Ability of Serum to Decrease Cellular AcylCoA:Cholesterol Acyl Transferase Activity Predicts Cardiovascular Outcomes

Share Embed


Descrição do Produto

Molecular Cardiology Ability of Serum to Decrease Cellular AcylCoA:Cholesterol Acyl Transferase Activity Predicts Cardiovascular Outcomes Julio A. Chirinos, MD; Juan P. Zambrano, MD; Simon Chakko, MD; Alan Schob, MD; Ronald B. Goldberg, MD; Guido Perez, MD; Armando J. Mendez, PhD Background—We evaluated whether cholesterol efflux activity of serum is associated with the presence of angiographic coronary artery disease (CAD) and the risk of major adverse cardiovascular events (MACE) and death. Methods and Results—We studied 168 men undergoing coronary angiography. Cholesterol efflux activity was measured in vitro by incubation of patient serum with human skin fibroblasts and defined as the ability of serum to decrease the pool of cholesterol available for esterification by the acylCoA:cholesterol acyl transferase (ACAT) reaction. We evaluated whether this activity was associated with the presence of CAD and the risk of MACE and death during a 4.5-year follow-up. Serum-induced changes in ACAT activity did not correlate with HDL levels or the presence of CAD. Patients in the highest tertile of change in ACAT activity had a significantly higher risk for MACE (HR, 2.15; 95% CI, 1.36 to 3.39; P⫽0.001) and death (HR, 2.23; 95% CI, 1.17 to 4.26; P⫽0.01). These correlations were independent of other risk markers including LDL, HDL, and C-reactive protein levels. Conclusions—Serum-induced depletion of cellular cholesterol available for esterification by ACAT was a strong, independent predictor of MACE and death. We speculate that the ability of serum to decrease ACAT activity depends on ATP binding cassette transporter A1 (ABCA1)–mediated efflux. Furthermore, serum samples that induce larger changes in ACAT activity contain increased levels of HDL particles that preferentially interact with ABCA1 and that these particles accumulate in the serum of patients because of low activity of ABCA1 in vivo preventing or limiting the extent of apoA-I lipidation. (Circulation. 2005;112:2446-2453.) Key Words: cardiovascular diseases 䡲 lipoproteins 䡲 cholesterol efflux 䡲 morbidity 䡲 risk factors

L

ipid metabolism has long been known to play a central role in the development of atherosclerosis. Elevated LDL and decreased HDL cholesterol (HDL-C) have been identified as important risk factors for the development of coronary artery disease (CAD). The antiatherogenic property of HDL has been attributed at least in part to the ability of HDL to promote cholesterol removal (efflux) from cells, the first step in the reverse cholesterol transport pathway.1 The ability of HDL to remove excess cholesterol from cultured cells is well established.2 Although blood levels of HDL are readily measured, comparisons of HDL function were first demonstrated by de la Moya-Llera et al.3 Since then, several studies have compared the ability of HDL to promote cholesterol efflux in subjects with a variety of conditions.4,5 Efflux of cellular cholesterol occurs through at least 3 distinct pathways.6 – 8 First, cholesterol efflux by aqueous diffusion is a bidirectional, energy-independent process and involves equilibrium of cholesterol molecules between cellular membranes and any acceptor, including HDL, able to bind or sequester cholesterol.8,9 Second, scavenger

receptor B-I, which mediates the selective uptake of HDL cholesteryl esters into cells, also facilitates the passive efflux of cholesterol from cells to HDL.10 Third, efflux of phospholipids and cholesterol mediated by the ATP– binding cassette transporter A1 (ABCA1) is an active, energy-requiring process that requires the presence of extracellular lipid–poor apolipoproteins.11,12 The contribution of the various efflux pathways to reverse cholesterol transport, especially in vivo, remains unknown and it is likely that all pathways contribute to the overall extent of reverse cholesterol transport. In the present study, serum cholesterol efflux activity was measured in vitro by incubation of patient serum with cholesterol-enriched human skin fibroblasts; cells that express passive and active, ABCA1-mediated cholesterol efflux pathways. Using 2 methods to assess cholesterol efflux induced by serum, we evaluated whether serum cholesterol efflux activity is associated with the presence of angiographic CAD and whether it predicts the risk for cardiovascular events and all-cause mortality during a 4.5-year follow-up.

Received November 15, 2004; revision received April 27, 2005; accepted May 3, 2005. From the Miller School of Medicine, University of Miami (J.A.C., J.P.Z., S.C., A.S., R.B.G., G.P., A.J.M.), the Diabetes Research Institute (R.B.G., A.J.M.), and the Veterans Affairs Medical Center (S.C., A.S.), Miami, Fla. Correspondence to Armando J. Mendez, PhD, Miller School of Medicine, University of Miami, Diabetes Research Institute (R-134), 1450 NW 10th Ave, Miami, FL 33138. E-mail [email protected] © 2005 American Heart Association, Inc. Circulation is available at http://www.circulationaha.org

DOI: 10.1161/CIRCULATIONAHA.104.521815

2446 Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2016

Chirinos et al

Methods Study Population We studied a cohort of 168 male veterans undergoing coronary angiography at the Miami Veterans Administration Medical Center between October 1998 and May 1999. The study was approved by the hospital’s Institutional Review Board and written informed consent was obtained from all patients. Indications for angiography included stable angina, abnormal cardiac stress test, acute coronary syndromes, cardiomyopathy, and valvular heart disease. Data recorded at study entry included age, ethnicity, height, weight, blood pressure, ejection fraction (measured by ventriculography at the time of coronary angiography or echocardiography within 1 month of the date of cardiac catheterization), smoking history, previous myocardial infarction (MI), history of peripheral vascular disease, congestive heart failure (CHF), diabetes mellitus, stroke, or revascularization procedures (CABG or percutaneous coronary intervention), and family history of CAD. The indication for cardiac catheterization and medications that patients were receiving at that time were also recorded.

Angiographic Studies Coronary angiography was carried out and images of the coronary tree were obtained in routine standardized projections. Significant CAD was defined as at least 1 stenotic lesion of ⱖ50% of the lumen diameter in ⱖ1 coronary arteries as identified by the patient’s attending cardiologist, who was unaware of the patient’s experimental laboratory findings. The number of coronary vascular territories involved with significant CAD (1-, 2-, or 3-vessel disease) was recorded. Left main coronary artery lesions were categorized as 2-vessel disease.

Laboratory Analysis Peripheral blood samples were collected just before cardiac catheterization. Blood was allowed to clot for 30 minutes at room temperature and serum collected after centrifugation. Serum aliquots were stored at ⫺80°C until analyzed. The following measurements were performed by automated analyzers with commercially available reagents and following instructions by the manufacturers (inter- and intra-assay coefficients of variation [CV] are indicated in parentheses): Total cholesterol (CV ⬍2.0% and ⬍3.0%, respectively) and triglycerides (CV ⬍3.2% and 4.7%, respectively; Roche Diagnostics), high sensitivity C-reactive protein (CRP; CV 2.8% and 3.6%, respectively), apoA-I (CV 5.3% and 8.1%, respectively) and apoB (CV 4.6% and 6.4%, respectively; Dade-Behring), apoA-II (CV 2.7% and 3.8%, respectively), free cholesterol (CV 3.8% and 5.7%, respectively), and phospholipids (CV 2.6% and 4.3%, respectively; Wako). HDL lipids were measured after precipitation of apoBcontaining lipoproteins13 (CV 3.2% and 4.7%, respectively). Verylow-density lipoprotein and LDL-C were estimated by calculation.14

Assays for Cellular Cholesterol Efflux Incubation of cholesterol-loaded human skin fibroblasts with the patient’s serum was performed to measure serum-induced cellular cholesterol efflux. Near-confluent fibroblast cultures were labeled with 3H-cholesterol (0.5 ␮Ci/mL in DMEM containing 5% serum) for 48 hours, loaded with 30 ␮g/mL nonlipoprotein cholesterol as previously described.15 Cells were incubated with DMEM containing 1% patient serum (in triplicate) for 6 hours; control cultures received DMEM alone. Efflux was assessed by 2 methods. Total cholesterol efflux was measured by the appearance of 3H-cholesterol into the extracellular medium after incubation with patient serum; this was calculated as the percentage of 3H label appearing in the medium relative to total 3H-cholesterol counts after incubation and corrected for efflux to medium without added serum. This method measures efflux mediated by both passive and active (eg, ABCA1mediated) pathways, although the contribution by each cannot be distinguished. The ability of serum to deplete cholesterol available for esterification by acyl CoA:cholesterol acyl transferase (ACAT) activity was assessed as a measure of intracellular cholesterol efflux

HDL Function and Cardiovascular Outcomes

2447

as described in detail.15 Briefly, the pool of intracellular cholesterol available for esterification by ACAT was measured during a 1-hour incubation with medium containing 0.5 ␮Ci/mL 14C-oleate after the incubation with patient serum. Cholesterol esterification was calculated as nanomoles of 14C-oleate incorporated into 14C-cholesteryl esters per milligram of cell protein per hour, and the change induced by serum was expressed as the percentage decrease in activity relative to incubations without serum. Previous studies have established that efflux of intracellular pools of cholesterol available for esterification by ACAT depends on an active, apolipoprotein dependent mechanism15,16 presumably mediated by ABCA1 activity.12,17 To account for temporal variation in culture conditions, a pooled serum control (combined from the sera of 5 male and 8 female normolipidemic subjects; total cholesterol, triglycerides, and HDL-C were 192, 112, and 66 mg/dL, respectively), stored identically to patient samples and thawed only once before use, was included in each assay as a normalizing reference to better compare samples analyzed in different assay runs.3 The efflux induced by the serum of each patient is expressed as relative total efflux and relative change in ACAT activity. Values ⬎1 had greater ability, and values ⬍1 had lesser ability to promote total or ABCA1-mediated cholesterol efflux than the reference sample. Efflux of 3H-cholesterol by the reference plasma was 6.7⫾0.6% (CV 8.9%) and the decrease in cholesterol esterification was 27.5⫾3.6% (CV 13.1%) over the course of the project (8 individual experiments, 6 to 8 replicates per experiment). Control experiments were performed to test whether depletion of cholesterol available for esterification by ACAT requires functional ABCA1 activity (Figure 1). The ability of serum samples obtained from control subjects to promote cholesterol efflux from cholesterol-enriched fibroblasts from a normal subject and a patient with Tangier disease were compared, the latter cells lacking ABCA1 activity.18,19 The sera tested had HDL levels ranging from 29 to 61 mg/dL and triglyceride levels from 135 to 493 mg/dL. When incubated with normal fibroblasts, all of the samples reduced the pool of cholesterol available for esterification by ACAT, although the extent varied among the samples without apparent association to serum lipid levels (Figure 1A). In contrast, incubation of the same samples with Tangier disease fibroblasts either had no measurable effect or caused an increase in ACAT activity. These data indicate that efflux of cholesterol from the ACAT substrate pool requires functional ABCA1 activity. To further test this concept, we examined ABCA1– dependent cholesterol efflux from normal fibroblasts treated with the ABCA1 inhibitor glyburide20 (Figure 1B). Glyburide treatment had no effect on basal ACAT activity compared with control cells (2785⫾132 versus 2572⫾215 pmol of oleate incorporated into cholesterol ester per milligram of cell protein, respectively). Glyburide inhibited the ability of serum to decrease ACAT activity for all of the samples tested (mean inhibition 95%, range 72% to 112%). 3H-Cholesterol efflux was inhibited by 25% in glyburide-treated cells compared with control incubations (not shown), similar to previous reports.21,22 These data further suggest that depletion of intracellular cholesterol occurs by a glyburide-sensitive pathway consistent with the concept that functional ABCA1 activity is required, whereas cholesterol efflux from other cellular pools is less affected. In another series of experiments, we compared the ability of serum and serum depleted of apolipoprotein B-containing lipoproteins (depleted of apoB by dextran sulfate precipitation)13 to promote cholesterol efflux. These results show that depletion of the very-lowdensity lipoprotein and LDL fractions had no detectable effect on the ability of serum to decrease ACAT activity (Figure 1C). In contrast, when total cholesterol efflux was measured, depletion of the apoBcontaining lipoproteins reduced efflux by 32% to 50% (Figure 1D), demonstrating a significant contribution of the apoB-containing lipoproteins on efflux of labeled cholesterol from cells to serumcontaining medium. In addition, the effect of the apoB-containing lipoproteins on the incorporation of oleate into cellular triacylglycerols was determined (Figure 1E). In general, incubation of cells with serum was mostly indistinguishable from serum-free control for the availability of oleate incorporation into triglycerides. These results

Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2016

2448

Circulation

October 18, 2005

Figure 1. Cellular cholesterol efflux and changes in ACAT activity mediated by serum or apoB-depleted serum. A, Human skin fibroblasts from a normal subject and a patient with Tangier disease enriched with cholesterol (see Methods) and change in cholesterol available for esterification by ACAT determined. Cells incubated with medium alone (control) or containing 1% serum from subjects with indicated serum lipid levels. B, Normal human skin fibroblasts treated with various serum samples as in A and with 5 ␮g/mL purified apoA-I, except that parallel cultures pretreated with 10 ␮mol/L glyburide for 30 min before addition of serum and during 6-hr incubation with serum. C, D, E, Normal human skin fibroblasts treated as in A and incubated with media alone (DMEM), patient serum or patient serum that had been depleted of apoB-containing lipoproteins by precipitation with dextran sulfate-Mn2⫹. After incubation of cells with serum, (C) ACAT activity determined by oleate incorporation in cholesteryl esters (CE), (D) 3H-cholesterol efflux from cells to medium, and (E) oleate incorporation into triglycerides (TG) determined. For all graphs, data expressed as mean⫾SD of triplicate incubations.

suggest that incubation of fibroblasts, under conditions used to assess cholesterol efflux, does not alter oleate uptake and utility for triglyceride synthesis, which is consistent with the assumption that cellular pools of oleate are labeled to relatively high and constant specific activities.23

Definitions of Events and Follow-Up Events were documented by patient interview and review of electronic hospital records. The primary combined end point was the first occurrence of any of the following major adverse cardiac events (MACE): death from any cause, myocardial infarction, unstable angina, revascularization with either PTCA or CABG (if these procedures were performed at least 30 days after study entry), and stroke. The secondary end point was all-cause mortality. The diagnosis of MI was made by the presence of suggestive symptoms, with either electrocardiographic evidence (new Q waves in ⱖ2 leads) or cardiac-marker evidence of infarction, according to the standard TIMI and American College of Cardiology definition. Unstable

angina was defined as ischemic discomfort at rest for at least 10 minutes prompting rehospitalization, combined with one of the following: ST segment or T wave changes, cardiac-marker elevations that were above the upper limit of normal but did not meet the criteria for MI, or a second episode of ischemic chest discomfort lasting ⬎10 minutes and that was distinct from the episode that had prompted hospitalization. Seven patients were lost for follow-up. One patient died 1 day after cardiac catheterization and was excluded from the analysis. The final analysis was performed with data from 160 patients. The mean follow-up among patients who did not have a MACE was 52.6 months (median 55.2; interquartile range 57.9 to 54.6).

Statistical Analysis Normally distributed continuous variables are expressed as mean ⫾ SD and were compared between groups with the unpaired t test; non-normally distributed continuous variables are expressed as median (interquartile range) and were compared with the Mann-

Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2016

Chirinos et al TABLE 1.

HDL Function and Cardiovascular Outcomes

2449

Baseline Characteristics of Patient Population (nⴝ160) No CAD (n⫽29)

Age, y, ⫾SD*

55.6⫾11

CAD (n⫽131) 63.5⫾9

Ethnicity, n (%) White

18 (62.1)

98 (74.8)

Black

6 (20.7)

16 (12.2)

5 (17.2)

17 (13.0)

Hispanic Body mass index, mean⫾SD

30.1⫾4

29.2⫾5

Systolic blood pressure, mm Hg (IQR)

130 (114–139)

134 (120–150)

Diastolic blood pressure, mm Hg (IQR)

72 (67–81)

70 (65–80)

Ejection fraction (IQR)

50 (35–60)

50 (35–55)

Smoking history, n (%)

4 (13.8)

Diabetes, n (%)†

5 (17.2)

49 (37.4)

History of stroke, n (%)

2 (6.9)

12 (9.2)

32 (24.4)

Peripheral vascular disease, n (%)

1 (3.4)

18 (13.7)

CHF, n (%)

5 (17.2)

24 (18.3)

Diuretics

5 (17.2)

30 (22.9)

␤-Blockers*

7 (24.1)

82 (62.6)

Calcium channel blockers

9 (31.0)

43 (32.8)

ACE inhibitors

13 (44.8)

73 (55.7)

Aspirin†

16 (55.2)

99 (75.6)

Statins

7 (24.1)

55 (42.0)

Other lipid-lowering drugs

1 (3.4)

Medication use, n (%)

3 (2.3)

Total cholesterol, mg/dL (IQR)

183 (152–221)

183 (164–218)

Triglycerides, mg/dL (IQR)†

161 (142–246)

211 (167–268)

Phospholipids, mg/dL (IQR)

203 (183–232)

214 (189–241)

Free cholesterol, mg/dL (IQR)

57 (44–74)

52 (45–62)

HDL-C, mg/dL (IQR)

35 (32–44)

37 (31–45)

ApoA-I, mg/dL (IQR)

118 (88–142)

115 (97–142)

ApoA-II, mg/dL (IQR)

27 (22–33)

28 (25–32)

ApoB, mg/dL (IQR)

99 (79–125)

106 (88–124)

LDL-C, mg/dL (IQR)

112 (77–142)

Serum-mediated change in ACAT activity

1.05 (0.64–1.40)

99 (83–126) 1.13 (0.82–1.42)

IQR indicates interquartile range. *P⬍0.001; †P⬍0.05.

Whitney U test. Continuous variables between ⬎2 groups were compared with one-way analysis of variance or the Kruskall-Wallis test as appropriate. Proportions were expressed as percentages and were compared with the ␹2 test. For survival analyses, we compared the highest tertile of the ABCA1-mediated efflux activity or the total efflux activity with the lower 2 tertiles. The cumulative incidence of event-free survival was assessed by the Kaplan-Meier method and curves compared with the log-rank statistic. Other univariate and multivariate survival analyses were performed with Cox regression. Analyses were done separately for the primary and secondary end points. Multivariate analyses were done incorporating all univariate predictors of the outcome and other potential confounders. All probability values are 2-tailed. Values of P⬍0.05 were considered statistically significant. Analyses were performed with the statistical package NCSS for Windows.

Results Baseline Angiographic and Clinical Correlations With Cholesterol Efflux Potential of Serum Baseline characteristics of patients with and without significant CAD in the study cohort are shown in Table 1. Patients with CAD were older, had a higher prevalence of diabetes mellitus, more frequent ␤-blocker and aspirin use, and higher levels of triglycerides. We found no significant differences in serum-mediated cell cholesterol efflux or the ability to decrease cellular ACAT activity between patients with and without significant CAD, nor was there a correlation between total efflux or changes in ACAT activity induced by serum and the number of vessels involved

Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2016

2450

Circulation

October 18, 2005 TABLE 2. Baseline Characteristics Patients in Tertiles of ABCA1–Mediated Cholesterol Efflux Activity Tertile 1

Tertile 2

Tertile 3

63.6

61.3

61.4

White

35 (66)

41 (76)

40 (76)

Black

7 (13)

8 (15)

7 (13)

11 (21)

5 (9)

6 (11)

30.0

29.3

28.7 132

ACAT Activity Age, y Ethnicity, n (%)

Hispanic Body mass index Systolic blood pressure, mm Hg

132

133

Diastolic blood pressure, mm Hg

70

72

70

0.5

0.5

13 (25)

12 (22)

11 (21)

Diabetes, n (%)

18 (34)

17 (32)

19 (36)

History of MI, n (%)

12 (23)

8 (15)

18 (34)

History of stroke, n (%)

5 (9)

1 (2)

8 (15)

History of percutaneous coronary intervention, n (%)

6 (11)

8 (15)

8 (15)

History of CABG, n (%)

8 (15)

8 (15)

11 (21)

Peripheral vascular disease, n (%)

7 (13)

3 (6)

9 (17)

12 (23)

9 (17)

8 (15)

Diuretics

15 (28)

10 (19)

10 (19)

␤-blockers

32 (60)

26 (48)

31 (59)

Left ventricular ejection fraction Smoking history, n (%)

CHF, n (%)

0.5

Medication use, n (%)

Calcium channel blockers

17 (32)

20 (37)

15 (28)

ACE inhibitors

28 (53)

28 (52)

30 (57)

Aspirin

40 (76)

38 (70)

37 (70)

Statins

13 (25)

27 (50)

22 (42)

0

15 (28)

13 (24)

16 (30)

1

11 (21)

11 (20)

9 (17)

2

8 (15)

13 (24)

8 (15)

3

No. of diseased vessels (%)

19 (36)

17 (32)

20 (38)

Total cholesterol, mg/dL*

193

195

176

Triglycerides, mg/dL†

236

197

192

Phospholipids, mg/dL

215

216

203

Free cholesterol, mg/dL

54

52

52

HDL-C, mg/dL

34

35

40

ApoA-I, mg/dL

109

124

109

ApoA-II, mg/dL

27

28

29

ApoB, mg/dL‡

111

107

95

LDL-C, mg/dL

102

107

95

Subjects in tertile 1 have highest, whereas those in tertile 3 have lowest ability to decrease cellular ACAT activity. *ANOVA P⫽0.046; †P⫽0.006; ‡P⬍0.001.

(P⬎0.05). Table 2 shows baseline characteristics of patients stratified in tertiles of serum-induced changes in ACAT activity. Patients in the higher tertile of activity (ie, those with the greatest ability to decrease cellular ACAT activity) had significantly higher levels of triglycerides and apoB and a lower proportion of them were using statins at baseline. Patients in the lowest tertile

of ACAT activity had lower total cholesterol levels than patients in the 2 upper tertiles. Levels of HDL-C and LDL-C, ApoA-I, ApoA-II, and free cholesterol were not significantly different between the groups. There were no other significant differences between these groups among the variables shown in Table 2.

Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2016

Chirinos et al

HDL Function and Cardiovascular Outcomes

2451

Figure 2. Predictors of the combined end point and all-cause mortality by univariate and multivariate analyses (Cox regression). A, For multivariate analyses, we used 3 different models. Model 1 included all univariate predictors of combined end point (ACAT activity tertile, age, ejection fraction [EF], CHF). Model 2 included all variables in model 1 plus systolic blood pressure, diastolic blood pressure, body mass index, statin use, ␤-blocker use, ACE inhibitor use, LDL, HDL, diabetes mellitus, smoking, and CRP. Model 3 included efflux tertile, ApoA-I, ApoA-II, HDL, HDL free cholesterol, and HDL phospholipids. B, For multivariate analysis, model 1 included all univariate predictors of all-cause mortality (efflux tertile, age, EF, CRP). Model 2 included all variables in model 1 plus systolic blood pressure, diastolic blood pressure, body mass index, statin use, ␤-blocker use, ACE inhibitor use, LDL, HDL, diabetes mellitus, CHF, and smoking. Model 3 included ACAT activity tertile, ApoA-I, ApoA-II, HDL, HDL free cholesterol, and HDL phospholipids.

Prospective Follow-Up During the follow-up period, 12.5% of patients had a new MI or unstable angina, 9.4% had a percutaneous coronary intervention, 4.4% had CABG, and 1.9% of patients had a new stroke. Death occurred in 23.1% of patients and 46.9% of patients had at least 1 MACE. Significant predictors of MACE on univariate analysis (Figure 2A) included age (HR/10-year increase 1.31; 95% CI 1.02 to 1.69; P⫽0.03), left ventricular ejection fraction (HR/10% increase 0.82; 95% CI 0.71 to 0.94; P⫽0.005), number of vascular territories affected with significant CAD (HR per vessel involved 1.23; 95% CI 1.02 to 1.5; P⫽0.03), and presence of CHF (HR 2.34; 95% CI 1.40 to 3.91; P⫽0.001). HDL levels did not predict MACE (HR 1.0; 95% CI 0.98 to 1.02; P⫽0.93). Patients in the highest tertile of ACAT activity had a significantly higher risk for MACE (Figures 2A and 3A, HR 2.15; 95% CI 1.36 to 3.39; P⫽0.001). ACAT activity, handled as a continuous variable in Cox regression, also predicted MACE (HR per unit increase 1.81; 95% CI 1.27 to 2.59; P⫽0.001). When adjusted for all of the univariate predictors of the combined end point, patients in the highest ACAT activity tertile had a significantly increased risk of MACE (HR 2.04; 95% CI 1.26 to 3.30; P⫽0.003). After adjustment for other potential confounders, including systolic blood pressure, diastolic blood pressure, body mass index, statin use, ␤-blocker use, angiotensin-converting enzyme (ACE) inhibitor use, LDL, HDL, diabetes, smoking, and CRP, the correlation persisted (HR 2.08; 95% CI 1.24 to 3.49; P⫽0.005). Similarly, after adjusting for HDL-C, ApoA-I, ApoA-II, HDL free cholesterol, and HDL phospholipids, the ability of serum to decrease cellular ACAT activity remained a significant predictor of MACE (HR 1.91; 95% CI 1.16 to 3.17; P⫽0.01). Significant predictors of all-cause mortality by univariate analysis (Figure 2B) included age (HR/10-year increase 1.71; 95% CI 1.15 to 2.54; P⫽0.007), ejection fraction (HR/10% increase 0.69; 95% CI 0.57 to 0.85; P⫽0.005), and CRP (HR/mg/dL increase 1.15; 95% CI 1.003 to 1.33; P⫽0.04). HDL levels did not predict all-cause mortality (HR 0.98; 95% CI 0.96 to 1.02; P⫽0.93).

Patients in the highest tertile of ACAT activity had a significantly higher risk of death (Figures 2B and 3B; HR 2.23; 95% CI 1.17 to 4.26; P⫽0.01). This correlation was independent of all univariate predictors of mortality (HR

Figure 3. Kaplan-Meier survival plots for patients in highest tertile of serum-induced decrease in cellular ACAT activity compared with those in lower 2 tertiles of activity. A, Event-free survival for combined end point. B, Corresponding curves for allcause mortality. Event-free survival was significantly lower in patients in highest ACAT activity tertile compared with 2 lower tertiles (log-rank P for combined end point 00008; log-rank P for all-cause mortality 0.01).

Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2016

2452

Circulation

October 18, 2005

2.21; 95% CI 1.12 to 4.36, P⫽0.02). After adjustment for other potential confounders, including the presence of diabetes, CHF, smoking, number of vessels involved with significant CAD, systolic blood pressure, diastolic blood pressure, body mass index, statin use, ␤-blocker use, ACE inhibitor use, LDL, HDL, and CRP, the correlation persisted (HR 2.28; 95% CI 1.06 to 4.88; P⫽0.03). Total cholesterol efflux activity did not correlate with the risk of MACE or death. A trend was observed for a higher risk of MACE in the highest tertile of total cholesterol efflux (HR 1.48; 95% CI 0.94 to 2.35; P⫽0.09) or when total efflux activity was handled as a continuous variable (HR per unit increase 2.85; 95% CI 0.91 to 9.0; P⫽0.07). This trend disappeared after adjusting for serum-mediated changes in cellular ACAT activity. Because previous studies have suggested that smaller, lipiddepleted HDL particles serve as better cholesterol acceptors from cells by the ABCA1-mediated efflux pathway,24 we estimated the relative size distribution of HDL particles by calculating the molar ratio of HDL-C/(apoA-I ⫹ apoA-II). This estimate of HDL size was highly correlated with both the fractional catabolic rate of HDL and HDL size in previous studies.25,26 We found a significant correlation between this estimate of HDL size and the change in ACAT activity in response to incubation with serum (r⫽0.44; P⬍0.0001). We also found a significant inverse correlation between estimated HDL size and serum triglyceride levels (r⫽⫺0.40, P⬍0.0001), consistent with previous results26; however, there was no correlation between estimated HDL size and MACE or death.

Discussion We tested the hypothesis that cellular cholesterol efflux mediated by serum would predict the extent of CAD in a group of subjects undergoing coronary angiography and whether it would predict adverse outcomes on follow-up. Cholesterol efflux was measured by the appearance of labeled cholesterol into medium containing serum (total efflux) or by the ability of serum to deplete cellular pools of cholesterol available for esterification by ACAT in cultured fibroblasts. The latter measure of efflux is assumed to require cholesterol acceptors that promote cholesterol efflux through apolipoprotein-dependent mechanisms mediated by ABCA1 activity.11,15,27 Although there was a broad response in the ability of different serum samples to promote cholesterol efflux by either method, there were no relationships between efflux activities and the number of coronary vessel territories with hemodynamically significant CAD. The study subjects were observed prospectively for 4.5 years to evaluate the incidence of MACE and all-cause mortality. Evaluation of these data showed a strong relationship between the ability of patient serum to reduce cellular ACAT activity and outcomes. Patients in the highest tertile of ACAT activity, by definition representing one third of our population, demonstrated double the risk of adverse outcomes, suggesting that this marker may account for a significant proportion of the risk.28 This is the first study to demonstrate that a test of HDL function predicts cardiovascular events and mortality. The predictive ability of patient serum to deplete cellular ACAT activity in an in vitro assay was independent of other documented risk factors including HDL, LDL, and CRP

levels. The fact that patients with the greatest ability to deplete intracellular pools of cholesterol available for esterification by the ACAT reaction had a significantly greater risk of MACE seems somewhat counterintuitive, if we assume that the ability of HDL to promote cholesterol efflux plays an important role in reverse cholesterol transport and prevention of CAD. Our original hypothesis, although not supported by these data, was that subjects with the greatest HDL functionality would exhibit lower CAD burden, based on the assumption that high efflux capacity would prevent formation or progression of atherosclerosis. To explain these apparently paradoxical observations, we speculate that depletion of intracellular cholesterol depends on ABCA1–dependent cholesterol efflux pathways. Patient sera with the greatest ability to affect this activity in vitro, reflected by the greatest ability to decrease ACAT activity, contain increased levels of HDL particles that preferentially interact with ABCA1. Data from several studies have suggested that these particles are likely to be small, lipid-deficient apoA-I-containing particles.8 We further speculate that these particles accumulate in the serum of patients caused either by low activity of ABCA1 in vivo preventing or limiting the extent of apoA-I lipidation or possibly by a relative increase in generation of such particles via the cholesteryl ester transfer protein–hepatic lipase reaction.29 This conclusion is consistent with previous studies demonstrating increased levels of pre␤-migrating HDL in patients with documented CAD.30 In this study, estimated HDL size using the molar ratio of HDL-C/(apoA-I ⫹ apoA-II) modestly correlated with ABCA1-mediated efflux activity but did not correlate with adverse outcomes. Because this calculation estimates an average HDL particle size, it may not accurately reflect the presence or relative accumulation of smaller HDL that may most avidly participate in ABCA1-mediated cholesterol efflux. Conversely, other factors not related to HDL size may also influence the extent of ABCA1-mediated cholesterol efflux. Additional studies are needed to elucidate the role of HDL particle size distribution in risk prediction and its influence on cholesterol efflux pathways. Reverse cholesterol transport may not only have an effect on the development and progression of atherosclerosis but also may have an impact on plaque vulnerability. The vulnerability of an atherosclerotic plaque is dependent on the magnitude of its lipid core, the degree of inflammation within the plaque, and the smooth muscle content of its fibrous cap.31 Thrombosis often occurs in moderately stenosed or angiographically normal coronary arteries.32 HDL function in vitro may correlate with plaque instability; the inability of HDL-mediated cholesterol efflux to reduce plaque lipid content, resulting from low levels of ABCA1 activity in vivo, could explain the increase in MACE in the subjects with the highest ABCA1– dependent cholesterol efflux activity despite an apparent lack of correlation with angiographic CAD. ABCA1 plays a rate-limiting role in reverse cholesterol transport and HDL formation. The absence of ABCA1 in humans and mice leads to a profound reduction in plasma HDL levels and accumulation of lipid in macrophage-rich tissue, demonstrating the physiological importance of the ABCA1 pathway in cholesterol clearance.17 Elevated levels of unesterified cholesterol are toxic to cells.33,34 Studies by Feng and Tabas35 have shown that free cholesterol overaccumulation, although initially increasing ABCA1 levels, leads to free cholesterol–induced accelerated degradation of the transporter before causing cell death. Consequently, ABCA1-

Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2016

Chirinos et al mediated cholesterol efflux activity is decreased, thus implicating loss of ABCA1 activity as a critical step in accelerating macrophage cell death.35 Our study is limited by the relatively small number of patients from a selected population of men referred for coronary angiography at a tertiary medical institution, which is not representative of the entire population at risk for cardiovascular events. It does prove, however, the important concept that HDL quality rather than only quantity is critical in protecting against cardiovascular events. Additional studies are required to confirm this finding in other populations and to assess whether manipulation of HDL efflux potential represents a therapeutic target for the prevention and treatment of atherosclerotic disease.

Acknowledgments This work was funded by support from the American Heart Association, Grant in Aid (Grant #9950534N to Dr Mendez) and The Retirement Research Foundation (Drs Goldberg and Mendez).

HDL Function and Cardiovascular Outcomes

17. 18.

19.

20.

21.

22.

23.

References 1. Rader DJ. Regulation of reverse cholesterol transport and clinical implications. Am J Cardiol 2003;92:42J– 49J. 2. Phillips MC, Johnson WJ, Rothblat GH. Mechanisms and consequences of cellular cholesterol exchange and transfer. Biochim Biophys Acta. 1987;906:223–276. 3. de la Llera MM, Atger V, Paul JL, Fournier N, Moatti N, Giral P, Friday KE, Rothblat G. A cell culture system for screening human serum for ability to promote cellular cholesterol efflux. Relations between serum components and efflux, esterification, and transfer. Arterioscler Thromb. 1994;14:1056 –1065. 4. Brites FD, Cavallero E, De Geitere C, Nicolaiew N, Jacotot B, Rosseneu M, Fruchart JC, Wikinski RL, Castro GR. Abnormal capacity to induce cholesterol efflux and a new LpA-I pre-beta particle in type 2 diabetic patients. Clin Chim Acta. 1999;279:1–14. 5. Autran D, Attia N, Dedecjus M, Durlach V, Girard-Globa A. Postprandial reverse cholesterol transport in type 2 diabetic patients: effect of a lipid lowering treatment. Atherosclerosis. 2000;153:453– 460. 6. Tall AR, Costet P, Wang N. Regulation and mechanisms of macrophage cholesterol efflux. J Clin Invest. 2002;110:899 –904. 7. Yancey PG, Bortnick AE, Kellner-Weibel G, de la Llera-Moya M, Phillips MC, Rothblat GH. Importance of different pathways of cellular cholesterol efflux. Arterioscler Thromb Vasc Biol. 2003;23:712–719. 8. Yancey PG, Kawashiri M, Moore R, Glick JM, Williams DL, Connelly MA, Rader DJ, Rothblat GH. In vivo modulation of HDL phospholipid has opposing effects on SR-BI- and ABCA1-mediated cholesterol efflux. J Lipid Res. 2004;45:337–346. 9. Phillips MC, Gillotte KL, Haynes MP, Johnson WJ, Lund-Katz S, Rothblat GH. Mechanisms of high density lipoprotein-mediated efflux of cholesterol from cell plasma membranes. Atherosclerosis. 1998;137:S13–S17. 10. Ji Y, Jian B, Wang N, Sun Y, de la Llera-Moya M, Phillips MC, Rothblat GH, Swaney JB, Tall AR. Scavenger receptor BI promotes high density lipoproteinmediated cellular cholesterol efflux. J Biol Chem. 1997;272:20982–20985. 11. Oram JF. HDL apolipoproteins and ABCA1: partners in the removal of excess cellular cholesterol. Arterioscler Thromb Vasc Biol. 2003;23:720–727. 12. Oram JF, Lawn RM. ABCA1: the gatekeeper for eliminating excess tissue cholesterol. J Lipid Res. 2001;42:1173–1179. 13. Grauholt AM, Grande P, Horby-Petersen J, Jensen J, Rostgaard M, Meinertz H. High-density lipoprotein cholesterol assay by magnesium dextransulphate precipitation. Scand J Clin Lab Invest. 1986;46:715–721. 14. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18:499 –502. 15. Mendez AJ. Cholesterol efflux mediated by apolipoproteins is an active cellular process distinct from efflux mediated by passive diffusion. J Lipid Res. 1997;38:1807–1821. 16. Oram JF, Mendez AJ, Slotte JP, Johnson TF. High density lipoprotein apolipoproteins mediate removal of sterol from intracellular pools but not

24.

25.

26.

27.

28. 29.

30.

31. 32.

33.

34.

35.

2453

from plasma membranes of cholesterol-loaded fibroblasts. Arterioscler Thromb. 1991;11:403– 414. Oram JF. ATP-binding cassette transporter A1 and cholesterol trafficking. Curr Opin Lipidol. 2002;13:373–381. Mendez AJ, Lin G, Wade DP, Lawn RM, Oram JF. Membrane lipid domains distinct from cholesterol/sphingomyelin-rich rafts are involved in the ABCA1-mediated lipid secretory pathway. J Biol Chem. 2001;276: 3158–3166. Oram JF, Mendez AJ, Lymp J, Kavanagh TJ, Halbert CL. Reduction in apolipoprotein-mediated removal of cellular lipids by immortalization of human fibroblasts and its reversion by cAMP: lack of effect with Tangier disease cells. J Lipid Res. 1999;40:1769 –1781. Becq F, Hamon Y, Bajetto A, Gola M, Verrier B, Chimini G. ABC1, an ATP binding cassette transporter required for phagocytosis of apoptotic cells, generates a regulated anion flux after expression in Xenopus laevis oocytes. J Biol Chem. 1997;272:2695–2699. Nieland TF, Chroni A, Fitzgerald ML, Maliga Z, Zannis VI, Kirchhausen T, Krieger M. Cross-inhibition of SR-BI- and ABCA1-mediated cholesterol transport by the small molecules BLT-4 and glyburide. J Lipid Res. 2004;45:1256 –1265. Burgess JW, Kiss RS, Zheng H, Zachariah S, Marcel YL. Trypsinsensitive and lipid-containing sites of the macrophage extracellular matrix bind apolipoprotein A-I and participate in ABCA1-dependent cholesterol efflux. J Biol Chem. 2002;277:31318 –31326. Goldstein JL, Dana SE, Brown MS. Esterification of low density lipoprotein cholesterol in human fibroblasts and its absence in homozygous familial hypercholesterolemia. Proc Natl Acad Sci U S A. 1974;71:4288 – 4292. Favari E, Lee M, Calabresi L, Franceschini G, Zimetti F, Bernini F, Kovanen PT. Depletion of pre-␤-high density lipoprotein by human chymase impairs ATP-binding cassette transporter A1- but not scavenger receptor class B type I-mediated lipid efflux to high density lipoprotein. J Biol Chem. 2004;279:9930 –9936. Brinton EA, Eisenberg S, Breslow JL. Increased apo A-I and apo A-II fractional catabolic rate in patients with low high density lipoprotein-cholesterol levels with or without hypertriglyceridemia. J Clin Invest. 1991;87:536–544. Velez-Carrasco W, Lichtenstein AH, Li Z, Dolnikowski GG, Lamon-Fava S, Welty FK, Schaefer EJ. Apolipoprotein A-I and A-II kinetic parameters as assessed by endogenous labeling with [2H3]leucine in middle-aged and elderly men and women. Arterioscler Thromb Vasc Biol. 2000;20:801– 806. Oram JF, Mendez AJ, Slotte JP, Johnson TF. High density lipoprotein apolipoproteins mediate removal of sterol from intracellular pools but not from plasma membranes of cholesterol-loaded fibroblasts. Arterioscler Thromb. 1991;11:403– 414. Manolio T. Novel risk markers and clinical practice. N Engl J Med. 2003;349:1587–1589. Borggreve SE, De Vries R, Dullaart RP. Alterations in high-density lipoprotein metabolism and reverse cholesterol transport in insulin resistance and type 2 diabetes mellitus: role of lipolytic enzymes, lecithin: cholesterol acyltransferase and lipid transfer proteins. Eur J Clin Invest. 2003;33:1051–1069. Asztalos BF, Roheim PS, Milani RL, Lefevre M, McNamara JR, Horvath KV, Schaefer EJ. Distribution of ApoA-I-containing HDL subpopulations in patients with coronary heart disease. Arterioscler Thromb Vasc Biol. 2000;20:2670 –2676. Price MJ, Shah PK. New strategies in managing and preventing atherosclerosis: focus on HDL. Rev Cardiovasc Med. 2002;3:129 –137. Bezerra HG, Higuchi ML, Gutierrez PS, Palomino SA, Silvestre JM, Libby P, Ramires JA. Atheromas that cause fatal thrombosis are usually large and frequently accompanied by vessel enlargement. Cardiovasc Pathol. 2001;10:189 –196. Warner GJ, Stoudt G, Bamberger M, Johnson WJ, Rothblat GH. Cell toxicity induced by inhibition of acyl coenzyme A:cholesterol acyltransferase and accumulation of unesterified cholesterol. J Biol Chem. 1995; 270:5772–5778. Tabas I, Marathe S, Keesler GA, Beatini N, Shiratori Y. Evidence that the initial up-regulation of phosphatidylcholine biosynthesis in free cholesterol-loaded macrophages is an adaptive response that prevents cholesterol-induced cellular necrosis. Proposed role of an eventual failure of this response in foam cell necrosis in advanced atherosclerosis. J Biol Chem. 1996;271:22773–22781. Feng B, Tabas I. ABCAl-mediated cholesterol efflux is defective in free cholesterol-loaded macrophages. J Biol Chem. 2002;277:43271– 43280.

Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2016

Ability of Serum to Decrease Cellular AcylCoA:Cholesterol Acyl Transferase Activity Predicts Cardiovascular Outcomes Julio A. Chirinos, Juan P. Zambrano, Simon Chakko, Alan Schob, Ronald B. Goldberg, Guido Perez and Armando J. Mendez Circulation. 2005;112:2446-2453 doi: 10.1161/CIRCULATIONAHA.104.521815 Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 2005 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7322. Online ISSN: 1524-4539

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circ.ahajournals.org/content/112/16/2446

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation is online at: http://circ.ahajournals.org//subscriptions/

Downloaded from http://circ.ahajournals.org/ by guest on February 15, 2016

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.