Dysregulated RANK Ligand/RANK Axis in Hyperhomocysteinemic Subjects
Effect of Treatment With B-Vitamins
Background and Purpose— Homocysteine has been linked to increased risk of ischemic stroke and other cardiovascular events. Matrix degradation and inflammation play an important role in these disorders, and we have demonstrated increased levels of matrix-degrading enzymes and inflammatory cytokines in hyperhomocysteinemic individuals. Recent studies suggest that RANK ligand (RANKL) through interaction with its receptor RANK can modulate matrix degradation and inflammation. The present study aimed to examine the role of the RANKL/RANK axis in hyperhomocystinemia.
Methods— RANKL/RANK was measured on protein or mRNA level before and after B-vitamin supplementation in hyperhomocysteinemic individuals. We also examined the in vitro effects of soluble RANKL in peripheral blood mononuclear cells from hyperhomocysteinemic individuals.
Results— Our main findings were: (1) compared to peripheral blood mononuclear cells from controls, cells from hyperhomocysteinemic individuals had significantly higher gene expression of RANKL and RANK; (2) folic acid treatment for 6 weeks in an open, uncontrolled study significantly reduced gene expression of RANKL/RANK in peripheral blood mononuclear cells from these individuals; (3) compared to placebo, treatment with folic acid, vitamin B12, and vitamin B6 for 3 months in a randomized, double-blind trial significantly lowered serum levels of soluble RANKL in hyperhomocysteinemic individuals; and (4) in vitro, soluble RANKL markedly increased the release of matrix metalloproteinase-9 and inflammatory cytokines from peripheral blood mononuclear cells in hyperhomocysteinemic subjects.
Conclusions— Our findings suggest a dysregulated RANKL/RANK axis in hyperhomocysteinemic subjects. Based on their role in atherogenesis, this enhanced expression of RANKL and RANK could contribute to the increased risk of cardiovascular disease in hyperhomocystinemia. Moreover, treatment with B-vitamins may have beneficial implications for plaque stability in these individuals.
Data from epidemiological studies suggest that raised homocysteine levels are associated with an increased risk of ischemic stroke, myocardial infarction, and other vascular disorders.1–4 However, several large clinically controlled trials have now shown that lowering homocysteine with vitamin-B therapy does not reduce risk of cardiovascular disease (CVD) or mortality among subjects with known CVD.5–7 Regarding stroke, however, a recent meta-analysis showed that folic acid supplementation can effectively reduce the risk of stroke in primary prevention, with a particularly marked effect in those who achieved a decrease in the homocysteine concentration of >20%.8 Thus, the role of homocysteine in CVD is still debated, and the precise mechanism by which hyperhomocystinemia is related to atherogenesis remains unclear.9–11
Increasing evidence supports the involvement of inflammation and matrix degradation in the pathogenesis of atherosclerosis and plaque destabilization. We have previously shown that hyperhomocysteinemic subjects are characterized by raised serum levels of markers of inflammation and matrix degradation like IL-6, epithelial cell-derived neurophil-activating peptide (ENA)-78, and matrix metalloproteinase (MMP)-9, potentially promoting atherogenesis and plaque instability.12–14 The receptor activator of nuclear factor-kappa B (RANK) and osteoprotegerin are members of the tumor necrosis factor receptor superfamily.15–16 Osteoprotegerin is a critical modulator of RANK ligand (RANKL) activity through its action as a decoy receptor for RANKL, thereby blocking interaction of RANKL with its receptor RANK.15–16 The RANKL/RANK axis has been shown to have pleiotropic effects on bone metabolism, endocrine function, and the immune system, and recently this axis has also been implicated in the development of atherosclerotic disorders.16–18 Thus, RANKL has been detected in atherosclerotic lesions, potentially promoting inflammation, endothelial cell activation, and matrix degradation.19 In addition, a novel role of RANKL as a vascular permeability factor was recently reported.20 Furthermore, raised serum levels of soluble RANKL (sRANKL) have been reported in patients with heart failure,19,21,22 and a recent large-scale epidemiological study supports a role for RANKL in CVD by the ability of sRANKL to predict vascular risk in apparently healthy individuals.23 In this latter study, the authors also found a strong association between serum levels of sRANKL and forthcoming stroke and transient ischemic attack. Moreover, symptomatic carotid plaques have been shown to express high levels of osteoprotegerin, further supporting the involvement osteoprotegerin /RANKL/RANK axis in cerebrovascular diseases.
The aim of the present study was to examine the RANKL and RANK in hyperhomocystinemia. Several approaches were used. First, we compared gene expression of RANKL and RANK in peripheral blood mononuclear cells (PBMC) from hyperhomocysteinemic subjects and matched healthy normohomocysteinemic volunteers. Second, we studied the effect of folic acid therapy on these parameters in an open, uncontrolled study. Third, the effect of folic acid, vitamin B12 and vitamin B6 (TrioBe) on sRANKL levels was examined in a randomized placebo-controlled, double-blind trial. Finally, the in vitro effect of sRANKL on PBMC from hyperhomocysteinemic individuals was investigated.
Subjects and Study Design
Hyperhomocysteinemic Subjects and Healthy Control Subjects
Fourteen adults aged younger than 70 years with hyperhomocystinemia (fasting plasma total homocysteine concentration >15 μmol/L) were consecutively recruited at the Lipid Clinic, Rikshospitalet University Hospital, Norway (Table 1). Eight of these patients were included in an open, uncontrolled folic acid supplementation study. Briefly, subjects with hyperhomocystinemia received folic acid according to the following schedule: 5 mg every day for the first week and 1 mg every day for the next 5 weeks. All subjects had previously been instructed by a nutritionist to follow the National Cholesterol Education Program Step I Diet, and all remained on this diet throughout the study. Control subjects were 12 sex- and age-matched healthy health care workers with no history of hypertension, diabetes, CVD, or other acute or chronic illness, consecutively recruited in the same period and from the same area of Norway (eastern part; Table 1).
Vitamin B12, B6, and Folic Acid (TrioBe) Therapy
Forty-two adults 18 to 70 years of age with hyperhomocystinemia (fasting plasma total homocysteine concentration >15 μmol/L at screening) were recruited at the Lipid Clinic and the Department of Medical Biochemistry, Rikshospitalet University Hospital, and at Department of Clinical Chemistry, Ullevål University Hospital, Oslo, Norway (Table 2). The patients were randomized in a double-blind fashion to receive therapy with TrioBe (Recip AB) or placebo. Inclusion criteria were women and men aged 18 to 70 years of age, plasma homocysteine levels ≥15 μmol/L, and written consent. Exclusion criteria were the presence of cancer or other coexisting serious disease, reduced renal function, treatment with B-vitamins or untreated vitamin B12 deficiency, treatment with phenobarbital, phenytoin, methotrexate, or trimethoprim, pregnant or breast-feeding women, or inability to follow the protocol, as judged by the investigator. The study was designed to examine inflammatory parameters in hyperhomocystinemia. The allocation of patients to treatment groups was performed by randomization using the method of randomly permuted blocks. The randomization list and sealed envelopes were prepared by specialized staff at Recip AB, who was not involved in the study. All study investigators, personnel, and participants were unaware of the treatment assignments. The participants were randomly allocated to receiving either TrioBe (folic acid, 0.8 mg; cyanocobalamin, 0.5 mg; and pyridoxine hydrochloride,3.0 mg; 1 tablet per day; Recip AB) or an identical-appearing placebo tablet (1 tablet per day; Recip AB) for 3 months. Compliance as judged by pill count of returned unused pills was 90%, 86% to 95%, and 88%, 84% to 91% (P=0.11) in the TrioBe and placebo groups, respectively. Assessment of dietary intake was performed by self-administered food questionnaire (SmartDiet; Lipid Clinic)24 at baseline and after 3 months of treatment. Thirty-eight of the 42 participants completed the study (20 in the TrioBe and 18 in the placebo group). The reasons for not completing the study were: a new cardiovascular event (ie, stroke; n=1); side effects (n=1); start of new medication during the study period (n=1); and stopping to take pills (n=1).
All parts of the study were approved by the Regional Committee of Medical Ethics. The intervention studies were approved by The Norwegian Medicine Control Authorities. Informed consent was obtained from all subjects.
Blood Sampling Protocol
Venous blood samples were collected after an overnight fast and without medication ingestion the morning of sampling. Plasma and serum were processed and stored at −80°C. All analyses, except the routine laboratory assays, were performed after last patients had completed the treatment period. To avoid run-to-run variability, all samples from a given subject were analyzed on the same day.
Cell Isolation and Culturing
Peripheral blood mononuclear cells (PBMCs) were obtained from heparinized blood by gradient centrifugation in Isopaque-Ficoll (Lymphoprep; Nycomed). PBMC pellets for RNA analysis were immediately frozen and stored at −80°C. Freshly isolated PBMCs were incubated for 24 hours in 96-well trays (Costar; 2×106 cells/mL) in medium (RPMI-1640 medium; Sigma Chemical; containing 2 mmol/L L-glutamine and 5% autologous serum) with or without recombinant sRANKL (1.5 μg/mL; R&D Systems). Cell-free supernatants were harvested and stored at −80°C until analysis. The endotoxin levels of all stimulants and culture media were <10 pg/mL (Limulus Amebocyte Assay; BioWhittaker).
Real-Time Quantitative Reverse-Transcription Polymerase Chain Reaction
Total RNA was isolated from PBMC pellets as described previously.25 To detect gene expression of RANK and RANKL, 1 μg total RNA from each sample was reverse-transcribed by TaqMan high-capacity reverse transcription reagent kit (Applied Biosystems). For quantitative real-time reverse-transcription polymerase chain reaction amplification, sequence-specific polymerase chain reaction primers were designed using the Primer Express software version 1.5 (Applied Biosystems) for RANK (forward primer [FP]: 5′-CCCGTTGCAGCTCAACAAG-3′; reverse primer [RP]: 5′- GCATTTGTCCGTGGAGGAA-3′) and RANKL (FP: 5′-GTGCAAAAGGAATTACAACATATCGT-3′; RP: 5′-AACCATGAGCCATCCACCAT-3′). The ß-actin was used as a housekeeping gene for normalization (Applied Biosystems).
Concentrations of sRANKL were quantified by Enzyme Immunoassays from Biomedica GmBH (Vienna, Austria). Concentrations of MMP-9, ENA-78, and IL-6 were measured by Enzyme Immunoassays (R&D Systems).
Routine Laboratory Assays
Concentrations of P-homocysteine were measured on the Abbott IMx analyzer, S-folate and S-vitamin B12 were measured on the Wallac AutoDelfia analyzer, and P-total cholesterol, P-LDL-cholesterol, P-HDL-cholesterol, P-triglycerides, and P-creatinine were measured using the Modular-P platform (Roche).
Data are given as median (25–75 percentiles) if not otherwise stated. Data from patients and controls and differences in changes between groups were compared by the Mann-Whitney U test. For categorical data, Pearson χ2 test was used. Changes in variables within groups were analyzed by the Wilcoxon signed-rank test. Multiple linear regression analysis was used to test dependent and independent variables. Probability values (2-sided) were considered significant at values of P≤0.05.
PBMC Expression of RANKL and RANK in Hyperhomocysteinemic Subjects and Healthy Control Subjects
Table 1 shows characteristics of the control and patient populations. No significant differences were observed for age, total cholesterol, or creatinine, whereas low-density lipoprotein cholesterol was significantly higher in hyperhomocysteinemic patients compared to controls (Table 1). The expected differences were observed with regard to homocysteine, folate, and vitamin B12 (Table 1).
When comparing PBMCs from the hyperhomocysteinemic individuals and healthy controls, we found markedly enhanced gene expression of RANKL and its corresponding receptor, RANK, in cells from those with hyperhomocystinemia (Figure 1). Moreover, although the groups were imbalanced with regard to smoking and statin use, with higher proportions of these characteristics in those with hyperhomocystinemia (Table 1), there were no significant differences in gene expression of RANKL and RANK between smokers (n=9) and nonsmokers (n=5; P=0.74 and P=0.84, respectively). However, statin users showed higher gene expression of RANKL and RANK (n=5) compared to statin nonusers (n=5; P=0.072 and P=0.014, respectively). Eight of the hyperhomocysteinemic subjects were included in an open, uncontrolled folic acid study, and folic acid supplementation for 6 weeks resulted in a significant decrease in plasma levels of homocysteine (from 27, 22–40 μmol/L to 10, 9–11 μmol/L, n=8; P<0.02) and a significant increase in serum levels of folate (from 5.7, 4.3–7.7 nmol/L to 35, 25–49 nmol/L, n=8; P<0.02). Notably, these changes during folic acid supplementation were accompanied by a decrease in mRNA levels in PBMCs for both RANKL and RANK (Figure 1 A,B).
Effect of TrioBe Therapy on Soluble RANKL in Hyperhomocystinemia
RANKL also exists in circulation as a biological active molecule.17 Therefore, we next measured serum levels of sRANKL in subjects with hyperhomocystinemia and explored the effect of folic acid, vitamin B12, and vitamin B6 (TrioBe) supplementation on this mediator in a randomized, double-blind, placebo-controlled trial. Characteristics of the participants are given in Table 2, showing no significant differences between the treatment groups. Whereas 3 months of TrioBe treatment increased serum levels of folate from 7.6 (5.2–8.7 nmol/L) to 38 (30–50 nmol/L, n=20; P<0.001) and of vitamin B12 from 225 (181–245 pmol/L) to 388 (300–555 pmol/L, n=20; P<0.001), the plasma concentration of homocysteine was reduced from 19 (17–26 μmol/L) to 10 (8–12 μmol/L, n=20; P<0.001). No significant differences in these parameters occurred within the placebo group (n=18). Moreover, whereas no significant changes were observed within the placebo group (n=18; P=0.26), TrioBe significantly reduced serum concentration of sRANKL (Figure 2), resulting in a significant difference in changes between the 2 treatment groups (P=0.005). Although there was no differences in baseline characteristics between the 2 treatment groups (Table 2), those allocated to TrioBe had higher baseline levels of sRANKL (P=0.03; Figure 2). Although not significant, the TrioBe group tended to have a higher proportion of smokers and statin users (Table 2), and whereas there were no significant difference in sRANKL between smokers (114; 83–300 pg/mL; n=19) and nonsmokers (79; 66–142 pg/mL; n=19; P=0.23) at baseline, statin users showed significantly higher pretreatment sRANKL (122; 85–495 pg/mL; n=16) compared to nonusers (84; 60–197 pg/mL; n=22; P=0.04), potentially reflecting more severe lipid disturbances in the former group. However, in a linear regression analysis, “treatment group” remained a significant determinant (β=−0.35; P=0.025) for changes in sRANKL also after adjusting for statin use. Nevertheless, we cannot exclude that an imbalance in smoking and lipid disturbances between the different study populations could have influenced some of our data.
Effect of RANKL In Vitro on PBMCs From Hyperhomocystinemia
RANKL has been shown to induce MMP activation and inflammation,17,19 and we have previously found that hyperhomocysteinemic subjects have increased plasma levels of MMP-9 as well as several inflammatory mediators such as IL-6 and ENA-78.12–14 To explore any possible pathogenic role of sRANKL in hyperhomocystinemia, we next examined the in vitro effect of sRANKL on the release of MMP-9, IL-6, and ENA-78 in PBMCs after culturing for 24 hours. The cells were isolated from 10 hyperhomocysteinemic subjects (plasma levels of homocysteine 18, 17–22 μmol/L; serum levels of folate 8.5, 7.0–11.0 nmol/L). As shown in Figure 3, recombinant sRANKL significantly enhanced the release of MMP-9, ENA-78, and IL-6, with particularly marked effects on the 2 latter mediators.
A role for the RANKL/RANK axis in atherogenesis and plaque destabilization has recently been reported by several groups, potentially also involving promotion of stroke and transient ischemic attack.16–23 In the present study we demonstrate that hyperhomocysteinemic subjects are characterized by enhanced gene expression of RANKL and its receptor RANK in PBMCs as well as increased serum levels of the soluble form of RANKL. We also show the ability of folic acid and B-vitamins to downregulate these mediators in hyperhomocysteinemic individuals. RANKL has been found to exert inflammatory effects in various cell types, and in the present study we show that sRANKL potently induced the release MMP-9 as well as the inflammatory cytokines ENA-78 and IL-6 in PBMCs from hyperhomocysteinemic subjects. Taken together, these findings may suggest that a dysregulated RANKL/RANK axis could be involved in the pathogenesis of hyperhomocystinemia-related CVD, and that B-vitamins may have potential implications for plaque stabilization in these individuals.
Previously, enhanced RANKL and RANK levels have been reported in patients with chronic heart failure, and several studies have demonstrated increased expression of these tumor necrosis factor-related mediators in atherosclerotic disorders.19–23 In the present study we demonstrate also that hyperhomocysteinemic individuals are characterized by increased expression of these mediators with enhanced expression of both the ligand (RANKL) and its corresponding receptor (RANK) in PBMCs as a major finding. Previously, RANKL has been detected within atherosclerotic plaques from abdominal aortas and carotid arteries in humans. We have recently shown enhanced expression of RANKL and RANK in both clinical and experimental atherosclerosis with particularly high levels in those with unstable disease, potentially contributing to the transition from a stable to an unstable plaque phenotype.19 Our findings in the present study suggest that RANKL and RANK should be added to the list of inflammatory mediators with relation to atherogenesis that are increased during hyperhomocystinemia. Moreover, several epidemiological observations have linked hyperhomocystinemia to increased risk for stroke and, interestingly, high RANKL levels have been shown to predict forthcoming cerebrovascular events in healthy subjects.23 It is therefore tempting to hypothesize that the association between hyperhomocystinemia and enhanced levels of RANK/RANKL also could be of importance for the hyperhomocystinemia-related increase in cerebrovascular disorders.
Remodeling of advanced atherosclerotic lesions and changes in plaque morphology are multifactorial processes, in which MMPs are suggested to play important roles.26–28 Previously, we have showed an altered MMP-9/tissue inhibitors of matrix metallo-proteinases (TIMP-1) balance in hyperhomocysteinemic individuals, which may contribute to the increased risk of ischemic stroke and other cardiovascular diseases in these subjects.14,25 We have also reported increased levels of the CXC chemokine ENA-78 and IL-6 in hyperhomocysteinemic subjects, potentially contributing to systemic and vascular inflammation in these individuals.12,13 RANKL has been shown to induce matrix degrading and inflammatory effects in various cells, and in the present study we show that sRANKL is a potent inducer of MMP-9, ENA-78, and IL-6 in PBMCs from hyperhomocysteinemic individuals. If this RANKL-mediated interaction also is operating in vivo in hyperhomocysteinemic individuals, then it could contribute to the matrix degrading and inflammatory and proatherogenic phenotype in these individuals, and in those with established atherosclerosis, these inflammatory interactions could potential promote plaque destabilization with subsequent development of transient ischemic attack, stroke, and myocardial infarction. It could be argued that a high concentration of RANKL was used in the in vitro experiment as compared with serum levels of this ligand. However, although we have no data, it is not inconceivable that such concentrations could exist within an inflammatory microenvironment such as within an atherosclerotic lesion consisting of several RANKL-expressing cells.
An important finding in the present study was that supplementation with folic acid and B-vitamins reduced gene expression of RANKL and its receptor RANK in PBMCs, as well as lowered serum levels of sRANKL. The recently reported improvement in stroke mortality observed after folic acid fortification in the United States and Canada, but not in England and Wales (where fortification is not mandatory), is consistent with the hypothesis that folic acid fortification helps to reduce deaths from stroke.29 These findings are supported by a recent meta-analysis, showing that folic acid supplementation can effectively reduce the risk of stroke in primary prevention.8 Interestingly, a significantly greater beneficial effect was seen in those trials with treatment duration of >36 months, and with a decrease in homocysteine concentration of >20%.8 Notably, a recent large-scale epidemiological study supports a role for sRANKL in stroke, myocardial infarction, and vascular death.23 The predictive significance of sRANKL was particularly marked for acute vascular syndromes like myocardial infarction and ischemic stroke.23 Based on these issues, it is tempting to speculate that the ability of B-vitamins to downregulate RANKL/RANK could contribute to their presumable beneficial effects in CVD.
The present study has some limitations. First, the differences in dose and duration in our 2 treatment studies made comparison of data somewhat difficult, although the doses of folic acid were in the same range (1 mg/day from week 2 on in the open study vs 0.8 mg/day in the TrioBe study). Second, it would have been an advantage if gene expression and gene product were measured in the same treatment study. Moreover, relatively few patients were included, and some of our data could have been influenced by differences in the study populations that were not related to homocysteine levels, such as smoking and statin use, and the role of RANKL/RANK in the hyperhomocystinemia-related CVD should be further investigated in larger study populations.
In conclusion, although there were some limitations, the present study suggests that hyperhomocysteinemic subjects are characterized by a dysregulation in the RANKL/RANK axis that at least partly may be restored by B-vitamin supplementation. Based on their role in atherogenesis, matrix degradation, and inflammation, this enhanced expression of RANKL and RANK in hyperhomocysteinemic subjects could contribute to the increased risk of CVD in these individuals, underlining the importance of restoration by B-vitamins.
The authors thank Liv Astrid Prestøy for excellent technical assistance.
Sources of Funding
This work was supported by grants from the Norwegian Foundation for Health and Rehabilitation, the Norwegian Association of Heart and Lung Patients, Nycomed Pharma, Oslo, Norway, and Recip AB, Årsta, Sweden.
M.S.N. and K.B.H. have received grant support from Nycomed Pharma, Oslo, Norway, and Recip AB, Årsta, Sweden.
- Received April 14, 2008.
- Accepted June 12, 2008.
Wald DS, Law M, Morris JK. Homocysteine and cardiovascular disease: evidence on causality from a meta-analysis. BMJ. 2002; 325: 1202–1206.
Toole JF, Malinow MR, Chambless LE, Spence JD, Pettigrew LC, Howard VJ, Sides EG, Wang CH, Stampfer M. Lowering homocysteine in patients with ischemic stroke to prevent recurrent stroke, myocardial infarction, and death: the Vitamin Intervention for Stroke Prevention (VISP) randomized controlled trial. JAMA. 2004; 291: 565–575.
Lonn E, Yusuf S, Arnold MJ, Sheridan P, Pogue J, Micks M, McQueen MJ, Probstfield J, Fodor G, Held C, Genest J Jr; Heart Outcomes Prevention Evaluation (HOPE) 2 Investigators. Homocysteine lowering with folic acid and B vitamins in vascular disease. N Engl J Med. 2006; 354: 1567–1577.
Kokame K, Kato H, Miyata T. Homocysteine-respondent genes in vascular endothelial cells identified by differential display analysis. GRP78/BiP and novel genes. J Biol Chem. 1996; 271: 29659–29665.
Upchurch GR Jr, Welch GN, Fabian AJ, Freedman JE, Johnson JL, Keaney JF Jr, Loscalzo J. Homocyst(e) ine decreases bioavailable nitric oxide by a mechanism involving glutathione peroxidase. J Biol Chem. 1997; 272: 17012–17017.
Holven KB, Aukrust P, Holm T, Ose L, Nenseter MS. Folic acid treatment reduces chemokine release from peripheral blood mononuclear cells in hyperhomocysteinemic subjects. Arterioscler Thromb Vasc Biol. 2002; 22: 699–703.
Holven KB, Halvorsen B, Bjerkeli V, Damås JK, Retterstøl K, Mørkrid L, Ose L, Aukrust P, Nenseter MS. Impaired inhibitory effect of interleukin-10 on the balance between matrix metalloproteinase-9 and its inhibitor in mononuclear cells from hyperhomocysteinemic subjects. Stroke. 2006; 37: 1731–1736.
Collin-Osdoby P. Regulation of vascular calcification by osteoclast regulatory factors RANKL and osteoprotegerin. Circ Res. 2004; 95: 1046–1057.
Min JK, Kim YM, Kim SW, Kwon MC, Kong YY, Hwang IK, Won MH, Rho J, Kwon YG. TNF-related activation-induced cytokine enhances leukocyte adhesiveness: induction of ICAM-1 and VCAM-1 via TNF receptor-associated factor and protein kinase C-dependent NF-kappaB activation in endothelial cells. J Immunol. 2005; 175: 531–540.
Sandberg WJ, Yndestad A, Øie E, Smith C, Ueland T, Ovchinnikova O, Robertson AK, Müller F, Semb AG, Scholz H, Andreassen AK, Gullestad L, Damås JK, Frøland SS, Hansson GK, Halvorsen B, Aukrust P. Enhanced T-cell expression of RANK ligand in acute coronary syndrome: possible role in plaque destabilization. Arterioscler Thromb Vasc Biol. 2006; 26: 857–863.
Min JK, Cho YL, Choi JH, Kim Y, Kim JH, Yu YS, Rho J, Mochizuki N, Kim YM, Oh GT, Kwon YG. Receptor activator of nuclear factor (NF)-kappaB ligand (RANKL) increases vascular permeability: impaired permeability and angiogenesis in eNOS-deficient mice. Blood. 2007; 109: 1495–1502.
Ueland T, Yndestad A, Øie E, Florholmen G, Halvorsen B, Frøland SS, Simonsen S, Christensen G, Gullestad L, Aukrust P. Dysregulated osteoprotegerin/RANK ligand/RANK axis in clinical and experimental heart failure. Circulation. 2005; 111: 2461–2468.
Kiechl S, Schett G, Schwaiger J, Seppi K, Eder P, Egger G, Santer P, Mayr A, Xu Q, Willeit J. Soluble receptor activator of nuclear factor-kappa B ligand and risk for cardiovascular disease. Circulation. 2007; 116: 385–391.
Nikkari ST, O'Brien KD, Ferguson M, Hatsukami T, Welgus HG, Alpers CE, Clowes AW. Interstitial collagenase (MMP-1) expression in human carotid atherosclerosis. Circulation. 1995; 92: 1393–1398.
Schonbeck U, Mach F, Sukhova GK, Murphy C, Bonnefoy JY, Fabunmi RP, Libby P. Regulation of matrix metalloproteinase expression in human vascular smooth muscle cells by T lymphocytes: a role for CD40 signaling in plaque rupture? Circ Res. 1997; 81: 448–454.
Yang Q, Botto LD, Erickson JD, Berry RJ, Sambell C, Johansen H, Friedman JM. Improvement in stroke mortality in Canada and the United States, 1990 to 2002. Circulation. 2006; 113: 1335–1343.