Sally P. A. McCormick

Aims: Cohort studies have demonstrated associations between calcific aortic valve disease (CAVD) and Lp(a). As Lp(a) is almost entirely genetically determined, in this study, we aim to determine whether Lp(a), when predicted from genetic data, is associated with CAVD and major adverse cardiovascular events (MACEs). Methods and results: Patients undergoing coronary angiography between January 2012 and May 2013 were invited to participate in the study. Of 752 analysable participants, 446 had their Lp(a) measured and 703 had a calculable LPA genetic risk score (GRS). The primary outcomes were the presence of CAVD at baseline and MACE over a 7-year follow-up. The GRS explained 45% of variation in Lp(a). After adjustment for cardiac risk factors and coronary artery disease (CAD), the odds of CAVD increased with increasing Lp(a) [odds ratio (OR) 1.039 per 10-unit increase, 95% confidence interval (CI) 1.022–1.057, P < 0.001] and GRS (OR 1.054 per 10-unit increase, 95% CI 1.024–1.086; P < 0.001). Lipoprotein(a) and the GRS as continuous variables were not associated with subsequent MACEs. A dichotomized GRS (>54) was associated with MACE, but this relationship became non-significant when CAD classification was added into the model (OR 1.333, 95% CI 0.927–1.912; P = 0.12). Conclusion: An LPA GRS can explain 45% of variation in Lp(a) levels, and both Lp(a) and the GRS are associated with CAVD. An elevated GRS is associated with future cardiac events in a secondary risk setting, but, if the CAD status is known, it does not provide additional prognostic information.

BACKGROUND: High levels of Lp(a) (lipoprotein(a)) are associated with multiple forms of cardiovascular disease. Lp(a) consists of an apoB ₁₀₀ -containing particle attached to the plasminogen homologue apo(a). The pathways for Lp(a) clearance are not well understood. We previously discovered that the plasminogen receptor PlgRKT (plasminogen receptor with a C-terminal lysine) promoted Lp(a) uptake in liver cells. Here, we aimed to further define the role of PlgRKT and to investigate the role of 2 other plasminogen receptors, annexin A2 and S100A10 (S100 calcium-binding protein A10) in the endocytosis of Lp(a). METHODS: Human hepatocellular carcinoma (HepG2) cells and haploid human fibroblast-like (HAP1) cells were used for overexpression and knockout of plasminogen receptors. The uptake of Lp(a), LDL (low-density lipoprotein), apo(a), and endocytic cargos was visualized and quantified by confocal microscopy and Western blotting. RESULTS: The uptake of both Lp(a) and apo(a), but not LDL, was significantly increased in HepG2 and HAP1 cells overexpressing PlgRKT, annexin A2, or S100A10. Conversely, Lp(a) and apo(a), but not LDL, uptake was significantly reduced in HAP1 cells in which PlgRKT and S100A10 were knocked out. Surface binding studies in HepG2 cells showed that overexpression of PlgRKT, but not annexin A2 or S100A10, increased Lp(a) and apo(a) plasma membrane binding. Annexin A2 and S100A10, on the other hand, appeared to regulate macropinocytosis with both proteins significantly increasing the uptake of the macropinocytosis marker dextran when overexpressed in HepG2 and HAP1 cells and knockout of S100A10 significantly reducing dextran uptake. Bringing these observations together, we tested the effect of a PI3K (phosphoinositide-3-kinase) inhibitor, known to inhibit macropinocytosis, on Lp(a) uptake. Results showed a concentration-dependent reduction confirming that Lp(a) uptake was indeed mediated by macropinocytosis. CONCLUSIONS: These findings uncover a novel pathway for Lp(a) endocytosis involving multiple plasminogen receptors that enhance surface binding and stimulate macropinocytosis of Lp(a). Although the findings were produced in cell culture models that have limitations, they could have clinical relevance since drugs that inhibit macropinocytosis are in clinical use, that is, the PI3K inhibitors for cancer therapy and some antidepressant compounds.

Background: Cohort studies have demonstrated associations between calcific aortic valve disease (CAVD) and Lp(a). As Lp(a) is almost entirely genetically determined, with the increasing clinical availability of genetic information, we aimed to determine if Lp(a), when predicted from genetic data, is associated with CAVD and major adverse cardiovascular events (MACE). Methods: Patients undergoing coronary angiography were invited to participate in the study. Of 752 analysable participants, 446 had Lp(a) measured, and 703 had a calculable LPA genetic risk score (GRS). CAVD was categorized as absent/present and by severity. The primary outcomes were presence of CAVD at baseline, and MACE over seven years follow-up. Results: The GRS explained 45% of the variation in Lp(a). After adjustment for cardiac risk factors and coronary artery disease, the odds of CAVD increased with Lp(a) (OR 1.039 per 10 unit increase, 95% C.I. 1.022 – 1.057, pLPA GRS can explain 45% of the variation in Lp(a) levels, and both Lp(a) and the GRS are associated with CAVD. An elevated GRS is associated with future cardiac events in a secondary risk setting, but, if coronary artery disease status is known, it does not provide additional prognostic information.

The expression of mammalian genes in transgenic mice is routinely undertaken to study the functional consequences of overexpressing specific proteins and to understand the DNA sequences controlling gene expression. However, until relatively recently, expression of intact copies of large genes (>40 kb) and/or clusters of genes in transgenic animals has been difficult because of the limitations in the cloning capacity of commonly used vectors, such as plasmids, cosmids, and bacteriophage λ. For many years, investigators using transgenic animals have learned to live with the inherent limitations of these cloning vectors by constructing more compact “minigene” constructs from cDNA and genomic clones. However, minigene constructs invariably force the investigator to make assumptions about the regions of a gene that are functionally important for gene expression. The recent development of the P1 bacteriophage and yeast artificial chromosome (YAC) cloning systems has allowed investigators to generate transgenic animals with intact copies of larger genes. P1 bacteriophages (Sternberg, 1992) contain inserts of 80 to 90 kb, whereas the YACs can accommodate inserts ranging from 100 kb to more than 1Mb. We have used P1 clones to generate transgenic mice for the purpose of studying important issues in mammalian lipoprotein metabolism.

Endocytosis is the process by which molecules are actively transported into cells. It can take on a variety of forms depending on the cellular machinery involved ranging from specific receptor-mediated endocytosis to the less selective and actin-driven macropinocytosis. The plasma lipoproteins, which deliver lipids and other cargo to cells, have been intensely studied with respect to their endocytic uptake. One of the first molecules to be visualised undergoing endocytosis via a receptor-mediated, clathrin-dependent pathway was low-density lipoprotein (LDL). The LDL molecule has subsequently been shown to be internalised through multiple endocytic pathways. Dissecting the pathways of lipoprotein endocytosis has been crucial to understanding the regulation of plasma lipid levels and how lipids enter cells in the arterial wall to promote atherosclerosis. It has also aided understanding of the dysregulation that occurs in plasma lipid levels when molecules involved in uptake are defective, as is the case in familial hypercholesterolemia (FH). The aim of this review is to outline the many endocytic pathways utilised for lipoprotein uptake. It explores the various experimental approaches that have been applied to visualise lipoprotein endocytosis with an emphasis on LDL and its more complex counterpart, lipoprotein(a) [Lp(a)]. Finally, we look at new developments in lipoprotein visualisation that hold promise for scrutinising endocytic pathways to finer detail in the future.

We recently found that plasminogen receptors regulate the plasma membrane binding and uptake of Lp(a) via macropinocytosis. In this study, we sought to further define lipoprotein(a) [Lp(a)] macropinocytosis, discovering an unexpected role for antidepressants and serotonin in the regulation of this process. We found that the tricyclic antidepressant imipramine enhanced Lp(a) uptake, in contradiction of its published role as a macropinocytosis inhibitor. Extending these experiments to the commonly used serotonin uptake inhibitors (SSRIs) citalopram and sertraline, we found that citalopram stimulated, while sertraline inhibited, Lp(a) uptake. Imipramine and citalopram enhanced cell surface binding of Lp(a) rather than upregulating macropinocytosis. Consistent with imipramine and citalopram boosting extracellular serotonin levels, serotonin itself also enhanced Lp(a) surface binding and uptake. Imipramine and serotonin increased expression of the plasminogen receptor with a C-terminal lysine (PlgRKT), a receptor known to enhance cell surface binding of Lp(a), likely accounting for their effects on Lp(a) uptake. Finally, imipramine and citalopram increased Lp(a) delivery into Rab11 recycling endosomes, but not degradative pathways in the cell. These findings indicate citalopram and imipramine may have utility as a potential Lp(a)-lowering therapeutic in people suffering from depression who often have elevated Lp(a) levels and an increased risk of cardiovascular disease.

Ribose-cysteine is a synthetic compound designed to increase glutathione (GSH) synthesis. Low levels of GSH and the GSH-dependent enzyme, glutathione peroxidase (GPx), is associated with cardiovascular disease (CVD) in both mice and humans. Here we investigate the effect of ribose-cysteine on GSH, GPx, oxidised lipids and atherosclerosis development in apolipoprotein E-deficient (apoE-/-) mice. Female 12-week old apoE-/- mice (n = 15) were treated with 4–5 mg/day ribose-cysteine in drinking water for 8 weeks or left untreated. Blood and livers were assessed for GSH, GPx activity and 8-isoprostanes. Plasma alanine transferase (ALT) and lipid levels were measured. Aortae were quantified for atherosclerotic lesion area in the aortic sinus and brachiocephalic arch and 8-isoprostanes measured. Ribose-cysteine treatment significantly reduced ALT levels (p<0.0005) in the apoE-/- mice. Treatment promoted a significant increase in GSH concentrations in the liver (ppppp50% in both the aortic sinus and brachiocephalic branch (p<0.05). Ribose-cysteine promotes a significant GSH-based antioxidant effect in multiple tissues as well as an LDL-lowering response. These effects are accompanied by a marked reduction in atherosclerosis suggesting that ribose-cysteine might increase protection against CVD.