Journal of Human Hypertension (2015), 1–6 © 2015 Macmillan Publishers Limited All rights reserved 0950-9240/15 www.nature.com/jhh
ORIGINAL ARTICLE
Desphospho-uncarboxylated matrix Gla protein is associated with increased aortic stiffness in a general population O Mayer Jr1,2, J Seidlerová1,2, P Wohlfahrt3,4, J Filipovský1,2, J Vaněk1, R Cífková3,4, J Windrichová5, O Topolčan5, MHJ Knapen6, NEA Drummen6 and C Vermeer6 Matrix Gla protein (MGP), a natural inhibitor of calcification, strongly correlates with the extent of coronary calcification. Vitamin K is the essential cofactor for the activation of MGP. The nonphosphorylated-uncarboxylated isoform of MGP (dp-ucMGP) reflects the status of this vitamin. We investigated whether there is an association between dp-ucMGP and stiffness of elastic and musculartype large arteries in a random sample from the general population. In a cross-sectional design, we analyzed 1087 subjects from the Czech post-MONICA study. Aortic and femoro-popliteal pulse wave velocities (PWVs) were measured using a Sphygmocor device. Dp-ucMGP concentrations were assessed in freshly frozen samples by enzyme-linked immunosorbent assay methods using the InaKtif MGP iSYS pre-commercial kit developed by IDS and VitaK. Aortic PWV significantly (P o 0.0001) increased across the dp-ucMGP quartiles. After adjustment for all potential confounders, aortic PWV independently correlated with dp-ucMGP (with beta coefficient (s.d.) 11.61 (5.38) and P-value = 0.031). In a categorized manner, subjects in the top quartile of dp-ucMGP (⩾ 671 pmol l−1) had a higher risk of elevated aortic PWV, with corresponding adjusted odds ratio (95% confidence interval) 1.73 (1.17–2.5). In contrast, no relation between dp-ucMGP and femoro-popliteal PWV was found. In conclusion, increased dp-ucMGP, which is a circulating biomarker of vitamin K status and vascular calcification, is independently associated with aortic stiffness, but not with stiffness of distal muscular-type arteries. Journal of Human Hypertension advance online publication, 28 May 2015; doi:10.1038/jhh.2015.55
INTRODUCTION Arteriosclerosis is a typical manifestation of vascular aging. It is characterized by remodeling (lumen enlargement and wall thickening) and stiffening (progressive loss of elastic properties) of large elastic arteries.1 Arteriosclerosis can coexist and interfere with atherosclerosis ( = development of intimal atheromatous plaques) despite the fact that both pathological processes are different. Progressive arterial stiffening can be detected clinically as increased pulse wave velocity (PWV), and noninvasive techniques for direct measurement of PWV are currently available. The media of large arteries is mainly composed of smooth muscle cells (30–50%), elastic lamellae formed by elastin (ca 25%) and more rigid collagen fibers (ca 35%).2 The key feature of vascular aging is that repeated fractures, fragmentation and thinning of elastin lamellae transfer the pulse wave pressure to more rigid collagen fibers, resulting in loss of elasticity of the arterial wall.3 However, also other processes in arterial media are responsible for stiffening of vessel wall, such as increase of collagen content, accumulation of advanced glycation end products and calcification of elastic elements (elastocalcinosis).2,4–6 The majority of pathophysiological processes in arterial stiffening is highly age related and thus represents the 'natural course' of vascular aging. However, in some subjects this process is accelerated.
Calcification of arterial media was once regarded as a passive degenerative process and a clinical marker of advanced or endstage diseases such as renal failure, diabetes mellitus, advanced atherosclerosis and so on. However, calcification is currently considered an active and highly regulated process, sharing many features with embryonic bone formation. Surprisingly, it might be at least potentially reversible.7–10 Matrix Gla protein (MGP) is a potent natural inhibitor of vascular calcification. It is secreted by smooth muscle cells in arterial media.11,12 Vitamin K has a crucial role in the synthesis of mature MGP, serving as a cofactor for the enzyme γ-glutamate carboxylase that converts glutamate residues into γ-carboxyglutamate (Gla).13 Different MGP isoforms are found in the circulation, but in the present study, we focused on desphospho-uncarboxylated MGP (dp-ucMGP).14–16 Because of loss of its own anticalcemic activity, the dp-ucMGP isoform may serve as a perfect surrogate biomarker for vascular vitamin K status (inversely)17 and extent of coronary calcification (positively).18 Moreover, circulating dp-ucMGP was found to be associated with mortality risk in the general population, in chronic kidney disease patients and in patients with stable manifest coronary heart disease.17,19,20 To our knowledge, no previous studies investigated whether dp-ucMGP correlates with vascular calcification in non-coronary localization and in healthy subjects. In the present paper, we investigated the association between circulating dp-ucMGP and
1 2nd Department of Internal Medicine, Medical Faculty of Charles University and University Hospital, Pilsen, Czech Republic; 2Biomedical Centre, Medical Faculty of Charles University, Pilsen, Czech Republic; 3Centre for Cardiovascular Prevention of the First Faculty of Medicine, Charles University and Thomayer Hospital, Prague, Czech Republic; 4 International Clinical Research Centre, St. Anne’s University Hospital, Brno, Czech Republic; 5Department of Immunodiagnostics, University Hospital, Pilsen, Czech Republic and 6 VitaK, Cardiovascular Research Institute, Maastricht University, Maastricht, The Netherlands. Correspondence: Professor O Mayer, 2nd Department of Internal Medicine, Medical Faculty of Charles University and University Hospital, 13 E Beneše Street, Pilsen 320 00 Czech Republic. E-mail: [email protected] Received 2 February 2015; revised 16 March 2015; accepted 7 April 2015
Desphospho-uncarboxylated MGP and aortic stiffness O Mayer Jr et al
2 aortic (carotid–femoral) PWV, as a proposed manifestation of medial elastocalcinosis. SUBJECTS AND METHODS Study population The study population consisted of a random general population sample. A survey of risk factors was undertaken in 2008 as a part of Czech postMONICA study.21 One percent of the Pilsen residents aged 25–75 years were selected from the General Health Insurance Registry. A total of 1417 subjects (663 males and 754 females, mean age 55.2 years) responded to the survey (that is, more than 68% of the invited residents). Data from PWV measurements were available in 1131 of them. We further excluded 44 subjects from the final analysis because dp-ucMGP could not be determined (missing samples or insufficient plasma volumes, highly lipemic plasma and so on). The informed consent was obtained from all subjects and all personal data were stored under the provisions of the Czech Data Protection Act.
Examinations and materials All study procedures were performed according to Good Clinical Practice regulation and study protocol approved by the central Ethical Committee of the Institute for Clinical and Experimental Medicine Prague and by the local Ethical Committee of the University Hospital Pilsen. All responders were interviewed and examined by standardized methods using the WHO MONICA standard manual. Methods of interview are described in detail elsewhere.21 Briefly, information on personal and demographic characteristics, personal and family history of coronary heart disease, life style and current pharmacotherapy were obtained. The following standardized examinations were performed: height and weight in light indoor clothes without shoes by DETECTO 20 (Webb City, MO, USA) scale and measuring stick. Blood pressure (BP) was measured three times on the right arm in sitting position using standard mercury sphygmomanometers to the nearest 2 mm Hg; appropriate cuff size was used. The mean arterial pressure derived from the office BP measurement was calculated as the diastolic pressure plus one-third of the pulse pressure. We methodologically followed the Expert Consensus Document from 2006 (ref. 2) to measure large-artery properties. We used a Sphygmocor device (AtCor Medical Ltd, West Ryde, NSW, Australia) to quantify PWV. With the patient in supine position, we measured aortic PWV (aPWV) between carotid and femoral arteries and distal PWV (dPWV) between femoral and dorsalis pedis or tibialis posterior arteries. Registrations of the pulse waves were electrocardiography gated and thus the time shift between the appearance of the wave at the first and the second sites could be calculated. The distance between the two sites was measured on the body surface. To obtain travel distance along the aorta, we measured the distance from the jugular fossa to the pulsation of the femoral artery in the groin, and we subtracted the distance from the jugular fossa to carotid pulsation from this distance. PWV was calculated as the ratio of the travel distance in meters to the transit time in seconds. Venous blood samples were drawn after at least 12 h of overnight fasting. All these study procedures were done on the same day and at about the same time during the examination campaign. Frozen samples, stored at − 80 °C, were used for biochemical laboratory analyses. The laboratory examinations, including assessment of total cholesterol and high-density lipoprotein cholesterol, triglycerides and creatinine were provided by the central laboratories of the post-MONICA survey (Institute of Clinical and Experimental Medicine, Prague, Czech Republic) from serum samples using a Cobas Mira/ROCHE analyser (Basel, Switzerland) and commercially available kits of the same provenience. Glucose levels were analyzed by enzymatic methods in the same laboratory using LACHEMA (Brno, Czech Republic) standard kits. Lowdensity lipoprotein cholesterol was calculated by the Friedewald equation, that is, low-density lipoprotein cholesterol = total cholesterol − high-density lipoprotein cholesterol − (triglyceride/2.22). Creatinine clearance was estimated by the Cocroft-Gault formula, that is, [(140 − age) × body weight × 1.23 (or 1.04 for women)/serum creatinine]. 25-hydroxyvitamin D was determined in the research laboratory of Department of Immunodiagnostics, University Hospital, Pilsen, Czech Republic, from serum samples by using an ACCESS 2 analyzer (Beckman-Coulter, Brea, CA, USA) and commercial kits of the same company. Dp-ucMGP was quantified in citrate plasma samples using the InaKtif MGP iSYS kit (IDS, Boldon, UK), a pre-commercial automated assay based on the sandwich Journal of Human Hypertension (2015), 1 – 6
(dual antibody) ELISA kits developed by VitaK (Maastricht University, Maastricht, The Netherlands). The intra- and inter-assay variation were 4.0 and 5.2%, respectively.
Data analysis The study had a cross-sectional design. Statistical analyses of the data were performed using STATISTICA 9 (StatSoft Inc., Tulsa, OK, USA) and STATA 6 (Stata Corporation, College Station, TX, USA) software packages. Power calculations were performed using s.d., ascertained in our previous studies. Conventional risk factors were dichotomized by usual cutoff points (see relevant section of tables). Overt diabetes was defined as fasting plasma glucose greater than or equal to 7.0 mmol l − 1 and/or the use of antidiabetic medications. To define moderate kidney insufficiency, we used the cutoff point of calculated creatinine clearance o60 ml min − 1 (none of the subjects had clearance less than 30 ml min − 1). Low vitamin D status was defined as 25-hydroxyvitamin D o 40.9 ng ml − 1 (that is, upper limit of the bottom quartile). Dp-ucMGP was subdivided by its quartiles: ⩽ 383, 384–523, 524–670 and ⩾ 671 pmol l − 1. The lower limit of the top quartile was used as a cutoff point of increased dp-ucMGP. As increased aPWV were defined all values ⩾ 9 m s − 1. Increased dPWV was defined as ⩾ 12.6 m s − 1, that is, 4th quartile of this parameter ('normal' values for dPWV were not established). For statistical analysis we used conventional statistical methods, namely descriptive statistics, multivariate linear and logistic regressions to ascertain the association between dp-ucMGP and large-artery properties after adjustment for potential confounders (the methods are specified in relevant sections).
RESULTS The study included 1087 subjects, 513 males and 574 females, aged 25–75 years. The baseline characteristics are given in Table 1. aPWV, but not dPWV, significantly increased across increasing dp-ucMGP quartiles (Figure 1); dp-ucMGP and aPWV also have significant mutual correlation as continuous variables (with Spearman's correlation coefficient 0.373 and P-value o 0.0001). In multivariate linear regression analysis, we identified the following covariates as significant positive determinants of dp-ucMGP (as continuous variable): body mass index and treatment with warfarin, whereas current smoking and calculated creatinine clearance were negative determinants (Table 2). After adjustment for these confounders, aPWV as a continuous variable (model A), but not dPWV (model B), was positively associated with dp-ucMGP concentrations. Table 3 shows adjusted associations between aPWV and dpucMGP in a categorized manner. Using stepwise logistic regression, we identified the following variables as significant predictors of increased aPWV (⩾9 m s − 1) (model A): age ⩾ 50 years, obesity, raised BP, overt diabetes, moderate kidney insufficiency (creatinine clearance o60 ml min − 1) and the highest quartile of dp-ucMGP. In a separate analysis (model B), increased dPWV (⩾12.6 m s − 1) was positively associated with age ⩾ 50 years, male gender and raised BP, whereas increased dp-ucMGP did not enter this regression model. Figure 2 gives the fully adjusted odds ratios and 95% confidence intervals for increased dp-ucMGP as a predictor of high aPWV separately by pre-specified subgroups of potential covariates (black bars represent relative risk of high aPWV in the subject with a selected cofactor present). The highest difference in relative risk was found by age subgroup (odds ratio 1.69 versus 11.72). Moreover, increased dp-ucMGP predicts high aPWV significantly in subjects without a history of vascular disease, current smoking habit, raised BP, overt diabetes or moderate kidney insufficiency, whereas only nonsignificantly in subjects with these single characteristics. DISCUSSION In general population, increased plasma dp-ucMGP concentration is independently associated with increased aortic stiffness, © 2015 Macmillan Publishers Limited
Desphospho-uncarboxylated MGP and aortic stiffness O Mayer Jr et al
3 quantified as carotid–femoral PWV (aPWV). Subjects in the top quartile of dp-ucMGP had a significantly higher risk (by 70%) of increased aPWV. To our knowledge, this is the first study investigating the possible role of vitamin K status (reflected by dp-ucMGP) and arterial stiffness. Previous studies with dp-ucMGP focused on vascular calcification measured by X-ray techniques
Table 1. Baseline characteristics of the study sample (mean (s.d.) or factor proportion) N Age (years) Age ⩾ 50 years (%) Gender (% of males) History of vascular disease (%)a Current smoking (%) Body mass index (kg m −2) Body mass index ⩾ 30 kg m−2 (%) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Mean arterial pressure (mm Hg) Raised blood pressure (%)b Treatment with antihypertensives (%) Treatment with ACEi or ARBs (%) Total cholesterol (mmol l− 1) HDL-cholesterol (mmol l − 1) LDL-cholesterol (mmol l − 1) Triglycerides (mmol l − 1) Treatment with hypolipidemics (%) Treatment with statins (%) Fasting glycemia (mmol l − 1) Overt diabetes (%)c Treatment with antidiabetics (%) 25-Hydroxyvitamin D3 (ng ml − 1) dp-ucMGP (pmol l − 1) Treatment with warfarin (%) Creatinine clearance (ml min − 1) Creatinine clearance o60 ml min − 1 (%) Aortic pulse wave velocity (m s − 1) Distal pulse wave velocity (m s − 1)
1087 54.8 (13.0) 66.6 47.2 7.8 30.6 27.6 (4.9) 28.1 129.0 (17.3) 81.0 (9.3) 97.0 (10.8) 31.2 49.9 34.7 5.18 (1.01) 1.44 (0.37) 3.08 (0.91) 1.48 (1.01) 19.0 14.6 5.39 (1.21) 9.0 3.7 53.3 (17.9) 570.5 (330.8) 2.4 98.9 (28.2) 5.1 7.93 (2.32) 11.15 (2.38)
Abbreviations: ACEi, angiotensin-converting enzyme inhibitors; ARBs, angiotensin II receptor blockers; dp-ucMGP,desphospho-uncarboxylated matrix Gla protein; HDL, high-density lipoprotein; LDL, low-density lipoprotein. aHistory of myocardial infarction, ischemic stroke, peripheral vascular disease or atherosclerosis in other localizations, including all revascularization procedures. bSystolic blood pressure ⩾ 140 and/or diastolic blood pressure ⩾ 90 mm Hg. cFasting glycemia ⩾ 7 mmol l − 1 and/or treatment with antidiabetics.
and mainly in coronary arteries (that is, atherosclerosis). The aPWV technique reflects mostly changes affecting arterial media (arteriosclerosis) and on large elastic central arteries (mainly aorta). Indeed, we cannot exclude that vitamin K status influences the arterial stiffness not only by mediocalcinosis, but this association may be also atherosclerosis related (via sub/intimal sclerotic plaques). Numerous prospective studies identified aortic stiffness as a powerful predictor of future cardiovascular events. Laurent et al.22 reported that a 5-m s − 1 increase of carotid–femoral pulse wave velocity (a marker of aortic stiffness) was associated in hypertensive patients with a 2.35-times higher risk of cardiovascular death, and a similar causality is evident in various other settings (end-stage renal disease or diabetic patients, general populationbased samples, older subjects, patients with anginous symptoms, erectile dysfunction and so on). A meta-analysis by Vlachopoulos et al.23 included 17 cohort studies with pooled samples of 15 877 subjects. The top tertile of aPWV was associated with 2.26 and 1.61 higher risk of cardiovascular events and all-cause mortality, respectively. The causality of arterial stiffness (clinically manifesting as raised central BP) in global cardiovascular burden was also confirmed by the interventional CAFE trial.24 MGP is a natural inhibitor of calcification. Therefore, we assume that the observed association between dp-ucMGP and arterial stiffness reflects a distinct calcification of the central arterial wall. It is evident that MGP may exert its biological role both in coronary arteries and in the non-coronary vascular bed. The universal role of MGP in vascular anti-calcification defense was established in murine models using MGP-deficient mice, which exhibited extensive aortic calcifications.25 Warfarin, a functional vitamin K antagonist, effectively inactivates MGP synthesis and induces aortic calcifications resulting in functional damage (even rupture) in wild-type mice.26 In humans, Wang et al.27 observed an association between the genetic variants of MGP and calcification in the arterial wall including the aortic vascular bed (but not in sclerotic plaques). Similar results were reported by other research groups.28,29 From a clinical point of view, the direct biological link between MGP and vitamin K is potentially very important. Our current finding raises the question whether vitamin K supplementation might be effective in reducing arterial stiffness (and consequently in primary prevention of vascular events). Observational studies reported that low intake of K2 vitamin was consistently associated with the extent of vascular calcification and incidence of cardiovascular events.30–34 Nonetheless, rather limited data are available from interventional studies. In a rat model, a high dose of both vitamin K1 and K2 was shown to reverse warfarin-induced
Figure 1. Box and whisker plots of aortic and distal pulse wave velocities (median and intequartile range) by dp-ucMGP quartiles. The P-values were calculated by Kruskal–Wallis analysis of variance; limits of dp-ucMGP quartiles are as follows: ⩽ 383, 384–523, 524–670 and ⩾ 671 pmol l − 1. © 2015 Macmillan Publishers Limited
Journal of Human Hypertension (2015), 1 – 6
Desphospho-uncarboxylated MGP and aortic stiffness O Mayer Jr et al
4 Table 2.
Multivariate association between desphospho–uncarboxylated matrix Gla protein and pulse wave velocities adjusted for potential confounders
Age Male gender History of vascular disease Current smoking Body mass index Mean arterial pressure Treatment with antihypertensives LDL-cholesterol Treatment with hypolipidemics Fasting glycemia Treatment with antidiabetics 25-Hydroxyvitamin D Treatment with warfarin Creatinine clearance Aortic pulse wave velocity Distal pulse wave velocity Const.
Model A
P-value
0.11 (1.26) 38.66 (21.27) − 8.84 (39.58) − 65.75 (21.30) 22.93 (3.07) − 0.66 (0.87) 11.30 (24.60) − 5.64 (10.90) 27.09 (28.94) − 1.23 (9.05) − 34.35 (56.74) − 0.56 (0.56) 1012.01 (60.62) − 3.37 (0.64) 11.61 (5.38) — 192.652 (117.82)
0.93 0.07 0.82 0.002 o0.0001 0.45 0.65 0.61 0.35 0.89 0.55 0.32 o0.0001 o0.0001 0.031 — 0.10
Model B 0.89 36.18 1.45 − 62.26 24.11 0.16 12.59 − 4.32 22.19 2.69 − 59.35 − 0.44 1032.99 − 3.49 — − 2.44 341.51
(1.24) (21.68) (40.50) (21.49) (3.09) (0.91) (24.66) (10.91) (29.02) (9.90) (58.06) (0.57) (62.36) (0.64) (4.19) (141.88)
P-value 0.47 0.10 0.97 0.004 o0.0001 0.86 0.61 0.69 0.45 0.79 0.31 0.44 o0.0001 o0.0001 — 0.56 0.016
Abbreviation: LDL, low-density lipoprotein. Multivariate linear regression model with dp-ucMGP as dependent continuous variables. Arterial stiffness parameters were evaluated in two separate models: model A—aortic PWV and model B—distal PWV. Numbers are regression coefficients with s.e.
Table 3.
Predictors of increased aortic (aPWV) and distal (dPWV) pulse wave velocities
Dependent variable
Age ⩾ 50 years Male gender Body mass index ⩾ 30 kg m −2 Raised blood pressure Overt diabetes Creatinine clearance o 60 ml min − 1 dp-ucMGP ⩾ 671 pmol l − 1
aPWV ⩾ 9 m s − 1 (model A)
dPWV ⩾ 12.6 m s − 1 (model B)
OR (95% CI)
P-value
OR (95% CI)
P-value
8.47 (4.54–15.8) Not entered 1.88 (1.28–2.77) 2.89 (2.02-4.17) 1.95 (1.12–3.40) 2.09 (1.04–4.18) 1.65 (1.11–2.43)
o 0.0001 — 0.001 o 0.0001 0.019 0.038 0.001
1.60 (1.10–2.32) 1.64 (1.18–2.28) Not entered 2.45 (1.74–3.45) Not entered Not entered Not entered
0.015 0.003 — o0.0001 — — —
Abbreviations: CI, confidence interval; dp-ucMGP,desphospho-uncarboxylated matrix Gla protein; OR, odds ratio. Multiple stepwise logistic regressions with aPWV ⩾ 9 m s − 1 (model A) or dPWV ⩾ 12.6 m s − 1 (model B) as dependent variables were used. In addition, the following variables were initially included into the full models: history of vascular disease, current smoking, low-density lipoprotein cholestrol ⩾ 2.5 mmol l − 1, 25-hydroxyvitamin D o40.9 ng ml −1, treatment with any antihypertensives, with ACEi or ARBs only, with any lipid-lowering drugs, with statins only and with warfarin. 'Not entered' refers to the fact that the variable did not reach the pre-specified significance level (5%) for addition into the final regression model.
aortic calcifications (in contrast to only standard dose of vitamin K1).9 To our knowledge, three human interventional studies were performed testing the efficacy of vitamin K1. Braam et al.10 reported that supplementation of vitamins K1 plus D resulted in favorable elastic properties of carotid artery compared with only-D vitamin monotherapy or placebo. Shea et al.11 studied the effect of K1 on coronary calcification. In a subgroup of subjects who had ⩾ 85% adherence to treatment and preexisting coronary calcifications, the progression of coronary calcification was 6% smaller than in the control group. Recently published 3-year intervention study by Knapen et al.35 on 244 healthy postmenopausal women, in which low-dose vitamin K2 (180 μg) was tested versus placebo, confirmed our results. The outcomes of this study clearly show an improved vascular elasticity in the K2-treated group, not only compared with placebo but also compared with the within-group baseline values. In the present study, we also investigated the potential interaction between dp-ucMGP and factors known to influence arterial stiffening. Despite the fact that age is the most powerful determinant of arterial stiffening, subjects younger than 50 years of age who had increased circulating dp-ucMGP levels had more than 11-fold higher risk of increased PWV. We also observed that Journal of Human Hypertension (2015), 1 – 6
dp-ucMGP predicts increased PWV preferentially in low-risk subjects, that is, in nonsmokers, in subjects without manifest vascular disease, hypertension, overt diabetes or moderate kidney insufficiency. This finding might be of potential clinical importance. Indeed, increased dp-ucMGP correctly identified subjects as 'high risk' who were considered as a 'normal risk' population by conventional approaches. As opposed to aortic stiffness, the stiffness of peripheral, muscular-type arteries (quantified as femoro-popliteal PWV) was not associated with dp-ucMGP. This finding might probably be explained by the different structure and physiological properties of central and peripheral arteries. Properties of muscular arteries are affected more by smooth muscle tone than by extracellular matrix structural components (like in central arteries)36—this is supported also by the fact that stiffness of distal arteries depends in our sample on age to far lesser extent than stiffness of central arteries. Study limitations First, our study had a cross-sectional design with all its limitations. Only a large randomized interventional trial with vitamin K © 2015 Macmillan Publishers Limited
Desphospho-uncarboxylated MGP and aortic stiffness O Mayer Jr et al
5 increased risk of PWV>9 m/sec
decreased risk of PWV>9 m/sec
1.69
age>50 years
11.72 1.91
male gender
2.07 3.89
history of vascular disease
1.69 2.05
current smoking
1.79 2.71
BMI>30 kg/m2
1.51 1.67
raised blood pressure
2.35 1.34
overt diabetes
1.83 0.75
creatinine clearance 30-60 mL/min
1.86 2.23
low D vitamin status
1.75
0.10
1.00
YES (factor present)
10.00
100.00
NO (factor absent)
Figure 2. Adjusted relative risk of increased dp-ucMGP on increased aortic PWV by subgroups of potential covariates. Multiple stepwise logistic regression with aPWV ⩾ 9 m s − 1 used as a dependent variable. The same independent variables, as in the Table 3 model, were included, but only odds ratios and 95% confidence intervals for increased dp-ucMGP (⩾671 pmol l − 1) were expressed. Black bars represents relative risk in subgroups with a cofactor present.
supplementation could finally elucidate the role of impaired vitamin K status in age-related stiffening of large arteries in a general population. Second, both PWV and vitamin K status (reflected by dp-ucMGP) were assessed using one single measurement. On the other hand, the strength of our data lies in a well-defined sample of relatively healthy subjects, who were examined using a standardized examination procedure of the established long-term epidemiological project (MONICA study). CONCLUSION In our sample from the general population, we confirmed that elevated dp-ucMGP was independently associated with high aPWV. Therefore, we propose that this factor may serve as a biomarker and potentially modifiable risk factor of vascular aging. Vitamin K supplementation (preferably with the K2 isoform) might form a promising concept how to selectively influence the agerelated changes in vessel wall and even global cardiovascular risk. Therefore, further research in this field is needed. What is already known about the topic ● Matrix Gla protein (MGP) is a natural inhibitor of calcification; vitamin K is the essential cofactor for its activation. ● MGP strongly correlates with the extent of coronary calcification, but its role in stiffening of large arteries has not yet been studied ● The nonphosphorylated-uncarboxylated isoform of MGP (dpucMGP) reflects the vitamin K status. What this study adds ● Elevated dp-ucMGP was independently associated with high aortic stiffness in the general population. ● Therapeutic manipulation of dp-ucMGP by vitamin K supplementation might form a new concept to influence age-related arterial stiffening.
ACKNOWLEDGEMENTS The clinical part of the project was supported by the Internal Grant Agency of Czech Ministry of Health (NT13186) and by the Charles University Research Fund (project number P36). We are grateful to IDS Plc., providing free of charge, newly developed
© 2015 Macmillan Publishers Limited
kits for estimation of dp-ucMGP. We would like to acknowledge the hard work of all investigators in the Czech Republic who participated in the post-MONICA project, and last but not least all laboratory technicians, realizing extent laboratory assessments for the present study.
CONFLICT OF INTEREST The authors declare no conflict of interest REFERENCES 1 O’Rourke MF, Mancia G. Arterial stiffness. J Hypertens 1999; 17: 1–4. 2 Laurent S, Cockcroft J, Van Bortel L, Boutouyrie P, Giannattasio C, Hayoz D et al. European network for non-invasive investigation of large arteries. Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J 2006; 27: 2588–2605. 3 Dao HH, Essalihi R, Bouvet C, Moreau P. Evolution and modulation of age-related medial elastocalcinosis: impact on large artery stiffness and isolated systolic hypertension. Cardiovasc Res 2005; 66: 307–317. 4 Fornieri C, Quaglino D Jr, Mori G. Role of the extracellular matrix in age-related modifications of the rat aorta. Ultrastructural, morphometric, and enzymatic evaluations. Arterioscler Thromb 1992; 12: 1008–1016. 5 Cattell MA, Anderson JC, Hasleton PS. Age-related changes in amounts and concentrations of collagen and elastin in normotensive human thoracic aorta. Clin Chim Acta 1996; 245: 73–84. 6 Bruel A, Oxlund H. Changes in biomechanical properties, composition of collagen and elastin, and advanced glycation endproducts of the rat aorta in relation to age. Atherosclerosis 1996; 127: 155–165. 7 Keeley FW, Partridge SM. Amino acid composition and calcification of human aortic elastin. Atherosclerosis 1974; 19: 287–296. 8 Demer LL, Tintut Y. Vascular calcification: pathobiology of a multifaceted disease. Circulation 2008; 117: 2938–2948. 9 Schurgers LJ, Spronk HM, Soute BA, Schiffers PM, DeMey JG, Vermeer C. Regression of warfarin-induced medial elastocalcinosis by high intake of vitamin K in rats. Blood 2007; 109: 2823–2831. 10 Braam LA, Hoeks AP, Brouns F, Hamulyak K, Gerichhausen MJ, Vermeer C. Beneficial effects of vitamins D and K on the elastic properties of the vessel wall in postmenopausal women: a follow-up study. Thromb Haemost 2004; 91: 373–380. 11 Shea MK, O'Donnell CJ, Hoffmann U, Dallal GE, Dawson-Hughes B, Ordovas JM et al. Vitamin K supplementation and progression of coronary artery calcium in older men and women. Am J Clin Nutr 2009; 89: 1799–1807. 12 Shanahan CM, Proudfoot D, Farzaneh-Far A, Weissberg PL. The role of Gla proteins in vascular calcification. Crit Rev Eukaryot Gene Expr 1998; 8: 357–375.
Journal of Human Hypertension (2015), 1 – 6
Desphospho-uncarboxylated MGP and aortic stiffness O Mayer Jr et al
6 13 Price PA, Urist MR, Otawara Y. Matrix Gla protein, a new gamma-carboxyglutamic acid-containing protein which is associated with the organic matrix of bone. Biochem Biophys Res Commun 1983; 117: 765–771. 14 Schurgers LJ, Cranenburg ECM, Vermeer C. Matrix Gla-protein: the calcification inhibitor in need of vitamin K. Thromb Haemost 2008; 100: 593–603. 15 Cranenburg ECM, Koos R, Schurgers LJ, Magdeleyns EJ, Schoonbrood TH, Landewé RB et al. Characterisation and potential diagnostic value of circulating matrix Gla protein (MGP) species. Thromb Haemost 2010; 104: 811–822. 16 Schurgers LJ, Spronk HM, Skepper JN, Hackeng TM, Shanahan CM, Vermeer C et al. Post-translational modifications regulate matrix Gla protein function: importance for inhibition of vascular smooth muscle cell calcification. J Thromb Haemost 2007; 5: 2503–2511. 17 van den Heuvel EG, van Schoor NM, Lips P, Magdeleyns EJ, Deeg DJ, Vermeer C et al. Circulating uncarboxylated matrix Gla protein, a marker of vitamin K status, as a risk factor of cardiovascular disease. Maturitas 2014; 77: 137–141. 18 Dalmeijer GW, van der Schouw YT, Vermeer C, Magdeleyns EJ, Schurgers LJ, Beulens JW. Circulating matrix Gla protein is associated with coronary artery calcification and vitamin K status in healthy women. J Nutr Biochem 2013; 24: 624–628. 19 Schurgers LJ, Barreto DV, Barreto FC, Liabeuf S, Renard C, Magdeleyns EJ et al. The circulating inactive form of matrix gla protein is a surrogate marker for vascular calcification in chronic kidney disease: a preliminary report. Clin J Am Soc Nephrol 2010; 5: 568–575. 20 Mayer O Jr, Seidlerová J, Bruthans J, Filipovský J, Timoracká K, Vaněk J et al. Desphospho-uncarboxylated matrix Gla-protein is associated with mortality risk in patients with chronic stable vascular disease. Atherosclerosis 2014; 235: 162–168. 21 Cífková R, Skodová Z, Bruthans J, Adámková V, Jozífová M, Galovcová M et al. Longitudinal trends in major cardiovascular risk factors in the Czech population between 1985 and 2007/8. Czech MONICA and Czech post-MONICA. Atherosclerosis 2010; 211: 676–681. 22 Laurent S, Boutouyrie P, Asmar R, Gautier I, Laloux B, Guize L et al. A Aortic stiffness is an independent predictor of all-cause and cardiovascular mortality in hypertensive patients. Hypertension 2001; 37: 1236–1241. 23 Vlachopoulos C, Aznaouridis K, Stefanadis C. Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and meta-analysis. J Am Coll Cardiol 2010; 55: 1318–1327. 24 Williams B, Lacy PS, Thom SM, Cruickshank K, Stanton A, Collier D et al. Differential impact of blood pressure-lowering drugs on central aortic pressure and clinical outcomes: principal results of the Conduit Artery Function Evaluation (CAFE) study. Circulation 2006; 113: 1213–1225.
Journal of Human Hypertension (2015), 1 – 6
25 Luo G, Ducy P, McKee MD, Pinero GJ, Loyer E, Behringer RR et al. Spontaneous calcification of arteries and cartilage in mice lacking matrix GLA protein. Nature 1997; 386: 78–81. 26 Krüger T, Oelenberg S, Kaesler N, Schurgers LJ, van de Sandt AM, Boor P et al. Warfarin induces cardiovascular damage in mice. Arterioscler Thromb Vasc Biol 2013; 33: 2618–2624. 27 Wang Y, Chen J, Zhang Y, Yu W, Zhang C, Gong L et al. Common genetic variants of MGP are associated with calcification on the arterial wall but not with calcification present in the atherosclerotic plaques. Circ Cardiovasc Genet 2013; 6: 271–278. 28 Tuñón-Le Poultel D, Cannata-Andía JB, Román-García P, Díaz-López JB, Coto E, Gómez C et al. Association of matrix Gla protein gene functional polymorphisms with loss of bone mineral density and progression of aortic calcification. Osteoporos Int 2014; 25: 1237–1246. 29 Yoshikawa K, Abe H, Tominaga T, Nakamura M, Kishi S, Matsuura M et al. Polymorphism in the human matrix Gla protein gene is associated with the progression of vascular calcification in maintenance hemodialysis patients. Clin Exp Nephrol 2013; 17: 882–889. 30 Beulens JW, Bots ML, Atsma F, Bartelink ML, Prokop M, Geleijnse JM et al. High dietary menaquinone intake is associated with reduced coronary calcification. Atherosclerosis 2009; 203: 489–493. 31 Jie KS, Bots ML, Vermeer C, Witteman JC, Grobbee DE. Vitamin K intake and osteocalcin levels in women with and without aortic atherosclerosis: a population-based study. Atherosclerosis 1995; 116: 117–123. 32 Villines TC, Hatzigeorgiou C, Feuerstein IM, O'Malley PG, Taylor AJ. Vitamin K1 intake and coronary calcification. Coron Artery Dis 2005; 16: 199–203. 33 Geleijnse JM, Vermeer C, Grobbee DE, Schurgers LJ, Knapen MH, van der Meer IM et al. Dietary intake of menaquinone is associated with a reduced risk of coronary heart disease: the Rotterdam Study. J Nutr 2004; 134: 3100–3105. 34 Gast GC, de Roos NM, Sluijs I, Bots ML, Beulens JW, Geleijnse JM et al. A high menaquinone intake reduces the incidence of coronary heart disease. Nutr Metab Cardiovasc Dis 2009; 19: 504–510. 35 Knapen MH, Braam LA, Drummen NE, Bekers O, Hoeks AP, Vermeer C. Menaquinone-7 supplementation improves arterial stiffness in healthy postmenopausal women: double-blind randomised clinical trial. Thromb Haemost 2015; 113: 1135–1144. 36 Mitchell GF, Parise H, Benjamin EJ, Larson MG, Keyes MJ, Vita JA et al. Changes in arterial stiffness and wave reflection with advancing age in healthy men and women: the Framingham Heart Study. Hypertension 2004; 43: 1239–1245.
© 2015 Macmillan Publishers Limited
ORIGINAL ARTICLE
Desphospho-uncarboxylated matrix Gla protein is associated with increased aortic stiffness in a general population O Mayer Jr1,2, J Seidlerová1,2, P Wohlfahrt3,4, J Filipovský1,2, J Vaněk1, R Cífková3,4, J Windrichová5, O Topolčan5, MHJ Knapen6, NEA Drummen6 and C Vermeer6 Matrix Gla protein (MGP), a natural inhibitor of calcification, strongly correlates with the extent of coronary calcification. Vitamin K is the essential cofactor for the activation of MGP. The nonphosphorylated-uncarboxylated isoform of MGP (dp-ucMGP) reflects the status of this vitamin. We investigated whether there is an association between dp-ucMGP and stiffness of elastic and musculartype large arteries in a random sample from the general population. In a cross-sectional design, we analyzed 1087 subjects from the Czech post-MONICA study. Aortic and femoro-popliteal pulse wave velocities (PWVs) were measured using a Sphygmocor device. Dp-ucMGP concentrations were assessed in freshly frozen samples by enzyme-linked immunosorbent assay methods using the InaKtif MGP iSYS pre-commercial kit developed by IDS and VitaK. Aortic PWV significantly (P o 0.0001) increased across the dp-ucMGP quartiles. After adjustment for all potential confounders, aortic PWV independently correlated with dp-ucMGP (with beta coefficient (s.d.) 11.61 (5.38) and P-value = 0.031). In a categorized manner, subjects in the top quartile of dp-ucMGP (⩾ 671 pmol l−1) had a higher risk of elevated aortic PWV, with corresponding adjusted odds ratio (95% confidence interval) 1.73 (1.17–2.5). In contrast, no relation between dp-ucMGP and femoro-popliteal PWV was found. In conclusion, increased dp-ucMGP, which is a circulating biomarker of vitamin K status and vascular calcification, is independently associated with aortic stiffness, but not with stiffness of distal muscular-type arteries. Journal of Human Hypertension advance online publication, 28 May 2015; doi:10.1038/jhh.2015.55
INTRODUCTION Arteriosclerosis is a typical manifestation of vascular aging. It is characterized by remodeling (lumen enlargement and wall thickening) and stiffening (progressive loss of elastic properties) of large elastic arteries.1 Arteriosclerosis can coexist and interfere with atherosclerosis ( = development of intimal atheromatous plaques) despite the fact that both pathological processes are different. Progressive arterial stiffening can be detected clinically as increased pulse wave velocity (PWV), and noninvasive techniques for direct measurement of PWV are currently available. The media of large arteries is mainly composed of smooth muscle cells (30–50%), elastic lamellae formed by elastin (ca 25%) and more rigid collagen fibers (ca 35%).2 The key feature of vascular aging is that repeated fractures, fragmentation and thinning of elastin lamellae transfer the pulse wave pressure to more rigid collagen fibers, resulting in loss of elasticity of the arterial wall.3 However, also other processes in arterial media are responsible for stiffening of vessel wall, such as increase of collagen content, accumulation of advanced glycation end products and calcification of elastic elements (elastocalcinosis).2,4–6 The majority of pathophysiological processes in arterial stiffening is highly age related and thus represents the 'natural course' of vascular aging. However, in some subjects this process is accelerated.
Calcification of arterial media was once regarded as a passive degenerative process and a clinical marker of advanced or endstage diseases such as renal failure, diabetes mellitus, advanced atherosclerosis and so on. However, calcification is currently considered an active and highly regulated process, sharing many features with embryonic bone formation. Surprisingly, it might be at least potentially reversible.7–10 Matrix Gla protein (MGP) is a potent natural inhibitor of vascular calcification. It is secreted by smooth muscle cells in arterial media.11,12 Vitamin K has a crucial role in the synthesis of mature MGP, serving as a cofactor for the enzyme γ-glutamate carboxylase that converts glutamate residues into γ-carboxyglutamate (Gla).13 Different MGP isoforms are found in the circulation, but in the present study, we focused on desphospho-uncarboxylated MGP (dp-ucMGP).14–16 Because of loss of its own anticalcemic activity, the dp-ucMGP isoform may serve as a perfect surrogate biomarker for vascular vitamin K status (inversely)17 and extent of coronary calcification (positively).18 Moreover, circulating dp-ucMGP was found to be associated with mortality risk in the general population, in chronic kidney disease patients and in patients with stable manifest coronary heart disease.17,19,20 To our knowledge, no previous studies investigated whether dp-ucMGP correlates with vascular calcification in non-coronary localization and in healthy subjects. In the present paper, we investigated the association between circulating dp-ucMGP and
1 2nd Department of Internal Medicine, Medical Faculty of Charles University and University Hospital, Pilsen, Czech Republic; 2Biomedical Centre, Medical Faculty of Charles University, Pilsen, Czech Republic; 3Centre for Cardiovascular Prevention of the First Faculty of Medicine, Charles University and Thomayer Hospital, Prague, Czech Republic; 4 International Clinical Research Centre, St. Anne’s University Hospital, Brno, Czech Republic; 5Department of Immunodiagnostics, University Hospital, Pilsen, Czech Republic and 6 VitaK, Cardiovascular Research Institute, Maastricht University, Maastricht, The Netherlands. Correspondence: Professor O Mayer, 2nd Department of Internal Medicine, Medical Faculty of Charles University and University Hospital, 13 E Beneše Street, Pilsen 320 00 Czech Republic. E-mail: [email protected] Received 2 February 2015; revised 16 March 2015; accepted 7 April 2015
Desphospho-uncarboxylated MGP and aortic stiffness O Mayer Jr et al
2 aortic (carotid–femoral) PWV, as a proposed manifestation of medial elastocalcinosis. SUBJECTS AND METHODS Study population The study population consisted of a random general population sample. A survey of risk factors was undertaken in 2008 as a part of Czech postMONICA study.21 One percent of the Pilsen residents aged 25–75 years were selected from the General Health Insurance Registry. A total of 1417 subjects (663 males and 754 females, mean age 55.2 years) responded to the survey (that is, more than 68% of the invited residents). Data from PWV measurements were available in 1131 of them. We further excluded 44 subjects from the final analysis because dp-ucMGP could not be determined (missing samples or insufficient plasma volumes, highly lipemic plasma and so on). The informed consent was obtained from all subjects and all personal data were stored under the provisions of the Czech Data Protection Act.
Examinations and materials All study procedures were performed according to Good Clinical Practice regulation and study protocol approved by the central Ethical Committee of the Institute for Clinical and Experimental Medicine Prague and by the local Ethical Committee of the University Hospital Pilsen. All responders were interviewed and examined by standardized methods using the WHO MONICA standard manual. Methods of interview are described in detail elsewhere.21 Briefly, information on personal and demographic characteristics, personal and family history of coronary heart disease, life style and current pharmacotherapy were obtained. The following standardized examinations were performed: height and weight in light indoor clothes without shoes by DETECTO 20 (Webb City, MO, USA) scale and measuring stick. Blood pressure (BP) was measured three times on the right arm in sitting position using standard mercury sphygmomanometers to the nearest 2 mm Hg; appropriate cuff size was used. The mean arterial pressure derived from the office BP measurement was calculated as the diastolic pressure plus one-third of the pulse pressure. We methodologically followed the Expert Consensus Document from 2006 (ref. 2) to measure large-artery properties. We used a Sphygmocor device (AtCor Medical Ltd, West Ryde, NSW, Australia) to quantify PWV. With the patient in supine position, we measured aortic PWV (aPWV) between carotid and femoral arteries and distal PWV (dPWV) between femoral and dorsalis pedis or tibialis posterior arteries. Registrations of the pulse waves were electrocardiography gated and thus the time shift between the appearance of the wave at the first and the second sites could be calculated. The distance between the two sites was measured on the body surface. To obtain travel distance along the aorta, we measured the distance from the jugular fossa to the pulsation of the femoral artery in the groin, and we subtracted the distance from the jugular fossa to carotid pulsation from this distance. PWV was calculated as the ratio of the travel distance in meters to the transit time in seconds. Venous blood samples were drawn after at least 12 h of overnight fasting. All these study procedures were done on the same day and at about the same time during the examination campaign. Frozen samples, stored at − 80 °C, were used for biochemical laboratory analyses. The laboratory examinations, including assessment of total cholesterol and high-density lipoprotein cholesterol, triglycerides and creatinine were provided by the central laboratories of the post-MONICA survey (Institute of Clinical and Experimental Medicine, Prague, Czech Republic) from serum samples using a Cobas Mira/ROCHE analyser (Basel, Switzerland) and commercially available kits of the same provenience. Glucose levels were analyzed by enzymatic methods in the same laboratory using LACHEMA (Brno, Czech Republic) standard kits. Lowdensity lipoprotein cholesterol was calculated by the Friedewald equation, that is, low-density lipoprotein cholesterol = total cholesterol − high-density lipoprotein cholesterol − (triglyceride/2.22). Creatinine clearance was estimated by the Cocroft-Gault formula, that is, [(140 − age) × body weight × 1.23 (or 1.04 for women)/serum creatinine]. 25-hydroxyvitamin D was determined in the research laboratory of Department of Immunodiagnostics, University Hospital, Pilsen, Czech Republic, from serum samples by using an ACCESS 2 analyzer (Beckman-Coulter, Brea, CA, USA) and commercial kits of the same company. Dp-ucMGP was quantified in citrate plasma samples using the InaKtif MGP iSYS kit (IDS, Boldon, UK), a pre-commercial automated assay based on the sandwich Journal of Human Hypertension (2015), 1 – 6
(dual antibody) ELISA kits developed by VitaK (Maastricht University, Maastricht, The Netherlands). The intra- and inter-assay variation were 4.0 and 5.2%, respectively.
Data analysis The study had a cross-sectional design. Statistical analyses of the data were performed using STATISTICA 9 (StatSoft Inc., Tulsa, OK, USA) and STATA 6 (Stata Corporation, College Station, TX, USA) software packages. Power calculations were performed using s.d., ascertained in our previous studies. Conventional risk factors were dichotomized by usual cutoff points (see relevant section of tables). Overt diabetes was defined as fasting plasma glucose greater than or equal to 7.0 mmol l − 1 and/or the use of antidiabetic medications. To define moderate kidney insufficiency, we used the cutoff point of calculated creatinine clearance o60 ml min − 1 (none of the subjects had clearance less than 30 ml min − 1). Low vitamin D status was defined as 25-hydroxyvitamin D o 40.9 ng ml − 1 (that is, upper limit of the bottom quartile). Dp-ucMGP was subdivided by its quartiles: ⩽ 383, 384–523, 524–670 and ⩾ 671 pmol l − 1. The lower limit of the top quartile was used as a cutoff point of increased dp-ucMGP. As increased aPWV were defined all values ⩾ 9 m s − 1. Increased dPWV was defined as ⩾ 12.6 m s − 1, that is, 4th quartile of this parameter ('normal' values for dPWV were not established). For statistical analysis we used conventional statistical methods, namely descriptive statistics, multivariate linear and logistic regressions to ascertain the association between dp-ucMGP and large-artery properties after adjustment for potential confounders (the methods are specified in relevant sections).
RESULTS The study included 1087 subjects, 513 males and 574 females, aged 25–75 years. The baseline characteristics are given in Table 1. aPWV, but not dPWV, significantly increased across increasing dp-ucMGP quartiles (Figure 1); dp-ucMGP and aPWV also have significant mutual correlation as continuous variables (with Spearman's correlation coefficient 0.373 and P-value o 0.0001). In multivariate linear regression analysis, we identified the following covariates as significant positive determinants of dp-ucMGP (as continuous variable): body mass index and treatment with warfarin, whereas current smoking and calculated creatinine clearance were negative determinants (Table 2). After adjustment for these confounders, aPWV as a continuous variable (model A), but not dPWV (model B), was positively associated with dp-ucMGP concentrations. Table 3 shows adjusted associations between aPWV and dpucMGP in a categorized manner. Using stepwise logistic regression, we identified the following variables as significant predictors of increased aPWV (⩾9 m s − 1) (model A): age ⩾ 50 years, obesity, raised BP, overt diabetes, moderate kidney insufficiency (creatinine clearance o60 ml min − 1) and the highest quartile of dp-ucMGP. In a separate analysis (model B), increased dPWV (⩾12.6 m s − 1) was positively associated with age ⩾ 50 years, male gender and raised BP, whereas increased dp-ucMGP did not enter this regression model. Figure 2 gives the fully adjusted odds ratios and 95% confidence intervals for increased dp-ucMGP as a predictor of high aPWV separately by pre-specified subgroups of potential covariates (black bars represent relative risk of high aPWV in the subject with a selected cofactor present). The highest difference in relative risk was found by age subgroup (odds ratio 1.69 versus 11.72). Moreover, increased dp-ucMGP predicts high aPWV significantly in subjects without a history of vascular disease, current smoking habit, raised BP, overt diabetes or moderate kidney insufficiency, whereas only nonsignificantly in subjects with these single characteristics. DISCUSSION In general population, increased plasma dp-ucMGP concentration is independently associated with increased aortic stiffness, © 2015 Macmillan Publishers Limited
Desphospho-uncarboxylated MGP and aortic stiffness O Mayer Jr et al
3 quantified as carotid–femoral PWV (aPWV). Subjects in the top quartile of dp-ucMGP had a significantly higher risk (by 70%) of increased aPWV. To our knowledge, this is the first study investigating the possible role of vitamin K status (reflected by dp-ucMGP) and arterial stiffness. Previous studies with dp-ucMGP focused on vascular calcification measured by X-ray techniques
Table 1. Baseline characteristics of the study sample (mean (s.d.) or factor proportion) N Age (years) Age ⩾ 50 years (%) Gender (% of males) History of vascular disease (%)a Current smoking (%) Body mass index (kg m −2) Body mass index ⩾ 30 kg m−2 (%) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Mean arterial pressure (mm Hg) Raised blood pressure (%)b Treatment with antihypertensives (%) Treatment with ACEi or ARBs (%) Total cholesterol (mmol l− 1) HDL-cholesterol (mmol l − 1) LDL-cholesterol (mmol l − 1) Triglycerides (mmol l − 1) Treatment with hypolipidemics (%) Treatment with statins (%) Fasting glycemia (mmol l − 1) Overt diabetes (%)c Treatment with antidiabetics (%) 25-Hydroxyvitamin D3 (ng ml − 1) dp-ucMGP (pmol l − 1) Treatment with warfarin (%) Creatinine clearance (ml min − 1) Creatinine clearance o60 ml min − 1 (%) Aortic pulse wave velocity (m s − 1) Distal pulse wave velocity (m s − 1)
1087 54.8 (13.0) 66.6 47.2 7.8 30.6 27.6 (4.9) 28.1 129.0 (17.3) 81.0 (9.3) 97.0 (10.8) 31.2 49.9 34.7 5.18 (1.01) 1.44 (0.37) 3.08 (0.91) 1.48 (1.01) 19.0 14.6 5.39 (1.21) 9.0 3.7 53.3 (17.9) 570.5 (330.8) 2.4 98.9 (28.2) 5.1 7.93 (2.32) 11.15 (2.38)
Abbreviations: ACEi, angiotensin-converting enzyme inhibitors; ARBs, angiotensin II receptor blockers; dp-ucMGP,desphospho-uncarboxylated matrix Gla protein; HDL, high-density lipoprotein; LDL, low-density lipoprotein. aHistory of myocardial infarction, ischemic stroke, peripheral vascular disease or atherosclerosis in other localizations, including all revascularization procedures. bSystolic blood pressure ⩾ 140 and/or diastolic blood pressure ⩾ 90 mm Hg. cFasting glycemia ⩾ 7 mmol l − 1 and/or treatment with antidiabetics.
and mainly in coronary arteries (that is, atherosclerosis). The aPWV technique reflects mostly changes affecting arterial media (arteriosclerosis) and on large elastic central arteries (mainly aorta). Indeed, we cannot exclude that vitamin K status influences the arterial stiffness not only by mediocalcinosis, but this association may be also atherosclerosis related (via sub/intimal sclerotic plaques). Numerous prospective studies identified aortic stiffness as a powerful predictor of future cardiovascular events. Laurent et al.22 reported that a 5-m s − 1 increase of carotid–femoral pulse wave velocity (a marker of aortic stiffness) was associated in hypertensive patients with a 2.35-times higher risk of cardiovascular death, and a similar causality is evident in various other settings (end-stage renal disease or diabetic patients, general populationbased samples, older subjects, patients with anginous symptoms, erectile dysfunction and so on). A meta-analysis by Vlachopoulos et al.23 included 17 cohort studies with pooled samples of 15 877 subjects. The top tertile of aPWV was associated with 2.26 and 1.61 higher risk of cardiovascular events and all-cause mortality, respectively. The causality of arterial stiffness (clinically manifesting as raised central BP) in global cardiovascular burden was also confirmed by the interventional CAFE trial.24 MGP is a natural inhibitor of calcification. Therefore, we assume that the observed association between dp-ucMGP and arterial stiffness reflects a distinct calcification of the central arterial wall. It is evident that MGP may exert its biological role both in coronary arteries and in the non-coronary vascular bed. The universal role of MGP in vascular anti-calcification defense was established in murine models using MGP-deficient mice, which exhibited extensive aortic calcifications.25 Warfarin, a functional vitamin K antagonist, effectively inactivates MGP synthesis and induces aortic calcifications resulting in functional damage (even rupture) in wild-type mice.26 In humans, Wang et al.27 observed an association between the genetic variants of MGP and calcification in the arterial wall including the aortic vascular bed (but not in sclerotic plaques). Similar results were reported by other research groups.28,29 From a clinical point of view, the direct biological link between MGP and vitamin K is potentially very important. Our current finding raises the question whether vitamin K supplementation might be effective in reducing arterial stiffness (and consequently in primary prevention of vascular events). Observational studies reported that low intake of K2 vitamin was consistently associated with the extent of vascular calcification and incidence of cardiovascular events.30–34 Nonetheless, rather limited data are available from interventional studies. In a rat model, a high dose of both vitamin K1 and K2 was shown to reverse warfarin-induced
Figure 1. Box and whisker plots of aortic and distal pulse wave velocities (median and intequartile range) by dp-ucMGP quartiles. The P-values were calculated by Kruskal–Wallis analysis of variance; limits of dp-ucMGP quartiles are as follows: ⩽ 383, 384–523, 524–670 and ⩾ 671 pmol l − 1. © 2015 Macmillan Publishers Limited
Journal of Human Hypertension (2015), 1 – 6
Desphospho-uncarboxylated MGP and aortic stiffness O Mayer Jr et al
4 Table 2.
Multivariate association between desphospho–uncarboxylated matrix Gla protein and pulse wave velocities adjusted for potential confounders
Age Male gender History of vascular disease Current smoking Body mass index Mean arterial pressure Treatment with antihypertensives LDL-cholesterol Treatment with hypolipidemics Fasting glycemia Treatment with antidiabetics 25-Hydroxyvitamin D Treatment with warfarin Creatinine clearance Aortic pulse wave velocity Distal pulse wave velocity Const.
Model A
P-value
0.11 (1.26) 38.66 (21.27) − 8.84 (39.58) − 65.75 (21.30) 22.93 (3.07) − 0.66 (0.87) 11.30 (24.60) − 5.64 (10.90) 27.09 (28.94) − 1.23 (9.05) − 34.35 (56.74) − 0.56 (0.56) 1012.01 (60.62) − 3.37 (0.64) 11.61 (5.38) — 192.652 (117.82)
0.93 0.07 0.82 0.002 o0.0001 0.45 0.65 0.61 0.35 0.89 0.55 0.32 o0.0001 o0.0001 0.031 — 0.10
Model B 0.89 36.18 1.45 − 62.26 24.11 0.16 12.59 − 4.32 22.19 2.69 − 59.35 − 0.44 1032.99 − 3.49 — − 2.44 341.51
(1.24) (21.68) (40.50) (21.49) (3.09) (0.91) (24.66) (10.91) (29.02) (9.90) (58.06) (0.57) (62.36) (0.64) (4.19) (141.88)
P-value 0.47 0.10 0.97 0.004 o0.0001 0.86 0.61 0.69 0.45 0.79 0.31 0.44 o0.0001 o0.0001 — 0.56 0.016
Abbreviation: LDL, low-density lipoprotein. Multivariate linear regression model with dp-ucMGP as dependent continuous variables. Arterial stiffness parameters were evaluated in two separate models: model A—aortic PWV and model B—distal PWV. Numbers are regression coefficients with s.e.
Table 3.
Predictors of increased aortic (aPWV) and distal (dPWV) pulse wave velocities
Dependent variable
Age ⩾ 50 years Male gender Body mass index ⩾ 30 kg m −2 Raised blood pressure Overt diabetes Creatinine clearance o 60 ml min − 1 dp-ucMGP ⩾ 671 pmol l − 1
aPWV ⩾ 9 m s − 1 (model A)
dPWV ⩾ 12.6 m s − 1 (model B)
OR (95% CI)
P-value
OR (95% CI)
P-value
8.47 (4.54–15.8) Not entered 1.88 (1.28–2.77) 2.89 (2.02-4.17) 1.95 (1.12–3.40) 2.09 (1.04–4.18) 1.65 (1.11–2.43)
o 0.0001 — 0.001 o 0.0001 0.019 0.038 0.001
1.60 (1.10–2.32) 1.64 (1.18–2.28) Not entered 2.45 (1.74–3.45) Not entered Not entered Not entered
0.015 0.003 — o0.0001 — — —
Abbreviations: CI, confidence interval; dp-ucMGP,desphospho-uncarboxylated matrix Gla protein; OR, odds ratio. Multiple stepwise logistic regressions with aPWV ⩾ 9 m s − 1 (model A) or dPWV ⩾ 12.6 m s − 1 (model B) as dependent variables were used. In addition, the following variables were initially included into the full models: history of vascular disease, current smoking, low-density lipoprotein cholestrol ⩾ 2.5 mmol l − 1, 25-hydroxyvitamin D o40.9 ng ml −1, treatment with any antihypertensives, with ACEi or ARBs only, with any lipid-lowering drugs, with statins only and with warfarin. 'Not entered' refers to the fact that the variable did not reach the pre-specified significance level (5%) for addition into the final regression model.
aortic calcifications (in contrast to only standard dose of vitamin K1).9 To our knowledge, three human interventional studies were performed testing the efficacy of vitamin K1. Braam et al.10 reported that supplementation of vitamins K1 plus D resulted in favorable elastic properties of carotid artery compared with only-D vitamin monotherapy or placebo. Shea et al.11 studied the effect of K1 on coronary calcification. In a subgroup of subjects who had ⩾ 85% adherence to treatment and preexisting coronary calcifications, the progression of coronary calcification was 6% smaller than in the control group. Recently published 3-year intervention study by Knapen et al.35 on 244 healthy postmenopausal women, in which low-dose vitamin K2 (180 μg) was tested versus placebo, confirmed our results. The outcomes of this study clearly show an improved vascular elasticity in the K2-treated group, not only compared with placebo but also compared with the within-group baseline values. In the present study, we also investigated the potential interaction between dp-ucMGP and factors known to influence arterial stiffening. Despite the fact that age is the most powerful determinant of arterial stiffening, subjects younger than 50 years of age who had increased circulating dp-ucMGP levels had more than 11-fold higher risk of increased PWV. We also observed that Journal of Human Hypertension (2015), 1 – 6
dp-ucMGP predicts increased PWV preferentially in low-risk subjects, that is, in nonsmokers, in subjects without manifest vascular disease, hypertension, overt diabetes or moderate kidney insufficiency. This finding might be of potential clinical importance. Indeed, increased dp-ucMGP correctly identified subjects as 'high risk' who were considered as a 'normal risk' population by conventional approaches. As opposed to aortic stiffness, the stiffness of peripheral, muscular-type arteries (quantified as femoro-popliteal PWV) was not associated with dp-ucMGP. This finding might probably be explained by the different structure and physiological properties of central and peripheral arteries. Properties of muscular arteries are affected more by smooth muscle tone than by extracellular matrix structural components (like in central arteries)36—this is supported also by the fact that stiffness of distal arteries depends in our sample on age to far lesser extent than stiffness of central arteries. Study limitations First, our study had a cross-sectional design with all its limitations. Only a large randomized interventional trial with vitamin K © 2015 Macmillan Publishers Limited
Desphospho-uncarboxylated MGP and aortic stiffness O Mayer Jr et al
5 increased risk of PWV>9 m/sec
decreased risk of PWV>9 m/sec
1.69
age>50 years
11.72 1.91
male gender
2.07 3.89
history of vascular disease
1.69 2.05
current smoking
1.79 2.71
BMI>30 kg/m2
1.51 1.67
raised blood pressure
2.35 1.34
overt diabetes
1.83 0.75
creatinine clearance 30-60 mL/min
1.86 2.23
low D vitamin status
1.75
0.10
1.00
YES (factor present)
10.00
100.00
NO (factor absent)
Figure 2. Adjusted relative risk of increased dp-ucMGP on increased aortic PWV by subgroups of potential covariates. Multiple stepwise logistic regression with aPWV ⩾ 9 m s − 1 used as a dependent variable. The same independent variables, as in the Table 3 model, were included, but only odds ratios and 95% confidence intervals for increased dp-ucMGP (⩾671 pmol l − 1) were expressed. Black bars represents relative risk in subgroups with a cofactor present.
supplementation could finally elucidate the role of impaired vitamin K status in age-related stiffening of large arteries in a general population. Second, both PWV and vitamin K status (reflected by dp-ucMGP) were assessed using one single measurement. On the other hand, the strength of our data lies in a well-defined sample of relatively healthy subjects, who were examined using a standardized examination procedure of the established long-term epidemiological project (MONICA study). CONCLUSION In our sample from the general population, we confirmed that elevated dp-ucMGP was independently associated with high aPWV. Therefore, we propose that this factor may serve as a biomarker and potentially modifiable risk factor of vascular aging. Vitamin K supplementation (preferably with the K2 isoform) might form a promising concept how to selectively influence the agerelated changes in vessel wall and even global cardiovascular risk. Therefore, further research in this field is needed. What is already known about the topic ● Matrix Gla protein (MGP) is a natural inhibitor of calcification; vitamin K is the essential cofactor for its activation. ● MGP strongly correlates with the extent of coronary calcification, but its role in stiffening of large arteries has not yet been studied ● The nonphosphorylated-uncarboxylated isoform of MGP (dpucMGP) reflects the vitamin K status. What this study adds ● Elevated dp-ucMGP was independently associated with high aortic stiffness in the general population. ● Therapeutic manipulation of dp-ucMGP by vitamin K supplementation might form a new concept to influence age-related arterial stiffening.
ACKNOWLEDGEMENTS The clinical part of the project was supported by the Internal Grant Agency of Czech Ministry of Health (NT13186) and by the Charles University Research Fund (project number P36). We are grateful to IDS Plc., providing free of charge, newly developed
© 2015 Macmillan Publishers Limited
kits for estimation of dp-ucMGP. We would like to acknowledge the hard work of all investigators in the Czech Republic who participated in the post-MONICA project, and last but not least all laboratory technicians, realizing extent laboratory assessments for the present study.
CONFLICT OF INTEREST The authors declare no conflict of interest REFERENCES 1 O’Rourke MF, Mancia G. Arterial stiffness. J Hypertens 1999; 17: 1–4. 2 Laurent S, Cockcroft J, Van Bortel L, Boutouyrie P, Giannattasio C, Hayoz D et al. European network for non-invasive investigation of large arteries. Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J 2006; 27: 2588–2605. 3 Dao HH, Essalihi R, Bouvet C, Moreau P. Evolution and modulation of age-related medial elastocalcinosis: impact on large artery stiffness and isolated systolic hypertension. Cardiovasc Res 2005; 66: 307–317. 4 Fornieri C, Quaglino D Jr, Mori G. Role of the extracellular matrix in age-related modifications of the rat aorta. Ultrastructural, morphometric, and enzymatic evaluations. Arterioscler Thromb 1992; 12: 1008–1016. 5 Cattell MA, Anderson JC, Hasleton PS. Age-related changes in amounts and concentrations of collagen and elastin in normotensive human thoracic aorta. Clin Chim Acta 1996; 245: 73–84. 6 Bruel A, Oxlund H. Changes in biomechanical properties, composition of collagen and elastin, and advanced glycation endproducts of the rat aorta in relation to age. Atherosclerosis 1996; 127: 155–165. 7 Keeley FW, Partridge SM. Amino acid composition and calcification of human aortic elastin. Atherosclerosis 1974; 19: 287–296. 8 Demer LL, Tintut Y. Vascular calcification: pathobiology of a multifaceted disease. Circulation 2008; 117: 2938–2948. 9 Schurgers LJ, Spronk HM, Soute BA, Schiffers PM, DeMey JG, Vermeer C. Regression of warfarin-induced medial elastocalcinosis by high intake of vitamin K in rats. Blood 2007; 109: 2823–2831. 10 Braam LA, Hoeks AP, Brouns F, Hamulyak K, Gerichhausen MJ, Vermeer C. Beneficial effects of vitamins D and K on the elastic properties of the vessel wall in postmenopausal women: a follow-up study. Thromb Haemost 2004; 91: 373–380. 11 Shea MK, O'Donnell CJ, Hoffmann U, Dallal GE, Dawson-Hughes B, Ordovas JM et al. Vitamin K supplementation and progression of coronary artery calcium in older men and women. Am J Clin Nutr 2009; 89: 1799–1807. 12 Shanahan CM, Proudfoot D, Farzaneh-Far A, Weissberg PL. The role of Gla proteins in vascular calcification. Crit Rev Eukaryot Gene Expr 1998; 8: 357–375.
Journal of Human Hypertension (2015), 1 – 6
Desphospho-uncarboxylated MGP and aortic stiffness O Mayer Jr et al
6 13 Price PA, Urist MR, Otawara Y. Matrix Gla protein, a new gamma-carboxyglutamic acid-containing protein which is associated with the organic matrix of bone. Biochem Biophys Res Commun 1983; 117: 765–771. 14 Schurgers LJ, Cranenburg ECM, Vermeer C. Matrix Gla-protein: the calcification inhibitor in need of vitamin K. Thromb Haemost 2008; 100: 593–603. 15 Cranenburg ECM, Koos R, Schurgers LJ, Magdeleyns EJ, Schoonbrood TH, Landewé RB et al. Characterisation and potential diagnostic value of circulating matrix Gla protein (MGP) species. Thromb Haemost 2010; 104: 811–822. 16 Schurgers LJ, Spronk HM, Skepper JN, Hackeng TM, Shanahan CM, Vermeer C et al. Post-translational modifications regulate matrix Gla protein function: importance for inhibition of vascular smooth muscle cell calcification. J Thromb Haemost 2007; 5: 2503–2511. 17 van den Heuvel EG, van Schoor NM, Lips P, Magdeleyns EJ, Deeg DJ, Vermeer C et al. Circulating uncarboxylated matrix Gla protein, a marker of vitamin K status, as a risk factor of cardiovascular disease. Maturitas 2014; 77: 137–141. 18 Dalmeijer GW, van der Schouw YT, Vermeer C, Magdeleyns EJ, Schurgers LJ, Beulens JW. Circulating matrix Gla protein is associated with coronary artery calcification and vitamin K status in healthy women. J Nutr Biochem 2013; 24: 624–628. 19 Schurgers LJ, Barreto DV, Barreto FC, Liabeuf S, Renard C, Magdeleyns EJ et al. The circulating inactive form of matrix gla protein is a surrogate marker for vascular calcification in chronic kidney disease: a preliminary report. Clin J Am Soc Nephrol 2010; 5: 568–575. 20 Mayer O Jr, Seidlerová J, Bruthans J, Filipovský J, Timoracká K, Vaněk J et al. Desphospho-uncarboxylated matrix Gla-protein is associated with mortality risk in patients with chronic stable vascular disease. Atherosclerosis 2014; 235: 162–168. 21 Cífková R, Skodová Z, Bruthans J, Adámková V, Jozífová M, Galovcová M et al. Longitudinal trends in major cardiovascular risk factors in the Czech population between 1985 and 2007/8. Czech MONICA and Czech post-MONICA. Atherosclerosis 2010; 211: 676–681. 22 Laurent S, Boutouyrie P, Asmar R, Gautier I, Laloux B, Guize L et al. A Aortic stiffness is an independent predictor of all-cause and cardiovascular mortality in hypertensive patients. Hypertension 2001; 37: 1236–1241. 23 Vlachopoulos C, Aznaouridis K, Stefanadis C. Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and meta-analysis. J Am Coll Cardiol 2010; 55: 1318–1327. 24 Williams B, Lacy PS, Thom SM, Cruickshank K, Stanton A, Collier D et al. Differential impact of blood pressure-lowering drugs on central aortic pressure and clinical outcomes: principal results of the Conduit Artery Function Evaluation (CAFE) study. Circulation 2006; 113: 1213–1225.
Journal of Human Hypertension (2015), 1 – 6
25 Luo G, Ducy P, McKee MD, Pinero GJ, Loyer E, Behringer RR et al. Spontaneous calcification of arteries and cartilage in mice lacking matrix GLA protein. Nature 1997; 386: 78–81. 26 Krüger T, Oelenberg S, Kaesler N, Schurgers LJ, van de Sandt AM, Boor P et al. Warfarin induces cardiovascular damage in mice. Arterioscler Thromb Vasc Biol 2013; 33: 2618–2624. 27 Wang Y, Chen J, Zhang Y, Yu W, Zhang C, Gong L et al. Common genetic variants of MGP are associated with calcification on the arterial wall but not with calcification present in the atherosclerotic plaques. Circ Cardiovasc Genet 2013; 6: 271–278. 28 Tuñón-Le Poultel D, Cannata-Andía JB, Román-García P, Díaz-López JB, Coto E, Gómez C et al. Association of matrix Gla protein gene functional polymorphisms with loss of bone mineral density and progression of aortic calcification. Osteoporos Int 2014; 25: 1237–1246. 29 Yoshikawa K, Abe H, Tominaga T, Nakamura M, Kishi S, Matsuura M et al. Polymorphism in the human matrix Gla protein gene is associated with the progression of vascular calcification in maintenance hemodialysis patients. Clin Exp Nephrol 2013; 17: 882–889. 30 Beulens JW, Bots ML, Atsma F, Bartelink ML, Prokop M, Geleijnse JM et al. High dietary menaquinone intake is associated with reduced coronary calcification. Atherosclerosis 2009; 203: 489–493. 31 Jie KS, Bots ML, Vermeer C, Witteman JC, Grobbee DE. Vitamin K intake and osteocalcin levels in women with and without aortic atherosclerosis: a population-based study. Atherosclerosis 1995; 116: 117–123. 32 Villines TC, Hatzigeorgiou C, Feuerstein IM, O'Malley PG, Taylor AJ. Vitamin K1 intake and coronary calcification. Coron Artery Dis 2005; 16: 199–203. 33 Geleijnse JM, Vermeer C, Grobbee DE, Schurgers LJ, Knapen MH, van der Meer IM et al. Dietary intake of menaquinone is associated with a reduced risk of coronary heart disease: the Rotterdam Study. J Nutr 2004; 134: 3100–3105. 34 Gast GC, de Roos NM, Sluijs I, Bots ML, Beulens JW, Geleijnse JM et al. A high menaquinone intake reduces the incidence of coronary heart disease. Nutr Metab Cardiovasc Dis 2009; 19: 504–510. 35 Knapen MH, Braam LA, Drummen NE, Bekers O, Hoeks AP, Vermeer C. Menaquinone-7 supplementation improves arterial stiffness in healthy postmenopausal women: double-blind randomised clinical trial. Thromb Haemost 2015; 113: 1135–1144. 36 Mitchell GF, Parise H, Benjamin EJ, Larson MG, Keyes MJ, Vita JA et al. Changes in arterial stiffness and wave reflection with advancing age in healthy men and women: the Framingham Heart Study. Hypertension 2004; 43: 1239–1245.
© 2015 Macmillan Publishers Limited
