Apolipoprotein A-I proteolysis in aortic valve stenosis: role of cathepsin S.

Aortic valve stenosis (AVS) is the most common valvular heart disease in the Western world. Therapy based on apolipoprotein A-I (apoA-I), the major protein component of high-density lipoproteins, results in AVS regression in experimental models. Nevertheless, apoA-I degradation by proteases might lead to suboptimal efficacy of such therapy. An activatable probe using a quenched fluorescently labeled full-length apoA-I protein was generated to assess apoA-I-degrading protease activity in plasma derived from 44 men and 20 women with severe AVS (age 65.0 ± 10.4 years) as well as from a rabbit model of AVS. In human and rabbit AVS plasma, apoA-I-degrading protease activity was significantly higher than in controls (humans: 0.038 ± 0.009 vs 0.022 ± 0.005 RFU/s, p < 0.0001; rabbits: 0.033 ± 0.016 vs 0.017 ± 0.005 RFU/s, p = 0.041). Through the use of protease inhibitors, we identified metalloproteinases (MMP) as exerting the most potent proteolytic effect on apoA-I in AVS rabbits (67%, p < 0.05 vs control), while the cysteine protease cathepsin S accounted for 54.2% of apoA-I degradation in human plasma (p < 0.05 vs control) with the maximum effect seen in women (68.8%, p < 0.05 vs men). Accordingly, cathepsin S activity correlated significantly with mean transaortic pressure gradient in women (r = 0.5, p = 0.04) but not in men (r = - 0.09, p = 0.60), and was a significant independent predictor of disease severity in women (standardized beta coefficient 0.832, p < 0.001) when tested in a linear regression analysis. ApoA-I proteolysis is increased in AVS. Targeting circulating cathepsin S may lead to new therapies for human aortic valve disease.

Beneficial Effects of High-Density Lipoproteins on Acquired von Willebrand Syndrome in Aortic Valve Stenosis.

BACKGROUND: Infusions of apolipoprotein A-I (apoA-I), the major protein component of high-density lipoproteins (HDL), result in aortic valve stenosis (AVS) regression in experimental models. Severe AVS can be complicated by acquired von Willebrand syndrome, a haemorrhagic disorder associated with loss of high-molecular-weight von Willebrand factor (vWF) multimers (HMWM), the latter being a consequence of increased shear stress and enhanced vWF-cleaving protease (ADAMTS-13) activity. Although antithrombotic actions of HDL have been described, its effects on ADAMTS-13 and vWF in AVS are unknown. METHODS AND RESULTS: We assessed ADAMTS-13 activity in plasma derived from a rabbit model of AVS (n = 29) as well as in plasma collected from 64 patients with severe AVS (age 65.0 ± 10.4 years, 44 males) undergoing aortic valve replacement (AVR). In both human and rabbit AVS plasma, ADAMTS-13 activity was higher than that in controls (p < 0.05). Accordingly, AVS patients had less HMWM than controls (66.3 ± 27.2% vs. 97.2 ± 24.1%, p < 0.0001). Both ADAMTS-13 activity and HMWM correlated significantly with aortic transvalvular gradients, thereby showing opposing correlations (r = 0.3, p = 0.018 and r = -0.4, p = 0.003, respectively). Administration of an apoA-I mimetic peptide reduced ADAMTS-13 activity in AVS rabbits as compared with the placebo group (2.0 ± 0.5 RFU/sec vs. 3.8 ± 0.4 RFU/sec, p < 0.05). Similarly, a negative correlation was found between ADAMTS-13 activity and HDL cholesterol levels in patients with AVS (r = -0.3, p = 0.045). CONCLUSION: Our data indicate that HDL levels are associated with reduced ADAMTS-13 activity and increased HMWM. HDL-based therapies may reduce the haematologic abnormalities of the acquired von Willebrand syndrome in AVS.

Low and high density lipoprotein metabolism in primary cultures of hepatic cells from normal and apolipoprotein E knockout mice.

Apolipoprotein E (apoE) plays a major role in lipoprotein metabolism by mediating the binding of apoE-containing lipoproteins to receptors. The role of hepatic apoE in the catabolism of apoE-free lipoproteins such as low density lipoprotein (LDL) and high density lipoprotein-3 (HDL(3)) is however, unclear. We analyzed the importance of hepatic apoE by comparing human LDL and HDL(3) metabolism in primary cultures of hepatic cells from control C57BL/6J and apoE knockout (KO) mice. Binding analysis showed that the maximal binding capacity (Bmax) of LDL, but not of HDL(3), is increased by twofold in the absence of apoE synthesis/secretion. Compared to control hepatic cells, LDL and HDL(3) holoparticle uptake by apoE KO hepatic cells, as monitored by protein degradation, is reduced by 54 and 77%, respectively. Cleavage of heparan sulfate proteoglycans (HSPG) by treatment with heparinase I reduces LDL association by 21% in control hepatic cells. Thus, HSPG alone or a hepatic apoE-HSPG complex is partially involved in LDL association with mouse hepatic cells. In apoE KO, but not in normal hepatic cells, the same treatment increases LDL uptake/degradation by 2.4-fold suggesting that in normal hepatic cells, hepatic apoE increases LDL degradation by masking apoB-100 binding sites on proteoglycans. Cholesteryl ester (CE) association and CE selective uptake (CE/protein association ratio) from LDL and HDL(3) by mouse hepatic cells were not affected by the absence of apoE expression. We also show that 69 and 72% of LDL-CE hydrolysis in control and apoE KO hepatic cells, respectively, is sensitive to chloroquine revealing the importance of a pathway linked to lysosomes. In contrast, HDL(3)-CE hydrolysis is only mediated by a nonlysosomal pathway in both control and apoE KO hepatic cells. Overall, our results indicate that hepatic apoE increases the holoparticle uptake pathway of LDL and HDL(3) by mouse hepatic cells, that HSPG devoid of apoE favors LDL binding/association but impairs LDL uptake/degradation and that apoE plays no significant role in CE selective uptake from either human LDL or HDL(3) lipoproteins.

Low- and high-density lipoprotein metabolism in HepG2 cells expressing various levels of apolipoprotein E.

To determine the importance of hepatic apolipoprotein (apo) E in lipoprotein metabolism, HepG2 cells were transfected with a constitutive expression vector (pRc/CMV) containing either the complete or the first 474 base pairs of the human apoE cDNA inserted in an antisense orientation, for apoE gene inactivation, or the full-length human apoE cDNA inserted in a sense orientation for overexpression of apoE. Stable transformants were obtained that expressed 15, 24, 226, and 287% the apoE level of control HepG2 cells. The metabolism of low-density lipoprotein (LDL) and high-density lipoprotein-3 (HDL(3)), two lipoprotein classes following both holoparticle and cholesteryl esters (CE)-selective uptake pathways, was compared between all these cells. LDL-protein degradation, an indicator of the holoparticle uptake, was greater in low apoE expressing cells than in control or high expressing cells, while HDL(3)-protein degradation paralleled the apoE levels of the cells (r(2) = 0.989). LDL- and HDL(3)-protein association was higher in low apoE expressing cells compared to control cells. In opposition, LDL- and HDL(3)-CE association was not different from control cells in low apoE expressing cells but rose in high apoE expressing cells. In consequence, the CE-selective uptake (CE/protein association ratio) was positively correlated with the level of apoE expression in all cells for both LDL (r(2) = 0.977) and HDL(3) (r(2) = 0.998). We also show that, although in normal and low apoE expressor cells, 92% of LDL- and 80% HDL(3)-CE hydrolysis is sensitive to chloroquine suggesting a pathway linked to lysosomes for both lipoproteins, cells overexpressing apoE lost 60% of chloroquine-sensitive HDL(3)-CE hydrolysis without affecting that of LDL-CE. Thus, the level of apoE expression in HepG2 cells determines the fate of LDL and HDL(3).

Low density lipoprotein uptake: holoparticle and cholesteryl ester selective uptake.

Low density lipoproteins (LDL) contain apolipoprotein B-100 and are cholesteryl ester-rich, triglyceride-poor macromolecules, arising from the lipolysis of very low density lipoproteins. This review will describe the receptors responsible for uptake of whole LDL particles (holoparticle uptake), and the selective uptake of LDL cholesteryl ester. The LDL-receptor mediates the internalization of whole LDL through an endosomal-lysosomal pathway, leading to complete degradation of LDL. Increasing LDL-receptor expression by pharmacological intervention efficiently reduces blood LDL concentrations. The lipolysis stimulated receptor and LDL-receptor related protein may also lead to complete degradation of LDL in presence of free fatty acids and apolipoprotein E- or lipase-LDL complexes, respectively. Selective uptake of LDL cholesteryl ester has been demonstrated in the liver, especially in rodents and humans. This activity brings five times more LDL cholesteryl ester than the LDL-receptor to human hepatoma cells, suggesting that it is a physiologically significant pathway. The lipoprotein binding site of HepG2 cells mediates this process and recognizes all lipoprotein classes. Scavenger receptor class B type I and CD36, which mediate the selective uptake of high density lipoprotein cholesteryl ester, are potentially involved in LDL cholesteryl ester selective uptake, since they both bind LDL with high affinity. It is not known whether they are identical to the uncloned lipoprotein binding site and if the selective uptake of LDL cholesteryl ester produces a less atherogenic particle. If this is verified, pharmacological up-regulation of LDL cholesteryl ester selective uptake may become another therapeutic approach for reducing blood LDL-cholesterol levels and the risk of atherosclerosis.

Selective uptake of cholesteryl esters from various classes of lipoproteins by HepG2 cells.

Selective uptake of cholesteryl esters (CE) from lipoproteins by cells has been extensively studied with high density lipoproteins (HDL). It is only recently that such a mechanism has been attributed to intermediate and low density lipoproteins (IDL and LDL). Here, we compare the association of proteins and CE from very low density lipoproteins (VLDL), IDL, LDL and HDL3 to HepG2 cells. These lipoproteins were either labelled in proteins with 125I or in CE with 3H-cholesteryl oleate. We show that, at any lipoprotein concentration, protein association to the cells is significantly smaller for IDL, LDL, and HDL3 than CE association, but not for VLDL. At a concentration of 20 microg lipoprotein/mL, these associations reveal CE-selective uptake in the order of 2-, 4-, and 11-fold for IDL, LDL, and HDL3, respectively. These studies reveal that LDL and HDL3 are good selective donors of CE to HepG2 cells, while IDL is a poor donor and VLDL is not a donor. A significant inverse correlation (r2 = 0.973) was found between the total lipid/protein ratios of the four classes of lipoproteins and the extent of CE-selective uptake by HepG2 cells. The fate of 3H-CE of the two best CE donors (LDL and HDL3) was followed in HepG2 cells after 3 h of incubation. Cells were shown to hydrolyze approximately 25% of the 3H-CE of both lipoproteins. However, when the cells were treated with 100 microM of chloroquine, a lysosomotropic agent, 85 and 40% of 3H-CE hydrolysis was lost for LDL and HDL3, respectively. The fate of LDL and HDL3-CE in HepG2 cells deficient in LDL-receptor was found to be the same, indicating that the portion of CE hydrolysis sensitive to chloroquine is not significantly linked to LDL-receptor activity. Thus, in HepG2 cells, the magnitude of CE-selective uptake is inversely correlated with the total lipid/protein ratios of the lipoproteins and CE-selective uptake from the two best CE donors (LDL and HDL3) appears to follow different pathways.

Selective uptake of cholesteryl ester from low density lipoprotein is involved in HepG2 cell cholesterol homeostasis.

Low density lipoprotein (LDL) can follow either a holoparticle uptake pathway, initiated by the LDL receptor (LDLr), and be completely degraded, or it can deliver its cholesteryl esters (CE) selectively to HepG2 cells. Although high density lipoprotein-CE selective uptake has been shown to be linked to cell cholesterol homeostasis in nonhepatic cells, there is no available information on the effect of LDL-CE selective uptake on hepatic cell cholesterol homeostasis. In order to define the role of the LDL-CE selective uptake pathway in hepatic cell cholesterol homeostasis, we used a cellular model that expresses constitutively a LDLr antisense mRNA and that shows LDLr activity at 31% the normal level (HepG2-all cells). The addition of a specific antibody anti-LDLr (IgG-C7) reduces LDL protein degradation (LDLr activity) to 7%. This cellular model therefore reflects, above all, LDL-CE selective uptake activity when incubated with LDL. The inactivation of LDLr reduces LDL-protein association by 78% and LDL-CE association by only 43%. The LDL-CE selective uptake was not reduced by the inactivation of LDLr. The activities of the various enzymes involved in cell cholesterol homeostasis were measured in normal and LDLr-deficient cells during incubation in the absence or presence of LDL as a cholesterol source. Essentially, 3-hydroxy-3-methylglutaryl coenzyme A reductase and acyl coenzyme A:cholesterol acyltransferase (ACAT) activities responded to LDL in LDLr-deficient cells as well as in normal HepG2 cells. Inhibition of lysosomal hydrolysis with chloroquine abolished the effect measured on ACAT activity in the presence of LDL, suggesting that CE of LDL, but not free cholesterol, maintains cell cholesterol homeostasis. Thus, in HepG2 cells, when LDLr function is virtually abolished, LDL-CE selective uptake is coupled to cell cholesterol homeostasis.

Uptake and fate of class B scavenger receptor ligands in HepG2 cells.

Class B scavenger receptors (SR-Bs) interact with native, acetylated and oxidized low-density lipoprotein (LDL, AcLDL and OxLDL), high-density lipoprotein (HDL3) and maleylated BSA (M-BSA). The aim of this study was to analyze the catabolism of CD36- and LIMPII-analogous-1 (CLA-1), the human orthologue for the scavenger receptor class B type I (SR-BI), and CD36 ligands in HepG2 (human hepatoma) cells. Saturation binding experiments revealed moderate-affinity binding sites for all the SR-B ligands tested with dissociation constants ranging from 20 to 30 microg.mL-1. Competition binding studies at 4 degrees C showed that HDL and modified and native LDL share common binding site(s), as OxLDL competed for the binding of 125I-LDL and 125I-HDL3 and vice versa, and that only M-BSA and LDL may have distinct binding sites. Degradation/association ratios for SR-B ligands show that LDL is very efficiently degraded, while M-BSA and HDL3 are poorly degraded. The modified LDL degradation/association ratio is equivalent to 60% of the LDL degradation ratio, but is three times higher than that of HDL3. All lipoproteins were good cholesteryl ester (CE) donors to HepG2 cells, as a 3.6-4.7-fold CE-selective uptake ([3H]CE association/125I-protein association) was measured. M-BSA efficiently competed for the CE-selective uptake of LDL-, OxLDL-, AcLDL- and HDL3-CE. All other lipoproteins tested were also good competitors with some minor variations. Hydrolysis of [3H]CE-lipoproteins in the presence of chloroquine demonstrated that modified and native LDL-CE were mainly hydrolyzed in lysosomes, whereas HDL3-CE was hydrolyzed in both lysosomal and extralysosomal compartments. Inhibition of the selective uptake of CE from HDL and native modified LDL by SR-B ligands clearly suggests that CLA-1 and/or CD36 are involved at least partially in this process in HepG2 cells.

Effect of reduced low-density lipoprotein receptor level on HepG2 cell cholesterol metabolism.

Low-density lipoproteins (LDL) are taken up by both LDL receptor (LDLr)-dependent and -independent pathways. In order to determine the importance of these pathways in the activity of the various enzymes that are important in maintaining the cellular cholesterol level in hepatic cells, we created HepG2 cells expressing lower levels of LDLr. Thus HepG2 cells were transfected with a constitutive expression vector (pRc/CMV) containing a fragment of LDLr cDNA inserted in an antisense manner. Stable transformants were obtained that showed significant reductions of 42, 72 and 85% of LDLr protein levels compared with the control, as demonstrated by immunoblotting and confirmed by the LDL binding assay. The best inactivation was achieved with the construct containing the first 0.7 kb of LDLr cDNA. Incubating the different HepG2 cell subtypes with LDL showed similar association of apolipoprotein B (apo B) or cholesteryl esters from LDL with the cells, indicating that the LDLr deficiency did not significantly affect LDL uptake by the cell. However, apoB degradation was reduced significantly by 71-82% in the most LDLr-deficient HepG2 cells. We also found that 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCoA red) activity is significantly increased by 32-35% in HepG2 cells expressing very low levels of LDLr that also demonstrate a significant decrease of 20% in acyl-CoA:cholesterol acyltransferase (ACAT) activity. However, these effects are moderate compared with those observed when cells were incubated in lipoprotein-depleted medium, where a >900% increase in HMGCoA red activity and a loss of 60% of ACAT activity was observed. Thus, in HepG2 cells, different levels of LDLr affect LDL-apoB degradation, but have very little effect on LDL association, HMGCoA red and ACAT activities, revealing that LDLr is more important in the clearance of LDL-apoB than in HepG2 cell cholesterol homoeostasis, a role that should be attributable to both LDLr-dependent and -independent pathways.