Polymorphonuclear Leukocytes May Impair Endothelial Function
Results of Crossover Randomized Study of Lipid-Lowering Therapies
Objectives— To examine whether polymorphonuclear leukocytes (PMNs) in hypercholesterolemia (HC) are activated to generate large amount of superoxide in vivo and hence impair endothelial function and, if so, whether statins, which possess anti-inflammatory properties, may restore PMN-mediated endothelial dysfunction.
Methods and Results— At baseline, subjects with HC showed impaired endothelial function (P<0.001), estimated by flow-mediated vasodilation of the brachial artery, and increased susceptibility of low-density lipoprotein (LDL) to oxidation (P<0.0001) compared with control subjects. PMNs obtained from HC produced greater amount of superoxide (P<0.0001), showed higher adhesiveness to cultured endothelial cells (HUVECs) (P<0.0001), and impaired endothelial nitric oxide synthase (eNOS) Ser1177 phosphorylation of HUVECs compared with controls (P<0.001). Crossover administration of fluvastatin or colestimide for 3 months lowered LDL to the same levels (P<0.001 for both). Endothelial function was restored (P<0.0001). LDL oxidation (P<0.0001) and superoxide release from PMNs (P<0.0001) were diminished only in fluvastatin but not in colestimide arm. Fluvastatin attenuated PMN adhesion to HUVECs (P<0.0001) and restored eNOS Ser1177 phosphorylation of HUVECs (P<0.001).
Conclusion— Statins may improve endothelial function at least in part by inactivating neutrophils independently of LDL reduction. Our results raise a novel concept that polymorphonuclear leukocytes may attack endothelia and play a pivotal role in the pathogenesis of atherosclerosis.
Several lines of evidence indicate that inflammation plays a pivotal role in the process of atherosclerosis.1 Although local inflammation of vulnerable plaque is responsible for the development of acute coronary syndrome,2,3 systemic inflammatory process as estimated by highly sensitive C-reactive protein (hsCRP) is closely associated with later cardiovascular events in subjects with hypercholesterolemia.4 Although precise mechanisms that link systemic inflammation to vascular diseases remain to be elucidated, reactive oxygen species (ROS) play a pivotal role in inflammation-mediated vascular injury.
Polymorphonuclear leukocytes (PMNs) are ubiquitous effector cells in numerous inflammatory conditions.5 Relevance of PMN function is usually thought in terms of host defense in bacterial infections.6 PMNs produce a large amount of ROS through the respiratory burst and cause lipid peroxidation.7 In addition, on stimulation, PMNs release cytokines that promote endothelial cell activation and hence cell-to-cell adhesion.8 Collectively, these findings suggest a possible role of PMN in the development of vascular diseases via impairment of endothelial function. Widespread activation of PMN was demonstrated in subjects with hypercholesterolemia.9,10
Three-hydroxyl-3-methylglutanyl coenzyme A reductase inhibitors (statins) significantly decrease cardiovascular mortality associated with hypercholesterolemia.11 In vitro studies have shown that statins exert beneficial effects on vascular diseases by inhibiting leukocyte rolling and adhesion12 and by stabilizing endothelial nitric oxide synthase (eNOS) post-translationally.13 Several clinical trials revealed that statins decrease markers of inflammation/oxidative stress14 and restore endothelium-dependent vasodilation15,16 in the placebo-controlled design. However, whether the restorations of endothelial function by statin are attributable to LDL reduction17 or to its pleiotropic effects remain to be tested by the alternative active treatment arm.
Accordingly, we hypothesized that PMN in hypercholesterolemia are activated, leading to oxidative stress in vivo and hence impair endothelial function, and, if so, statins may restore the PMN-mediated endothelial dysfunction independently of low-density lipoprotein (LDL) reduction. To test these hypotheses, we measured PMN activity, oxidative stress, and endothelial function in subjects with hypercholesterolemia and their age-matched controls, and we conducted a crossover trial using 2 different classes of lipid-lowering agents, fluvastatin and colestimide. Moreover, to investigate the direct interrelation between PMNs and endothelia, cell adhesion and eNOS phosphorylation as a process of eNOS activation in cultured human umbilical endothelial cells (HUVECs) were investigated before and after ex vivo exposure of PMNs obtained from hypercholesterolemic subjects with or without fluvastatin treatment.
Eleven hypercholesterolemic patients with both total serum cholesterol >240 mg/dL and LDL cholesterol >160 mg/dL and 8 age-matched healthy subjects as controls were enrolled (Table). There were no significant differences in age, blood pressure, plasma glucose, or body mass index between hypercholesterolemic subjects and normocholesterolemic controls (Table). All female subjects were postmenopausal. No patients had symptomatic coronary artery diseases or family history of premature cardiovascular diseases. All subjects were nonsmokers, nonobese, nondiabetic, and normotensive. No subjects were using medication or vitamin supplements. The reproducibility of measurement, ie, coefficient of variation (CV) has been <10%, and the expected difference between drug was estimated ≈20%. Based on this statistical background, the number for >90% of statistical power is estimated as 10 to 12 in the crossover design.
The protocol was explained and written informed consent was obtained from each subject. The study was approved by Ethical Committee for Human Study in our institution. The study was performed with subjects in a supine position and in an air-conditioned room at 22 to 23°C. Patients were randomized to have fluvastatin 20 mg or colestimide 3 grams daily in a crossover design. Before and after 12 weeks of each treatment, endothelial function and biochemical and hematologic parameters were evaluated. The design of this study was a crossover trial in which drugs were not blinded, whereas the measurements of endothelial function or assay were completely blinded.
Measurements of Vascular Function
Flow-mediated vasodilation of the brachial artery was measured by a previously described noninvasive technique to assess endothelial function.18,19 Briefly, after a 10-minute equilibration period, with the use of a 10-MHz linear-array transducer and the SSA-380A system (Toshiba), the brachial artery was longitudinally imaged 5 cm proximal to the antecubital crease, twice at baseline, and then from 1 to 15 minutes continuously after the release of 4.5-minute upper arm arterial occlusion at 250 mm Hg of pressure with a 12.5-cm-wide cuff. Photographic images of end-diastolic frames were obtained and analyzed by a single investigator blinded to the subjects and sequences. The arterial diameter was measured by a caliper at the single most equivalently imaged site with side-by-side presentation. Flow-mediated vasodilation, ie, endothelium-dependent vasodilation, was determined as the maximal percent diameter change of the post-occlusion arterial diameter measurement relative to the mean of the corresponding 2 baseline measurements. The inter-observer and intra-observer variations of 2 baseline measurements were 2.8% and 1.6%, respectively, in our laboratory. Blood flow velocity was measured by Doppler at baseline and immediately after the release of cuff occlusion. Arterial blood flow was determined as arterial cross-sectional area times mean Doppler velocity. The magnitude of reactive hyperemia was calculated as the maximum flow divided by the baseline flow. As an internal control, we measured changes in the brachial diameter, ie, endothelium-independent vasodilation, induced by sublingual nitroglycerin (300 mg; Myocol Spray; Toa Eiyo Co) given 15 minutes after the measurement of flow-mediated vasodilation. Five minutes after nitroglycerin administration, the scan was performed to assess endothelium-independent vasodilation.
Activity of PMNs
Activity of PMNs was estimated by superoxide generation during respiratory burst.6 Briefly, PMNs were separated from platelets and mononuclear cells by Ficoll-Hypaque density-gradient centrifugation. Purified PMNs were activated with 2 μg/mL of phorbol myristate acetate for 20 minutes and superoxide dismutase-inhibitable cytochrome c reduction, as an index of maximal superoxide generation by NADPH oxidase, was measured by the calorimetric method as previously described.20 NADPH oxidase is responsible for the rapid generation of superoxide from PMNs, ie, respiratory burst. The activity of this enzyme is not only determined by the subunits such as gp91phox or p22phox but also determined by their assembly. Because it has been considered that “total NADPH oxidase activity” is assessed by phorbol ester-induced superoxide release, we adopted this method.
As a marker of systemic oxidative stress, susceptibility of LDL to oxidation was assessed by LDL lag time as previously described.21 LDL (1.019 <d <1.063 g/mL) was prepared from plasma by sequential ultracentrifugation. EDTA and salt were removed from LDL using prepacked columns filled with Bio-Gel P6 as desalting gel. CuSO4 was added to LDL and the 234-nm absorption was measured for a period of 5 hours.
It is difficult to examine if changes in PMN function are directly linked to endothelial function in vivo because statin would improve both PMN function and endothelium function. To resolve this issue, we examined their link ex vivo: effects of patient-derived neutrophils on HUVECs and on eNOS activation of HUVECs before and after statin. With this approach, we were at least able to examine whether the improved PMN function after statin affects the link between PMN and endothelium.
Binding assay was performed with 3×108 of separated PMNs coincubated with HUVECs in 6-well plates for 60 minutes at 37°C. Nonadherent cells were removed by gentle washing with phosphate-buffered saline 3 times. Adherent cells were counted by microscopy in a blinded manner.
eNOS Ser1177 Phosphorylation
To investigate the interrelation between PMNs and endothelia more directly, we performed additional ex vivo experiments before and after fluvastatin treatment (n=8). As an index of eNOS activation, the phosphorylation of Ser1177 on eNOS was assessed by Western blot analysis22 when HUVECs were coincubated with or without PMNs. HUVECs, which were cultured in the 6-well dishes to be confluent, were coincubated with or without 3×105 of PMNs for 60 minutes in 5% CO2 at 37°C. Then cells were stimulated with the calcium ionophore, A23187, for 3 minutes at concentration of 1 μmol, and they were washed twice with ice-cold phosphate-buffered saline. Total cell lysates were prepared by scraping the cells in lysis buffer (62.5 mmol/L Tris-HCl, 2% SDS, 10% glycerol, 50 nmol DTT, 0.01% BPB). After centrifugation for 5 minutes, samples were loaded onto 7.5% to 15% polyacrylamide gel (Readygels J; Bio-Rad) and electrophorased. Gels were transferred to a nitrocellulose membrane, then the reaction was blocked by incubation in Tris-buffered saline (10 mmol/L Tris, pH 7.5, 100 mmol/L NaCl) containing 0.1% (v/v) Tween 20 and 5% (v/v) nonfat dry milk for 1 hour, followed by 2-hour incubation with anti–phospho-eNOS-Ser1177 (Cell Signaling Technology) or anti-eNOS (Santa Cruz Biotechnology). Bands were visualized with the appropriate anti-rabbit horseradish peroxidase-conjugated secondary antibody using enhanced chemiluminescence substrate (Amersham Pharmacia Biotech) and Kodak X-OMAT film. The relative band densities were quantified using National Institutes of Health Image 1.62. The phosphorylations of eNOS were calculated as the ratio of phosphorylated eNOS to eNOS.
Results were expressed as mean±SEM, and a value of P<0.05 was considered to be statistically significant. Comparisons between normal controls and hypercholesterolemic subjects were analyzed by Student t test. The lipid profiles, vascular function, cell adhesiveness, LDL lag time, and PMN activity, before and after treatment with fluvastatin or colestimide, were analyzed by 2-way ANOVA for repeated measurements and, if appropriate, followed by Schaeffe post hoc analyses.
Baseline Results: Hypercholesterolemia Versus Control
Lipids profiles at baseline, total cholesterol (P<0.001), and LDL cholesterol (P<0.001) were significantly higher in the hypercholesterolemic subjects than in normocholesterolemic controls, whereas high-density lipoprotein cholesterol and triglycerides were similar between the groups (Figure 1).
Both flow-mediated vasodilation (P<0.001) and nitroglycerin-induced vasodilation (P<0.001) were impaired in hypercholesterolemic patients compared with normocholesterolemic controls (Figure 2). The ratio of flow-mediated vasodilation to nitroglycerin-induced vasodilation was lower in hypercholesterolemic patients, indicating that endothelial function was predominantly impaired (P<0.001; Figure 2). The baseline diameter of brachial artery (3.86±0.57 versus 3.80±0.48 mm, not significant) and hyperemic flow response (308±64 versus 305±60%, not significant) were similar between controls and hypercholesterolemic patients
The number of adherent PMNs to HUVECs after 60-minute exposure was significantly greater in hypercholesterolemic subjects (P<0.0001; Figure 3).
Unless stimulation by phorbol myristate acetate, superoxide production of PMN was not detectable. We measured superoxide from PMNs stimulated with phorbol myristate acetate. The production of superoxide from PMNs was significantly greater in hypercholesterolemic subjects (P<0.0001; Figure 4).
Susceptibility of LDL to oxidation assessed by LDL lag time was curtailed in hypercholesterolemic patients (P<0.0001; Figure 4).
Effects of Lipid-Lowering Therapies: A Crossover Trial
Lipids profiles after 12-week treatment of fluvastatin or colestimide and total and LDL cholesterol were significantly decreased to the similar levels (P<0.0001 for both by ANOVA; Figure 1). The reductions of total or LDL cholesterol were not different between fluvastatin and colestimide groups. High-density lipoprotein cholesterol and triglycerides levels were not altered.
After treatment with fluvastatin, flow-mediated vasodilation was restored to the control levels (P<0.0001 by ANOVA; Figure 2). Nitroglycerin-induced vasodilation was also improved in fluvastatin-treated subjects (P<0.05 by ANOVA; Figure 2), although the magnitude was rather smaller than that of flow-mediated vasodilation. Thus, the ratio was greater after fluvastatin treatment, indicating that endothelial function was predominantly improved (P<0.001 by ANOVA; Figure 2). However, colestimide treatment had no effect on vascular functions despite its lipid-lowering effects (Figure 2). The baseline diameter of brachial artery (fluvastatin 3.73±0.42 versus colestimide 3.79±0.55 mm, not significant) and hyperemic responses (fluvastatin 292±74 versus colestimide 290±56%, not significant) were not altered by either treatment.
After treatment with fluvastatin, superoxide production was significantly attenuated compared with pretreatment, whereas colestimide had no effect on superoxide production by PMNs (P<0.001 by ANOVA; Figure 4). Furthermore, there was a significant inverse correlation between superoxide production by PMNs and flow-mediated vasodilation (r=−0.49, P<0.03).
After treatment with fluvastatin, the LDL lag time was significantly prolonged compared with pretreatment, whereas colestimide had no effect on the LDL lag time (P<0.001 by ANOVA, Figure 4) Furthermore, there was a significant positive correlation between the LDL lag-time and flow-mediated vasodilation (r=0.74, P<0.0002).
For ethical reasons, our study had no washout period. By 2-way ANOVA for repeated measurements, we found no order effects on endothelial function and PMN activity. In addition, by post hoc analyses, which compared the effects of the same lipid-lowering agents between the statin–resin group and the resin–statin group, there was no difference on drug effects irrespective of the order.
Ex Vivo Interrelation Between PMN and Endothelia: Cell Adhesiveness
The number of adherent PMNs obtained from fluvastatin-treated subjects to HUVECs was significantly less compared with pretreatment (P<0.0001 by ANOVA; Figure 3). Colestimide treatment did not influence the PMN–HUVEC adhesiveness.
eNOS Ser1177 Phosphorylation
The exposure of PMNs obtained from hypercholesterolemic subjects impaired eNOS phosphorylation of HUVECs (control 0.88±0.08, hypercholesterolemia 0.66±0.05; P<0.001). Fluvastatin restored the phosphorylation of eNOS to the nearly normal levels (P<0.001; Figure 5).
The salient findings of this study are presented herein. In subjects with hypercholesterolemia, endothelial function was impaired and LDL oxidation was increased compared with controls. PMNs obtained from subjects with hypercholesterolemia generated greater amounts of superoxide and augmented cell adhesiveness to cultured endothelial cells compared with controls. Despite administration of fluvastatin or colestimide that lowered LDL to the similar levels, endothelium-dependent vasodilation and PMN–HUVEC adhesiveness were restored only in fluvastatin-treated subjects. Fluvastatin diminished the susceptibility of LDL to oxidation and attenuated superoxide release from PMNs, whereas colestimide had no effects. The exposure of PMNs obtained from hypercholesterolemic subjects on HUVECs impaired Ser1177 phosphorylation of eNOS, whereas chronic fluvastatin treatment normalized eNOS Ser1177 phosphorylation. These findings suggest that activated inflammatory cells in hypercholesterolemia may impair endothelial function through oxidative stress, and fluvastatin may improve endothelial function by its anti-inflammatory properties against inflammatory cells independently of LDL reduction.
Oxidative Stress in Hypercholesterolemia
Oxidative stress estimated by LDL lag time was significantly increased in hypercholesterolemic subjects compared with controls. Although a number of indices of oxidative stress are advocated, LDL lag time, an index of susceptibility of LDL oxidation, has been widely used as a gold standard of the assessment of oxidative stress in humans.21 There are several sources of ROS in vivo such as mitochondrion, xanthine oxidase, catecholamine, or arachidonic acid. Among them, NADPH oxidase in phagocytic leukocytes generates a large amount of superoxide on stimuli, ie, respiratory burst.23 Therefore, we measured superoxide production as an index of NADPH oxidase activity by PMNs as a possible source of oxidative stress in vivo. Likewise, LDL lag time and superoxide productions from PMNs were greater in hypercholesterolemic subjects than in normal controls, suggesting that activated PMNs may, at least in part, play a role in increased oxidative stress.
Pleiotropic Effect Beyond LDL Reduction
Several studies have shown that treatments with statins restore endothelial function15,16 and decrease the magnitude of inflammation or oxidative stress14 in hypercholesterolemic subjects. The paradigm, “pleiotropic effects of statins beyond LDL reduction” has been advocated and accepted extensively.24 However, one has to be cautious to interpret these previous results because most of them were placebo-controlled studies that lacked the active control arm. We previously demonstrated that acute removal of plasma LDL by LDL apheresis normalized endothelial function in subjects with familial hypercholesterolemia.17 Therefore, “pleiotropic effects of statins beyond LDL reduction” in the aforementioned studies may be associated with the LDL reduction itself. In the present crossover study, 3-month treatment with fluvastatin or colestimide lowered plasma LDL levels in a similar manner without affecting other lipid profiles. Nevertheless, only fluvastatin treatment showed significant improvement of endothelial function, attenuation of superoxide production by PMNs, and reduction of lipid peroxidation. These results may indicate that fluvastatin restored the cross-talk between neutrophils and endothelia independently of LDL reduction. It is possible that cholesterol and LDL levels in PMNs are altered in hypercholesterolemic subjects, resulting in activation of those cells, and that fluvastatin (but not colestimide) may also ameliorate PMN functions by ameliorating lipid levels in the cells.
Endothelium–Polymorphonuclear Leukocytes Interaction
In this study, we tried to link the forearm endothelial vasodilator functions measured under physiological conditions with PMN functions measured under nonphysiological conditions. There was a close link between physiological flow-mediated vasodilation and phorbol ester–stimulated respiratory burst of PMNs as an index of PMN activity. At basal conditions, superoxide production estimated by cytochrome c reduction, which is a gold standard for in vitro measurement for ROS release from leukocytes, was not detected, indicating that PMNs were not activated without stimuli, which is consistent with previous reports.20 Therefore, we estimated the potential activity of PMNs during respiratory burst. In ex vivo experiments we counted the adhered PMNs to HUVECs in unstimulated conditions, indicating that PMNs may activate endothelium even in the absence of stimuli, ie, physiological conditions. Improvement of the response to NO was closely related to diminished superoxide release from PMNs. Several mechanisms can be attributed to this interrelationship. Superoxide released from PMNs may directly impair eNOS activity, partially caused by oxidation of tetrahydrobiopterin, an essential cofactor of eNOS, which causes unstabilization of dimeric structure of eNOS, leading to superoxide release by eNOS per se. In addition, endothelium-derived NO can be inactivated by superoxide-forming peroxinitrate, which also impairs biological effects of NO. We have demonstrated the close links among endothelial function, superoxide production by PMNs, and the LDL lag time before and after fluvastatin treatment in vivo. These findings suggest that oxidative stress-mediated aggression of PMN on endothelia may be attenuated by fluvastatin treatment. To investigate the interrelation between PMNs and endothelia more directly, we performed additional ex vivo experiments. The exposure of PMNs obtained from hypercholesterolemic subjects on HUVECs increased the adhesiveness between them and impaired calcium-ionophore–induced Ser1177 phosphorylation of eNOS in HUVECs, a molecular marker of eNOS activation, suggesting that PMNs from hypercholesterolemia directly activate endothelia and impair the cellular signaling for NO production; however, fluvastatin decreased PMN–HUVEC adhesiveness and augmented eNOS Ser1177 phosphorylation. These findings may suggest that fluvastatin treatment hampers the attack of PMN against endothelia in humans. The molecular mechanisms of the anti-PMN activity of fluvastatin are beyond the scope of this study. However, it has been demonstrated that statins inhibit superoxide anion formation produced by NADPH oxidase in neutrophils because of a blockade of isoprenoid synthesis in vitro.25 Therefore, the beneficial effects of statins in the prevention of cardiovascular events could be caused in part by the manipulation of the regulatory system of the mevalonate pathway in leukocytes unrelated to cholesterol biosynthesis.
We demonstrated the close link between endothelial function and PMN activity in the former half of this study, which does not necessarily indicate the direct causal role of neutrophils in endothelial dysfunction or effects of the specific treatment. To resolve this issue, we examined their link ex vivo: effects of patient-derived neutrophils on normal cultured endothelial cells and on their eNOS phosphorylation. With this approach, we demonstrated PMN obtained from hypercholesterolemic subjects: (1) increased adhesiveness, ie, endothelial activation, between cultured endothelial cell and PMNs; and (2) impaired the phosphorylation of eNOS, an index of eNOS activity. These ex vivo results may suggest the attack of PMNs against endothelial cells. The drawback of this study is that we failed to dissect PMN activation from endothelial activation in our patients with mild hypercholesterolemia. Increased PMN adhesion to the endothelium could be caused by endothelial activation with hypercholesterolemia alone or with activation of other blood cells, such as monocytes and platelets. Thus, the increased PMN adhesion to the endothelium does not necessarily indicate PMN activation in this study. To support our conclusion, we should show direct evidence for PMN activation and the positive correlation between endothelial dysfunction and PMN activation under physiological conditions. To investigate the precise role of neutrophils on endothelial dysfunction, further studies, such as leukapheresis on endothelial dysfunction, may be required. In the present study, we did not present the data of hsCRP, because our preliminary results in a small size of samples failed to demonstrate a significant relationship between the cholesterol levels and hsCRP. We estimated PMN activity by measuring superoxide dismutase-inhibitable cytochrome c reduction during phorbol ester stimulation.20 Although the maximal NADPH oxidase activity can be anticipated from this method, unstimulated or receptor-stimulated activity of PMNs needs to be determined by alternative methods.5,6 Hypochlorous acid generated by myeloperoxidase also plays a pivotal role in PMN-mediated tissue injury. Recent studies have shown that plasma myeloperoxidase is elevated in patients with acute coronary syndrome, and its levels are associated with their prognosis.26 The role of myeloperoxidase in PMN-mediated endothelial dysfunction remains to be tested in further studies.
Our results raise a novel concept that inflammatory cells may attack endothelia and play a pivotal role in the pathogenesis of atherosclerosis. Recent large-scale trials have shown that statins improve clinical outcome in acute coronary syndrome,27 in which widespread activation of PMN has been demonstrated.26 Furthermore, PMNs may be associated with postprandial endothelial dysfunction via inflammation/oxidative stress.28 Thus, PMNs may become a novel target as the therapeutic strategy for cardiovascular diseases.
This study was supported in part by grants from Academic Frontier Projects, grant-in-aid for Scientific Research (H.M.) and grant-in-aid for Encouragement of Young Scientists (R.S.) from the Ministry of Education, Science, and Culture, Tokyo, and from Japan Heart Foundation Grant for Research on Hypertension and Vascular Disease, Tokyo.
- Received July 18, 2004.
- Accepted February 15, 2005.
Maseri A, Fuster V. Is there a vulnerable plaque? Circulation. 2003; 107: 2068–2071.
Ridker P. Clinical application of C-reactive protein for cardiovascular disease detection and prevention. Circulation. 2003; 107: 363–369.
Baggiolini M, Boulay F, Badwey J, Curnutte J. Activation of neutrophil leukocytes: chemoattractant receptors and respiratory burst. FASEB J. 1993; 7: 1004–1010.
Cai H, Harrison D. Endothelial dysfunction in cardiovascular diseases: the role of oxidant stress. Circ Res. 2000; 87: 840–844.
Ludwig P, Hunninghake D, Hoidal J. Increased leucocyte oxidative metabolism in hyperlipoproteinaemia. Lancet. 1982; 8294: 348–350.
Yasunari K, Maeda K, Nakamura M, Yoshikawa J. Oxidative stress in leukocytes is a possible link between blood pressure, blood glucose, and C-reacting protein. Hypertension. 2002; 39: 777–780.
Gould A, Rossouw J, Santanello N, Heyse J, Furberg C. Cholesterol reduction yields clinical benefit: impact of statin trials. Circulation. 1998; 97: 946–952.
Scalia R, Gooszen ME, Jones SP, Hoffmeyer M, Rimmer DM 3rd, Trocha SD, Huang PL, Smith MB, Lefer AM, Lefer DJ. Simvastatin exerts both anti-inflammatory and cardioprotective effects in apolipoprotein E-deficient mice. Circulation. 2001; 103: 2598–2603.
Wolfrum S, Jensen K, Liao J. Endothelium-dependent effects of statins. Arterioscler Thromb Vasc Biol. 2003; 23: 729–736.
Egashira K, Hirooka Y, Kai H, Sugimachi M, Suzuki S, Inou T, Takeshita A. Reduction in serum cholesterol with pravastatin improves endothelium-dependent coronary vasomotion in patients with hypercholesterolemia. Circulation. 1994; 89: 2519–2524.
Tamai O, Matsuoka H, Itabe H, Wada Y, Kohno K, Imaizumi T. Single LDL apheresis improves endothelium-dependent vasodilatation in hypercholesterolemic humans. Circulation. 1997; 95: 76–82.
Miyazaki H, Matsuoka H, Itabe H, Usui M, Ueda S, Okuda S, Imaizumi T. Hemodialysis impairs endothelial function via oxidative stress: effects of vitamin E-coated dialyzer. Circulation. 2000; 101: 1002–1006.
Morales-Ruiz M, Lee MJ, Zollner S, Gratton JP, Scotland R, Shiojima I, Walsh K, Hla T, Sessa WC. Sphingosine 1-phosphate activates Akt, nitric oxide production, and chemotaxis through a Gi protein/phosphoinositide 3-kinase pathway in endothelial cells. J Biol Chem. 2001; 276: 19672–19677.
Takemoto M, Liao J. Pleiotropic effects of 3-hydroxy-3-methylglutaryl coenzyme a reductase inhibitors. Arterioscler Thromb Vasc Biol. 2001; 21: 1712–1719.
Bokoch G, Prossnitz V. Isoprenoid metabolism is required for stimulation of the respiratory burst oxidase of HL-60 cells. J Clin Invest. 1992; 89: 402–408.
Schwartz GG, Olsson AG, Ezekowitz MD, Ganz P, Oliver MF, Waters D, Zeiher A, Chaitman BR, Leslie S, Stern T. Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIRACL) Study Investigators. Effects of atorvastatin on early recurrent ischemic events in acute coronary syndromes: the MIRACL study: a randomized controlled trial. JAMA. 2001; 285: 1711–1718.
van Oostrom AJ, Sijmonsma TP, Verseyden C, Jansen EH, de Koning EJ, Rabelink TJ, Castro Cabezas M. Postprandial recruitment of neutrophils may contribute to endothelial dysfunction. J Lipid Res. 2003; 44: 576–583.