important role in lipoprotein metabolism and reverse cholesterol transport, but its in vivo metabolism is poorly understood. In order to gain insight into factors modulating apoA-IV metabolism in humans, the in vivo kinetics of the two major human apoA-IV isoproteins apoA-IV-1 and apoA-IV-2 were investigated in normolipidemic human subjects. 131I-apoA-IV-1 and 125I-apoA-IV-2 were reassociated with autologous plasma and injected into study subjects. Analysis of the kinetic data revealed a rapid mean fractional catabolic rate (FCR) for apoA-IV-1 of 2.42 +/- 0.11 d-1. The mean production, or transport, rate of apoA-IV-1 was 16.3 +/- 1.4 mg/kg per d. Plasma apoA-IV concentrations were highly correlated with apoA-IV production rate (r = 0.84, P < 0.001) and not correlated with apoA-IV fractional catabolic rate (r = 0.25, P = NS). The mean FCR of apoA-IV-2 was 2.21 +/- 0.10 d-1. In the ten subjects in whom 131I-apoA-IV-1 and 125I-apoA-IV-2 were simultaneously injected, the FCR of apoA-IV-2 was significantly slower by paired t test (P = 0.003). The FCR of apoA-IV-2 in an apoA-IV-2/2 homozygote was only 1.49 d-1, substantially slower than in all other subjects. We conclude that: (a) apoA-IV is a rapidly catabolized apolipoprotein in humans, with a fractional catabolic rate more than 10 times greater than that of apoA-I; (b) apoA-IV has a high absolute transport rate similar to that of apoA-I; (c) plasma levels of apoA-IV are primarily determined by apoA-IV production rate in normolipidemic subjects; and (d) the fractional catabolic rate of the common variant apoA-IV-2 is slower than that of the wild-type apoA-IV-1.

We have identified the molecular defect in two siblings presenting with classical clinical and biochemical features of Fish Eye disease (FED), including corneal opacities, HDL cholesterol < 10 mg/dl, normal plasma cholesteryl esters, and elevated triglycerides. In contrast to previously reported patients with FED who are unable to esterify HDL-associated cholesterol, our patients' plasma lecithin-cholesterol acetyltransferase (alpha-LCAT)-specific activities assayed using an HDL-like proteoliposome substrate were 12.7-25.7 nmol/micrograms (19.5 +/- 1.8 in controls). In addition, significant residual cholesterol esterification was present in VLDL/LDL-depleted plasma, confirming the presence of HDL-associated alpha-LCAT activity. DNA sequence analysis of the proband's LCAT gene identified deletion of the triplet coding for leu300, which resulted in the loss of a restriction site for MlnI. Digestion of PCR-amplified DNA using MlnI established that both siblings are homozygous for this defect. Expression of LCAT300-del. in human embryonic kidney-293 cells revealed normal mRNA and intracellular LCAT concentrations. However, reduced amounts of LCAT300-del., which had a normal specific alpha-LCAT activity, were present in the media. In summary, we report the first case of FED associated with a mutant enzyme that has a normal alpha-LCAT-specific activity. The functional significance of this LCAT gene defect has been established in an in vitro expression system, which demonstrates that very small amounts of this functional LCAT mutant enzyme accumulate in the media. Characterization of LCAT300-del. established that selective alpha-LCAT deficiency is not a prerequisite for the development of FED. On the basis of our combined results, we propose that the residual amounts of total plasma LCAT activity and not its distribution on lipoproteins primarily determines the heterogeneity in phenotypic expression observed in familial LCAT deficiency syndromes.

Lipoprotein lipase, a lipolytic enzyme essential for normal hydrolysis of triglycerides in very low density lipoprotein (VLDL) and chylomicrons, is found in several cell types, including macrophages. The role of lipoprotein lipase in mediating the uptake of normal VLDL triglycerides into human cultured monocyte-derived macrophages was studied using macrophage cells from a functionally lipoprotein lipase-deficient patient and macrophages of cells from a normal subject. After incubation with VLDL, massive accumulation of phase refractile (lipid) inclusions were noted by phase contrast microscopy within the normal, but not within the lipoprotein lipase-deficient, macrophages. Chemical determinations of intracellular lipid confirmed massive triglyceride accumulation within normal macrophages, but not in lipoprotein lipase-deficient macrophages. VLDL-derived cholesterol did not accumulate in either cell. These results confirm an additional role of lipoprotein lipase, that of mediating triglyceride accumulation into macrophages from normal human VLDL. Human monocyte-macrophages genetically deficient in a functional lipoprotein lipase will be useful to determine the role of lipoprotein lipase in macrophage accumulation of lipid from other forms of triglyceride-carrying lipoproteins, including hypertriglyceridemic VLDL, beta-VLDL, and chylomicrons.

O-linked glycosylation is a common post-translational modification of apolipoproteins, but no structural or functional role for it has been identified. We examined the biosynthesis of apolipoprotein (apo) A-II in Hep G2 cells and in glycosylation-defective Chinese hamster ovary (CHO) cell mutants transfected with apoA-II cDNA. Three monomeric isoforms of apoA-II with an apparent molecular mass of 8.5, 9.8, and 11.4 kDa were synthesized by Hep G2 cells and transfected wild-type CHO cells. The 9.8- and 11.4-kDa isoforms were sialylated but not the 8.5-kDa isoform. Transfected 1dlD cells, which are defective in the biosynthesis of galactose and N-acetylgalactosamine, only produced the 8.5-kDa isoform; however, when grown in media supplemented with these sugars, ldlD cells produced all three isoforms of apoA-II. Pulse-chase analysis of ldlD cells showed that glycosylation was not necessary for secretion of apoA-II. Glycosylation did modify the association of apoA-II with nascent high density lipoprotein (HDL) secreted by Hep G2 cells. The sialylated isoforms were lipid-poor and were present in the lipoprotein-deficient density range, whereas the nonsialylated 8.5-kDa isoform was associated with LpA-I, A-II lipoprotein particles in the HDL density range. ApoA-II from transfected ldlD cells, regardless of glycosylation, were lipid-poor. When preincubated with HDL from serum, however, sialylated apoA-II from both ldlD cells and Hep G2 cells associated with lipoprotein particles within the HDL3 density, whereas nonsialylated apoA-II was found throughout the HDL density range. In summary, O-linked glycosylation is not necessary for the secretion of apoA-II but does modify the association of apoA-II to HDL and may, therefore, play an important role in the metabolism of HDL.

Lipoprotein(a) [Lp(a)] is an atherogenic lipoprotein which is similar in structure to, but metabolically distinct from, LDL. Factors regulating plasma concentrations of Lp(a) are poorly understood. Apo(a), the protein that distinguishes Lp(a) from LDL, is highly polymorphic, and apo(a) size is inversely correlated with plasma Lp(a) level. Even within the same apo(a) isoform class, however, plasma Lp(a) concentrations vary widely. A series of in vivo kinetic studies were performed using purified radiolabeled Lp(a) in individuals with the same apo(a) isoform but different Lp(a) levels. In a group of seven subjects with a single S4-apo(a) isoform and Lp(a) levels ranging from 1 to 13.2 mg/dl, the fractional catabolic rate (FCR) of 131I-labeled S2-Lp(a) (mean 0.328 day-1) was not correlated with the plasma Lp(a) level (r = -0.346, P = 0.45). In two S4-apo(a) subjects with a 10-fold difference in Lp(a) level, the FCR's of 125I-labeled S4-Lp(a) were very similar in both subjects and not substantially different from the FCRs of 131I-S2-Lp(a) in the same subjects. In four subjects with a single S2-apo(a) isoform and Lp(a) levels ranging from 9.4 to 91 mg/dl, Lp(a) concentration was highly correlated with Lp(a) production rate (r = 0.993, P = 0.007), but poorly correlated with Lp(a) FCR (mean 0.304 day-1). Analysis of Lp(a) kinetic parameters in all 11 subjects revealed no significant correlation of Lp(a) level with Lp(a) FCR (r = -0.53, P = 0.09) and a strong correlation with Lp(a) production rate (r = 0.99, P < 0.0001). We conclude that the substantial variation in Lp(a) levels among individuals with the same apo(a) phenotype is caused primarily by differences in Lp(a) production rate.

The molecular defects in the lecithin:cholesterol acyltransferase (LCAT) gene have been identified in a 52-year-old patient with classic LCAT deficiency, presenting with corneal clouding and proteinuria. Plasma total cholesterol was normal, triglycerides were elevated, whereas high density lipoprotein (HDL) cholesterol (8 mg/dl) and plasma cholesteryl esters (6% of total cholesterol) were markedly reduced. Plasma cholesterol esterification rate (pCER) was zero, alpha-LCAT activity, assayed using an HDL-like proteoliposome substrate was reduced to 1.6% of control, and LCAT mass was 3.7% of normal plasma levels. DNA sequence analysis of the proband's LCAT gene identified a C to A substitution, converting tyr83 to a stop codon, and a T to A transition, replacing tyr156 by asn. Restriction analysis of PCR-amplified DNA from the proband, a control and his four children using the enzymes Acc I and Rsa I established that the patient is a compound heterozygote for both mutations. The two children, heterozygous for the stop codon defect, were phenotypically indistinguishable from the two with the tyr156 defect. In vitro expression of LCAT (tyr156–>asn) in human embryonic kidney-293 cells established the functional significance of this mutation. The secreted translation product had only 6% of control mass and no detectable CER; however, the residual LCAT mass of the in vitro expressed LCAT (tyr156–>asn) demonstrated a specific alpha-LCAT activity of 30% of control, suggesting that this amino acid substitution results in a mutant enzyme that retains some enzymic activity, but may be rapidly catabolized. In summary, we have identified two unique defects in the LCAT gene that lead to the expression of classic LCAT deficiency in this kindred.

Fibric acid derivatives (fibrates) and β-hydroxy-β-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors are hypolipidaemic agents with contrasting mechanisms of action. Fibrates have multiple effects on capillary and hepatic enzymes involved in fatty acid and triglyceride metabolism. The principal lipid-lowering effect of fibrates appears to be mediated through increased activity of lipoprotein lipase, but other hepatic enzymes are also affected by these drugs. By increasing the activity of lipoprotein lipase, fibrates have a pronounced triglyceride-lowering effect; the response of plasma cholesterol to fibrates, however, varies with the nature of the underlying lipid abnormality. Some elevation of low density lipoprotein (LDL) may occur in hypertriglyceridaemic patients, apparently as a result of reduced non-receptor-mediated clearance of LDL. In hypercholesterolaemic patients, fibrates reduce triglyceride and, to a lesser extent, plasma LDL levels; the reduction in LDL may be related to stimulation of receptor-mediated clearance. Reductase inhibitors are the most potent cholesterol-lowering drugs available and have a clearly defined mode of action. By inhibiting HMG-CoA reductase, the rate-limiting step in hepatic cholesterol synthesis, these agents induce a compensatory increase in the synthesis and expression of hepatic LDL receptors, resulting in reduction of plasma LDL levels. There is also evidence that reductase inhibitors inhibit the synthesis of LDL and apolipoprotein (apo) B, the principal surface protein in LDL. This inhibitory action may be especially beneficial in patients with combined hyperlipidaemia, in whom there is evidence of overproduction of LDL and apo B. Reductase inhibitors also reduce triglyceride levels, but to a lesser extent than LDL; their triglyceride-lowering efficacy is not as marked as that of fibrates. LDL is regarded as the principal atherogenic lipoprotein and preference may, therefore, be given to reductase inhibitors when contemplating lipid regulation for the prevention of CHD. Fibrates, however, offer advantages in the management of patients with predominant or exclusive hypertriglyceridaemia.

Lipoprotein lipase (LPL), a key enzyme which initiates the hydrolysis of triglycerides present in chylomicrons and very low density lipoproteins, consists of multiple functional domains which are necessary for normal activity. The catalytic domain of LPL mediates the esterase function of the enzyme but separate lipid binding sites have been proposed to be involved in the interaction of LPL with emulsified lipid substrates at the water-lipid interface. Like pancreatic lipase (PL), LPL contains a surface loop covering the catalytic pocket that may modulate access of the substrate to the active site of the enzyme. Secondary structural analysis of this loop reveals a helix-turn-helix motif with two short amphipathic helices that have hydrophobic moments of 0.64 and 0.68. In order to investigate the role of the loop in the initial interaction of LPL with its substrate, we utilized site-directed mutagenesis to generate eight constructs in which the amphipathic properties of the loop were altered and expressed them in human embryonal kidney-293 cells. Reducing the amphiphilicity without changing the predicted secondary structure of the loop abolished the ability of the lipase to hydrolyze emulsified, long chain fatty acid triglycerides (triolein) but not the water soluble substrate tributyrin. Replacing the loop of LPL with the loop of hepatic lipase, which differs in 15 of 22 amino acids but is also amphiphilic, led to the expression of an enzyme that retained both triolein and tributyrin hydrolyzing activity. Substitution of the LPL loop by a short four amino acid peptide, which may allow more direct access to the active site than the 22 amino acid loop, enhanced hydrolysis of short chain fatty acid triglycerides by more than 2-fold, while the ability to hydrolyze emulsified substrates was abolished. Thus, disruption of the amphipathic structure of the LPL loop selectively decreases the hydrolysis of emulsified lipid substrate without affecting the esterase or catalytic function of the enzyme. These studies establish that the loop with its two amphipathic helices is essential for hydrolysis of long chain fatty acid substrate by LPL providing new insight into the role of the LPL loop in lipid-substrate interactions. We propose that the interaction between the lipoprotein substrates and the amphipathic helices within this loop may in part determine lipase substrate specificity.

Apolipoprotein (apo) E-4Philadelphia is a double mutant of apoE in which residue 13 of the mature protein, glutamic acid (GAG), is replaced by lysine (AAG) and amino acid 145, arginine (CGT), is converted to cysteine (TGT). These mutations result in two restriction fragment length polymorphisms for the enzymes AvaI and BbvI, a smaller apparent molecular weight of apoE-4Philadelphia on sodium dodecyl polyacrylamide gels, and severe type III hyperlipoproteinemia (HLP) in a 24-year-old homozygous female (Lohse, P., Mann, W. A., Stein, E. A., and Brewer, H. B., Jr. (1991) J. Biol. Chem. 266, 10479-10484). In the current study, we have extended our analysis to include nine additional family members of the Philadelphia kindred spanning four generations. DNA and protein analysis demonstrated that the originally described propositus is a true homozygote for the epsilon-4Philadelphia allele and that six of the nine family members are heterozygous for the mutated allele and the normal epsilon-3 allele or, in one case, the epsilon-4 allele. Heterozygosity for apoE-4Philadelphia leads to the expression of a moderate form of type III HLP without clinical manifestations. These results are consistent with a dominant mode of inheritance of this dyslipoproteinemia. The simultaneous presence of unaffected individuals, heterozygotes, and a homozygote in the Philadelphia kindred makes it possible for the first time to demonstrate that the mutant apoE exhibits an incomplete or partial dominance of type III HLP. Heterozygosity for the normal epsilon-3 allele appears to have an influence on the expression of type III HLP, resulting in a phenotype intermediate between that of the two homozygous states.