The Association of Serum Osteocalcin Levels with Metabolic Parameters and Inflammation in Postmenopausal Women with Metabolic Syndrome

Abstract

Introduction:

Although adipose tissue largely plays a role in the etiopathogenesis of metabolic syndrome (MS), which is an inflammatory process, the skeleton may also contribute to this process through osteocalcin (OC), which is a bone-derived protein. In this study, we aimed to evaluate OC levels in postmenopausal women with MS and to investigate the association of OC levels with the metabolic and inflammatory parameters.

Methods:

Thirty-five postmenopausal women diagnosed with MS were recruited for the study. Sixteen postmenopausal women without any of the MS criteria formed the control group. Body weight, height, and waist and hip circumference of all of the subjects were measured and body mass indices (BMIs) were calculated. Levels of serum glucose, insulin, C-peptide, triglyceride, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, albumin, creatinine, calcium, phosphorus, total alkaline phosphatase, parathormone, and as inflammatory parameters, erythrocyte sedimentation rate, fibrinogen, and high-sensitive C-reactive protein (hsCRP) were studied from fasting venous blood samples of all the subjects. Homeostatic model assessment for insulin resistance (HOMA-IR) was calculated. Serum total OC levels were studied from all of the subjects. Bone mineral densities were also measured.

Results:

Serum OC levels of the group with MS median (5.37 ng/mL) were lower than the OC levels of the group without MS (P < 0.01). Serum OC levels significantly and negatively correlated with fasting blood glucose (r = −0.310, P < 0.05), insulin (r = −0.343, P < 0.05), and HOMA-IR (r = −0.384, P < 0.01) values. Serum OC levels showed a significant and negative correlation with body weight (r = −0.293, P < 0.05), BMI (r = −0.333, P < 0.05), and waist-to-hip ratio (r = −0.384, P < 0.05). The inflammatory markers in the patient group were significantly higher than the control group. We found a negative association between serum OC levels and hsCRP levels in all cases (r = −0.283, P < 0.05).

Conclusion:

In the presence of MS, OC levels are significantly low and display a close association with glucose metabolism and adipose tissue. In addition, OC may play a regulatory role in subclinical systematic inflammatory response.

Introduction

Metabolic syndrome (MS) is increasingly drawing more attention and is one of the most studied subjects in the literature with worldwide increase in the prevalence of obesity today. Both MS itself and each of its components carry an increased risk for cardiovascular complications. The main pathophysiological characteristic of MS is increased insulin resistance, and it creates a proinflammatory and prothrombotic environment.¹

In recent years, it has been suggested that not only adipose tissue but also skeletal tissue may contribute to the components of MS, which is an inflammatory process. Among skeleton-derived products, the osteocalcin (OC) protein has received the most attention. OC is the product of osteoblasts and makes up the majority of proteins that are not collagen in the bone.²

Genetic cell-based and biochemical analyses have shown that OC stimulated insulin and adiponectin expression by beta cell proliferation. In animal-based studies, an abnormal increase in visceral fat, obesity, and glucose intolerance and a decrease in beta cell proliferation are demonstrated in OC knockout mice, and an improved glucose tolerance and an improved insulin secretion are observed when OC is given to these mice.³ This suggests that the skeleton may be effective in regulating energy metabolism, and that OC may have a potential effect on obesity development. If this novel potential role of OC in energy metabolism can be understood better, it may assist the prevention and treatment of obesity and diabetes in humans.

There are very few clinical studies analyzing the association of OC with MS parameters. In the present study, we aimed to analyze serum OC levels in postmenopausal women with MS components and to determine their relationship with metabolic and inflammatory markers.

Methods

Thirty-five female patients diagnosed with MS and 16 female patients who had none of the criteria of MS, according to the 2005 International Diabetes Federation (IDF) guidelines, were included into the study. All of the patients were selected among women who were in the postmenopausal period to standardize the effects of bone metabolism and turnover. We took into consideration that the patient group and the control group were all euthyroid and that the study consisted of subjects who did not use any medication affecting bone turnover and who had no signs of infection. All of the women were in the natural postmenopausal period and none of them was receiving hormone replacement therapy or any other medication for osteoporosis treatment that could alter the serum OC levels.

The study protocol was approved by the Ethics Committee of Eskisehir Osmangazi University Medical Faculty with its resolution dated September 2, 2010, and numbered 2010/166. The participants of the study were informed about the purpose and the method of the study. Written informed consent was obtained from all of the subjects in the patient and the control groups.

The height, weight, and waist and hip circumference of all the subjects were measured. Their body mass indices (BMIs) were calculated as kg/m². Venous blood samples were taken from the antecubital vein following an at least 10 hr-fasting overnight from the subjects. From these blood samples, the erythrocyte sedimentation rate (ESR) was measured by the Western grain method using the VACUETTE SRS 100 device, fibrinogen by the Clothing method using the Beckman Coulter ACL TOP device and HemosIL fibrinogen C XL kits, and high-sensitive C-reactive protein (hsCRP) by the nephelometry method using ProSpec and Siemens CardioPhase kit. Serum triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C) levels were assayed by enzymatic colorimetry, creatinine by Jaffe (kinetic colorimetry), albumin and total alkaline phosphatase (ALP) by immunoturbidimetry, calcium and phosphorus by colorimetric endpoint, glucose by enzymatic colorimetric method using the Roche Hitachi Modular autoanalyzer, and insulin, parathormone (PTH), and C-peptide by the solid-phase two-site (sequential chemiluminescence immunoassay) method using Siemens Immulite 2000.

Homeostatic model assessment for insulin resistance (HOMA-IR) values were calculated to determine the presence of insulin resistance. Glucose (mg/dL) × insulin × (μU/mL)/405 formula was used to calculate HOMA-IR.⁴

Bone mineral density (BMD) was measured from lumbar vertebrae and femoral neck by X-ray absorptiometry (DXA) method using the Hologic 4500 WQDR device in all of the subjects in the patient and control groups. T and Z scores were recorded.

Venous blood samples for serum OC levels from all of the subjects were taken into centrifuge tubes containing no anticoagulants. Samples were centrifuged at +4°C at 4000 rpm for 3 min. Obtained serum samples were assayed by the chemiluminescence immunometric method (Immulite, 2000; SIEMENS).

Statistical analysis

SPSS for Windows 16.6 package program (SPSS, Inc., Chicago, IL) was used for the statistical analysis of data. To determine the distribution forms of variables (whether or not they are normally distributed), Shapiro–Wilk's test was used. Data are expressed as mean ± SD. For intergroup comparison of variables that are normally distributed, the independent-samples t test was used. On the contrary, for intergroup comparison of variables that are not normally distributed, the Mann–Whiney U test was used. In determining the relationships among variables, the Pearson and Spearman correlation coefficients were calculated. P < 0.05 was considered statistically significant.

Results

While the age range of the patient group was 39–71 (55.31 ± 7.0), the age range of the control group was 42–62 (52.93 ± 4.76). There was no statistically significant difference between the ages of the groups. Nineteen of the participants from the patient group were diagnosed with type 2 diabetes mellitus (DM).

When the patient group was compared with the control group, there was a statistically significant difference between the groups in terms of body weight, height, BMI, waist circumference, waist/hip ratio, and systolic and diastolic blood pressure. The values of these variables were found to be significantly higher in the patient group with MS compared with the control group (P < 0.001, P < 0.05, P < 0.001, P < 0.001, P < 0.001, P < 0.01, P < 0.001, and P < 0.001, respectively). The anthropometric and clinical data of the patient group and the control group are shown in Table 1.

Table 1.

The Anthropometric and Clinical Data of the Patient and Control Groups

	Patient group, (n:35)	Control group, (n:16)	P
Age (year)^a	55.31 ± 7.0 (39–71)	52.93 ± 4.76 (42–62)	>0.05
Body weight (kg)^a	83.95 ± 14.73 (56–110)	67.25 ± 6.70 (58–80)	<0.001
Body mass index (kg/m²)^a	34.41 ± 6.12 (23–47)	25.73 ± 2.90 (20–25)	<0.001
Waist (cm)^a	103.48 ± 12.84 (75–125)	85.25 ± 9.62 (71–99)	<0.001
Hip (cm)^a	116.57 ± 12.72 (98–145)	103.87 ± 7.32 (93–118)	<0.001
Waist/hip^a	0.88 ± 0.07 (0.73–1.04)	0.81 ± 0.07 (0.66–0.94)	<0.01
Systolic BP (mmHg)^b	140 (90–200)	110 (80–120)	<0.001
Diastolic BP (mmHg)^b	90 (50–110)	77.5 (60–90)	<0.001

Data which are statistically significant are in bold.

Parametric data are shown as mean ± SD. For the intergroup comparison, independent-samples t test is used.

Nonparametric data are shown as median (minimum–maximum). For the intergroup comparison, the Mann–Whitney U test is used.

In the patient group, serum glucose, insulin, C-peptide, and HOMA-IR values were found to be significantly higher compared with the control group (P values <0.001, <0.01, <0.01, and <0.01, respectively). When lipid profile was analyzed, there was no significant difference in serum LDL-C levels (P > 0.05) between the patient group and the control group. However, in the patient group, HDL-C levels were significantly lower (P < 0.001), and TG levels were significantly higher (P < 0.001) (Table 2).

Table 2.

Biochemical Parameters and Bone Mineral Density Data of the Patient and Control Groups

	Patient group, (n:35)	Control group, (n:16)	P
Glucose (mg/dL)^a	114 (76–308)	90 (76–99)	<0.001
Insulin (μU/mL)^a	11.10 (3.43–130)	6.25 (2.77–18.6)	<0.01
C peptide (ng/mL)^a	2.65 (0.26–8.27)	1.69 (1.02–3.34)	<0.01
HOMA-IR^a	2.86 (0.77–81)	1.38 (0.62–7.33)	<0.01
HDL-C (mg/dL)^b	44.42 ± 8.78 (28–66)	56 ± 11.14 (40–74)	<0.001
LDL-C (mg/dL)^b	128.20 ± 37.02	129.06 ± 43.97	>0.05
Triglyceride (mg/dL)^a	169 (64–681)	94.50 (45–145)	<0.001
Creatinine (mg/dL)^a	0.70 (0.5–1.0)	0.80 (0.7–1.0)	>0.05
Albumin (g/dL)^b	4.61 ± 0.23	4.63 ± 0.27	>0.05
Calcium (mg/dL)^b	9.58 ± 0.52	9.83 ± 0.48	>0.05
Phosphorus (mg/dL)^b	3.68 ± 0.49	3.82 ± 0.45	>0.05
ALP (U/L)^b	213.17 ± 52.63	187.93 ± 24.89	>0.05
PTH (pg/mL)^a	52.60 (29.5–66)	60.95 (35.1–66.9)	>0.05
Fibrinogen (mg/dL)^b	341.71 ± 47.53	313.56 ± 40.18	<0.05
ESR (mm/hr)^a	15 (6–45)	10.50 (2–22)	<0.05
hsCRP (mg/L)^a	5.07 (0–38)	0.2 (0–6.21)	<0.01
Osteocalcin (ng/mL)	5.37 (1.9–21.1)	9.28 (2.88–13.9)	<0.01
L1–L4 T score^b	−1.76 ± 1.7	−1.60 ± 0.89	>0.05
Femoral neck T score^a	−1.19 (−2.87–1.62)	−1.37 (−2.18–0.31)	>0.05

Data which are statistically significant are in bold.

Nonparametric data are shown as median (minimum–maximum). For the intergroup comparison, the Mann–Whitney U test is used.

Parametric data are shown as mean ± SD. For the intergroup comparison, the independent-samples t test is used.

ALP, alkaline phosphatase; ESR, erythrocyte sedimentation rate; HDL-C, high-density lipoprotein cholesterol; HOMA-IR, homeostatic model assessment for insulin resistance; hsCRP, high-sensitive C-reactive protein; LDL-C, low-density lipoprotein cholesterol; PTH, parathormone.

There was no significant difference in serum calcium, phosphorus, ALP, and PTH levels between the patient and the control groups (P > 0.05). Bone biomarkers and BMD measurement values of the patient and the control groups are shown in Table 2.

When the inflammatory markers were evaluated, fibrinogen levels, ESR, and hsCRP levels in the patient group were significantly higher than the inflammatory parameters in the control group (P values <0.05, <0.05, and <0.01, respectively) (Table 2). Serum OC levels in the patient group were significantly lower than the serum OC level in the control group (P < 0.01) (Table 2),

In the patient group, while OC was negatively correlated with HOMA-IR and insulin levels (respectively, r = −0.386, P < 0.05; r = −0.359, P < 0.05), it was positively correlated with ALP, one of the bone markers, (r = 0.336, P < 0.05). In all cases, serum OC levels were negatively associated with fasting blood glucose (r = −0.310, P < 0.05), insulin (r = −0.343, P < 0.05), and HOMA-IR (r = −0.384, P < 0.01). Again, in all cases, serum OC levels had a negative relationship with body weight (r = −0.283, P < 0.05), BMI (r = −0.333, P < 0.05), and waist/hip ratio (r = −0.293, P < 0.05).

In the patient group, there was a statistically significant difference between serum OC levels of the cases with diabetes and serum levels of the cases without diabetes (P < 0.01). Serum OC levels of the cases with diabetes were found to be lower. The mean serum OC levels of the patients with diabetes were 4.59 ± 2.47 ng/mL, and in the cases without diabetes, the mean serum OC levels were 8.13 ± 4.65 ng/mL.

When we evaluated inflammatory markers, we found a significantly negative association between OC levels and hsCRP levels in both the patient and the control groups. In all cases, we observed that fibrinogen was positively related with BMI (r = 0.411, P < 0.01) and waist circumference (r = 0.421, P < 0.01), and that the ESR was positively related with body weight (r = 0.340, P < 0.05), BMI (r = 0.415, P < 0.01), waist circumference (r = 0.442, P < 0.001), hip circumference (r = 0.371, P < 0.05), waist/hip ratio (r = 0.277, P < 0.05), serum glucose (r = 0.277, P < 0.05), fibrinogen (r = 0.445, P < 0.001), and hsCRP (r = 0.376, P < 0.01) levels.

Similarly, hsCRP levels were positively associated with body weight (r = 0.435, P < 0.001), BMI (r = 0.460, P < 0.001), waist circumference (r = 0.419, P < 0.01), hip circumference (r = 0.325, P < 0.05), waist/hip ratio (r = 0.321, P < 0.05), systolic blood pressure (r = 0.297, P < 0.05), diastolic blood pressure (r = 0.327, P < 0.05), HOMA-IR (r = 0.328, P < 0.05), TG (r = 0.494, P < 0.001), fibrinogen (r = 0.390, P < 0.01), and ESR (r = 0.376, P < 0.01).

Discussion

In this study, we compared serum OC levels of the postmenopausal women with MS with the postmenopausal women with none of the MS criteria and evaluated the relationship of OC with metabolic and inflammatory markers. Serum OC levels of the postmenopausal patients with MS were found to be significantly higher than the patients without MS criteria. We also showed that serum OC levels had a significant negative association with glucose metabolism and markers of adipose tissue and inflammation.

MS is an inflammatory condition in which visceral adipose tissue plays an important role. However, there are also some data suggesting that the skeleton may contribute to this process through OC, a bone- derived protein, as well.^5
–7

In vivo studies demonstrate that insulin secretion increases when pancreas β cells are cultured together with osteoblasts. Accordingly, it is argued that an osteoblast-driven factor in the circulation regulates β cell function.⁸ It was demonstrated that in OC knockout mice, glucose intolerance develops, insulin secretion is impaired, and insulin resistance occurs, and when recombinant OC was given to these mice, insulin secretion was found to be increased.^3,8

In the present study, we found serum OC levels in postmenopausal women with MS to be significantly lower compared with women without MS. In the literature, there are very few studies on how OC levels change in different clinical situations in which insulin resistance takes part in the pathogenesis. There are studies showing that serum OC levels are lower in cases with type 2 DM compared with cases without.^5,7,9 In this study, when we classified the patient group with MS into subgroups according to the presence of type 2 DM, we found that serum OC levels in the patients with type 2 DM were lower than serum OC levels in the patients without type 2 DM. Consistent with this finding, very few studies on patients with MS report that OC levels are lower in patients with MS compared with normal subjects.¹⁰

Although serum OC levels in the patient group were found to be lower compared with those in the control group, no statistically significant difference was found in ALP levels, also a bone formation marker, between the two groups. While OC levels showed a positive correlation with serum ALP levels in the study, there was no difference in ALP levels between the groups. This makes us think that the skeleton specifically plays a role in MS through OC.

In our study, there was a negative correlation between serum OC levels and parameters of glucose metabolism. In a study by Pittas et al., it was reported that serum OC levels were inversely associated with fasting plasma glucose, insulin, and HOMA-IR levels.⁶ They argue that this inverse association was connected with the decrease in adiponectin secretion. In another study on obese children, it was found that serum OC levels were significantly and negatively associated with HOMA-IR and in the follow-up period when those children were treated with diet and exercise, the change in OC showed an inverse association with HOMA-IR.¹¹

There are also studies that report a negative relationship of OC levels with fasting plasma glucose, insulin levels, and HOMA-IR values in patients with MS.^12
–14 Our findings also confirm this relationship of OC with glucose metabolism and insulin resistance.

In animal studies conducted in OC knockout mice, it was demonstrated that body mass and adipose cell number increased, and these mice became abnormally obese.³ In a study conducted on obese women, it was reported that OC levels significantly increased in women who lost large amounts of weight following a diet and exercise therapy.¹⁴ In another study, OC levels were found to be lower in obese postmenopausal women compared with normal-weight women, and serum OC and body weight were reported to be negatively correlated .^15,16 We also found OC levels to be negatively correlated with body weight, BMI, and waist/hip ratio. This finding is a clinical proof showing that OC decreases in the case of obesity.

OC is a protein specifically secreted by osteoblasts and exposed to posttranslational modification by vitamin K-mediated γ carboxylation. Vitamin K was shown to reduce proinflammatory cytokine production and display an anti-inflammatory effect.¹⁷

It can be argued that OC also indirectly has anti-inflammatory effects. When the relationship between vitamin K and proinflammatory biomarkers in the circulation was investigated, no relationship between uncarboxylated OC percent and inflammatory parameters was found.¹⁸ Accordingly, it was argued that vitamin K modulated inflammation by a different mechanism other than γ carboxylation. In another study, vitamin K suppressed inflammation by regulating nuclear factor κB in macrophages at the gene level when bone forming functions of OC increased.¹⁹ There are very few studies on the role of OC in inflammation in the literature.

Pittas et al., in their study on individuals with dysmetabolic phenotype, found that serum OC levels negatively correlated with serum hs-CRP and IL-6 levels.⁶ Based on this, they claimed that OC might affect other components of dysmetabolic phenotype such as inflammation and adiposity as OC played a role in the regulation of insulin sensitivity.

In the present study, fibrinogen, ESR, and hsCRP levels in the patients with MS were found to be significantly higher compared with the control group and showed a negative association with adiposity indices, glucose metabolism, and lipids. Accordingly, our findings support the view that MS is an inflammatory disease. In addition, as we found that serum OC levels were significantly lower in patients with MS compared with patients without MS, we think that OC levels may change in an inflammatory situation or that the OC level itself may be effective in the control of inflammation.

While we found no relationship between OC levels and hsCRP levels in the patients with MS, we found a negative association in both the patient and the control groups. This makes us to think that OC counteracts with inflammation, independent of its effects on glucose and adipose metabolism.

There are some limitations of our study. First of all, our study is cross sectional in nature, includes only postmenopausal women, and the study population is small. Also, as there are reports suggesting that the undercarboxylated form of OC was the main metabolically active form, it would be better to study the undercarboxylated form of OC as well.²⁰ However, we could only measure the total OC levels in our study.

In conclusion, in the present study, we found that serum OC levels were low in postmenopausal women with MS, which itself is a chronic inflammatory process. We also found that OC was negatively associated with body weight, BMI, waist/hip ratio, fasting blood glucose, insulin, and HOMA-IR. According to these findings, OC, and thereafter the bone tissue, appears to be effective in the regulation of glucose metabolism and body adiposity.

Accordingly, we can claim that OC plays a role in the etiopathogenesis of MS. Apart from this, OC may be a protein that plays a role in the inflammatory process, other than its roles in glucose and adipose mechanisms. Studies on different clinical models, which will evaluate the cause and effect relationship as to whether or not OC can be used as an inflammation determinant or a therapeutic agent, need to be conducted more comprehensively.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

References

Grundy

. Metabolic syndrome: A multiplex cardiovascular risk factor. J Clin Endocrinol Metab, 2007; 92:399–404.

Komori

. What is the function of osteocalcin?. J Oral Biosci, 2020; 62:223–227.

Lee

, Sowa

, Hinoi

, et al. Endocrine regulation of energy metabolism by the skeleton. Cell, 2007; 130:456–469.

Wallace

, Levy

, Matthews

. Use and abuse of HOMA modeling. Diabetes Care, 2004; 27:1487–1495.

Kanazawa

, Yamaguchi

, Yamamoto

, et al. Serum osteocalcin level is associated with glucose metabolism and atherosclerosis parameters in type 2 diabetes mellitus. J Clin Endocrinol Metab, 2009; 94:45–49.

Pittas

, Harris

, Eliades

, et al. Association between serum osteocalcin and markers of metabolic phenotype. J Clin Endocrinol Metab, 2009; 94:827–832.

, Yu

, Jeon

, et al. Relationship between osteocalcin and glucose metabolism in postmenopausal women. Clin Chim Acta, 2008; 396:66–69.

Ferron

, Hinoi

, Karsenty

, et al. Osteocalcin differentially regulates beta cell and adipocyte gene expression and affects the development of metabolic diseases in wild-type mice. Proc Natl Acad Sci U S A, 2008; 105:5266–5270.

Zhou

, Ma

, Li

, et al. Serum osteocalcin concentrations in relation to glucose and lipid metabolism in Chinese individuals. Eur J Endocrinol, 2009; 161:723–729.

10.

Guney

, Sener-Simsek

, Tokmak

, et al. Assessment of the relationship between serum vitamin D and osteocalcin levels with metabolic syndrome in non-osteoporotic postmenopausal women. Geburtshilfe Frauenheilkd, 2019; 79:293–299.

11.

Reinehr

, Roth

. A new link between skeleton, obesity and insulin resistance: Relationships between osteocalcin, leptin and insulin resistance in obese children before and after weight loss. Int J Obes (Lond), 2010; 34:852–858.

12.

Lee

, Jo

, Kim

, et al. Association between obesity, metabolic risks and serum osteocalcin level in postmenopausal women. Gynecol Endocrinol, 2012; 28:472–477.

13.

Lee

, Jo

, Kim

, et al. Association between osteocalcin and metabolic syndrome in postmenopausal women. Arch Gynecol Obstet, 2015; 292:673–681.

14.

Fernández-Real

, Izquierdo

, Ortega

, et al. The relationship of serum osteocalcin concentration to insulin secretion, sensitivity, and disposal with hypocaloric diet and resistance training. J Clin Endocrinol Metab, 2009; 94:237–245.

15.

Holecki

, Zahorska-Markiewicz

, Janowska

, et al. The influence of weight loss on serum osteoprotegerin concentration in obese perimenopausal women. Obesity (Silver Spring), 2007; 15:1925–1929.

16.

Welsh

, Rutherford

, James

, et al. The acute effects of exercise on bone turnover. Int J Sports Med, 1997; 18:247–251.

17.

Shearer

, Newman

. Metabolism and cell biology of vitamin K. Thromb Haemost, 2008; 100:530–547.

18.

Shea

, Booth

, Massaro

, et al. Vitamin K and vitamin D status: Associations with inflammatory markers in the Framingham Offspring Study. Am J Epidemiol, 2008; 167:313–320.

19.

Katsuyama

, Otsuki

, Tomita

, et al. Menaquinone-7 regulates the expressions of osteocalcin, OPG, RANKL and RANK in osteoblastic MC3T3E1 cells. Int J Mol Med, 2005; 15:231–236.

20.

Vergnaud

, Garnero

, Meunier, et al. Undercarboxylated osteocalcin measured with a specific immunoassay predicts hip fracture in elderly women: The EPIDOS Study. J Clin Endocrinol Metab, 1997; 82:719–724.