Donate Help Contact The AHA Sign In Home
American Heart Association
Arteriosclerosis, Thrombosis, and Vascular Biology
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Arteriosclerosis, Thrombosis, and Vascular Biology. 1996;16:1052-1062

This Article
Right arrow Abstract Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jin, F.-Y.
Right arrow Articles by Kashyap, M. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jin, F.-Y.
Right arrow Articles by Kashyap, M. L.
(Arteriosclerosis, Thrombosis, and Vascular Biology. 1996;16:1052-1062.)
© 1996 American Heart Association, Inc.

Articles

Gemfibrozil Stimulates Apolipoprotein A-I Synthesis and Secretion by Stabilization of mRNA Transcripts in Human Hepatoblastoma Cell Line (Hep G2)

Fu-You Jin; Vaijinath S. Kamanna; Mei-Yu Chuang; Kengathevy Morgan; Moti L. Kashyap

the Cholesterol Center, Medical Service, Long Beach (Calif) Department of Veterans Affairs Medical Center, and the University of California, Irvine.

Correspondence to Moti L. Kashyap, MD, Director, Cholesterol Center, Department of Veterans Affairs Medical Center, 5901 E Seventh St (111GE), Long Beach, CA 90822.


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Gemfibrozil is a widely used drug that elevates plasma HDL and lowers triglycerides and LDL. The mechanism of action of this pharmacological agent on HDL metabolism is not established. Since the liver is the major organ involved in HDL production and removal, we assessed the effect of gemfibrozil on the modulation of apoA-I (a major protein of HDL)–containing particles by a human hepatoblastoma cell line (Hep G2). Incubation of Hep G2 cells with gemfibrozil resulted in the following statistically significant findings: (1) increased accumulation of apoA-I in the medium without affecting uptake of radiolabeled HDL-protein or HDL–apoA-I; (2) accelerated incorporation of [3H]leucine and [35S]methionine into apoA-I; (3) equivalent increases in [3H]leucine incorporation into HDL particles without and with apoA-II (LpA-I and LpA-I+A-II, respectively); (4) equal efflux of fibroblast cholesterol by harvested LpA-I and LpA-I+A-II particles; (5) increased steady state apoA-I mRNA without affecting apoA-I transcription; and (6) increased apoA-I mRNA half-life (2.2-fold). These data indicate that gemfibrozil stabilizes apoA-I mRNA transcripts, resulting in increased translation of functional apoA-I–containing particles capable of effluxing cellular cholesterol, thus defining a major mechanism by which gemfibrozil increases HDL.


Key Words: atherosclerosis • coronary artery disease • reverse cholesterol transport • cellular cholesterol efflux • apolipoprotein A-I gene expression


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
It is generally accepted that the plasma concentration of HDL bears an inverse relationship to atherosclerotic cardiovascular disease. Although the exact reasons for this association are not entirely clear, the following properties of HDL and/or its components are considered to be cardioprotective. ApoA-I, a major protein of HDL, initiates cholesterol efflux and thereby facilitates removal of excess tissue (eg, arterial) cholesterol by the process of reverse cholesterol transport.1 2 Thus, apoA-I has been found to be more powerful as a marker in some studies for coronary disease than other HDL components (eg, cholesterol).3 4 Dramatic evidence of its antiatherogenic properties has been demonstrated by the relative lack of development of atherosclerosis in an atherosclerosis-prone mouse model made transgenic for human apoA-I.5 HDLs have been shown to prevent LDL oxidation, a process that may potentiate its atherogenicity.6 Apo A-I enhances fibrinolysis7 and antiplatelet activity,8 thus possibly having an antithrombotic role in coronary disease prevention. Additional studies have indicated that the increased apoA-I gene expression in apoE-deficient mice markedly suppressed atherosclerosis, further supporting a protective role for apoA-I.9

Agents that increase HDL levels have been found to be associated with decreased coronary disease risk as assessed by epidemiological studies (eg, ethanol),10 clinical trials (eg, niacin11 and fibric acid derivatives such as gemfibrozil12 ), and angiographic studies (eg, niacin13 ). The significant cardioprotection in premenopausal women is considered to result from higher HDL levels in females, probably due to estrogens.14 Parallel to these clinical observations, earlier studies have shown that the treatment of human hepatoma cells with estrogen increased apoA-I mass in the culture medium.15

Gemfibrozil has been shown in a primary prevention study to significantly reduce coronary events.12 16 Two large-scale national secondary prevention trials are under way to determine whether raising HDL in patients with coronary disease can decrease coronary events.17 18 19 In spite of this clinical information and resources allocated to expensive clinical trials, the mechanism(s) by which gemfibrozil and other fibrates increase HDL is unclear.

Using HDL turnover (kinetic) techniques, we have shown previously20 that gemfibrozil increased the production rate (ie, transport rate) of apos A-I and A-II in patients with low HDL levels associated with primary hypertriglyceridemia without affecting the fractional catabolic rate. These observations formed the rationale for studying in greater detail the effect of gemfibrozil on hepatic production of apoA-I HDL particles. Because the liver is the major organ for its synthesis, we used Hep G2 cells, a human hepatoma cell line that has been shown to be a useful model for studying hepatic lipoprotein metabolism.21 22 Additionally, Forte et al23 have characterized the secretion of HDL particles by Hep G2 cells. The specific aims of this study were to assess the effect of gemfibrozil on Hep G2 cells with respect to (1) production of apoA-I; (2) incorporation of radiolabeled amino acids ([3H]leucine and [35S]methionine) into apoA-I; (3) properties of secreted material to functionally efflux cellular cholesterol; (4) production of LpA-I and LpA-I+A-II; and (5) steady state mRNA for apoA-I, its transcription rate, and half-life. The data presented indicate that gemfibrozil increases the steady state levels of mRNA for apoA-I by stabilizing the transcript and thus increasing the translation and secretion of functionally potent apoA-I particles capable of effluxing cellular cholesterol, thereby defining at the molecular level one major mechanism of action of this drug.


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Tissue-culture materials and media were obtained from Sigma Chemical Company unless otherwise noted. FBS was obtained from Hyclone Laboratories. L-[4,5-3H]Leucine, [3H]cholesterol, and 32P-labeled nucleotides were purchased from Amersham Corporation. Hep G2 cells, human fibroblast cells, and cDNA probe for human apoA-I were obtained from American Type Culture Collection. Nucleic acid labeling and detection kit, Genius 1, Lumi-Phos 530, and polyclonal antibody for human apoA-I and apoA-II were obtained from Boehringer Mannheim Biochemicals. All other chemicals used were of analytical grade. Gemfibrozil was obtained as a gift from Parke-Davis (Morris Plains, NJ).

HDL Isolation
Blood samples were collected from fasting healthy human volunteers; serum was isolated by centrifugation and pooled for lipoprotein isolation. Total HDL and HDL3 were isolated by sequential density ultracentrifugation at final densities of 1.21 g/mL and 1.121 to 1.21 g/mL, respectively.24 The purity of lipoprotein was monitored by agarose gel electrophoresis and its protein content measured by the method of Lowry et al 25 using bovine serum albumin as a standard.

Cell Culture
A human hepatoblastoma cell line, Hep G2, was grown in T-75 flasks with 15 mL of high-glucose DMEM containing 10% FBS, 1% glutamine-penicillin-streptomycin and 1% fungizone. Cells were grown in a humidified incubator at 37°C in an atmosphere of 5% CO2 and 95% air. Subcultures were made from confluent stock cultures by trypsinization with PBS containing 0.5 mmol/L EDTA. Human fibroblast cells were cultured in T-75 flasks with 15 mL of DMEM medium at 37°C in a humidified incubator with 5% CO2. Subcultures were made (passages 18 through 20) as described above.

Studies on Secretion of Apo A-I by Hep G2 Cells
The cells were plated in 60-mm culture Petri dishes at a concentration of 4x106 cells/dish in 4 mL DMEM and grown for 3 to 4 days until they attained 75% to 80% confluence. The studies examining the dose response of gemfibrozil on apoA-I secretion were performed by incubating Hep G2 cells with various amounts of gemfibrozil (0 to 400 µmol/L) at 37°C for 72 hours. The effect of incubation time on apoA-I secretion was examined by incubating cells with 200 µmol/L gemfibrozil at 37°C for varying times (8 to 72 hours). At the termination of the incubation, culture medium from each flask was removed and the cell monolayer washed with PBS and collected for cellular DNA measurement.26 A 50-µL sample of culture medium was assayed for apoA-I by an enzyme-linked immunosorbent assay (ELISA), using a human apoA-I–specific monoclonal antibody (HB-22) developed and characterized in our laboratory.27 The concentration of apoA-I was expressed as nanograms per microgram of cellular DNA.

The HB-22 monoclonal antibody against human apoA-I showed no cross-reactivity with human apoA-II, LDL, VLDL, albumin, or bovine apoA-I. The ELISA utilized was linear over a concentration range of 0.1 to 20 µg/mL of apoA-I, with a correlation coefficient of .995. The aliquots of medium used in the assays for this study corresponded to a range of concentrations that were within the linear range of the assay. Pooled human plasma served as a source for the apoA-I standard that was calibrated against the CDC reference plasma by a CDC lipid/apolipoprotein standardization program at the apolipoprotein laboratory, University of Cincinnati.

De Novo Synthesis of Apo A-I
Studies examining the effect of various doses of gemfibrozil and time of incubation on the de novo synthesis of apoA-I by Hep G2 cells were performed by measuring the incorporation of radiolabeled leucine into apolipoproteins secreted into the medium. Hep G2 cells (4x106) were incubated with varying concentrations of gemfibrozil (0 to 400 µmol/L) in high-glucose DMEM containing 10% FBS for 48 hours at 37°C in a humidified incubator. After the incubation, the medium was replaced with leucine-poor DMEM (5% of the leucine of normal medium) without FBS, containing the corresponding amounts of gemfibrozil and [3H]leucine (5 µCi/mL), and cells were incubated for 18 hours at 37°C. At the end of the incubation, the medium was collected and used for immunoprecipitation. The cell monolayer was washed with PBS and collected for DNA measurement. The effect of incubation time on apoA-I synthesis was examined by incubating Hep G2 cells with gemfibrozil (200 µmol/L) at 37°C in a humidified incubator. The medium was then replaced with leucine-poor DMEM without FBS, containing corresponding amounts of gemfibrozil and [3H]leucine (5 µCi/mL), and cells were incubated at 37°C for varying periods (4 to 24 hours). After the incubation, the media and cells were collected for immunoprecipitation and cellular DNA measurement, respectively.

The incorporation of radiolabeled leucine into apoA-I was measured by immunoprecipitation, using monospecific antibodies for apoA-I.27 In brief, 100 µL of the medium was added to an Eppendorf tube containing 200 µL PBS and 100 µL antihuman apoA-I antiserum (1:50 dilution). After a 48-hour incubation at 4°C, 100 µL of anti-IgG (1:10 dilution) was added to each tube and incubated for 24 hours at 4°C. The precipitated apoA-I immunocomplex was separated by centrifugation, the pellets were washed five times with PBS containing 0.1% SDS dissolved in 100 µL of 1 N NaOH, and an aliquot was used to measure the radioactivity. The incorporation of [3H]leucine into albumin was measured by an immunoprecipitation procedure exactly as described above using anti-human albumin antiserum. The incorporation of [3H]leucine into apoA-I, or albumin, was expressed as counts per minute per microgram cellular DNA. The specificity of this immunoprecipitation technique was examined by analyzing apoA-I immunocomplex by SDS-polyacrylamide gel electrophoresis. These studies indicated that all the radioactivity of apoA-I immunocomplex was present in a band corresponding to apoA-I but not other apolipoproteins.

Additional studies examining the de novo synthesis of apoA-I by Hep G2 cells were performed by using [35S]methionine. Hep G2 cells were incubated with varying concentrations of gemfibrozil (0 to 200 µmol/L) for 48 hours. After the incubation, medium was replaced with fresh methionine-free DMEM containing the respective concentrations of gemfibrozil and [35S]methionine (150 µCi/mL). After 15 minutes of pulsing, the medium was replaced with fresh DMEM containing 15 µg/mL of methionine and incubated for 1 hour. At the end of the incubation period, culture medium and cells were collected. The radiolabeled apoA-I in medium and cell lysate was assayed by immunoprecipitation and expressed in terms of total cellular protein.

Separation of LpA-I and LpA-I+A-II Particles
Experimental protocols for these studies were exactly the same as described for de novo synthesis of apoA-I. After the incubation of Hep G2 cells with gemfibrozil and [3H]leucine, the medium was collected and used to isolate LpA-I and LpA-I+A-II particles by immunoaffinity column chromatography. In brief, affinity columns specific for apoA-I or apoA-II were prepared by coupling polyclonal antibodies for human apoA-I or apoA-II to CNBr-activated sepharose 4B (Pharmacia) according to the procedure described in the instruction manual. Aliquots of culture medium (250 µL) were loaded onto the apoA-I affinity column and incubated at 4°C for 16 to 18 hours to allow binding of apolipoprotein particles to specific antibody. The affinity column was then washed with 0.5 mol/L NaCl and retained apoA-I–containing particles were eluted with 3 mol/L NaSCN, pH 6.0.28 An aliquot of eluted fraction was counted for radioactivity and represents LpA-I with and without A-II particles. Similarly, another aliquot (250 µL) of culture medium was subjected to apoA-II–specific immunoaffinity column chromatography and an aliquot of NaSCN-eluted retained fraction counted for radioactivity. This retained fraction on apoA-II–specific affinity column represents the contribution of apoA-II in LpA-I+A-II particles. The incorporation of [3H]leucine into LpA-I particles (without A-II) was measured by the difference in radioactivity between retained fractions on apoA-I affinity column and apoA-II affinity column. The procedure described above used to separate LpA-I and LpA-I+A-II allowed us to prepare relatively larger amounts of these particles (which were required for immunoprecipitation and cholesterol efflux studies) than the conventional method of separation of LpA-I and LpA-I+A-II, using successive passing of starting material over an apoA-II column and then passing the unretained fraction through an apoA-I column. Furthermore, the validation of the specificity of separation of HDL subfractions by immunoaffinity columns was assessed by Western blot analysis of eluants (LpA-I+A-II and LpA-I) obtained from anti–apoA-II and anti–apoA-I columns using normal human serum as a starting material. The results indicated that the NaSCN-eluted retained fraction of the anti–apoA-II column demonstrated immunopositive bands for apoA-I and apoA-II. The unretained fraction from the anti–apoA-II column (containing LpA-I and other proteins) was concentrated, dialyzed, and passed through an anti–apoA-I column. The retained fraction obtained by NaSCN elution from this anti–apoA-I column showed an immunopositive band for apoA-I but not for apoA-II. Thus, these results validate the specificity of the immunoaffinity column procedure used for isolation of LpA-I and LpA-I+A-II HDL particles.

Uptake of HDL or Apo A-I by Hep G2 Cells
Studies examining the uptake by Hep G2 cells were performed by using radiolabeled HDL total protein or apoA-I-HDL. Radioiodination of HDL total protein was carried out by incubating freshly isolated HDL3 with carrier-free 125I as described earlier by McFarlane.29 After the iodination, unreacted 125I was removed by gel filtration followed by exhaustive dialysis against PBS. Specific activity of 125I-HDL protein was 186 cpm/ng protein. Radioiodination of apoA-I was performed by incubating apoA-I (350 µg in 1.0 mol/L glycine, pH 7.4) with 125I (1 mCi) and 1,3,4,6-tetrachloro-3{alpha},6{alpha}-diphenylglycouril (Iodo-Gen, Pierce Chemical Company; 10 µg in methylene chloride) for 30 minutes at room temperature. After the incubation, the 125I–apoA-I was purified by passing through Sephadex G-25 columns and then dialyzed. 125I–apoA-I was then incubated with HDL3 for 1 hour at 37°C to reassociate apoA-I with HDL in vitro.30 At the termination of the incubation, 125I–apoA-I–HDL was isolated by ultracentrifugation at a density of 1.21 g/mL. The radiolabeled apoA-I–HDL was exhaustively dialyzed and the specific activity was 62 cpm/ng protein. Uptake studies were initiated by preincubating Hep G2 cells with varying concentrations of gemfibrozil (0 to 400 µmol/L) for 48 hours at 37°C. The medium was replaced with fresh DMEM containing FBS (5 mg/mL) and either 125I-HDL or 125I–apoA-I–HDL (50 µg protein) was added. After 16 hours of incubation at 37°C, cell monolayers were washed thoroughly (four to five times with PBS) and digested with 1 N NaOH. An aliquot was used for radioactivity measurement. The uptake of radiolabeled HDL particles by Hep G2 cells was expressed in terms of cellular protein.

Measurement of Cholesterol Efflux
Experimental protocols for these studies were exactly as described for secretion of apoA-I. After the incubation of Hep G2 cells with gemfibrozil, the medium was collected and used for cholesterol efflux measurement. An aliquot of culture medium (5 mL) was concentrated to 1 mL by lyophilization and dialyzed against DMEM to remove excess salt present in the concentrated sample. The ability of these concentrated samples to efflux free cholesterol was measured by a previously described modified procedure31 32 of Fielding and Fielding33 and Rothblatt et al,34 using [3H]cholesterol-labeled human fibroblasts. To ensure specific and homogenous incorporation of radiolabeled cholesterol throughout the cytoplasmic matrix of fibroblasts, we incubated fibroblasts with [3H]cholesterol for 72 hours at 37°C, washed them with PBS, and then incubated them with fresh DMEM containing 1% fetal bovine albumin for 16 hours at 37°C. Cholesterol efflux assay was initiated by incubating concentrated culture medium with [3H]cholesterol-labeled fibroblasts for 20 hours at 37°C in a humidified incubator. Quantitative analysis of the Hep G2 cell culture medium (in presence or absence of gemfibrozil) was assessed by measuring the radioactivity appearing in the medium per milliliter of incubation medium per milligram of fibroblast cellular protein.

Northern Blot Analysis
Experimental protocols for these studies were exactly as described for secretion of apoA-I. Total RNA was isolated from Hep G2 cells using the protocol of Chomczynski and Sacchi.35 In brief, cells were homogenized with 4 mol/L guanidinium thiocyanate, total protein and DNA were removed with water-saturated phenol, and the RNA was precipitated by using isopropanol. After washing with ethanol, the samples were dried under vacuum centrifugation and the amount of RNA was quantitated by measuring the absorbance at 260 nm with a spectrophotometer. Total RNA (20 µg) was loaded into individual wells of a 1.2% agarose gel containing formaldehyde, and electrophoresis was performed.36 The RNA from the gel was transferred onto MSI nylon membranes and UV cross-linked using a stratalinker (Stratagene). The membrane was then hybridized overnight with the digoxigenin-labeled cDNA probe for human apoA-I.37 Messenger RNA detection was performed by a chemiluminescent method, using a Lumi-Phos 530 kit and autoradiography.38 39 Blots were then rehybridized with human ß-actin cDNA probe as an internal control to assess RNA quantity and integrity. Quantification of mRNA signals was performed by densitometric scanning of autoradiographic bands and normalized with ß-actin mRNA signals, using the LKB laser densitometer (Pharmacia LKB Biotech).

Measurement of Apo A-I mRNA Half-life
Hep G2 cells were incubated in the presence of actinomycin D (1 µmol/L) at 37°C for 0, 2, 4, 8, 16, 24, and 48 hours. For experimental sets, Hep G2 cells were preincubated with gemfibrozil (200 µmol/L) for 48 hours at 37°C. After the preincubation, the medium was replaced with fresh DMEM containing gemfibrozil (200 µmol/L) and actinomycin D (1 µmol/L) and incubated for the same time intervals as for control Hep G2 cells. At the termination of the incubation, cells were washed and collected for RNA isolation. Northern blot analysis was performed as described earlier, and the half-life for apoA-I transcript was calculated from the density of the autoradiographic bands at each time point of incubation of cells in the presence of actinomycin D as described earlier.40 In brief, apoA-I mRNA half-life was measured by constructing a percent decay curve of mRNA signals (based on 0 hours' incubation with actinomycin D as 100% signal) by using incubation of Hep G2 cells with actinomycin D for varying time periods of 2 to 48 hours. mRNA half-life was calculated by extrapolating the incubation time in hours at which 50% of mRNA decayed. Preliminary experiments were performed to examine the effect of actinomycin D on cellular viability. These studies indicated that the incubation of Hep G2 cells with actinomycin D (1 µmol/L) for 0 to 48 hours did not significantly alter Hep G2 cell viability (93% to 98% of control), as assessed by the trypan blue exclusion method.

Measurement of Newly Transcribed Apo A-I mRNA
Hep G2 cells were grown in DMEM+10% FBS for 4 to 5 days in 175 cm2 flasks. For experiments, cells were incubated with 200 µmol/L gemfibrozil for 48 hours, whereas control cells were incubated without gemfibrozil for the same period. After incubation, cells were washed with ice-cold PBS and gently dislodged from the plastic surface by scraping. Cells were then lysed in NP-40 lysis buffer (10 mmol/L Tris-HCl, pH 7.4, 10 mmol/L NaCl, 3 mmol/L MgCl2, and 0.5% NP-40). The nuclei were isolated by centrifugation (500g for 5 minutes at 4°C), resuspended in glycerol storage buffer (50 mmol/L Tris-HCl, pH 8.3, 40% glycerol, 5 mmol/L MgCl2, and 0.1 mmol/L EDTA), and stored in liquid nitrogen until use.

PKT 218 plasmid DNA containing human apoA-I cDNA was prepared as follows: 100 µg plasmid DNA was linearized with restriction enzyme BamHI. Linearized cDNA was then denatured by treatment with 1 mol/L NaOH and neutralized with 6xSSC. cDNA plasmid (5 µg ) was spotted onto nitrocellulose membrane filters by using a slot blot apparatus. The filters were air-dried and UV cross-linked and stored in a vacuum desiccator until use. The thawed nuclei (5x107 cells) were added to an equal volume of reaction buffer (10 mmol/L Tris-HCl, pH 8.0, 5 mmol/L MgCl2, and 0.3 mol/L KCl) containing 1 mmol/L ATP, CTP, and GTP plus 10 µL of 10 mCi/mL [{alpha}-32P]UTP. After a 30-minute incubation at 30°C, DNA was digested with DNase, followed by proteinase K digestion. The newly formed 32P-labeled RNA was then purified by phenol-chloroform extraction and precipitated by isopropanol. An aliquot was counted and equal counts of labeled nuclear RNA were hybridized with the cDNA for apoA-I on nitrocellulose filters. Comparison was made by using a cDNA for ß-actin. After 36 hours of hybridization, the filters were washed and exposed to x-ray film for autoradiography.41 42

Statistical Analysis
In studies examining apoA-I accumulation in the medium (secretion), de novo synthesis of apoA-I, uptake of HDL or apoA-I, and cholesterol efflux experiments, triplicate Hep G2 cell plates were used for each incubation condition, and an average value was calculated for these triplicates in each experiment. Three such experiments were performed, and their mean value was used to calculate the SEM and for statistical analysis. In studies examining apoA-I mRNA expression, mRNA half-life, and nuclear runoff experiments, four to five separate experiments were performed, using one Hep G2 cell plate for each incubation, and a mean value was used to calculate the SEM and for statistical analysis. Statistical significance was calculated by using Student's t test, and a value of P<.05 was considered significant.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 *Results
down arrowDiscussion
 down arrowReferences
 
The incubation of various amounts of gemfibrozil (0 to 400 µmol/L) for 72 hours with Hep G2 cells showed a dose-dependent increase in apoA-I accumulation (secretion) in the medium as measured by ELISA. A significant increase in apoA-I secretion by Hep G2 cells was noted at 100 µmol/L of gemfibrozil (apoA-I concentration, in ng/µg cell DNA: control, 173.4±31.9; gemfibrozil [100 µmol/L], 211.0±31.3; P=.04), and the maximum effect was observed at 200 µmol/L (apoA-I concentration, in ng/µg cell DNA: control, 173.4±31.9; gemfibrozil [200 µmol/L], 275.6±40.7; P=.002). Incubation of Hep G2 cells with gemfibrozil at 25- to 200-µmol/L concentrations for 72 hours did not alter the morphology or the viability of cells. However, the addition of gemfibrozil at 300- and 400-µmol/L concentrations decreased the viability of Hep G2 cells by {approx}10% to 12% compared with controls.

The effect of incubation time on apoA-I accumulation in medium of Hep G2 cells in the presence of 200 µmol/L of gemfibrozil was examined by measuring apoA-I concentration in medium by ELISA (Fig 1ADown). At 48 hours of incubation of gemfibrozil with Hep G2 cells, a significant increase in apoA-I mass compared with the control was noted, and the maximal effect was observed at 72 hours of incubation (Fig 1ADown). Similarly, a time-dependent increase in the de novo synthesis of apoA-I was observed by Hep G2 cells in presence of 200 µmol/L of gemfibrozil (Fig 1BDown). In these experiments, Hep G2 cells were preincubated with gemfibrozil (200 µmol/L) for 48 hours prior to the addition of [3H]leucine and incubation with gemfibrozil for an additional 4 to 24 hours. Incubation of gemfibrozil with Hep G2 cells for as little as 4 hours (total incubation with gemfibrozil including preincubation, 52 hours) increased the apoA-I de novo synthesis, as measured by the incorporation of radiolabeled leucine into apoA-I, and the maximum effect was noted at 24 hours of incubation (total incubation with gemfibrozil including preincubation, 72 hours; Fig 1BDown). Based on the total incubation period of Hep G2 cells with gemfibrozil, the last incubation point (72 hours) in both Fig 1A and 1BDownDown had similar induction of apoA-I accumulation in the medium, as determined by either ELISA or [3H]leucine incorporation into apoA-I (1.5- to 1.7-fold) compared with respective controls.




View larger version (45K):
[in this window]
[in a new window]
  		Figure 1. Effect of gemfibrozil on secretion and de novo synthesis of apoA-I by Hep G2 cells at varying incubation times. A, Cells were incubated with or without 200 µmol/L of gemfibrozil at 37°C for 8 to 72 hours. At each time point of the incubation, culture media were assayed for apoA-I by ELISA, using monoclonal antibody to apoA-I, and results were expressed in terms of total cellular DNA. B, Hep G2 cells were incubated with 200 µmol/L of gemfibrozil for 48 hours. Medium was then changed to fresh leucine-poor DMEM with the respective concentrations of gemfibrozil containing 5 µCi/mL of [3H]leucine and incubated for 4 to 24 hours. At each time point of the incubation, culture media were assayed for radiolabeled apoA-I by immunoprecipitation and results were expressed in terms of total cellular DNA. Each point represents the mean of three separate experiments done in triplicate. Statistical significance was compared with results of control. Data are mean±SEM, n=3. *P<.05; **P<.01.

Since the accumulation of apoA-I in the culture medium represents both secretion and reuptake by Hep G2 cells, further studies were performed to examine the effect of gemfibrozil on uptake of HDL particles by Hep G2. The preincubation of Hep G2 cells with gemfibrozil (0 to 400 µmol/L) for 48 hours did not significantly alter the uptake of either 125I-HDL protein or 125I–apoA-I–HDL compared with the respective control. The uptake of 125I-HDL protein by Hep G2 cells treated with gemfibrozil (200 µmol/L) was 23 218±354 cpm/mg cell protein compared with control cell uptake of 22 816±417 cpm/mg cell protein. Similarly, the uptake of 125I–apoA-I–HDL by Hep G2 cells treated with gemfibrozil (200 µmol/L) and control was 11 404±188 and 11 235±314 cpm/mg cell protein, respectively. These results suggest that the accumulation of apoA-I in the Hep G2 culture medium was primarily due to secretion. In these studies, cellular uptake of radiolabeled HDL particles represented {approx}0.6% of the total radioactivity of HDL particles added to the incubation medium.

Additional experiments were designed to examine the de novo synthesis of apoA-I by measuring the incorporation of [3H]leucine into newly synthesized apoA-I secreted into the medium. Data from these studies show that the incorporation of radiolabeled leucine into apoA-I increased in a dose-dependent manner by Hep G2 cells incubated with gemfibrozil (Fig 2ADown). A significant increase in the incorporation of radiolabeled leucine into immunoprecipitable apoA-I was noted in the presence of as little as 50 µmol/L of gemfibrozil. At higher concentrations of gemfibrozil (100 to 200 µmol/L), the elevation in the incorporation of radiolabeled leucine into apoA-I was persistent, with a maximum effect noted at 200 µmol/L of gemfibrozil compared with control.



View larger version (47K):
[in this window]
[in a new window]
  		Figure 2. Effect of gemfibrozil on the incorporation of [3H]leucine into newly synthesized apoA-I (A) and albumin (B) by Hep G2 cells. Cells were incubated with varying concentrations of gemfibrozil (0 to 400 µmol/L) for 48 hours at 37°C. Medium was then changed to fresh leucine-poor DMEM with the respective concentrations of gemfibrozil containing 5 µCi/mL of [3H]leucine and incubated for 18 hours at 37°C. At the end of the incubation, culture media were assayed for radiolabeled apoA-I and albumin by immunoprecipitation and results were expressed in terms of total cellular DNA. Each point represents the mean of three separate experiments done in triplicate. Statistical significance was compared with results of control. Data are mean±SEM, n=3. *P<.05; **P<.01.

Additional experiments were performed to determine whether gemfibrozil specifically alters apoA-I synthesis or whether the observed effects are a consequence of alterations in total protein synthesis. Results from these studies indicated that the incubation of gemfibrozil (0 to 400 µmol/L) with Hep G2 cells did not alter the de novo synthesis of albumin, as measured by the incorporation of radiolabeled leucine into immunoprecipitable albumin in the culture medium (Fig 2BUp). Similarly, the treatment of Hep G2 cells with gemfibrozil did not alter the incorporation of radiolabeled leucine into the total trichloroacetic acid–precipitable protein (data not shown).

Further confirmatory pulse-chase experiments were performed to examine the incorporation of [35S]methionine into apoA-I in Hep G2 cells preincubated with varying concentrations of gemfibrozil (0 to 200 µmol/L). Results from these studies indicated that the preincubation of Hep G2 cells with gemfibrozil resulted in a significant increase in radiolabeled methionine incorporation into both cellular and secreted apoA-I (Fig 3A and 3BDownDown).



View larger version (38K):
[in this window]
[in a new window]
  		Figure 3. Effect of gemfibrozil on the incorporation of [35S]methionine into newly synthesized apoA-I by Hep G2 cells. Cells were incubated with varying concentrations of gemfibrozil (0 to 200 µmol/L) for 48 hours at 37°C. Medium was then changed to fresh methionine-free DMEM containing the respective concentrations of gemfibrozil and [35S]methionine (150 µCi/mL). After 15 minutes of pulsing, the medium was replaced with fresh DMEM containing 15 µg/mL of methionine and incubated for 1 hour at 37°C. At the end of the incubation period, culture media were collected, and cells were harvested and lysed by lysis buffer. The radiolabeled apoA-I in medium and cell lysate was assayed by immunoprecipitation and results were expressed in terms of total cellular protein. A, [35S]Methionine incorporation into apoA-I in cell lysate. B, [35S]Methionine incorporation into apoA-I secreted into the medium. Each point represents the mean of three separate experiments done in triplicate. Statistical significance was compared with results of control. Data are mean±SEM, n=3. *P<.05; **P<.01.

To examine the effect of gemfibrozil on the de novo synthesis of LpA-I and LpA-I+A-II HDL particles in Hep G2 cells, aliquots of medium from de novo synthesis experiments were fractionated into LpA-I and LpA-I+A-II particles by immunoaffinity column chromatography. Results from these experiments revealed that the incubation of gemfibrozil with Hep G2 cells increased in a dose-dependent manner the secretion of newly synthesized LpA-I and LpA-I+A-II HDL particles into the medium (Fig 4Down). Gemfibrozil stimulated the de novo synthesis of both LpA-I and LpA-I+A-II particles to the same extent (Fig 4Down, 19% to 31% of control).



View larger version (36K):
[in this window]
[in a new window]
  		Figure 4. Effect of gemfibrozil on the incorporation of [3H]leucine into newly synthesized LpA-I and LpA-I+A-II by Hep G2 cells. Cells were incubated with varying concentrations of gemfibrozil (0 to 400 µmol/L) for 48 hours at 37°C. Medium was then changed to fresh leucine-poor DMEM with the respective concentrations of gemfibrozil containing 5 µCi/mL of [3H]leucine and incubated for 18 hours at 37°C. At the end of the incubation, culture media were used to isolate LpA-I and LpA-I+A-II particles by immunoaffinity chromatography, as described in "Methods." Radioactivity incorporated into LpA-I and LpA-I+A-II was measured and expressed in terms of total cellular DNA. Each point represents the mean of three separate experiments done in triplicate. Statistical significance was compared with results of control. Data are mean±SEM, n=3. *P<.05; **P<.01.

The ability of gemfibrozil-induced secretion of apoA-I–containing lipoprotein particles to efflux cholesterol was examined by using [3H]cholesterol-labeled fibroblasts. Cholesterol efflux studies performed with conditioned medium obtained from Hep G2 cells treated with varying amounts of gemfibrozil showed a dose-dependent increase in cholesterol efflux, as measured by the release of [3H]cholesterol from fibroblasts into the culture medium (Fig 5Down). Experiments were then performed to assess the ability of LpA-I and LpA-I+A-II particles isolated from conditioned medium to efflux fibroblast cholesterol. Apo A-I mass was measured (by ELISA) in both particle preparations, and efflux was normalized to reflect radioactivity in the medium per unit of apoA-I mass. In control medium, both particles showed similar efflux per unit mass of apoA-I (LpA-I, 27.7±0.7 cpm·µg cell protein-1·µg apoA-I-1; LpA-I+A-II, 26.8±0.5 cpm·µg cell protein-1·µg apoA-I-1); the efflux property per unit apoA-I was essentially the same in particles isolated from gemfibrozil-incubated Hep G2 cells (LpA-I, 26.4±1.1 cpm·µg cell protein-1·µg apoA-I-1; LpA-I+A-II, 25.1±0.5 cpm·µg cell protein-1·µg apoA-I-1). Thus, gemfibrozil increased concentration of LpA-I and LpA-I+A-II, both of which increased fibroblast cholesterol equally.



View larger version (64K):
[in this window]
[in a new window]
  		Figure 5. Effect of gemfibrozil-induced apoA-I–containing particles from Hep G2 cells to efflux cholesterol from cultured human fibroblasts. Hep G2 cells were incubated with varying concentrations of gemfibrozil (0 to 400 µmol/L) for 72 hours at 37°C. After the incubation, an aliquot of medium was concentrated (fivefold) and added to [3H]cholesterol-labeled human fibroblast cultures to measure its ability to efflux cholesterol, as described in "Methods." Radioactivity appearing in the culture medium (as a measure of cholesterol efflux) was expressed in terms of cellular protein. Each point represents the mean of three separate experiments done in triplicate. Data are mean±SEM, n=6. *P<.05; **P<.01.

Northern blot analysis was performed to examine the effect of gemfibrozil on apoA-I gene expression by Hep G2 cells. Incubation of varying amounts of gemfibrozil with Hep G2 cells induced in a dose-dependent fashion the steady state mRNA transcripts (0.9 kb) for apoA-I (Fig 6Down). Quantitative analysis of the apoA-I mRNA message, as measured by densitometric scanning of blots (after normalizing with ß-actin mRNA message as an internal standard), indicated that the treatment of Hep G2 cells with gemfibrozil at 50-, 100-, 200-, or 400-µmol/L concentrations significantly increased steady state apoA-I mRNA levels by 1.4±0.11-, 1.48±0.07-, 2.04±0.13-, and 2.03±0.1-fold, respectively, compared with control.



View larger version (71K):
[in this window]
[in a new window]
  		Figure 6. Representative study examining the effect of gemfibrozil on the steady state apoA-I mRNA levels by Hep G2 cells. Cells were incubated at 37°C with varying concentrations of gemfibrozil (0 to 400 µmol/L) for 72 hours. Aliquots of total RNA (20 µg) isolated from Hep G2 cells were electrophoresed and Northern blot analysis was performed using human apoA-I cDNA probe, as described in "Methods." Lanes 1 to 5 correspond to gemfibrozil concentrations at 0, 50, 100, 200, and 400 µmol/L, respectively. A, ApoA-I mRNA; B, ß-Actin mRNA (internal control). The blots are representative of four separate experiments. In each incubation, one Hep G2 cell plate was used, and four such separate experiments were performed. Quantitative analysis was performed by densitometric scanning of blots and normalized with ß-actin. Arbitrary densitometric units from four experiments were used to calculate SEM and for statistical analysis.

Because steady state mRNA expression reflects both transcription and transcript degradation, we examined the effect of gemfibrozil on the newly transcribed message and transcript degradation of apoA-I in Hep G2 cells. As shown in Fig 7Down, the incubation of gemfibrozil (200 µmol/L) with Hep G2 cells for 72 hours had no effect on apoA-I mRNA transcription rate, as measured by nuclear runoff assays (arbitrary densitometric values after normalization with ß-actin: control, 2.44±0.23, gemfibrozil treatment, 2.39±0.19). The mRNA half-life studies, in which new RNA production was inhibited by actinomycin D, showed that treatment of Hep G2 cells with gemfibrozil inhibited degradation of apoA-I mRNA transcripts compared with untreated Hep G2 cells (Fig 8DownA1 and 8DownA2). At 16 and 24 hours of incubation of cells with actinomycin D, gemfibrozil-treated Hep G2 cells had {approx}58% to 60% of mRNA transcript abundance (Fig 8DownA1, lanes 6 and 7), while control Hep G2 cells retained about 40% of mRNA transcript message (Fig 8DownA2, lanes 6 and 7) compared with their respective controls.



View larger version (52K):
[in this window]
[in a new window]
  		Figure 7. Nuclear runoff study of the effect of gemfibrozil on apoA-I transcription by Hep G2 cells. Cells were incubated in the presence or absence of 200 µmol/L of gemfibrozil for 48 hours. After the incubation, cells were washed, nuclei isolated by centrifugation at 500g, and a nuclear runoff assay was performed, as described in "Methods" using human apoA-I cDNA and ß-actin (internal standard) cDNA probes. The blot is representative of five separate experiments. In each incubation, one Hep G2 cell plate was used, and five separate experiments were performed (three with nonisotopic cDNA labeling and two experiments with 32P-labeled UTP). Quantitative analysis of apoA-I transcription was performed by densitometric scanning of blots and normalized with ß-actin. Arbitrary densitometric units from five experiments were used to calculate SEM and for statistical analysis.



View larger version (18K):
[in this window]
[in a new window]
  		Figure 8. Representative study examining the effect of gemfibrozil on apoA-I transcript half-life. Hep G2 cells were incubated in the presence of 1 µmol/L actinomycin D only or 1 µmol/L actinomycin D plus 200 µmol/L gemfibrozil for 2 to 48 hours. The cells were then lysed and the total RNA was extracted and hybridized with a cDNA probe for apoA-I, as described. A1, Apo A-I mRNA transcript degradation profile at varying times of incubation of control Hep G2 cells with actinomycin D (top) and GAPDH signal (bottom). A2, Apo A-I mRNA transcript degradation profile at varying times of incubation of Hep G2 cells with gemfibrozil and actinomycin D (top) and GAPDH signal (bottom). Lanes 1 to 7 correspond to 0, 2, 4, 8, 16, 24, and 48 hours of incubation with actinomycin D. B, Apo A-I mRNA decay curves of control and gemfibrozil-treated cells in the presence of actinomycin D for varying times of incubation. Apo A-I mRNA half-life was calculated according to the procedure described in "Methods." C, Histogram showing the effect of gemfibrozil on apoA-I mRNA half-life. Open bar, control; shaded bar, gemfibrozil-treated cells. Data are mean±SEM of four separate experiments, and each incubation contained one Hep G2 cell plate. Quantitative analysis of mRNA levels was performed by densitometric scanning of blots and normalized with GAPDH. Arbitrary densitometric units from four separate experiments were used to calculate SEM and for statistical analysis.

On the basis of quantitative scanning of autoradiographic apoA-I mRNA bands, we constructed a curve to calculate apoA-I transcript degradation rate and half-life according to the procedure described previously.39 Statistical analysis of the data from four separate experiments indicated that the calculated apoA-I mRNA half-life in untreated Hep G2 cells was {approx}14.3±0.46 hours (Fig 8B and 8CUpUp). When Hep G2 cells were treated with gemfibrozil, the apoA-I mRNA half-life was prolonged to {approx}31.6±3.05 hours, indicating that the gemfibrozil treatment stabilized apoA-I mRNA transcripts (Fig 8B and 8CUpUp, P=.01).


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 *Discussion
down arrowReferences
 
The data indicate that gemfibrozil significantly increased accumulation of apoA-I in the medium of Hep G2 cells (Fig 1AUp). Since apoA-I is produced and degraded by the liver, apoA-I accumulation was the net result of these two processes. Our data also confirm that in the Hep G2 system, the accumulation of apoA-I in the medium largely represents secretion. We found the uptake of HDL protein (mainly apoA-I) by Hep G2 cells to be 0.26 µg·mg cell protein-1·24 h-1, whereas the accumulation in the medium was 6.98 µg·mg cell protein-1·24 h-1, indicating that uptake was <4% of net accumulation. These data are in accord with previous observations of this system.43

Gemfibrozil had no effect on the uptake of 125I–HDL-protein or 125I–HDL–apoA-I, indicating that it does not influence removal of HDL. Thus, we inferred that gemfibrozil stimulated the synthesis and secretion of apoA-I. This observation is consistent with in vivo turnover studies previously reported by us indicating that this drug has no effect on the fractional catabolic rate of apoA-I or apoA-II.20

The accumulation of secreted apoA-I by gemfibrozil was dose dependent, with the maximal effect at a 200-µmol/L concentration. At a 400-µmol/L concentration, accumulation decreased, probably indicating an in vitro toxic effect. Time-course experiments at this concentration showed progressively increasing accumulation, the mean increase being statistically significant at 48 hours. This finding formed the rationale for future experiments in which the cells were preincubated for at least 48 hours prior to all other studies. The reason for an apparent delay in significant accumulation is unclear and may represent the time for conversion of gemfibrozil to an active metabolite or some other processes. Similar observations have been reported with another fibrate, fenofibrate, which was found to affect significant enhancement of apoA-I secretion rate at 72 hours of incubation.44 These investigators indicated that the incubation of Hep G2 cells with fenofibrate for 4 days significantly increased apoA-I mass in the culture medium (twofold compared with controls), a similar observation made with gemfibrozil in our studies. Furthermore, these studies44 by pulse-chase immunoprecipitation experiments showed that the increased apoA-I secreted mass in medium was due to an increase in apoA-I synthesis.

That gemfibrozil increased de novo synthesis of apoA-I is indicated by increased incorporation of [3H]leucine into apoA-I in a dose-dependent manner, with maximal effect at 200 µmol/L (Fig 2Up) and an increased rate of synthesis with gemfibrozil (Fig 3Up). The effect was specific, as [3H]leucine incorporation into albumin was unaffected (Fig 2Up). These observations were confirmed by pulse-chase experiments using [35S]methionine, which showed its increased incorporation into apoA-I both in medium and in cell lysate (Fig 3Up), and are in accord with a previous report.45 In preliminary studies using pulse-chase experiments, we have observed no significant intracellular degradation of apoA-I in control cells. Since apoA-I is not normally processed through intracellular degradation prior to secretion (unlike apoB), it is unlikely that the increased accumulation of apoA-I in cell lysate compared with medium seen at 200 µmol/L gemfibrozil (Fig 3Up) would represent decreased intracellular apoA-I degradation in gemfibrozil-treated cells. However, it is possible that during 1 hour of chase period, all of the apoA-I synthesized in gemfibrozil-treated cells may not be secreted into the medium and may require additional incubation time.

Because the liver produces both LpA-I and LpA-I+A-II and since LpA-I has been claimed to be more antiatherogenic than LpA-I+A-II,46 47 we assessed the effect of gemfibrozil on the production of these particles. Incorporation of [3H]leucine into both types of particles separated by immunoaffinity techniques showed similar rates of incorporation into both. Thus, gemfibrozil increased production of both LpA-I and LpA-I+A-II, but not selectively.

Accumulated apoA-I in the medium was functionally potent in effluxing cellular cholesterol. This was confirmed by incubating concentrated medium harvested from Hep G2 cells with [3H]cholesterol-labeled fibroblasts (Fig 5Up). Medium from gemfibrozil-incubated Hep G2 cells potently effluxed cholesterol in a dose-dependent fashion. Efflux of fibroblast cholesterol by secreted LpA-I or LpA-I+A-II was similar. As indicated earlier, it has been suggested that LpA-I particles are more potent in effluxing cholesterol than LpA-I+A-II, and the evidence supporting this suggestion was found in adipocytes.48 However, Cheung et al49 and Johnson et al50 have reported that there are no significant differences in cholesterol efflux by LpA-I or LpA-I+A-II in various cell types, including rat hepatoma cells, human skin fibroblasts, and rabbit aortic smooth muscle cells. Our results are in line with the latter report. However, other evidence unrelated to cellular efflux indicates that LpA-I is associated with decreased atherosclerosis compared with LpA-I+A-II. For example, transgenic mice producing human apoA-I have significantly less atherosclerosis8 than those producing LpA-I+A-II47 when fed atherogenic diets. Patients with specific increases in LpA-I are less prone to atherosclerosis than those with LpA-I+A-II reductions.46 Because of the variability in cholesterol efflux properties of LpA-I and LpA-I+A-II particles, it is not clearly understood whether LpA-I particles without A-II would exert greater antiatherosclerotic properties. Our data indicate that gemfibrozil does increase LpA-I production, which may contribute to its cardioprotective properties, although there is increased LpA-I+A-II production as well.

Although gemfibrozil increased the steady state mRNA for apoA-I (Fig 6Up), nuclear runoff studies indicated no change in apoA-I transcription (Fig 7Up). However, more specific methods (eg, transfection experiments with apoA-I gene promoter) are required to conclusively assess apoA-I gene transcription. The apoA-I mRNA half-life studies performed using actinomycin D to inhibit transcription indicated that the treatment of Hep G2 cells with gemfibrozil significantly increased apoA-I mRNA half-life from {approx}14 to 31 hours. These data indicate that the molecular site of action of gemfibrozil (or its metabolites) is at some point in the degradation of apoA-I mRNA. Since not much is known regarding apoA-I mRNA degradation, additional research will be necessary to address this question. Although these studies showed that gemfibrozil had no effect on apoA-I transcription by Hep G2 cells, previous studies have indicated that the treatment of Hep G2 cells with other fibrates (eg, fenofibrate) demonstrated a negative effect on apoA-I gene transcription,51 which was probably independent of the PPAR. However, fenofibrate increased Hep G2 cell apoA-II gene transcription through the interaction of PPAR with apoA-II promotor PPRE.52 Thus, it appears that fenofibrate and its analogues (eg, gemfibrozil) may exert varied effects on the transcription of apoA-I and A-II. In this regard, studies are warranted to examine the effect of gemfibrozil on PPAR-mediated apoA-I and A-II gene transcription by the human hepatocyte. Such studies were beyond the scope of this investigation.

How do the results of this study relate to the action of gemfibrozil on plasma HDL levels in vivo? The data in this study strongly indicate that the fibrates stimulate hepatic apoA-I production. This finding is consistent with our previous report that gemfibrozil increases the transport rates of apoA-I and A-II in patients with low HDL levels.20 Malmendier and Delcroix53 also observed that the treatment of patients (familial type IIA hypercholesterolemia) with fenofibrate increased both transport (synthetic) rate and fractional catabolic rate of apoA-I without alterations in plasma apoA-I levels. Of course, after initial secretion in the hepatic extracellular space, remodeling of the nascent particles occurs as HDL matures to the heterogenous form in the intravascular compartment.

Recently we reported that VLDL isolated from fasting normotriglyceridemic plasma inhibited the accumulation of apoA-I and the incorporation of [3H]leucine into apoA-I by cultured Hep G2 cells.54 This observation suggested that VLDL (or a subfraction) inhibits hepatic apoA-I production. Gemfibrozil potently reduces circulating VLDL, as evidenced by its well-known hypotriglyceridemic effect. This appears to result from both an inhibition of VLDL transport rate (production) reported in in vivo kinetic studies55 and its catabolism via increased lipoprotein lipase activity.20 Thus, reduction of VLDL levels could result in increased hepatic apoA-I production. This mechanism could be important in hypertriglyceridemic patients, since it has been found that the increase in HDL by gemfibrozil is positively related to pretreatment triglyceride concentrations.56

It is evident from the above considerations that gemfibrozil can increase HDL apoA-I production by three mechanisms: (1) direct hepatic synthesis, (2) diminishing the inhibitory effect of VLDL on hepatic apoA-I synthesis, and (3) enhancing transfer of apoA-I and other components from chylomicrons to HDL during lipoprotein lipase–mediated lipolysis. The extent to which these mechanisms could contribute to final steady state HDL levels is unclear, but they help explain the variable clinical effects that gemfibrozil has on HDL (and its constituents; eg, apoA-I versus cholesterol) concentrations in different patients. Additional factors to be considered in extrapolating from the in vitro to the in vivo situation include the following. First, although Hep G2 cells have been used extensively in studying hepatic modulation of lipoproteins, there could be significant differences with primary hepatocytes, and the findings in this report need to be confirmed in human hepatocyte cultures if possible. Second, hepatic lipoprotein modulation in vivo is also under control of other humoral factors (eg, insulin, growth hormone, and albumin levels) that could influence apoA-I production. Third, recent studies by Rea et al57 indicate that rabbit hepatocytes are inhibited by a distinct factor secreted by nonparenchymal hepatic cells, which line the sinusoids. The existence of such a paracrine control of apoA-I is unclear in the human liver and raises the possibility of additional multifactorial influences on apoA-I production in vivo. In addition to direct hepatic production, gemfibrozil stimulates extrahepatic lipoprotein lipase,20 an enzyme that catabolizes triglyceride-rich chylomicrons (which contain apoA-I of intestinal origin) and VLDL that do not have apoA-I. Lipolysis of chylomicrons would result in apoA-I (and other byproducts) to transfer to HDL,58 thus adding to the pool of circulating apoA-I. Likewise, lipolysis of VLDL would yield donor material (lipids and C apolipoproteins) to HDL.

In summary, our data indicate that gemfibrozil stabilizes hepatic apoA-I transcription, thereby increasing production of functional apoA-I–HDL particles, which mediate reverse cholesterol transport. The data thus define one major mechanism by which gemfibrozil, and possibly certain other fibric acid derivatives, prevents atherosclerotic cardiovascular complications.


*    Selected Abbreviations and Acronyms
 
DMEM = Dulbecco's modified Eagle's medium
FBS = fetal bovine serum
Hep G2 = human hepatocyte cell line
LpA-I = HDL subfraction without apoA-II
LpA-I+A-II = HDL subfraction with apoA-II
PPAR = peroxisome proliferator–activated receptor
PPRE = peroxisome proliferator responsive element


*    Acknowledgments
 
This work was supported in part by a grant from Warner-Lambert, Parke-Davis Co, and by the Long Beach Research Foundation. We thank Tom Wagner for excellent typing assistance, and Marian B. Berman and Mark Hubbard for art work.


*    Footnotes
 
Previously published as preliminary results in abstract form (J Invest Med. 1995;43[suppl 2]:321A).

Received November 8, 1995; revision received March 13, 1996;
*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 up arrowDiscussion
 *References
 

  1. Kashyap ML. Basic considerations in the reversal of atherosclerosis: significance of high density lipoproteins in stimulating reverse cholesterol transport. Am J Cardiol. 1989;63:56H-59H.[Medline] [Order article via Infotrieve]
  2. Fielding CJ, Fielding PE. Cholesterol transport between cells and body fluids. Med Clin North Am. 1982;66:363-373.[Medline] [Order article via Infotrieve]
  3. Maciejko JJ, Holmes DR, Kottke BA. Apolipoprotein A-I is a marker of angiographically assessed coronary-artery disease. N Engl J Med. 1983;309:385-389.[Abstract]
  4. Mendoza SG, Zerpa A, Carrasco H, Colmenares O, Rangel A, Gartside PS, Kashyap ML. Estradiol, testosterone, apolipoproteins, lipoprotein cholesterol, and lipolytic enzymes in men with premature myocardial infarction and angiographically assessed coronary occlusion. Artery. 1983;12:1-23.[Medline] [Order article via Infotrieve]
  5. Rubin EM, Krauss RM, Spangler EA, Verstuyft JG, Clift SM. Inhibition of early atherogenesis in transgenic mice by human apolipoprotein A-I. Nature. 1991;353:265-267.[Medline] [Order article via Infotrieve]
  6. Henrikson T, Evenson SA, Carlander B. Injury to cultured endothelial cells induced by LDL: protection of high density lipoproteins. Scand J Clin Lab Invest. 1979;3:369-375.
  7. Saku K, Ahmad M, Glas-Greenwalt P, Kashyap ML. Activation of fibrinolysis by apolipoproteins of high density lipoproteins in man. Thromb Res. 1985;39:1-8.[Medline] [Order article via Infotrieve]
  8. Yoi Y. Serum prostacyclin stabilizing factor is identical to apolipoprotein A-I (apoA-I). J Clin Invest. 1988;82:803-807.
  9. Plump AS, Scott CJ, Breslow JL. Human apolipoprotein A-I gene expression increases high density lipoprotein and suppresses atherosclerosis in the apolipoprotein E-deficient mouse. Proc Natl Acad Sci U S A. 1994;91:9607-9611.[Abstract/Free Full Text]
  10. Klatsky AL, Armstrong AM, Griedman GD. Alcohol and mortality. Ann Intern Med. 1992;117:646-654.
  11. Canner PL, Berge KG, Wanger NK. Fifteen-year mortality in coronary drug project patients: long-term benefit with niacin. J Am Coll Cardiol. 1986;8:1245-1255.[Abstract]
  12. Frick MH, Elo O, Kaapa K, Heinonen OP, Heinsalmi P, Helo P, Huttunen JK, Kaitaniemi P, Koskinen P, Manninen V, Malkonen M, Manttari M, Norola S, Pasternack A, Pikkarainen J, Romo M, Sjoblom T, Nikkila EA. Helsinki Heart Study: primary-prevention trial with gemfibrozil in middle-aged men with dyslipidemia: safety of treatment, changes in risk factors, and incidence of coronary heart disease. N Engl J Med. 1987;317:1237-1245.[Abstract]
  13. Brown G, Albers JJ, Fisher LF. Regression of coronary artery disease as a result of intensive lipid-lowering therapy in men with high levels of apolipoprotein B. N Engl J Med. 1990;323:1289-1298.[Abstract]
  14. Stampfer MJ, Colditz GA. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med. 1991;20:47-63.[Medline] [Order article via Infotrieve]
  15. Tam S-P, Archer TK, Deeley RG. Effects of estrogen on apolipoprotein secretion by the human hepatocarcinoma cell line, HepG2. J Biol Chem. 1985;260:1670-1675.[Abstract/Free Full Text]
  16. Kashyap ML. Effects of drugs on high density lipoprotein apoprotein metabolism. In: Miller NE, ed. High Density Lipoproteins, Reverse Cholesterol Transport, and Coronary Heart Disease. Selected proceedings. Princeton, NJ: Excerpta Medica (Elsevier) Publishing Co; 1989:60-68.
  17. Bloomfield-Rubins H, Robins SJ, Iwane MK, Boden WE, Fye CL, Gordon DJ, Schaefer EJ, Schechtman G, Wittes JT, for the Department of Veterans Affairs HIT Study Group. Rationale and design of the Department of Veterans Affairs high-density lipoprotein cholesterol intervention trial (HIT) for secondary prevention of coronary artery disease in men with low high-density lipoprotein cholesterol and desirable low-density lipoprotein cholesterol. Am J Cardiol. 1993;71:45-52.[Medline] [Order article via Infotrieve]
  18. Bloomfield-Rubins H, Robins SJ, Collins D, Iranmanesh A, Wilt TJ, Mann D, Mayo-Smith M, Fass FH, Elam MB, Rutan GH, Anderson JW, Kashyap M, Schechtman G. Distribution of lipids in 8500 men with coronary artery disease. Am J Cardiol. 1995;75:1196-1201.[Medline] [Order article via Infotrieve]
  19. The Bezafibrate Infarction Prevention (BIP) Study Group Israel. Lipids and lipoproteins in symptomatic coronary heart disease: distribution, intercorrelations, and significance for risk classification in 6700 men and 1500 women. Circulation. 1992;86:839-848.[Abstract/Free Full Text]
  20. Saku K, Gartside PS, Hynd BA, Kashyap ML. Mechanism of action of gemfibrozil on lipoprotein metabolism. J Clin Invest. 1985;75:1702-1712.
  21. Zannis VI, Breslow JL, San Giacomo TR, Aden DP, Knowles BB. Characterization of the major apolipoproteins secreted by two human hepatoma cell lines. Biochemistry. 1981;20:2089-2096.
  22. Thrift R, Forte TM, Cahoon BE, Shore VG. Characterization of lipoproteins produced by the human liver cell line, Hep G2, under defined conditions. J Lipid Res. 1986;27:236-250.[Abstract]
  23. Forte TM, McCall MR, Knowles BB, Shore VG. Isolation and characterization of lipoproteins produced by human hepatoma-derived cell lines other than HepG2. J Lipid Res. 1989;30:817-829.[Abstract]
  24. Havel RJ, Eder HA, Bragdon JH. The distribution and characterization of ultracentrifugally separated lipoproteins in human serum. J Clin Invest. 1955;34:1345-1353.
  25. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951;193:265-275.[Free Full Text]
  26. Maniatis T, Fritsch EF, Sambrook T. Analysis and cloning of eukaryotic genomic DNA. In: Nolan C, ed. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory; 1989:14-23 ch 2, section 9.
  27. Rothwell TC, Kamanna VS, Jin FY, Koren E, Foley T, Kashyap ML. Characterization of a monoclonal antibody (HB-22) and development of an enzyme-linked immunosorbent assay for human apolipoprotein A-I. Clin Chem. 1995;41:1150-1158.[Abstract/Free Full Text]
  28. Malmendier CL, Lontie JF, Lagrost L, Delcroix C, Dubois DY, Gambert P. In vivo metabolism of apolipoproteins A-IV and A-I associated with high density lipoprotein in normolipidemic subjects. J Lipid Res. 1991;32:801-808.[Abstract]
  29. McFarlane AS. Efficient trace-labeling of proteins with iodine. Nature. 1958;182:53.[Medline] [Order article via Infotrieve]
  30. Shepherd J, Gotto AM, Taunton OD, Claslake MJ, Farish E. The in vitro interaction of human apolipoprotein A-I and high density lipoproteins. Biochim Biophys Acta. 1977;489:486-501.[Medline] [Order article via Infotrieve]
  31. Gupta AK, Ross EA, Meyers JN, Kashyap ML. Increased reverse cholesterol transport in athletes. Metabolism. 1993;42:684-690.[Medline] [Order article via Infotrieve]
  32. Selam JL, Kashyap ML, Gupta AK, Turner D, Wong ND, Lozan JL, Charles MA. Alterations in reverse cholesterol transport associated with programmable implantable intraperitoneal insulin delivery. Metabolism. 1994;43:655-669.[Medline] [Order article via Infotrieve]
  33. Fielding CJ, Fielding PE. Evidence for a lipoprotein carrier in human plasma catalyzing sterol efflux from cultured fibroblasts and its relationship to lecithin:cholesterol acyltransferase. Proc Natl Acad Sci U S A. 1981;78:3911-3914.[Abstract/Free Full Text]
  34. Rothblatt GH, Bamberger MM, Phillips MC. Reverse cholesterol transport. In: Colowich SP, Kaplan NO, eds. Methods in Enzymology. Orlando, Fla: Academic Press; 1986;128:628-644.
  35. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem. 1987;162:156-159.[Medline] [Order article via Infotrieve]
  36. Brown T. Analysis of RNA by Northern and slot blot hybridization. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K,, eds. Current Protocols in Molecular Biology. Brooklyn, NY: John Wiley & Sons; 1993:1-14 ch 4, section 9.
  37. Breslow JL, Ross D, McPherson J, Williams H, Kurnit D, Nussbaum AL, Karathanasis SK, Zannis VI. Isolation and characterization of cDNA clones for human apolipoprotein A-I. Proc Natl Acad Sci U S A. 1982;79:6861-6865.[Abstract/Free Full Text]
  38. Schaap AP, Akhavan H, Romanol J. Chemiluminescent substrates for alkaline phosphatase: application to ultrasensitive enzyme-linked immunoassays and DNA probes. Clin Chem. 1989;35:1863-1864.[Free Full Text]
  39. Hu RM, Chuang MY, Prins B, Kashyap ML, Frank HJL, Pedram A, Levin ER. High density lipoproteins stimulate the production and secretion of endothelin-1 from cultured bovine aortic endothelial cells. J Clin Invest. 1994;93:1056-1062.
  40. Hu RM, Levin ER, Pedram A, Frank HJL. Atrial natriuretic peptide inhibits the production and secretion of endothelin from cultured endothelial cells. J Biol Chem. 1992;267:17384-17389.[Abstract/Free Full Text]
  41. Greenberg ME. Identification of newly transcribed RNA. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K, eds. Current Protocols in Molecular Biology. Brooklyn, NY: John Wiley & Sons; 1993:1-11, ch 4, section 10.
  42. Greenberg ME, Ziff EB. Stimulation of 3T3 cells induces transcription of the c-fos proto-oncogene. Nature. 1984;311:433-438.[Medline] [Order article via Infotrieve]
  43. Dashti N, Wolfbauer G, Alaupovic P. Binding and degradation of human high-density lipoproteins by human hepatoma cell line Hep G2. Biochim Biophys Acta. 1985;833:100-110.[Medline] [Order article via Infotrieve]
  44. Hahn SE, Goldberg DM. Modulation of lipoprotein production in Hep G2 cells by fenofibrate and clofibrate. Biochem Pharmacol. 1992;43:625-633.[Medline] [Order article via Infotrieve]
  45. Tam S. Effects of gemfibrozil and ketoconazole on human apolipoprotein A-I, B, and E levels in two hepatoma cell lines, Hep G2 and Hep 3B. Atherosclerosis. 1991;91:51-61.[Medline] [Order article via Infotrieve]
  46. Puchois P, Kandoussi A, Fievet P, Fourrier JL, Bertr M, Koren E, Fruchart JC. Apolipoprotein A-I–containing lipoproteins in coronary artery disease. Atherosclerosis. 1987;68:35-40.[Medline] [Order article via Infotrieve]
  47. Warden CH, Kedrick CC, Qiao JH, Castellani LW, Lusis AJ. Atherosclerosis in transgenic mice overexpressing apolipoprotein A-II. Science. 1993;261:469-472.[Abstract/Free Full Text]
  48. Barbaras R, Puchois P, Fruchart JC, Ailhaud G. Cholesterol efflux from cultured adipose cells is mediated by Lp A-I particles but not by Lp A-I:A-II particles. Biochem Biophys Res Commun. 1987;142:63-69.[Medline] [Order article via Infotrieve]
  49. Cheung MC, Lum KD