Chemicals and reagents

The WXZZ decoction (Beijing Kangrentang Pharmaceutical Co., Ltd., China) was dissolved in heated (60 °C) deionized water to obtain a 3.0 g/mL stock solution and stored at 4 °C. Metformin was purchased from Zhongmei Shanghai Squibb Pharmaceutical Co., Ltd. Dehydroepiandrosterone sulfate (DHEA) was purchased from Beijing Abifan Biotechnology Co., Ltd.

Animal

SPF-grade female C57BL/6 mice, aged 3 weeks and weighing 10–12 g, were obtained from Spivey Biotechnology Co., Ltd (Beijing, China). Upon arrival, the animals were housed in the Animal Experiment Center of Beijing University of Chinese Medicine. A total of 40 mice were randomly assigned to 8 cages, with 5 mice per cage. The animals were maintained under controlled environmental conditions, with a room temperature of 20–25 °C, relative humidity of 40–60%, and a 12-hour light/dark cycle. Food and water were provided ad libitum.

All experimental procedures involving animals were reviewed and approved by the Laboratory Animal Ethics Sub-committee of the Academic Committee of Beijing University of Chinese Medicine (Approval No. BUCM-2023030104-1105).

DHEAS and high-fat diet-induced IR and anovulatory mice

Animals were randomly assigned to control or experimental group. IR and anovulatory mice model was induced by DHEA and high-fat diet (HFD) [24].The experimental group (n = 30) was subjected to a daily subcutaneous injection of 0.2 mL DHEA (6 mg/kg) [24], administered into the dorsal neck region. The DHEA was prepared in a solution with soybean oil. In addition, these animals were fed a HFD (4.73 kcal/g; Beijing Huafukang Biotechnology Co., Ltd.), with an energy distribution of 20% protein, 35% carbohydrate, and 45% fat. The control group (n = 10) received an equivalent subcutaneous injection of 0.2 mL soybean oil, administered under identical conditions, and were fed a standard diet (3.85 kcal/g; Beijing Huafukang Biotechnology Co., Ltd.) with an energy distribution of 20% protein, 70% carbohydrate, and 10% fat.

The injections and diet regimen were administered continuously for 3 weeks. Throughout the study, all animals were weighed daily at approximately 8:00 a.m. by the same researcher who recorded using an electronic balance, with measurements expressed in grams. Vaginal smears were collected daily at approximately 9:00 a.m. for 12 consecutive days, starting from the 10th day of the modeling phase. The collected cells were preserved using 4% paraformaldehyde fixation to ensure optimal preservation of cellular morphology for subsequent analysis of estrous cycle.

At the end of the treatment period, animals were fasted for 12 h, with only water provided, before blood collection. Fasting blood glucose (FBG) levels were measured using blood obtained from the tail vein. Following this, blood was collected from the retro-orbital sinus under anesthesia to separate serum for the determination of fasting insulin (FINS) levels and gonadal hormones. Then 7 mice (3 in the control group and 4 in the model group) were killed by cervical decertification method to observe the structural changes of ovarian tissue, so as to ensure the successful establishment of PCOS mouse model.

Insulin resistance in this study was evaluated using HOMA-IR [25]. The HOMA-IR index was calculated using the formula:

The mice model was defined as a significant increase in body weight, serum T, and LH levels compared to the control group, along with vaginal smears showing abnormal estrous cycle and increasing ovarian follicles. Additionally, a HOMA-IR index should exceed 1.96 standard deviations above the mean of the control group.

Experimental grouping and treatment

Mice in experimental group were randomly assigned to three groups, thus a total of four groups entered this phase of study. The control group (Con, n = 7) and the model group (DHEA + HFD, n = 9) received 0.2 mL/10 g of distilled water by gavage. The metformin group (DHEA + HFD + Met, n = 9) was administered metformin hydrochloride (200 mg/kg/day) by gavage [26], while the WXZZ group (DHEA + HFD + WXZZ, n = 8) received a WXZZ solution (270 mg/kg/day) by gavage, calculated based on the “Equivalent Doses for Animals and Humans Based on Body Surface Area” algorithm [27], with a conversion factor of 12.3:1 for mice. The dosage for WXZZ and metformin calculate based on the body weight of mice on that day, then dissolving in 0.2 ml distilled water by gavage.

All experimental groups were maintained on HFD, while the control group received a standard diet. The treatments were administered once daily at approximately 10:00 a.m. for a duration of 2 weeks.

Assessment of estrous cycle

The estrous cycle in mice, typically lasting 4–5 days, was monitored in this study by collecting vaginal exfoliated cells daily at approximately 9:00 a.m. over 12 consecutive days. During the modeling phase, collections began on the 10th day, while in the treatment phase, collections began on the third day. Vaginal smears were obtained by flushing the vaginal canal with sterile saline, spreading the fluid onto glass slides, and fixing the samples in 4% paraformaldehyde. The smears were then stained with HE to visualize the cells [28]. Microscopic analysis of the HE-stained smears allowed for the identification of the estrous cycle stage based on the predominant cell types (Table 1).

Table 1 Characteristics of vaginal exfoliated cells in different estrous cyclesMeasurement of FBG and other serum parameters

FPG levels were measured using an Accu-Chek Performa glucometer (Roche, Korea). To obtain the blood sample, a small incision was made at the tail tip of the mice using sterile medical scissors. A drop of blood was then applied to a test strip for glucose measurement, with results expressed in mmol/L.

For serum analysis, samples were collected from the mice’s eyeballs. The collected blood was allowed to clot at 4 °C for 2 h, followed by centrifugation at 3000 rpm for 10 min to separate the serum, then the sample was stored at -80 °C until further analysis. Serum levels of T, estradiol (E2), progesterone (P), LH, and FSH were measured using enzyme-linked immunosorbent assay (ELISA) kits (Wuhan Eliot Biotechnology Co., China). Serum NEFA levels was analyzed by a fully automated biochemical analyzer (Roche, Germany). Serum irisin (MALLBIO, MBE12290, JL21442-48 T; assay range: 2.5 ng/mL − 80 ng/mL; sensitivity: 0.1 ng/mL) and FINS (MALLBIO, MBE10122, JL10692-48 T; assay range: 1.25 mU/L − 40 mU/L; sensitivity: 0.1 mIU/L) levels were determined using double antibody one-step sandwich ELISA kits, following the manufacturer’s protocols. Absorbance was measured at 450 nm using a microplate reader, and concentrations were calculated based on standard curves.

Histomorphological observations

Following the completion of blood sampling, the mice were anesthetized via intraperitoneal injection of 2% pentobarbital sodium solution (0.2 mL/100 g). Bilateral gastrocnemius muscle and abdominal adipose tissues were rapidly excised. Half of tissue specimen was used for hematoxylin-eosin (HE) staining. The remaining half was stored at -80 °C for subsequent protein and mRNA expression analyses. The tissues were processed uniformly across all groups.

Gastrocnemius muscle tissue samples were fixed in 4% paraformaldehyde for a minimum of 24 h. Following fixation, the tissues underwent gradient dehydration using a series of ethanol concentrations, followed by clearing in xylene. The cleared tissues were then embedded in paraffin wax for 4 h. Tissue sections were prepared with a thickness of 4 μm and stained sequentially with HE. The stained sections were sealed with neutral balsam and analyzed under a light microscope (Nikon, 400x magnification) for histological analysis.

Adipose tissue was stained with Oil Red O. Lipid droplet content was analyzed using an adipogenesis assay kit (cell-based: Abcam, ab133102, USA) according to the manufacturer’s instructions. Briefly, cells were washed twice with Lipid Droplet Analysis Wash Solution, then Lipid Droplet Analysis Oil Red O solution was added to the cells, and the cells were incubated at room temperature for 20 min before staining was observed microscopically. Stained lipid droplets were detected by reading the absorbance at 490 nm using enzyme standards.

Western blot analysis

Western blotting was employed to detect the protein levels of AMPK, PGC-1α, FNDC5, and irisin in the gastrocnemius muscle tissues and CaMKK, AMPK, PGC1-α, and UCP1 in subcutaneous fatty tissue [29]. Proteins were extracted from the target tissues by homogenizing the samples in RIPA Tissue Cell Rapid Lysis Solution (R0020, Beijing Solepol) containing protease inhibitors. The homogenates were lysed thoroughly, and the supernatant was collected following centrifugation. The protein concentration of the extracted samples was measured using the BCA Protein Assay Kit (PICPI23223, Thermo Fisher Scientific). The extracted proteins were separated by SDS-PAGE (S1010, Beijing Solepol) and subsequently transferred onto nitrocellulose (NC) membranes (HATF00010, Millipore) through electroblotting. The membranes were blocked overnight at 4 °C in a 5% skim milk powder (D8340, Solepol, Beijing) blocking solution to prevent non-specific binding.

Immunoblotting was performed by incubating the membranes with primary antibodies against AMPK, PGC1-α, FNDC5, irisin, CaMKK, UCP1 (Table 2) for 24 h at 4 °C. After incubation, the membranes were washed and then incubated with the corresponding secondary antibodies. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control. Enhanced chemiluminescence signals were detected using a chemiluminescence imaging system (chemq3400mini, China), and the gray values of the protein bands were quantitatively analyzed using ImageJ software. The process was repeated at least three times to ensure reproducibility and reliability of the results.

Table 2 Details of the antibodies used in Western blottingRT-PCR assay

Reverse transcription-polymerase chain reaction (RT-PCR) was utilized to quantify mRNA expression of AMPK, PGC1-α, FNDC5, and irisin in gastrocnemius muscle tissues and CaMKK, AMPK, PGC1-α, annd UCP1 in subcutaneous fatty tissue [30]. Total RNA was extracted from tissue samples using TRIzol reagent (1596-026, Invitrogen, USA). To remove genomic DNA, RNA was treated with DNase I according to the instructions provided with the reverse transcription kit (Fermentas, #K1622, USA). Complementary DNA (cDNA) was synthesized from the purified RNA. cDNA was amplified using the SYBR Green PCR kit (#K0223, Thermo Fisher Scientific) and analyzed with a Real-Time PCR detector (ABI-7300, ABI Corp.). The PCR conditions were set as follows: denaturation at 95 °C for 15 s, annealing at 55 °C for 45 s, and extension at 72 °C for 30 s, for a total of 40 cycles. The relative expression levels were calculated using the 2−ΔΔCt method, with GAPDH serving as the internal control for normalization. The specific primers used are detailed in Table 3.

Table 3 Primer sequence used for RT-PCRStatistical analysis

Statistical analyses were conducted using SPSS version 26.0 and GraphPad Prism version 9.0.0. All data are presented as mean ± standard deviation (SD) and were validated for normal distribution using the shapiro-wilk test before subsequent statistical analyses. Intergroup differences were compared using the student’s t-test for two groups and one-way ANOVA for three or more groups. For non-normally distributed data, the kruskal-wallis test was employed. Correlation analyses were performed using Pearson’s correlation coefficient or normally distributed variables and Spearman’s correlation coefficient for non-normally distributed variables. P < 0.05 was considered as statistically significant.