| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Medicine (A.H., M.J.T.), Helsinki University Central Hospital; and Department of Molecular Medicine (M.J.), National Public Health Institute, Biomedicum, FIN-00290 Helsinki, Finland
Address all correspondence and requests for reprints to: Anna Höckerstedt, M.D., Department of Medicine, Division of Cardiology, Helsinki University Central Hospital, 00290 Helsinki, Finland. E-mail: anna.hockerstedt{at}hus.fi.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To be able to function as potential antioxidants in lipoproteins, estrogens first need to be converted to their lipophilic fatty acyl derivatives, estrogen esters, in a reaction catalyzed by plasma lecithin/cholesterol acyltransferase (LCAT) (27, 28, 29, 30, 31, 32, 33, 34). After esterification, estrogens are able to incorporate in HDL, after which they can be transferred to LDL in a process that is at least partly mediated by cholesterol ester transfer protein (30, 33). The fact that human plasma is endowed with a number of antioxidant defense mechanisms involving many free radical scavengers, reducing agents, and antioxidant enzymes suggests that the oxidation of lipoproteins occurs in subendothelial space rather than plasma (35). In view of this, estrogens incorporated in the lipoprotein particles could provide antioxidant protection in the arterial intima.
The aim of the present study was to provide direct evidence that the estradiol esters generated by the function of LCAT and incorporated in HDL increase the antioxidant potential of this lipoprotein fraction. Experimental conditions were designed using supraphysiological concentrations of estradiol because the aim was to specifically investigate basic mechanisms underlying the role of this principal estrogen in an in vitro model system.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Blood was collected from healthy volunteers (five men and five women). The subjects fasted overnight, after which blood was collected into EDTA-containing tubes and centrifuged at 2300 x g for 10 min at +10 C. Lipoprotein subclasses were immediately isolated from fresh plasma by sequential ultracentrifugation using potassium bromide solution for density adjustments. The study protocol was approved by the ethics committee of the Department of Medicine, Helsinki University Central Hospital.
Isolation of HDL
Each plasma sample was adjusted to a density of 1.063 g/ml and ultracentrifuged at 170,000 x g for 20 h at +10 C in a Beckman LE-80K ultracentrifuge (Beckman, Palo Alto, CA) using Ti 50.4 rotor according to the method of Havel et al. (36). After removal of the top layer of apoliptoeint B-containing lipoproteins, density of the bottom fraction was adjusted to 1.21 g/ml and reultracentrifuged at 270,000 x g for 24 h at +10 C. Total HDL in the top layer was removed and used for experiments immediately. EDTA was present in all density solutions to prevent oxidation during isolation of HDL.
Purification of HDL
EDTA-free HDL was prepared by using size-exclusion chromatography on a Sephadex G25 column in PBS [20 mmol/liter sodium phosphate (pH 7.4) containing 150 mmol/liter NaCl]. The prepared HDL was immediately used for experiments. Protein concentration was determined by the method of Lowry et al. (37).
Purification and measurement of LCAT
LCAT was purified from fresh lipoprotein-deficient plasma (density > 1.21 g/ml) by a combination of phenyl-Sepharose CL-4B, ion exchange (a quaternary methylamine anion exchanger), and hydroxyl-apatite chromatographies as described previously (31, 38, 39), and LCAT activity was analyzed by a radiometric assay using a proteoliposome substrate (39). LCAT activity in purified preparations varied between 60 and 440 nmol cholesterol esterified per hour per milliliter, whereas no cholesterol ester transfer protein, phospholipid transfer protein, lipoprotein lipase, or hepatic lipase activities were detected. LCAT activity in isolated HDL was measured before and after addition of exogenous LCAT. The added LCAT activity was the same within each experiment.
Incubations
Estradiol-17ß [concentration of estradiol (E2) before incubation step was 300 µmol/liter in the final experimental solution] was added to HDL (1 or 2 mg total protein) in total sample volume of 1.5–3 ml, and the mixture was incubated at +37 C for 24 h in the absence or presence of exogenous purified LCAT [activity 16.9 nmol/h·ml ± 1.2 (SEM), range 1.6–44.7 nmol/h·ml in all incubations] as well as in the absence or presence of the LCAT inhibitor, dithionitrobenzoic acid (DTNB) (final concentration 3 mmol/liter; Sigma, St. Louis, MO). After incubation, HDL was isolated by size-exclusion chromatography on a Sephadex G-25 column as above to remove small molecular weight substances not associated with HDL. HDL-associated estradiol and its fatty acyl derivatives in the samples were measured in a single experiment by the recently published method of Vihma et al. (40). Simultaneous experiments were made with labeled 17ß-estradiol [2,4,6,7-3H(N) (17ß-E2)] (specific activity of 72 Ci/mmol; NEN Life Science Products, Boston, MA) in 0.5 mol/liter HEPES buffer (pH 7.4), added to HDL to give a total radioactivity of 2x 105 to 2x 106 dpm. Radioactivity in the eluted fractions was determined by liquid scintillation counting (Rack-ß; Wallac, Turku, Finland).
Lipoprotein oxidation
Oxidation of HDL lipids generates conjugated dienes that absorb light at 234 nm (41). The kinetics of HDL oxidation was monitored by measuring the formation of conjugated dienes according to the method of Esterbauer et al. (41). The protein concentrations in the samples were adjusted to 100 µg/ml protein by dilution with PBS. Copper-initiated oxidation was promoted in the samples by adding freshly prepared copper sulfate to a final concentration of 10 µmol/liter. Oxidation took place at +37 C, and changes in A234 were monitored for 600 min at 3-min intervals in the spectrophotometer HT-Soft (PerkinElmer, Boston, MA).
Statistical analysis
A two-tailed, unpaired Student’s t test was used to determine the statistical significance of differences in lag times, and P < 0.05 was considered significant. Data are expressed as the mean ± SEM. The relation between LCAT activity and lag time was calculated using linear regression analysis (Pearson’s correlation coefficient). All statistical analyses were carried out with SPSS (version 9.0 program for Windows; SPSS Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Figure 1
represents the copper-induced oxidation of HDL as a function of time. Consistent with a number of previous reports, the curves display three characteristic consecutive time periods: lag phase, propagation phase, and decomposition phase, respectively. The period of inhibited oxidation, termed the lag time, is partially due to the radical scavenging reactions mediated by the endogenous antioxidants in HDL particles. Increase in the antioxidative capacity of HDL causes prolongation of lag time and a shift to the right of the oxidation curves.
|
). For comparison, control incubations were performed in the presence of E2 and catalytically inactive LCAT (n = 2), which showed no prolongation of lag times, compared with incubation with E2 alone (data not shown). Also, when the incubations were made in the presence of DTNB, no prolongation of lag time was detected when compared with incubations of native HDL (Table 1). Although a slight prolongation was observed in the mean of the lag times between native HDL isolated from women, compared with that of men (28 vs. 22 min, respectively), the difference did not reach statistical significance.
|
shows the mean lag times after incubation of: 1) HDL alone; 2) HDL in the presence of exogenous E2; 3) HDL in the presence of exogenous LCAT; 4) HDL in the presence of both exogenous E2 and LCAT; and 5) HDL in the presence of exogenous E2, LCAT, and the LCAT inhibitor DTNB. There was a significant increase in the lag time of HDL when both E2 and LCAT were present in the incubation (P < 0.00001). Also, incubations in the presence of either E2 alone (P = 0.013) or LCAT alone (P = 0.002) resulted in a prolongation of lag time, compared with native HDL. These data emphasize the antioxidant role of esterified E2 associated with HDL but also demonstrate some antioxidant activity for LCAT alone.
Before addition of exogenous LCAT, endogenous LCAT activity in isolated HDL ranged between 1.6 and 21.2 nmol/h·ml (mean ± SEM: 10.5 ± 0.9 nmol/h·ml). The amount of added exogenous LCAT activity varied between 8.7 and 26.9 nmol/h·ml (mean ± SEM: 14.6 ± 0.9 nmol/h·ml), resulting in a final activity of 16.9 ± 1.2 nmol/h·ml (range 1.6–44.7 nmol/h·ml) during all incubations. Figure 2 demonstrates the correlation between LCAT activity in the incubation mixture and the lag time of HDL oxidation. When HDL was incubated with exogenous LCAT but without addition of E2, no significant correlation between the lag time and LCAT activity could be observed (Fig. 2A: r = 0.341; P = ns, Pearson’s correlation analysis). However, when E2 also was added to the incubation, the lag time became even more prolonged, and there was a significant correlation between LCAT activity and lag time (Fig. 2B, r = 0.608; P < 0.01). These results provide further evidence that HDL-associated catalytically active LCAT provides antioxidant protection by E2 esterification in our in vitro system.
|
Because our data suggested an antioxidative function for estrogen fatty acyl esters in HDL particles, we next quantitated free (nonesterified) and esterified E2 in the HDL fraction by a recently developed method (40) in one of our experiments. These measurements were repeated three times. The addition of E2 and exogenous LCAT (final activity 22 nmol/h·ml) into the HDL incubation mixture induced the formation of 380 pmol E2 esters per 1 mg HDL protein. Addition of DTNB to this incubation mixture caused a significant reduction in both LCAT activity by 80% and in the concentration of E2 esters by 99%. When no exogenous E2 was added to the HDL incubation mixture, the E2 ester concentrations were 0.33 pmol per 1 mg HDL protein with added LCAT and 0.17 pmol per 1 mg HDL protein without LCAT. These data are in line with the duration of lag times of HDL oxidation (Table 1), suggesting that the antioxidant capacity of HDL increases along with greater contents of esterified estradiol.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Very little is known about the factors that influence the expression of the antioxidant effects of estrogens in vivo. However, studies by Shwaery et al. (28, 34) suggested an important role for conversion of estrogens to lipophilic derivatives, which occurs in plasma. Fatty acyl esterification of estrogens is considered a prerequisite for their incorporation into lipoproteins (27, 46) and even a key event for their antioxidant actions. There is indeed previous evidence that the reaction is catalyzed by LCAT, a HDL-associated enzyme responsible for the generation of the major pool of circulatory cholesterol esters (31, 32). However, the studies on the antioxidant efficacy of E2 esters (27, 28, 34) have only presumed a role for LCAT in E2 esterification basing it on the following: 1) E2 esterification did not occur in incubations with isolated LDL, but it occurred in incubations with plasma (28, 34) and during coincubation of LDL and HDL (27); and 2) E2 esterification was inhibited by an enzyme inhibitor, DTNB (28, 34), indirectly suggesting inhibition of LCAT. Although these studies have provided valuable information, they have not documented direct proof that E2 esters are formed in HDL in a reaction catalyzed by LCAT and that these E2 esters provide protection of HDL particles against oxidation. In addition, we show for the first time that the presence of E2 in the incubation mixture resulted in a significant positive correlation between LCAT activity and HDL oxidation lag time, suggesting that increase in E2 esterification HDL further improved antioxidant efficacy.
In our experiments the concentrations of E2 esters in HDL after incubation and purification of HDL have been calculated per 1 mg total HDL protein. This amount of total HDL protein corresponds with approximately 0.5 ml human plasma. From this we can estimate that the amount of HDL-associated E2 esters observed in our experiments could be translated to 760 pmol/ml in plasma. Thus, the E2-ester concentrations presented here are more than a 1000-fold excess of physiological E2-ester levels in plasma, which have been measured recently by Vihma et al. (40) in pregnant women 196–750 pmol/liter. To our knowledge the highest concentrations of endogenous E2 esters have been determined in human ovarian follicular fluid (mean 106 pmol/ml, range 57–262 pmol/ml) (48). However, the purpose of the present study was not to investigate physiological plasma levels but instead to study the basic mechanism underlying the formation and function of HDL-associated estradiol fatty acid esters. Esterbauer’s method used in our experiments is very applicable for this kind of first-step analysis, but although regarded as a gold standard, it is relatively insensitive and requires sufficient amounts of antioxidants present necessitating the use of high concentrations of E2. E2, although the principal human estrogen, is only one of many estrogens, which may contribute to the antioxidant protection of HDL and possible difference between female and male HDL.
In our experiments we used exogenous, purified LCAT, but because a significant portion of catalytically active endogenous LCAT remained associated with the HDL particles after ultracentrifugation, a mixture of endogenous and exogenous LCAT participated in the esterification reaction. The amount of endogenous HDL-associated LCAT activity varied considerably, resulting in large variations in total LCAT activity. We observed that functional endogenous LCAT provided antioxidant efficacy by enhancing exogenous E2 esterification to some extent in the absence of added LCAT. Incorporation of estradiol fatty acyl esters in HDL was confirmed by quantitating the free and esterified estradiol in HDL with a recently developed sensitive method (40). In some experiments in which pharmacological E2 amounts were incubated with HDL under conditions of low LCAT activity, free E2 also adhered to the HDL particles. On the other hand, we previously demonstrated that under physiological conditions nonesterified estradiol has low binding to serum lipoproteins (40). In the present study, greater E2 ester contents in HDL were associated with prolonged oxidation lag times. The fact that addition of LCAT alone gave rise to increased antioxidant capacity is in line with results of Vohl et al. (49), who proposed that scavenging free radicals was the underlying mechanism for LCAT-induced antioxidation. Thus, we suggest two different mechanisms explaining the increased resistance to oxidation of HDL, one being the incorporation of E2 esters and the other the antioxidant activity of LCAT itself. We hypothesize that E2 esterification and LCAT activity could have physiological roles together with the antioxidative enzymes carried by HDL, the platelet-activating factor acetylhydrolase and paraoxonase (50, 51). These mechanisms may contribute to blocking the build-up of oxidized lipids in HDL, thus preserving its ability to inhibit LDL oxidation (52, 53). Appropriate in this context is also our previous observation demonstrating that E2 esters formed in HDL are transported to LDL by a mechanism that is at last partly dependent on cholesterol ester transfer protein (30). Thus, E2 esters formed in HDL might be important in protection of both HDL and LDL against oxidation and the activity of both LCAT and cholesterol ester transfer protein might have an important role in the antioxidative capacity of LDL.
A recent study has proposed that HDL binds to the scavenger receptor B class I, which is also expressed in vascular endothelial cells and delivers estrogen to endothelial nitric oxide synthase (eNOS), thereby stimulating the enzyme (54). The authors, however, did not measure esterified E2, and it is not clear how the E2 activated eNOS. Recent observations by Nofer et al. (55) have presented critical aspects stating that native E2 concentrations in HDL are by far too low to explain eNOS activation via scavenger receptor B class I route. However, it is quite possible that certain membrane domains of the endothelium may cluster molecules like E2 esters from native HDL and form high local concentrations, thus being able to facilitate/initiate relevant physiological responses. Interestingly, earlier data on lipoprotein-bound nonestrogenic steroid esters have suggested that they can be taken up into cells via lipoprotein receptors and further hydrolyzed into free steroid forms (56, 57, 58). Nevertheless, it is quite evident that the possible physiological role of HDL receptor-facilitated endothelial entry of HDL associated estrogen esters must be assessed critically in future studies.
In conclusion, we have demonstrated in an in vitro model system unambiguously that LCAT facilitates the formation of hydrophobic estradiol fatty acid esters that become incorporated in HDL particles in which they significantly increase its antioxidant potential. In physiological terms, estrogen esters contained in HDL could enter the arterial subendothelial compartment and could, in theory, modulate several mechanisms crucial in the development of atherosclerosis. Whereas our experimental model suggests the possibility of this type of an in vivo mechanism, further studies using physiological estrogen ester concentrations are required to clarify this issue.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: DTNB, Dithionitrobenzoic acid; E2, estradiol; 17ß-E2, 17ß-estradiol; eNOS, endothelial nitric oxide synthase; HDL, high-density lipoprotein; LCAT, lecithin/cholesterol acyltransferase; LDL, low-density lipoprotein.
Received January 27, 2004.
Accepted July 21, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, n = 9) or absence (
, n = 7) of exogenous LCAT. Similar incubations were carried out without addition of 17ß-E2 in the presence of LCAT (
, n = 9) and with native HDL (—, n = 10). Endogenous LCAT activity in isolated HDL was 10.5 ± 0.9 nmol/h·ml (range 1.6–21.2 nmol/h·ml). Added LCAT activity was 14.6 ± 0.9 nmol/h·ml (range 8.7–26.9 nmol/h·ml), resulting in a final total activity of 22 ± 2.3 nmol/h·ml (range 11.3–44.7 nmol/h·ml) in incubations containing exogenous LCAT. After incubation, HDL was isolated by size-exclusion chromatography on a Sephadex G-25 column and after adjusting protein concentration to 100 µg/ml, oxidation was started with 10 µM CuSO4 (final concentration per incubation), and diene formation was monitored with a PerkinElmer HT-Soft spectrophotometer at OD 234 nm. The data are expressed as mean ± SEM.