Author + information
- Received January 31, 2006
- Revision received December 20, 2006
- Accepted February 9, 2007
- Published online June 5, 2007.
- Cesaria Prontera, PhD⁎,
- Nicola Martelli, Tech†,
- Virgilio Evangelista, MD†,
- Etrusca D’Urbano, Tech‡,
- Stefano Manarini, Tech†,
- Antonio Recchiuti, PhD⁎,
- Alfredo Dragani, MD§,
- Cecilia Passeri, MD§,
- Giovanni Davì, MD⁎ and
- Mario Romano, MD†,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Mario Romano, Dipartimento di Scienze Biomediche, Ce.S.I., Universitá G. D’Annunzio, 66013 Chieti, Italy.
Objectives This study evaluated the impact of hyperhomocysteinemia (HHcy) on the CD40/CD40 ligand (CD40L) dyad in vivo and in vitro.
Background Hyperhomocysteinemia is associated with an increased incidence of atherothrombosis, although the molecular mechanisms of this association are incompletely defined. The CD40L pair triggers inflammatory signals in cells of the vascular wall, representing a major pathogenetic pathway of atherosclerosis.
Methods We used a commercially available enzyme-linked immunosorbent assay kit to evaluate circulating levels of soluble (s) CD40L in 24 patients with HHcy and 24 healthy subjects. We also used real-time polymerase chain reaction and flow cytometry to determine expression levels of CD40 and vascular cell adhesion molecule (VCAM)-1 in human umbilical vein endothelial cells (HUVECs) and of CD40L in human platelets.
Results The sCD40L levels were significantly increased in HHcy patients (median [interquartile range] 8.0 [0.7 to 10.5] ng/ml vs. 2.1 [1.9 to 2.3] ng/ml, p = 0.0001). Positive correlations were noted between log sCD40L and log homocysteine (Hcy) (R = 0.68, p < 0.0001) or log sVCAM-1 (R = 0.41, p < 0.005). Homocysteine significantly stimulated CD40 mRNA expression in HUVECs (p = 0.033). Consistently, 24-h exposure to Hcy increased the percentage of CD40-expressing cells (p = 0.00025). Homocysteine also significantly enhanced CD40L expression in platelets (p = 0.025) to a comparable extent as that of thrombin. Notably, Hcy increased VCAM-1 protein expression induced by CD40L in HUVECs (p = 0.0046).
Conclusions The present results uncover a potential molecular target of Hcy, namely the CD40/CD40L dyad. Collectively, they indicate that upregulation of CD40/CD40L signaling may represent a link between HHcy and an increased risk of cardiovascular disease.
Elevated homocysteine (Hcy) concentration, determined by genetic or dietary factors, is recognized as an independent risk factor for cardiovascular disease (1). Homocysteine may promote vascular damage and atherothrombosis by a number of mechanisms, including release of proinflammatory mediators, induction of oxidative and endoplasmic reticulum stress, and activation of apoptotic pathways in vascular cells (2). However, the relative contribution of these processes to the causal relationship between hyperhomocysteinemia (HHcy) and atherothrombosis is still under debate.
Accumulating evidence supports the involvement of CD40/CD40 ligand (CD40L) signaling in atherosclerosis. Both CD40 and CD40L are expressed by vascular cells, macrophages, and platelets (3,4). The CD40/CD40L engagement on the surface of endothelial cells, smooth muscle cells, or macrophages triggers a potent inflammatory response, characterized by the release of inflammatory cytokines (interleukins 1β, 6, 8, 12) and chemokines (monocyte chemoattractant protein-1), expression of adhesion molecules (E-selectin, vascular cell adhesion molecule [VCAM]-1, intercellular adhesion molecule-1, P-selectin), activation of matrix metalloproteinases, and procoagulant tissue factor (3–7). Antibody blockade or genetic disruption of CD40L in Apolipoprotein-E−/−mice provides direct evidence of the involvement of CD40/CD40L signaling in atherosclerosis progression (8,9).
A soluble form of CD40L (sCD40L) is rapidly released by T cells and activated platelets (10,11). The sCD40L levels are increased in a number of pathological conditions characterized by cardiovascular damage, i.e., unstable angina (11), acute coronary syndromes (12), hypercholesterolemia (13,14), arterial hypertension (15), and diabetes (16). Thus, measurement of sCD40L is now regarded as an index of platelet activation and inflammatory vascular damage.
Because HHcy is associated with signs of vascular inflammation and platelet activation (17,18), we tested the hypothesis that Hcy may up-regulate the CD40/CD40L system. Here we provide the first evidence of increased sCD40L circulating levels in HHcy patients and of CD40 up-regulation by Hcy in human umbilical vein endothelial cells (HUVECs) and of CD40L in platelets.
Rabbit anti-CD40 and anti-CD40L polyclonal immunoglobulin (Ig) G were from Santa Cruz Biotechnology (Santa Cruz, California). Mouse anti-VCAM-1 (CD106) phycoerythrin-conjugated was from BioLegend (San Diego, California). Anti-rabbit fluorescein-conjugate IgG was from Calbiochem (Milan, Italy). The spin/vacuum total RNA isolation system was from Promega (Milan, Italy). Reagents for real-time analysis were from Applied Biosystems (Milan, Italy). Human thrombin (2,000 NIH U/mg protein); N-2 hydroxyethyl piperazine-N 1-2-ethanesulfonic acid (HEPES); ethylene glycol-bis (b-aminoethyl ether)-N, N, N′, N′,-tetraacetic acid (EGTA); prostaglandin E1(PGE1); DL-homocysteine; L-cysteine; and all other chemicals were purchased from Sigma-Aldrich (Milan, Italy). Thrombin was dissolved in saline at 20 μmol/l (50 U/ml) and stored at −20°C until use.
Patients, genotyping, and measurements
We selected 24 subjects carrying the 5,10 methylene tetrahydrofolate reductase (MTHFR) C677T genotype with HHcy (>15 μmol/l) and 24 age- and gender-matched subjects, also expressing the MTHFR C677T polymorphism, but with normal homocysteinemia (<15 μmol/l). Exclusion criteria were represented by a recent history of thrombotic events (<6 months), pregnancy or delivery in the previous 6 months, hypercholesterolemia, diabetes, current medication for birth control or hormone replacement therapy, and recent use of aspirin, ticlopidine, clopidogrel, anti-inflammatory drugs, vitamin supplements, or anticoagulant agents. Among the 48 patients, 21 (44%) had clinical evidence of vascular disease, in particular, angina pectoris (n = 4), myocardial infarction (n = 3), transient ischemic attack (n = 2), stroke (n = 1), and peripheral artery disease (n = 2). Nine patients had suffered deep venous thrombosis or pulmonary embolism. Patient characteristics are summarized in Table 1.
Subjects were studied as outpatients after a 12-h fast. Blood samples were obtained in the morning. Informed consent was obtained from each subject after approval of the protocol by the local institutional ethics committee.
Analysis of the MTHFR C677T mutation was carried out by digestion with the restriction enzyme HinfI. The section of the gene containing the mutation was amplified by polymerase chain reaction as described previously (19).
Fasting plasma total Hcy (the sum of free and protein-bound forms plus cysteine-homocysteine mixed disulfide) was measured in 51 ethylenediaminetetraacetic acid (EDTA)-anticoagulated blood samples immediately refrigerated to prevent in vitro total Hcy formation. Plasma was stored at −80°C. The total Hcy was measured using the Imx Hcy assay (Abbott Park, Illinois). Plasma sCD40L and sVCAM-1 levels were measured using specific immunoassays (R & D Systems) according to the instructions of the manufacturer. Intra-assay and inter-assay coefficients of variation were <6%.
Human umbilical vein endothelial cells were isolated from umbilical cords obtained from randomly selected healthy mothers delivering at the Chieti University Hospital, using 0.1% collagenase at 37°C. Cells were grown on 1.5% gelatin-coated plates in medium Dulbecco’s modified Eagle’s medium (D-MEM)/M-199 (50:50) supplemented with 20% heat-inactivated fetal calf serum, 10 μg/ml heparin, 50 μg/ml endothelial cell growth factor (ECGF), 50 mg/ml penicillin/streptomycin in 5% CO2at 37°C, and used within 4 passages. Before treatment, HUVECs were made quiescent with D-MEM/M-199 (50:50) medium supplemented with 1% bovine serum albumin (BSA) and 50 μg/ml ECGF for 20 h. Homocysteine was delivered to cells in D-MEM/F12 (50:50) with 2% fetal calf serum and ECGF.
For platelet isolation, blood was collected from healthy volunteers who had not received any medication for at least 2 weeks. Platelet-rich plasma was prepared by centrifugation at 200gfor 15 min. Platelets were isolated by centrifugation at 1,100gfor 15 min, after addition to platelet-rich plasma of 1 μmol/l PGE1. The pellet was suspended in HEPES-tyrode containing 1 μmol/l PGE1and 5 mmol/l EGTA and centrifuged at 1,100gfor 10 min. Platelets were suspended with HEPES-tyrode buffer containing 1 mmol/l Ca2+at the concentration of 1 × 108/ml.
Real-time polymerase chain reaction
Total cellular RNA was extracted using the SV total RNA Isolation System (Promega). Polyadenosine RNA was reverse-transcribed for 60 min at 42°C with StrataScript II (50 U/ml) (Stratagene, Milan, Italy). Real-time measurements of CD40 were carried out using the Assay-on-Demand Hs00374176 from Applied Biosystems, in the ABI PRISM 7900 HT apparatus, according to the instructions of the manufacturer. Glyceraldehyde-3-phosphate dehydrogenase was used as the housekeeping gene. Data were elaborated with the SDS 2.0 software (Applied Biosystems, Foster City, California). Relative gene expression was evaluated using the 2−ΔΔCtformula.
Flow cytometric analysis
The HUVECs were harvested with phosphate-buffered saline (PBS)/EDTA, washed with PBS/0.5% BSA, and incubated with a polyclonal antibody against human CD40 (1:25 dilution, 1 μg for 5 × 105cells) for 30 min at 4°C. Cells were then exposed to a secondary fluorescein-conjugate anti-rabbit antibody for 30 min at 4°C. Samples were washed with PBS/0.5% BSA and analyzed in a Becton Dickinson FACScan flow cytometer (Becton Dickinson, Milan, Italy). Data were analyzed using the CellQuest software. Washed platelets (1 × 108/ml) were exposed to stimuli for 5 min at 37°C, and then incubated with anti-CD40L polyclonal IgG (1 μg for 5 × 106cells) or control rabbit polyclonal IgG for 30 min at 4°C. Samples were then exposed to a fluorescein-conjugate goat anti-rabbit IgG antibody (4°C, 30 min), fixed with 2% paraformaldehyde for 1 h at room temperature and analyzed by FACScan. The CD40L specific mean fluorescence intensity (MFI) was calculated by subtracting the value of the isotype-matched control antibody from that of the specific antibody. For evaluation of VCAM-1 expression, HUVECs were grown in EGM-2MV medium (Cambrex Bio Science, Walkersville, Maryland) supplemented with growth factors. At subconfluence, cells were washed twice with PBS and incubated with vehicle or Hcy for 24 h. Cells were then exposed to sCD40L (Bender MedSystem, Burlingame, California) for 6 h. The HUVECs were harvested and washed as above, suspended in buffer containing PBS + BSA (0.5%), and incubated with anti-human VCAM-1 (CD106) conjugated with phycoerythrin (20 μl/106cells) or isotype-matched IgG1κ-phycoerythrin (Sigma Aldrich, Milan, Italy) in the dark for 30 min at 4°C. Cells were analyzed in a Becton Dickinson FacsCalibur flow cytometer. A total of 10,000 events were acquired. The VCAM-1–specific MFI was calculated using the CellQest analysis software by subtracting the value of the isotype-matched antibody from that of the specific antibody.
Western blot analysis
Proteins were separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS/PAGE) under reducing conditions and blotted onto nitrocellulose membranes. These were exposed, in 10% defatted dry milk, tris-buffered saline 1×, 0.1% tween-20, to a rabbit polyclonal anti-CD40 IgG (1:200 dilution for 3 h), followed by a secondary anti-rabbit IgG peroxidase-conjugate antibody (1:5,000 dilution for 1 h). Immunoblots were visualized by the enhanced chemoluminescence system, and CD40 protein levels were normalized with β-actin.
Differences between individuals with plasma Hcy >15 μmol/l and individuals with Hcy <15 μmol/l were evaluated using the 2-sample ttest, Mann-Whitney Utest, and chi-square test, as appropriate. In view of the skewed distribution of homocysteine (skewness = 2.9; after log transformation skewness = 0.6), sVCAM-1 (skewness = 1.3; after log transformation skewness = 0.7) and sCD40L (skewness = 1.2; after log transformation skewness = −0.6), these variables were log-transformed for correlation analysis and for subsequent multiple-linear regression analysis. For correlation analysis, the Pearson correlation coefficient was calculated (nonparametrically distributed variables examined after log transformation). Data are presented as mean (±SD) or as median and interquartile range (25th, 75th percentile). Only values of p < 0.05 were regarded as statistically significant. All tests were 2-tailed, and analyses were performed using a computer software package (Statistica 6, StatSoft Inc., Tulsa, Oklahoma, or Statistical Package for the Social Sciences, version 13, SPSS Inc., Chicago, Illinois).
In vitro results are reported as mean ± SD. Statistical analyses were performed using the Student ttest. Values of p < 0.05 were considered statistically significant.
HHcy is associated with increased levels of sCD40L and sVCAM-1
To determine whether the CD40/CD40L system is regulated by Hcy levels, we measured sCD40L in 2 groups of subjects carrying the MTHFR C677T mutation, but with different plasma Hcy levels. As shown in Table 1, the incidence of cardiovascular risk factors was comparable in the 2 groups, thus excluding potentially confounding factors. Individuals with plasma Hcy >15 μmol/l had higher sCD40L concentrations compared with subjects with Hcy <15 μmol/l (median [interquartile range]: 8.0 [6.7 to 10.5] vs. 2.1 [1.9 to 2.3] ng/ml, p = 0.0001) (Fig. 1A).The HHcy patients also showed increased levels of sVCAM-1 (644 [567 to 767] vs. 519 [477 to 625] ng/ml, p < 0.02) (Fig. 1B).
Consistently, plasma log Hcy levels positively correlated with both log sCD40L (R = 0.68, R2= 0.45, p < 0.0001) (Fig. 2A)and log sVCAM-1 (R = 0.30, R2= 0.09, p < 0.04) (results not shown). A weak correlation was also noted between log sCD40L and log sVCAM-1 (R = 0.41, R2= 0.16, p < 0.005) (Fig. 2B).
To further evaluate the relationship between plasma sCD40L levels and laboratory and clinical variables, we carried out a multiple linear regression analysis. This analysis confirmed that log Hcy and log sVCAM-1 were the major determinants of log sCD40L, independently from the other risk factors (Table 2).
Hcy regulates CD40 expression in HUVECs
Next, we analyzed the effect in vitro of Hcy on CD40 expression in HUVECs. The Hcy, at concentrations compatible with levels reached in HHcy patients, determined a time- and concentration-dependent increase in CD40 protein expression. At these concentrations of Hcy, HUVECs viability exceeded 95%. The CD40 expression was maximally stimulated by concentrations of Hcy between 100 and 500 μM (Fig. 3A).As reported in Figure 3C, flow cytometric analysis showed that after 12-h exposure to 50 or 500 μM Hcy, the percentage of HUVECs expressing detectable levels of CD40 increased from 8.9 ± 0.4 to 12.4 ± 1.8 and 19.3 ± 8.1 (p = 0.017 and p = 0.045 vs. controls, respectively). Stimulation was maintained up to 24-h incubation. At this time, the percentage of CD40-expressing cells increased from 14.7 ± 0.06 in controls to 20.3 ± 1.2 or 29.3 ± 2.1 in cells treated with 50 or 500 μM Hcy, respectively (p = 0.02 or p = 0.00025 vs. controls). Similar results were obtained when we monitored CD40 protein expression by western blotting (Fig. 3D). Homocysteine also increased CD40 mRNA levels. Real-time polymerase chain reaction measurements showed a peak increment of ∼183% (1.7 ± 0.69 vs. 0.6 ± 0.29 relative abundance, p = 0.033) after 6-h incubation (Fig. 4).Thus, Hcy significantly up-regulates CD40 expression in HUVECs.
Hcy potentiates CD40L-induced up-regulation of VCAM-1 in endothelial cells
To determine whether up-regulation of CD40 expression in endothelial cells was associated with changes in CD40L activity, we exposed HUVECs to Hcy (100 μM, 24 h) followed by sCD40L (500 ng/ml, 6 h) and evaluated VCAM-1 protein expression by flow cytometry. Consistent with previous results with human aortic endothelial cells (HAECs) (20), Hcy increased basal VCAM-1 expression levels from 5.1 ± 2.5 MFI to 11.3 ± 2.6 MFI. The sCD40L greatly enhanced VCAM-1 expression (58.6 ± 11.8 MFI). When cells were pretreated with Hcy, a significant increase in sCD40L-induced VCAM-1 expression was observed (from 58.6 ± 11.8 MFI to 78.4 ± 11.0 MFI) (Fig. 5).We obtained similar results with HAECs (data not shown). These findings indicate that Hcy can potentiate CD40L-induced responses in endothelial cells.
CD40 expression is also induced by oxidant stress and cysteine
Oxidant species are generated during the auto-oxidation of Hcy to homocysteine (21). On the other hand, trans-sulfuration of Hcy generates cysteine, which is also considered a cardiovascular risk factor (22). Therefore, we asked whether oxidative stress and/or cysteine may be involved in the mechanism of CD40 up-regulation by Hcy. For this purpose, we exposed HUVECs to hydrogen peroxide (H2O2) or L-cysteine. We found that both H2O2and L-cysteine significantly up-regulated CD40 expression in HUVECs. In particular, H2O2(300 μM) increased the percentage of CD40-expressing cells from 27.6 ± 7.7 to 43.6 ± 7.5, p = 0.031, whereas L-cysteine (500 μM) increased this percentage from 8.2 ± 1.3 to 22.9 ± 12.2, p = 0.05) (Fig. 6).We also observed a modest, not statistically significant increase in CD40 expression when HUVECs were exposed to methionine (results not shown).
Hcy enhances CD40L expression in human platelets
Because sCD40L originates mainly from circulating platelets, and Hcy levels are directly correlated with sCD40L levels in patients carrying the MTHFR C677T mutation (Fig. 2), we reasoned that Hcy may regulate CD40L expression in platelets. To verify this hypothesis, we exposed washed human platelets to an increasing concentration of Hcy for varying times. Homocysteine up-regulated platelet CD40L expression, with a maximum when platelets were incubated with 500 μM Hcy for 5 min (14.1 ± 4.8 MFI to 31.4± 6.2 MFI (p = 0.025) (Fig. 7).Remarkably, thrombin, which is considered a potent inducer of platelet CD40L, stimulated CD40L expression to a similar extent as Hcy (33.3 ± 5.0 MFI, p = 0.044) (Fig. 7), indicating that results with Hcy may be pathophysiologically relevant.
In the present report, we present the first evidence of a significant increase in sCD40L levels in individuals with HHcy expressing the MTHFR C677T genotype (Fig. 1A). In addition, we show that Hcy, at concentrations that may be reached in patients with HHcy, up-regulates CD40 expression in HUVECs and CD40L in platelets (Figs. 3, 4, and 7). These results unravel a novel potential pathogenetic mechanism of HHcy-associated atherothrombosis, namely the up-regulation of CD40 signaling in vascular cells.
The involvement of the CD40/CD40L dyad in vascular inflammation and atherogenesis is now widely recognized (23). Increased sCD40L levels have been found in patients with a variety of cardiovascular disorders (11–16). In these conditions, activated platelets may represent a major source of sCD40L. It has been documented that sCD40L levels are reduced by antiplatelet drugs (24), and positively correlate with 11d-TXB2, an index of in vivo platelet activation (25). This also may be the case with HHcy subjects. We have previously documented in vivo platelet activation in HHcy (18) and report here that Hcy stimulated CD40L expression in isolated human platelets to a similar extent as thrombin (Fig. 7). Thus, CD40L up-regulation by Hcy may be a consequence of platelet activation. This is consistent with early observations showing a very low level of CD40L expression in resting platelets (4). On the other hand, increased CD40 expression in vascular cells of atherosclerotic lesions had been clearly documented (8). Engagement of endothelial CD40 by platelet CD40L triggers a number of proinflammatory responses, including release of inflammatory cytokines and expression of adhesion molecules, such as VCAM-1, which promotes recruitment of leukocytes and progression of atherosclerosis (4). Thus, Hcy by increasing circulating sCD40L levels (Fig. 1) and expression of CD40 in endothelial cells (Figs. 3 and 4) and of CD40L in platelets (Fig. 6) may promote CD40 signaling. Indeed, we observed that exposure of HUVECs to Hcy significantly potentiated VCAM-1 expression induced by sCD40L (Fig. 5). This seems to be consistent with the observation that subjects with Hcy concentrations above 15 μmol/l displayed elevated sVCAM-1 (Fig. 1). Although Hcy can directly up-regulate VCAM-1 expression in endothelial cells (20) (Fig. 5), the contribution of a CD40-related circuit may be relevant in vivo. Results in Figure 2, showing a significant positive correlation between sCD40L and sVCAM-1 levels, support this concept.
Homocysteine may alter endothelial cell functions by multiple and still incompletely elucidated mechanisms. We have recently reported that severe HHcy is associated with increased peroxidation of arachidonic acid in vivo (18). Others have shown that exposure to Hcy triggers generation of reactive oxygen species (26,27) and that Hcy impairs the antioxidant potential of endothelial cells (28). Thus, some Hcy bioactions may be related to the induction of oxidative damage. Our present results, showing that hydrogen peroxide significantly increases CD40 expression in HUVECs (Fig. 5), suggest that oxidative events may be involved in Hcy-induced CD40 up-regulation. This is an intriguing aspect. It is, in fact, known that CD40 activation by its ligand stimulates production of reactive oxygen species (29). Thus, up-regulation of CD40 by oxidant damage may represent a mechanism of amplification of CD40L-induced proinflammatory responses. However, the molecular events involved in CD40 stimulation by Hcy in HUVECs remain to be fully elucidated. Because CD40 expression in vascular cells is under the control of the transcription factor nuclear factor (NF)-κΒ (30) and Hcy stimulates NF-κΒ activity in HUVECs via generation of superoxide anion (27), it may be reasoned that CD40 up-regulation by Hcy may proceed through NF-κB activation. On the other hand, products generated during Hcy metabolism also may play a role in the pathogenesis of atherothrombosis. In particular, cysteine is considered a cardiovascular risk factor (22), whereas a recent report emphasizes the role of methionine in atherogenesis (31). In our experimental conditions, L-cysteine was as potent as Hcy at inducing CD40 expression (Fig. 5), whereas results with methionine were less consistent, although some increase in CD40 expression could be observed. Thus, it may be hypothesized that at least part of the Hcy effects on CD40 expression may be related to its metabolic conversion to cysteine. On the other hand, cysteine is known to potentiate the capability of Hcy to oxidate low-density lipoprotein (32), suggesting that both Hcy and cysteine may converge on pro-oxidant pathways to up-regulate CD40.
In conclusion, our present results showing a previously unappreciated relationship between Hcy and the CD40/CD40L dyad in vivo and in vitro provide novel evidence of the role that HHcy may play in the pathogenesis of atherothrombosis.
The authors thank Angela Falco for assistance with patients.
Supported in part by a grant from the Italian Ministry of Research to the Center of Excellence on Aging of the University of Chieti (to Drs. Davì and Romano).
- Abbreviations and Acronyms
- bovine serum albumin
- CD40 ligand
- ethylenediaminetetraacetic acid
- N-2 hydroxyethyl piperazine-N 1-2-ethanesulfonic acid
- human umbilical vein endothelial cells
- mean fluorescence intensity
- 5,10 methylene tetrahydrofolate reductase
- nuclear factor
- phosphate-buffered saline
- vascular cell adhesion molecule
- Received January 31, 2006.
- Revision received December 20, 2006.
- Accepted February 9, 2007.
- American College of Cardiology Foundation
- Mach F.,
- Schonbeck U.,
- Sukhova G.K.,
- et al.
- Dechanet J.,
- Grosset C.,
- Taupin J.L.,
- et al.
- Aukrust P.,
- Muller F.,
- Ueland T.,
- et al.
- Garlichs C.D.,
- John S.,
- Schmeisser A.,
- et al.
- Cipollone F.,
- Mezzetti A.,
- Porreca E.,
- et al.
- Damas J.K.,
- Otterdal K.,
- Yndestad A.,
- et al.
- Varo N.,
- Vicent D.,
- Libby P.,
- et al.
- Davi G.,
- DiMinno G.,
- Coppola A.,
- et al.
- Silverman M.D.,
- Tumuluri R.J.,
- Davis M.,
- Lopez G.,
- Rosenbaum J.T.,
- Lelkes P.I.
- El-Khairy L.,
- Ueland P.M.,
- Refsum H.,
- Graham I.M.,
- Vollset S.E.,
- European Concerted Action Project
- Nannizzi-Alaimo L.,
- Alves V.L.,
- Phillips D.R.
- Zeng X.,
- Dai J.,
- Remick D.G.,
- Wang X.
- Au-Yeung K.K.,
- Woo C.W.,
- Sung F.L.,
- Yip J.C.,
- Siow YL O.K.
- Weiss N.,
- Zhang Y.Y.,
- Heydrick S.,
- Bierl C.,
- Loscalzo J.
- Urbich C.,
- Dernbach E.,
- Aicher A.,
- Zeiher A.M.,
- Dimmeler S.
- Wagner A.H.,
- Gebauer M.,
- Pollok-Kopp B.,
- Hecker M.
- Troen A.M.,
- Lutgens E.,
- Smith D.E.,
- Rosenberg I.H.,
- Selhub J.