High-fat diet (D12451, comprising 45% fat, 35% carbohydrates, and 20% protein) purchased from Shenyang Maohua Biotechnology Co., Ltd. The authenticated crude drugs of Rhei Radix et Rhizoma (202210653–3) and Coptidis Rhizoma (231002) were procured from the Affiliated Hospital of Gansu University of Chinese Medicine. Metformin hydrochloride tablets (ABZ4625) was obtained from Sino-American Shanghai Squibb Pharmaceuticals Ltd.

Rat INS ELISA Kit (YJ985241) Rat FFA ELISA KIT (YJ524122) were purchased from Shanghai Yuanju Biotechnology Center. p-AMPKα (14099) antibody SIRT1 (27523) antibody PGC-1α (27583) antibody were purchased from Signalwayantibody Co., Ltd. PPARγ (YT3836) antibody was obtained from ImmunoWay Biotechnology Company. PRDM16 (DF13303) antibody was purchased from affinity biosciences. UCP1 (GB112174) antibody was obtained from Wuhan Servicebio Technology Co., Ltd. β-tubulin (380628) Goat-anti-Mouse IgG-HRP (511103) were purchased from Chengdu Zen-Bioscience Co., Ltd.

The Dahuang-Huanglian herbal preparation was processed according to Chinese Pharmacopoeia specifications and prepared by the Pharmaceutical Preparation Laboratory of Gansu University of Chinese Medicine Affiliated Hospital. Based on our research group’s previous studies, a dosage of DHHL was set at 10 g for each component, in a 1:1 ratio (14). Using a multifunctional extraction tank, the first extraction involved soaking crude drugs in 10-fold volume of distilled water for 30 minutes, followed by conventional decoction twice (40 minutes per cycle). The filtered solutions were combined and concentrated to achieve 1g crude drug per mL. After high-temperature sterilization and cooling, the extract was stored at 4°C. For experimental use, the stock solution was diluted proportionally prior to administration.

Eighty SPF-grade male SD rats, aged 7 weeks old, were obtained by the Gansu University of Chinese Medicine (Laboratory animal production license No. SYXK(Gan)2020-0001, Ethical Review Approval No. SY2024-232). The rats were fed with standard rodent food and water and under stable light-dark (12h-12h) cycles and constant room temperature. The Animal Welfare Law of China is strictly observed during animal experiments, and the supervision and inspection of the Laboratory Animal Ethics Committee is accepted at all times.

After one week of acclimatization feeding, all rats were randomly divided into two groups using a random number table method: the normal diet (ND) group (n = 16) fed standard chow and the high-fat diet (HFD) group (n = 64) fed fat-enriched chow. Obesity modeling was assessed following 8 weeks of HFD feeding, with rats exceeding 20% of the ND group’s average body weight defined as obese model animals (30), while non-qualified rats were excluded. To validate successful modeling, 6 rats each from ND and HFD groups were randomly selected for additional evaluation. Obese rats confirmed as successful models continued HFD feeding and were randomly allocated into five groups (n = 8 each): Model group, metformin (Met) group, low-dose Dahuang-Huanglian (LDHHL) group, medium-dose Dahuang-Huanglian (MDHHL) group, and high-dose Dahuang-Huanglian (HDHHL) group, while ND rats served as the Control group (n = 8). The HDHHL, MDHHL, and LDHHL groups received daily gavage of 3.6 g/kg, 1.8 g/kg, and 0.9 g/kg DHHL extract respectively. The Met group was administered 0.18 g/kg/day metformin hydrochloride suspension according to pharmaceutical guidelines. Both Control and Model groups received equivalent volumes of saline daily. All treatments were administered once daily for 4 consecutive weeks.

100 μL liquid sample was added to a 1.5 mL centrifuge tube with 400 μL solution (acetonitrile: methanol = 1:1(v:v)) containing four internal standards (0.02 mg/mL L-2-chlorophenylalanine, etc.) to extract metabolites. The samples were mixed by vortex for 30 s and low-temperature sonicated for 30 min (5°C, 40 KHz). The samples were placed at -20°C for 30 min to precipitate the proteins. Then the samples were centrifuged for 15 min (4°C, 13000 g). The supernatant was removed and blown dry under nitrogen. The sample was then re-solubilized with 100 μL solution (acetonitrile: water = 1:1) and extracted by low-temperature ultrasonication for 5 min (5°C, 40 KHz), followed by centrifugation at 13000 g and 4°C for 10 min. The supernatant was transferred to sample vials for LC-MS/MS analysis.

The LC-MS/MS analysis of sample was conducted on a UHPLC-Orbitrap Exploris 240 system equipped with an ACQUITY HSS T3 column (100 mm × 2.1 mm i.d., 1.8 μm; Waters, USA) at Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). The mobile phases consisted of 0.1% formic acid in water:acetonitrile (2:98, v/v) (solvent A) and 0.1% formic acid in acetonitrile (solvent B). The flow rate was 0.40 mL/min and the column temperature was 40°C. The injection volume was 5 μL (31).

MS conditions: The UPLC system was coupled to a UHPLC-Orbitrap Exploris 240 system Mass Spectrometer equipped with an electrospray ionization (ESI) source operating in positive mode and negative mode. The optimal conditions were set as followed: source temperature at 400°C; sheath gas flow rate at 40 arb; Aux gas flow rate at 10 arb; ion-spray voltage floating (ISVF) at -2800V in negative mode and 3500V in positive mode, respectively; Normalized collision energy, 20-40-60V rolling for MS/MS. Data acquisition was performed with the Data Dependent Acquisition (DDA) mode. The detection was carried out over a mass range of 70–1050 m/z.

Progenesis QI (Waters Corporation, Milford, USA) software was used for peak extraction, retention time correction and peak alignment of raw data, and metabolites were identified by matching HMDB, Metlin and self-built database; After screening valid variables by 80% rule, fill missing values with minimum value, normalize sum and eliminate variables with RSD>30% of QC samples. PCA/OPLS-DA model analysis (7 cross validation) was performed by using R language ropls package after log10 transformation to screen differential metabolites with VIP>1 and p<0.05. Thanks for the pathway annotation based on KEGG database and Fisher exact test enrichment analysis via Python’s scipy.stats package, all performed on the Majorbio.com platform.

Body weight measurements were determined at weekly intervals across experimental cohorts. Prior to tissue collection under anesthesia, final body weight and nose-to-anus length were quantified for each animal. Lee’s adiposity index was calculated using the formula: [cube root of body weight (g) × 1000]/body length (cm).

Fasting blood glucose (FBG) monitoring was performed weekly across experimental groups. Following 12-hour fasting with free water access, blood samples were obtained via caudal vein puncture using aseptic techniques. Glucose quantification was conducted immediately with Roche Accu-Chek monitoring systems and manufacturer-specified test strips. Post-sampling wound care involved prompt antisepsis using povidone-iodine solution, followed by hemostatic compression. All procedures maintained circadian consistency, with measurements conducted between 8:00-10:00 AM to minimize diurnal variation interference.

Serum free fatty acids (FFA) and fasting serum insulin (FINS) levels were measured using ELISA kits. The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated as follows: HOMA-IR = [FINS (μU/mL) × FBG (mmol/L)]/22.5.

Rats in each group were fasted for 12 h prior to the intraperitoneal glucose tolerance test (IPGTT). Following body weight measurement, 50% glucose solution was intraperitoneally administered at a dose of 2 g/kg. Blood samples were obtained from the tail vein at 0, 15, 30, 60, and 120 min post-injection for glucose determination using a glucometer. IPGTT curves were plotted, and the area under the curve (AUC) was calculated through trapezoidal integration.

For the insulin tolerance test (ITT), animals were fasted for 6 h before receiving intraperitoneal insulin injection (0.75 U/kg) following body weight measurement. Blood glucose levels were assessed at identical time intervals as IPGTT to construct ITT curves and compute AUC values.

Serum biochemical parameters including triglycerides (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-c), and high-density lipoprotein cholesterol (HDL-c) were quantified using a fully automatic biochemical analyzer (Roche cobas c 311).

Adipose tissue specimens were immersion-fixed in 4% paraformaldehyde, subsequently processed through dehydration and paraffin embedding, and sectioned at 5 μm thickness. Tissue sections were subjected to hematoxylin and eosin (H&E) staining for comprehensive histological evaluation.

Total RNA was extracted from rat adipose tissue using TRIzol reagent (Invitrogen), with optimized tissue mass (20–50 mg) adjusted for low RNA abundance. Following homogenization (60 Hz, 60 s) and phase separation with chloroform, RNA was precipitated with isopropanol and washed with 75% ethanol. RNA integrity was confirmed by Agilent 2100 Bioanalyzer (RNA 6000 Nano kit) and concentration quantified via NanoDrop 2000. Libraries were constructed from ≥1 μg total RNA using the NEBNext Ultra II Directional RNA Library Prep Kit for Illumina, involving poly(A) mRNA enrichment, fragmentation, cDNA synthesis, end repair/adapter ligation, and size selection (400–500 bp). Library quality was assessed by Agilent 2100 High Sensitivity DNA Kit and quantified by QPCR (StepOnePlus). Multiplexed libraries were sequenced on the Illumina platform (PE150). Raw data were filtered by removing adapter sequences and low-quality reads (Q20 cutoff). Clean reads were aligned to the rat reference genome (rn6) using HISAT2 v2.0.5, followed by gene expression quantification (HTSeq v0.9.1, FPKM normalization) and differential expression analysis (DESeq v1.20.0; |log₂FC| >1, p < 0.05). Functional enrichment of differentially expressed genes was analyzed via GO (topGO) and KEGG (clusterProfiler v3.4.4).

Paraffin-embedded tissue sections underwent deparaffinization with xylene and graded ethanol series. Antigen retrieval was performed using EDTA-based antigen retrieval buffer followed by blocking with bovine serum albumin. Sequential incubations with primary and secondary antibodies were conducted, accompanied by nuclear counterstaining with DAPI. Autofluorescence was eliminated using commercial quenching reagent before fluorescence-preserving mounting. Specimens were examined under fluorescence microscopy, and fluorescence intensity was quantified through ImageJ analysis software.

The thoroughly homogenized tissues were lysed in RIPA lysis buffer supplemented with protease and phosphatase inhibitors on ice for 30 min. After centrifugation at 12000 × g for 15 min, the supernatant was collected for protein quantification using BCA Protein Assay Kit. Equal amounts of protein samples mixed with loading buffer were denatured by boiling at 100°C for 10 min. Proteins were separated through SDS-PAGE and electrophoretically transferred to nitrocellulose membranes. Membranes were blocked with 5% skim milk for 1 h, followed by sequential incubation with specific primary antibodies (overnight at 4°C) and HRP-conjugated secondary antibodies (1 h at room temperature). β-tubulin served as an internal control. Protein bands were visualized using enhanced chemiluminescence substrate and quantitatively analyzed with ImageJ software.

Tissue fragments were promptly immersed in electron microscopy fixative and subjected to light-protected fixation at room temperature for 2 hours, followed by storage at 4°C. The fixed specimens were subsequently embedded, ultrathin sections were prepared, and ultrastructural observations were performed using transmission electron microscopy (TEM) with image acquisition.

Deparaffinized sections were subjected to heat-induced antigen retrieval using 10 mM citrate buffer. After 15-minute incubation with 3% hydrogen peroxide at room temperature, non-specific binding was blocked with bovine serum albumin. Sequential incubations with specific primary antibodies (4°C overnight) and HRP-conjugated secondary antibodies (room temperature for 1 hour) were performed, followed by chromogenic development using 3,3’-diaminobenzidine (DAB) substrate. Nuclei were counterstained with hematoxylin, differentiated with acidic ethanol solution, and blued in ammonia water. Sections were dehydrated through an ethanol gradient, cleared in xylene, and permanently mounted with neutral resin. Stained specimens underwent microscopic examination, with positive signals quantitatively analyzed using ImageJ software.

All statistical analyses were performed using SPSS 26.0 software. Continuous variables with normal distribution were expressed as mean ± standard deviation (SD), while non-normally distributed data were presented as median (interquartile range) [M (P25, P75)]. Between-group comparisons for normally distributed variables utilized independent samples t-test, whereas the wilcoxon rank-sum test was employed for non-normally distributed data. Multiple group comparisons were conducted using one-way ANOVA, with post-hoc analyses applying Least Significant Difference (LSD) method when homogeneity of variance was confirmed, and Tamhane’s T2 method when heterogeneity existed. A threshold of P < 0.05 indicated statistical significance.

The chemical constituents of DHHL medicinal materials were identified and analyzed via UPLC-MS/MS, yielding the metabolite names, retention time (RT) in chromatography, mass-to-charge ratio (M/Z), molecular formula, ion detection mode (Mode), quality control (QC), and their botanical origins. Through non-targeted UPLC-MS/MS profiling coupled with database matching against MJBIOTCM (Majorbio’s specialized TCM metabolome database), we identified 777 distinct compounds as listed in Supplementary ( Supplementary Table S1 ). From this comprehensive chemical inventory, we prioritized 10 herb-specific markers (e.g., emodin/berberine) based on two criteria: established bioactivity in Pharmacopoeia monographs, and documented relevance to metabolic regulation. Comprehensive analytical of these representative constituents are presented in ( Table 1 ), providing a focused phytochemical signature of DHHL’s therapeutic potential. Subsequently, total ion chromatograms (TIC) were generated for all components under both positive and negative ion modes ( Figures 1A, B ). Based on subclass categorization from the identification results, pie charts illustrating the proportional distribution of compositional abundance (content percentage) and constituent quantity (subclass count) within different phytochemical categories were constructed ( Figures 1C, D ).

An imbalance between energy intake and energy expenditure represents a fundamental pathological mechanism underlying obesity. Numerous studies have confirmed that HFD feeding significantly increases body weight, FBG, and other metabolic parameters, leading to systemic metabolic disorders (32). Our research also shows that after 8 weeks of HFD feeding, various physiological parameters were measured in both the ND group and the HFD group. Compared with rats in the ND group, those in the HFD group exhibited significantly increased body weight and FBG levels ( Figures 2A, B ). Further analysis using Lee’s index showed that the HFD group had a markedly higher Lee’s index than the ND group ( Figure 2C ). The results of the ITT demonstrated that the HFD group had severely impaired glucose handling capacity ( Figures 2D–G ). Moreover, rats fed with HFD demonstrated considerable elevation in serum concentrations of FFA and FINS when compared with their ND counterparts ( Figures 2H, I ). Concurrently, the HOMA-IR was substantially heightened in the HFD cohort ( Figure 2J ). These findings indicate that HFD feeding induces obesity and elevated FBG in rats.

To further assess the impact of HFD on lipid metabolism, serum biochemical analysis was performed. The results showed that the TG levels were significantly elevated in HFD rats, whereas TC, HDL-c, with LDL-c profiles exhibiting statistical equivalence between the rodent populations subjected to either HFD or ND interventions ( Figures 3A–D ). As shown by the gross morphology of adipose tissues, the volumes of iWAT, epididymal white adipose tissue (eWAT), and BAT were markedly increased in HFD rats. H&E staining revealed that the adipocyte size in HFD rats was significantly larger than that in the ND group ( Figure 3E ). The extended mathematical analysis established that participants exposed to HFD manifested substantially higher values for both adipose tissue weight and adipocyte area relative to the ND group ( Figures 3F–J ).

Except for the ND group, all other groups continued to receive HFD feeding. Metformin was used as a positive control drug, and different doses of DHHL decoction were administered to the DHHL treatment groups. After 4 weeks of treatment, relevant indicators were measured. TCM offers inherent strengths in managing chronic metabolic disorders such as obesity. Existing studies have documented the significant efficacy of Rhei Radix et Rhizoma and Coptidis Rhizoma in ameliorating obesity, dyslipidemia, and glucose dysregulation (6, 13). Our prior research further confirms that the DHHL herb pair effectively improves metabolic parameters in genetically obese diabetic (db/db) mice (14). Rats in the model group showed continuous weight gain, which was significantly higher than that in the control group, whereas treatment with metformin and HDHHL significantly suppressed HFD-induced weight gain ( Figure 4A ). The assessment of Lee’s index revealed a substantial increase in rats assigned to the model group; however, intervention with either metformin or DHHL effectively diminished this parameter, indicative of obesity improvement ( Figure 4C ). Rats in the model cohort demonstrated notably increased FBG measurements; however, intervention with DHHL therapy resulted in a considerable reduction of these elevated glucose levels ( Figure 4B ). The results of IPGTT and ITT assessments indicated substantial compromise in glucose utilization capacity and insulin sensitivity parameters among subjects in the model group; these impairments were notably ameliorated following DHHL administration, indicating improvements in glucose and insulin tolerance ( Figures 4D–G ). Analysis of serum samples demonstrated that model group subjects manifested substantially higher levels of FFA, FINS, and HOMA-IR compared to the control cohort; subsequent DHHL therapy effectively reduced these elevated biochemical markers ( Figures 4H–J ).

As shown in the figure below, the TG content in the model group was significantly higher than that in the control group, whereas DHHL administration significantly reduced serum TG levels in rats ( Figure 5A ). However, only slight differences were observed among groups regarding serum levels of TC, LDL-c, and HDL-c ( Figures 5B–D ). Gross morphology of adipose tissues revealed that iWAT, eWAT, and the model group displayed substantially enlarged and heavier BAT relative to the control group, with different dosages of DHHL treatment producing variable degrees of diminution in both the size and weight of adipose tissue ( Figures 5E–H ). H&E staining was used to analyze adipose tissues from each group. Data indicated that adipocytes were substantially larger in the model group, and treatment with various doses of DHHL resulted in graduated diminution of adipocyte dimensions ( Figures 5I–K ). In summary, DHHL effectively improved body fat composition in obese rats, and its therapeutic effect exhibited a dose-dependent trend.

The aforementioned experiments have confirmed that DHHL can significantly improve body weight and glucose metabolism disorders in obese rats, exhibiting a dose-dependent therapeutic effect. To further elucidate the mechanism by which DHHL effectively alleviates obesity in rats, we performed transcriptomic sequencing analysis on iWAT from rats in the Model group and the HDHHL group. Visualization via volcano plot demonstrated a considerable quantity of genes with differential expression patterns when comparing the Model and HDHHL cohorts ( Figure 6A ). In alignment with these findings, hierarchical clustering visualization via heatmap revealed pronounced distinctions in iWAT transcriptomic signatures between the compared groups ( Figure 6B ). To investigate the potential biological pathways involved, we constructed KEGG enrichment bar charts covering signaling pathways, the endocrine system, and lipid metabolism ( Figure 6C ), as well as KEGG enrichment bubble plots ( Figure 6D ). Both figures showed significant enrichment in pathways related to WAT browning, notably the AMPK and PPAR signaling cascades. Gene set enrichment analysis (GSEA) revealed that the gene sets of AMPK signaling pathway ( Figure 6E ), PPARγ-associated pathways ( Figure 6G ), adaptive thermogenesis ( Figure 6I ), and mitochondrial fission ( Figure 6K ) exhibited significant upregulation in the HDHHL group compared with the Model group. Furthermore, differential heatmap analysis visually confirmed that key regulators of adaptive thermogenesis and white adipose tissue browning, including PPARγ and PRDM16, exhibited significantly reduced expression in the Model group with concurrent marked elevation in the HDHHL intervention group ( Figure 6J ).

The above transcriptomic analysis results indicate that DHHL may exert a regulatory effect on genes associated with WAT browning. Hallmark features of WAT browning include increased mitochondrial abundance within adipocytes and elevated expression of thermogenesis-related proteins such as PPARγ, PRDM16, and UCP1. Furthermore, several studies have confirmed that rhubarb and coptis can promote WAT browning. Berberine, the primary active constituent of coptis, enhances WAT browning in both 3T3-L1 murine preadipocyte cells and obese mice (13), thereby ameliorating obesity. Dietary supplementation with rhubarb similarly induces WAT browning, augmenting thermogenesis and energy expenditure to counteract metabolic syndrome (33). Therefore, this study further examined the regulatory role of DHHL in promoting WAT browning in rats. Transmission electron microscopy revealed that, in contrast to the Control group, a substantial diminution in mitochondrial population was observed within the Model group; following DHHL treatment, mitochondrial abundance was partially restored ( Figure 7A ). Western blotting (WB) and immunohistochemistry (IHC) analyses showed that thermogenesis-related proteins were highly expressed in the adipose tissue of rats in the Control group. The HFD-induced obese rodent cohort, however, displayed substantially reduced levels of these proteins when iWAT tissue was examined. Notably, DHHL administration markedly reversed the HFD-induced suppression of thermogenesis-related protein expression in iWAT ( Figures 7B–I ). These findings collectively suggest that DHHL promotes WAT browning in obese rats.

Earlier investigations have revealed the essential function of the AMPK/SIRT1/PGC-1α pathway in modulating the browning process of white adipose tissue (28, 29). Consistently, transcriptomic analyses in the present study also indicated that the therapeutic effect of DHHL on obese rats is closely associated with activation of the AMPK pathway. We sought additional confirmation regarding DHHL’s obesity-alleviating effects through WAT browning enhancement via the AMPK/SIRT1/PGC-1α pathway by evaluating crucial protein expressions within iWAT across different groups employing fluorescence immunohistochemistry (FIHC) ( Figures 8A–C ) and WB ( Figure 8D ), followed by quantitative statistical analysis presented as bar graphs ( Figures 8E–J ). Examination of iWAT revealed that subjects in the Model group displayed considerably reduced levels of phosphorylated AMPKα (p-AMPKα), SIRT1, and PGC-1α expression compared to their Control counterparts. Notably, DHHL treatment markedly restored the expression of these proteins. These findings suggest that DHHL may promote iWAT browning by activating the AMPK/SIRT1/PGC-1α signaling pathway, which could be a key mechanism underlying its anti-obesity effects.