2025 Volume 32 Issue 12 Pages 1558-1570
Aim: Serum syndecan-1 (SDC-1) concentration is a biomarker for endothelial glycocalyx (EG) degradation, which is elevated in type 2 diabetes (T2D). EG degradation is an early step in vascular endothelial dysfunction. This study investigated the association between serum SDC-1 concentration and visceral fat accumulation, which is closely related to vascular endothelial dysfunction, in people with and without T2D.
Methods: This was a cross-sectional study with two independent groups, one including 219 individuals without diabetes (ND) and the other including 203 individuals with T2D. Visceral fat accumulation was assessed as the visceral fat area (VFA) using computed tomography (CT) in ND or dual bioelectrical impedance analysis (BIA) in T2D. Multivariate analyses were performed for ND and T2D to assess the association between VFA and serum SDC-1 concentrations.
Results: The medians of serum SDC-1 concentration were 16.0 ng/mL and 26.5 ng/mL in ND and T2D, respectively. In the univariate analysis, both CT-VFA in the ND group and BIA-VFA in the T2D group were positively correlated with serum SDC-1 concentration. Moreover, the association between VFAs and serum SDC-1 concentration was independent of other covariates in multivariate analysis for each group. However, neither the body mass index nor subcutaneous fat area were associated with serum SDC-1 concentrations in either group.
Conclusions: CT-VFA and BIA-VFA were independently associated with serum SDC-1 concentrations. Our findings suggest that visceral fat accumulation is involved in the degradation of EG irrespective of the presence of T2D.
Syndecan-1 (SDC-1) is a transmembrane proteoglycan expressed on the luminal side of vascular endothelial cells1). SDC-1 binds to glycosaminoglycans, such as heparan sulfate and hyaluronan, to form a gel-like layer called the endothelial glycocalyx (EG). EG plays a protective role in endothelial cells by preventing the adhesion of circulating molecules or leukocytes and maintaining vascular permeability2). The circulating form of SDC-1 reflects the shedding of its transmembrane form of SDC-1 on endothelial cells and is widely used as a biomarker of EG degradation due to inflammation or oxidative stress3). Elevated circulating SDC-1 levels are predictive of mortality and the need for renal replacement therapy in patients with sepsis, a representative condition of acute inflammation4). EG degradation has been observed not only in acute inflammatory conditions but also in chronic diseases such as type 2 diabetes (T2D)5), hypertension6), and chronic kidney disease7). In individuals with T2D, circulating SDC-1 concentrations are not only elevated compared to those without T2D8), but are also associated with albuminuria9), a key marker closely linked to vascular endothelial dysfunction. The association between albuminuria or proteinuria and SDC-1 has been reported in various kidney diseases such as nephrotic syndrome10), minimal change disease11) and type 1 diabetes3) as well as T2D, suggesting that EG degradation is an early step in the course of vascular endothelial dysfunction12, 13).
Intra-abdominal or visceral fat accumulation is more closely associated with various metabolic disorders, such as T2D and dyslipidemia14, 15) and subclinical atherosclerosis16), than subcutaneous fat accumulation. Vascular endothelial dysfunction is observed in individuals with obesity17), especially those with visceral fat accumulation18). It is reasonable to speculate that the degradation of EG, which triggers vascular endothelial dysfunction, progresses with visceral fat accumulation. However, the relationship between EG degradation and obesity or visceral fat accumulation has only been investigated in a few animal studies. Diet-induced obese mice show reduced EG thickness and attenuated reactive vascular dilatation to blood flow19), while EG thickness of mesenteric arteries passing through visceral fat tissue in obese mice was reported to be smaller than that of arteries passing through the subcutaneous fat tissue20). These animal studies suggest a close relationship between visceral fat accumulation and EG degradation.
We investigated the association between visceral fat accumulation and serum SDC-1 concentrations in individuals with and without T2D.
The cross-sectional study included two independent groups: individuals with T2D and those without diabetes (ND). For the T2D group, we utilized a registry database (Approval No. 308) from the Diabetes Center at Osaka Metropolitan University Hospital, comprising 735 participants enrolled between October 2011 and April 2017. Among 488 individuals hospitalized between February 2013 and April 2017, information on visceral fat accumulation (VFA) measurement using bioelectrical impedance analysis (BIA) was provided consecutively. A total of 336 individuals consented and underwent the measurement. After excluding 27 cases with repeated measurements and 60 cases with insufficient stored serum samples (＜200 µL), 249 participants remained. Further exclusions were made for estimated glomerular filtration rate (eGFR) ＜30 mL/min/1.73m2 7, 21) (n = 24), acute infections4) (n = 5), or missing physiological or laboratory data to be used as explanatory variables (n = 17), leaving 203 participants for the final analysis (Fig.1).
The flow of the participant selection in this study. FPG, fasting plasma glucose; OGTT, oral glucose tolerance test; CT, computed tomography; BIA, bioelectrical impedance analysis; VFA, visceral fat area.
For the ND group, we used the MedCity21 Health Examination Registry (Approval no. 2927) from individuals undergoing comprehensive medical checkups that included visceral fat measurement at MedCity21, Osaka Metropolitan University Hospital Advanced Medical Center for Preventive Medicine. From the 438 individuals enrolled in the registry between June 2015 and November 2017, a total of 251 participants with available data on diabetes treatment, fasting glucose, HbA1c, or oral glucose tolerance test results were selected in order, starting from the most recently registered individuals, to obtain a sample size comparable to that of the T2D group. After excluding 27 individuals based on the American Diabetes Association22) and Japan Diabetes Society23) diabetes criteria, as well as 2 without CT-VFA measurements and 3 with missing data, 219 participants remained for the final analysis (Fig.1).
3.2 Ethical ConsiderationsThis study was conducted in compliance with the Declaration of Helsinki (2013 amendment) and the Ethical Guidelines for Medical Research Involving Human Subjects (Japanese Ministry of Health, Labour and Welfare, 2014). This study was approved by the Ethical Committee of Osaka Metropolitan University Graduate School of Medicine (Approval No. 2021-194). All participants provided written informed consent for the previous studies, including future use of their stored sample (Approval No. 2927 for ND in MedCity21 and No. 308 for T2D in Diabetes Center of Osaka Metropolitan University Hospital). In this study, serum SDC-1 concentrations were measured after providing participants with the opportunity to opt-out of participation, rather than requiring a formal consent form, on the department’s website (https://www.omu.ac.jp/med/interm2/research/optout/). This opt-out approach ensured that individuals could choose not to participate without the need for active consent, in accordance with ethical guidelines for voluntary involvement.
3.3 Clinical ParametersBody mass index (BMI) was calculated as weight in kilograms divided by the square of the height in meters (kg/m2). Blood pressure (BP) was measured using the conventional cuff method with an automatic sphygmomanometer at rest. Complications of cardiovascular diseases (CVD) included history of coronary artery disease, cerebrovascular disease, or peripheral artery disease. Biochemical parameters were analyzed using standard protocols at the Central Laboratory of Osaka Metropolitan University Hospital24, 25) or MedCity21 26) with blood samples drawn after an overnight fast. Glycated hemoglobin A1c (HbA1c) was estimated as the National Glycohemoglobin Standardization Program equivalent value (%), using the conversion formula established by the Japan Diabetes Society23). The formula established by the Japanese Society of Nephrology27) suitable for the Japanese population was used to calculate eGFR.
3.4 Measurement of Serum Syndecan-1 ConcentrationSerum SDC-1 concentration was measured using an enzyme-linked immunosorbent assay kit (Diaclone, Besançon, France)4) using frozen samples stored without thawing until use. The intra- and inter-assay coefficients of variation for human SDC-1were 6.2% and 10.2%, respectively.
3.5 Assessment of Visceral Fat AccumulationThe visceral fat accumulation was assessed as the visceral fat area (VFA) and subcutaneous fat area (SFA) using computed tomography (CT) (Supria Grande, Hitachi, Ltd., Tokyo, Japan) in the ND group, as described previously28, 29). VFA and SFA at the level of the umbilicus were automatically calculated using fatPointer software package, ver.2 (Hitachi, Ltd., Tokyo, Japan). On the other hand, VFA and SFA were measured by dual bioelectrical impedance analysis (BIA) instrument (HDS-2000, Omron Healthcare Co. Ltd. Kyoto, Japan) in the T2D group. Dual BIA estimates the cross-sectional area of intra-abdominal fat at the umbilical level using two distinct impedance measurements, as described in previous studies30, 31). A close correlation between VFA measured using dual BIA and VFA obtained from CT scans has been previously established32, 33).
3.6 Statistical AnalysisData are expressed as number (%), mean±standard deviation, or median (interquartile range), as appropriate. Comparisons between the ND and T2D groups were performed using the unpaired t-test, Wilcoxon rank-sum test, or chi-square test, as appropriate. The propensity score was estimated using a logistic regression model that included age, sex, eGFR, and BMI as covariates. Serum SDC-1 concentrations were converted to a logarithmic scale before univariate and multivariate analyses. The univariate correlations of clinical parameters including BMI, SFAs, and VFAs with serum SDC-1 concentration were performed using Pearson’s correlation analyses. Comparisons between SDC-1 concentration and categorical variables were analyzed using Wilcoxon’s rank sum tests. Multivariate regression analyses were performed to examine the associations of BMI, SFAs, and VFAs with log [SDC-1] with adjustments for the following variables: age, sex, systolic BP, HbA1c, low-density lipoprotein cholesterol (LDL-C), eGFR, C-reactive protein (CRP), complications of CVD, smoking status, use of 3-hydroxy-3-methyl-glutaryl-coenzyme A (HMG-CoA) reductase inhibitors, and use of renin-angiotensin-aldosterone system (RAS) inhibitors. Next, multiple regression analyses were performed using the same explanatory and dependent variables as in the above analysis for individuals without impaired kidney function (eGFR ＞60 mL/min/1.73m2) to minimize the influence of kidney dysfunction on serum SDC-1 concentration.
Statistical significance was set at p＜0.05. Statistical analyses were performed using the JMP®13 (SAS Institute Inc., Cary, NC, USA).
A diagram showing the process of participant selection is presented in Fig.1. The clinical characteristics of 219 participants in ND group and 203 in T2D group are summarized in Table 1. Participants in the T2D group were older and had higher BMI and blood pressures than those in the ND group. The proportion of the sexes did not differ between the T2D and ND groups. The levels of glucose, uric acid, triglycerides, and CRP were higher and the levels of eGFR, high-density lipoprotein cholesterol and LDL-C were lower in people with T2D than those with ND. The proportion of HMG-CoA reductase inhibitor users, RAS inhibitor users, and patients with CVD complications was higher in the T2D group than that in the ND group. No difference was noted in the proportion of smokers between the two groups.
| No diabetes | Type 2 diabetes | p value | |
|---|---|---|---|
| N (male/female) | 219 (115/104) | 203 (120/83) | 0.173 |
| Age (year) | 54.9±12.5 | 60.2±13.8 | ＜0.001 |
| Duration of diabetes (year) | N/A | 10 (2 - 18) | N/A |
| BMI (kg/m2) | 22.9±3.3 | 26.8±5.2 | ＜0.001 |
| Systolic BP (mmHg) | 120±15 | 133±19 | ＜0.001 |
| Diastolic BP (mmHg) | 74±11 | 79±11 | ＜0.001 |
| Complications of CVD (n, %) | 10 (4.6) | 26 (12.8) | 0.003 |
| HMG-CoA reductase inhibitors (n, %) | 27 (12.3) | 72 (35.5) | ＜0.001 |
| RAS inhibitors (n, %) | 17 (7.8) | 69 (34.0) | ＜0.001 |
| Smoking (n, %) | 117 (53.4) | 112 (55.2) | 0.719 |
| Fasting plasma glucose (mg/dL) | 100±9 | 130±37 | ＜0.001 |
| HbA1c (%) | 5.6±0.3 | 8.6±1.9 | ＜0.001 |
| Uric acid (mg/dL) | 5.5±1.4 | 6.1±1.5 | ＜0.001 |
| eGFR (mL/min/1.73m2) | 77.6±14.0 | 73.8±21.6 | 0.032 |
| Total cholesterol (mg/dL) | 203±34 | 185±43 | ＜0.001 |
| Triglyceride (mg/dL) | 88 (63 - 125) | 114 (86 - 163) | ＜0.001 |
| HDL-C (mg/dL) | 61±17 | 44±13 | ＜0.001 |
| LDL-C (mg/dL) | 121±30 | 112±38 | 0.009 |
| CRP (mg/dL) | 0.09±0.17 | 0.16±0.29 | 0.001 |
| Urinary albumin to creatinine ratio (mg/g) | N/A | 12 (6 - 38) | N/A |
| SDC-1 (ng/mL) | 16.0 (9.6 - 23.5) | 26.5 (18.7 - 40.6) | ＜0.001 |
| CT-VFA (cm2) | 75±45 | N/A | N/A |
| CT-SFA (cm2) | 139±59 | N/A | N/A |
| BIA-VFA (cm2) | N/A | 92±49 | N/A |
| BIA-SFA (cm2) | N/A | 202±101 | N/A |
Data are expressed as mean±standard deviation, median (interquartile range) or n (percentage) as appropriate. P-values were calculated using unpaired t-test, Wilcoxon rank sum test, or Chi-square test as appropriate.
Abbreviations: BMI, body mass index; BP, blood pressure; CVD, cardiovascular disease; HMG-CoA, 3-hydroxy-3-methyl-glutaryl-coenzyme A; RAS, renin-angiotensin-aldosterone system; HbA1c, glycated hemoglobin A1c; eGFR, estimated glomerular filtration rate; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; CRP, C-reactive protein; SDC-1, syndecan-1; VFA, visceral fat area; SFA, subcutaneous fat area; CT, computed tomography; BIA, bioelectrical impedance analysis; N/A, not available.
The median values of serum SDC-1 concentration were 16.0 ng/mL in ND and 26.5 ng/mL in T2D (Table 1). In a propensity score-matched comparison of 116 participants from both groups, the serum SDC-1 concentrations were significantly higher in participants in the T2D group than in those in the ND group (Supplementary Table 1).
| No diabetes | Type 2 diabetes | p value | |
|---|---|---|---|
| N (male/female) | 116 (74/42) | 116 (70/46) | 0.588 |
| Age (year) | 58.8±11.7 | 59.2±13.3 | 0.838 |
| BMI (kg/m2) | 24.4±3.4 | 24.4±3.7 | 0.964 |
| Systolic BP (mmHg) | 124±15 | 131±20 | 0.004 |
| Diastolic BP (mmHg) | 77±12 | 78±11 | 0.326 |
| Complications of CVD (n, %) | 5 (4.3) | 16 (13.8) | 0.012 |
| HMG-CoA reductase inhibitors (n, %) | 14 (12.1) | 32 (27.6) | 0.003 |
| RAS inhibitors (n, %) | 17 (14.7) | 39 (33.6) | ＜0.001 |
| Smoking (n, %) | 69 (59.5) | 62 (53.5) | 0.354 |
| Fasting plasma glucose (mg/dL) | 103±8 | 131±38 | ＜0.001 |
| HbA1c (%) | 5.7±0.3 | 8.6±1.9 | ＜0.001 |
| Uric acid (mg/dL) | 5.9±1.3 | 5.9±1.5 | 0.691 |
| eGFR (mL/min/1.73m2) | 76.4±13.7 | 74.5±20.8 | 0.412 |
| Total cholesterol (mg/dL) | 203±33 | 187±45 | 0.002 |
| Triglyceride (mg/dL) | 98 (75 - 132) | 106 (83 - 157) | 0.070 |
| HDL-C (mg/dL) | 56±14 | 46±15 | ＜0.001 |
| LDL-C (mg/dL) | 122±30 | 113±39 | 0.040 |
| CRP (mg/dL) | 0.09±0.15 | 0.13±0.32 | 0.236 |
| SDC-1 (ng/mL) | 16.5 (9.8 - 26.2) | 24.6 (17.6 - 43.5) | ＜0.001 |
Data are expressed as mean±standard deviation, median (interquartile range) or n (percentage) as appropriate. P-values were calculated using unpaired t-test, Wilcoxon rank sum test, or Chi-square test as appropriate.
Abbreviations: BMI, body mass index; BP, blood pressure; CVD, cardiovascular disease; HMG-CoA, 3-hydroxy-3-methyl-glutaryl-coenzyme A; RAS, renin-angiotensin-aldosterone system; HbA1c, glycated hemoglobin A1c; eGFR, estimated glomerular filtration rate; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; CRP, C-reactive protein; SDC, syndecan; VFA, visceral fat area; SFA, subcutaneous fat area; CT, computed tomography; BIA, bioelectrical impedance analysis; N/A, not available.
The mean VFA and SFA measured using CT in the ND group was 75 cm2 and 139 cm2, respectively. The mean VFA and SFA values measured using BIA in the T2D group were 92 cm2 and 202 cm2, respectively. Univariate analysis confirmed that the CT-VFA was strongly and positively correlated with BMI (r = 0.655, p＜0.001) and CT-SFA (r = 0.349, p＜0.001) in the ND group. Similarly, in the T2D group, BIA-VFA exhibited a strong positive correlation with BMI (r = 0.822, p＜0.001) and BIA-SFA (r = 0.798, p＜0.001).
4.3 Association of BMI, SFAs, and VFAs with SDC-1In the univariate analysis, BMI in ND (r = 0.177, p = 0.009), BMI in T2D (r = 0.145, p = 0.039) (Fig.2.A, D) and CT-VFA in ND (r = 0.283, p＜0.001), BIA-VFA in T2D (r = 0.218, p = 0.002) (Fig.2.B, E) were significantly and positively correlated with log [SDC-1]. In contrast, neither CT-SFA (r = 0.058, p = 0.394) in ND nor BIA-SFA in T2D (r = 0.087, p = 0.216) showed a significant correlation with log [SDC-1] (Fig.2.C, F). In addition to fat accumulation, systolic BP, LDL-C, and eGFR, but not CRP, showed significant correlations with log [SDC-1] in T2D. In the intergroup comparison, differences in SDC-1 concentration were observed depending on sex and smoking status in the ND group, and on sex, the use of HMG-CoA reductase inhibitors, and smoking status in the T2D group (Table 2).
Correlations of BMI and fat accumulation with log-transformed serum SDC-1 concentrations in individuals without diabetes (A-C) and in those with type 2 diabetes (D-F) using Pearson’s correlation analyses. Fat accumulation was measured using CT in individuals without diabetes (B and C) and BIA in people with type 2 diabetes (E and F). Correlation coefficients are shown as r values.
BMI, body mass index; SDC-1, syndecan-1; VFA, visceral fat area; SFA, subcutaneous fat area; CT, computed tomography; BIA, bioelectrical impedance analysis.
| (1) Comparison between subgroups | ||||||
| Subgroup | No diabetes | Type 2 diabetes | ||||
| N |
Median SDC-1 (interquartile range) |
p | N |
Median SDC-1 (interquartile range) |
p | |
| Male | 115 | 19.0 (12.1 - 28.8) | 120 | 29.5 (20.1 - 47.1) | ||
| vs Female | 104 | 11.9 (8.6 - 18.6) | ＜0.001 | 83 | 22.7 (15.9 - 31.5) | 0.004 |
| Complications of CVD (+) | 10 | 13.8 (5.4 - 21.9) | 26 | 27.6 (18.1 - 42.2) | ||
| vs Complications of CVD (–) | 209 | 16.3 (9.6 - 23.6) | 0.289 | 177 | 26.3 (18.7 - 40.6) | 0.974 |
| HMG-CoA reductase inhibitors (+) | 27 | 14.6 (9.8 - 21.4) | 72 | 21.9 (16.8 - 30.9) | ||
| vs HMG-CoA reductase inhibitors (–) | 192 | 16.2 (9.5 - 23.8) | 0.684 | 131 | 28.8 (21.4 - 47.5) | ＜0.001 |
| RAS inhibitors (+) | 17 | 17.4 (9.0 - 28.8) | 69 | 31.1 (20.7 - 49.5) | ||
| vs RAS inhibitors (–) | 202 | 16.0 (9.6 - 23.4) | 0.881 | 134 | 25.4 (17.9 - 37.6) | 0.071 |
| Smoking (+) | 117 | 18.4 (11.3 - 29.1) | 112 | 27.7 (20.6 - 45.7) | ||
| vs Smoking (–) | 102 | 14.2 (8.7 - 19.6) | ＜0.001 | 91 | 23.4 (15.9 - 34.6) | 0.033 |
| (2) Correlation with other variables | ||||||
| Variables | No diabetes | Type 2 diabetes | ||||
| Correlation coefficient | p | Correlation coefficient | p | |||
| Age | –0.084 | 0.217 | –0.115 | 0.103 | ||
| Systolic BP | 0.077 | 0.256 | 0.191 | 0.006 | ||
| HbA1c | –0.059 | 0.383 | –0.019 | 0.785 | ||
| LDL-C | 0.001 | 0.991 | 0.143 | 0.042 | ||
| eGFR | –0.064 | 0.345 | –0.206 | 0.003 | ||
| CRP | 0.101 | 0.138 | –0.050 | 0.478 | ||
P-value was calculated using Wilcoxon’s rank sum test or Pearson’s correlation analysis as appropriate.
Abbreviations: SDC-1, syndecan-1; CVD, cardiovascular disease; HMG-CoA, 3-hydroxy-3-methyl-glutaryl-coenzyme A; RAS, renin-angiotensin- aldosterone system; BP, blood pressure; HbA1c, glycated hemoglobin A1c; LDL-C, low-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate; CRP, C-reactive protein.
Multiple regression analyses were performed to examine the independent association between VFAs and serum SDC-1 concentration in each group (Table 3). CT-VFA was independently associated with log [SDC-1] (β = 0.260, p = 0.002) in ND (Model 2). A significant association between BIA-VFA and log [SDC-1] was also found in T2D (β = 0.152, p = 0.030) along with age, eGFR, and the use of HMG-CoA reductase inhibitors (Model 2). On the other hand, neither BMI in ND (β = 0.070, p = 0.368), CT-SFA in ND (β = 0.083, p = 0.246), BMI in T2D (β = 0.094, p = 0.207) nor BIA-SFA in T2D (β = 0.049, p = 0.507) was significantly associated with log [SDC-1] (Model 1 and 3 in ND and T2D). Because serum SDC-1 levels are influenced by kidney function7, 21), an interaction analysis was conducted to determine whether eGFR modifies the association between serum SDC-1 levels and VFAs. A significant interaction was not observed in either the ND group (p for interaction = 0.661) or the T2D group (p for interaction = 0.464). Furthermore, a supplementary analysis limited to participants with normal kidney function in the ND group (n = 197) demonstrated that CT-VFA was significantly associated with log [SDC-1] (β = 0.205, p = 0.025), as well as in the analysis with overall participants (Supplemental Table 2. Model 2). Similarly, in T2D group with normal kidney function (n = 148), BIA-VFA showed the positive trend of association with log [SDC-1] (β = 0.165, p = 0.052), although the association was not statistically significant (Supplemental Table 2. Model 2).
| No diabetes | Type 2 Diabetes | |||||
|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | Model 1 | Model 2 | Model 3 | |
| Sex (male = 1, female = 0) | 0.179＊ | 0.123 | 0.208＊＊ | 0.110 | 0.093 | 0.107 |
| Age (years) | –0.093 | –0.145 | –0.090 | –0.223＊ | –0.219＊＊ | –0.247＊＊ |
| Systolic BP (mmHg) | 0.044 | –0.012 | 0.043 | 0.141＊ | 0.126 | 0.150＊ |
| HbA1c (%) | –0.067 | –0.084 | –0.068 | 0.001 | 0.003 | 0.002 |
| LDL-C (mg/dL) | 0.049 | 0.032 | 0.047 | 0.021 | 0.021 | 0.026 |
| eGFR (mL/min/1.73m2) | –0.142 | –0.146 | –0.144 | –0.326＊＊ | –0.320＊＊ | –0.330＊＊ |
| CRP (mg/dL) | 0.052 | 0.022 | 0.053 | –0.064 | –0.073 | –0.057 |
| Complications of CVD (Yes = 1, No = 0) | –0.081 | –0.104 | –0.079 | 0.016 | 0.015 | 0.016 |
| HMG-CoA reductase inhibitors (Yes = 1, No = 0) | –0.020 | –0.037 | –0.020 | –0.276＊＊ | –0.285＊＊ | –0.271＊＊ |
| RAS inhibitors (Yes = 1, No = 0) | –0.011 | –0.028 | –0.012 | 0.113 | 0.110 | 0.121 |
| Smoking (Yes = 1, No = 0) | 0.177＊ | 0.132 | 0.184＊ | 0.055 | 0.043 | 0.058 |
| BMI (kg/m2) | 0.070 | 0.094 | ||||
| CT-VFA (cm2) | 0.260＊＊ | |||||
| CT-SFA (cm2) | 0.083 | |||||
| BIA-VFA (cm2) | 0.152＊ | |||||
| BIA-SFA (cm2) | 0.049 | |||||
| R2 | 0.148＊＊ | 0.182＊＊ | 0.150＊＊ | 0.252＊＊ | 0.265＊＊ | 0.248＊＊ |
Values are standardized regression coefficients (β) determined by using multiple regression analysis. R2, coefficient of determination. ＊, p＜0.05; ＊＊,p＜0.01.
Abbreviations: SDC-1, syndecan-1; BP, blood pressure; HbA1c, glycated hemoglobin A1c; LDL-C, low-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate; CRP, C-reactive protein; RAS, renin-angiotensin-aldosterone system; BMI, body mass index; VFA, visceral fat area; SFA, subcutaneous fat area; CT, computed tomography; BIA, bioelectrical impedance analysis.
| No diabetes (n = 197) | Type 2 Diabetes (n = 148) | |||||
|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | Model 1 | Model 2 | Model 3 | |
| Sex (male = 1, female = 0) | 0.225＊＊ | 0.178＊ | 0.258＊＊ | 0.077 | 0.055 | 0.076 |
| Age (years) | –0.091 | –0.134 | –0.089 | –0.243＊ | –0.243＊ | –0.265＊ |
| Systolic BP (mmHg) | 0.094 | 0.051 | 0.094 | 0.186＊ | 0.175＊ | 0.197＊ |
| HbA1c (%) | –0.089 | –0.094 | –0.089 | –0.061 | –0.050 | –0.058 |
| LDL-C (mg/dL) | 0.051 | 0.039 | 0.048 | 0.019 | 0.008 | 0.020 |
| eGFR (mL/min/1.73m2) | –0.102 | –0.107 | –0.104 | –0.254＊＊ | –0.250＊＊ | –0.260＊＊ |
| CRP (mg/dL) | 0.048 | 0.029 | 0.050 | –0.099 | –0.110 | –0.099 |
| Complications of CVD (Yes = 1, No = 0) | –0.044 | –0.060 | –0.042 | 0.020 | 0.022 | 0.017 |
| HMG-CoA reductase inhibitors (Yes = 1, No = 0) | –0.048 | –0.068 | –0.049 | –0.258＊＊ | –0.268＊＊ | –0.255＊＊ |
| RAS inhibitors (Yes = 1, No = 0) | 0.047 | 0.036 | 0.047 | 0.166＊ | 0.165＊ | 0.166＊ |
| Smoking (Yes = 1, No = 0) | 0.167＊ | 0.138 | 0.174＊ | 0.016 | 0.005 | 0.019 |
| BMI (kg/m2) | 0.080 | 0.106 | ||||
| CT-VFA (cm2) | 0.205＊ | |||||
| CT-SFA (cm2) | 0.093 | |||||
| BIA-VFA (cm2) | 0.165 | |||||
| BIA-SFA (cm2) | 0.073 | |||||
| R2 | 0.191＊＊ | 0.209＊＊ | 0.194＊＊ | 0.241＊＊ | 0.255＊＊ | 0.237＊＊ |
Values are standardized regression coefficients (β) determined by using multiple regression analysis. R2, coefficient of determination. ＊, p＜0.05; ＊＊,p＜0.01.
Abbreviations: SDC-1, syndecan-1; BP, blood pressure; HbA1c, glycated hemoglobin A1c; LDL-C, low-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate; CRP, C-reactive protein; RAS, renin-angiotensin-aldosterone system; BMI, body mass index; VFA, visceral fat area; SFA, subcutaneous fat area; CT, computed tomography; BIA, bioelectrical impedance analysis.
The present study explored the association between visceral fat accumulation and EG degradation in individuals with and without T2D. The results demonstrated that both CT-VFA in the ND group and BIA-VFA in the T2D group were significantly associated with EG degradation, as assessed by serum SDC-1 concentrations in multivariate analyses. In contrast, neither BMI in ND, CT-SFA in ND, BMI in T2D, nor BIA-SFA in T2D were associated with serum SDC-1 concentration.
Previous studies have indicated that visceral fat accumulation is more closely linked to vascular endothelial dysfunction than is subcutaneous fat accumulation18, 34). The diameter of the visceral adipose tissue (VAT) measured with ultrasound shows a negative correlation with vascular endothelial function, assessed as flow-mediated dilatation (FMD) in obese individuals18). In the same study, a positive correlation was observed between the diameter of the subcutaneous adipose tissue (SAT) and FMD, which was opposite to that of VAT. In another study, an increase in VAT, but not SAT, was associated with a decrease in FMD when healthy individuals with normal weight gained weight through an additional diet for 8 weeks34). These studies suggest that visceral fat accumulation causes damage to vascular endothelial cells. In the present study, among obesity-related indices, although VFAs were strongly correlated with BMI and SFAs, only VFAs showed an independent association with serum SDC-1 concentrations. Because no previous study has investigated the association between visceral fat accumulation and EG degradation, the present study is the first to demonstrate an association of visceral fat accumulation, but not subcutaneous fat accumulation or BMI, with EG degradation.
However, the mechanism by which visceral fat accumulation leads to EG degradation has not yet been fully understood. Oxidative stress increases chronically in individuals with obesity35) due to overproduction of pro-inflammatory cytokines, such as tumor necrosis factor-α and interleukin-6 36) and underproduction of antioxidants37) in the adipose tissue. A previous study showed that damage to the EG during acute hyperglycemia induced by intravenous glucose infusion was abolished by the simultaneous infusion of an antioxidant38), suggesting that oxidative stress causes the degradation of EG. Oxidative stress was also shown to be increased in individuals with mildly decreased eGFR (from 90 to 60 mL/min/1.73m2)39). In the present study, although its decline was mild (mean, 73.8 mL/min/1.73m2), eGFR was independently and inversely association with the serum SDC-1 concentration in T2D (Table 2), which is consistent with the previous study39). Taken together, it is possible to speculate that oxidative stress derived from VAT contributes to EG degradation, although oxidative stress markers could not be measured in the present study.
Another possible mechanism linking visceral fat accumulation and EG degradation is the involvement of matrix metalloproteinase-9 (MMP-9), which is produced by adipose tissue. MMP-9 is an enzyme that sheds the transmembrane form of SDC-1 on the vascular endothelial cells40), thereby increasing the circulating form of SDC-1. The expression of MMP-9 mRNA in the adipose tissue has been shown to be positively correlated with BMI in individuals with obesity41). In addition, plasma MMP-9 concentration was increased in individuals with obesity42) and T2D43) and was decreased in those receiving HMG-CoA reductase inhibitors44). Participants with T2D in the present study had poor glycemic control (mean HbA1c, 8.6%) and were overweight (mean BMI, 26.8 kg/m2), suggesting an increase in the MMP-9 expression in the adipose tissue. Multiple regression analyses in the present study showed that the use of HMG-CoA reductase inhibitors was a negative determinant of serum SDC-1 concentration in people with T2D (Table 2). Our findings are in line with those of prior studies showing a decrease in plasma MMP-9 concentration following the administration of HMG-CoA reductase inhibitors44, 45).
This study demonstrated that VFAs, but not SFAs, was a determinant of serum SDC-1 concentration. This finding is consistent with those of previous studies suggesting that oxidative stress is greater in the VAT than in the SAT. Mesenchymal stem cells (MSCs) derived from human VAT express higher levels of oxidative stress-related genes than those derived from SAT46). Clinically, stronger association of oxidative stress biomarkers with VFA compared to that with SFA were observed in individuals with metabolic syndrome47) and those undergoing bariatric surgery48). Additionally, the expression of antioxidant proteins differs between the VAT and SAT in obese individuals49).
This study demonstrated an association between the visceral fat accumulation and EG degradation irrespective of the presence or absence of T2D. Previous studies8), including ours9) showed that serum SDC-1 concentrations were higher in individuals with T2D than in those without T2D, and our results suggest an association between elevated VFA and serum SDC-1 concentrations, both of which are elevated in T2D. In addition, the current study showed that CT-VFA and SDC-1 were associated with each other at lower levels in people with ND than in people with T2D, suggesting that these relationships may be generalizable. Future studies exploring the effect of weight loss on serum SDC-1 concentrations in individuals with visceral fat accumulation would be of interest.
The present study had several limitations. First, and most importantly, methods for measuring VFA differed between the ND and T2D groups, and BIA was used in the T2D group instead of CT, which is the gold standard. The correlation between BIA-VFA and CT-VFA could not be measured in the participants in this study, however, previous studies have shown a good correlation between the VFA obtained using CT and that using BIA32, 33). Second, this study was a cross-sectional study, and causality could not be determined. Third, a portion of SDC-1 in serum could originate from sources other than vascular endothelial cells, such as neutrophils, especially during inflammation50). Forth, the possibility cannot be excluded that kidney function may affect the association between SDC-1 and VFAs, as the excretion of SDC-1 is partially dependent on renal clearance21). In this study, we performed an interaction analysis and confirmed that eGFR did not influence the relationship between SDC-1 and VFAs. Furthermore, a supplementary analysis restricted to participants with normal renal function demonstrated a consistent direction of association. Finally, selection bias may exist because no specific criteria were established for deciding whether to perform VFA measurements in either group.
VFAs were independently associated with serum SDC-1 concentration in individuals with and without T2D. Our data suggests that EG degradation is a consequence of the damage to endothelial cells induced by visceral fat accumulation, irrespective of the presence of T2D.
The authors acknowledge the valuable assistance of all the data entry staff at the Department of Metabolism, Endocrinology, and Molecular Medicine, Osaka Metropolitan University Graduate School of Medicine.
The authors declare that there is no conflict of interest.
The authors disclose the receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Osaka Kidney Foundation (Grant number OKF21-0012) and Bayer Academic Support (Grant number BASJ20210415007).
