A Common Promoter Polymorphism in the Hepatic Lipase Gene (LIPC-480C>T) Is Associated With an Increase in Coronary Calcification in Type 1 Diabetes

Type 1 diabetic subjects (ages 20–55 years) were referred for participation in a pilot study of coronary calcification in type 1 diabetes. All subjects were asymptomatic for coronary disease and had no history of coronary artery bypass or angioplasty. DNA was genotyped in 97 subjects. Results from 91 Caucasian subjects are presented in this study. All subjects provided informed consent, and the protocol was approved by the Institutional Review Board.

Blood was collected from individuals who had fasted for at least 12 h and placed in tubes containing 0.1% EDTA. The sample was centrifuged immediately at 4°C. Plasma was maintained at 4°C until assayed for lipids. White cells were stored at 4°C before DNA extraction.

Cholesterol and triglyceride levels were measured using enzymatic methods, HDL was separated using dextran sulfate, and LDL cholesterol was calculated based on the Friedewald equation (LDL = total cholesterol -- HDL -- triglycerides/5). If the triglyceride level was >400 mg/dl, LDL cholesterol was calculated as the difference between the cholesterol in the d <1.006 g/ml fraction minus the cholesterol in the dextran sulfate supernatant.

Genomic DNA was extracted from leukocytes after erythrocyte lysis with 13.1 mmol/l ammonium chloride and 0.9 mmol/l ammonium bicarbonate by salt deproteinization, and resuspended in 10 mmol/l Tris-HCl, 0.2 mmol/l Na₂ EDTA (pH 7.5). The LIPC-480 promoter site (21,29) was genotyped in the context of a larger multiplex assay, the basis of which has been previously described (30). The multiplex assay included single nucleotide polymorphisms (SNPs) in the following genes in addition to LIPC: ApoAIV, ApoB, ApoCIII, ApoE, β3Ad-R, CETP, LDL-R, LTA, Lp(a), LpL, PON1, PON2, and PPARγ. None of these SNPs were significantly related to CAC in univariate analysis (all P > 0.10). CETP-TaqIB, genotyped separately, was related to CAC (P = 0.02).) In brief, biotinylated primer pools were used to simultaneously amplify multiple targets, and the amplified alleles were detected colorimetrically using sequence-specific oligonucleotide probes that had been immobilized in a linear array. LIPC primers defining a 145-bp product were one pair of a total of 24 primer pairs in a single reaction. The specificity of the immobilized probes was confirmed using genomic targets of known genotype, as determined by sequencing.

High-resolution, noncontrast, contiguous 3-mm tomographic images were acquired at a 100-ms exposure time using an Imatron C-150XLP EBCT scanner (Imatron, San Francisco, CA). Scanning started from near the lower margin of bifurcation of the main pulmonary artery and proceeded caudally until the entire epicardial system was imaged during a single held breath, typically 35–45 s in duration. Prospective electrocardiogram triggering was set at 80% of the R-R interval. The threshold for CAC was set at a density of 130 Hounsfield units (HU) in at least one pixel (an area >0.51 mm²) (2). The region of interest was encircled within each coronary artery, and computer-driven measurement of the lesion area and maximum density were recorded according to the method described by Agatston. A score for each region is calculated by multiplying the area by density score (one for 130–199 HU, two for 200–299 HU, three for 300–399 HU, and four for >399 HU). A total CAC score in Agatston units (AU) was calculated by adding up scores for all slices (typically 35–40 images acquired) and for left main coronary artery, left anterior descending coronary artery, circumflex artery, and right coronary artery.

Allele frequency was determined by counting, and Hardy-Weinberg equilibrium was tested using χ² analysis. Differences in mean values of quantitative traits were determined by two-tailed Student’s t test or Mann-Whitney test if the quantitative data were not normally distributed. Triglyceride levels were transformed (natural logarithm) for statistical analysis (untransformed data are presented). Sex differences were assessed by χ² analysis (Yates corrected). Significance of the percent of subjects with CAC by LIPC-480 genotype and allele frequencies among categories of CAC were determined using Cochrane-Armitage χ² test for linear trend. CAC categories were defined as no detectable calcium (CAC = 0 AU; n = 40), minimal calcium (CAC >0 to 20 AU; n = 36), moderate calcium (CAC >21 to 100 AU; n = 6), and extensive calcium (CAC >100 AU, n = 9). Although the cut points of CAC = 0 AU and CAC = 100 AU are well established, there is no consensus for cut points for minimal and moderate calcium (ranging from 5 to 50 AU). Analysis of LIPC-480 T allele frequencies using different cut points for minimum and moderate calcium did not alter the relation between LIPC-480 T allele and extent of CAC.

The a priori hypothesis of a relationship between the LIPC-480C>T genotypes and CAC was tested using multiple logistic regression with CAC yes/no as the dependent variable. LIPC-480 genotype was coded as LIPC-480CC = 1, CT = 2, and TT = 3. Significance level was set at α = 0.05.

A total of 97 subjects with type 1 diabetes provided samples for the current study; results are reported on all Caucasian subjects (n = 91) (Table 1). These subjects had a mean age of 38 years (range 23–56 years) and a duration of diabetes of 22 years. The LIPC-480 allele frequency was 0.24, similar to reports from other Caucasian populations (21,23).

Coronary calcification was detectable in 51 (56%) of these type 1 diabetic subjects. The distribution of CAC was highly skewed (Fig. 1); 84% of the subjects had CAC values <20 AU, and 10% of these subjects had extensive CAC, with values >100 AU.

Factors that were related to the presence of coronary calcification in type 1 diabetes are presented in Table 2. Subjects with CAC >0 AU were older (40.0 ± 1.0 vs. 36.1 ± 1.3 years; P = 0.02) and had a significantly longer duration of diabetes (26.2 ± 1.3 vs. 17.8 ± 1.4 years; P < 0.001). In addition, diastolic blood pressure (79.7 ± 1.0 vs. 76.0 ± 1.4 mmHg, P = 0.04) and triglyceride levels (101 ± 17.3 vs. 66 ± 7.6 mg/dl; P = 0.04) were significantly higher, and HDL cholesterol (55.7 ± 2.4 vs. 61.0 ± 2.5 mg/dl; P = 0.05) was significantly lower in CAC subjects. HbA_1c, insulin dosage, systolic blood pressure. and LDL cholesterol were not related to the presence of CAC in this study population.

The LIPC-480 T allele frequency was 0.24 in the population (Table 1), and the LIPC-480 genotypes did not deviate from Hardy-Weinberg expectations (P = 0.92). The LIPC-480 T allele was significantly more common in subjects with coronary calcification (31 vs. 14%; P = 0.006) (Table 2). There was a significant difference in the proportion of subjects with CAC >0 AU by LIPC-480 genotype (P = 0.02 by Fisher’s exact test). The percent of subjects with CAC >0 AU increased from 44% (24/54) among subjects with the LIPC-480CC genotype to 71% (22/31) in LIPC-480CT genotype to 83% (5/6) in homozygous LIPC-480TT subjects (P < 0.007 for Cochrane-Armitage test of linear trend) (Fig. 2). There was no significant difference in total cholesterol, LDL cholesterol, HDL cholesterol, or triglyceride levels or in blood pressure among LIPC-480 genotypes.

Subjects were categorized into four groups based on the extent of their coronary calcification (Fig. 3). The LIPC-480 T allele was greater within the group with higher CAC scores (test for trend, P = 0.007). The major increase in the LIPC-480 T allele frequency (a doubling from 0.14 to 0.29) was from the category with no coronary calcification to the group with CAC >0 to 20 AU. Subjects with CAC of >20 to 100 AU and >100 AU had even higher LIPC-480 T allele frequencies (0.31 and 0.39, respectively).

To determine the independent effect of the LIPC-480 allele on the presence of coronary calcification, multiple logistic regression analysis was performed (Table 3). The LIPC-480 genotype was significantly, independently related to the presence of CAC (OR = 2.90, 95% CI 1.22–6.92, P < 0.05) after accounting for the effects of age, duration of diabetes, sex, diastolic blood pressure, HDL, and triglycerides. Duration of diabetes was the only other factor that was independently related to CAC, whereas age, sex, diastolic blood pressure, HDL, and triglycerides were not independently related to CAC.

This study is, to our knowledge, the first to show that a common polymorphism in the promoter region of the hepatic lipase gene (LIPC-480C>T) is associated with a greater than twofold increase in the presence of CAC among subjects with type 1 diabetes. This relation is independent of HDL cholesterol levels. In addition, there is a striking dosage-response relation. The LIPC-480 T allele frequency is higher as the extent of CAC is greater. These results suggest that the LIPC-480 T allele increases the susceptibility to subclinical coronary disease in type 1 diabetes.

It is important to note that the less-frequent allele (LIPC-480 T) is associated with greater CAC. This is despite the known anti-atherogenic effects of this allele on HDL₂ and LDL size. The polymorphism is associated with lower hepatic lipase (21–24) and higher HDL levels (21,22), primarily HDL₂(23,25,26). In addition, this polymorphism is associated with less small, dense LDL (23). Despite these reported favorable effects on lipoproteins, the results presented here suggest the LIPC-480 T allele is associated with more subclinical coronary disease, perhaps indicating alternate pro-atherogenic effects of the LIPC promoter variant.

One possible explanation for the pro-atherogenic effect of the LIPC-480 T allele may relate to nonlipolytic functions of hepatic lipase. Hepatic lipase appears to have a role in mediating remnant lipoprotein uptake (31,32). Higher hepatic lipase is associated with lower levels of intermediate-density lipoproteins (IDLs), even when the enzyme is lipolytically inactive (33). This suggests an important function for hepatic lipase in remnant clearance in addition to its function as a lipase. Therefore, the lower hepatic lipase levels associated with the decreased expression by the LIPC promoter variant may lead to an increase in IDL, an atherogenic lipoprotein class (34–36).

Previous data have also suggested a pro-atherogenic role for low hepatic lipase. It has been reported that the extent of coronary atherosclerosis is inversely associated with hepatic lipase levels (37), and that atherosclerotic lesion progression is associated with lower hepatic lipase (38). In high-risk subjects with familial hypercholesterolemia, hepatic lipase was inversely associated with the amount of coronary calcification (39). On the other hand, a decrease in hepatic lipase with intensive lipid-lowering therapy is associated with less angiographic lesion progression (40). The exact nature of the relation between hepatic lipase and atherosclerosis remains unclear.

Evidence is emerging that the LIPC promoter polymorphisms are related to coronary atherosclerosis. In the REGRESS study, the LIPC-480 T allele was significantly higher among coronary disease patients compared with asymptomatic control subjects (21). As pointed out by others (28), the genotype distribution at LIPC-480 was not in Hardy-Weinberg equilibrium in the REGRESS population. This could be explained by the marginally smaller number of LIPC-480CC subjects in the coronary disease population (P = 0.06), suggesting that the LIPC-480CC genotype is associated with decreased susceptibility to coronary disease. Others have not been able to demonstrate a significant difference in LIPC promoter allele frequencies between coronary disease patients and control subjects (24,41). Recent preliminary data presented in abstract form have indicated an increase in CHD associated with the LIPC-480 T allele in Caucasian (42,43) and Hispanic subjects (43). Dugi et al. (44) have shown that the extent of coronary atherosclerosis by angiography is inversely related to hepatic lipase levels and is higher in subjects with the LIPC promoter rare allele. A recent report indicated that the LIPC-480 T allele is associated with a decrease in coronary flow reserve (45), a measure of endothelial dysfunction (46). In addition, the LIPC promoter genotype predicts changes in coronary disease stenosis. Homozygous TT subjects have an increased progression of coronary stenosis on lipid-lowering therapy, whereas those with the CC genotype show significant regression in coronary stenosis on the same therapy (47). The current study adds to the literature suggesting an increase in CHD risk associated with the LIPC-480 T allele.

Despite the relatively small sample size in the current study (n = 91), there was a statistically significant relation between the LIPC-480 T allele and coronary calcification. This relation is attributable in part to the strength of the association (a greater than twofold increase in risk of coronary calcification for the presence of one LIPC-480 T allele) and the relatively common frequency of the LIPC-480 T allele (0.24). In addition, this study was performed in a population at high risk for vascular disease; in fact, >50% of these type 1 diabetic subjects had detectable coronary calcification. However, it is possible that these results are attributable to chance alone, so it will be important to verify these results with additional investigations. Support for these findings are emerging from studies of the LIPC promoter polymorphism in other populations (21,42–44).

It is possible that the LIPC-480 variant itself does not contribute to the increase in CAC, but that this marker is in linkage disequilibrium with another genetic variant that has physiological significance. This marker is known to be in complete linkage disequilibrium with three other LIPC promoter variants in Caucasian populations (29). Although the LIPC-480 T allele is associated with a decrease in hepatic lipase in vivo, it is unclear from in vitro studies which of these LIPC promoter variants alters gene expression (48–50). Moreover, these LIPC promoter variants may be in linkage disequilibrium with an as yet undiscovered polymorphism that is responsible for the decrease in hepatic lipase levels.

Patients with type 1 diabetes are at a significant increased risk for premature coronary disease. There is also a significant increase in subclinical CHD, as measured by coronary calcification (2). This study identified a common genetic polymorphism in the hepatic lipase gene (LIPC-480C>T) that is associated with a greater than twofold increase in subclinical CHD in patients with type 1 diabetes. In addition, the LIPC-480 T allele is associated with the extent of coronary calcification in a dosage-dependent manner. Future studies will be required to determine if this polymorphism predicts progression of coronary calcification and clinical CHD events in this high-risk population.

These studies were performed at the Adult Clinical Research Center at the University of Colorado Health Sciences Center, and at the Barbara Davis Center for Childhood Diabetes. Support for these studies was provided by the National Heart, Lung, and Blood Institute (NIH/NHL BI-RO1-HL61753), and a research grant from Merck.