Roux-en-Y gastric bypass promotes white adipose tissue browning and systemic metabolic recovery in obese type 2 diabetic mice Authors: Tengfei Qi, Ru Ji, Liangping Wu, Jipei He, Hongbin Zhang, Zhengyong Xie Published: 2026 ABSTRACT ObjectiveRoux-en-Y gastric bypass (RYGB) is a highly effective therapy for obesity and type 2 diabetes mellitus (T2DM), but its underlying mechanisms, particularly regarding adipose tissue remodeling, remain incompletely understood. This study investigated the hypothesis that RYGB induces a phenotypic shift in white adipose tissue (WAT) towards a brown-like (beige) state, which contributes to its systemic metabolic benefits.MethodsAn obese T2DM model was established in C57BL/6J mice using a high-fat diet and low-dose streptozotocin. Mice were allocated to control (CON), sham-operated (SHAM), or RYGB groups. Metabolic parameters were monitored for 8 weeks post-surgery. Terminal analyses included body composition assessment via micro-CT, plasma lipid profiling, and histological evaluation of liver and multiple adipose depots (epididymal, inguinal, perirenal, and brown adipose tissue).ResultsRYGB induced sustained weight loss, normalized glycemia, improved insulin sensitivity, and corrected dyslipidemia. Body composition analysis revealed a preferential reduction in visceral and subcutaneous fat mass, with lean mass preserved. Histologically, RYGB markedly alleviated hepatic steatosis. Crucially, RYGB induced a profound remodeling of WAT, characterized by a significant reduction in adipocyte size in both epididymal and inguinal depots, indicating improved adipose tissue health. Concomitantly, brown adipose tissue exhibited a reversal of obesity-associated “whitening,” displaying a more active morphology with smaller lipid droplets.ConclusionBeyond its established effects on weight and metabolism, RYGB induces morphological changes in white adipose tissue consistent with improved metabolic health. This structural and phenotypic reprogramming of fat depots likely represents a key cellular mechanism underpinning the surgery’s potent anti-diabetic and hepatoprotective effects. Our findings highlight adipose tissue plasticity as a central therapeutic target of RYGB.Clinical trial numberNot applicable.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12902-026-02319-6. Introduction The global epidemic of obesity is a primary driver of type 2 diabetes mellitus (T2DM) [1], with adipose tissue dysfunction playing a central role in the pathogenesis of insulin resistance and metabolic syndrome [2]. In obesity, white adipose tissue (WAT) undergoes pathological expansion characterized by adipocyte hypertrophy, hypoxia, and chronic low-grade inflammation, which disrupts systemic metabolic homeostasis [3–5]. This adipose tissue inflammation is a key contributor to systemic insulin resistance and the progression of T2DM [6]. Notably, BAT, which exhibits greater resistance to inflammation compared to WAT [7, 8], is often found to be reduced in volume and functionally impaired in individuals with obesity and T2DM [9]. Bariatric surgery, particularly Roux-en-Y gastric bypass (RYGB), has proven to be a highly effective treatment not only for severe obesity but also for T2DM. Since Pories et al. first reported the long-term remission of diabetes following RYGB [10], subsequent research has shown that its benefits extend beyond weight loss to include a substantial reduction in adipose tissue mass [11], a decrease in adipocyte size [12], alterations in adipokine secretion [13], and a mitigation of chronic low-grade inflammation [14]. Accurately assessing changes in body composition is essential for evaluating the metabolic effects of interventions like RYGB. Traditional metrics like body mass index (BMI) are limited in their ability to differentiate between fat and lean mass or to delineate fat distribution [15, 16]. In contrast, computed tomography (CT) allows for precise, three-dimensional quantification of adipose tissue. Single-slice CT imaging at the level of the third lumbar vertebra (L3) has been validated as a reliable and practical method, showing high correlation with whole-body adipose tissue volumes, and is therefore widely used in body composition analysis [17, 18]. Based on this evidence, the aim of the present study was to comprehensively investigate the effects of RYGB on a range of metabolic parameters and tissue-specific alterations in a murine model of obese T2DM. We specifically focused on its impact on body weight, glucose and insulin metabolism, plasma lipid profile, hepatic steatosis, and systemic fat distribution and cellular morphology. Materials & methods Experimental animals Male C57BL/6 mice (n = 30, 8-week-old, 23–26 g) were purchased from Shanghai Southern Model Biological Co., Ltd. [SCXK (Shanghai) 2018-0007] and housed under Specific Pathogen-Free (SPF) conditions (22 ± 2 °C, 60 ± 5% humidity, 12/12-h light/dark cycle) with free access to standard chow and water. All animal experiments were approved by the Animal Ethics Committee of the General Hospital of Southern Theater Command and conducted in accordance with the ARRIVE guidelines. Experimental design Animal grouping and T2DM model induction After a one-week acclimation period, 18 mice completed the 8-week protocol and were divided into three groups (n = 6 each): control (CON), sham-operated (SHAM), and Roux-en-Y gastric bypass (RYGB). Type 2 diabetes mellitus (T2DM) was induced in the SHAM and RYGB groups using an established method combining a high-fat diet (HFD; Dyets, Bethlehem, PA, USA; HF60, 5.26 kcal/g) from 8 to 14 weeks of age, followed by intraperitoneal injection of streptozotocin (STZ; Sigma-Aldrich, St. Louis, MO, USA; S0130-100MG, 40 mg/kg for 5 consecutive days) [19, 20]. Mice with fasting blood glucose (FBG) levels persistently > 16.7 mmol/L were considered successful T2DM models. The CON group was maintained on a standard diet. Surgical procedures Mice in the RYGB group underwent Roux-en-Y gastric bypass surgery under anesthesia (Zoletil 50 (Virbac, Carros, France; FBS060) and Xylazine hydrochloride (Sigma-Aldrich, St. Louis, MO, USA; X1251-5G), 20 mg/kg each, i.p.). Briefly, a proximal gastric pouch (≈ 10% of total volume) was created. The small intestine was transected 4 cm distal to the Treitz ligament, and anastomoses were constructed to establish the RYGB anatomy [21]. Mice in the SHAM group received identical anesthesia and surgical trauma, including gastrotomy and intestinal incision with suturing, but without altering the gastrointestinal anatomy. The CON group received no surgery. Sample collection and biochemical analysis At 8 weeks after surgery, mice were euthanized via isoflurane inhalation. Blood was collected via cardiac puncture. Plasma was isolated and stored at -80 °C for subsequent analysis. Tissues (liver, epididymal, brown, inguinal, and perirenal fat) were harvested and preserved. Plasma levels of total cholesterol (TC), triglycerides (TG), high-density lipoprotein (HDL), and low-density lipoprotein (LDL) were measured using ELISA according to the manufacturer’s instructions. Fasting blood glucose was measured using a glucometer (Roche) after 6 h of food deprivation. Plasma insulin levels were determined by ELISA according to the manufacturer’s instructions. Insulin resistance was assessed by the homeostasis model assessment of insulin resistance (HOMA-IR) using the formula: [fasting glucose (mmol/L) × fasting insulin (mU/L)] / 22.5. Quantitative real-time PCR was performed to assess the mRNA expression of inflammatory markers (IL-6, TLR4, TNFα) and the thermogenic marker UCP1. GAPDH was used as an internal control, and relative expression levels were calculated using the 2^−ΔΔCt method. Micro-CT imaging Body composition was analyzed at the L3 vertebral level using micro-CT (scanning parameters: 120 kVP, 5.0 mm thickness). Adipose tissue was defined within a threshold of -195 to -45 Hounsfield Units [22]. Visceral fat area (VFA), subcutaneous fat area (SFA), and lean body mass area were quantified using RadiAnt DICOM viewer software (Medixant, Poland). Plasma biochemical analysis by ELISA Fasting blood glucose was measured using a glucometer (Roche) after 6 h of food deprivation. Blood samples were collected into EDTA-coated tubes, centrifuged at 4 °C for 15 min at 4,500 rpm, and plasma was stored at -80 °C. Plasma insulin levels were measured using a mouse insulin ELISA kit (MEIMIAN, Cat# MM-45303M2) following the manufacturer’s protocol. Plasma levels of total cholesterol (TC), triglycerides (TG), high-density lipoprotein (HDL), and low-density lipoprotein (LDL) were measured using commercial ELISA kits (MEIMIAN; Cat# MM-0632M2 for TC, MM-0631M2 for TG, MM-43693M1 for HDL, and MM-44068M2 for LDL). All assays were performed according to the manufacturer’s instructions. Insulin resistance was assessed by the homeostasis model assessment of insulin resistance (HOMA-IR) using the formula: [fasting glucose (mmol/L) × fasting insulin (mU/L)] / 22.5. Quantitative real-time PCR analysis Total RNA was extracted from liver tissue using TRIzol™ Reagent (Invitrogen, Carlsbad, CA, USA; 15596026CN) according to the manufacturer’s protocol. RNA concentration and purity were assessed by spectrophotometry (NanoDrop 2000, Thermo Fisher Scientific). Reverse transcription was performed using the HiScript III RT SuperMix kit (Vazyme, Cat# R323) with 1 µg of total RNA. Quantitative real-time PCR was performed using ChamQ Universal SYBR qPCR Master Mix (Vazyme, Cat# Q711) on a QuantStudio 5 System (Applied Biosystems). The 20 µL reaction mixture contained 10 µL SYBR Master Mix, 0.4 µL each of forward and reverse primers (10 µM), 2 µL cDNA, and 7.2 µL DEPC water. Cycling conditions were: 95 °C for 30 s, followed by 40 cycles of 95 °C for 10 s and 60 °C for 30 s. Melt curve analysis was performed to verify primer specificity. The following genes were analyzed: IL-6, TLR4, TNF-α and UCP1. GAPDH was used as an internal control. Primer sequences are listed in Table 1. Relative mRNA expression levels were calculated using the 2^−ΔΔCt method and expressed as fold change relative to the CON group. Table 1Primer sequences for quantitative real-time PCRGenePrimer Sequence (5’ to 3’)IL-6-FCTCCCAACAGACCTGTCTATACIL-6-RCCATTGCACAACTCTTTTCTCATLR4-FATGGCATGGCTTACACCACCTLR4-RGAGGCCAATTTTGTCTCCACATNF-α-FCGAGTGACAAGCCTGTAGCCCTNF-α-RGTCTTTGAGATCCATGCCGTTGUCP1-FCCAGGCTTCCAGTACTATTAGGUCP1-RGTACTCTTGGACCGTATCGTAGGAPDH-FAACGGCACAGTCAAGGCTGAGAPDH-RACGCCAGTAGACTCCACGACAT Table 1Primer sequences for quantitative real-time PCRGenePrimer Sequence (5’ to 3’)IL-6-FCTCCCAACAGACCTGTCTATACIL-6-RCCATTGCACAACTCTTTTCTCATLR4-FATGGCATGGCTTACACCACCTLR4-RGAGGCCAATTTTGTCTCCACATNF-α-FCGAGTGACAAGCCTGTAGCCCTNF-α-RGTCTTTGAGATCCATGCCGTTGUCP1-FCCAGGCTTCCAGTACTATTAGGUCP1-RGTACTCTTGGACCGTATCGTAGGAPDH-FAACGGCACAGTCAAGGCTGAGAPDH-RACGCCAGTAGACTCCACGACAT Primer sequences for quantitative real-time PCR Histological analysis Liver and adipose tissue samples (epididymal, inguinal, perirenal, and brown adipose tissue) were collected immediately after euthanasia, rinsed in ice-cold PBS, and fixed in 4% paraformaldehyde for 24 h at 4 °C. For hematoxylin and eosin (H&E) staining, tissues were dehydrated through a graded ethanol series, cleared in xylene, embedded in paraffin, and sectioned at 5 μm thickness using a rotary microtome (Leica RM2235). Sections were deparaffinized, rehydrated, stained with hematoxylin (Sigma-Aldrich, Cat# MHS16) for 5 min, rinsed, and counterstained with eosin (Sigma-Aldrich, Cat# HT110116) for 2 min, then dehydrated and mounted with permanent mounting medium. For Oil Red O staining, fresh liver tissues were embedded in OCT compound, frozen at -80 °C, and sectioned at 8 μm thickness using a cryostat (Leica CM1950). Sections were fixed in 10% formalin, rinsed, immersed in 60% isopropanol, and stained with Oil Red O working solution (Sigma-Aldrich, Cat# O1391) for 15 min, followed by hematoxylin counterstaining and mounting with aqueous mounting medium. Statistical analysis All data were recorded using Microsoft Excel software and analyzed with SPSS 26.0. The normality of continuous variables was assessed using the Shapiro–Wilk test. Data following a normal distribution are presented as mean ± standard deviation (SD), while non-normally distributed data are expressed as median with interquartile range (Q1, Q3). For comparisons between two groups, Student’s t-test was used for normally distributed data, and the Mann–Whitney U test was applied for non-normally distributed data. For multiple group comparisons, one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test was performed for normally distributed data with homogeneous variances; if variances were unequal, Welch’s ANOVA with Games–Howell post hoc test was used. For non-normally distributed data, the Kruskal–Wallis test with Dunn’s post hoc test was employed. Categorical variables are presented as frequencies and percentages. A two-tailed P value < 0.05 was considered statistically significant. Graphical illustrations were generated using GraphPad Prism 9. Results Effect of RYGB on body weight of mice modeling type 2 diabetes mellitus As shown in Fig. 1, body weight increased progressively in the CON group. Before the STZ intervention (weeks 8–14), the SHAM and RYGB groups showed comparable and significant weight gain compared to the CON group (P > 0.05 for SHAM vs. RYGB). Post-STZ, body weight in the SHAM and RYGB groups decreased and then stabilized. Following surgery, the RYGB group exhibited a sharp decline in body weight within the first week, which continued until week 6 before stabilizing. At the endpoint, the body weight of the RYGB group was significantly lower than that of both the SHAM and CON groups (P < 0.0001 for both). The SHAM group also remained significantly heavier than the CON group (P < 0.0001). Fig. 1Body weight changes over time Body weight changes over time Body weight was monitored weekly from baseline (8 weeks of age) to 8 weeks post-surgery. Mice in SHAM and RYGB groups were fed a high-fat diet (HFD) from weeks 8–14 and received low-dose streptozotocin (STZ) injections (40 mg/kg for 5 consecutive days) to induce type 2 diabetes mellitus (T2DM). CON group mice were maintained on standard chow throughout the study. RYGB or sham surgery was performed at week 17. Data are presented as mean ± SD (n = 6 per group). At the endpoint (8 weeks post-surgery, week 25), body weight in the RYGB group was significantly lower than both the SHAM and CON groups (P < 0.01 for both). The SHAM group remained significantly heavier than the CON group (P < 0.01). ns, not significant; *P < 0.05, **P < 0.01, ***P < 0.001 indicate significant differences between indicated groups at specific time points Effects of RYGB on general condition and fasting blood glucose in mice modeled with type 2 diabetes mellitus As illustrated in Fig. 2, fasting blood glucose (FBG) levels were comparable among all groups before STZ injection. Post-STZ, FBG levels in the SHAM and RYGB groups surged to > 16.7 mmol/L, confirming successful T2DM induction. After surgery, the RYGB group showed a rapid and significant improvement in FBG compared to the SHAM group (P < 0.0001), with levels declining until week 4 and stabilizing thereafter. In contrast, FBG in the SHAM group remained persistently high (> 20 mmol/L) throughout the study. Fig. 2Fasting blood glucose levels over time. Fasting blood glucose was measured weekly after 6 h of food deprivation. T2DM was induced in SHAM and RYGB groups by HFD feeding (weeks 8–14) followed by STZ injections (40 mg/kg for 5 consecutive days). RYGB or sham surgery was performed at week 17. Data are presented as mean ± SD (n = 6 per group). ****P < 0.0001 for RYGB vs. SHAM at indicated time points Fasting blood glucose levels over time. Fasting blood glucose was measured weekly after 6 h of food deprivation. T2DM was induced in SHAM and RYGB groups by HFD feeding (weeks 8–14) followed by STZ injections (40 mg/kg for 5 consecutive days). RYGB or sham surgery was performed at week 17. Data are presented as mean ± SD (n = 6 per group). ****P < 0.0001 for RYGB vs. SHAM at indicated time points RYGB improves insulin sensitivity assessed by insulin tolerance test and HOMA-IR in mice modeling type 2 diabetes mellitus To evaluate the effects of RYGB on insulin sensitivity, we performed an insulin tolerance test (ITT) at 8 weeks after surgery and calculated the homeostasis model assessment of insulin resistance (HOMA-IR) based on fasting glucose and insulin levels. As shown in Fig. 3A, the ITT revealed significantly higher blood glucose levels at all time points (0–120 min) and a greater area under the curve (AUC) in the SHAM group compared to the CON group, confirming severe insulin resistance in the T2DM model. Conversely, the RYGB group demonstrated markedly improved glucose disposal and a significantly lower AUC than the SHAM group (P < 0.0001), indicating enhanced systemic insulin sensitivity. Consistently, fasting insulin levels were significantly lower in the RYGB group compared to the SHAM group (P < 0.001). HOMA-IR analysis (Fig. 3B) showed that the SHAM group had significantly elevated HOMA-IR values compared to the CON group (P < 0.001), reflecting profound insulin resistance. RYGB surgery dramatically reduced HOMA-IR to levels comparable to the CON group (P < 0.001 vs. SHAM; P > 0.05 vs. CON). These data demonstrate that RYGB effectively reverses insulin resistance in obese T2DM mice, as evidenced by both dynamic (ITT) and static (HOMA-IR) assessments of insulin sensitivity. Fig. 3AInsulin tolerance test (ITT) at 8 weeks post-surgery. After 4 h of fasting, mice received an intraperitoneal injection of insulin (0.75 U/kg). Blood glucose levels were measured at 0, 15, 30, 45, 60, 90 and 120 min post-injection and expressed as percentage of baseline glucose. Data are presented as mean ± SD (n = 6 per group). ****P < 0.0001 for RYGB vs. SHAM at each time point Insulin tolerance test (ITT) at 8 weeks post-surgery. After 4 h of fasting, mice received an intraperitoneal injection of insulin (0.75 U/kg). Blood glucose levels were measured at 0, 15, 30, 45, 60, 90 and 120 min post-injection and expressed as percentage of baseline glucose. Data are presented as mean ± SD (n = 6 per group). ****P < 0.0001 for RYGB vs. SHAM at each time point Fig. 3BHomeostasis model assessment of insulin resistance (HOMA-IR) at 8 weeks post-surgery. HOMA-IR was calculated from fasting glucose and fasting insulin levels using the formula: [fasting glucose (mmol/L) × fasting insulin (mU/L)] / 22.5. Data are presented as mean ± SD with individual data points shown (n = 6 per group). ***P < 0.001; ns, not significant; one-way ANOVA with Tukey’s post hoc test Homeostasis model assessment of insulin resistance (HOMA-IR) at 8 weeks post-surgery. HOMA-IR was calculated from fasting glucose and fasting insulin levels using the formula: [fasting glucose (mmol/L) × fasting insulin (mU/L)] / 22.5. Data are presented as mean ± SD with individual data points shown (n = 6 per group). ***P < 0.001; ns, not significant; one-way ANOVA with Tukey’s post hoc test Effects of metabolic bariatric surgery on body composition of mice Body composition analysis via micro-CT demonstrated that RYGB surgery induced a profound reduction in adiposity (Fig. 4A). Compared to the SHAM group, RYGB significantly decreased both subcutaneous (SWAT, P < 0.001) and visceral (VWAT, P < 0.01) white adipose tissue areas (Fig. 4B). Notably, the reduction in VWAT was so pronounced that the RYGB group exhibited significantly lower levels than even the lean CON group (P < 0.05), highlighting a preferential targeting of metabolically harmful visceral fat. Crucially, these changes occurred without any significant loss of lean body mass across all groups. Fig. 4ARepresentative micro-CT images at the L3 vertebral level. Images were acquired at 8 weeks post-surgery from CON, SHAM, and RYGB groups. Adipose and lean tissues were segmented based on Hounsfield Unit (HU) thresholds using RadiAnt DICOM viewer software. Adipose tissue was identified within the range of -195 to -45 HU, while muscle tissue was identified within the range of -29 to + 150 HU, according to established protocols. Following segmentation, tissues were pseudo-colored for visualization: muscle (red), visceral white adipose tissue (VWAT, green), and subcutaneous white adipose tissue (SWAT, yellow). Images shown are representative of n = 3 mice per group Representative micro-CT images at the L3 vertebral level. Images were acquired at 8 weeks post-surgery from CON, SHAM, and RYGB groups. Adipose and lean tissues were segmented based on Hounsfield Unit (HU) thresholds using RadiAnt DICOM viewer software. Adipose tissue was identified within the range of -195 to -45 HU, while muscle tissue was identified within the range of -29 to + 150 HU, according to established protocols. Following segmentation, tissues were pseudo-colored for visualization: muscle (red), visceral white adipose tissue (VWAT, green), and subcutaneous white adipose tissue (SWAT, yellow). Images shown are representative of n = 3 mice per group Fig. 4BQuantitative analysis of adipose tissue and muscle areas at the L3 vertebral level. Subcutaneous white adipose tissue (SWAT), visceral white adipose tissue (VWAT), and muscle areas were quantified from micro-CT images using RadiAnt DICOM viewer software. Adipose tissue was identified using Hounsfield Unit thresholds (-195 to -45 HU), and muscle tissue was identified within the range of -29 to + 150 HU. Data are presented as mean ± SD with individual data points shown (n = 3 per group). *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant; one-way ANOVA with Tukey’s post hoc test Quantitative analysis of adipose tissue and muscle areas at the L3 vertebral level. Subcutaneous white adipose tissue (SWAT), visceral white adipose tissue (VWAT), and muscle areas were quantified from micro-CT images using RadiAnt DICOM viewer software. Adipose tissue was identified using Hounsfield Unit thresholds (-195 to -45 HU), and muscle tissue was identified within the range of -29 to + 150 HU. Data are presented as mean ± SD with individual data points shown (n = 3 per group). *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant; one-way ANOVA with Tukey’s post hoc test Histological analysis of different tissues Effects of RYGB on the liver of mice modeling type 2 diabetes mellitus Histological analysis confirmed severe hepatic steatosis in the SHAM group, which was effectively reversed by RYGB. H&E staining showed disorganized, lipid-engorged hepatocytes in SHAM mice, whereas liver architecture in the RYGB group was restored to a state comparable to the CON group (Fig. 5A). Quantification of Oil Red O staining confirmed a significant increase in lipid droplets in SHAM mice (P < 0.0001 vs. CON), which was normalized to control levels after RYGB (P < 0.001 vs. SHAM; Fig. 5B). Fig. 5ARepresentative hematoxylin and eosin (H&E) staining of liver sections. Liver sections from CON, SHAM, and RYGB groups were collected at 8 weeks post-surgery and stained with H&E to assess hepatic morphology. Scale bar = 200/50 µm Representative hematoxylin and eosin (H&E) staining of liver sections. Liver sections from CON, SHAM, and RYGB groups were collected at 8 weeks post-surgery and stained with H&E to assess hepatic morphology. Scale bar = 200/50 µm Fig. 5BOil Red O staining and quantification of hepatic steatosis. Representative Oil Red O staining of liver sections from CON, SHAM, and RYGB groups. Lipid droplets appear red. Scale bar = 200/50 µm. quantification of Oil Red O-positive area (%) measured using ImageJ software on five random fields per section. Data are presented as mean ± SD with individual data points shown (n = 6 per group). ***P < 0.001, ****P < 0.0001; ns, not significant; one-way ANOVA with Tukey’s post hoc test Oil Red O staining and quantification of hepatic steatosis. Representative Oil Red O staining of liver sections from CON, SHAM, and RYGB groups. Lipid droplets appear red. Scale bar = 200/50 µm. quantification of Oil Red O-positive area (%) measured using ImageJ software on five random fields per section. Data are presented as mean ± SD with individual data points shown (n = 6 per group). ***P < 0.001, ****P < 0.0001; ns, not significant; one-way ANOVA with Tukey’s post hoc test Effect of RYGB on adipose tissue in mouse model of type 2 diabetes mellitus Histomorphometric analysis across all adipose depots revealed consistent and dramatic effects of RYGB on adipocyte size, a key indicator of metabolic health. Epididymal WAT: Adipocytes in the SHAM group were markedly hypertrophied compared to the CON group (P < 0.0001). RYGB not only reversed this hypertrophy (P < 0.0001 vs. SHAM) but resulted in adipocytes that were significantly smaller and more compact than those in the CON group (P < 0.01), suggesting a profound remodeling beyond simple normalization (Fig. 6A). RYGB resulted in significantly smaller and more compact adipocytes compared to both SHAM and CON groups (P < 0.01 vs. CON), indicating profound remodeling of this depot. Inguinal WAT: A similar pattern was observed in the subcutaneous inguinal depot. SHAM group adipocytes were significantly enlarged (P < 0.0001 vs. CON), and RYGB induced a robust reduction in cell size (P < 0.0001 vs. SHAM), yielding smaller and more uniform adipocytes than in the CON group (P < 0.05; Fig. 6B). Perirenal WAT: In this visceral depot, RYGB again significantly ameliorated adipocyte hypertrophy observed in the SHAM group (P < 0.001 vs. CON), leading to a dense population of small adipocytes in the RYGB group (P < 0.0001 vs. SHAM and P < 0.05 vs. CON; Fig. 6C). Brown Adipose Tissue (BAT): RYGB also significantly remodeled BAT. The SHAM group showed enlarged, lipid-laden adipocytes, a sign of “whitening” (P < 0.01 vs. CON). In contrast, the RYGB group exhibited a striking reversal, with adipocytes containing numerous, smaller lipid droplets, higher staining intensity, and significantly reduced cell size (P < 0.0001 vs. SHAM and P < 0.05 vs. CON), indicating reversal of obesity-associated BAT whitening (Fig. 6D). Fig. 6AHistomorphology of epididymal white adipose tissue (eWAT). Representative H&E staining of eWAT from CON, SHAM, and RYGB groups at 8 weeks post-surgery. Scale bar = 200/50 µm. Quantitative analysis of adipocyte cross-sectional area measured using ImageJ software. Data are presented as mean ± SD with individual data points showing the average adipocyte area per mouse (n = 6 per group). **P < 0.01, ****P < 0.0001; one-way ANOVA with Tukey’s post hoc test Histomorphology of epididymal white adipose tissue (eWAT). Representative H&E staining of eWAT from CON, SHAM, and RYGB groups at 8 weeks post-surgery. Scale bar = 200/50 µm. Quantitative analysis of adipocyte cross-sectional area measured using ImageJ software. Data are presented as mean ± SD with individual data points showing the average adipocyte area per mouse (n = 6 per group). **P < 0.01, ****P < 0.0001; one-way ANOVA with Tukey’s post hoc test Fig. 6BHistomorphology of epididymal white adipose tissue (iWAT). Representative H&E staining of iWAT from CON, SHAM, and RYGB groups at 8 weeks post-surgery. Scale bar = 200/50 µm. Quantitative analysis of adipocyte cross-sectional area measured using ImageJ software. Data are presented as mean ± SD with individual data points showing the average adipocyte area per mouse (n = 6 per group). **P < 0.01, ****P < 0.0001; one-way ANOVA with Tukey’s post hoc test Histomorphology of epididymal white adipose tissue (iWAT). Representative H&E staining of iWAT from CON, SHAM, and RYGB groups at 8 weeks post-surgery. Scale bar = 200/50 µm. Quantitative analysis of adipocyte cross-sectional area measured using ImageJ software. Data are presented as mean ± SD with individual data points showing the average adipocyte area per mouse (n = 6 per group). **P < 0.01, ****P < 0.0001; one-way ANOVA with Tukey’s post hoc test Fig. 6CHistomorphology of epididymal white adipose tissue (pWAT). Representative H&E staining of pWAT from CON, SHAM, and RYGB groups at 8 weeks post-surgery. Scale bar = 200/50 µm. Quantitative analysis of adipocyte cross-sectional area measured using ImageJ software. Data are presented as mean ± SD with individual data points showing the average adipocyte area per mouse (n = 6 per group). **P < 0.01, ****P < 0.0001; one-way ANOVA with Tukey’s post hoc test Histomorphology of epididymal white adipose tissue (pWAT). Representative H&E staining of pWAT from CON, SHAM, and RYGB groups at 8 weeks post-surgery. Scale bar = 200/50 µm. Quantitative analysis of adipocyte cross-sectional area measured using ImageJ software. Data are presented as mean ± SD with individual data points showing the average adipocyte area per mouse (n = 6 per group). **P < 0.01, ****P < 0.0001; one-way ANOVA with Tukey’s post hoc test Fig. 6DHistomorphology of epididymal white adipose tissue (BAT). Representative H&E staining of BAT from CON, SHAM, and RYGB groups at 8 weeks post-surgery. Scale bar = 200/50 µm. Quantitative analysis of adipocyte cross-sectional area measured using ImageJ software. Data are presented as mean ± SD with individual data points showing the average adipocyte area per mouse (n = 6 per group). **P < 0.01, ****P < 0.0001; one-way ANOVA with Tukey’s post hoc test Histomorphology of epididymal white adipose tissue (BAT). Representative H&E staining of BAT from CON, SHAM, and RYGB groups at 8 weeks post-surgery. Scale bar = 200/50 µm. Quantitative analysis of adipocyte cross-sectional area measured using ImageJ software. Data are presented as mean ± SD with individual data points showing the average adipocyte area per mouse (n = 6 per group). **P < 0.01, ****P < 0.0001; one-way ANOVA with Tukey’s post hoc test Effects of RYGB on lipid profiles in type 2 diabetic model mice Serum lipid analysis revealed pronounced dyslipidemia in the SHAM group, characterized by elevated triglycerides (TG), total cholesterol (TC), and low-density lipoprotein (LDL) levels compared to the CON group (P < 0.001 for all, Fig. 7). RYGB surgery effectively reversed these abnormalities, significantly reducing TG, TC, and LDL levels while increasing high-density lipoprotein (HDL) compared to the SHAM group (P < 0.001 for all). Fig. 7RYGB corrects dyslipidemia in obese T2DM mice. Plasma levels of total cholesterol (TC), triglycerides (TG), high-density lipoprotein (HDL) and low-density lipoprotein (LDL), measured by ELISA at 8 weeks post-surgery. Data are presented as mean ± SD with individual data points shown (n = 6 per group). *P < 0.05, **P < 0.01, ***P < 0.001; one-way ANOVA with Tukey’s post hoc test RYGB corrects dyslipidemia in obese T2DM mice. Plasma levels of total cholesterol (TC), triglycerides (TG), high-density lipoprotein (HDL) and low-density lipoprotein (LDL), measured by ELISA at 8 weeks post-surgery. Data are presented as mean ± SD with individual data points shown (n = 6 per group). *P < 0.05, **P < 0.01, ***P < 0.001; one-way ANOVA with Tukey’s post hoc test RYGB attenuates hepatic inflammation in obese T2DM mice Chronic low-grade inflammation is a hallmark of obesity-associated metabolic dysfunction, and the liver is a key organ affected by inflammatory stress in the setting of non-alcoholic fatty liver disease (NAFLD). To investigate whether RYGB modulates the inflammatory state of the liver, we measured the mRNA expression levels of pro-inflammatory cytokines (IL-6, TNFα) and the innate immune receptor TLR4 in liver tissue by quantitative real-time PCR at 8 weeks post-surgery (Fig. 8). The SHAM group exhibited significantly elevated mRNA expression of IL-6, TLR4, and TNFα compared to the CON group (P < 0.001 for all; Fig. 14A–C), confirming the presence of obesity-induced hepatic inflammation. Remarkably, RYGB surgery robustly suppressed the expression of all three inflammatory markers. The mRNA levels of IL-6, TLR4, and TNFα in the RYGB group were significantly lower than those in the SHAM group (P < 0.001) and were restored to levels comparable to or even lower than those in the CON group. These results demonstrate that RYGB effectively attenuates hepatic inflammation, which likely contributes to the improvement in liver histology and systemic insulin sensitivity observed after surgery. Fig. 8RYGB attenuates hepatic inflammation. Relative mRNA expression of pro-inflammatory mediators in liver tissue at 8 weeks post-surgery, determined by quantitative real-time PCR. IL-6, TLR4, and TNFα mRNA levels were normalized to GAPDH. Data are presented as mean ± SD with individual data points shown (n = 6 per group). **P < 0.01, ***P < 0.001; one-way ANOVA with Tukey’s post hoc test RYGB attenuates hepatic inflammation. Relative mRNA expression of pro-inflammatory mediators in liver tissue at 8 weeks post-surgery, determined by quantitative real-time PCR. IL-6, TLR4, and TNFα mRNA levels were normalized to GAPDH. Data are presented as mean ± SD with individual data points shown (n = 6 per group). **P < 0.01, ***P < 0.001; one-way ANOVA with Tukey’s post hoc test RYGB upregulates UCP1 mRNA expression in brown and white adipose tissue To investigate whether RYGB induces thermogenic gene expression, we measured UCP1 mRNA levels by quantitative real-time PCR in interscapular brown adipose tissue (BAT) and inguinal white adipose tissue (iWAT) at 8 weeks post-surgery (Fig. 9). In BAT, UCP1 mRNA expression was markedly suppressed in the SHAM group compared to CON (P < 0.001), reflecting obesity-induced BAT dysfunction. RYGB surgery significantly upregulated UCP1 expression, which was significantly higher than SHAM (P < 0.001) but remained significantly lower than CON (P < 0.001). This partial restoration indicates that RYGB partially reverses BAT whitening. In iWAT, a similar pattern was observed. UCP1 mRNA levels were significantly reduced in SHAM compared to CON (P < 0.001). RYGB increased UCP1 expression, which was significantly higher than SHAM (P < 0.001) and not significantly different from CON (P > 0.05), suggesting that RYGB effectively restores browning capacity in subcutaneous white adipose tissue. These results demonstrate that RYGB robustly upregulates UCP1 expression in both BAT and iWAT, providing molecular evidence of enhanced thermogenic potential and reversal of obesity-induced adipose dysfunction. Fig. 9RYGB upregulates UCP1 mRNA expression in adipose tissues. Relative UCP1 mRNA levels in (A) interscapular brown adipose tissue (BAT) and (B) inguinal white adipose tissue (iWAT) at 8 weeks post-surgery, measured by qPCR and normalized to GAPDH. Values are expressed as fold change relative to the CON group (mean ± SD, n = 6 per group). ***P < 0.001; ns, not significant; one-way ANOVA with Tukey’s post hoc test RYGB upregulates UCP1 mRNA expression in adipose tissues. Relative UCP1 mRNA levels in (A) interscapular brown adipose tissue (BAT) and (B) inguinal white adipose tissue (iWAT) at 8 weeks post-surgery, measured by qPCR and normalized to GAPDH. Values are expressed as fold change relative to the CON group (mean ± SD, n = 6 per group). ***P < 0.001; ns, not significant; one-way ANOVA with Tukey’s post hoc test Discussion This study demonstrates that Roux-en-Y gastric bypass (RYGB) induces multifaceted metabolic improvements in obese T2DM mice, with three primary findings: (1) rapid and sustained weight loss beginning within one week, peaking at week 6 (P < 0.0001), (2) resolution of dyslipidemia, characterized by reductions in triglycerides (TG), total cholesterol (TC), and low-density lipoprotein (LDL), as well as an increase in high-density lipoprotein (HDL) compared to SHAM controls (P < 0.001), and (3) significant adipose remodeling, evidenced by a reduction in adipocyte diameter (P < 0.0001) and a micro-CT-confirmed decrease in subcutaneous adipose tissue (SWAT) and visceral white adipose tissue (VWAT). These outcomes are consistent with established findings from human RYGB studies, which consistently show substantial weight loss, resolution of dyslipidemia, and improved metabolic health in patients with obesity and T2DM [23–25]. Under specific physiological conditions, white adipose tissue (WAT) undergoes a transformation known as “browning,” acquiring features characteristic of brown adipocytes [26, 27]. This browning process is increasingly acknowledged for its potential metabolic benefits, particularly in the context of obesity and type 2 diabetes mellitus (T2DM). Recent studies have demonstrated that the browning of adipose tissue is closely associated with levels of inflammation [28]. Brown adipose tissue (BAT) is inherently more resistant to inflammation and exhibits significantly lower infiltration of inflammatory cells compared to WAT [7]. Specifically, in humans, visceral adipose tissue (VAT) in obesity is characterized by a high proportion of pro-inflammatory immune cells, such as Th1 and Tc1 cells [29], as well as CD20 + T cells [30], contributing to both local and systemic inflammation. In individuals with obesity and T2DM, the size and functionality of BAT are significantly reduced, accompanied by elevated levels of inflammation. This shift results in the loss of BAT’s characteristic brown color, which is replaced by a white, inflammatory state typical of WAT [31]. Emerging research suggests that promoting adipose tissue browning can effectively enhance glucose uptake and consumption, thereby improving insulin sensitivity. In a related experimental context, mice subjected to Roux-en-Y gastric bypass (RYGB) experienced substantial weight loss within six weeks, although the rate of weight loss began to decelerate after the sixth week. Compared to the SHAM group, the RYGB group exhibited significant improvements in lipid metabolism, as evidenced by decreased serum levels of triglycerides (TG), total cholesterol (TC), and low-density lipoprotein cholesterol (LDL). Furthermore, the high-density lipoprotein (HDL) level was markedly increased compared to the SHAM group. CT scan analyses at eight weeks after surgery revealed that the RYGB procedure resulted in notable reductions in both fat content and overall body weight. This sustained weight loss can primarily be attributed to a decrease in fat volume, corroborating findings from previous studies [32]. Importantly, a higher lean body mass in individuals with obesity is correlated with a lower inflammatory profile, improved insulin sensitivity, and enhanced metabolic health [2]. Thus, weight loss achieved through metabolic bariatric surgery, which successfully reduces adiposity while preserving lean body mass, represents a highly favorable outcome for patients battling obesity. The experimental results demonstrated a significant reduction in fat content localized in the trunk region, suggesting that metabolic bariatric surgery can alter fat distribution patterns. In humans, the distribution of body fat is a crucial determinant of metabolic and cardiovascular risk, with excessive abdominal and hip fat linked to a higher incidence of obesity-related diseases and increased mortality rates [33]. The accumulation of subcutaneous abdominal fat is particularly notorious for promoting insulin resistance [34], while visceral obesity significantly contributes to the onset of metabolic syndrome [35, 36]. Beyond the observed reduction in adipocyte size and total fat mass, our histomorphometric analyses revealed a profound structural remodeling of adipose depots following RYGB. Critically, adipocytes in RYGB-treated mice exhibited morphological changes suggestive of browning on H&E staining. While these features are consistent with a browning-like phenotype, it is important to note that definitive confirmation would require specific immunohistochemical markers such as UCP1 or perilipins. Nonetheless, this morphological remodeling is now directly supported by molecular evidence: RYGB significantly upregulated UCP1 mRNA expression in both brown adipose tissue (BAT) and inguinal white adipose tissue (iWAT) (Fig. 9), confirming enhanced thermogenic capacity. This UCP1 upregulation was accompanied by a marked reduction in adipose tissue inflammation cytokines (IL-6, TNFα), collectively indicates a shift toward a more oxidative and thermogenic profile in adipose depots. Notably, brown adipose tissue (BAT) in RYGB mice exhibited reversal of obesity-associated whitening, as evidenced by smaller lipid droplets and reduced adipocyte size. This UCP1 upregulation was accompanied by a marked reduction in adipose tissue inflammation, as evidenced by suppressed expression of pro-inflammatory cytokines. This morphological improvement in BAT, together with the upregulation of UCP1 expression and reduced inflammation in adipose tissue, suggests a broader systemic shift toward a more oxidative metabolic state following RYGB. The anti-inflammatory effect of RYGB was further supported by molecular analysis of liver tissue. Hepatic mRNA expression of pro-inflammatory cytokines IL-6 and TNFα, as well as the innate immune receptor TLR4, was significantly elevated in SHAM mice and normalized after RYGB (Fig. 8). This attenuation of hepatic inflammation aligns with the histological improvement in steatosis (Fig. 5A and B) and likely contributes to the enhanced systemic insulin sensitivity observed in RYGB-treated mice. These findings provide direct evidence that RYGB mitigates obesity-associated inflammation and induces adipose tissue browning, extending beyond simple morphological improvement in adipocyte size. Therefore, the observed reduction in adipose tissue content in mice following RYGB may not only reflect a protective physiological response to the surgery but could also be instrumental in improving insulin resistance post-intervention. This underscores the potential of RYGB as a transformative approach for individuals with severe obesity, offering a multifaceted strategy for enhancing metabolic health and reducing disease risk. Several limitations of this study should be acknowledged. First, although we have provided molecular evidence of UCP1 mRNA upregulation, we did not perform immunohistochemical staining for UCP1 protein localization within adipose tissue, which would provide spatial confirmation of the browning phenotype. Second, direct measurements of thermogenic function—such as energy expenditure, oxygen consumption, or core body temperature—were not performed. Such functional assays would be necessary to confirm that the observed UCP1 upregulation truly translates into increased whole-body thermogenesis, rather than simply reflecting reduced lipid storage secondary to weight loss or reduced caloric intake. Future studies incorporating pair-fed control groups are needed to disentangle the contribution of caloric restriction from the intrinsic effects of RYGB. Third, although we measured fasting insulin and calculated HOMA-IR, we did not perform glucose tolerance tests or assess dynamic insulin secretion, which would provide a more comprehensive evaluation of β-cell function. Despite these limitations, our data provide important insights into the metabolic improvements induced by RYGB and highlight adipose tissue remodeling as a key therapeutic target. Future studies incorporating protein-level confirmation of browning markers and functional thermogenic assessments are needed to definitively establish the underlying mechanisms. Conclusions RYGB has shown substantial efficacy in addressing obesity and associated metabolic disorders in a murine model. The procedure not only led to significant weight reduction but also reversed the onset of diabetes mellitus and improved insulin sensitivity. Notably, RYGB ameliorated hepatic steatosis, suggesting a beneficial effect on liver health. Additionally, the surgery resulted in a reduction of total body fat content and positively affected body fat distribution, indicating that RYGB could promote a more favorable body composition. Importantly, the procedure induced morphological changes in adipose tissue consistent with a browning-like phenotype, which, together with UCP1 upregulation and reduced inflammation, may contribute to increased energy expenditure and improved metabolic health. However, direct functional studies are required to confirm enhanced thermogenesis. An improvement in the morphology of adipose tissue across multiple body regions was also observed, reflecting overall enhancements in adipose tissue function. These findings support the potential of RYGB not only as a weight loss intervention but also as a therapeutic strategy to mitigate obesity-related complications, including diabetes and hepatic steatosis. Future studies may further investigate the underlying mechanisms and long-term effects of RYGB on metabolic health. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1