|Home | About | Journals | Submit | Contact Us | Français|
To compare the abilities of human wild type apoA-I (WT apoA-I) and human apoA-IMilano (apoA-IM) to promote macrophage reverse cholesterol transport (RCT) in apoA-I–null mice infected with adeno-associated virus (AAV) expressing either WT apoA-I or apoA-IM.
WT apoA-I– or apoA-IM–expressing mice were intraperitoneally injected with [H3]cholesterol-labeled J774 mouse macrophages. After 48 hours, no significant difference was detected in the amount of cholesterol removed from the macrophages and deposited in the feces via the RCT pathway between the WT apoA-I and apoA-IM groups. Analysis of the individual components of the RCT pathway demonstrated that the apoA-IM-expressing mice promoted ATP-binding cassette transporter A1 (ABCA1)-mediated cholesterol efflux as efficiently as WT apoA-I but that apoA-IM had a reduced ability to promote cholesterol esterification via lecithin cholesterol-acyltransferase (LCAT). This resulted in reduced cholesteryl ester (CE) and increased free cholesterol (FC) levels in the plasma of mice expressing apoA-IM compared to WT apoA-I. These differences did not affect the rate of delivery of labeled cholesterol to the liver via SR-BI–mediated selective uptake or its subsequent excretion in the feces.
Within the limits of the in vivo assay, WT apoA-I and apoA-IM are equally efficient at promoting macrophage RCT, suggesting that if apoA-IM is more atheroprotective than WT apoA-I it is not due to an enhancement of macrophage RCT.
Apolipoprotein A-I (apoA-I) is the major protein component of HDL.1,2 Plasma apoA-I and HDL cholesterol levels are inversely associated with the risk of cardiovascular disease.3 The antiatherogenic properties of apoA-I are thought to arise, at least in part, from its central role in reverse cholesterol transport (RCT), the pathway by which excess cholesterol is effluxed from peripheral tissues and transported to the liver, where it is subsequently excreted from the body.3–5
The first natural mutant of apoA-I identified was the apoA-IMilano (apo A-IM) mutation (R173C).6 The apoA-IM mutation is located in the N-terminal helix bundle domain of the protein.7,8 Naturally occurring variants of apoA-I between residues 121 and 186 are often associated with low apoA-I levels.9 Individuals heterozygous for the apoA-IM mutation have very low plasma apoA-I and HDL cholesterol levels as well as moderately elevated triglycerides.10 Despite a lipid profile that is usually associated with a high risk of premature cardiovascular disease, apoA-IM carriers display no increase in cardiovascular disease or events.10–12 This has led to speculation that apoA-IM is a gain-of-function mutation that has enhanced cardio-protective effects,13–19 whereas others believe that wild-type (WT) apoA-I and apoA-IM are functionally equivalent.20,21 A clinical trial of repeated intravenous infusions of apoA-IM–phospholipid complexes demonstrated regression of existing atheromas after 5 weekly treatments,22,23 whereas other clinical trials focused on the ability of WT apoA-I/phosphatidylcholine disks to increase endothelial function.24,25
To date it has not been determined how the R173C substitution affects RCT in vivo. To investigate this question we performed the first head-to-head comparison of in vivo macrophage-RCT between WT apoA-I and apoA-IMilano using an assay developed in our laboratory.
The cDNAs of interest were expressed in Escherichia coli and the resulting proteins purified to greater than 95% purity by gel filtration chromatography.7 All proteins were stored at −80°C in lyophilized form and before use were dissolved in the appropriate buffer containing 6 mol/L Gdn HCl and dialyzed extensively before use.
Wt apoA-I cDNA was mutated using the Quikchange Site-Directed Mutagenesis Kit from Stratagene. The resulting cDNA were sequenced to confirm the presence of the intended mutation and submitted to the University of Pennsylvania Vector Core for use in creating the liver-specific apoA-I adeno-associated virus (AAV) serotype 8.26
Experiments were performed in male apoA-I–null mice obtained from Jackson Labs and fed a chow diet. For each experiment, 18 mice (n=6/group) received i.p. injection of AAV8 (1×1012 GC) containing either WT apoA-I, apoA-IM, or LacZ cDNA. On day 42 after vector injection, each animal received intraperitoneal injections of [3H]cholesterol-labeled J774 cells. Blood was collected at 2, 6, 24 and 48 hours. Feces was collected from 0 to 48 hours, and after exsanguinations at 48 hours bile and liver samples were collected as previously described.27
The endogenous LCAT cholesterol esterification rate (CER) in whole plasma was measured using a modified Stokke and Norum procedure.28–30 Seventy-five microliters of fresh non-radioactive plasma were incubated with a BSA solution containing 3×106 dpm of [3H]cholesterol on ice overnight to allow the [3H]cholesterol tracer to equilibrate evenly across the entire spectrum of lipoproteins. The plasma was then incubated at 37°C for 30 minutes as duplicates while a third aliquot was maintained at 4°C as a control. Lipids were extracted by TLC. Endogenous LCAT activity was expressed as nanomoles of CE formed per mL of plasma during the 30 minutes incubation.
After [3H]cholesterol labeling the cells (J774 or mouse peritoneal macrophage [MPM]), media containing the appropriate acceptor was added for up to 24 hours. In some experiments the ACAT inhibitor CP113 818 (2 µg/mL) was added to the media. To determine the cholesterol efflux, media were sampled at indicated times, filtered, and counted by liquid scintillation counting to determine the [3H] released. [3H] in the media was compared with total [3H] at time zero to determine the percent release of [3H]cholesterol.31–33 ABCA1-mediated efflux was calculated by %efflux+cAMP–%efflux−cAMP.
Rat Fu5AH hepatoma cells were prepared as described previously34 and incubated with 20% serum containing [3H]cholesterol and [3H]cholesteryl ester from mice expressing either WT apoA-I or apoA-IM. After 8 hours, the cells were washed 3 times with PBS, the cell lipids were extracted with isopropyl alcohol as previously described,35 and the levels of tritium label determined.
Data are from representative experiments and are expressed as mean±SD. Statistical test for significance was done using an unpaired t test or 1-way Anova followed by a Tukey test for pairwise comparisons.
ApoA-I–null mice (n=6 per group) were IP injected with AAV-expressing WT apoA-I, apoA-IM, or LacZ control at a dose of 1×1012 genome copies (GC). Six weeks after infection, WT apoA-I– and apoA-IM–expressing mice roughly reproduced the relative apoA-I and lipid levels of the human apoA-IM carriers and their controls36; apoA-IM–expressing mice had lower plasma apoA-I and substantially lower HDL-C levels compared with WT apoA-I-expressing mice (Table). Differences between the groups were not attributable to differential gene expression, as hepatic mRNA levels of WT apoA-I and apoA-IM were similar (supplemental Figure I). At week 6 after AAV injection, cholesterol-labeled J774 cells were injected for a macrophage RCT study. The [3H]-cholesterol counts in plasma, expressed as a percentage of total labeled cholesterol injected, at each time point were significantly higher for WT apoA-I compared to apoA-IM or LacZ mice (Figure 1A). Expression of human WT apoA-I significantly increased the transport of macrophage-derived cholesterol into the feces compared to the LacZ control (Figure 1B). However, there was no significant difference in counts in either the liver and bile (supplemental Table I) or the feces (Figure 1B) between WT apoA-I and apoA-IM mice despite the lower plasma levels of apoA-I and HDL cholesterol for apoA-IM–expressing mice. FPLC profiles of pooled plasma samples from the 48-hour time point indicated that the HDL formed by WT apoA-I and apoA-IMilano–expressing mice were similar (Supplemental Figure II).
In a subsequent experiment the dose of WT apoA-I AAV was reduced 5-fold, whereas the apoA-IM dose was held constant in an attempt to equalize the level of HDL-C between WT apoA-I– and apoA-IM–expressing mice. This treatment resulted in substantially reduced apoA-I levels in mice expressing WT apoA-I and similar levels of HDL-C in the WT apoA-I and apoA-IM mice, (Table). Compared with the lacZ control mice, expression of low dose WT apoA-I promoted significantly increased plasma counts and fecal excretion of tracer (Figure 1C and 1D), although to a lesser extent than the higher dose experiment. However, there remained no significant difference between apoA-IMilano–expressing and WT apoA-I– expressing mice in fecal tracer excretion despite the much lower plasma levels of WT apoA-I.
To gain further insight into the effect of apoA-IM compared with WT apoA-I on RCT, we investigated each of the steps of RCT individually; these steps include ABCA1-mediated cholesterol efflux, the conversion of free cholesterol to cholesteryl ester by LCAT, and SR-BI–mediated selective uptake of cholesterol and cholesteryl ester from HDL particles.3
To measure ABCA1-mediated cholesterol efflux, J774 macrophages (the cells used in the macrophage RCT experiments) and MPM were labeled with [3H]-cholesterol. After an equilibration period, 2.5% serum from either WT apoA-I– or apoA-IM–expressing mice (from RCT #1) was added for 4 hours. As previously reported for MPMs32 there were no differences in ABCA1-mediated cholesterol efflux (3.6±0.9% for WT apoA-I and 4.6±1.5% for apoA-IM). This finding was confirmed with J774 macrophages which were used in the macrophage RCT experiments described above (7.9±0.8% for WT apoA-I and 8.5±1.2% for apoA-IM; Figure 2A). This was a somewhat surprising observation given the lower concentration of apoA-IM compared to WT apoA-I in the serum samples (Table). To address this issue, recombinant WT apoA-I and apoA-IM (in the monomeric and homodimeric states) were used at different concentrations (0 to 20 µg/mL) as acceptors for ABCA1-mediated efflux of [3H]-cholesterol from J774 macrophages (Figure 2B). From these experiments, the kinetic constants of the ABCA1 reactions were determined and the Km values for cholesterol efflux to WT apoA-I, monomeric, and dimeric apoA-IMilano were 4.5±0.5, 2.1±0.3, and 4.8±0.9 µg apoA-I/ml, respectively. The VMax values of the reaction for WT apoA-I, monomeric, and dimeric apoA-IMilano were 7.5±0.3, 5.2±0.2, and 5.8±0.4% cholesterol efflux/4 hour, respectively.
The second step in the RCT process is the conversion of FC to CE by LCAT in the plasma. The cholesterol esterification rates (CER), as determined in the Materials and Methods, for apoA-IM– and LacZ-expressing mice were 45% and 18%, respectively, of that of the WT apoA-I–expressing mice (Figure 3A). Corresponding to the reduction in the CER, plasma CE levels in apoA-IM– and LacZ-expressing mice were significantly reduced compared to the WT apoA-I–expressing animals (Figure 3B).
The third step of the RCT pathway is the SR-BI–mediated selective uptake of cholesterol from HDL in the plasma into the liver. Serum samples from the 48-hour time point of the RCT experiment in Figure 2C, which contain [3H]-cholesterol and [3H]-cholesteryl ester, were diluted in media and applied to Fu5AH rat liver hepatoma cells that express high levels of SR-BI37 and incubated for 8 hours. After the 8-hour incubation, the cells were processed as described in the Materials and Methods to determine the amount of label that was taken up from the media into the cells (Figure 4). Serum from mice expressing either WT apoA-I or apoA-IM was equally effective at promoting SR-BI–mediated selective CE and FC uptake into these liver cells.
Carriers of the apoA-IM variant have reduced levels of plasma apoA-I and HDL, but no corresponding increase in cardiovascular disease compared to individuals with the WT apoA-I gene.10–12 One theory used to explain this inconsistency is that HDL containing apoA-IM are more efficient at promoting RCT. This study is the first direct comparison of the ability of WT apoA-I and apoA-IM to promote macrophage RCT in vivo.
ApoA-I–null mice were injected with AAV expressing either WT apoA-I, apoA-IM, or the LacZ control. The mice were then used in macrophage RCT experiments. In the first macrophage RCT experiment, the relative amounts of apoA-I and HDL-C levels in WT apoA-I– and apoA-IM–expressing mice closely approximated the levels seen in humans,36 with lower apoA-I and HDL-C levels in mice expressing apoA-IM (Table). When the mice were assayed for the ability to promote macrophage RCT, there was no significant difference between the WT apoA-I and apoA-IM groups in fecal excretion of tracer (Figure 1A and 1B) despite the lower HDL-C levels in the latter group. In the second experiment, the amount of WT apoA-I AAV used was reduced so that the resulting HDL-C levels in the WT apoA-I and apoA-IM mice were equal (Table). Surprisingly, despite much lower plasma apoA-I levels in the mice expressing WT apoA-I, there was still no significant difference in fecal tracer excretion between mice expressing WT apoA-I and mice expressing apoA-IM. This result suggests that apoA-IM is not more effective than WT apoA-I at promoting macrophage RCT as assessed by this model. To gain greater insight, we dissected the following 3 steps of the RCT pathway, ABCA1-mediated cholesterol efflux, LCAT-mediated conversion of FC to CE, and SR-BI–mediated selective uptake of cholesterol.
Serum from mice expressing apoA-I at levels similar to those listed in Table, RCT 1, was used to measure ABCA1-mediated cholesterol efflux (the transporter responsible for a majority of the cholesterol efflux37) from J774 macrophages, the cell type used in the macrophage RCT experiments. Both WT apoA-I and apoA-IM serum promoted efflux to similar extents (Figure 2A) despite lower levels of apoA-I in the apoA-IM mice compared to the WT apoA-I mice.32 To address this issue, we measured the effects of apoA-I concentration on ABCA1-mediated cholesterol efflux from J774 cells (Figure 2B) and determined that Km of the reactions were in the range of 2 to 5 µg/mL of apoA-I for WT and apoA-IM. In the serum of the WT apoA-I and apoA-IM mice, if one conservatively estimates that 5% of the circulating apoA-I is in the lipid-free or lipid-poor state,38,39 the concentrations would be 77 and 61 µg/mL, respectively. These levels are well above the Km of the ABCA1 reaction so that the reaction would be operating near maximal levels. Even if the fact that the levels of apoA-I present in interstitial fluid would be lower than in the serum40 is taken into account, the concentration of apoA-I would still be well above the Km of the reaction. The level of lipid-free or lipid-poor apoA-I in the second RCT experiment would also have been well above the saturation point of the ABCA1-mediated reaction. It should also be noted that measurements of pre-β HDL in human apoA-IM carriers determined that 15% of the apoA-I was in the lipid-free or lipid-poor state,36 consistent with saturating conditions for the apoA-I/ABCA1 reaction.41,42 It is also worth noting that although ABCA1 is responsible for transferring cholesterol to lipid-poor apoA-I, ABCG1 can transfer cholesterol to lipidated apoA-I and that the Km of the ABCG1 reaction is approximately 50 µg/mL HDL protein.43 The concentration of WT apoA-I and apoA-IM in the mice is above the Km of the ABCG1 reaction.
The second step of the RCT process, the conversion of FC to CE by LCAT in the plasma, was measured in plasma using well established methods.29 The results demonstrated that apoA-IM was unable to activate mouse LCAT to the same extent as WT apoA-I (Figure 3). This result with plasma HDL confirmed earlier work,44 showing that recombinant HDL made with dimeric apoA-IM was 70% less efficient as a substrate for human LCAT than recombinant HDL prepared with WT apoA-I. Structural analysis of the apoA-IM mutation45 determined that the loss of the arginine and the gain of the cysteine residue at position 173 destabilized the N-terminal helix bundle of the apoA-I molecule, a region that has been determined to be of particular importance for LCAT activation.9
The third step of RCT is the selective uptake of plasma HDL cholesterol by SR-BI into the liver. Serum from WT apoA-I and apoA-IM mice was equally effective at promoting SR-BI selective uptake (Figure 4). The apoA-IM mouse serum had a higher FC/CE ratio compared to the WT apoA-I serum (Figure 3B) because of the reduced LCAT activation by apoA-IM compared to WT apoA-I. Previous work has shown that SR-BI can mediate the selective uptake of both CE and FC from HDL, with the rate constant for the latter being some 60% higher.46 Thus, the alteration in FC/CE ratio between the WT and apoA-IM HDL is unlikely to be harmful to net hepatic selective uptake of HDL cholesterol. Furthermore, the Km for high affinity selective uptake from HDL by SR-BI is approximately 1 µg apoA-I/mL.47 Thus, the concentrations of apoA-I in the WT apoA-I and apoA-IM mouse serum samples are far above the Km that the reaction is saturated and any HDL compositional or quantity differences do not affect SR-BI–mediated selective uptake.
Despite lower apoA-I and HDL-C levels and a reduced ability to activated LCAT, expression of apoA-IM is as efficient at promoting RCT as expression of WT apoA-I as assessed by this mouse tracer assay. However, even when WT apoA-I is expressed at a lower level (plasma levels one-fourth of those in apoA-IM mice) to generate similar levels of HDL-C in both groups of mice, expression of apoA-IM is not more effective in promoting RCT. Thus, if apoA-IMilano does have an enhanced cardioprotective ability, it is likely attributable to other properties of this mutant protein; for example increased antioxidant activity15 arising from the presence of a cysteine residue in the primary structure of apoA-I or the ability of apoA-IMilano to modulate vascular function.12,25
Sources of Funding
This work was supported by AHA grant 0625372U and NIH grants HL22633, HL063768, and HL59407.
Reprints: Information about reprints can be found online at http://www.lww.com/reprints