Isolation, quantitation, and characterization of a stable complex formed by Lp[a] binding to triglyceride-rich lipoproteins.

Lipoprotein [a] (Lp[a]) is a cholesterol-rich lipoprotein resembling LDL to which a large polymorphic glycoprotein, apolipoprotein [a] (apo[a]), is covalently coupled. Lp[a] usually exists as a free-standing particle in normolipidemic subjects; however, it can associate noncovalently with triglyceride-rich lipoproteins in hypertriglyceridemic (HTG) subjects. In this study, 10-78% of the Lp[a] present in five HTG subjects was found in the triglyceride-rich lipoprotein (TRL) fraction. The Lp[a]-TRL complex was resistant to dissociation by ultracentrifugation (UCF) alone, but was quantitatively dissociated by UCF in the presence of 100 mM proline. Of this dissociated Lp[a], 70-88% was in the form of a lipoprotein resembling conventional Lp[a]. Incubation of Lp[a]-depleted TRL with native Lp[a] resulted in a reconstituted Lp[a]-TRL complex that closely resembled the native isolates in all examined properties. Complex formation was inhibited by several compounds in the order proline > tranexamate > epsilon-aminocaproate >> arginine > lysine. Neither plasminogen nor LDL inhibited binding of Lp[a] to TRL. We observed the preferential binding of Lp[a] containing higher apparent molecular weight apo[a] polymorphs to TRL both in native and reconstituted Lp[a]-TRL complexes. A disproportionate amount of Lp[a] was bound to the larger TRL particles. Although most apo[a] bound to TRL was in the form of conventional Lp[a] particles, lipid-free recombinant apo[a] was observed to bind TRL. These results provide unequivocal evidence of the existence of an Lp[a]-TRL complex under pathophysiologic conditions. The metabolic fate of the Lp[a]-TRL complex, which is more abundant in hypertriglyceridemia, may be different from that of conventional Lp[a], and may contribute uniquely to the progression or severity of cardiovascular disease.

Evaluation of Lp[a] and other independent risk factors for CHD in Asian Indians and their USA counterparts.

Conventional risk factors for coronary heart disease (CHD) do not completely account for the observed increase in premature CHD in people from the Indian subcontinent or for Asian Indians who have immigrated to the USA. The objective of this study was to determine the effect of immigration to the USA on plasma levels of lipoprotein [a] (Lp[a]) and other independent risk factors for CHD in Asian Indians. Three subject groups were studied: group 1, 57 subjects living in India and diagnosed with CHD (CHD patients); group 2, 46 subjects living in India and showing no symptoms of CHD (control subjects); group 3, 206 Asian Indians living in the USA. Fasting blood samples were drawn to determine plasma levels of triglyceride (TG), total cholesterol (TC), low density lipoprotein [LDL cholesterol (LDL-Chol)], high density lipoprotein [HDL cholesterol (HDL-Chol)], apolipoprotein B-100 (apoB-100), and Lp[a]. Apolipoprotein [a] (apo[a]) size polymorphism was determined by immunoblotting. Plasma TG, apoB-100, and Lp[a] concentrations were higher in CHD patients than in control and USA groups. CHD patients had higher levels of TC and LDL-Chol and lower HDL-Chol than control subjects. However, the USA population had higher levels of TC, LDL-Chol, and apoB-100 and lower HDL-Chol than control subjects. Plasma Lp[a] levels were inversely correlated with the relative molecular weight of the more abundant of each subject's two apo[a] isoforms (MAI), and CHD patients showed higher frequencies of lower relative molecular weights among MAI. Our observed changes in lipid profiles suggest that immigrating to the USA may place Asian Indians at increased risk for CHD. This study suggests that elevated plasma Lp[a] confers genetic predisposition to CHD in Asian Indians, and nutritional and environmental factors further increase the risk of CHD. This is the first report implicating MAI size as a predictor for development of premature CHD in Asian Indians. Including plasma Lp[a] concentration and apo[a] phenotype in screening procedures may permit early detection and preventive treatment of CHD in this population.

Cytosolic enzymes from rat tissues that activate the cooked meat mutagen metabolite N-Hydroxyamino-1-methyl-6- phenylimidazo[4,5-b]pyridine (N-OH-PhIP).

Heterocyclic amines are formed during the cooking of foods rich in protein and can be metabolically converted into cytotoxic and mutagenic compounds. These "cooked-food mutagens" constitute a potential health hazard because DNA damage arising from dietary exposure to heterocyclic amines can modify cell genomes and thereby affect future organ function. To determine enzymes responsible for heterocyclic amine processing in mammalian tissues, we performed studies to measure genotoxic activation of the N-hydroxy form of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) --a common dietary mutagen. O-Acetyltransferase, sulfotransferase, kinase, and amino-acyl synthetase activities were assayed using substrate-specific reactions and cytosolic enzymes from newborn and adult rat heart, liver, spleen, kidney, brain, lung, and skeletal muscle. The resultant enzyme-specific DNA adduct formation was quantified via (32)P-postlabeling techniques. In biochemical assays with rat tissue cytosolic proteins, O-acetyltransferases were the enzymes most responsible for N-hydroxy-PhIP (N-OH-PhIP) activation. Compared to O-acetyltransferase activation, there was significantly less kinase activity and even lesser amounts of sulfotransferase activity. Proyl-tRNA synthetase activation of N-OH-PhIP was not detected. Comparing newborn rat tissues, the highest level of O-acetyltransferase mutagen activation was observed for neonatal heart tissue with activities ranked in the order of heart > kidney > lung > liver > skeletal muscle > brain > spleen. Enzymes from cultured neonatal myocytes displayed high O-acetyltransferase activities, similar to that observed for whole newborn heart. This tissue specificity suggests that neonatal cardiac myocytes might be at greater risk for damage from dietary heterocyclic amine mutagens than some other cell types. However, cytosolic enzymes from adult rat tissues exhibited a different O-acetyltransferase activation profile, such that liver > muscle > spleen > kidney > lung > brain > heart. These results demonstrated that enzymes involved in catalyzing PhIP-DNA adduct formation varied substantially in activity between tissues and in some tissues, changed significantly during development and aging. The results further suggest that O-acetyltransferases are the primary activators of N-OH-PhIP in rat tissues.

Bioactivation of the proximal food mutagen 2-hydroxyamino-1-methyl-6-phenylimidazo[4,5-b]pyridine (N-OH-PhIP) to DNA-binding species by human mammary gland enzymes.

We have investigated phase II activation of the food-derived mutagen 2-hydroxyamino-1-methyl-6-phenyl[4,5-b]pyridine (N-OH-PhIP) by cytosolic acetyltransferase, sulfotransferase, and tRNA synthetase/kinase enzymes from human breast tissue. Cytosol from homogenates of mammary gland tissue obtained from breast-reduction surgery or mastectomy was incubated with and without enzyme-specific cofactors, and mutagen binding of calf thymus DNA was quantified by 32P-postlabeling. In addition, microsomal fractions of mammary epithelial cells from some individuals were examined for prostaglandin H synthetase activation of N-OH-PhIP. Our results show that all four enzymes can participate in activating N-OH-PhIP, thus inducing PhIP-DNA adduct formation in human mammary cells. However, not all individuals exhibited all these activities; instead each individual showed a combination of one or more activation pathways. The present findings demonstrate that the human mammary gland has the capacity to metabolically activate a dietary mutagen by several enzyme systems, including acetyltransferase, sulfotransferase, tRNA synthetase/kinase, and prostaglandin hydroperoxidase catalysis.

Introduction, distribution, and removal of 7-methylguanine in different liver chromatin fractions of young and old mice.

The distribution and elimination of 7-methylguanine (m7Gua) from different liver DNA chromatin fractions has been studied after treating young and old mice with N-methyl-N-nitrosourea (MNU). Guanine methylation kinetics was first studied in total liver DNA following intraperitoneal injections of 25 mg/kg and 50 mg/kg MNU does. MNU-induced DNA alkylation, as measured by m7Gua levels, was dose-dependent in liver tissues of young (9-11 month) and old mice (28-29 month). However, liver DNA in old mice incurred approximately 50% more damage than young mice for weight-normalized doses of MNU. The kinetics of adduct removal from total DNA was biphasic. A more rapid phase of m7Gua removal was observed during the first 24 h to 48 h following MNU administration; thereafter, the remaining m7Gua adducts were hydrolyzed much more slowly. Similar amounts of m7Gua were removed by 24 h at both the 25 mg/kg and 50 mg/kg MNU doses for a single age-group, but old liver tissue removed significantly more m7Gua than young liver tissue during this initial phase. Following a single injection of carcinogen (50 mg/kg), liver nuclei were isolated and chromatin was sheared by limited Micrococcal nuclease digestion. Chromatin was separated into nuclease-soluble, low-salt, high-salt and nuclear matrix fractions. All four fractions of young liver chromatin were methylated to the same degree. In contrast, there were differences in m7Gua levels between old liver chromatin fractions. DNA in the nuclease-sensitive fraction was most heavily alkylated, whereas nuclear matrix sequences were modified the least. Removal of m7Gua occurred at relatively uniform rates in all chromatin fractions regardless of age, indicating that m7Gua was not preferentially repaired in different nuclease-susceptible regions of chromatin. These results suggest that the N-methylpurine-DNA glycosylase responsible for eliminating m7Gua from the mammalian genome is not deficient in senescent liver tissue. However, there may be age-related changes in chromatin composition or structure that make some genomic sequences more accessible to alkylating agents in liver tissue of older animals.

Lipoprotein(a) in plasma, arterial wall, and thrombus from patients with aortic aneurysm.

The plasma concentration of lipoprotein(a) [Lp(a)] is highly correlated with the incidence of cardiovascular and peripheral vascular disease. A positive physiological role for Lp(a) has not yet been clearly identified, although elevated plasma levels in pregnant women, long-distance runners, subjects given growth hormone, patients after cardiovascular surgery, and patients with cancer, diabetes, or renal disease suggest its involvement in tissue synthesis and repair. The hypothesis that Lp(a) is involved in repair/reinforcement of the aorta was tested in 38 patients undergoing surgery for aortic aneurysm. In 29 patients 1 day before surgery, the mean plasma Lp(a) protein level was 10.7 mg/dl. At about 1, 2, and 8 weeks after surgery, the level was 14.1, 15.1, and 15.2 mg/dl, respectively. These levels are significantly higher than those of a comparable group of normal subjects (6.4 mg/dl; n = 274). Specimens of resected aortic aneurysm showed extensive medial degeneration, discontinuous elastic fibers, and deposition of mucopolysaccharides; these specimens were treated with a detergent-containing buffer to extract entrapped lipoproteins. The mean Lp(a) protein level in aortic wall extracts was 14.6 ng/mg tissue; these individual values were significantly associated with plasma Lp(a) levels before surgery (r2 = 0.31, p = 0.0003). The mean Lp(a) protein level in aortic thrombus extracts was substantially higher at 69.6 ng/mg tissue; these individual levels also were significantly associated with plasma Lp(a) concentrations before surgery (r2 = 0.68, p < 0.0001). The observations that: (i) plasma Lp(a) protein is about 1.7-fold higher in patients with aortic aneurysms than in normal subjects; and (ii) that Lp(a) protein in the aneurysmic thrombus is about 4.8-fold higher than in the aortic wall suggest that this lipoprotein plays a significant and direct role in thrombus formation and in reinforcement of the aneurysmic aortic wall.

Aging affects the levels of DNA damage in postmitotic cells.

Cells are continuously exposed to DNA damaging agents that may cause mutations or lead to cell death. To counter this constant, ubiquitous attack on the genetic material, cells possess highly diverse and efficient systems to repair a variety of DNA lesions. For cells that are nondividing and are expected to remain functionally viable for many years, it is important that damage not accumulate in those genes that are essential to maintaining differentiated gene expression. If damage were to accumulate slowly in working genes, then the outcomes might appear as biological changes typically associated with senescence. Estimates on the types of DNA damage believed to arise spontaneously suggest that methylation of N7-guanine is one of the more frequently occurring events, exceeded only by single-strand breaks and possibly depurination. Previous studies have shown that the steady-state levels of m7Gua increase during aging of postmitotic mammalian tissues. To test for the possibility that repair of m7Gua might decline in senescent animals, we induced methyl adducts in young and old mice with single doses of MNU, and determined the kinetics of adduct removal. Liver, kidney and brain all exhibited some active repair of m7Gua as characterized by the rapid removal of the adduct from DNA. However, a fraction of damage was refractory to repair and was lost from DNA much more slowly. This repair-resistant fraction of damage was greater in DNA from the old tissues, but the interpretation of the data is not straightforward, because different amounts of damage were induced in young and old tissues with the same weight-normalized dose of MNU. Although old cells had higher levels of persistent adducts, initial repair rates were similar between age-matched tissues. Furthermore, experiments indicated that mRNA levels for 3-methyladenine glycosylase repair enzyme did not change with age. Our working hypothesis is that repair enzymes are present and active in senescent postmitotic tissues, but changes have occurred in old chromatin that have affected the ability of repair enzymes to efficiently process these adducts.

Interaction of LDL and Lp[a] with human skin fibroblasts.

We have studied the interaction of LDL and Lp[a] with fibroblasts. Our studies suggest that Lp[a] does not effectively compete with LDL for binding to the LDL receptor, and does not efficiently suppress the activity of the intracellular enzyme HMG-CoA reductase. However, Lp[a-], formed by reduction of the disulfide bond between apo[a] and apoB, behaves much like homologous LDL, whether or not apo[a] is removed from the mixture, and in spite of the fact that one or more apoB disulfides may also have been cleaved. In our studies we also noted that Lp[a] often enhanced binding of 125I-LDL by fibroblasts. Further investigation has suggested that this interaction is time-dependent. Experiments in receptor-negative fibroblasts indicate that the enhancement is not related to the presence of the LDL receptor; however, it is inhibited by the removal of calcium from the medium. The presence of sialic acid at millimolar concentrations in the medium inhibits much of the Lp[a]-enhanced binding of 125I-LDL to the cells. These studies suggest that Lp[] may in some way enhance LDL binding to cells, perhaps via interaction with cell surface glycosaminoglycans or proteoglycans or with collagen.