Introduction

Obesity has become a major public health concern worldwide.1 Importantly, the adverse health consequences of obesity are determined less by total adiposity than by visceral adiposity.2,3 Visceral obesity, a major form of ectopic fat deposition, is closely linked to insulin resistance, cardiovascular disease, and multiple metabolic complications.4,5 Therefore, identifying biomarkers that reflect visceral fat accumulation and its related metabolic disturbances is of considerable importance for early risk stratification and clinical intervention.

Visceral adipose tissue is known to express and secrete a variety of adipokines, such as leptin, adiponectin, chemerin and apelin. These adipokines have been associated with visceral fat burden and related metabolic risk in humans.6,7 The release of adipokines by adipocytes can induce a chronic inflammatory state, which may play a central role in the development of insulin resistance and visceral obesity.8,9 Growth differentiation factor 3 (GDF3), a member of the transforming growth factor-β superfamily, is an adipokine that has recently attracted increasing attention. It may participate in visceral adipose proliferation and adiposity, metabolic disturbances, and immune responses.10,11 In mice models of obesity, aging, or inflammation, Gdf3 mRNA expression is increased in white adipose tissue, including visceral depots.12–15 Animal studies GDF3 may signal through the activin receptor-like kinase 7(ALK7) to promote adipose accumulation, particularly in visceral fat depots, and to contribute to insulin resistance.16,17 Conversely, acute loss of GDF3 function reduces adipose tissue lipolysis and improves insulin sensitivity without changes in body weight.14 In addition, GDF3 has been shown to promote adipose tissue inflammation by stimulating adipocytes to secrete monocyte chemoattractant protein-1 (MCP-1), tumor necrosis factor-α (TNF-α), and interleukin-6 (IL-6), thereby enhancing macrophage recruitment and M1 polarization.18 Recently, a study have shown that serum GDF3 levels are elevated in patients with nonalcoholic steatohepatitis.19 However, serum GDF3 levels in individuals with visceral obesity have not yet been well studied. Therefore, further investigation is warranted to clarify the associations of circulating GDF3 levels with visceral obesity and related metabolic and inflammatory disorders in humans.

In this study, we conducted both a cross-sectional and a pharmacological intervention study to investigate the association between serum GDF3 levels and visceral obesity in adults. Our findings highlight the potential of GDF3 as a biomarker for visceral obesity and related insulin resistance and inflammatory disturbances.

Materials and Methods Study Design and Participants

This study comprised two parts: a cross-sectional observational study and an embedded self-controlled interventional study. In the cross-sectional component, we used non-probability convenience sampling; eligible adults were enrolled consecutively as they presented. From December 2022 to July 2024, we recruited adults aged 18–65 years from the Obesity Clinic and the Health Management Center of The First Affiliated Hospital of University of South China. After applying the exclusion criteria, a total of 282 participants were included, comprising 120 individuals with visceral obesity and 162 individuals without visceral obesity. All participants were non-diabetic had stable body weight (no notable fluctuation) during the 3 months prior to enrollment.

An embedded self-controlled longitudinal component was conducted. During the study period (2022), semaglutide use for weight management in China was off-label. With written informed consent, 23 adults with visceral obesity and BMI >30 kg/m2 received a standardized 12-week, once-weekly semaglutide treatment and completed prespecified baseline and follow-up assessments. Semaglutide was prescribed for obesity management rather than for glycaemic control. Eligibility for semaglutide pharmacotherapy required documented failure of lifestyle-based weight loss, and treatment was initiated in routine clinical practice after clinical evaluation and shared decision-making in participants with a sustained willingness to receive medication. This practice was aligned with the contemporaneous international evidence base and regulatory/guideline support for obesity pharmacotherapy available at that time. Semaglutide has been shown to confer significant benefits in reducing visceral fat among individuals with obesity.20

All participants provided written informed consent before enrollment. This study has been recorded by the China Clinical Trial Registry (No. ChiCTR2200059056). The protocol was approved by the Ethics Committee of The First Affiliated Hospital of University of South China (No. 2022110901003).

Definitions and Exclusion Criteria

Visceral obesity was defined as a visceral fat area (VFA) ≥ 100 cm2.21,22 In this study, VFA was assessed using a multi-frequency bioelectrical impedance analysis (BIA) system (HDS-2000, OMRON). Previous studies have demonstrated good agreement between BIA-derived VFA and abdominal CT measurements (r = 0.88),23 and BIA have been widely used in epidemiological research.24

To minimize potential confounding, individuals with the following conditions were excluded: secondary metabolic disorders (eg, Cushing syndrome, polycystic ovary syndrome), active infection (acute or recurrent), severe cardiovascular/cerebrovascular, hepatic, or renal disease, malignancy, recent (< 30 days) use of immunosuppressants, diagnosed diabetes mellitus, pregnancy or lactation, and use of anti-obesity medications or prior bariatric surgery within the past 3 months.

Collection of Medical and Laboratory Data

General information was collected via standardized interviews, including age, sex, smoking status, alcohol consumption, medical history (eg, diabetes mellitus, cardiovascular/cerebrovascular disease), and recent medication and surgical history. Current smoking was defined as use of combustible tobacco on ≥ 1 day in the past 30 days. Alcohol consumption was categorized as non-heavy drinking or heavy drinking (≥ 30 g/day for men and ≥ 20 g/day for women).25 Blood pressure was measured three times using a mercury sphygmomanometer, and systolic blood pressure (SBP) and diastolic blood pressure (DBP) were calculated as the average of all available measurements.

After an overnight fast (≥ 8 h), blood samples were obtained between 8:00 and 9:00 a.m. All assays were performed according to the instructions of manufacturers in a single certified laboratory. Serum biochemical parameters were measured using an automated biochemical analyzer, including alanine aminotransferase (ALT), aspartate aminotransferase (AST), uric acid (UA), triglycerides (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C). Fasting plasma glucose (FPG) was determined by the hexokinase method. Glycated hemoglobin A1c (HbA1c) was measured by high-performance liquid chromatography. High-sensitivity C-reactive protein (hs-CRP) was quantified on a Hitachi 7600 analyzer (Japan). Fasting insulin (FINS) was measured by enzyme-linked immunosorbent assay (ELISA) (DRG, Marburg, Germany). The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated as follows: HOMA-IR = FINS (μU/mL) × FPG (mmol/L)/22.5.

Measurement of Serum GDF3 and Inflammatory Cytokines by ELISA

Serum GDF3 levels were measured using commercial ELISA kits from Novus Biologicals Technology (USA). Serum IL-6, TNF-α, and MCP-1 levels were measured using ELISA kits from Reed Biotech Ltd (China). The measurements were performed following the manufacturers’ instructions. All samples were assayed in duplicate and in random order. Standards and serum samples were added to antibody-precoated microplates, followed by incubation with biotinylated detection antibodies and horseradish peroxidase-conjugated reagents. After washing, substrate solution was added for color development, and the reaction was terminated with stop solution. Optical density was measured at 450 nm using a microplate reader. For the GDF3 assay, the sensitivity was 46.88 pg/mL and the detection range was 78.13–5000 pg/mL. According to the manufacturer, no significant cross-reactivity or interference was observed, and the coefficient of variation was <10%. Standard curves were generated using serially diluted standards, and sample concentrations were calculated from the standard curve after blank correction. For diluted samples, the measured concentrations were multiplied by the corresponding dilution factors to obtain the final values.

Statistical Analysis

The normality of continuous variables was assessed using the Kolmogorov–Smirnov test. Normally distributed data are presented as mean ± standard deviation (SD), whereas non-normally distributed data are presented as median with interquartile range (IQR). Categorical variables are summarized as frequencies and percentages (n, %).

Comparisons of categorical variables were performed using the χ2-test. Continuous variables with a normal distribution were compared using the independent-samples t test, whereas non-normally distributed variables were compared using the Mann–Whitney U-test. Multivariable logistic regression was used to evaluate the association between serum GDF3 levels and visceral obesity. Serum GDF3 was modeled as both a continuous variable and a categorical variable (tertiles, with the lowest tertile as the reference). A test for linear trend was conducted by assigning the median value of each tertile and modeling it as a continuous variable. Based on prior reports,26,27 the final multivariable model was adjusted for age, sex, SBP, DBP, FPG, ALT, AST, and TG. Receiver operating characteristic (ROC) curve analysis was performed to assess the discriminative performance of serum GDF3 for visceral obesity status.

Spearman correlation was used to assess correlations between serum GDF3 levels and metabolic parameters, and a false discovery rate (FDR)-corrected threshold was applied to address multiple comparisons and define statistical significance. Multivariate linear regression analysis was employed to identify independent predictors of serum GDF3 concentrations. In addition, within-subject changes before and after semaglutide intervention were evaluated using the paired t test or the Wilcoxon signed-rank test, as appropriate.

Mediation analysis (R package “mediation”) was conducted to evaluate the potential mediating effect of serum GDF3 on the associations between VFA and metabolic–inflammatory phenotypes. This approach estimates the total effect, direct effect, indirect effect, and proportion mediated. Statistical significance of the mediation effect was assessed using 5,000 bootstrap iterations. Covariates included in the mediation models were consistent with those used in the logistic regression analyses.

Missing values were present in a small proportion of covariates (TG and SBP; maximum missing rate < 5%) and were imputed using the median. Except for mediation analysis, all statistical analyses were performed using SPSS version 27.0 (IBM Corp., Armonk, NY, USA). A two-sided P-value < 0.05 was considered statistically significant.

Sample size and power calculations were performed using PASS 2025 (NCSS, Kaysville, UT). For both the case–control comparison and the self-controlled subcohort analysis, the estimated power (1−β) exceeded 0.90 at a two-sided α of 0.05.

Results Basic Characteristics

The basic characteristics of the study subjects are shown in Table 1. Consistent with previous reports, compared with participants without visceral obesity, those with visceral obesity had significantly higher BMI, VFA, FINS, HOMA-IR, ALT, AST, UA, TG, LDL-C, hs-CRP, MCP-1, TNF-α and IL-6 (all P < 0.05). In contrast, HDL-C was significantly lower in the visceral obesity group than in the non-visceral obesity group (P < 0.001). There were no significant differences in smoking status, alcohol consumption, FPG, HbA1c, or TC between the two groups (all P > 0.05).

Table 1 Clinical and Biochemical Features in Study Subjects

Serum GDF3 Levels of Study Participants

In addition, compared with the non-visceral obesity group [1246.23 (1005.97, 1586.71) pg/mL], the visceral obesity group [1682.83 (1356.86, 2336.92) pg/mL] had markedly higher serum GDF3 levels (P < 0.001; Figure 1A). Sex-stratified analyses showed the same pattern in men and in women (Figure 1B and C). There was no significant difference in serum GDF3 between men and women (P > 0.05; Figure 1D).

Figure 1 Comparison of serum GDF3 levels between non-visceral obesity and visceral obesity. (A) Non-visceral obesity vs visceral obesity(overall). (B) Men only: Non-visceral obesity vs visceral obesity. (C) Women only: Non-visceral obesity vs visceral obesity. (D) Men vs women (overall). Visceral obesity was defined by visceral fat area (VFA) ≥ 100 cm2. Data are shown as the median (IQR). Comparisons used the Mann–Whitney U-test (two-sided). Significance: ns = not significant, ***P < 0.001.

Association Between Serum GDF3 Levels and Visceral Obesity

The associations of serum GDF3 levels with visceral obesity are presented in Table 2. Multivariable logistic regression analysis showed that serum GDF3 concentrations were positively associated with the presence of visceral obesity (P < 0.05). When serum GDF3 was modeled as a categorical variable, the association remained significant and showed a dose–response relationship (P for trend < 0.05). Compared with the lowest tertile, the highest tertile of serum GDF3 was associated with a higher prevalence of visceral obesity (adjusted OR = 6.322; 95% CI: 2.918–13.695).

Table 2 Logistic Regression Analysis of the Association of Serum GDF3 with Visceral Obesity

Furthermore, ROC curve analysis indicated that serum GDF3 could reasonably discriminate visceral obesity status (Figure 2), with an area under the curve of 0.739 (95% CI: 0.681–0.798). The sensitivity and specificity were 0.750 and 0.636, respectively, and the optimal cut-off value was 1366 pg/mL.

Figure 2 ROC curve analysis was performed for the prediction of visceral obesity according to the GDF3 levels.

Modulatory Effects of Semaglutide Therapy on VFA and Serum GDF3 Levels

After 12 weeks of semaglutide treatment, participants showed a significant reduction in VFA, a decrease of 7.5% from baseline (P < 0.05). Improvements were also observed in other obesity- and metabolism-related parameters, including BMI, lipid profile, and inflammatory markers. Notably, serum GDF3 levels declined significantly from 2051.12 ± 665.21 pg/mL at baseline to 1434.86 ± 346.12 pg/mL after treatment (P < 0.05; Table 3), representing a 30.05% reduction. The percentage reduction in serum GDF3 was positively correlated with the percentage decrease in VFA (r = 0.437, P < 0.05; Figure S1), and this association remained significant after adjustment for change in body weight, although the correlation coefficient was slightly attenuated.

Table 3 Changes in Anthropometric and Biochemical Parameters After 12-Week GLP-1RA Intervention

Associations Between Serum GDF3 Levels with Insulin Resistance and Inflammatory Markers

Spearman correlation analyses showed that serum GDF3 was positively correlated with a metabolic parameter (HOMA-IR) and inflammatory markers (hs-CRP, IL-6, MCP-1, and TNF-α) (all P < 0.001; Figure 3). After further adjustment for demographic characteristics and clinical biochemical parameters, multivariate linear regression analysis indicated that VFA, HOMA-IR, MCP-1, and TNF-α were independently associated with serum GDF3 levels (all P < 0.05; Table 4).

Table 4 Correlation of Serum GDF3 Levels with Clinical Variables in All Subjects

Figure 3 Correlations of serum GDF3 with anthropometric indices, glucose-lipid metabolism and inflammatory markers. (A) BMI. (B) WC. (C) VFA. (D) HOMA-IR. (E) MCP-1. (F) TNF-α. Each dot represents one participant; solid lines represent fitted linear regression lines and dotted lines 95% confidence intervals. r and P values are from Spearman’s rank correlation.

Abbreviations: BMI, body mass index; WC, waist circumference; VFA, visceral fat area; HOMA-IR, homeostatic mode assessment of insulin resistance; MCP-1, monocyte chemoattractant protein-1; TNF-α, tumor necrosis factor-α.

Potential Mediating Role of Serum GDF3 in the Associations of VFA with MCP-1 and HOMA-IR

Mediation analyses suggested that serum GDF3 partially mediated the associations between VFA and metabolic and inflammatory outcomes. The proportion mediated by serum GDF3 was 32.6% for HOMA-IR and 16.7% for MCP-1, whereas no significant mediation effect was observed for TNF-α. (Figure 4).

Figure 4 Mediation analyses assessing serum GDF3 as a mediator of the associations between VFA and metabolic–inflammatory indices. (A) VFA–MCP-1, (B) VFA–TNF-α and (C) VFA–HOMA-IR. IE, the estimate of the indirect effect; DE, the estimate of the direct effect; Proportion of mediation = IE/(DE + IE).

Abbreviations: VFA, visceral fat area; GDF3, growth differentiation factor 3; TNF-α, tumor necrosis factor-alpha; MCP-1, monocyte chemoattractant protein-1; HOMA-IR, homeostasis model assessment of insulin resistance.

Discussion

To our knowledge, this study is the first to evaluate the association between serum GDF3 levels and visceral obesity in a general population. We found that serum GDF3 levels were significantly elevated in individuals with visceral obesity and reduced after semaglutide treatment in association with decreased visceral fat, suggesting that GDF3 may serve as a potential biomarker of visceral fat accumulation. In addition, mediation analyses indicated that serum GDF3 may partially mediate the associations of VFA with HOMA-IR and MCP-1. Collectively, these findings provide new evidence supporting the potential role of GDF3 as a biomarker of visceral obesity.

Among individuals with obesity, those with visceral obesity typically have a higher metabolic risk.28 GDF3 is an adipokine that has recently been implicated in several conditions, including myocardial infarction, metabolic dysfunction-associated steatohepatitis, obesity, and aging.16,19,29,30 Animal studies suggest that GDF3 is upregulated in white adipose tissue under obese conditions.31 Mechanistically, GDF3 may serve as a ligand in ALK7 signaling in adipocytes, thereby promoting fat accumulation. Deficiency of either GDF3 or ALK7 reduced adipose tissue expansion and partially protected against diet-induced obesity.12,16 Collectively, these findings from animal models support a close biological link between GDF3 and obesity-related adipose expansion, including visceral fat accumulation.

In our study, serum GDF3 levels were elevated in individuals with higher VFA. In addition, we conducted a semaglutide intervention study to further explore the relationship between GDF3 and VFA. Semaglutide represents a cornerstone therapy for obesity and has been shown to substantially reduce visceral fat among individuals with obesity.32,33 Consistent with previous findings, we confirmed that a 12-week semaglutide intervention significantly reduced body weight and VFA, and improved lipid profiles and insulin resistance. Notably, serum GDF3 levels also decreased significantly following semaglutide treatment, in parallel with the reduction in VFA, and changes in GDF3 were positively correlated with changes in VFA. These findings support a close association between circulating GDF3 and visceral adiposity in humans. However, due to the lack of control groups in the study, we cannot exclude the possibility that semaglutide therapy directly regulates GDF3 expression. It should also be noted that all participants in the semaglutide subgroup were non-diabetic adults with obesity (BMI >30 kg/m2), and semaglutide was prescribed for obesity management rather than for glycaemic control. During the study period in 2022, such use remained off-label in China. However, treatment was initiated in routine clinical practice in individuals who had shown an inadequate response to lifestyle-based weight loss, and this prescribing approach was aligned with the contemporaneous international evidence base and regulatory/guideline support for obesity pharmacotherapy. These considerations should be taken into account when interpreting the interventional findings.

Visceral adipose tissue is closely linked to insulin resistance and inflammatory.34,35 We found that serum GDF3 was independently associated with HOMA-IR, indicating that GDF3 may contribute to obesity-related insulin resistance. This finding is consistent with recent studies.13,14 Importantly, because all participants in the present study were non-diabetic, the observed association of serum GDF3 with HOMA-IR suggests that GDF3 may reflect early insulin resistance in the absence of overt type 2 diabetes. This feature may confer potential clinical relevance beyond that of some established adipokines, as GDF3 appears to capture both visceral adiposity and its related early metabolic dysfunction. However, this potential advantage requires further validation in longitudinal and mechanistic studies.

Furthermore, the positive correlations between circulating GDF3 and inflammatory markers, including hs-CRP, MCP-1, TNF-α and IL-6, highlight its potential role in obesity-associated chronic low-grade inflammation. Mechanistically, GDF3 has been reported to modulate immune cell function, particularly by promoting the polarization of adipose tissue macrophages toward a pro-inflammatory M1 phenotype.18 These M1-polarized macrophages accumulate in adipose tissue and secrete large amounts of inflammatory cytokines, including TNF-α, IL-6, and MCP-1, thereby reinforcing the pro-inflammatory microenvironment.36 Our analysis further showed that serum GDF3 was independently associated with MCP-1 and TNF-α. TNF-α is a key mediator of insulin resistance and impairs insulin signaling through inflammation-related kinase pathways.37 MCP-1 acts as a chemotactic factor that recruits additional immune cells into adipose tissue and amplifies the inflammatory microenvironment, thereby establishing a self-perpetuating cycle that contributes to obesity pathogenesis.38 Notably, mediation analyses showed a significant partial mediation effect of GDF3 in the association between VFA and MCP-1, whereas no comparable effect was observed for TNF-α. This pattern suggests that GDF3 may be more involved in early inflammatory initiation (eg, macrophage recruitment) and in the development of insulin resistance rather than in later stages of inflammatory amplification. It should be noted that mediation analysis based on observational data represents a statistical decomposition under specific assumptions and does not establish causal mediation. Overall, our findings suggest a potential “visceral fat–GDF3–insulin resistance/inflammation axis”.

There are certain limitations that should be attention. First, the cross-sectional design does not allow for any causal inferences. Second, participants were recruited from a single region, and the overall sample size, particularly that of the intervention subcohort, was limited, which may reduce the generalisability of the findings. In our cohort, participants with visceral obesity were younger than those without visceral obesity. Although we adjusted for age in multivariable models, residual confounding may remain. Third, VFA was assessed using BIA rather than abdominal CT. Although BIA is feasible in clinical practice, it may introduce measurement error. In summary, we hope to conduct larger prospective studies in the future and validate our findings through mechanism-based experiments.

Conclusion

In conclusion, these findings support the potential utility of GDF3 as a circulating biomarker of visceral obesity. Serum GDF3 level is associated with insulin resistance and inflammation in non-diabetic adults.

Data Sharing Statement

The data supporting the findings of this study are available from the corresponding author on reasonable request.

Ethics Approval

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee of the First Affiliated Hospital of University of South China (protocol code 2022ll0901003, Date: 1 September 2022).

Acknowledgments

We thank all the research team members and participants in this cohort study for their contribution to this research.

Author Contributions

Zhaozhi Li: Conceptualization, Methodology, Software, Writing – Original Draft.

Jiaoyang Li: Conceptualization, Formal Analysis, Software, Writing – Original Draft.

Yun Ouyang: Data Curation, Investigation, Methodology, Writing – Original Draft.

Liyan Jiang: Data Curation, Investigation, Writing – Review & Editing.

Zhezhen Liao: Formal Analysis, Methodology, Validation, Writing – Review & Editing.

Li Ran: Data Curation, Methodology, Resources, Writing – Review & Editing.

Fei Yang: Project Administration, Visualization, Validation, Writing – Review & Editing.

Xin-hua Xiao: Conceptualization, Funding Acquisition, Project Administration, Supervision, Writing – Review & Editing.

Ya-di Wang: Conceptualization, Resources, Supervision, Writing – Review & Editing.

All authors gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This research was funded by grants from the National Natural Science Foundation of China (Grant No. 82470910 and No. 82571012), the Natural Science Foundation of Hunan province (Grant No. 2025JJ60483), Hunan Provincial Health Commission’s 2023 National Key Clinical Specialty Major Scientific Research Project (Grant No. z2023066), Hunan Provincial Health High-Level Talent Scientific Research Project (Grant No. R2023134) and Postgraduate Scientific Research Innovation Project of Hunan Province (Grant No. CX20251487).

Disclosure

The authors report no conflicts of interest in this work.

References

1. Phelps NH, Singleton RK, Zhou B, et al. Worldwide trends in underweight and obesity from 1990 to 2022: a pooled analysis of 3663 population-representative studies with 222 million children, adolescents, and adults. Lancet. 2024;403(10431):1027–12.

2. Hamdy O, Porramatikul S, Al-Ozairi E. Metabolic obesity: the paradox between visceral and subcutaneous fat. Curr Diabetes Rev. 2006;2(4):367–373. doi:10.2174/1573399810602040367

3. NHLBI Obesity Education Initiative Expert Panel on the Identification, Evaluation, and Treatment of Obesity in Adults (US). Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults--the evidence report. Obes Res. 1998;6(Suppl 2):51s–209s.

4. Neeland IJ, Ross R, Despres JP, et al. Visceral and ectopic fat, atherosclerosis, and cardiometabolic disease: a position statement. Lancet Diabetes Endocrinol. 2019;7(9):715–725. doi:10.1016/S2213-8587(19)30084-1

5. Wander PL, Boyko EJ, Leonetti DL, McNeely MJ, Kahn SE, Fujimoto WY. Change in visceral adiposity independently predicts a greater risk of developing type 2 diabetes over 10 years in Japanese Americans. Diabetes Care. 2013;36(2):289–293. doi:10.2337/dc12-0198

6. Haan YC, Oudman I, Diemer FS, et al. Creatine kinase as a marker of obesity in a multi-ethnic population. Mol Cell Endocrinol. 2017;442:24–31. doi:10.1016/j.mce.2016.11.022

7. Yu L, Bao S, Zhu F, et al. Serum lactate dehydrogenase is a novel predictor for the severity in the patients with MAFLD: a cross-sectional study in Hefei, China. Diabetes Metab Syndr Obes. 2025;18:345–361. doi:10.2147/DMSO.S492153

8. Antuna-Puente B, Feve B, Fellahi S, Bastard JP. Adipokines: the missing link between insulin resistance and obesity. Diabetes Metab. 2008;34(1):2–11. doi:10.1016/j.diabet.2007.09.004

9. Fontana L, Eagon JC, Trujillo ME, Scherer PE, Klein S. Visceral fat adipokine secretion is associated with systemic inflammation in obese humans. Diabetes. 2007;56(4):1010–1013. doi:10.2337/db06-1656

10. Ibáñez CF. Regulation of metabolic homeostasis by the TGF-β superfamily receptor ALK7. Febs j. 2021;289(19):5776–5797. doi:10.1111/febs.16090

11. Choi E, Lee J, Kim H, Kim Y-J, Kim SH. TGF-β superfamily-induced transcriptional activation pathways establish the RAD52-dependent ALT machinery during malignant transformation of MPNSTs. Sci Rep. 2024;14(1):26475. doi:10.1038/s41598-024-76732-z

12. Shen JJ, Huang L, Li L, Jorgez C, Matzuk MM, Brown CW. Deficiency of growth differentiation factor 3 protects against diet-induced obesity by selectively acting on white adipose. Mol Endocrinol. 2009;23(1):113–123. doi:10.1210/me.2007-0322

13. Hall JA, Ramachandran D, Roh HC, et al. Obesity-linked PPARγ S273 phosphorylation promotes insulin resistance through growth differentiation factor 3. Cell Metab. 2020;32(4):665–675.e6. doi:10.1016/j.cmet.2020.08.016

14. Kotikalapudi N, Ramachandran D, Vieira D, et al. Acute regulation of murine adipose tissue lipolysis and insulin resistance by the TGFβ superfamily protein GDF3. Nat Commun. 2025;16(1):4432. doi:10.1038/s41467-025-59673-7

15. Camell CD, Sander J, Spadaro O, et al. Inflammasome-driven catecholamine catabolism in macrophages blunts lipolysis during ageing. Nature. 2017;550(7674):119–123. doi:10.1038/nature24022

16. Andersson O, Korach-Andre M, Reissmann E, Ibáñez CF, Bertolino P. Growth/differentiation factor 3 signals through ALK7 and regulates accumulation of adipose tissue and diet-induced obesity. Proc Natl Acad Sci U S A. 2008;105(20):7252–7256. doi:10.1073/pnas.0800272105

17. Yogosawa S, Mizutani S, Ogawa Y, Izumi T. Activin receptor-like kinase 7 suppresses lipolysis to accumulate fat in obesity through downregulation of peroxisome proliferator-activated receptor γ and C/EBPα. Diabetes. 2013;62(1):115–123. doi:10.2337/db12-0295

18. Hu X, Dong X, Li G, et al. Brd4 modulates diet-induced obesity via PPARγ-dependent Gdf3 expression in adipose tissue macrophages. JCI Insight. 2021;6(7):e143379. doi:10.1172/jci.insight.143379

19. Xiang L, Li X, Luo Y, et al. A multi-omic landscape of steatosis-to-NASH progression. Life Metab. 2022;1(3):242–257. doi:10.1093/lifemeta/loac034

20. Kadowaki T, Isendahl J, Khalid U, et al. Semaglutide once a week in adults with overweight or obesity, with or without type 2 diabetes in an east Asian population (STEP 6): a randomised, double-blind, double-dummy, placebo-controlled, phase 3a trial. Lancet Diabetes Endocrinol. 2022;10(3):193–206. doi:10.1016/S2213-8587(22)00008-0

21. Zheng J, Hu Y, Xu H, et al. Normal-weight visceral obesity promotes a higher 10-year atherosclerotic cardiovascular disease risk in patients with type 2 diabetes mellitus-a multicenter study in China. Cardiovasc Diabetol. 2023;22(1):137. doi:10.1186/s12933-023-01876-7

22. Chen Y, Jia H, Ye R, et al. Low visceral adipose tissue predicts the outcome of neoadjuvant chemotherapy for colorectal liver metastases: a multicenter real-world study. J Cachexia Sarcopenia Muscle. 2025;16(2):e13785. doi:10.1002/jcsm.13785

23. Ryo M, Maeda K, Onda T, et al. A new simple method for the measurement of visceral fat accumulation by bioelectrical impedance. Diabetes Care. 2005;28(2):451–453. doi:10.2337/diacare.28.2.451

24. Yao F, Zeng L, Hua M, et al. Association of epicardial and visceral adipose tissue in relation to subclinical cardiac dysfunction in Chinese: Danyang study. BMJ Open. 2023;13(12):e075576. doi:10.1136/bmjopen-2023-075576

25. Zhang JJ, Yu HC, Li Y, et al. Association between serum 25-hydroxy vitamin D concentrations and mortality among individuals with metabolic dysfunction-associated fatty liver disease: a prospective cohort study. Am J Clin Nutr. 2022;116(5):1409–1417. doi:10.1093/ajcn/nqac260

26. Jeon HH, Lee YK, Kim DH, Pak H, Shin SY, Seo JH. Risk for metabolic syndrome in the population with visceral fat area measured by bioelectrical impedance analysis. Korean J Intern Med. 2021;36(1):97–105. doi:10.3904/kjim.2018.427

27. Lee YC, Lee YH, Chuang PN, Kuo CS, Lu CW, Yang KC. The utility of visceral fat level measured by bioelectrical impedance analysis in predicting metabolic syndrome. Obes Res Clin Pract. 2020;14(6):519–523. doi:10.1016/j.orcp.2020.09.008

28. Hanlon CL, Yuan L. Nonalcoholic fatty liver disease: the role of visceral adipose tissue. Clin Liver Dis. 2022;19(3):106–110. doi:10.1002/cld.1183

29. Masurkar N, Bouvet M, Logeart D, et al. Novel cardiokine GDF3 predicts adverse fibrotic remodeling after myocardial infarction. Circulation. 2023;147(6):498–511. doi:10.1161/CIRCULATIONAHA.121.056272

30. Wang W, Yang Y, Meng Y, Shi Y. GDF-3 is an adipogenic cytokine under high fat dietary condition. Biochem Biophys Res Commun. 2004;321(4):1024–1031. doi:10.1016/j.bbrc.2004.07.058

31. Bu Y, Okunishi K, Yogosawa S, et al. Insulin regulates lipolysis and fat mass by upregulating growth/differentiation factor 3 in adipose tissue macrophages. Diabetes. 2018;67(9):1761–1772. doi:10.2337/db17-1201

32. Pham H, Hui H, Morvaridi S, et al. A bitter pill for type 2 diabetes? The activation of bitter taste receptor TAS2R38 can stimulate GLP-1 release from enteroendocrine L-cells. Biochem Biophys Res Commun. 2016;475(3):295–300. doi:10.1016/j.bbrc.2016.04.149

33. Nauck M. Incretin therapies: highlighting common features and differences in the modes of action of glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase-4 inhibitors. Diabetes Obesity Metab. 2016;18(3):203–216. doi:10.1111/dom.12591

34. Gastaldelli A, Miyazaki Y, Pettiti M, et al. Metabolic effects of visceral fat accumulation in type 2 diabetes. J Clin Endocrinol Metab. 2002;87(11):5098–5103. doi:10.1210/jc.2002-020696

35. Xu H, Barnes GT, Yang Q, et al. Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance. J Clin Invest. 2003;112(12):1821–1830. doi:10.1172/JCI200319451

36. Chawla A, Nguyen KD, Goh YPS. Macrophage-mediated inflammation in metabolic disease. Nat Rev Immunol. 2011;11(11):738–749. doi:10.1038/nri3071

37. Hotamisligil GS, Arner P, Caro JF, Atkinson RL, Spiegelman BM. Increased adipose tissue expression of tumor necrosis factor-alpha in human obesity and insulin resistance. J Clin Invest. 1995;95(5):2409–2415. doi:10.1172/JCI117936

38. Kanda H. MCP-1 contributes to macrophage infiltration into adipose tissue, insulin resistance, and hepatic steatosis in obesity. J Clin Investig. 2006;116(6):1494–1505. doi:10.1172/JCI26498