Functional Characterization of 4 Polymorphisms in Promoter Region of Hepatic Lipase Gene

Subjects

The proximal promoter of the HL gene was sequenced with the use of DNA samples from 25 healthy subjects with either low, intermediate, or high postheparin plasma HL activity.¹⁹ Genotyping for the polymorphisms encountered was subsequently performed in a total of 186 apparently healthy men, aged 35 to 45 years (mean±SD 40.3±3.5 years). They were selected at random from a registry containing all permanent residents in Stockholm County. Of those initially invited, 81% agreed to participate in the research program. All the men were interviewed to exclude individuals with a history of cardiovascular disease. Other exclusion criteria were the presence of severely impaired renal function, arteritis, collagenosis, diabetes mellitus, a history of alcohol abuse or other forms of addiction, and non-Swedish origin of the subject. All subjects gave informed consent to their participation in the study, the protocol of which had been approved by the ethics committee of the Karolinska Hospital.

Gene Sequencing

For DNA procedures, nucleated cells from frozen whole blood were prepared according to Sambrook et al,²⁰ and DNA was extracted by a salting-out method.²¹ For the nucleotide sequencing of the promoter of the HL gene, a 1400-bp section of the proximal promoter, spanning from positions −1267 to 133, was amplified by polymerase chain reaction (PCR) with the forward primer (5′-CTCTGCTTAAAATCCCTACCTTC) and the reverse primer (5′-ACCTGGTTTCAGGCTTTGTC). This PCR fragment was used as a template for further amplifications as part of the Taq DyeDeoxy Terminator Cycle sequencing system (Perkin-Elmer, Applied Biosystems Division). Nested primers, designed on the basis of the published sequence of the promoter of the HL gene, were used for the analysis of overlapping sections of 200 to 300 bp in both directions.

Genotyping

Genotyping for the −250G/A and −514C/T polymorphisms was performed with use of a PCR fragment amplified with the forward primer (5′-GGATCACCTCTCAATGGGTC) and the reverse primer (5′-ACCTGGTTTCAGGCTTTGTC), followed by digestion with the restriction enzymes DraI and NlaIII, respectively. The −710T/C and −763A/G polymorphisms were genotyped with use of a PCR fragment amplified with the forward primer (5′-TCCTGGCCAGAAATCTCTTC) and the reverse primer (5′-GACCCATTGAGAGGTGATCC), followed by digestion with the restriction enzymes AvaII and SphI, respectively.

Electrophoretic Mobility Shift Assay

Nuclear extracts were prepared according to Alksnis et al.²² All buffers were freshly supplemented with leupeptin (0.7 μg/mL), aprotinin (16.6 μg/mL), phenylmethylsulfonyl fluoride (0.2 mmol/L), and 2-mercaptoethanol (0.33 μL/mL). The protein concentrations in the extracts were estimated by the method of Kalb and Bernlohr.²³ Incubation for EMSA was conducted as described,²⁴ and the reaction products were applied to 7% (wt/vol) polyacrylamide gel (acrylamide/N,N′-methylene-bis-acrylamide weight ratio 80:1), and then electrophoresis was performed in 22.5 mmol/L Tris/22.5 mmol/L boric acid/0.5 mmol/L EDTA buffer for 2.5 hours at 200 V. Nonradioactive competitor DNAs, which were either identical, of the opposite allelic variant, or of nonspecific origin, were added in 100-fold excess of the labeled DNA.

DNA Constructs

Two sets of double-stranded oligonucleotides were constructed, constituting the 30-bp sequence around the −710T/C polymorphism, flanked by BamHI and BglII ends. The double-stranded oligonucleotides were ligated head to tail into a BamHI-digested HCAT vector.²⁵ The correct sequence and orientation of the inserts were tested by DNA sequencing. pCAT plasmids were constructed with use of 945-bp promoter fragments, spanning from −895 to 50, ligated into a pCAT-basic vector as described by the supplier (Promega Corp). The promoter fragments were obtained by PCR amplification of DNA samples from subjects homozygous for the −250A/−514T/−710C/−763G haplotype or homozygous for the wild-type haplotype with use of the forward primer (5′-AACTGCAG-TCCTGGCCAGAAATCTCTTC) and the reverse primer (5′-GCTCTAGAGTCCATTTCTCCGTTTCACC). The correct sequence of the inserts was tested by DNA sequencing. Plasmids specific for the −250A, −514T, −710C, and −763G mutations were generated with the QuickChange site-directed mutagenesis kit (Stratagene Cloning Systems) with use of the wild-type pCAT plasmid described above.

Transient Transfection Assay

Human hepatoblastoma (HepG2) cells were cultured in 90-mm dishes in DMEM supplemented with 10% FCS. Confluent cells were transfected by the calcium-phosphate DNA coprecipitation method, essentially as described.²⁰ The pSV-β-galactosidase gene (Promega) was cotransfected as an internal control. In all experiments, 5 μg of the CAT construct and 5 μg of the β-galactosidase plasmid were added to the medium. CAT activity was analyzed by the method of Sambrook et al²⁰ and quantified with a PhosphorImager (Fuji Photo Film Co). β-Galactosidase activity was determined as described by the supplier (Promega). CAT levels were expressed in arbitrary units after standardization for β-galactosidase activity. All constructs were tested in triplicate in 4 to 8 independent transfection experiments.

Statistical Methods

Allele frequencies were estimated by gene counting.

Detection of 4 Common Polymorphisms in Promoter Region of HL Gene

A 1267-bp section of the proximal promoter of the HL gene was sequenced in both directions with use of DNA samples from 25 subjects with either low, intermediate, or high postheparin plasma HL activities. The sequence analysis of the present study agreed completely with the nucleotide sequence determined by Oka et al,²⁶ whereas several differences were observed with other published nucleotide sequences.^{8 9} In agreement with the report by Guerra et al,¹⁵ 4 polymorphisms were found, located at positions −250 (G-to-A substitution), −514 (C-to-T substitution), −710 (T-to-C substitution), and −763 (A-to-G substitution). No additional polymorphisms were discovered. Several homozygotes for the rare alleles were found among subjects with low postheparin HL activity. Heterozygotes were found in subjects with either low or intermediate postheparin HL activity, whereas none of the subjects with high postheparin HL activity displayed any of the 4 rare alleles.

Assays were then developed for all 4 polymorphisms in the promoter of the HL gene, and DNA samples from 186 healthy middle-aged men were analyzed. All polymorphisms were found to be in Hardy-Weinberg equilibrium. The frequencies of the rare −250A, −514T, −710C, and −763G alleles were 20.7%, 19.9%, 20.7%, and 20.7%, respectively. Complete allelic association was observed for the −250G/A, −710T/C, and −763A/G polymorphisms. The −514C/T polymorphism was also in strong linkage disequilibrium with the other polymorphisms, but in DNA samples from 3 subjects, the rare −250A/−710C/−763G haplotype was found in association with the common −514C allele.

Allele-Specific Binding of Nuclear Protein(s) to the −710T/C Polymorphic Site

The potential physiological roles of the −250G/A, −514C/T, −710T/C, and −763A/G polymorphisms in the promoter of the HL gene were analyzed by EMSA. No evidence was found for differences in binding patterns of nuclear protein(s) when the wild-type and the mutant alleles of the −250G/A, −514C/T, and −763A/G polymorphisms were analyzed (data not shown). However, a distinct difference in binding pattern was observed between the wild-type −710T and the mutant −710C fragments (Figure 1A⇓). The protein-DNA complex indicated with an arrow in Figure 1A⇓ was present at considerably higher concentrations when the mutant −710C fragment was compared with the wild-type −710T fragment. Quantitative analyses of this complex demonstrated significant differences between the mutant and wild-type fragments at all nuclear extract concentrations tested (Figure 1B⇓). Competition studies showed that a 100-fold excess of the unlabeled −710T or −710C fragment substantially reduced the interaction of the labeled −710T fragment with the nuclear proteins (Figure 1C⇓). Similar results were obtained in competition studies that used the −710C fragment. It is noteworthy that the mixture of labeled and unlabeled −710C and −710T fragments generated an additional protein-DNA complex in these competition experiments (Figure 1C⇓, lanes 4 and 8). The nature of this complex is presently unknown. Taken together, the results of the EMSA studies provided evidence of a difference between the −710T and the −710C fragments regarding the binding of ≥1 nuclear proteins.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Differential binding of nuclear proteins to the −710T/C polymorphic site. A, EMSA of nuclear extract derived from HepG2 cells bound to a 30-bp DNA fragment containing either the −710T site (lanes 1 to 4, indicated with N) or the −710C site (lanes 5 to 8, indicated with M) of the HL promoter. Arrow refers to the −710C-allele predominant factor. F denotes free DNA. Lanes are as follows: 1 and 5, without extract; 2 and 6, 50 μg/mL HepG2 extract; 3 and 7, 100 μg/mL HepG2 extract; and 4 and 8, 200 μg/mL HepG2 extract. B, Quantitative analysis of the protein-DNA complex indicated with arrow in panel A. The protein-DNA complex was quantified in 3 independent experiments by PhosphorImager. The intensity of the protein-DNA complex was expressed as pixels/mm² (PSL/mm²). Significant differences in intensities were observed at all 3 HepG2 extract concentrations tested (P<0.01). C, EMSA of nuclear extract derived from HepG2 cells bound to the −710T site (lanes 1 to 5, indicated with N) or the −710C site (lanes 6 to 9, indicated with M) in the presence of unlabeled DNA as competitor. Arrow refers to the −710T-allele predominant factor. F denotes free DNA. Lanes are as follows: 1, without extract; 2 and 6, 200 μg/mL HepG2 extract in the absence of competitor; and 3 to 5 and 7 to 9, 200 μg/mL HepG2 extract in the presence of 100-fold excess of unlabeled DNA as competitor. Competitors used were as follows: −710T site (lanes 3 and 8, indicated with N), −710C site (lanes 4 and 7, indicated with M), and nonrelated 30-bp fragment (lanes 5 and 9, indicated with X).

−710T/C Polymorphism and Transcription of HL Gene

Transient transfection studies in HepG2 cells were conducted to explore whether the −710T/C polymorphism influences the rate of transcription of the HL gene in vitro. CAT activities were compared between constructs harboring either 1 or 2 tandemly arranged 30-bp fragments of the HL promoter containing either the −710T or the −710C sites. As shown in Figure 2⇓, no differences in CAT activities were observed between constructs containing either of the 2 sites. These results indicate that the difference in binding pattern of nuclear protein(s) between the −710T and −710C alleles, as observed in EMSA, does not affect the rate of transcription of the HL gene.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

No difference in transcriptional activity of the −710C allele. CAT activities of constructs harboring 1 or 2 tandemly arranged 30-bp fragments with either the −710T or the −710C site were compared in transfection studies with HepG2 cells. Constructs were tested in triplicate in 4 independent experiments. The CAT activities of the −710C-1x and −710C-2x constructs were expressed relative to the CAT activities of the −710T-1x and −710T-2x constructs, respectively, with the latter jointly indicated as wild type (WT). No significant differences in CAT activities between the different constructs were observed.

Presence of Rare Alleles of All 4 HL Promoter Polymorphisms Does Not Influence Basal Rate of Transcription of HL Gene

In subsequent transfection experiments, the potential effects of the promoter polymorphisms on the basal rate of transcription of the HL gene were analyzed with the use of 945-bp promoter constructs. CAT activities were compared between wild-type promoter constructs and promoter constructs with the mutant haplotype, eg, constructs containing the rare alleles of all 4 promoter polymorphisms (−250A/−514T/−710C/−763G haplotype). Considerable variation in the results of these experiments was observed, which made it difficult to determine the definite roles of the promoter polymorphisms in terms of their influence on transcription of the HL gene. Two different plasmid constructs were therefore analyzed; each was isolated independently in 2 preparations and tested separately in 2 independent experiments. As shown in Figure 3⇓, no difference in CAT activities was observed between the mutant haplotype construct and the wild-type construct in 8 independent experiments.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

No difference in transcriptional activity between WT and mutant (M) HL promoter constructs. CAT activities of promoter constructs containing the WT haplotype (−250G/−514C/−710T/−763A) or the M haplotype (−250A/−514T/−710C/−763G) were compared in transfection studies with HepG2 cells. Constructs were tested in triplicate in 8 independent experiments. The CAT activity of the M construct was expressed relative to the CAT activity of the WT construct.

Individual Contribution of −250A, −514T, −710C, and −763G Mutations to Basal Rate of Transcription of HL Gene

Additional transfection experiments were conducted to evaluate the contribution of each of the 4 polymorphisms to the basal rate of transcription of the HL gene. To this end, promoter constructs with specific mutations were generated by using site-directed mutagenesis. All 4 promoter constructs were compared with the wild-type construct in transient transfection studies in HepG2 cells. As shown in Figure 4⇓, no significant differences were observed between the −250A, −514T, −710C, and −763G constructs and the wild-type construct. In all, the transfection studies indicated that none of the 4 polymorphisms of the HL promoter has a major influence on the basal rate of transcription of the HL gene.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

No difference in transcriptional activity between the 4 HL promoter polymorphisms. CAT activities of promoter constructs containing the WT haplotype (−250G/−514C/−710T/−763A) or any of the −250A, −514T, −710C, or −763G mutations generated by site-directed mutagenesis were compared in transfection studies with HepG2 cells. Constructs were tested in triplicate in 4 independent experiments. The CAT activities of the mutant constructs were expressed relative to the activity of the WT construct. No significant differences between the CAT activities of the different constructs were observed.