The role of hepatic lipase in lipoprotein metabolism

Clinica Chimica Acta 286 (1999) 243–255

The role of hepatic lipase in lipoprotein metabolism Philip W. Connelly

a,b ,

*

a

b

Department of Medicine, St. Michael’ s Hospital, University of Toronto, Toronto, Ont., Canada Departments of Biochemistry and Laboratory Medicine and Pathobiology University of Toronto, Toronto, Ont., Canada Received 20 November 1998; accepted 7 January 1999

Abstract Hepatic lipase (HL) is one of two major lipases released from the vascular bed by intravenous injection of heparin. HL hydrolyzes phospholipids and triglycerides of plasma lipoproteins and is a member of a lipase superfamily that includes lipoprotein lipase and pancreatic lipase. The enzyme can be divided into an NH 2 -terminal domain containing the catalytic site joined by a short spanning region to a smaller COOH-terminal domain. The NH 2 -terminal portion contains an active site serine in a pentapeptide consensus sequence, Gly–Xaa–Ser–Xaa–Gly, as part of a classic Ser–Asp–His catalytic triad, and a putative hinged loop structure covering the active site. The COOH-terminal domain contains a putative lipoprotein-binding site. The heparin-binding sites may be distributed throughout the molecule, with the characteristic elution pattern from heparin– sepharose determined by the COOH-terminal domain. Of the three N-linked glycosylation sites, Asn-56 is required for efficient secretion and enzymatic activity. HL is hypothesized to directly couple HDL lipid metabolism to tissue / cellular lipid metabolism. The potential significance of the HL pathway is that it provides the hepatocyte with a mechanism for the uptake of a subset of phospholipids enriched in unsaturated fatty acids and may allow the uptake of cholesteryl ester, free cholesterol and phospholipid without catabolism of HDL apolipoproteins. HL can hydrolyze triglyceride and phospholipid in all lipoproteins, but is predominant in the conversion of intermediate density lipoproteins to LDL and the conversion of post-prandial triglyceride-rich HDL into the post-absorptive triglyceride-poor HDL. It has been suggested that enzymatically inactive HL can play a role in hepatic lipoprotein uptake forming a ‘bridge’ by binding to the lipoprotein and to the cell surface. This raises the interesting possibility that production and secretion of mutant inactive HL could promote clearance of VLDL remnants. We have described a rare family with HL deficiency. Affected patients are compound heterozygotes for a mutation of Ser267Phe that causes an inactive enzyme and a mutation of Thr383Met that results in impaired *Corresponding address: J. Alick Little Lipid Research Laboratory, 38 Shuter Street, Toronto, Ont. M5B 1A6, Canada. Tel.: 1 1-416-864-6023; fax: 1 1-416-864-5870. 0009-8981 / 99 / $ – see front matter  1999 Elsevier Science B.V. All rights reserved. PII: S0009-8981( 99 )00105-9

244

P.W. Connelly / Clinica Chimica Acta 286 (1999) 243 – 255

secretion of HL and reduced specific activity. Human HL deficiency in the context of a second factor causing hyperlipidemia is strongly associated with premature coronary artery disease.  1999 Elsevier Science B.V. All rights reserved. Keywords: Hepatic lipase; Lipoproteins; Lipoprotein lipase; Triglyceride; Phospholipid; Heparin; Lipoprotein receptors

1. Introduction Hepatic lipase (HL) is one of two major lipases released from the vascular bed by intravenous injection of heparin [1], the other being lipoprotein lipase (LPL). HL is distinguished from LPL by its resistance to inhibition by 1 M NaCl or protamine sulfate and the absence of a requirement for an apolipoprotein activator. HL is synthesized primarily by hepatocytes (and also found in adrenal gland and ovary) and hydrolyzes phospholipids and triglycerides of plasma lipoproteins. This discussion of HL will cover the structure and evolutionary neighbors of HL, the lipid and lipoprotein substrates of HL, the question of whether HL functions as a ‘bridge’ in lipoprotein uptake, the phenotype of the HL ‘knock out’, and finally future prospects for research into HL.

2. Normal HL structure The amino acid sequence of mature human HL has been deduced from DNA sequence data to consist of 476 amino acids, comprising a protein with a calculated molecular weight of 53 431 Da [2,3]. The human HL gene is on chromosome 15q21 and is 35 kilobases (kb) in size, with nine exons that encode a mRNA of 1.5 kb [4]. HL requires glycosylation at asparagine-56 for activity and secretion [5]. HL can be divided into two domains, an N-terminal domain that contains an Asp–His–Ser catalytic triad and a C-terminal domain. The active site serine is a pentapeptide consensus sequence, Gly–Xaa–Ser–Xaa– Gly. The heparin binding sites may be in either domain; however, the behavior on elution from heparin – sepharose columns using a salt gradient is determined primarily by the C-terminal domain. This domain contains the major lipoprotein binding site(s). Hill et al. [6] have proposed that human HL functions as a dimer with the monomers in a head-to-tail arrangement (Fig. 1). Dugi et al. used chimeric enzymes in which a putative ‘loop’ domain that covers the active site of HL was exchanged with the homologous domain from LPL [7]. These elegant studies showed that the ‘loop’ covering the active site has lipid binding characteristics that determine the relative balance between triglyceride and

P.W. Connelly / Clinica Chimica Acta 286 (1999) 243 – 255

245

Fig. 1. Schematic representation of the domain structure of HL in the homodimeric form.

phospholipid hydrolysis. These studies were extended by Kobayashi et al. [8] to in vivo experiments in mice using adenovirus constructs to express the chimeric enzymes in mice with the native HL gene knocked out. The most significant lowering of plasma phospholipid and cholesterol levels was observed for intact human HL and human LPL containing the HL ‘loop’ domain.

3. HL evolutionary neighbors The cloning and sequencing of three major lipases, HL, LPL and pancreatic lipase made it clear that they constitute a family of homologous enzymes. A gapped BLAST 2.0 search [9] of the NIH sequence database identified HL as a member of a lipase family that includes LPL, pancreatic lipase, pancreatic lipase-related protein-1, pancreatic lipase-related protein-2 and hornet phospholipase A 1 , a distant relationship with fly yolk proteins was also noted (Fig. 2). The homology is highest with LPL and is seen throughout the sequence, except for a small portion of the N-terminus. Pancreatic lipase shares significant homology except at the N-terminus and the C-terminus. The pancreatic lipaserelated proteins 1 and 2, which have yet to have a clearly identified physiological role, share homology across nearly the entire length of HL, but are distinguished by several large gaps within the alignment. The hornet phospholipase A 1 shares homology over a central region of HL and lacks homology in the C-terminal domain. This is consistent with the lack of binding of this lipase to lipoprotein substrates and the proposal that the C-terminal domain of HL is primarily important as a lipid-binding domain.

246

P.W. Connelly / Clinica Chimica Acta 286 (1999) 243 – 255

Fig. 2. Alignment of lipase sequences with human HL. Solid bars indicate homologous sequences and hatched bars indicate gaps in homology. LPL, lipoprotein lipase; PL, pancreatic lipase; PL-RP1, pancreatic lipase-related protein 1; PL-RP2, pancreatic lipase-related protein 2; PLA-1, phospholipase A 1 .

4. Lipid and lipoprotein substrates of HL HL hydrolyzes both phospholipid (PL) and triglyceride (TG). It functions as a phospholipase A 1 and hydrolyzes fatty acids from the 1 and 3 positions of TG. It preferentially hydrolyzes phosphatidylethanolamine (PE) and phosphatidylcholine (PC) containing the unsaturated fatty acids linoleate and arachidonate [10]. Early studies by Nilsson and colleagues suggested that HL plays a special role in vivo in the partitioning and uptake of chylomicron and HDL phospholipids by the liver [11]. Because HL is a phospholipase A 1 , the products of phospholipid hydrolysis are 2-acyl lysophosphatidylcholine and 2-acyl lysophosphatidylethanolamine. These lysophospholipids are likely rapidly reacylated in the hepatocyte, providing a mechanism for the conservation of unsaturated fatty acids. On the basis of animal studies and studies of human HL deficiency, we previously summarized the dual role of HL in intravascular lipoprotein metabolism (Fig. 3). The evidence in 1986 pointed to a role for HL in the hydrolysis of the TG that could accumulate in the post-prandial HDL and we proposed that the net balance between HDL subfractions called HDL 2 and HDL 3 was a result of a competition between HL and the enzyme lecithin cholesterol acyltransferase (LCAT) for HDL [12]. HL was also known to be important in the final steps in the lipolytic conversion of very low density lipoprotein (VLDL) to low density lipoprotein (LDL) by hydrolysis of the LDL triglyceride.

P.W. Connelly / Clinica Chimica Acta 286 (1999) 243 – 255

247

Fig. 3. Metabolic scheme showing the multiple steps in lipoprotein metabolism catalyzed by HL. Abbreviations: MG, monoglycerides; lysoPL, lysophospholipid; FFA, free fatty acid; PL, phospholipid; TG, triglyceride; apo, apolipoproteins; LP, lipoprotein, LPL, lipoprotein lipase; LCAT, lecithin cholesterol acyltransferase; HL, hepatic lipase.

I will highlight a refinement and revision to these concepts of the role of HL in HDL metabolism. Through the work of a number of laboratories, it has been shown that extensive hydrolysis of HDL by HL in vitro can result in the formation of a cholesterol-poor phospholipid–apolipoprotein AI complex called preb-HDL. Preb-HDL is thought to be the most active HDL fraction in accepting cholesterol from cells, but it may also be a rapidly catabolized form of apolipoprotein AI. Cheung et al. [13] have purified a lipase inhibitor from the serum of patients with nephrotic syndrome and characterized it as preb-HDL. These combined observations suggest that a product of HL action, preb-HDL, is also an inhibitor of HL, which would feed back on HL activity and regulate the further production of preb-HDL (Fig. 4). Since LCAT can use preb-HDL as a substrate, the balance between the concentrations of HDL subfractions including preb-HDL would be determined by the competition of LCAT, HL and mechanisms that enrich HDL in TG. In a state such as nephrotic syndrome, it appears that LCAT is relatively ineffective and the net result is an accumulation of preb-HDL that inhibits both LPL and HL contributing to the hypertriglyceridemia common in this syndrome.

248

P.W. Connelly / Clinica Chimica Acta 286 (1999) 243 – 255

Fig. 4. Metabolic scheme showing the putative roles of HL and LCAT in the metabolism of preb-HDL with the inhibition of HL by preb-HDL.

5. Does HL have a ‘bridge’ function in the clearance of remnant lipoproteins? The role of HL in the clearance of chylomicron remnants and VLDL remnants is not well established. This is perhaps because there appear to be multiple steps in the clearance of remnant lipoproteins by the liver. These have been well summarized and critically evaluated by Cooper [14]. The current concepts consist of a two-step process beginning with the sequestration of remnant lipoproteins in the liver Space of Disse by binding to heparan, LDL-receptor related protein (LRP) and / or HL. The binding is dependent upon the particles containing apolipoprotein E or acquiring apolipoprotein E upon entering the Space of Disse. The sequestration step may be rapid with internalization of the remnant particle occurring as a slower, second step. It has been suggested that HL does not need to be enzymatically active to participate in the sequestration step and that when both the lipoprotein-binding function and the heparanbinding function of HL are intact, the enzyme can function as a ‘bridge’ [8,15]. Obtaining convincing evidence of this non-catalytic function in vivo is difficult. Huff et al. [16] have studied determinants of cholesteryl ester accumulation from VLDL uptake by a model hepatocyte, the HepG2 cell, which secretes and binds HL. Typical human VLDL requires a lipolytic processing step involving LPL before significant cholesteryl ester accumulation by HepG2 cells is observed. Once lipolyzed, the accumulation requires heparan. About 50% of the accumula-

P.W. Connelly / Clinica Chimica Acta 286 (1999) 243 – 255

249

tion of cholesteryl ester can be inhibited by the lipase inhibitor, tetrahydrolipstatin. Thus, in this model, about one-half of the uptake of lipoprotein can be attributed to the enzymatic activity of HL. Transgenic mice have also been used to study the effects of catalytically active and catalytically inactive HL. Dichek et al. [15] chose the apolipoprotein E knockout mouse, which accumulates remnant apolipoprotein B48-containing particles and expresses the mouse HL. Mouse HL is unusual in that about 50% of the enzyme is in plasma and 50% is bound to hepatocytes. An 8-fold overexpression of catalytically inactive human HL reduced the concentration of plasma cholesterol from 13.3 to 6.36 mmol / l, with the major part of the reduction due to a lowering of VLDL cholesterol concentrations. A 25-fold overexpression of normal human HL reduced the concentration of plasma cholesterol to 6.78 mmol / l. These experiments show that, under supraphysiological conditions, the known lipoprotein- and heparan-binding properties of HL can be used to reduce plasma remnant lipoprotein concentrations independent of apolipoprotein E. A greater challenge for the future lies in demonstrating an effect of catalytically inactive HL when expressed at levels approximating the physiologically relevant concentrations.

6. Human and mouse knock-outs of HL: redundancy or measuring the wrong phenotype? We have studied a kindred with HL deficiency due to compound heterozygosity for mutations in HL [17,18]. The mutation S267F (in which the wild-type serine is replaced by phenylalanine) results in an inactive enzyme, while the mutation T383M (in which the wild-type threonine is replaced by methionine) results in an enzyme that is poorly secreted from cells and has a low activity [19]. No other isolated HL deficiency of this magnitude has been reported in humans, making detection of complete HL deficiency one of the rarest of lipoprotein disorders. Biochemical data for the proband (B1) and his siblings are shown in Table 1. When OHLD extended family members (excluding unrelated spouses) were classified according to the presence or absence of S267F and T383M, there were four categories of subjects: normal (n 5 3), simple T383M heterozygotes (n 5 5), simple S267F heterozygotes (n 5 4) and compound heterozygotes (n 5 3) [18]. Compared with the unaffected family members, the HL-deficient subjects had no significant differences in VLDL cholesterol, VLDL triglyceride or LDL cholesterol. There were subjects in OHLD who were not HL deficient (e.g. B4) and who had the normal HL sequence at codons 267 and 383, but who had an abnormal lipoprotein phenotype. This suggests that a second dyslipidemia,

P.W. Connelly / Clinica Chimica Acta 286 (1999) 243 – 255

250

Table 1 Lipid and lipoprotein concentrations in the OHLD proband and siblings a

Sex Age HL genotype TC (mmol/l) TG (mmol/l) VLDL-C VLDL-TG LDL-C LDL-TG HDL-C HDL-TG ApoAI (g/l) ApoB (g/l) HL LPL

B1

B2

B3

B4

B5

B6

M 51 F267/M383 8.3 8.9 5.5 5.8 2.0 1.4 0.8 0.6 1.59 1.92 0 9.4

M 53 F267/M383 7.1 5.7 2.1 3.1 3.3 2.0 1.7 0.7 1.96 1.92 0 13.8

M 62 F267/M383 6.2 2.7 0.7 0.8 3.5 1.1 2.1 0.6 2.02 1.42 0 16.0

M 60 S267/T383 6.8 3.2 1.5 2.7 3.8 0.4 1.6 0.2 1.83 1.51 8.6 11.5

F 51 S267/M383 8.1 10.4 4.7 9.6 2.6 1.3 1.14 0.97 2.10 2.13 3.2 12.4

M 65 F267/T383 5.7 2.3 0.9 1.1 3.5 0.8 1.3 0.3 1.51 1.46 2.6 7.1

a HL, hepatic lipase; LDL, low density lipoprotein; HDL, high density lipoprotein; VLDL, very low density lipoprotein; HL, hepatic lipase; LPL, lipoprotein lipase; PL, phospholipid; TG, triglyceride; OHLD, Ontario Hepatic Lipase Deficient kindred; apo, apolipoprotein; F267, phenylalanine267; S267, serine 267; M383, methionine 383; T383, threonine 383; C, cholesterol; TG, triglyceride. All lipid concentrations in mmol / l. Enzyme activities expressed as mmol free fatty acids released / ml plasma per h.

unrelated to HL deficiency, segregates through OHLD, particularly through the ‘unaffected’ branch comprising the offspring of B4. The small number of subjects available for study precludes a more definitive analysis. All compound heterozygotes for HL deficiency have hyperalphatriglyceridemia and hyperbetatriglyceridemia when they are defined as levels of LDL-TG or HDL-TG exceeding the 95th percentile for age and sex. No subject other than the three compound heterozygotes had detectable b-VLDL, the presence of which appeared to be directly related to degree of hyperlipidemia. Therefore, the lipoprotein fractions that varied significantly according to degree of HL deficiency were LDL-TG and HDL-TG. The phenotype of heterozygotes involves varying degrees of TG-enrichment of LDL and / or HDL. S267F appears to be associated with TG-rich LDL and HDL in older female heterozygotes. T383M appears to be associated with TG-rich LDL and HDL, but only in the oldest heterozygotes studied. Thus, age and gender in heterozygotes may affect phenotypes, such as HL activity, and / or the TG-content of LDL and HDL. Plasma post-heparin HL activities varied significantly according to HL

P.W. Connelly / Clinica Chimica Acta 286 (1999) 243 – 255

251

genotypic class with activities highest in normal subjects, lower in T383M heterozygotes, lower still in S267F heterozygotes, and absent in compound heterozygotes. This is consistent with the deficiency imparted by S267F being clinically more severe than the deficiency imparted by T383M. Among T383M heterozygotes, there was considerable variability in HL activity. For example, two young male T383M heterozygotes (C2 and C3) had normal HL activity. This suggests that the HL deficiency imparted by T383M might be modulated by secondary factors (e.g. age, gender or other variants of other genes). Since T383M has reduced, but detectable in vitro activity, it is possible that simple T383M heterozygotes may in most clinical situations have adequate lipolytic activity and no phenotypic abnormality. However, in the presence of secondary factors that modulate HL (e.g. estrogen or alcohol) and thus impair expression of an HL mutant with marginal activity, there could be a significant impact at the clinical level.

7. Heart disease in OHLD family members The three compound heterozygotes (B1, B2 and B3) have had coronary artery disease. As mentioned, the proband B1 had a fatal MI at age 51. B2 was symptomatic with angina at age 50 with severe multiple vessel coronary atherosclerosis requiring coronary angioplasty at age 53. B2 then suffered a severe MI at age 58 despite treatment with lovastatin. Repeat angiography in B2 in 1992 revealed severe diffuse multiple vessel occlusive disease. Treatment includes lovastatin 80 mg daily. B3 had a history of angina since his mid-50s with three-vessel coronary bypass surgery at age 57. B5, a sister of the proband who was a T383M simple heterozygote, developed angina at age 51 with subsequent two-vessel coronary bypass surgery. None of the simple heterozygote children of the proband or the proband’s siblings has yet had symptoms of coronary artery disease; however, the eldest studied was under 40 years of age. Thus, the evidence from OHLD is consistent with other data that suggest that HL plays a significant role in the conversion of VLDL to LDL of normal composition. The earlier steps in the conversion of VLDL to VLDL remnants are shared with LPL, and HL is not obligatory. HL appears to have a major role in the conversion of apoE-containing VLDL remnants and IDL into particles of LDL density without apoE. While the hydrolysis of PL and TG by HL is the basis for the change in particle size and density, the major functional change that occurs is the conversion of particles from a form where apoE is the major ligand for receptor-dependent clearance of the particles from the circulation to a form where apoB100 is the major ligand for receptor-dependent clearance [20].

252

P.W. Connelly / Clinica Chimica Acta 286 (1999) 243 – 255

8. ApoAI- and AII-containing lipoproteins in HL deficiency The conversion of VLDL to LDL requires the loss of TG and surface constituents consisting of apolipoproteins C and E, PL and cholesterol [21]. This ‘excess’ surface material is transferred to the HDL fraction [21,22] and preferentially contributes to the mass of HDL 2 . HDL can be separated into particles containing apoAI and apoAII (AI,AII) and particles containing apoAI without apoAII (AI, desAII) [23]. The AI,AII and AI, des AII are equal acceptors of TG and PL during the metabolism of post-prandial lipoproteins resulting in the increase in the concentration of HDL 2 . In the compound heterozygote HL-deficient subject B2, the HDL 2 has an abnormally high apoAII content, while the HDL 3 has a normal AII content (Fig. 5). Both the HDL 2 and HDL 3 are TG rich. The HDL 2 mass has been observed to increase post-prandially with a reciprocal decrease in the HDL 3 mass. This is consistent with AI,AII particles undergoing sufficient TG-enrichment to shift their density to the HDL 2 range and with HL being of primary importance in the removal of TG and the subsequent density shift of AI,AII particles to the HDL 3 range. Furthermore, in subject B2, the AI,AII HDL particles are abnormally distributed (constituting about 70% of the HDL 2 compared with 10% of the HDL 2 in normal subjects). HL acts preferentially on the TG-rich AI,AII HDL particles, converting AI,AII HDL to its post-absorption composition and density [24,25], consistent with the experimental results of Mowri and co-workers [24–26] summarized in Fig. 6. In contrast to the studies in human, a mouse model of HL deficiency is characterized by a mild dyslipidemia [27]. HDL subfraction analysis has not been done.

Fig. 5. The distribution of apolipoprotein AII between HDL subfractions in control subjects and a single HL-deficient patient followed in 1991, 1992 and 1993.

P.W. Connelly / Clinica Chimica Acta 286 (1999) 243 – 255

253

Fig. 6. Metabolic scheme for the effect of post-prandial lipemia on the apolipoprotein-specific subfractions of HDL. Abbreviations: AI /AII, lipoprotein AI, AII; AI, lipoprotein AI des AII. Other abbreviations as in Fig. 2.

9. Sunrise or sunset for HL? The aggregate of the data available to date are consistent with HL having a modest catalytic role in determining the plasma concentration of remnant lipoproteins. However, the association of lower HL activity with higher remnant lipoprotein concentration appears to require an independent cause for elevation of VLDL and / or VLDL remnant lipoproteins. A significant amount of data point to HL playing a major role in HDL phospholipid and TG metabolism with the lipoprotein subclass defined as apolipoprotein AI, AII particles being a preferred substrate. The future of HL looks promising for application of molecular techniques to identify the relationship of enzyme structural elements to function. The use of chimeric constructs, hybrids of HL and LPL, can be anticipated to lead to a new understanding of both enzymes. A major issue to be addressed, which was not covered here, is why is HL expression regulated by hormones? It is well established that hormones such as estrogen reduce HL expression, yet there is no obvious metabolic necessity for this response. Understanding the hormonal regulation of HL is likely to provide new insight into its biological function(s). No animal models of HL deficiency have addressed the potentially important role of HL in lipoprotein PE clearance. Is PE metabolism abnormal in HL knock-out mice or is there a compensatory mechanism or redundancy in this pathway? Finally, is there a special role for HL in the delivery of PE and PC containing linoleic acid and arachidonic acid to tissues? These PL species have unique metabolic fates and are also susceptible to oxidation. Can the benefit of HL be due to the efficient partitioning of these PL species to the liver?

254

P.W. Connelly / Clinica Chimica Acta 286 (1999) 243 – 255

It is hoped that this brief summary of the enzyme HL will stimulate the development of novel concepts about the role HL. Perhaps the sun will rise on understanding the function(s) of HL beyond the delivery of HDL cholesterol to cells.

Acknowledgements The invaluable contributions of Dr J. Alick Little, Dr Robert A. Hegele, Camilla Vezina, Graham Maguire and Maureen Lee are gratefully acknowledged.

References [1] Lipoprotein lipase and hepatic lipase. In: Jackson RL, Boyer P, editors, The enzymes, New York: Academic Press, 1983, pp. 141–81. [2] Stahnke G, Sprengel R, Augustin J, Will H. Human hepatic triglyceride lipase: cDNA cloning, amino acid sequence and expression in a cultured cell line. Differentiation 1987;35:45–52. [3] Komaromy MC, Schotz MC. Cloning of rat hepatic lipase cDNA: evidence for a lipase gene family. Proc Natl Acad Sci USA 1987;84:1526–30. [4] Hide WA, Chan L, Li WH. Structure and evolution of the lipase superfamily. J Lipid Res 1992;33:167–78. [5] Wolle J, Jansen H, Smith LC, Chan L. Functional role of N-linked glycosylation in human hepatic lipase: asparagine-56 is important for both enzyme activity and secretion. J Lipid Res 1993;34:2169–76. [6] Hill JS, Davis RC, Yang D, Wen J, Philo JS, Poon PH, Phillips ML, Kempner ES, Wong H. Human hepatic lipase subunit structure determination. J Biol Chem 1996;271:22931–6. [7] Dugi KA, Dichek HL, Talley GD, Brewer Jr. HB, Santamarina-Fojo S. Human lipoprotein lipase: the loop covering the catalytic site is essential for interaction with lipid substrates. J Biol Chem 1992;267:25086–91. [8] Kobayashi J, Applebaum-Bowden D, Dugi KA, Brown DR, Kashyap VS, Parrott C, Duarte C, Maeda N, Santamarina-Fojo S. Analysis of protein structure-function in vivo. Adenovirusmediated transfer of lipase lid mutants in hepatic lipase-deficient mice. J Biol Chem 1996;271:26296–301. [9] Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 1997;25:3389–402. [10] Landin B, Nilsson A, Twu JS, Schotz MC. A role for hepatic lipase in chylomicron and high density lipoprotein phospholipid metabolism. J Lipid Res 1984;25:559–63. [11] Nilsson A, Landin B, Schotz MC. Hydrolysis of chylomicron arachidonate and linoleate ester bonds by lipoprotein lipase and hepatic lipase. J Lipid Res 1987;28:510–7. [12] Little JA, Connelly PW. Familial hepatic lipase deficiency. Adv Exp Med Biol 1986;201:253–60.

P.W. Connelly / Clinica Chimica Acta 286 (1999) 243 – 255

255

[13] Cheung AK, Parker CJ, Ren K, Iverius PH. Increased lipase inhibition in uremia: identification of pre-beta-HDL as a major inhibitor in normal and uremic plasma. Kidney Int 1996;49:1360–71. [14] Cooper AD. Hepatic uptake of chylomicron remnants. J Lipid Res 1997;38:2173–92. [15] Dichek HL, Brecht W, Fan J, Ji ZS, McCormick SP, Akeefe H, Conzo L, Sanan DA, Weisgraber KH, Young SG et al. Overexpression of hepatic lipase in transgenic mice decreases apolipoprotein B-containing and high density lipoproteins. Evidence that hepatic lipase acts as a ligand for lipoprotein uptake. J Biol Chem 1998;273:1896–903. [16] Huff MW, Miller DB, Wolfe BM, Connelly PW, Sawyez CG. Uptake of hypertriglyceridemic very low density lipoproteins and their remnants by HepG2 cells: the role of lipoprotein lipase, hepatic triglyceride lipase, and cell surface proteoglycans. J Lipid Res 1997;38:1318– 33. [17] Connelly PW, Maguire GF, Lee M, Little JA. Plasma lipoproteins in familial hepatic lipase deficiency. Arteriosclerosis 1990;10:40–8. [18] Hegele RA, Little JA, Vezina C, Maguire GF, Tu L, Wolever TS, Jenkins DJ, Connelly PW. Hepatic lipase deficiency. Clinical, biochemical, and molecular genetic characteristics. Arterioscler Thromb 1993;13:720–8. [19] Durstenfeld A, Ben-Zeev O, Reue K, Stahnke G, Doolittle MH. Molecular characterization of human hepatic lipase deficiency. In vitro expression of two naturally occurring mutations. Arterioscler Thromb 1994;14:381–5. [20] Rubinstein A, Gibson JC, Paterniti JRJ, Kakis G, Little A, Ginsberg HN, Brown WV. Effect of heparin-induced lipolysis on the distribution of apolipoprotein E among lipoprotein subclasses. Studies with patients deficient in hepatic triglyceride lipase and lipoprotein lipase. J Clin Invest 1985;75:710–21. [21] Tall AR. Metabolic and genetic control of HDL cholesterol levels. J Intern Med 1992;231:661–8. [22] Deckelbaum RJ, Bramkrishnan R, Eisenberg S, Olivecrona T, Bengtsson-Olivecrona G. Triacylglycerol and phospholipid hydrolysis in human plasma lipoproteins: role of lipoprotein and hepatic lipase. Biochemistry 1992;31:8544–51. [23] Cheung MC, Albers JJ. Characterization of lipoprotein particles isolated by immunoaffinity chromatography: particles containing A-I and A-II and particles containing A-I but no A-II. J Biol Chem 1984;259:12201–9. [24] Mowri HO, Patsch W, Smith LC, Gotto Jr. AM, Patsch JR. Different reactivities of high density lipoprotein 2 subfractions with hepatic lipase. J Lipid Res 1992;33:1269–79. [25] Mowri HO, Patsch JR, Ritsch A, Foger B, Brown S, Patsch W. High density lipoproteins with differing apolipoproteins: relationships to postprandial lipemia, cholesteryl ester transfer protein, and activities of lipoprotein lipase, hepatic lipase, and lecithin: cholesterol acyltransferase. J Lipid Res 1994;35:291–300. [26] Mowri HO, Patsch JR, Gotto Jr. AM, Patsch W. Apolipoprotein A-II influences the substrate properties of human HDL 2 and HDL 3 for hepatic lipase. Arterioscler Thromb Vasc Biol 1996;16:755–62. [27] Homanics GE, de Silva HV, Osada J, Zhang SH, Wong H, Borensztajn J, Maeda N. Mild dyslipidemia in mice following targeted inactivation of the hepatic lipase gene. J Biol Chem 1995;270:2974–80.