Potential conflict of interest: Nothing to report.
A major atheroprotective functionality of high-density lipoproteins (HDLs) is to promote “reverse cholesterol transport” (RCT). In this process, HDLs mediate the efflux and transport of cholesterol from peripheral cells and its subsequent transport to the liver for further metabolism and biliary excretion. We have previously demonstrated in cultured hepatocytes that P2Y13 (purinergic receptor P2Y, G protein–coupled, 13) activation is essential for HDL uptake but the potential of P2Y13 as a target to promote RCT has not been documented. Here, we show that P2Y13-deficient mice exhibited a decrease in hepatic HDL cholesterol uptake, hepatic cholesterol content, and biliary cholesterol output, although their plasma HDL and other lipid levels were normal. These changes translated into a substantial decrease in the rate of macrophage-to-feces RCT. Therefore, hallmark features of RCT are impaired in P2Y13-deficient mice. Furthermore, cangrelor, a partial agonist of P2Y13, stimulated hepatic HDL uptake and biliary lipid secretions in normal mice and in mice with a targeted deletion of scavenger receptor class B type I (SR-BI) in liver (hypomSR-BI–knockoutliver) but had no effect in P2Y13 knockout mice, which indicate that P2Y13-mediated HDL uptake pathway is independent of SR-BI–mediated HDL selective cholesteryl ester uptake. Conclusion: These results establish P2Y13 as an attractive novel target for modulating RCT and support the emerging view that steady-state plasma HDL levels do not necessarily reflect the capacity of HDL to promote RCT. (HEPATOLOGY 2010)
The risk of developing atherosclerosis, a leading cause of death in industrialized countries, is directly related to the plasma concentration of low-density lipoprotein (LDL) cholesterol and inversely related to high-density lipoprotein (HDL) cholesterol levels. Although HDL particles directly protect the vascular wall, their beneficial effect is mostly attributed to their central functions in “reverse cholesterol transport” (RCT). This is a process in which excess cholesterol from peripheral tissue and macrophages/foam cells is taken up and processed in HDL particles and later delivered to the liver, before being excreted in the bile as free cholesterol or after transformation into bile acids.1 This process, the principal way by which the body eliminates cholesterol, relies on specific interactions between HDL particles and peripheral cells (cholesterol efflux) on the one hand and hepatocytes (HDL cholesterol uptake) on the other hand. Hepatic HDL uptake can involve either the selective uptake of cholesteryl ester (CE) from HDL particles into hepatocytes (i.e., uptake of CE without HDL proteins) or holoparticle HDL endocytosis (i.e., uptake of both HDL protein and lipid moieties).
The scavenger receptor class B, type I (SR-BI) is a HDL receptor that mediates selective uptake in vivo. In SR-BI–deficient mice, HDL selective CE uptake by the liver decreases by more than 90% but HDL holoparticle uptake is unchanged, suggesting that holoparticle uptake does not involve SR-BI.2, 3 A likely candidate mechanism mediating HDL holoparticle uptake in the absence of SR-BI is the cell-surface complex related to mitochondrial F1-adenosine triphosphatase (ATPase), namely ecto-F1-ATPase.4 When HDL binds to apolipoprotein A-I (apoA-I), this enzyme generates extracellular adenosine diphosphate (ADP) which then specifically activates the P2Y13 receptor which is a G-protein-coupled receptor. This results in HDL holoparticle endocytosis mediated by as-yet unidentified low-affinity binding sites. In addition, cangrelor, which acts as a partial agonist of the P2Y13 receptor, stimulates HDL endocytosis by human hepatocytes and perfused mouse liver.5
In this study, we generated P2Y13-null mice (−/−) in order to investigate the potential contribution of P2Y13 to RCT.
abca1, ATP-binding cassette, subfamily A, member 1; abcb4, ATP-binding cassette b4; abcb11/bsep, ATP-binding cassette b11/bile salt export pump; abcg1, ATP-binding cassette g1; abcg5/abcg8, ATP-binding cassette g5/g8; apoA-I, apolipoprotein A-I; ATP, adenosine triphosphate; HDL, high-density lipoprotein; hmgcr, 3-hydroxy-3-methylglutaryl-coenzyme A reductase; mRNA, messenger RNA; ntcp, Na+-taurocholate cotransporting polypeptide; oatp, organic anion transport polypeptide; P2Y13, purinergic receptor P2Y, G-protein-coupled 13; RCT, reverse cholesterol transport; SR-BI, scavenger receptor class B, type I.
Materials and Method
For a detailed description of materials and methods used, please see Supporting Information.
To investigate the role of P2Y13 in cholesterol metabolism in vivo, we generated P2Y13-null mice (−/−) by a gene-targeting strategy (Fig. 1) and their wild-type littermates (+/+) were used as controls. Physical examination of the (−/−) mice on a chow diet indicated that P2Y13 deletion had no detrimental effect on development, fertility, or any hematological or biochemical parameters. Of note, we found no difference in plasma total cholesterol, HDL cholesterol, and triglyceride levels between the (+/+) and (−/−) mice maintained on chow diet (Table 1). Hepatic lipids were also similar, except that free cholesterol content was significantly lower in (−/−) mice than in (+/+) mice (Table 1), suggesting that the hepatic cholesterol metabolism is impaired in mice lacking the P2Y13 gene.
Table 1. Plasma, Hepatic, and Biliary Lipid Values in Wild-Type and P2Y13-KO Mice
Blood samples were collected from mice fasted for 3 hours.
Values are expressed as means ± SEM; n ≥ 10 mice per group for plasma values and n ≥ 6 for liver and biliary lipid values.
Indicates significant difference (P < 0.05) from wild-type (+/+) mice.
To examine HDL uptake mediated by P2Y13, we allowed endocytosis of fluorescent DyLight549-HDL in primary hepatocytes isolated from P2Y13 (−/−) and (+/+) mice. After 30 minutes of endocytosis, fluorescent DyLight549-HDL showed a punctate endosomal pattern in hepatocytes from (+/+) mice (Fig. 2A). In contrast, the staining was strikingly different in hepatocytes from P2Y13 (−/−) mice, showing very few vesicular structures positive for fluorescent DyLight549-HDL (Fig. 2B), which suggests that receptor-mediated HDL uptake is markedly impaired when P2Y13 is not expressed. We next compared the uptake of radiolabeled HDL by livers of P2Y13 (−/−) and (+/+) mice. Hepatic uptake of 125I-HDL was significantly lower in livers from P2Y13 (−/−) mice compared to (+/+) (−20.3% ± 5.6%; Fig. 2C). Remarkably, the P2Y13 partially agonistic compound cangrelor (10 μM) induced marked stimulation of 125I-HDL hepatic uptake in livers from (+/+) mice (+34.9% ± 6.4% versus phosphate-buffered saline control) but had no effect in livers from P2Y13 (−/−) mice (Fig. 2C). This confirms that cangrelor acts specifically via P2Y13 activation and suggests that P2Y13 can be activated in vivo to promote hepatic HDL uptake. Moreover, the stimulation of 125I-HDL hepatic uptake by cangrelor was also observed in livers from mice with a targeted deletion of SR-BI in liver (hypomSR-BI–KOliver,6). The level of stimulation was the same as in P2Y13 (+/+) mice (Fig. 2C), suggesting that the P2Y13-mediated HDL uptake pathway is independent of SR-BI–mediated HDL selective CE uptake.
Because biliary sterols are supposedly derived mainly from HDL cholesterol,7, 8 we next investigated whether the lower free cholesterol liver content in P2Y13-KO mice would be translated into altered biliary lipid secretion rates (Table 1). Bile flow and bile acid secretion were somewhat lower following P2Y13 inactivation. Biliary secretion of cholesterol and phospholipids were significantly lower in P2Y13-KO mice compared to wild type controls (P < 0.05), indicating that hepatobiliary cholesterol transport is impaired in these mice. Conversely, cangrelor (8 nmol/kg) treatment dramatically increased biliary total cholesterol, bile acid and phospholipid secretion together with bile flow in wild-type and hypomSR-BI–KOliver mice but not in P2Y13-KO mice (Fig. 3). In both wild-type and P2Y13-KO mice, plasma lipid levels remained unchanged over 4 hours after cangrelor treatment (data not shown). This suggests that pharmacological activation of P2Y13 might stimulate biliary lipid secretion in vivo with unchanged plasma lipid levels. This occurred independently of SR-BI, previously reported to mediate biliary cholesterol secretion, without concomitant changes in either biliary bile acid or phospholipid secretion.9, 10
To test the possibility that P2Y13 is involved in RCT in vivo, [3H]cholesterol-loaded peritoneal macrophages from C57BL6/J mice were injected intraperitoneally in P2Y13 (−/−) and (+/+) mice. The kinetics of [3H] counts in plasma were not significantly changed in P2Y13 (−/−) mice versus (+/+) control mice (Fig. 4A). Likewise, there was no difference in [3H]cholesterol count in the liver at the end of the 48-hour period (Fig. 4B), which is consistent with most RCT experiments showing that the tracer amount recovered in the liver after 48 hours is essentially unchanged.11 Nonetheless, absence of P2Y13 resulted in a significant reduction in overall macrophage-to-feces RCT, as reflected by lower counts in feces from both the bile acid and neutral sterol fractions (Fig. 4C).
Hepatic expression of genes that are involved in cholesterol transport and metabolism was determined to assess potential changes due to P2Y13 deletion (Table 2). Hepatic messenger RNA (mRNA) expression of ATP-binding cassette transporters abca1, which plays a pivotal role in the efflux of cholesterol to apoA-I, and abcg1, which has a similar role in the efflux to HDL,12, 13 were significantly lower in P2Y13 (−/−) mice compared to (+/+) mice (Table 2; abca1 −56% ± 5% and abcg1 −67% ± 7%), suggesting that hepatic HDL biogenesis might be impaired in these mice. However, P2Y13 deletion had no significant effect on the hepatic mRNA expression of biliary lipid transport proteins abcg5/abcg8, bile salt export pump (abcb11/bsep), sodium taurocholate cotransporting polypeptide (ntcp), and organic anion transport polypeptide (oatp). This suggests that the observed lower biliary lipid output in P2Y13 (−/−) mice was not due to modulation of expression of these lipid transporters. The mRNA levels of the bile acid synthesis enzymes (cytochrome P450, family 7, subfamily A, polypeptide 1 [cyp7a1]; cyp27a1; and cyp8b1) were unchanged by P2Y13 deficiency, indicating that the lower biliary cholesterol and phospholipid secretion observed in P2Y13 (−/−) mice was not due to altered bile acid synthesis. SR-BI expression was also unaffected by the absence of P2Y13 which, together with the fact that cangrelor still stimulates both hepatic HDL uptake and biliary lipid secretion in mice lacking hepatic SR-BI as in wild-type mice, support the idea of independent P2Y13 and SR-BI–mediated HDL uptake pathways. No significant change was observed in the sterol regulatory element binding protein 2 target genes, low-density lipoprotein receptor, or 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase mRNA levels in the liver of P2Y13 (−/−) mice, suggesting that cholesterol derived from HDL and entering the liver via P2Y13 is not accessible to the endoplasmic reticulum cholesterol pool which regulates hepatic cholesterol synthesis.14
Table 2. Effect of P2Y13 Deletion on Hepatic mRNA Expression of Genes Involved in Lipid Homeostasis
Real-time PCR was performed on separate livers (n ≥ 8) of 3-hour-fasted wild-type and P2Y13-KO mice (males, 8-10 weeks old, chow diet). For all genes scored, the fold-change was calculated by dividing the KO value by the wild-type value (e.g., a drop of 60% from wild-type is reported as 0.40). Comparison of gene expression values between wild-type and P2Y13-KO mice was done with the use of the Student t test (two-tailed). The threshold for fold-change was set above 1.5 or below 0.5 and
P < 0.01.
1.02 ± 0.04
ATP synthase, mitochondrial F1 complex, beta polypeptide
cytochrome P450, family 7, subfamily A, polypeptide 1
1.12 ± 0.09
cytochrome P450, family 27, subfamily A, polypeptide 1
1.14 ± 0.08
cytochrome P450, family 8, subfamily B, polypeptide 1
0.97 ± 0.16
ATP-binding cassette, subfamily G, member 5
1.19 ± 0.09
ATP-binding cassette, subfamily G, member 8
0.75 ± 0.05
ATP-binding cassette, subfamily B (MDR/TAP), member 4
1.06 ± 0.06
ATP-binding cassette, subfamily B (MDR/TAP), member 11
0.82 ± 0.09
solute carrier family 10 (sodium/bile acid cotransporter family), member 1
0.89 ± 0.09
solute carrier organic anion transporter family, member 1A2
0.78 ± 0.12
3-hydroxy-3-methylglutaryl-coenzyme A reductase
1.19 ± 0.09
sterol regulatory element binding transcription factor 2
These data indicate that P2Y13 activity is involved in macrophage-to-feces RCT. In situ liver perfusion and gallbladder cannulation experiments suggest that this occurs by promoting hepatic HDL uptake and biliary lipid secretion, independently of SR-BI–mediated HDL selective CE uptake. Surprisingly, although hallmark features of RCT are impaired in P2Y13-deficient mice, these mice have unchanged HDL cholesterol levels compared to wild-type mice. Thus, the P2Y13-dependent RCT pathway might contribute only minimally to the overall plasma HDL levels. In concordance with this observation, ABCA1 expression in macrophages does not seem to be a major contributing factor to plasma HDL levels although ABCA1 seems to contribute considerably to the efflux of excess cholesterol from these cells.15 An alternate idea is that compensatory processes might explain why plasma HDL levels are unaltered in P2Y13-null mice. Most HDL contributed to the plasma pool is hepatic in origin, mainly through the lipidation of nascent apoA-I by ABCA1,12, 16 the expression of which contributes up to approximately 70% to plasma HDL biogenesis in vivo. Recent studies also demonstrate that ABCG1, implicated in cholesterol efflux toward HDL, might also contribute to plasma HDL levels.13 A possibility is that circulating HDL levels in P2Y13-null mice are unaffected, because the decreased HDL uptake is balanced by a decrease in hepatic HDL formation. On the one hand, HDL uptake and thereby catabolism is decreased in P2Y13-null mice and, on the other hand, HDL formation in these mice might be lower as reflected by the decreased hepatic ABCA1 and ABCG1 expression. Although not formally tested, these findings are consistent with the overall concept of the role of hepatic ABCA1/ABCG1 in HDL formation.12, 16 The phenotype observed in P2Y13-null mice suggests that plasma HDL cholesterol may not always be the most reliable marker for assessing the potential utility of new therapeutic agents targeting HDL. It is thus important to develop effective approaches to evaluate HDL functionality. Our results show that pharmacological activation of P2Y13 has the potential to raise biliary lipid secretion by enhancing hepatic HDL uptake. An interesting feature is that P2Y13 activation not only drives biliary cholesterol secretion but also the secretion of PL and bile acids, which suggest that pharmacological activation of P2Y13 would not induce cholestasis. If P2Y13 has similar activity in humans, it may become an attractive target for RCT promotion.
We thank the technical service of the animal (Genotoul Anexplo Platform) and lipidomic (Metatoul platform) facilities of the Bio-Medical Research Federative Institute of Toulouse (IFR150), Michel Nauze and Corinne Rolland for technical assistance, and Dr. Florence Bietrix (University of Amsterdam, the Netherlands) for support and guidance.