Plasma lecithin:cholesterol acyltransferase and carotid intima-media thickness in European individuals at high cardiovascular risk.

Lecithin:cholesterol acyltransferase (LCAT) is the enzyme responsible for cholesterol esterification in plasma. LCAT is a major factor in HDL remodeling and metabolism, and it has long been believed to play a critical role in macrophage reverse cholesterol transport (RCT). The effect of LCAT on human atherogenesis is still controversial. In the present study, the plasma LCAT concentration was measured in all subjects (n = 540) not on drug treatment at the time of enrollment in the multicenter, longitudinal, observational IMPROVE study. Mean and maximum intima-media thickness (IMT) of the whole carotid tree was measured by B-mode ultrasonography in all subjects. In the entire cohort, LCAT quartiles were not associated with carotid mean and maximum IMT (P for trend 0.95 and 0.18, respectively), also after adjustment for age, gender, HDL-cholesterol (HDL-C), and triglycerides. No association between carotid IMT and LCAT quartiles was observed in men (P=0.30 and P=0.99 for mean and maximum IMT, respectively), whereas carotid IMT increased with LCAT quartiles in women (P for trend 0.14 and 0.019 for mean and maximum IMT, respectively). The present findings support the concept that LCAT is not required for an efficient reverse cholesterol transport and that a low plasma LCAT concentration and activity is not associated with increased atherosclerosis.

to explore the relationship between LCAT and preclinical atherosclerosis, as assessed by carotid IMT.

Participants
A total of 540 individuals free of any pharmacological treatments were enrolled among those recruited in the IMPROVE study, a multicenter, longitudinal, observational study, funded by the European Union within the Fifth Framework Programme. The IMPROVE study ( 18 ) started in March 2004, involved seven centers in fi ve European countries (Finland, France, Italy, the Netherlands, and Sweden), and recruited a total of 3,711 individuals (1,050 in Finland, 501 in France, 1,095 in Italy, 532 in the Netherlands, and 533 in Sweden) who met the eligibility criteria. The criteria were as follows: 54-to 79-year-old men and women who had at least three cardiovascular risk factors (dyslipidemia, hypertension, diabetes, smoking, or family history of cardiovascular disease), who were asymptomatic for cardiovascular disease, and who were free of any conditions that might limit longevity or carotid IMT visualization ( 18 ). All individuals not on drug therapy at the time of recruitment were included in the present analysis. All patients gave written, informed consent.

Biochemical analyses
Blood was collected after an overnight fast. Plasma samples were kept at -80°C prior to shipment to Milano and were not thawed before analysis. Plasma total and unesterifi ed cholesterol, HDL-C, and triglyceride levels were determined with certifi ed enzymatic methods by using a Roche diagnostics Integra 400 autoanalyzer. Plasma LDL-cholesterol was calculated with the Friedewald's formula. Plasma LCAT concentration was measured by an immunoenzymatic assay developed in our laboratory ( 19 ). Briefl y, this is a competitive ELISA using an anti-peptide rabbit antibody generated against the peptide corresponding to residues 393-416 of human LCAT as primary antibody and using peroxidase-conjugated goat anti-rabbit immunoglobulins as secondary antibody. The intra-assay and inter-assay CVs are 3% and 9%, respectively.

Ultrasonographic variables
Carotid ultrasonography was performed as described ( 18 ). Briefl y, the far walls of the left and right common carotids, bifurcations, and internal carotids were visualized in anterior, lateral, and posterior projections and recorded on sVHS videotapes. Carotid IMT measurements were performed in a centralized laboratory (Department of Pharmacological Sciences, University of Milan, Italy) using a dedicated software (M'Ath, Metris S.R.L., France). All carotid measurements were averaged to calculate the IMT mean for each subject; the highest IMT value among the three carotid segments was defi ned as the IMT max . Repeatability of carotid IMT measurements was previously described ( 18 ).

Statistical analyses
Numerical variables were summarized as means and 95% confi dence intervals (CI). Data with skewed distributions (triglycerides and carotid IMT measurements) were summarized as geometric means and 95% CI and log-transformed before further analyses. Categorical variables were summarized as percentages. Clinical and anthropometric variables were compared between genders by unpaired Students' t -test or Chi-square test, as appropriate. Linear associations between LCAT and carotid IMT measurements, crude or adjusted for potential confounders, were reaction accounts for the synthesis of most of the plasma CE. The human LCAT gene (4.5 kb) is localized on chromosome 16 (region 16q22) and encodes a glycoprotein of 416 residues, with a molecular weight of 63 kDa. In blood, LCAT preferentially binds to high density lipoproteins (HDL), where it is activated by apolipoprotein A-I; a minority of plasma LCAT circulates bound to apoB-containing lipoproteins ( 1 ).
LCAT has a key place in HDL metabolism, and it has long been believed to play a critical role in macrophage reverse cholesterol transport (RCT) by maintaining the cholesterol gradient between the cell membrane and extracellular acceptors ( 2 ). However, recent fi ndings have questioned this concept. Human LCAT overexpression in mice remarkably increases plasma HDL levels but does not enhance macrophage RCT in vivo ( 3 ). Conversely, LCATdefi cient mice display a preserved macrophage RCT in vivo despite the severe plasma HDL reduction ( 3 ). These fi ndings are consistent with data in humans, showing that sera from genetic carriers of LCAT defi ciency can effi ciently remove cholesterol from macrophages ( 4 ) and that HDL is capable of directly delivering large amounts of unesterifi ed cholesterol to the liver via the scavenger receptor type BI (SR-BI) without LCAT-mediated conversion into CE ( 5,6 ).
The effect of LCAT on atherogenesis is controversial. Investigations in various animal models overexpressing or lacking LCAT provided inconsistent results (recently reviewed in Ref. 7 ). LCAT overexpression in mice or rabbits remarkably increased plasma HDL levels and was associated with protection against diet-induced atherosclerosis in rabbits but not in mice. Mice lacking the LCAT gene showed the expected reductions in plasma HDL levels but had less or more atherosclerosis than their counterparts with active LCAT, mostly depending on the plasma levels of apoB-containing lipoproteins.
The relationship between LCAT and vascular disease in humans has been investigated in a limited number of studies, also with confl icting results. Early cross-sectional investigations reported either decreased ( 8,9 ) or increased ( 10 ) LCAT activity in patients with angiographically proven coronary artery disease (CAD). More recently, higher plasma LCAT activity was associated with increased incidence of CAD in men participating in the populationbased PREVEND study ( 11 ). However, no association was found between plasma LCAT concentration, which strongly correlates with LCAT activity, and future cardiovascular events in a larger cohort of men from the EPIC-Norfolk population study ( 12 ). Notably, in this latter study, higher plasma LCAT levels were instead associated with a greater risk of CAD in women ( 12 ). Carotid intima-media thickness (IMT) is one of the best established and most commonly used measures of preclinical atherosclerosis ( 13 ). It correlates with many cardiovascular risk factors, including low plasma HDL-C levels ( 14,15 ), and with the severity of CAD ( 16 ). Most importantly, in prospective studies, carotid IMT predicts clinical CAD ( 17 ). In the present study, we measured plasma LCAT concentration in European men and women at high cardiovascular risk Plasma LCAT concentration and carotid IMT Table 2 shows the characteristics of participants stratifi ed by quartiles of plasma LCAT concentration. In the whole group of subjects, plasma LCAT concentration was not associated with carotid IMT mean and IMT max ( Table 3 ). Adjustment for age, gender, HDL-C, and triglycerides did not change the results. No association between LCAT concentration and carotid IMT was observed in men, either before or after adjustment for age, HDL-C, and triglycerides ( Table 3 ). In contrast, carotid IMT increased with LCAT quartiles in women ( Table 3 ). The association between LCAT and IMT max was positive and signifi cant, although losing full signifi cance after adjustment for age, HDL-C, and triglycerides. Nevertheless, the interaction gender × LCAT was not signifi cant ( P = 0.19). The association between LCAT and IMT mean was also positive but not signifi cant.

DISCUSSION
The present study shows that low plasma levels of LCAT are not associated with increased carotid IMT in individuals at high cardiovascular risk. However, gender-specifi c analysis shows that low plasma LCAT levels are associated, although with borderline signifi cance, with decreased carotid IMT in women. These results are in agreement with a recent study showing that low LCAT activity is associated with reduced carotid IMT in a small group of patients with the metabolic syndrome ( 20 ). Carotid IMT, one of the best established and most commonly used surrogate markers of human atherosclerosis ( 13 ), correlates with the severity of coronary heart disease (CHD) ( 16 ) and predicts clinical CHD in prospective studies ( 17 ). We have not analyzed cardiovascular events in the studied population, but the relation we have observed between LCAT and carotid IMT likely refl ects a similar relation between LCAT and events. Indeed, recent fi ndings have shown that low plasma LCAT levels are not associated with increased vascular tested by ANCOVA. All tests were two-sided, and P values below 0.05 were regarded as signifi cant. In gender-specifi c analysis of the relationship between LCAT and IMT, the signifi cance threshold was set at 0.025 to account for multiple comparisons. The sample size of 540 subjects yielded 80% statistical power to detect as signifi cant a correlation coeffi cient greater than 0.12 or greater than 0.17 in the analysis stratifi ed by gender. All analyses were performed using the SAS statistical package (SAS Institute Inc., Cary, NC) version 9.13.

Baseline characteristics of study participants
A total of 540 subjects, 247 women and 293 men, were enrolled in the study. Almost half of the participants were hypertensive, 56% of them were current or former smokers, and 11% had diabetes ( Table 1 ). On average, plasma total and LDL-cholesterol levels were slightly above the upper normal limit, while HDL-C and triglyceride levels were within the normal range ( Table 1 ). Plasma LCAT concentration was also within the normal range (3.1-6.7 g/ml).
The prevalence of current/former smokers and diabetes was lower in women than men ( Table 1 ). Women had signifi cantly higher total, LDL-, and HDL-cholesterol levels than men, but they had lower triglycerides. The average plasma LCAT concentration was similar in women and men ( Table 1 ).

Plasma LCAT concentration and lipid levels
In the whole group of examined subjects, plasma LCAT concentration correlated positively with HDL-C ( R = 0.146, P = 0.0007) and negatively with triglycerides ( R = Ϫ 0.086, P = 0.046). As expected, plasma LCAT concentration also correlated signifi cantly and negatively with the unesterifi ed/total cholesterol ratio ( R = Ϫ 0.140, P = 0.002), confi rming that, in the absence of genetic defects, LCAT concentration refl ects enzyme activity. the free cholesterol gradient between cell membranes and plasma acceptors ( 2 ). However, recent evidence has questioned this concept. First, sera from humans and mice lacking LCAT activity can effi ciently remove cholesterol from macrophages via the ABCA1 transporter ( 3,4 ), likely the major player in cholesterol removal from macrophages ( 23,24 ), in a process less sensitive to cholesterol gradient than passive diffusion. Second, LCATdefi cient mice display a preserved macrophage-to-feces RCT in vivo ( 3 ), in agreement with the observation that unesterifi ed cholesterol within HDL may be effi ciently delivered to the liver and excreted into the bile without conversion into CE ( 5, 6 ). Thus, although LCAT is clearly important for HDL remodeling and metabolism, it may not be as critical in maintaining a normal rate of macrophage RCT.

Limitations
The studied population only included high-risk patients, thus limiting the extrapolation of the present fi ndings to the general population. Nevertheless, the major findings of the study are fully consistent with those obtained in the EPIC-Norfolk general population ( 12 ), events in the general population ( 11,12 ) and that low plasma LCAT levels are associated with decreased vascular events in women ( 12 ).
The present results are fully consistent with those of our recent study of individuals with genetic LCAT defi ciency ( 21 ), where carotid IMT was measured in 40 carriers of LCAT gene mutations (12 carriers of two mutant LCAT alleles and 28 heterozygotes) from Italian LCAT-defi cient families ( 22 ) and then compared with 80 matched healthy controls. Carriers of two mutant LCAT alleles had extremely reduced HDL-C levels and heterozygotes had lower HDL-C levels than matched controls. However, the mean and maximum IMT values in carriers were 0.07 mm and 0.21 mm smaller than in controls; moreover, the inheritance of a mutated LCAT genotype had a remarkable gene-dose-dependent effect in reducing carotid IMT ( 21 ).
The observation that reduced plasma LCAT concentrations leading to reduced HDL-C levels do not result in increased atherosclerosis is quite surprising because of the critical role that LCAT has been believed to play in RCT. LCAT has long been believed to be necessary for macrophage cholesterol effl ux and RCT by maintaining  indicating that the relationship between LCAT and atherosclerosis persists over a wide range of cardiovascular risk. Such a conclusion is strengthened by inclusion in the present study of individuals from all over Europe, thus minimizing the impact of genetic and environmental factors.

CONCLUSION
The present study, together with the recent analysis of the EPIC-Norfolk population ( 12 ) and in agreement with our study of Italian LCAT-defi cient subjects ( 21 ), supports the concept that a low plasma LCAT concentration and/ or activity is not associated with increased preclinical atherosclerosis and cardiovascular risk. This is true despite the low plasma HDL-C levels of subjects with low LCAT concentration, as shown here and in carriers of genetic LCAT defi ciency ( 21 ). Furthermore, gender-specifi c analysis in the present study and of the EPIC-Norfolk populations ( 12 ) indicate that, in women, low plasma LCAT concentrations are associated with decreased preclinical atherosclerosis and cardiovascular risk. Altogether, these fi ndings challenge the long-held belief that LCAT is atheroprotective, and they support the concept that LCAT is not required for an effi cient RCT ( 3,4 ) and that a low plasma LCAT concentration and activity is not associated with increased atherosclerosis ( 21 ).