A novel allele in the promoter of the hepatic lipase is associated with increased concentration of HDL-C and decreased promoter activity

Hepatic lipase (HL) is a lipolytic enzyme involved in the metabolism of plasma lipoproteins, especially HDLs. Association of the polymorphisms in the promoter region of the LIPC gene to post-heparin plasma HL activity and the plasma HDL-C concentration has been investigated thoroughly, but to date little is known about this in the Chinese. In the present study, we analyzed the polymorphisms in the promoter region of LIPC gene in Chinese patients with coronary artery disease (CAD) using denaturing high performance liquid chromatography (DHPLC) and DNA sequencing. As the result, a novel single nucleotide polymorphism (cid:2) 586T-to-C was identiﬁed and no linkage of this variant with other polymorphisms in the promoter was found. Compared with the nonsymptomatic control subjects, excess of carriers of the (cid:2) 586T/C substitution were detected in the CAD patients (43% vs. 31%, (cid:3) 2 (cid:4) 4.597, degree of freedom (cid:4) 2, P (cid:4) 0.032).The –586C allele as (cid:10) 2 values for allele associations, were estimated by the Estimating Haplotype-Frequen-cies software program (31); LD coefﬁcients D (cid:8) (cid:12) D/D max were calculated. The lipid phenotypic data among different genotypes were statistically analyzed using the Student’s t -test.

Human hepatic lipase (HL, triacylglycerol, EC3.3.3.3) is a glycoprotein synthesized primarily in hepatocytes. Following secretion, the enzyme binds to the hepatic sinusoidal endothelial surface, where it hydrolyzes triglycerides and phospholipids in plasma lipoproteins (1)(2)(3). HL has been suggested to play a role in HDL metabolism. Evidence demonstrated that HDL-cholesterol (HDL-C) level was at least partly regulated by HL level and on this basis it had been thought that lowering HL would increase HDL-C (4). HL deficiency leads to elevation in HDL-C, increased levels of triglyceride in HDL and LDL, and impaired metabolism of post-prandial glyceride-rich lipoproteins (5)(6)(7)(8), and all of these are considered to be risk factors for premature atherosclerosis. Although HL seems to be an important enzyme with multiple functions, the exact role of HL in lipoprotein metabolism has not yet been established.
There is a considerable body of evidence demonstrating that the rare alleles of some common promoter polymorphisms including Ϫ 250G/A, Ϫ 514C/T, Ϫ 710T/C, and Ϫ 763A/G are associated with marked decreased plasma HL activity and increased HDL-C levels. And some studies showed that these variants have a significant effect on the basal rate of transcription of the LIPC gene (25,26); however, a recent study did not support this conclusion (27). We present evidence of a novel promoter single nucleotide polymorphism (SNP), which is associated with plasma HDL-C concentration and at the same time has significant effect on the transcription of LIPC gene.

Subjects
One hundred sixty male patients ( р 65-years-old) with coronary artery disease were from the West China Hospital, Sichuan University. All of them were examined by coronary angiography using the Judkins technique. Patients having any main coronary artery branch (left anterior descending, left circumflex artery, right coronary artery) with at least one stenosis of Ͼ 60% were included in the study. Meanwhile, 150 unrelated age-matched male subjects selected via health screening at the same hospital, free of any clinical and biochemical signs of CAD, were used as controls for the study.

Measurement of lipids and lipoproteins
Blood samples were collected at baseline from patients and controls after an overnight fast. Plasma separated from cells by centrifugation at 500 g for 10 min at room temperature was used immediately for lipid and lipoprotein analysis. The levels of plasma cholesterol and triglyceride were determined with an enzymatic kit (Boehringer Mannheim) and calibrated with a serum calibrator. HDL-C was measured in the supernatant after precipitation of apoB-containing lipoproteins with a 4% sodium phosphotungstate solution after centrifugation. LDL-C was calculated by use of the Friedewald formula (28).

DNA preparation and PCR amplification
Genomic DNA was isolated from peripheral blood leukocytes using the "salting-out" procedure (29) and stored at 4 Њ C. A 668 bp segment extending from nt ϩ 29 to Ϫ 639 (30) was amplified by PCR with primers 5 Ј -CAGTCCTCTACACAGCTGGAAC-3 Ј and 5 Ј -CGGGGTCCAGGCTTTCTTGGT-3 Ј . Each PCR amplification mixture contained 0.1 g of genomic DNA, 40 pmol of each primer, 25 pmol dNTPs, and standard PCR buffer in a total volume of 50 l. The reaction mixture was preheated at 94 Њ C for 4 min. Subsequently, 0.4 units of Taq polymorase was added. The 30 cycles of PCR amplification was performed with a temperature profile consisting of denaturation for 45 s at 94 Њ C, annealing for 30 s at 62 Њ C, and extension for 30 s at 72 Њ C. The reactions were carried out in a Perkin Elmer GeneAmp 9600 PCR System (Perkin Elmer).

Denaturing high performance liquid chromatography
Denaturing high performance liquid chromatography (DHPLC) screening for base variant was performed on an automated HPLC instrument (Hewlett Packard). The support for the stationary phase was made of a specially prepared wide-pore silica with a dense layer of an aliphatic organosilane packed into a 50 mm-X4.6-mmId dsDNA-analysis column (Hewlett Packard). The mobile phase was 0.1 mol/l trietylammonium acetate (PE BioSystems) buffer at pH 7.0 containing ethylenediamine tetraacetic acid (0.1 mmol/l).
PCR products were eluted with a linear acetonitrile gradient. The start and end points of the gradient were adjusted according to the size of the PCR products. The temperature required for successful resolution of heteroduplex molecules was predicted by the DHPLC algorithm available at http://insertion.stanford.edu/melt.html. In the present study, the appropriate temperature of analysis for amplicon was determined empirically by running it at different temperatures until a good resolution between homo-and heteroduplexes was obtained. The temperature of DHPLC for this amplified fragment is 63 Њ C. The heteroduplex molecules are generally eluted ahead of homoduplex, therefore the appearance of additional peaks or shoulders during DHPLC was interpreted as indicative of a single base mismatch in heteroduplex DNA fragments and, therefore, reason for need of sequencing.

DNA sequencing and genotype assay
The location and chemical nature of the mismatch was confirmed by sequencing of the re-amplified product. The heterozygous and homozygous samples were cloned in T-Easy vector (Promega), then sequenced in both directions on the ALFexpress DNA automated sequencer, using the dye-terminator cycle thermal sequenase sequencing kit (USB company).
A T-to-C substitution at nucleotide Ϫ 586, introducing a new Acc I restriction site, was confirmed by restriction analysis. Genotyping for the Ϫ 250G/A and Ϫ 514C/T polymorphisms was performed by digesting the 668 bp PCR fragment with Dra I and Nla III, respectively. Electrophoresis was performed on a 2% agarose gel containing ethidium bromide.

Generation of luciferase reporter constructs
For in vitro promoter studies, the 5 Ј -flanking region of the LIPC gene was amplified with upstream primer 5 Ј -TCATGGTAC-CCAGTCCTCTACACAGCTGGAAC-3 Ј and downstream primer 5 Ј -TAGTCTCGAGCGGGGTCCAGGCTTTCTTGGT-3 Ј , each of them contains a Kpn I and Xho I site introduced to the 5 Ј end (restriction sites underlined), respectively. The T allele and the C allele were amplified from individuals who differed in LIPC promoter structure, respectively.
The PCR products were digested with Kpn I and Xho I and purified by electrophoresis on a 1.5% low-melting-point agarose gel followed by magic PCR preps DNA purification system (Promega). The purified mutant and wild-type LIPC promoter sequences were cloned into promoterless pGL-3 enhancer vector (Promega). The vector containing either T allele or C allele was designated as pGL3-T or pGL3-C, respectively. Plasmid DNA was obtained by transforming the vector into JM109 cells and subsequent largescale plasmid preparation using Qiagen plamid kit (Qiagen). Reporter constructs were sequenced prior to use in reporter assay. The statistical significance of differences between control subjects and patients was tested by Pearson 2 analysis.

Cell culture and transient transfection
The human hepatoma cell line HepG2 was plated into 45 mm culture dishes at 20% confluence in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum (GibcoBRL, the Netherlands). In each experiment, three different luciferase reporter plasmids were transfected: a ) pGL3-T; b ) pGL3-C; and c ) pGL3-control (Promega), which contains SV40 promoter and enhancer sequences. The DNA mixture for transfection was composed of test plasmid (0.75 g) and pSV-␤ -galactosidase control vector (Promega; 0.04 g) that served as internal control to normalize activities of luciferase. Transfection was carried out using the liposome method (Gene Therapy Systems, San Diego, CA). Luciferase activity was measured with a luminometer (Model TD-20/20; Promega), and the ␤ -galactosidase activity was determined as described by the suppler (Promega). To correct for transfection efficiency, light units from the luciferase assay were divided by the absorbance reading from the ␤ -galactosidase assay. The corrected LIPC promoter-driven luciferase activity is expressed as a percentage of the pGL3-control SV40 promoterdriven luciferase activity that served as the positive control in every transfection experiment. Luciferase activity was expressed as relative luciferase units. The promoterless pGL3-basic vector (Promega) lacking promoter and enhancer was used as a negative control in each of the transfection experiment. All constructs were tested in triplicate in three independent transfection experiments.

Statistical analysis
The assumption of Hardy-Weinberg equilibrium was tested by means of gene counting and 2 analysis. Haplotype frequencies for pairs of alleles, as well as 2 values for allele associations, were estimated by the Estimating Haplotype-Frequencies software program (31); LD coefficients DЈ ϭ D/D max were calculated. The lipid phenotypic data among different genotypes were statistically analyzed using the Student's t-test.

A novel polymorphism T-to-C (HL-586) in the LIPC promoter
Screening for base variant of the 668 bp region in the 5Ј flanking of the LIPC gene with DHPLC in CAD patients and controls revealed that there is a variation in some samples. As is known, any mismatched base pair in a heteroduplex molecule is generally eluted ahead of the homoduplex, resulting in one additional DHPLC peak (data not shown). The character of varied base was then identified by sequence analysis. As the result, a new base variation, namely Ϫ586T-to-C transition was discovered (Fig. 1).

Allele frequencies and the linkage disequilibrium of the LIPC promoter
To determine the prevalence of the Ϫ586T/C, Ϫ514C/T, and Ϫ250G/A polymorphisms, we screened them in all the 160 CAD patients and 150 controls. The genotype distribution and allele frequencies are listed in Table 1. No deviation from Hardy-Weinberg equilibrium was noted in both CAD and control groups. As the result, excess of carriers of the Ϫ586T/C substitution were detected in the CAD patients compared with the nonsymptomatic control subjects (43% vs. 31%, 2 ϭ 4.597, df ϭ 2, P ϭ 0.032). The prevalence of the -586C allele was significantly higher in the CAD patients than in control subjects ( 2 ϭ 4.575, df ϭ 1, P ϭ 0.032). The extent of D in pairwise combinations of alleles at the LIPC promoter locus was estimated by means of maximum likelihood from the frequency of diploid genotypes in the CAD and control groups. Haplotype frequencies and the disequilibrium statistics D, DЈ, and D max or D min are given in Table 2. It is clear that the disequilibrium values for Ϫ586/Ϫ514 and Ϫ586/Ϫ250 pairs do not differ significantly from zero. So the Ϫ586 variant is a novel mutation.

Association between T-to-C substitution and plasma lipids
Studies on the relation between T-to-C (HL-586) and plasma lipid showed that neither cholesterol and triglyceride nor LDL-C was significantly different between subjects with or without this gene variant. However, HDL-C levels did differ among different genotypes(P ϭ 0.011). The subjects homozygous for the the C allele (HL-586) had the highest HDL-C values (1.13 Ϯ 0.24 mmol/l), and subjects homozygous for the Ϫ586T allele had the lowest (0.91 Ϯ 0.34 mmol/l), while the heterozygote had the intermediate value (0.98 Ϯ 0.43 mmol/l) ( Table 3).

Effects of Ϫ586 T/C polymorphism on promoter activity
To examine the potential effects of the Ϫ586T/C polymorphism on the rate of transcription of the LIPC gene in vitro, a 668-bp promoter fragment of LIPC gene (nt ϩ29 to Ϫ639) carrying either T or C allele was inserted upstream of the luciferase gene in the pGL3 promoterless enhancer plasmid vector. The activity of LIPC T/C promoter-luciferase reporter gene constructs were assessed in transient transfection assay in HepG2 human hepatoma cells. Triplicate experiments were performed using DNA from different plasmid preparations. The luciferase light units in these assay ranged between 13,000 to 16,000 for the Ϫ586T construct and from 7,200 to 8,900 for the Ϫ586C construct. As shown in Fig. 2, significantly lower luciferase activities were observed for the pGL3-C construct as compared with the pGL3-T construct (a 56% decrease; P Ͻ 0.001). These results indicate that Ϫ586T/C polymorphism has a major influence on the basal transcription rate of the LIPC gene.

DISCUSSION
In the present study, a novel base variation (Ϫ586T/C) in the LIPC promoter region was found by DHPLC and DNA sequencing. This polymorphism was present in about 43% of patients with angiographically established coronary artery disease and in about 31% of nonsymptomatic control subjects. The T-to-C allele was significantly more frequent in the patient with CAD than in the control subjects.
There is considerable evidence that HL activity is an important determinant of plasma HDL-C concentrations. Clinical studies have consistently found an inverse relationship between HL activity measured in post heparin plasma and plasma HDL-C concentrations (22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37); and drugs, such as anabolic steroids, that increase HL activity cause a proportional reduction in plasma levels of HDL-C (38). Association studies showed that the Ϫ586T/C variation may account for the variation in plasma HDL-C concentration, at least in the tested Chinese. Since we did not measure the HL activity in the present study, we can only speculate that the Ϫ586 polymorphism may affect the activity of this enzyme and thereby influence the plasma HDL-C. Given the well established inverse relationship between HL activity and HDL-C concentrations, however, it seems very likely that the Ϫ586C is associated with low HL activity by directly affecting HL expression or through linkage disequilibrium with another polymorphism that directly decreases the enzyme activity. Since no linkage of the Ϫ586T/C variant with other polymorphisms in the first 668 bp of the LIPC promoter was found, it suggests that the base substitution may lead to a lowered HL expression. To determine whether LIPC allele containing Ϫ586C does influence the transcription rate of LIPC gene, the transient transfection studies in HepG2 cells were conducted. The results from the study strongly suggested that the T/C (HL-586) base substitution did have a significant contribution to the basal rate of transcription of the LIPC gene. This observation provided new strong evidence that the single nucleotide polymorphism in HL gene may be one of major factors in determination of plasma HDL-C levels.
The promoter sequence variant Ϫ514T in the HL gene has been shown to be significantly associated with low post-heparin HL activity (18,(21)(22)(23)(24). Some studies have also found that the Ϫ514T variant is associated with elevation in plasma HDL-C (17)(18)(19)(20)(21)23). It has generally been assumed that these associations are due to the change in rate of transcription of the LIPC gene because of the promoter polymorphisms. However, recent studies in vitro showed contradictory results (25)(26)(27). We tested for associations of the same LIPC Ϫ514T with plasma lipoprotein traits in Chinese CAD patients and normal controls. The LIPC Ϫ514T allele frequencies in these two groups were 0.231 and 0.300, respectively. No significant association was found between LIPC Ϫ514T and plasma HDL-C after adjusting for covariates including gender and body mass index, although the plasma HL activity was not available for analyses. There was no consistent relationship between the population mean plasma HDL-C concentration and the population LIPC Ϫ514T frequency. Our findings are consistent with the results from the Canadian populations (39), suggesting that the common promoter variation in LIPC, which has been reported to be associated with variation in post heparin HL activity and HDL triglyceride concentration, is not always associated with variation in plasma HDL-C concentration, possibly due to yet unspecified environmental or genetic factors.
Little is known about how the expression of the LIPC gene is regulated. The promoter region of the human LIPC gene has been characterized, and possible regulatory elements were identified by searching for consensus sequences. Transfection studies have indicated that multiple elements in the proximal promoter influence the transcription of the LIPC gene (30,40,41), but none of these elements have been identified and/or analyzed yet in detail.
In summary, we show here that a novel variant in the LIPC promoter is functional. The molecular mechanism and the transcription factors involved remain to be established.
This study was funded by National Natural Science Foundation of China grant 39993420 (S.Z.); and the State 863 High Tech-