Increased Dephospho-uncarboxylated Matrix Gla-Protein Is Associated With Lower Axial Skeletal Muscle Mass in Patients With Hypertension

Abstract

Background

Matrix Gla-protein (MGP) is a well-established inhibitor of vascular calcification that is activated by vitamin K-dependent carboxylation. In the setting of vitamin K2 deficiency, dephospho-uncarboxylated MGP (dpucMGP) levels increase, and have been associated with large artery stiffening. Vitamin K2 is also a mitochondrial electron carrier in muscle, but the relationship of vitamin K2 deficiency and dpucMGP with muscle mass is not well understood. We therefore aimed to examine the association of vitamin K2 deficiency and dpucMGP with skeletal muscle mass in patients with hypertension.

Methods

We studied 155 hypertensive adults without heart failure. Axial skeletal muscle mass was measured using magnetic resonance imaging from axial steady-state free precession images. DpucMGP was measured with ELISA. Carotid–femoral pulse wave velocity (CF-PWV) was measured from high-fidelity arterial tonometry recordings.

Results

We found an inverse relationship between dpucMGP levels and axial muscle mass, with progressively rising dpucMGP levels correlating with decreasing axial muscle mass. In an unadjusted linear regression model, correlates of dpucMGP included axial skeletal muscle area factor (β = −0.32; P < 0.0001) and CF-PWV (β = 0.31; P = 0.0008). In adjusted analyses, independent correlates of dpucMGP included axial skeletal muscle area factor (β = −0.30; P = 0.0003) and CF-PWV (β = 0.20; P = 0.019).

Conclusions

In hypertensive adults, dpucMGP is independently associated with lower axial muscle mass, in addition to increased large artery stiffness. Further studies are required to investigate the role of vitamin K supplementation in this population.

Graphical Abstract

Graphical Abstract.

Matrix Gla-protein (MGP) has a well-established role as an inhibitor of vascular calcification, and is activated by vitamin K-dependent carboxylation.¹ In vitamin K-deficient states (particularly, vitamin K2 deficiency), carboxylation is reduced, resulting in increased circulating levels of dephospho-uncarboxylated MGP (dpucMGP). Increased levels of dpucMGP have been associated with increased large artery stiffness in disease states such as heart failure, diabetes, and primary hyperaldosteronism.^1–5 Interestingly, vitamin K2 is also a mitochondrial electron carrier in muscle.⁶

Reduced skeletal health, particularly reduced muscle mass, is increasingly recognized as an important component of aging and a determinant of prognosis.⁷ Moreover, a reduced muscle mass has been shown to be associated with hypertension in older adults.⁸ However, common biologic mechanisms that may underlie hypertension and sarcopenia are unknown. Given the role of dpucMGP in large artery stiffening (which causes an increased pulse pressure and isolated systolic hypertension), as well as its role in mitochondria, we aimed to test the hypothesis that circulating dpucMGP (a surrogate of vitamin K2 deficiency) is associated not only with increased large artery stiffness, but also with reduced muscle mass in patients with hypertension.

METHODS

We studied 155 adults with hypertension, defined as >130/80 mm Hg per the 2017 Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure,⁹ enrolled in a cohort study of patients referred for a cardiac magnetic resonance study at the Corporal Michael J. Crescenz VA Medical Center. We excluded participants with a history of heart failure or the absence of hypertension. The study protocol was approved by the Philadelphia VA Medical Center Institutional Review Board and all subjects provided informed consent. The cardiac magnetic resonance protocol included sagittal and axial stacks of the chest and upper abdomen. Axial skeletal muscle mass was measured from axial steady-state free precession images using previously published methods.¹⁰ Briefly, thoracic skeletal muscles were traced bilaterally on axial images, and cross-sectional areas were obtained. Factor analysis was performed and we identified a latent factor that underlies the shared variability in the cross-sectional area of the muscles studied in the analysis, and this factor was utilized as a continuous measurement of axial skeletal muscle mass.¹¹ DpucMGP was measured with ELISA (VitaK, The Netherlands). Carotid–femoral pulse wave velocity (CF-PWV), the gold standard noninvasive test for large artery stiffness, was measured in a subset of participants (n = 112) with high-fidelity tonometry recordings of the carotid and femoral pulses, using the R wave of the electrocardiogram as a fiducial point (SphygmoCor Px Device, AtCor Medical, Lisle, IL).

Continuous variables are presented as mean ± SD or median (interquartile range [IQR]) and compared across tertiles of dpucMGP using analysis of variance or the Kruskal–Wallis test, as appropriate. Categorical variables are presented as frequencies and percentages and compared with the chi-square or Fisher’s exact text. Linear regression was performed to assess the relationship between dpucMGP levels and various parameters. Standardized regression coefficients along with 95% confidence intervals were computed. Box–Cox transformations were applied when appropriate to normalize the distribution of parameters and/or model residuals during analyses. All statistical tests were 2 sided. A P value threshold of 0.05 was used to define statistical significance. Analyses were performed using the Matlab statistics and machine learning toolbox (the Mathworks, Natwick, MA).

RESULTS

The characteristics of the study population were as follows: age: 62.4 ± 12.2 years; male sex: 89.6%; Caucasian: 53.5%; African-American: 39.4%; other race/ethnicity: 7.1%; diabetes mellitus: 41.9%; coronary artery disease: 26.1%; estimated glomerular filtration rate (eGFR) <60 ml/min/1.73 m²: 84.4%; systolic blood pressure: 143 mm Hg (IQR: 134–157); diastolic blood pressure: 84 mm Hg (IQR: 78–91); and CF-PWV: 10.4 ± 3.6 m/s.

Across tertiles of dpucMGP, there was a progressive increase in the prevalence of male sex (lowest tertile: 76.9%; middle tertile: 94.1%; highest tertile: 98.1%; P = 0.0008). Body mass index also increased with higher levels of dpucMGP (lowest tertile: 25.9 kg/m² [IQR: 24.5, 35.0]; middle tertile: 29.9 kg/m² [IQR: 26.0, 34.2]; highest tertile: 31 kg/m² [IQR = 28.9, 36.3]; P < 0.0061). Other characteristics, including race, systolic or diastolic blood pressure, the prevalence of coronary artery disease, current smoking status, chronic obstructive pulmonary disease, diabetes mellitus, eGFR, or medication use, were similar across all tertiles of dpucMGP levels. Figure 1a shows the mean dpucMGP levels in the lowest, middle, and highest tertiles of axial skeletal muscle mass. Overall, we found an inverse relationship between dpucMGP levels and axial muscle mass, with progressively rising dpucMGP levels correlating with decreasing axial muscle mass. There were significant differences in dpucMGP levels across all 3 tertiles of muscle mass (analysis of variance, P = 0.0005). In post hoc pairwise comparisons, significant differences in dpucMGP levels were found between the lowest and highest tertiles of muscle mass (614 vs. 325 pmol/l, P = 0.0005) and between the middle and highest tertiles of muscle mass (527 vs. 325 pmol/l, P = 0.012).

Figure 1.

(a) Circulating levels of dephospho-uncarboxylated matrix Gla-protein (dpucMGP) across tertiles of axial skeletal muscle area latent factor; (b) volcano plot showing the correlates of dpucMGP. The significance level (alpha = 0.05) is shown by the horizontal dashed line; nonsignificant correlates (blue circles) are not labeled for simplicity (see text); (c) linear regression model showing the independent correlates of dpucMGP; standardized regression coefficients and 95% confidence intervals are shown. Abbreviations: CAD, coronary artery disease; CF-PWV, carotid–femoral pulse wave velocity; DBP, diastolic blood pressure; GFR, glomerular filtration rate.

Correlates of dpucMGP assessed via linear regression are shown in Figure 1b as a volcano plot. We examined the general clinical characteristics mentioned above, as well as CF-PWV and the axial skeletal muscle area factor. The volcano plot shows standardized regression coefficients and −log10 (P values) for each examined variable in unadjusted analyses, highlighting statistically significant correlates of dpucMGP. Among general clinical characteristics, significant correlates of dpucMGP included age (β = 0.286; P = 0.0003), African-American ethnicity (β = −0.40; P < 0.0001), history of coronary artery disease (β = 0.20; P = 0.014), aspirin use (β = 0.20; P = 0.014), eGFR (β = −0.40; P < 0.0001), and diastolic blood pressure (β = −0.17; P = 0.034). In addition, in these unadjusted analyses, CF-PWV (β = 0.31; P = 0.0008) and axial skeletal muscle area factor (β = −0.32; P < 0.0001) were significant correlates of dpucMGP.

In a linear regression model that included skeletal muscle mass and CF-PWV as correlates of dpucMGP, both were independently associated with dpucMGP. In this model, CF-PWV was positively associated with dpucMGP (β = 0.24; P = 0.006) whereas axial skeletal muscle mass was negatively associated with dpucMGP (β = −0.31; P = 0.006).

We also built a linear regression model that included axial skeletal muscle mass, CF-PWV and all other significant clinical correlates of dpucMGP (Figure 1c). In this model, significant independent correlates of dpucMGP included CF-PWV (β = 0.20; P = 0.019), axial skeletal muscle mass (β = −0.30; P = 0.0003), African-American ethnicity (β = −0.22; P = 0.005), and eGFR (β = −0.44; P < 0.0001).

Discussion

In this study of a cohort of 155 hypertensive patients without heart failure, we measured dpucMGP levels and axial skeletal muscle areas, and assessed the relationship of dpucMGP levels with a variety of clinical variables. We found that dpucMGP levels are inversely related to axial skeletal muscle mass, and found that progressively increasing dpucMGP levels correlated with reductions in axial skeletal muscle mass. In unadjusted linear regression analyses, we found that dpucMGP levels were independently correlated with axial skeletal muscle mass, CF-PWV, age, African-American ethnicity, history of coronary artery disease, aspirin use, eGFR, and diastolic blood pressure. In adjusted analyses, axial skeletal muscle mass, CF-PWV, African-American ethnicity, and eGFR were all independent correlates of dpucMGP. Overall, our findings suggest that dpucMGP is an independent correlate of axial skeletal muscle mass, even after adjusting for various confounders, in patients with hypertension.

DpucMGP is traditionally thought of as being a marker of vascular vitamin K2 deficiency leading to arterial calcification and arterial stiffness. Our analyses support this relationship given the positive association of CF-PWV with dpucMGP in both unadjusted and adjusted analyses, which has been reported previously in various populations. However, our results indicate for the first time, a further relationship of vitamin K2 deficiency and dpucMGP with axial skeletal muscle mass. Whereas the mechanism of this association cannot be established from the current study, these findings are consistent with the important role of vitamin K2 in skeletal muscle mitochondria.¹² Whether vitamin K2 supplementation may have a role in enhancing mitochondrial function and/or preventing reductions in muscle mass cannot be addressed by our observational study, and should be the focus of future research.⁶

Our study should be considered in the context of its strengths and limitations. Strengths of this study include a well-characterized sample, which allowed adjustment with several possible confounders, as well as the concomitant measurement of large artery stiffness and axial skeletal muscle areas. Limitations include the cross-sectional study design, which can provide evidence of association but not causation. Second, our population included mostly male patients, which is reflective of the patient demographics at this VA Medical Center. Third, we defined hypertension as >130/80 mm Hg, and further studies are required to investigate whether these findings can be specifically applied to patients with stage II hypertension. Finally, we did not measure other laboratory tests indicative of nutritional status, which may also be associated with reductions in axial muscle mass.

In conclusion, our findings suggest that there is an inverse relationship between dpucMGP levels and axial muscle mass, with higher dpucMGP levels correlating with reductions in axial muscle mass in hypertensive patients. Our study is the first to provide evidence that dpucMGP is an independent correlate of axial muscle mass in hypertensive patients. These results suggest that vitamin K2 deficiency and increased circulating dpucMGP levels may lead to reductions in axial muscle mass in patients with hypertension, and could potentially be targeted with vitamin K supplementation, which should be investigated in future studies.

DISCLOSURE

J.A.C. is supported by NIH grants R01-HL 121510, R33-HL- 146390, R01HL153646, R01-AG058969, 1R01-HL104106, P01-HL094307, R03-HL146874, and R56-HL136730. He has recently consulted for Bayer, Sanifit, Fukuda-Denshi, Bristol Myers Squibb, JNJ, Edwards Life Sciences, Merck, and the Galway-Mayo Institute of Technology. He received University of Pennsylvania research grants from the NIH, Fukuda-Denshi, Bristol Myers Squibb, and Microsoft. He is named as an inventor in a University of Pennsylvania patent for the use of inorganic nitrates/nitrites in Heart Failure with Preserved Ejection Fraction. He has received research device loans from Atcor Medical, Fukuda-Denshi, Uscom, NDD Medical Technologies, Microsoft, and MicroVision Medical.