Elevated serum tartrate-resistant acid phosphatase isoform 5a levels in metabolic syndrome

Background Tartrate-resistant phosphatase isoform 5a is expressed in tumor-associated macrophages and is a biomarker of chronic inflammation. Herein, we correlated serum tartrate-resistant phosphatase isoform 5a levels with metabolic syndrome status and made comparisons with traditional markers of inflammation, including c-reactive protein and interleukin-6. Methods One hundred healthy volunteers were randomly selected, and cut-off points for metabolic syndrome related inflammatory biomarkers were determined using receiver operating characteristic curves. Linear and logistic regression models were subsequently used to correlate inflammatory markers with the risk of metabolic syndrome. Results Twenty-two participants met the criteria for metabolic syndrome, and serum tartrate-resistant phosphatase isoform 5a levels of >5.8 μg/L were associated with metabolic syndrome (c-statistics, 0.730; p = 0.001; 95% confidence interval, 0.618–0.842). In addition, 1 μg/L increases in tartrate-resistant phosphatase isoform 5a levels were indicative of a 1.860 fold increase in the risk of metabolic syndrome (p = 0.012). Conclusions Elevated serum tartrate-resistant phosphatase isoform 5a levels are associated with the risk of metabolic syndrome, with a cut-off level of 5.8 μg/L.


Demographic characteristics
Demographic characteristics of the 100 participants were stratified by sex and are presented in Table 1. The male to female ratio was 88:12, and twenty-two participants met the criteria for MetS. Components and distributions of MetS between male and female subjects were similar, except that blood pressure was predominantly higher in the male group. Age, body mass index, fasting glucose levels, and lipid profiles did not differ significantly between male and female study groups (p > 0.05).

Comparison of MetS to non-MetS patients
Although fasting glucose levels did not differ significantly between MetS and non-MetS subjects (p = 0.069), systolic blood pressure (p = 0.032), diastolic blood pressure (p = 0.001), waist circumference (p = 0.002), triglyceride levels (p < 0.001), and HDL levels (p < 0.001) differed significantly. IL-6 (p = 0.546) and CRP levels (p = 0.079) did not differ between patients with and without MetS, whereas TRACP-5a levels were significantly higher in subjects with MetS (p < 0.001; Figure 2, Table 3). No significant differences in blood urea nitrogen, creatinine, aspartate aminotransferase (AST), alanine aminotransferase (ALT), fasting glucose, and uric acid levels, or other blood cell parameters were identified between MetS and non-MetS patients.
In subsequent c-statistics analyzes, participants were classified in terms of MetS risk according to a TRACP-5a cut-off level of 5.8 μg/L. This cut-off point was accompanied by significant differences in AST and ALT levels, white blood cell counts, red blood cell counts, and hemoglobin levels. Moreover, blood pressure, waist circumference, and lipid profiles corresponded significantly with TRACP-5a levels.

Elevated TRACP-5a was associated with increased risk of MetS
In further analyses, binary logistic regression models were used to adjust parameters between MetS and non-MetS groups and to determine the risk of MetS (Table 4). Based on the data in Table 3 (column A), we controlled factors that differed significantly between MetS and non-MetS groups. In regression model 1, age, total cholesterol, and low-density lipoprotein (p = 0.007, 95% CI, 1.097 to 1.809) were covariates, whereas regression model 2 included the inflammatory parameters in model 1 with SBP, DBP, waist circumference, triglyceride, HDL, and fasting glucose (p = 0.012, 95% CI, 1.148-3.013). In models 1 and 2, single unit increases in TRACP-5a levels correlated with 1.409-and 1.860-fold increases in the risk for MetS. In contrast, IL-6 and CRP levels were not predictive of the risk for MetS.

Elevated blood sugar and triglyceride levels are associated with increments in TRACP-5a
We used linear regression models to identify MetS components that influence serum TRACP-5a concentrations (Table 5). In these analyzes, model 1 included sex only, model 2 included model 1 with AST, ALT, and uric acid levels, and model 3 comprised the parameters of model 2 with white and red blood cell counts and hemoglobin levels. After controlling for significant parameters from the TRACP-5a group (Table 3, column B), MetS components were not cumulatively associated with TRACP-5a levels until the total number of MetS components was ≥3 (Table 5), indicating that the presence of MetS is an independent predictor of TRACP-5a levels. However, further analyses of the effects of each component on TRACP-5a levels showed that impaired glucose tolerance and excess triglyceride levels are crucial determinants of TRACP-5a levels (p = 0.006 and p = 0.031, respectively).
atherosclerotic plaques [29]. Traditional risk factors for cardiovascular disease include sex, age, smoking history, hypertension, dyslipidemia, insulin resistance, and central obesity [30]. In addition, it is well documented that cardiovascular events and MetS are strongly associated with chronic inflammation [22,31,32].
In the present comparisons of MetS and non-MetS patient groups, TRACP-5a levels differed significantly, whereas IL-6 and CRP levels did not. In addition, among the present parameters, only age differed significantly between MetS and non-MetS groups. These data are in agreement with a previous cohort study of females showing that serum TRACP-5a levels increase with age [33]. However, age and TRACP-5a levels were not significantly correlated according to Pearson's coefficient (r = 0.192, p = 0.055), suggesting that increasing age is not strongly associated with differences in TRACP-5a levels between MetS and non-MetS patients.
AST, ALT, and uric acid levels, white and red blood cell counts, and hemoglobin levels were significantly stratified at the TRACP-5a cut-off of 5.8 μg/L (Table 3). However, in the present linear regression models of TRACP-5a levels, none of these parameters were significant independent correlates (p > 0.05 of β coefficient in each regression model). However, among patients that met at least three criteria for MetS, these variables were correlated with TRACP-5a levels ( Table 5).
After controlling for other variables, elevated triglyceride and blood sugar were predictive of TRACP-5a levels. Insulin resistance has been associated with both serum IL-6 and CRP levels in previous studies [34][35][36], and concomitant improvements (reductions) in these parameters have been observed following exercise training for 12 weeks in obese subjects, whereas TRACP-5a titers remained elevated [37]. In contrast, TRACP-5a, CRP, and IL-6 levels were decreased with weight loss twelve months after bariatric surgery (gastric bypass surgery or vertical banded gastric partition) in severely obese subjects [38]. Taken with the present observations, these data suggest that TRACP-5a secretion during macrophage infiltration is associated with circulating triglyceride levels and is affected by incremental changes in adipose tissue volumes in vessel endothelia. The present study was limited by sex bias, with male subjects comprising 80% of the cohort, although the proportion of males-to-females in MetS vs. non-MetS groups was equal (p = 0.519). This study was also limited by its cross-sectional design, which prevented evaluations of TRACP-5a responses to treatments such as lifestyle modification, exercise, and medication. Thus, future studies are required to confirm the present observations in larger cohorts of both sexes. Moreover, TRACP-5a accumulation in vascular endothelial cells needs to be verified and evaluated in autopsies from animal models. Finally, as the sample size was small, further studies are required to define relationships between TRACP-5a and individual MetS components, and the effects of treatments for insulin resistance and inflammatory markers.
In conclusion, this study warrants further consideration of increasing TRACP-5a levels as a more specific biomarker than the traditional MetS inflammatory IL-6 and CRP. Specifically, the present analyzes suggest that >5.8 μg/L TRACP-5a level is a potential cut-off point for the presence of MetS.

Study design, participants, and blood sample collection
This was a preliminary study for observing the levels and trends of inflammatory markers in patients with MetS. This study was conducted with the annual physical examination for civilians who received physical check-up every year at a single institution. Subjects with a history of coronary artery disease or medication for hypertension, dyslipidemia, blood sugar control, or malignancy were excluded. We requested for volunteers who were willing to participate in our study without selection and the expected final participant number was designated as 100 in the beginning of this study. From April 2015 to June 2015, 100 healthy civilian volunteers were enrolled. All participants received physical examinations and serum biochemical and hematologic analyzes, and inflammatory biomarker evaluations were performed to determine IL-6, CRP, and TRACP-5a titers. Blood samples were collected in the morning after at least 8 hours (h) of fasting and were stored at −80°C within 30 min of sampling. The study was approved by the Institutional Review Board of Tri-Service General Hospital (TSGH-IRB approval number: 1-103-05-065). Written informed consent was obtained from each participant prior to participation in the study. All laboratory methods were performed in accordance with the related guidelines.

Biochemical markers and serologic data
TRACP-5a titers were assessed using a two-site immunoassay as described previously [11]. Briefly, approximately 10 μL aliquots of serum were washed and diluted with 90 μL buffer (10 mM Tris-HCl, 150 mM NaCl, pH 7.5, containing 2% glycerol, 10 mM EDTA, and 0.05% Tween-20 solution) and were added in duplicate to 1 mg biotinylated mab220-coated streptavidin wells (Pierce Chemical Co.) that were specific for TRACP-5a. Diluted samples were covered with plastic wrap and were then incubated for 16 h (overnight) at 4°C. Streptavidin wells were then washed, and 100 μL aliquots of horseradish peroxidase-conjugated anti-TRACP mab162 diluted at 1:1000 were added and incubated for 1 h at room temperature. Streptavidin wells were washed again and o-phenylenediamine dihydrochloride and H 2 O 2 (horseradish peroxidase substrate) at pH 5.0 was added. Wells were then mixed and incubated for precisely 15 min and reactions were stopped by adding 50 μL aliquots of 2 M H 2 SO 4 . CRP levels were analyzed using enzymelinked sandwich immunoassays with rabbit polyclonal antiserum CRP (DAKO Denmark). Serum IL-6 titers were determined using a commercial immunoassay kit (RayBiotech, Inc.).

Statistical analysis
Statistical analyzes were performed using Statistical Product and Service Solutions (SPSS) software (version 18.0; SPSS, Chicago, IL, USA). Descriptive data were expressed as means ± standard deviations and categorical variables were compared using chi-square or Fisher's exact tests. Differences between continuous variables and categories were identified using student's t-test, and c-statistics were used to determine cut-off points for inflammatory markers. Significant associations of parameters with MetS were identified using linear and binary regression analyzes. Differences and associations were considered significant when p < 0.05.

Author contributions
Huang YJ, Huang TW, and Wu LW participated mainly in the manuscript writing. Chao TY and Sun YS performed the laboratory tests. Chen SJ, Chu DM, and Chen WL performed the statistical analysis and crucial review of the manuscript for language editing and article structure. Wu LW and Sun YS also promoted the study design and the financial funding.