Lily L. Wu

Higher Serum Bilirubin Is Associated With Decreased Risk for Early Familial Coronary Artery Disease

Mildly increased serum bilirubin has recently been suggested as a protective factor, possibly reducing the risk of coronary artery disease (CAD) by acting as an antioxidant. We tested this hypothesis by examining serum bilirubin concentrations and other coronary risk factors in 120 men and 41 women with early familial CAD and 155 control subjects. At screening, both cases and control subjects were 38 to 68 years old. Early familial CAD patients had experienced myocardial infarction, coronary artery bypass grafting, or coronary angioplasty by age 55 years for men and 65 for women and had another sibling similarly affected. The average total serum bilirubin concentration was 8.9 +/- 6.1 mumol/L in cases and 12.4 +/- 8.1 mumol/L in control subjects (P = .0001 for difference). In univariate analysis stratified by sex, serum bilirubin was strongly and inversely related to CAD risk, with relative odds of 0.4 to 0.1 (relative to the lowest quintile, P = .04 to .00001) in both men and women as bilirubin increased into the upper two quintiles. Multiple logistic regression analysis was performed including age, sex, smoking, body mass index, diabetes, hypertension, plasma measured LDL cholesterol, HDL cholesterol, triglycerides, and serum bilirubin as potential risk factors. Bilirubin entered as an independent protective factor with an odds ratio of 0.25 (P = .0015) for an increase of 17 mumol/L (1 mg/dL). The standardized logistic regression coefficient for bilirubin was -.33 compared with -.34 for HDL, suggesting that the protective effect of bilirubin on CAD risk in the population is comparable to that of HDL cholesterol. A history of cigarette smoking was associated with significantly lower serum bilirubin concentration and appeared to attenuate the protective effect of bilirubin.

Genotyping and sequence analysis of apolipoprotein E isoforms

Apolipoprotein E (apoE), a polymorphic plasma protein, is essential for catabolism of lipoproteins by receptor-mediated endocytosis. One of the apoE isoforms (E2) differs in its binding affinity to specific receptors and contributes to variations in lipoprotein metabolism. Diagnosis of apoE isoforms is done by isoelectric focusing, but it is hindered by various degrees of post-translational sialylation of the apoE protein. Electrophoretically silent structural variations may also escape detection by this technique. We describe a method for genotyping apoE based on hybridization of allele-specific oligonucleotides with enzymatically amplified genomic DNA, which permits unambiguous diagnosis of six common apoE phenotypes within 24 h. Among 100 E2 alleles present in 81 unrelated individuals genotyped by this technique, we found two rare structural mutants of apoE in addition to the common E2 form, E2(158Arg----Cys). Automated sequencing of amplified DNA identified the rare mutants as E2(136Arg----Ser) and E2(145Arg----Cys). The genotypic method may complement or even replace isoelectric focusing for routine determination of apoE phenotypes and for identification of rare structural variants.

A gene for high urinary kallikrein may protect against hypertension in Utah kindreds.

The inheritance of 12-hour overnight total urinary kallikrein excretion and its association with family history of essential hypertension were studied in 405 normotensive adults and 391 youths in 57 Utah pedigrees. Total urinary kallikrein excretion was highly familial with 51% of the total variance attributable to a dominant allele for high total urinary kallikrein excretion and 27% attributable to the combined effects of polygenes and shared family environment. An estimated 28% of the population has one or two copies of the dominant allele for high total urinary kallikrein excretion (2.3 SD units higher than the low homozygotes). About 83% of the population could be assigned to one of the two genotypic populations. Individuals with the high total urinary kallikrein excretion genotype were significantly less likely to have one or two hypertensive parents (relative odds = 0.56, p = 0.042). We conclude that a dominant allele expressed as high total urinary kallikrein excretion may be associated with decreased risk of essential hypertension. Further studies should be performed to confirm this finding and to test for interactions between this apparently protective gene and other genetic and environmental determinants of essential hypertension.

Hyperhomocysteinemia is a risk factor for cancer and a new potential tumor marker

Plasma homocysteine (Hcy), a well-known independent risk factor for coronary heart disease, is also a risk factor for cancer. Results from our studies indicate that Hcy could be used as a tumor marker. We found elevated circulating total homocysteine (tHcy) in cancer patients even though they were not treated with anti-folate drugs. In serial specimens from cancer patients undergoing treatment, the change of tHcy coincided with the concentration of tumor markers. The rapid proliferation of tumor cells contributed to the much higher concentrations of circulating tHcy. Both concentrations of tHcy and tumor marker would increase in parallel during the growth of tumor cell, but only the Hcy concentration would decline in response to tumor cell death. Several biochemical changes, including folate deficiency, oxidative stress, aberrant DNA methylation, and production of homocysteine thiolactone have been identified in association with hyperhomocysteinemia, which explained why elevated homocysteine eventually led to carcinogenesis. Conceivably, tHcy may be used as a more accurate tumor marker for monitoring cancer patients during treatment, and hyperhomocysteinemia as a risk factor for carcinogenesis.

Phenotypic expression of heterozygous lipoprotein lipase deficiency in the extended pedigree of a proband homozygous for a missense mutation.

Familial lipoprotein lipase (LPL) deficiency is a rare genetic disorder accompanied by well-characterized manifestations. The phenotypic expression of heterozygous LPL deficiency has not been so clearly defined. We studied the pedigree of a proband known to be homozygous for a mutation resulting in nonfunctional LPL. Hybridization of DNA from 126 members with allele-specific probes detected 29 carriers of the mutant allele. Adipose tissue LPL activity, measured previously, was reduced by 50% in carriers, but did not reliably distinguish them from noncarriers. Carriers were prone to the expression of a form of familial hypertriglyceridemia characterized by increased plasma triglyceride, VLDL cholesterol and apolipoprotein B, and decreased LDL and HDL cholesterol concentrations. These manifestations were age modulated, with conspicuous differences between carriers and noncarriers observed only after age 40. Several noncarriers exhibited similar lipid abnormalities, but without the inverse relationship between VLDL cholesterol and LDL cholesterol noted among carriers. In addition to age and carrier status, the potentially reversible conditions, obesity, hyperinsulinemia and lipid-raising drug use were contributory. Thus heterozygous lipoprotein lipase deficiency, together with age-related influences, may account for a form of familial hypertriglyceridemia.

Apolipoprotein, low density lipoprotein subfraction, and insulin associations with familial combined hyperlipidemia. Study of Utah patients with familial dyslipidemic hypertension.

Familial dyslipidemic hypertension (FDH) Is a syndrome recently described from sibshlps selected for early familial hypertension and found to have one or more of three fasting llpld abnormalities [high trlglycerldes, low high density lipoprotein (HDL) cholesterol, high low density lipoprotein (LDL) cholesterol]. In further analyses of these same 131 hypertensive subjects, apolipoprotein A-l and B, fasting plasma Insulin (adjusted for body mass Index), and detailed anthropometries were different In two subgroups of FDH. Of 63 FDH patients, 19 met the criteria for familial combined hyperlipidemia (FCHL); 44 did not, but still had high triglycerlde and/or low HDL cholesterol levels. When compared to 20 normollptdemlc hypertensive patients, the 19 hypertensive patients with FCHL had 196% higher very low density lipoprotein cholesterol (p=0.0001), 33% higher apolipoprotein B (p=0.0002), smaller LDL particles (p=0.007), and 73% higher fasting Insulin (p=0.003), but no significant differences in body mass Index or sklnfold thicknesses. The other 44 FDH patients without FCHL had 33% lower HDL (p=0.0001), with only 8% lower apolipoprotein A-l levels (p=0.20); significantly higher subscapular sklnfolds (p=0.02), weights (p=0.002), body mass Index (p=0.006), knee widths (p=0.0007), and wrist circumferences (p=0.0009); smaller, denser LDL subtractions (p=0.001); and Increased apolipoprotein B levels (p=0.01) compared to the normolipldemlc hypertensive group. Increased fasting Insulin levels were similar to the normollpldemlc group and significantly lower than the FCHL group after adjustment for body mass index, suggesting a relationship between obesity and fasting Insulin levels only in the non-FCHL group. We conclude that FDH consists of at least two subgroups: 1) FCHL with high apolipoprotein B, small LDL particles, and Increased fasting plasma Insulin levels, and 2) a less well-defined residual having upper central obesity with low HDL cholesterol and high triglycerlde levels. Elevated insulin levels found In both groups, but possibly originating through different physiological mechanisms, may provide the pathophyslologlcal connections between dysllpldemia, obesity, and hypertension.

Evaluation of Coronary Risk Factors in Patients With Heterozygous Familial Hypercholesterolemia

Age at onset of clinically manifested coronary artery disease (CAD) varies widely among patients with familial hypercholesterolemia (FH). A number of factors in addition to high low-density lipoprotein cholesterol (LDL) have been suggested as predictors of risk among patients with FH, but a comprehensive examination of their utility is lacking. We therefore measured plasma lipids, carotid intima-medial thickness, and a variety of coronary risk factors in 262 patients with FH > or = 30 years old (68 of whom had premature CAD). Age (p < 0.0001) and gender were the most important determinants of premature CAD risk, with men having 5.64 times the risk of women (p < 0.0001). In addition, cigarette smoking (odds ratio [OR] 2.71, p = 0.026), smaller LDL as determined by the LDL cholesterol/LDL apolipoprotein B ratio (OR 2.60, p = 0.014), and white blood cell count (p = 0.014) were also statistically significant risk factors. Lipoprotein(a) and the presence of xanthoma were associated with risk only in very early coronary cases. After correction for age, carotid intima-media thickness was not associated with CAD risk. Insulin, fibrinogen, homocysteine, plasma C-reactive protein, and the angiotensin-converting enzyme insertion/deletion polymorphism were unrelated to risk in this cohort. These results provide little justification for extensive investigation of risk factors among patients with FH, at least for the risk factors examined here. Rather, the inherent high LDL cholesterol of these patients should be the focus of preventive efforts. The novel finding of increased risk with smaller LDL may prove useful but needs further confirmation.

Lipoprotein(a) Interactions With Lipid and Nonlipid Risk Factors in Early Familial Coronary Artery Disease

An interaction between high plasma lipoprotein(a) [Lp(a)], unfavorable plasma lipids, and other risk factors may lead to very high risk for premature CAD. Plasma Lp(a), lipids, and other coronary risk factors were examined in 170 cases with early familial CAD and 165 control subjects to test this hypothesis. In univariate analysis, relative odds for CAD were 2.95 (P < .001) for plasma Lp(a) above 40 mg/dL. Nearly all the risk associated with elevated Lp(a) was found to be restricted to persons with historically elevated plasma total cholesterol (6.72 mmol/L [260 mg/dL] or higher) or with a total/HDL cholesterol ratio > 5.8. Nonlipid risk factors were also found to at least multiply the risk associated with Lp(a). When Lp(a) was over 40 mg/dL and plasma total/HDL cholesterol > 5.8, relative odds for CAD were 25 (P = .0001) in multiple logistic regression. If two or more nonlipid risk factors were also present (including hypertension, diabetes, cigarette smoking, high total homocysteine, or low serum bilirubin), relative odds were 122 (P < 1 x 10(-12)). The ability of nonlipid risk factors to increase risk associated with Lp(a) was dependent on at least a mildly elevated total/HDL cholesterol ratio. In conclusion, high Lp(a) was found to greatly increase risk only if the total/HDL cholesterol ratio was at least mildly elevated, an effect exaggerated by other risk factors. Aggressive lipid lowering in those with elevated Lp(a) therefore appears indicated.

Higher Plasma Homocyst(e)ine and Increased Susceptibility to Adverse Effects of Low Folate in Early Familial Coronary Artery Disease

To examine the graded risks for coronary artery disease (CAD) associated with plasma homocyst(e)ine [H(e)] and to evaluate the extent to which this risk is mediated by altered vitamin status, we measured plasma concentrations of H(e), vitamins B6 and B12, and folate as well as other coronary risk factors in subjects with early familial CAD and in control subjects. We studied 120 male and 42 female patients with early CAD who were unrelated to each other but were from families in which at least one other sibling had early CAD. Control subjects were 85 men and 70 women with the same age range (38 to 68) as the subjects with CAD at screening. Increasing H(e) was associated with graded increased risks of CAD that appeared consistent with a multiplicative model. Relative odds for CAD were approximately 12.8 in women when those with H(e) levels of 9 mumol/L and above were compared with those with H(e) levels of 9 mumol/L or less (P = .007). For men, the same comparison yielded relative odds of 13.8 (P = .0002). Plasma H(e) remained a strong, independent risk factor after adjustment for standard risk factors and plasma vitamin levels in multiple logistic regression (relative odds, 8.1 for a 10-mumol/L increase in H(e); 95% confidence interval, 3.2 to 20.4; P < .0001). In multivariate ANCOVA the slope of H(e) versus folate was much steeper in subjects with CAD than in control subjects (P = .0035). These data suggest that high plasma H(e) is an important, independent contributor to risk for early familial CAD.(ABSTRACT TRUNCATED AT 250 WORDS)

Genetic linkage between lipoprotein(a) phenotype and a DNA polymorphism in the plasminogen gene

Coronary heart disease risk correlates directly with plasma concentrations of lipoprotein(a) (Lp(a)), a low-density lipoprotein-like particle distinguished by the presence of the glycoprotein apolipoprotein(a) (apo(a)), which is bound to apolipoprotein B-100 (apoB-100) by disulfide bridges. Size isoforms of apo(a) are inherited as Mendelian codominant traits and are associated with variations in the plasma concentration of lipoprotein(a). Plasminogen and apo(a) show striking protein sequence homology, and their genes both map to chromosome 6q26-27. In a large family with early coronary heart disease and high plasma concentrations of Lp(a), we found tight linkage between apo(a) size isoforms and a DNA polymorphism in the plasminogen gene; plasma concentrations of Lp(a) also appeared to be related to genetic variation at the apo(a) locus. We found free recombination between the same phenotype and alleles of the apoB DNA polymorphism. This suggests that apo(a) size isoforms and plasma lipoprotein(a) concentrations are each determined by genetic variation at the apo(a) locus.