Severe Hypoalphalipoproteinemia in Mice Expressing Human Hepatic Lipase Deficient in Binding to Heparan Sulfate Proteoglycan*

Unlike human hepatic lipase (hHL) that is mainly cell surface-anchored via binding to heparan sulfate proteoglycans (HSPG), mouse HL (mHL) has a low affinity to HSPG and thus is largely blood-borne. The reduced HSPG binding of mHL is attributable to the C-terminal amino acids. To determine the functions of HSPG binding of hHL in vivo, we created adenovirus vectors encoding hHL or a chimeric protein (designated hHLmt) in which the C-terminal HSPG-binding sequences were replaced with the corresponding mouse sequences. Injecting hHLmt-expressing virus into C57BL/6J mice (1.8 × 1010 virus particles/mouse) resulted in a 3-fold increase in pre-heparin HL activity, whereas infection with an identical dose of hHL virus did not change pre-heparin HL activity. In hHLmt-expressing mice, the concentration of total cholesterol and phospholipids was inversely related to the hHL activity in pre-heparin plasma in a dose- and time-dependent manner, and the decrease was mainly attributable to high density lipoproteins (HDL) cholesterol and HDL phospholipids. The expression of hHL exhibited no change in plasma total cholesterol or phospholipid levels as compared with control mice infected with luciferase or injected with saline. The reduced HDL lipids in the hHLmt-expressing mice were accompanied by markedly decreased plasma and hepatic apolipoprotein (apo) A-I. In primary hepatocytes isolated from hHLmt-expressing mice, the concentration of cell-associated and secreted apoA-I was decreased by 2–3-fold as compared with hepatocytes isolated from control mice, whereas the levels of apoB and apoE were unaltered. Infection of primary hepatocytes with hHLmt virus ex vivo also resulted in reduced apoA-I secretion but had no effect on cell-associated apoA-I. These results suggest that expression of HSPG binding-deficient hHL has a profound HDL-lowering effect.

Low circulating concentrations of high density lipoproteins (HDL) 1 are the most common lipoprotein abnormality associated with coronary artery disease. Thus, HDL are considered as anti-atherothrombotic lipoproteins with cardioprotective properties. Proposed protective roles attributed to HDL include reverse cholesterol transport from extrahepatic tissues to the liver (1), antioxidant effects against oxidation of low density lipoproteins (2), inhibition of inflammation through actions of the associated protein paraoxonase (3), and enhancement of the production of nitric oxide (4). Formation, remodeling, and catabolism of HDL are achieved through multiple pathways and are regulated by a variety of proteins. In addition to the biosynthesis of apolipoprotein (apo) A-I and apoA-II that are the major structural components of HDL, the metabolism of HDL also involves the function of cell surface membrane-bound proteins such as ATP-binding cassette transporter 1 (ABCA1) (5) and scavenger receptor class B type 1 (6), as well as blood-borne proteins including cholesteryl ester transfer protein (7), phospholipid transfer protein (PLTP) (8), and lecithin:cholesterol acyl transferase (LCAT) (9). Moreover, several lipases, existing as both cell surface-anchored and blood-borne forms, play a pivotal role in the remodeling of HDL. They include endothelial lipase (EL), lipoprotein lipase (LPL), and hepatic lipase (HL) (10).
Human HL (hHL), a heparin-binding protein, is a member of the superfamily of lipases and phospholipases (EC 3.1.1.3) (11)(12)(13) that includes pancreatic lipase, LPL (14), EL (15,16), and the recently cloned lipase-H (17). Synthesized in the liver, hHL is secreted and bound to hepatocytes and hepatic endothelial surfaces. In humans, plasma hHL activity can be increased by 1,000-fold upon infusion of heparin that disrupts hHL binding to heparan sulfate proteoglycans (HSPG) (18). Active hHL exists as a homodimer (19) and has a broad substrate specificity, catalyzing the hydrolysis of fatty acyl chains at the sn-1 position of phospholipids and of mono-, di-, and triacylglycerol that are associated with a variety of lipoproteins including HDL (20 -24). The dual phospholipase and triglyceride lipase activity of hHL is in contrast to LPL and EL that are predominantly triglyceride lipase and phospholipase, respectively (25). In addition to its lipolysis activity, HL may also play a role in facilitating the binding and * This work was supported by Grant-in Aids of the Heart and Stroke Foundation of Ontario (to Z. Y.). Portions of this work were presented at the 5th Conference on Arteriosclerosis, Thrombosis, and Vascular Biology of the American Heart Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
¶ A recipient of a studentship from the Heart and Stroke Foundation of Canada.
ʈ A recipient of a studentship from the Ontario Graduate Scholarship. ¶ ¶ A recipient of the Career Investigator Award from the Heart and Stroke Foundation of Canada.
ʈʈ A Career Investigator of the Heart and Stroke Foundation of Canada. To whom correspondence should be addressed. Tel.: 613-798-5555 (ext. 18711); Fax: 613-761-5281; E-mail: zyao@ottawaheart.ca. uptake of lipoproteins (26 -28) as well as in selective uptake of cholesteryl esters from a variety of lipoproteins (29 -31). These non-lipolysis roles of hHL are thought to occur through bridging lipoproteins (to which HL binds via hydrophobic interactions or protein-protein interactions (32)) to cell surface receptors. Studies with chimeric proteins have demonstrated that the C-terminal domain of HL, like that of LPL, governs its affinity to heparin (33,34). Using a chimeric lipase (hHLmt) expressed in a cell culture system, we have demonstrated that a region encompassing the C-terminal 70 amino acids of hHL confers high affinity to HSPG (35).
The physiological significance of hHL binding to HSPG, especially with respect to its lipolysis activity, remains unclear. Previous in vitro and cell culture studies have demonstrated that hHL can be displaced from purified HSPG (36) and cell surfaces (37) by exogenous HDL and by apoA-I but not by other lipoproteins. The displacement of hHL resulted in increased hydrolysis of very low density lipoproteins. The plasma levels of cholesterol associated with low density lipoproteins are shown to be associated closely with the pre-heparin activities of HL and LPL (38), suggesting that the blood-borne lipase activity may play a role in the metabolism of apoB-containing lipoproteins. Thus, the possibility exists that the in vivo lipolysis function of HL may not require binding to HSPG on cell surfaces. In the current study, we have tested the hypothesis that increases in circulating hHL activity lead to enhanced catabolism of plasma lipoproteins. Expression of the HSPG bindingdeficient chimeric hHLmt in mice through adenovirus-mediated gene transfer resulted in elevated pre-heparin lipase activity and severe hypoalphalipoproteinemia in vivo.

EXPERIMENTAL PROCEDURES
Materials-Enzymes (KpnI, XbaI, PacI, PmeI, and T4 DNA ligase) used for subcloning were purchased from New England Biolabs (Mississauga, Canada). Triolein, cholesterol, fatty acid-free bovine serum albumin, fibronectin, fumed silica, heparin, and fetal bovine serum were purchased from Sigma. William's medium, Hepatozyme, penicillin, streptomycin, and Fungizone were purchased from Invitrogen. The Superose-6 columns and horseradish peroxidase-conjugated goat antirabbit IgG antibody were purchased from Amersham Biosciences. [ 3 H]Triolein was purchased from Dupont. The LIPO lipoprotein agarose gel electrophoresis system was purchased from Beckman Coulter. Rabbit polyclonal antibodies against mouse apoA-I, apoB, and apoE were purchased from Biodesign International (Saco, ME).
Generation of Adenoviruses-Adenoviruses encoding luciferase, wildtype hHL, and the chimeric hHLmt (35) were generated using the AdEasy adenoviral system (Q-Biogene, Carlsbad, CA) according to the manufacturer's instructions. Briefly, the respective cDNAs were subcloned into an E1/E3-deleted recombinant adenovirus vector backbone containing the cytomegalovirus promoter to generate pAd.luc, pAd.hHL, and pAd.hHLmt, respectively. The pAd plasmids were transfected into the 293 cells (American Type Culture Collection, Manassas, VA) cultured in Eagle's minimum essential medium containing 100 units/ml penicillin, 100 units/ml streptomycin, 250 ng/ml Fungizone, and 10% fetal bovine serum. Positive plaques containing adenoviruses encoding luciferase (Ad.luc), hHL (Ad.hHL), or hHLmt (Ad.hHLmt) were propagated using the 293 cells in confluent 150-mm dishes. The adenoviruses were purified by CsCl density gradient ultracentrifugation and dialyzed against phosphate-buffered saline. The desalted virus stock was adjusted to 4% sucrose, quantified by measuring the absorbance at 260 nm, and stored at Ϫ80°C prior to use.
Animals and Primary Hepatocyte Cultures-Female C57BL/6J mice (Charles River Laboratories, Wilmington, MA) at 6 -8 weeks old were maintained on a normal chow diet and a 12-h light/12-h dark cycle. Mice were injected with 200 l of saline or with up to 1.8 ϫ 10 10 virus particles (VP) encoding luciferase, hHL, or hHLmt via the tail vein. Blood was collected via the saphenous vein from 7-h fasted mice at various time points. One h later, mice were injected with heparin (500 units/kg) via the tail vein, and post-heparin blood was collected 5 min after heparin administration.
Primary hepatocytes were prepared from adenovirus-infected or uninfected mice according to previously described methods (39,40). The cells were plated at a density of 1-2 ϫ 10 6 cells/well in fibronectin-coated (25 g/well) 6-well plates with William's medium containing 100 units/ml penicillin, 100 units/ml streptomycin, 250 ng/ml Fungizone, and 10% fetal bovine serum. Six h after plating, cells were washed twice in serum-free William's medium and incubated in Hepatozyme medium for another 4 h. The conditioned media were collected and spun at 2,000 rpm for 10 min to remove cell debris. Fumed silica (12.5 mg) was mixed with 1 ml of media by incubation at 4°C for 16 h with continued rotating. The media were removed after centrifugation at 3,000 rpm for 5 min, and the fumed silica was washed twice with ice-cold phosphatebuffered saline. Following washes, a cell lysis/SDS-PAGE loading buffer (200 l of 10 mM Tris; 8 M urea; 2% SDS; 10% glycerol; 5% ␤-mercaptoethanol; 0.001% bromphenol blue; pH 8.3) was added, and samples were heated at 75°C for 20 min to elute bound proteins. The samples were centrifuged at 3,000 rpm for 5 min, and the supernatant was stored at Ϫ80°C prior to use. For the cell samples, the cell lysis/SDS-PAGE loading buffer (200 l) was mixed with the cells, and the mixture was stored at Ϫ80°C prior to use.
Ex vivo adenoviral infections of primary hepatocytes were achieved by adding adenovirus at a concentration of 20 VP/cell to the Hepatozyme medium for 24 h. Following infection, cells were washed twice with Hepatozyme medium and incubated in Hepatozyme medium for 4 h. Cells and media were collected and treated for apolipoprotein analysis as described above.
Plasma Lipid Analyses-Fresh plasma (3 l) were separated on 0.6% agarose gels for 30 min at 100 V of direct current in 1% sodium barbital, pH 8.6. Agarose gels were fixed for 30 min (with 54% ethanol; 19% acetic acid in double distilled H 2 O) and dried for 1 h at 65°C prior to staining with 7% Sudan black B in methanol. For separation of lipoproteins, pooled plasma (500 l) from saline, Ad.luc-, Ad.hHL-, or Ad.hHLmt-injected mice were loaded onto two in-series Superose-6 columns that had been equilibrated with phosphate-buffered saline and fractionated by FPLC. One-ml fractions were collected, and lipids were measured. Total cholesterol (TC) and phospholipid (PL) from plasma (2 l/measurement) or FPLC (30 l/measurement) were measured using commercial kits (Roche Applied Science) according to the manufacturer's instructions.
Protein Analyses-Plasma and primary hepatocyte samples were separated by SDS-PAGE (3-15% gel). Gels were transferred to nitrocellulose membranes, and the membranes were immunoblotted for different apoproteins. To detect plasma apoB, 2 l of plasma was used. Plasma was diluted 1:10,000 for detecting apoA-I.
Hepatic Lipase Activity Assay-The activity of HL from plasma was assayed using a [ 3 H]triolein emulsion as described previously (41).
Statistical Analyses-Data in which statistical values are provided were analyzed using the paired t test. Error bars on the data are Ϯ S.D.

Expression of hHLmt Increases Pre-heparin HL Activity-We
have previously generated a chimeric HL (hHLmt) by exchanging the C-terminal 70 amino acid residues with the corresponding sequence of mHL (Fig. 1A) and shown in vitro that the chimeric hHLmt retained the catalytic properties of hHL but shared the HSPG binding properties of mHL (35). To determine any functional differences between normal hHL and the HSPG binding-deficient chimeric hHLmt in vivo, we generated adenoviruses encoding hHL and hHLmt and infected them (1.8 ϫ 10 10 VP) into female C57BL/6J mice. Plasma HL activities were measured in both pre-and post-heparin plasma 7 days after infection, using [ 3 H]triolein as a substrate (Fig. 1B). The pre-heparin plasma HL activities in hHLmt-expressing mice were 3-fold higher than that in hHL-expressing mice (9.8 Ϯ 1.7 versus 3.1 Ϯ 0.5 mol FFA/h/ml plasma), indicating a reduced HSPG binding affinity of hHLmt. The pre-heparin plasma HL activity in hHL-expressing mice exhibited a background level similar to that in control mice injected with saline or infected with luciferase (Fig. 1B). The post-heparin plasma HL activities between hHL-and hHLmt-expressing mice were comparable (17.6 Ϯ 6.6 versus 25.4 Ϯ 7.6 mol FFA/h/ml plasma), suggesting similar amounts of the HL being expressed.
Expression of hHLmt Decreases HDL Lipids-Pre-heparin plasma from mice infected for 7 days with adenovirus encoding hHL or hHLmt (up to 1.8 ϫ 10 10 VP) were analyzed by agarose gel electrophoresis ( Fig. 2A). A marked reduction of lipid stain-ing for ␣-migrating lipoproteins was observed in mice expressing hHLmt in a dose-dependent manner. At the same doses of infection, mice expressing hHL displayed no changes in lipid staining of lipoproteins. Quantification of total cholesterol and phospholipids in the plasma of mice 7 days after infection (with 1.8 ϫ 10 10 VP) showed decreased total cholesterol (hHL, 127 Ϯ 58 mg/dl; hHLmt, 76 Ϯ 30 mg/dl) and phospholipids (hHL, 162 Ϯ 40 mg/dl; hHLmt, 63 Ϯ 43 mg/dl) as a result of the expression of the HSPG binding-deficient hHLmt.
Inverse Correlation between Plasma Total Cholesterol and Phospholipid Levels and Pre-heparin HL Activity-The concentrations of total cholesterol and phospholipid were contrasted between hHL-and hHLmt-expressing mice infected with respective viruses for up to 8 days. Both total cholesterol (Fig. 3A) and phospholipid (Fig. 3B) concentrations decreased to minimum at 4 days after infection, whereas the pre-heparin plasma hHLmt activity gradually increased to maximum (Fig. 3C). In contrast, there were no significant changes in plasma total cholesterol, phospholipids, or pre-heparin HL activity in mice expressing the wild-type HL throughout the entire post-infec-tion period examined. The average plasma total cholesterol (74.4 Ϯ 11.6 mg/dl), phospholipid levels (196.9 Ϯ 27.1 mg/dl), and pre-heparin HL activities Fractionation of lipoproteins in the fasted pre-heparin plasma by FPLC showed that the loss of total cholesterol and phospholipid was mainly attributable to decreased HDL. Data presented in Fig. 4 show lipid profiles of plasma samples pooled from three mice 7 days after infection with 1.8 ϫ 10 10 VP. The most prominent peak of total cholesterol in the plasma of hHL-expressing mice was for HDL (7.4 mg/dl), which was 5-fold lower (1.4 mg/dl) in the plasma of hHLmt-expressing mice (Fig. 4A). Likewise, the most prominent peak of phospholipid in the plasma of hHL-expressing mice was for HDL (12.0 mg/dl), and the level of HDL phospholipid in hHLmt-expressing mice was 5-fold lower (2.3 mg/dl) (Fig. 4B). These results suggest strongly that expression of the HSPG binding-deficient hHLmt has profoundly decreased plasma concentrations of HDL lipids. Expression of hHLmt Decreases Hepatic ApoA-I Secretion-We next determined whether the loss of HDL lipids is associated with changes in apoA-I, the major apolipoprotein constituent of HDL. Immunoblot analysis of pre-heparin plasma, obtained from mice infected with virus (1.8 ϫ 10 10 VP) encoding hHL or hHLmt, showed markedly diminished apoA-I in hHLmt-expressing mouse plasma 2 days after infection (Fig.  5A). There was no change in plasma apoA-I in mice infected with hHL or luciferase and in mice injected with saline. Semiquantitative analysis by scanning densitometry of the immunoblots showed comparable levels of apoA-I between hHL-expressing mice, luciferase-expressing mice, and saline-injected mice throughout the post-infection period (Fig. 5B).
Because the majority of the plasma apoA-I originates from the liver, we examined the effect of HL expression on the hepatic apoA-I levels in the infected mice. Immunoblot analysis of apoA-I present in total liver homogenates of three mice infected with 1.8 ϫ 10 10 VP revealed marked reduction of hepatic apoA-I in mice expressing hHLmt as compared with those expressing hHL (Fig. 6A). Scanning densitometry analysis of the hepatic apoA-I immunoblots showed a 10-fold decrease in mice expressing hHLmt (Fig. 6B). Thus, expression of HSPG binding-deficient hHLmt results in decreased apoA-I in the liver, which may underlie the reduced plasma apoA-I concentrations.
To ascertain that hHLmt expression indeed results in attenuated hepatic apoA-I expression and secretion, we pre-pared primary hepatocytes from mice that had been infected with virus encoding hHLmt or luciferase for 10 days. Cellassociated apoA-I and accumulation of apoA-I in the conditioned medium of cells cultured with a serum-free medium for 4 h was determined by immunoblot analysis. As shown in Fig. 7A, apoA-I levels in medium from hHLmt-infected hepatocytes were reduced by 3-fold (determined by scanning densitometry) as compared with medium from luciferase control hepatocytes. The decrease was specific to apoA-I as no appreciable changes were observed for apoE or apoB48. (The level of apoB100 was below detection.) The level of cell-associated apoA-I in hHLmt-expressing hepatocytes was reduced by 2-fold (determined by scanning densitometry) as compared with that in luciferase control cells (Fig. 7B). The hepatic apoE or apoB48 concentrations were not changed between hHLmt-expressing and luciferase control cells.
Reduced hepatic secretion of apoA-I was also observed with primary hepatocytes isolated from normal mice that were infected with virus encoding hHLmt ex vivo. One day after infection, the cells were incubated with serum-free medium for 4 h, and the accumulation of apoA-I at the end of incubation was determined by immunoblot analysis. As shown in Fig. 8A, accumulation of apoA-I in the conditioned medium of cells infected with hHLmt virus was decreased as compared with cells infected with either hHL or luciferase. There was no difference in accumulation of apoB48 in the medium between hHLmt-expressing cells and hHL-or luciferase-expressing cells. Unlike the primary hepatocytes isolated from mice infected with hHLmt in which hepatic apoA-I was reduced (Fig.  7B), the cell-associated apoA-I in cells transiently infected with hHLmt ex vivo did not show alterations as compared with cells infected with luciferase control (Fig. 8B). The in vivo and ex vivo data together suggest that the impaired hepatic apoA-I production may underlie the hypoalphalipoproteinemia effect of expression of the HSPG binding-defective hHLmt.

DISCUSSION
The present work attempts to examine the physiological significance of the cell surface anchorage of hHL in vivo. We have disrupted normal HSPG binding of hHL using recombinant DNA techniques and determined the effect of expression of the HSPG binding-deficient hHL (hHLmt) on lipoprotein metabolism in mice. The advantage of this animal model is that elevation of circulating HL levels in vivo is achieved without heparin administration that indiscriminately removes all HSPG-binding proteins from the endothelium surface. Our data have shown that the expression of the heparin bindingdeficient hHLmt results in a severe hypoalphalipoproteinemia phenotype in mice, as exemplified by markedly decreased HDL lipids and HDL proteins. These lipid and lipoprotein lowering effects of the HSPG binding-deficient hHLmt appear to be specific to HDL as apoB-containing lipoprotein levels were unaffected. The HDL lowering effect of hHLmt in mice is unlikely to be the artifact of an overly high level expression of the recombinant enzyme proteins because expression of the wildtype hHL at similar levels exerts no effect. Thus, the current studies have revealed an important functional significance for HSPG binding of hHL in the metabolism of lipoproteins, mainly HDL, that originate from the liver.
In humans, high plasma HL activity (measured using postheparin samples) is associated with reduced HDL cholesterol and small HDL particles (42). Conversely, genetic HL deficiency is associated with modestly elevated HDL cholesterol and large HDL particles (43,44). Reduced HDL lipid and HDL proteins observed in mice expressing the HSPG binding-deficient hHLmt suggest that the level of circulating HL may be catalytically more active than the cell surface-anchored HL in the catabolism of HDL. Factors that regulate the dissociation of HL from cell surfaces in vivo are unclear. We have recently observed that apoA-I and HDL can displace hHL from purified proteoglycans in vitro and from cultured cell surfaces (36,37) and that hHL is active only when it is free in solution. The present study with the HSPG binding-deficient hHLmt provides in vivo evidence that the catabolism of HDL may occur in the circulation by the blood-borne HL, whereas the cell surfacebound hHL is less involved. To date, no mutations within hHL have been reported that affect HSPG binding or affect HDL lipid and protein levels. A positive correlation has been suggested between the post-heparin hHL activity and the preheparin hHL activity (38). Thus, it is possible that the aforementioned inverse relationship between post-heparin hHL activity and plasma HDL may also reflect a relationship between pre-heparin hHL activity and HDL. It is noteworthy that a major difference in the experimental design between the present experiments and previous ones (for instance, Refs. 26, 45, and 46) is the low dose of adenoviruses (1.8 ϫ 10 10 VP) being used that resulted in an elevation of post-heparin hHL or hHLmt activity at maximum only 5-fold above control levels. The expression of recombinant HL activity at this low level would be of more physiological relevance than the massively increased HL activity (by 50 -100-fold) achieved in mice (26,45) or rabbits (46) by adenovirus infection or transgeneic technologies. These previous studies showed that hHL overexpression was invariably associated with reduced total cholesterol and phospholipid in plasma and HDL, indicating that HL and its activity play a major role in HDL metabolism. However, the possibility that excessive HL expression in the transgenic animals might render elevated free HL in circulation that in turn affects HDL metabolism was not examined previously.
An unexpected observation made in the present study is that decreased secretion of apoA-I from the hepatocytes of mice infected with virus encoding hHLmt (for 10 days) is accompanied by reduced hepatic apoA-I. The mechanism by which expression of the HSPG binding-deficient hHLmt decreases hepatic apoA-I production is unclear and remains to be defined. Reduction in hepatic apoA-I, however, was not observed with hepatocytes acutely infected with hHLmt ex vivo. Thus, the decreased hepatic apoA-I production appears to be a slow response to hHLmt expression. Whether or not there are extrahepatic factors induced by hHLmt expression that inhibit hepatic apoA-I synthesis remains to be determined. Nevertheless, decreased secretion of apoA-I from primary hepatocytes actually infected with virus ex vivo, in the face of normal cellular apoA-I (Fig. 8), suggests that the HDL-lowering effect of hHLmt observed in vivo is at least partly attributable to impaired hepatic secretion of apoA-I.
The severe hypoalphalipoproteinemia in mice infected with hHLmt may also suggest impaired formation of HDL in the plasma and/or accelerated HDL catabolism. Similar hypoalphalipoproteinemia phenotypes have been observed in other model systems such as lpl-deficient mice (47) and Pltp-deficient mice (48). However, major differences in lipoprotein profiles exist between the hHLmt-expressing mice and these knockout mice. For instance, in lpl-deficient mice, low HDL cholesterol is accompanied by severe hypertriglyceridemia, likely as a consequence of impaired lipolysis of triglyceride-rich lipoproteins (47). However in hHLmt-expressing mice, low HDL was associated with no apparent change in apoB-containing lipoproteins ( Fig. 2A), ruling out the possibility that hHLmt expression suppresses LPL-mediated lipolysis. It should be noted that unlike hHL, the post-heparin activity of which is negatively correlated with plasma HDL concentrations, post-heparin plasma LPL activity is directly correlated with HDL cholesterol levels (42). Expression of a heparin binding-deficient human LPL in mice results in elevated triglyceride and cholesterol associated with very low density lipoproteins (49). Thus, the mode of action is distinct between LPL and HL in regulating plasma lipoproteins including HDL. In Pltp-deficient mice that exhibit hypoalphalipoproteinemia as a result of hypercatabolism of HDL proteins, secretion of lipoproteins containing apoB is also diminished (50). The lack of an effect of hHLmt expression on apoB secretion (Figs. 7A and 8A) excludes the plausibility of altered PLTP activity in hHLmt-expressing mice. Naturally occurring mutations in other proteins involved in HDL metabolism, such as ABCA1 (51) and LCAT (52), have also been associated with hypoalphalipoproteinemia. The effect of hHLmt expression on the activity of ABCA1 and LCAT remains to be determined. Plasma factors that have an inhibitory effect on HL activity include apoC-I (53) and apoA-II (54). Whether or not hHLmt has altered structural information that renders the enzyme less sensitive to the inhibition of these apolipoproteins toward HDL catabolism also needs to be examined.
In summary, the present work shows that a relatively short term expression of the HSPG binding-deficient hHLmt results in severe hypoalphalipoproteinemia. Although genetic variation of the HL gene LIPC is an important source of variation in HDL levels in the general population (10), the relationship between HL and atherosclerosis is complex. Data support HL as being both pro-atherogenic and anti-atherogenic (55,56). Overexpression of LIPC in mice results in reduced atheroscle- FIG. 7. Expression of hHLmt decreases apoA-I secretion and cell-associated apoA-I in primary hepatocytes isolated from mice infected with hHLmt virus in vivo. Female C57BL/6J mice (6 -8 weeks old) were infected with virus (1.8 ϫ 10 10 VP) encoding luciferase (indicated by C) or hHLmt. Ten days after infection, primary hepatocytes were isolated from the infected mice and plated. Six h after initial plating, cells were washed and incubated with serum-free medium for 4 h. After incubation, media were collected, apolipoproteins were adsorbed to fumed silica and eluted, and cells were collected and lysed. A, the fumed silica-treated media samples were separated by SDS-PAGE (3-15% gel). Proteins were transferred to nitrocellulose and immunoblotted for mouse apoA-I, apoB, and apoE, respectively. B, cell samples were similarly separated by SDS-PAGE and analyzed by immunoblotting. rosis (22), whereas knockout of LIPC also shows reduced atherosclerosis in the APOE-null background (57). Since hypoalphalipoproteinemia is frequently associated with atherosclerosis, it will be of interest to determine whether or not a long term expression of hHLmt will pose a risk for atherosclerotic vascular diseases.