Correction of Hypertriglyceridemia and Impaired Fat Tolerance in Lipoprotein Lipase–Deficient Mice by Adenovirus-Mediated Expression of Human Lipoprotein Lipase
Abstract Humans homozygous or heterozygous for mutations in the lipoprotein lipase (LPL) gene demonstrate significant disturbances in plasma lipoproteins, including raised triglyceride (TG) and reduced HDL cholesterol levels. In this study we explored the feasibility of adenovirus-mediated gene replacement therapy for LPL deficiency. A total of 5×109 plaque-forming units (pfu) of an E1/E3–deleted adenovirus expressing either human LPL (Ad-LPL) or the bacterial β-galactosidase gene (Ad-LacZ) as a control were administered to mice heterozygous for targeted disruption in the LPL gene (n=57). Peak expression of total postheparin plasma LPL activity was observed at day 7 in Ad-LPL mice versus Ad-LacZ controls (834±133 vs 313±89 mU/mL, P<.01), and correlated with human-specific LPL activity (522±219 mU/mL) and mass (9214±782 ng/mL), a change that was significant to 14 and 42 days, respectively. At day 7, plasma TGs were significantly reduced relative to Ad-LacZ mice (0.17±0.07 vs 1.90±0.89 mmol/L, P<.01) but returned to endogenous levels by day 42. Ectopic liver expression of human LPL was confirmed by in situ hybridization analysis and from raised LPL activity and mass in liver homogenates. Analysis of plasma lipoprotein composition revealed a marked decrease in VLDL-derived TGs. Severely impaired oral and intravenous fat-load tolerance in LPL-deficient mice was subsequently corrected after Ad-LPL administration and closely paralleled that observed in wild-type mice. These findings suggest that liver-targeted, adenovirus-mediated LPL gene transfer offers an effective means for transient correction of altered lipoprotein metabolism and impaired fat tolerance due to LPL deficiency.
Reprint requests to Dr Michael R. Hayden, Department of Medical Genetics, 407-2125 E Mall, NCE Building, University of British Columbia, Vancouver, BC, Canada, V6T 1Z4.
- Received May 20, 1997.
- Accepted June 29, 1997.
Lipoprotein lipase is a critical enzyme in the catabolism of TG-rich CMs and VLDLs. Its pivotal role is most obvious in patients with complete LPL deficiency, who experience significant morbidity due to profound, chronic hypertriglyceridemia, often presenting in early infancy with colicky pain, hepatosplenomegaly, or failure to thrive. However, LPL deficiency may not be recognized until later childhood or even adulthood, with the development of abdominal pain, lactescent plasma, eruptive xanthomas, and lipemia retinalis.1 LPL deficiency occurs at the highest frequency in the French-Canadian population, in which the incidence in eastern Quebec is 1:5000, ≈100 to 200 times the estimated worldwide frequency of 1:106.2
Recently it has been appreciated that patients with mutations in the LPL gene that result in partial defects in LPL catalytic function are very common, occurring with a frequency between 5% and 7% in the general population.3 4 5 6 The clinical presentation may be evident only by marginally elevated TG levels in the nonstressed state, with profound hypertriglyceridemia triggered by factors such as normal pregnancy, obesity, or diabetes.7 8 Postprandial metabolic studies have been performed on individuals heterozygous for mutations in the LPL gene,9 10 demonstrating an “unmasking” of the lipolytic defect after a fat challenge and resulting in prolonged postprandial lipemia and significant disturbances in lipoprotein levels and composition.
Clearly, plasma lipoproteins represent one important risk factor for the development of coronary artery disease. The most obvious mechanism involving impaired LPL action is its impact on lipoprotein concentrations and particle composition, accompanied by raised TG and reduced levels of HDL-C in plasma. Lipoprotein profiles in humans homozygous11 and heterozygous4 for mutations in the LPL gene are consistent with an enhanced susceptibility toward atherosclerosis. In view of the high frequency of mutations in the LPL gene in the general population, it is suggested that LPL deficiency may engender a significant risk factor for hyperlipidemia and atherogenesis.
Conversely, increases in LPL activity by several mechanisms, including studies of transgenic mice that overexpress LPL, primarily in adipose, heart, and skeletal tissues, demonstrate a significantly improved lipid profile, including a reduction in plasma TG levels and a decreased TC/HDL-C ratio.12 13 14 15 Moreover, suppression of diet-induced atherosclerosis has been reported in LDL receptor–knockout mice overexpressing LPL.16 Similarly, administration of drugs (NO-1886 or fenofibrate)17 18 19 that promote the action of LPL result in lowered plasma TG levels, improved capacity to handle lipid loads, and increased HDL-C levels.
We hypothesized that the LPL gene, successfully delivered and expressed by an adenoviral vector, may be useful in correcting the lipolytic defects due to LPL deficiency. We have previously demonstrated in vitro that “the machinery” for secretion of active, functional LPL from typically nonexpressing, mature hepatic cells can be utilized after exogenous adenovirus-mediated LPL gene delivery20 and have also recently confirmed from studies in normal CD1 mice that the mature adult liver in vivo is able to efficiently express and appropriately process adenovirus-delivered LPL (unpublished data, 1997).
Encouraged by these preliminary studies, we have used a first-generation E1/E3–deleted adenoviral vector to evaluate the metabolic consequences and stability of liver-targeted LPL gene replacement in mice with a targeted disruption of the LPL gene.21 Complete deficiency of functional LPL protein is lethal in mice,21 22 presumably due to congestion of the peripheral and pulmonary circulation by large TG-rich lipoproteins. Mice heterozygous for LPL deficiency have an approximately threefold elevation in TG levels associated with decreased PHP LPL activity levels compared with their WT littermates. We adopted this model of murine heterozygous LPL deficiency to evaluate the effect on hypertriglyceridemia, lipoprotein profile, and tolerance to oral and intravenous fat loading in vivo by liver-targeted, adenovirus-mediated LPL gene transfer.
Recombinant Adenoviral Vectors
The construction and preparation of the recombinant adenoviruses Ad-LPL and Ad-LacZ have been described previously.20 23 In brief, they are based on an adenovirus serotype 5 (Ad5) backbone, containing a partial deletion in the E1 region and a complete deletion of E3. Ad-LPL carries the 1.6-kb cDNA of the human LPL gene and Ad-LacZ contains the bacterial β-galactosidase gene (LacZ), each fused with the simian virus 40 nuclear localization signal, each under control of the rous sarcoma virus–long-terminal repeat (RSV-LTR) and respectively inserted into the Ad5 E1–deleted region. The viruses were purified by two rounds of CsCl density centrifugation, followed by gel filtration (PD10, Pharmacia LKB Biotechnology Inc), and stored at −70°C in HEPES-buffered saline containing 10% glycerol. Each preparation of virus was titered by plaque assay on 293 cell monolayers.20
Murine Viral Infection and Sampling
A total of 57 male mice of a mixed C57BL/6 and 129 background, either heterozygous for targeted disruption in the LPL gene or WT littermates, were matched for age between 6 and 12 weeks, maintained on a normal chow diet, and then used in this study. All procedures involving experimental animals were performed in accordance with protocols from the Canadian Council on Animal Care and the University of British Columbia Animal Care Committee. Dose-response analysis predicted an effective in vivo dose of 5×109 pfu of purified recombinant adenovirus for intravenous injection (see “Results”). Therefore, 5×109 pfu of either Ad-LPL or Ad-LacZ were diluted to a 200-μL final volume in DMEM (GIBCO BRL) and infused intravenously through the tail vein to each mouse. Prior to blood sampling, the mice were fasted for 12 hours with free access to water. All blood was collected from the retro-orbital plexus by using heparin-coated capillary tubes (Scientific Products). To estimate LPL immunoreactive mass and activity, PHP was collected from mice following intravenous injection of 200 U/kg of sodium heparin (Sigma Chemical Co) via the tail vein. Blood was collected 10 minutes after heparin injection and placed immediately on ice, and plasma was separated and collected after microfuge centrifugation at 4°C. PHP samples were immediately frozen at −80°C. After animals were killed by cervical dislocation, liver, lung, and spleen tissues from both Ad-LPL–and Ad-LacZ–infected cohorts were placed in 10% formalin for ISH on days 7, 42, and 60 after vector administration. For day 7 liver samples only, 100 mg was placed on ice for immediate determination of LPL activity and mass from respective liver homogenates of Ad-LPL–(n=5) and Ad-LacZ–(n=5) infected animals.
Livers from both the Ad-LPL– and Ad-LacZ–treated animals were harvested 7 and 42 days after administration of each vector and at the end of the experiment on day 60. ISH was performed on tissue sections as follows. Paraffin-embedded tissue sections (3 mm) were cut onto Superfrost glass slides (Fisher), baked, dewaxed, and rehydrated in graded alcohols. The tissue was permeabilized with HCl, 2× SSC, and 1 mg/mL of proteinase K. Following dehydration in graded alcohols and drying, the hybridization solution containing a digoxigenin-labeled LPL strand–specific riboprobe (sense or antisense strand, Boehringer Mannheim) derived from a 1.6-kb human LPL cDNA sequence was applied. Siliconized coverslips were mounted and sealed with rubber cement. Slides were denatured and incubated overnight at 42°C. A posthybridization wash with 50% formamide overnight was followed by 2× SSC washes and blocked with lamb serum and antibody application (sheep anti-digoxigenin polyclonal antibody conjugated to alkaline phosphatase, Boehringer Mannheim). The phosphatase was then activated in a buffer of pH 9.5, and enzymatic development of substrate (nitroblue tetrazolium/5-bromo-4-chloro-3-indoyl phosphate, Sigma) was performed for 48 hours. Slides were subsequently counterstained with carmalum and examined quantitatively for reaction product by light microscopy.
Measurement of LPL Mass and Activity
PHP was obtained as described above and used specifically for measurement of LPL immunoreactive mass and activity. Collected liver tissues were homogenized in 9 volumes of extraction buffer and processed for LPL mass and activity analysis as previously described.20 The plasma samples and supernatants from liver tissue homogenates were stored separately at −80°C prior to analysis. LPL mass was measured by an ELISA method based on two monoclonal antibodies, 5F9 and 5D2, raised against purified bovine milk LPL.24 These monoclonal antibodies against LPL were generous gifts from Dr John Brunzell (University of Washington, Seattle). The 5D2 monoclonal antibody recognizes an epitope located at residue 400 of human LPL but does not recognize mouse LPL, whereas 5F9 recognizes an unknown epitope. 5F9 was routinely coated in 96-well plates to serve as the capture antibody, whereas horseradish peroxidase–conjugated 5D2 served as the detection antibody in a sandwich ELISA to assess total LPL immunoreactive mass.20 21 22 23 24 Total LPL activity in PHP plasma was measured in duplicate by using a [3H]triolein emulsion substrate as previously described.20 21 22 23 24 25 Human-specific LPL activity was obtained after a 2-hour, 4°C preincubation of plasma with the human-specific 5D2 monoclonal antibody, diluted 1:8. This approach inhibits human LPL activity by at least 85% but does not alter endogenous mouse LPL activity.20 21 22 23 24 One milliunit (mU) of lipase activity is equivalent to 1 nmol free fatty acid released per minute at 37°C.
Plasma Lipid and FPLC Analysis
To assess lipid levels, HDL-C was determined after LDL-C and VLDL-C were precipitated with an equal volume of 10% polyethylene glycol 8000 solution. TC and TG were determined by using commercial kits No. C236691 and No. 450032, respectively (Boehringer Mannheim). Plasma lipoproteins were separated by FPLC with two Superose 6 columns (Pharmacia LKB Biotechnology Inc) linked in series. Plasma was sampled and pooled from each mouse cohort at various time intervals. After 0.22 μm filtration, 200 μL of plasma was loaded onto the columns. The elution flow rate was 0.5 mL/min in a running buffer consisting of 0.15 mol/L NaCl, 1 mmol/L EDTA, and 0.02% NaN3, pH 8.2. Fractions of 0.5 mL were collected, and TC and TG contents were determined by enzymatic kit assay (Boehringer Mannheim) in 96-well plates.
Oral and Intravenous Fat-Tolerance Tests
On day 6 after Ad-LPL and Ad-LacZ infection, the mice were fasted for 12 hours. On day 7 three groups of mice, including WT uninfected controls (n=9), heterozygous Ad-LacZ–injected controls (n=9), and heterozygous Ad-LPL–injected mice (n=9), were given either 300 μL olive oil (Sigma) orally by gastric feeding tube (n=5 per cohort) or 250 μL of 20% Travamulsion (Clintec Nutrition Co, containing 20 g soybean oil and 1.2 g egg phosphatide per deciliter) intravenously via the tail vein (n=4 per cohort). Approximately 50 μL blood was collected retro-orbitally at the indicated times for a period of 24 hours. Plasma TG was measured enzymatically as described above.
Dose-Response Analysis of Ad-LPL Expression in WT Mice
To initially evaluate the ideal dosage and in vivo utility of our Ad-LPL vector, 12 WT littermates from the murine LPL gene–targeted kindred were evaluated for PHP LPL mass and activity according to varying pfu of administered Ad-LPL. The mice (n=3 per group) were intravenously injected with Ad-LPL at three different doses: 1×109, 2.5×109, or 5×109 pfu, or Ad-LacZ at 5×109 pfu. The dose-response relationship on day 7 according to both human-specific LPL activity and immunoreactive mass in PHP is shown in Fig 1⇓. Clearly, human-specific LPL mass and activity increased proportionally with increasing dosages of Ad-LPL, relative to negligible levels from the Ad-LacZ–injected control mice. Peak levels of LPL activity (388.7±68.0 mU/mL) and mass (7136.9±1193.6 ng/mL) were achieved in these WT mice with 5×109 pfu of Ad-LPL, which was therefore chosen as the maximum desirable dose for all subsequent mouse injections.
Hepatic expression of human LPL in vivo
We next systemically administered 5×109 pfu of Ad-LPL (n=9) or Ad-LacZ (n=9) by tail vein injection to mice heterozygous for targeted disruption of the LPL gene. Livers, lungs, and spleens from both Ad-LPL and Ad-LacZ animals were harvested 7 and 42 days after administration of each vector and at the end of the experiment on day 60. The adenovirus-mediated expression of the human LPL gene in mouse liver was initially confirmed on day 7 by ISH analysis using an LPL riboprobe (Fig 2B⇓), at which time it was obviously positive in the majority of hepatic cells. By comparison, negligible human LPL transcript was detected from either spleen or lung tissues on day 7, and no signal was detected with the sense control LPL riboprobe in any tissues (data not shown). No endogenous mouse LPL mRNA was detectable on day 7 in Ad-LacZ–infected mouse liver, confirming the riboprobe to be human-specific (Fig 2A⇓). On day 42 (Fig 2C⇓), LPL expression was significantly decreased and upon sacrifice on day 60 (Fig 2D⇓), no detectable Ad-LPL message was apparent in the liver.
Total lipase activity determined from the plasma and liver is due to the combination of HL and LPL. The respective activity for each enzyme is measured by performing the assay in the presence or absence of 1 mol/L NaCl, which, when present, differentially inhibits LPL but not HL activity. Thus, LPL activity is that portion of the lipolytic activity inhibited by 1 mol/L NaCl. As shown in the Table⇓, the total lipase activity from day 7 postinfection liver tissue homogenates of mice that received Ad-LPL was significantly elevated compared with Ad-LacZ controls (432.8±140.9 versus 188.7±17.3 mU/g liver, P=.02). The liver homogenates of both Ad-LacZ and Ad-LPL mice contained similar amounts of lipase activity not inhibited by 1 mol/L NaCl (P>.05) and represents HL. Therefore, the additional inhibited activity of 255.4±131.0 mU/g liver in the livers of mice that received Ad-LPL is consistent with human-specific LPL activity (P=.015). In association with this finding, human LPL immunoreactive mass from Ad-LPL–treated mice was 1306.9±814.4 ng/g liver versus undetectable LPL mass in Ad-LacZ controls (P=.02).
Detection of Human-Specific LPL in Mouse PHP
Ectopic expression of human LPL in the liver is clearly accessible to the plasma compartment, since intravenous injection of heparin resulted in a rapid release of human LPL into the bloodstream. As shown in Fig 3A⇓ and 3B⇓, on day 3 after Ad-LPL administration, both human LPL mass and activity in PHP could be detected (Fig 3B⇓), yet no significant change in total LPL activity (mouse plus human, Fig 3A⇓) versus Ad-LacZ controls was apparent. Subsequently, peak human LPL expression was achieved on day 7. Total LPL activity in PHP was 833.8±132.5 mU/mL in Ad-LPL mice, 2.7 times higher than Ad-LacZ controls, of which 70% was human specific. Human LPL immunoreactive mass was 9343±1974.7 ng/mL in PHP, 8.5 times higher than levels typically detected from normal human PHP measured in this laboratory by the same method (1100±100 ng/mL, n=10). Both human LPL activity and mass decreased to 393.7 mU/mL and 2600 ng/mL, respectively, on day 14 in accordance with a reduced total LPL activity of 573.2±304.9 mU/mL. The difference in total LPL activity in PHP was indistinguishable on or beyond day 28 between Ad-LPL–and Ad-LacZ–treated animals, owing to the disappearance of detectable human-specific LPL activity. However, human LPL immunoreactive mass remained significantly elevated at 478.0 ng/mL (P<.05) until at least 42 days after administration of Ad-LPL (Fig 3B⇓). By day 60, no detectable human LPL activity or protein was observed in PHP samples.
Correction of Hypertriglyceridemia in LPL +/− Mice
As a direct consequence of hepatic expression of human LPL, the moderate hypertriglyceridemia in heterozygous LPL +/− mice was corrected and inversely correlated with levels of PHP LPL (Fig 4A⇓). Plasma TG levels dropped sharply from 2.64 to 0.54 mmol/L 3 days after administration of Ad-LPL and decreased maximally to 0.17 mmol/L by day 7 (P<.001). These levels were approximately one fifth of uninfected WT levels (1.01±0.14, n=5). Peak expression of human LPL began to decline after day 7, and as a result, TG levels rose gradually, returning to the endogenous levels seen in Ad-LacZ control mice by days 42 and 60, correlating with the disappearance of human-specific LPL activity or mass. Plasma TG levels were unexpectedly reduced by ≈30% in Ad-LacZ mice 3 days after virus administration but stabilized thereafter. Surprisingly, plasma HDL-C levels were also reduced nearly 40% (from 1.21 to 0.79 mmol/L) by day 3 after Ad-LacZ injection yet remained unchanged at baseline after Ad-LPL administration (Fig 4B⇓). Similar changes in TC (Fig 4C⇓) likely reflect this alteration in HDL-C, since most of the plasma cholesterol in mice is carried in the HDL fraction.
As revealed by FPLC (Fig 5⇓), the plasma lipoprotein profile of mice heterozygous for targeted disruption of the LPL gene compared with their WT littermates is characterized by a much larger TG peak in the VLDL fraction (Fig 5A⇓ and 5B⇓). This peak was profoundly decreased 7 days after administration of Ad-LPL (Fig 5D⇓) relative to uninfected WT mice (Fig 5A⇓) or LPL +/− mice that received Ad-LacZ (Fig 5C⇓ and 5E⇓). This confirmed that the correction of hypertriglyceridemia by administration of Ad-LPL was due to an enormous reduction of CM/VLDL-derived TGs. The alteration of HDL-C in mice that received Ad-LPL, as analyzed by FPLC, was consistent with the levels of HDL-C as measured in plasma by the quantitative precipitation method. A broader and higher peak correlated with relative enrichment of a larger isoform in the HDL fraction and was evident by day 7 after Ad-LPL administration (Fig 5D⇓) but was resolved by day 60 (Fig 5F⇓). The nature of this larger HDL species, while compatible with increased LPL-mediated conversion to HDL2, was not identifiable by our current assay procedure and is under further investigation.
Correction of Impaired Fat Tolerance by Hepatic Expression of Human LPL
In preliminary experiments, we found that the response to orally fed fat was greatly impaired in heterozygous LPL mice compared with WT littermates (data not shown). Subsequently, both oral and intravenous fat-tolerance tests were performed on WT untreated or LPL +/− mice that received either Ad-LacZ or Ad-LPL (n=9, each group; Fig 6A⇓ and 6B⇓). As shown in Fig 6A⇓, after a bolus oral feeding of 0.3 mL olive oil, plasma TG levels in WT mice (n=5) increased and peaked at ≈2 hours (2.60 mmol/L) and returned to baseline levels at 4 hours. However, TG levels in LPL +/− mice that received Ad-LacZ (n=5) took a longer time to reach the peak, which was five times greater than that of the WT group (13.56 mmol/L at 4 hours). Although TG levels started to decline 6 hours after the oral fat load, levels throughout the 24-hour period of study remained significantly higher than the respective levels in either unifected WT or LPL. +/− mice treated with AD-LPL (n=5). When LPL +/− mice were injected intravenously with Ad-LPL, plasma TG levels on day 7 were greatly reduced compared with both uninfected WT and Ad-LacZ control mice. Therefore, although the response curve to an oral fat load paralleled that demonstrated by LPL +/− mice that received Ad-LacZ, postprandial TG levels were approximately 10-fold lower, nearing levels observed in WT mice.
Variability in the absorption of fat given orally is another factor that could influence response curves between WT and LPL +/− mice that received Ad-LacZ or Ad-LPL. Therefore, an intravenous fat-tolerance test, which bypasses the gastrointestinal absorption phase, was performed (n=4 in each cohort). Fig 6B⇑ confirms that the clearance of lipids injected intravenously via bolus infusion was significantly prolonged in LPL +/− mice administered Ad-LacZ compared with WT littermates. However, the clearance curve of Ad-LPL–treated heterozygous mice remarkably paralleled that observed in WT mice. Thus, the capacity to restore near-normal postprandial LPL–mediated hydrolysis of serum triacylglycerols is clearly apparent from these studies in heterozygous LPL mice after a single injection of Ad-LPL.
The present study demonstrates the therapeutic potential of adenovirus-mediated delivery of exogenous LPL to the liver in ultimately correcting the lipolytic defects associated with hypertriglyceridemia and impaired fat tolerance caused by heterozygous LPL deficiency in mice. Although transient expression is an intrinsic limitation of adenovirus-mediated gene transfer, a single intravenous injection of Ad-LPL successfully resulted in ectopic expression of functionally active human LPL for at least 14 days and expression of immunoreactive human LPL for at least 42 days. Fasting determinations of plasma TG levels remained significantly lowered for at least 1 month, returning to baseline measures by day 60. In the absence of measurable human LPL catalytic activity, this slow return of TGs to baseline levels may reflect in part the noncatalytic effect of LPL protein on the removal of lipoproteins from the circulation. In vitro studies have revealed that LPL may function as a “bridging” ligand to the LDL receptor, which serves to enhance the uptake of CM and VLDL remnants independent of its lipolytic activity.26 27
FPLC characterization of lipoproteins confirmed that the major contribution to significantly decreased plasma TG levels was derived from a reduction of TG content in the VLDL fraction due to the ectopic expression of human LPL by the liver. In contrast to the HDL-C–lowering effect associated with the Ad-LacZ vector, HDL-C levels remained at baseline after Ad-LPL administration. Presumably, overexpression of LPL appropriately compensated for the HDL-C–lowering effect, which possibly represents a nonspecific, transient adenovector-mediated effect on the hepatic or circulatory transport of TG-rich lipoproteins. FPLC analysis of plasma lipoproteins in mice that received Ad-LPL was consistent with a relative enrichment of a larger HDL isoform. The nature of this larger HDL species is compatible with enhanced LPL-mediated conversion to HDL2, a possibility that is in keeping with observations by Shimada et al15 in LPL-overexpressing transgenic mice, in which HDL2-C was increased 1.4-fold over nontransgenic controls.
Peak human LPL gene expression in the liver was noted by day 7 after parenteral adenovirus LPL infection. At this time, plasma TG levels in fasted animals decreased to approximately one tenth and one fifth of control levels as determined in heterozygous Ad-LacZ–injected mice and uninfected WT littermates, respectively. The magnitude of this drop was unexpected, since the increased mobilization and hepatic influx of fatty acids derived from hydrolysis of TG-rich lipoproteins normally stimulates increased hepatic VLDL-TG assembly and secretion back into the circulation.28 Conversely, transgenic animals overexpressing human LPL in extrahepatic tissues, such as adipose and muscle, hydrolyze TG locally in those tissues.12 13 Consequently, mobilized fatty acids are either resynthesized as TGs for storage in adipose tissue or utilized for energy consumption in muscle, thereby ultimately reducing plasma TG levels.
We have also demonstrated improved tolerance to oral and intravenous fat loading in LPL-deficient mice following Ad-LPL gene delivery. Humans spend ≈75% of their time in a postprandial state,29 which poses a more constant and serious challenge to the health of individuals with LPL deficiency. A fat-tolerance test provides an accepted physiological assessment of lipid clearance that may more accurately reflect disturbances in lipoprotein homeostasis compared with assessments performed in the fasting state.1 10 29 The amount of fat administered either orally or intravenously to the mice (300 or 50 mg per 25 g body weight, respectively) is proportional to 840 or 140 g consumed by a 70-kg human. Under such an extreme fat load, the clearance time of exogenous TGs was greatly prolonged in LPL-deficient mice relative to WT littermates. However, overexpressed liver-derived LPL in heterozygous mice that received Ad-LPL restored normal postprandial TG clearance, in keeping with the response elicited from uninfected WT mice. After ingestion of the oral fat load, the clearance time and curve were found to parallel those from Ad-LacZ–injected heterozygous mice. However, the Ad-LPL mice manifested 10-fold lower plasma TG levels in the range of normal, uninfected littermates. After the intravenous fat load, the clearance time and curve were completely normalized to those demonstrated by WT littermates. It is possible that the type of oral fat load administered (olive oil) and its different route of absorption relative to intravenous fat loading (Intralipid) could account for the difference in oral versus intravenous lipid clearance in Ad-LPL–infected mice relative to WT controls.
The relationship between LPL and atherogenesis is not mediated simply by the changes that occur in plasma lipoproteins. For example, expression of macrophage-derived LPL in the vessel wall may be proatherogenic.30 31 Studies of inbred murine strains have shown an association between high levels of LPL synthesis and secretion in macrophages, with increased susceptibility to atherosclerosis.32 When present in the arterial wall, LPL may promote binding and retention of LDL to the subendothelial matrix, where these lipoproteins may be converted to more atherogenic forms.33 The atherogenicity of LPL could depend on which cells are proportionally affected by increased LPL-mediated lipoprotein uptake; for example, liver cells or cells at the arterial wall.
Our studies have demonstrated that the metabolic effects of LPL in the regulation of plasma TG metabolism are not necessarily dependent on coordinated secretion from its constitutive peripheral tissue sources. Our findings also suggest that ectopically-expressed LPL from the liver can participate in plasma lipoprotein metabolism in a manner similar to its action at physiological sites in peripheral tissues. Furthermore, liver-expressed LPL will remove atherogenic TG-rich lipoproteins from the circulation, a finding theoretically compatible with ultimate protection against atherosclerosis. The demonstration of successful LPL gene expression and function resulting from adenoviral targeting of the murine host liver also supports the prospect for modifying alternate pathways of lipid metabolism by enhanced LPL expression. The feasibility of such a strategy has been previously demonstrated in mice presenting with an improved lipoprotein profile and decreased susceptibility to atherosclerosis on a genetic background created from crossing gene-targeted LDL receptor–deficient mice with LPL-overexpressing transgenic mice.16
The major limitations of currently developed adenoviral vectors, such as transience of expression and related immunogenicity, have been previously well described,34 35 but a number of recent developments appear promising for this vector system.36 37 Prevention of the adenovirus-induced immunogenic host response is the main obstacle to be overcome before this technology can render prolonged transgene expression and be applied therapeutically. Nonetheless, these studies have demonstrated that the level and duration of expression of the adenovirus-delivered human LPL transgene is sufficient to have significant effects on lipoprotein metabolism. The evidence provided by this study for correcting the lipolytic defects due to LPL deficiency by liver-targeted LPL gene delivery is encouraging and supports the assessment of other vector systems that might allow longer-term expression that is less immunogenic.
Selected Abbreviations and Acronyms
|FPLC||=||fast protein liquid chromatography|
|ISH||=||in situ hybridization|
This work was supported by grants to M.R. from the Medical Research Council (MRC) (Canada) and Gencell/Rhone-Poulenc Rorer. G.L. was supported by the Heart and Stroke Foundation of Canada. K.J.D.A.E. was supported by a National Science and Engineering Research Council (NSERC) studentship. M.R.H. is an established investigator of the British Columbia Children’s Hospital. We thank Dr John D. Brunzell (Department of Medicine, University of Washington, Seattle) for generously providing us with the 5D2 and 5F9 LPL monoclonal antibodies. We also appreciate the expert technical assistance provided by Adrienne Vair and Julie Chow (Department of Pathology, University of British Columbia).
Brunzell JD. Familial lipoprotein lipase deficiency and other causes of the chylomicronemia syndrome. In: Scriver CR, Beaudet AL, Sly D, Valle D, eds.Metabolic Basis of Inherited Disease. 7th ed. New York, NY: McGraw-Hill Book Co; 1995:1913-1932.
Gagne C, Brun LD, Julien P, Moorjani S, Lupien PJ. Primary lipoprotein lipase activity deficiency: clinical investigation of a French Canadian population. Can Med Assoc J. 1989;140:405-411.
Bijvoet S, Gagne E, Moorjani S, Gagne C, Henderson HE, Fruchart J-C, Dallongeville J, Alaupovic P, M. Prins, Kastelein JJ, Hayden MR. Alterations in plasma lipoproteins and apolipoproteins before the age of 40 in heterozygotes for lipoprotein lipase deficiency. J Lipid Res. 1996;37:640-650.
Mailly F, Fisher R.M, Nicaud V, Luong LA, Evans AE, Marques-Vidal P, Luc G, Arveiler D, Bard JM, Poirier O, Talmud PJ, Humphries SE. Association between the LPL-D9N mutation in the lipoprotein lipase gene and plasma lipid traits in myocardial infarction survivors from the ECTIM Study. Atherosclerosis. 1996;122:21-28.
Jukema JW, van Boven AJ, Groenmeijer B, Zwinderman AH, Reiber JH, Bruschke AV, Henneman JA, Molhoek GP, Bruin T, Jansen H, Gagne E, Hayden MR, Kastelein JJ. The Asp9Asn mutation in the lipoprotein lipase gene is associated with increased progression of coronary atherosclerosis. Circulation. 1996;94:1913-1918.
Ma Y, Liu M-S, Ginzinger D, Frohlich J, Brunzell JD, Hayden MR. Gene-environment interaction in the conversion of a mild to severe phenotype in a patient homozygous for a Ser172Cys mutation in the lipoprotein lipase gene. J Clin Invest. 1993;91:1953-1958.
Miesenbock G, Holzl B, Foger B, Brandstatter E, Paulweber B, Sandhofer F, Patsch JR. Heterozygous lipoprotein lipase deficiency due to a missense mutation as the cause of impaired triglyceride tolerance with multiple lipoprotein abnormalities. J Clin Invest. 1993;91:448-455.
Pimstone, SN, Clee SM, Gagne SE, Miao L, Zhang H, Stein EA, Hayden MR. A frequently occurring mutation in the lipoprotein lipase gene (Asn291Ser) results in altered postprandial chylomicron triglyceride and retinyl palmitate response in normolipidemic carriers. J Lipid Res. 1996;37:1-10.
Gnudi L, Jensen DR, Tozzo E, Eckel RH, Kahn BB. Adipose-specific overexpression of GLUT-4 in transgenic mice alters lipoprotein lipase activity. Am J Physiol. 1996;270:785-792.
Levak-Frank S, Radner H, Walsh A, Stollberger R, Knipping G, Hoefler G, Sattler W, Weisnstock PH, Breslow JL, Zechner R. Muscle-specific overexpression of lipoprotein lipase causes a severe myopathy characterized by proliferation of mitochondria and peroxisomes in transgenic mice. J Clin Invest. 1995;96:976-986.
Liu M-S, Jirik FR, Leboeuf RC, Henderson H, Castellani LW, Lusis AJ, Ma Y, Forsythe IJ, Zhang H, Kirk E, Brunzell JD, Hayden MR. Alteration of lipid profiles in plasma of transgenic mice expressing human lipoprotein lipase. J Biol Chem. 1994;269:11417-11424.
Shimada M, Shimano H, Gotoda T, Yamamoto K, Kawamura M, Inaba T, Yazaki Y, Yamada N. Overexpression of human lipoprotein lipase in transgenic mice. J Biol Chem. 1993;268:17924-17929.
Shimada M, Ishibashi S, Inaba T, Yagyu H, Harada K, Osuga J-I, Oshahi K, Yazaki Y, Yamada N. Suppression of diet-induced atherosclerosis in low density lipoprotein receptor knockout mice overexpressing lipoprotein lipase. Proc Natl Acad Sci U S A. 1996;93:7242-7246.
Tsutsumi K, Inoue Y, Shima A, Iwasaki K, Kawamura M, Murase T. The novel compound NO-1886 increases lipoprotein lipase activity with resulting elevation of high density lipoprotein cholesterol, and long-term administration inhibits atherogenesis in the coronary arteries of rats with experimental atherosclerosis. J Clin Invest. 1993;92:411-417.
Tsutsumi K, Inoue Y, Shima A, Murase T. Correction of hypertriglyceridemia with low high-density lipoprotein cholesterol by the novel compound NO-1886, a lipoprotein lipase-promoting agent, in STZ-induced diabetic rats. Diabetes. 1995;44:414-417.
Shepherd J. Fibrates and statins in the treatment of hyperlipidemia: an appraisal of their efficacy and safety. Eur Heart J. 1995;16:5-13.
Coleman T, Seip RL, Gimble JM, Lee D, Maeda N, Semenkovich CF. COOH-terminal disruption of lipoprotein lipase in mice is lethal in homozygotes, but heterozygotes have elevated triglycerides and impaired enzyme activity. J Biol Chem. 1995;270:12518-12525.
Weinstock PH, Bisgaier CL, Aalto-Setala K, Radner H, Ramakrishnan R, Levak-Frank S, Essenburg AD, Zechner R, Breslow JL. Severe hypertriglyceridemia, reduced high density lipoprotein and neonatal death in lipoprotein lipase knockout mice: mild hypertriglyceridemia with impaired very low density lipoprotein clearance in heterozygotes. J Clin Invest. 1995;65:2555-2568.
Stratford-Perricaudet LD, Makeh I, Perricaudet M, Briand P. Widespread long-term gene transfer to mouse skeletal muscles and heart. J Clin Invest. 1992;90:626-630.
Peterson J, Fujimoto WY, Brunzell JD. Human lipoprotein lipase: relationship of activity, heparin affinity and conformation as studied with monoclonal antibodies. J Lipid Res. 1992;33:1165-1170.
Nilsson-Ehle P, Schotz MC. A stable, radioactive substrate emulsion for assay of lipoprotein lipase. J Lipid Res. 1976;17:536-541.
Beisiegel U, Weber W, Bengtsson-Olivecrona G. Lipoprotein lipase enhances the binding of chylomicrons to low density lipoprotein receptor-related protein. Proc Natl Acad Sci U S A. 1991;88:8342-8346.
Skottova N, Savonen R, Lookene A, Hultin M, Olivecrona G. Lipoprotein lipase enhances removal of chylomicrons and chylomicron remnants by the perfused rat liver. J Lipid Res. 1995;36:1334-1344.
Coppack SW, Jensen MD, Miles JM. In vivo regulation of lipolysis in humans. J Lipid Res. 1994;35:177-193.
Sprecher DL, Knauer SL, Black DM, Kaplan LA, Akeson AA, Dusing M, Lattier D, Stein EA, Rymaszewski M, Wiginton DA. Chylomicron-retinyl palmitate clearance in type I hyperlipidemic families. J Clin Invest. 1991;88:985-994.
O’Brien KD, Gordon D, Deeb S, Ferguson M, Chait A. Lipoprotein lipase is synthesized by macrophage-derived foam cells in human coronary atherosclerotic plaques. J Clin Invest. 1992;89:1544-1550.
Mattsson L, Johansson H, Ottosson M, Bondjers G, Wiklund O. Expression of lipoprotein lipase mRNA and secretion in macrophages isolated from human atherosclerotic aorta. J Clin Invest. 1993;92:1759-1765.
Renier G, Skamene E, Desanctis JB, Radzioch D. High macrophage lipoprotein lipase expression and secretion are associated in inbred murine strains with susceptibility to atherosclerosis. Arterioscler Thromb. 1993;13:190-196.
Auerbach BJ, Bisgaier CL, Wolle J, Saxena U. Oxidation of low-density lipoproteins greatly enhances their association with lipoprotein lipase anchored to endothelial cell matrix. J Biol Chem. 1996;271:1329-1335.
Yeh P, Dedieu J-F, Orsini C, Vigne E, Denefle P, Perricaudet M. Efficient dual transcomplementation of adenovirus E1 and E4 regions from a 293-derived cell line expressing a minimal E4 functional unit. J Virol. 1996;70:559-565.