Dried stems of Gnetum parvifolium from Yulin, Guangxi, are sun-dried, pulverized to pass through a No. 3 sieve, and the powder is stored for later use.

Rutin reference standard (batch No. RP220522, >99% purity; Chengdu Medson Technology Co., Ltd.); DPPH solution (batch No. 20240514; Feijing Biotechnology Co., Ltd.); ascorbic acid (analytical grade, batch No. 2024002; Tianjin Beichen Fangzheng Reagent Factory); aluminum nitrate (analytical grade, batch No. 4104-20-9; Tianjin Kemiou Chemical Reagent Co., Ltd.); sodium nitrite (analytical grade, batch No. 1401061; Xilong Chemical Co., Ltd.); and sodium hydroxide (analytical grade, batch No. 2112131; Xilong Chemical Co., Ltd.). All other reagents were of analytical grade, and deionized water was used throughout the experiments.

TU-1950 double-beam UV-Vis spectrophotometer (Beijing Purkinje General Instrument Co., Ltd.); KQ-800KDV ultrasonic cleaner (Kunshan Jielimei Ultrasonic Instrument Co., Ltd.); FA1104N electronic analytical balance (Shanghai Sunny Hengping Scientific Instrument Co., Ltd.).

Accurately weigh 0.1 g of rutin reference standard into a 50 mL clean and dry volumetric flask, add 70% ethanol solution to dissolve and dilute to the mark to obtain a 0.02% rutin reference standard solution.

Accurately weigh 0.2 g of Gnetum parvifolium fine powder, add 20 mL of 70% ethanol solution to a 50 mL conical flask and mix well, then sonicate at room temperature for 20 min, perform suction filtration with a Büchner funnel, and the filtrate is the sample solution.

Aliquots of the standard solution (1.0 mL; Section “3.1”) and the sample solution (0.5 mL; Section “3.2”) were accurately pipetted into separate 25 mL volumetric flasks. To each flask, 1 mL of 5% sodium nitrite solution was added, and the contents were mixed and allowed to stand for 5 min. Subsequently, 1 mL of 10% aluminum nitrate solution was added, followed by mixing and an additional 5-minute standing period. Next, 5 mL of 4% sodium hydroxide solution was added. The solution was then brought to volume with 70% ethanol, mixed thoroughly, and allowed to stand for 15 min. A reagent blank was prepared in the same manner, excluding the standard or sample solution. The absorption spectrum for each solution was then recorded from 200 to 600 nm against the reagent blank.

Aliquots of a 0.02% rutin standard solution (0.40, 0.60, 0.80, 1.00, and 1.20 mL) were individually transferred into five clean, dry 25 mL volumetric flasks. Color was developed using the aluminum nitrate colorimetric method, and the absorbance of each solution was measured with a spectrophotometer at the characteristic absorption wavelength of 345 nm. A standard curve was then constructed by plotting absorbance against rutin mass concentration [9] .

In accordance with the guidelines of the Chinese Pharmacopoeia, six 0.2 g samples of Gnetum parvifolium fine powder were precisely weighed using an analytical balance (0.1 mg sensitivity). The samples were processed following the procedure described in Section 3.2. After color development via the aluminum nitrate colorimetric method, the absorbance of each solution was measured with a UV-visible spectrophotometer at a characteristic absorption wavelength of 345 nm. The total flavonoid content of each sample was then calculated, and the relative standard deviation (RSD) was determined. The resulting RSD was 1.67%, which falls within the industry standard of ≤2.0% for good repeatability. This demonstrates that the experimental method possesses high repeatability and reliability [10] . These results are presented in Table 1 .

A 1.0 mL aliquot of the standard solution described in Section 3.1 was precisely pipetted into a 25 mL clean, dry volumetric flask. A color development procedure was then performed according to the aluminum nitrate colorimetric method. The absorbance of the resulting solution was measured at 345 nm. Six replicate measurements were performed, and the data were recorded. Statistical analysis of the results yielded a relative standard deviation (RSD) of 1.32%, indicating that the instrument possesses good precision and meets the requirements of the experiment. These results are presented in Table 2 .

An aliquot of 0.5 mL of the sample solution was accurately pipetted into a 25 mL volumetric flask for color development using the aluminum nitrate colorimetric method. The absorbance of the solution was measured and recorded at a wavelength of 345 nm with a spectrophotometer. Readings were taken at 0.5-hour intervals over a continuous 3-hour period. Analysis of the absorbance data yielded a relative standard deviation (RSD) of 1.48%, indicating that the sample solution remained stable for 3 hours and met the requirements for subsequent analysis. These results are presented in Table 3 .

Six 0.1 g samples of fine Gnetum parvifolium powder were precisely weighed and transferred to 50 mL ground-mouth conical flasks. To each sample, 4.0 mg of rutin standard, dried to a constant weight, was added. Total flavonoids were extracted according to the procedure in Section “3.2”, followed by color development using the aluminum nitrate colorimetric method. The absorbance of the resulting solution was measured at 345 nm with a UV-visible spectrophotometer. Statistical analysis determined the average sample recovery rate to be 98.07% with a relative standard deviation (RSD) of 1.85%. This result indicates that the method provides good accuracy and reliable measurements, as detailed in Table 4 .

With ethanol as the extractant, the total flavonoid yield from Gnetum parvifolium was used as the evaluation index to compare three methods: room temperature cold maceration (24 h), water bath reflux (0.5 h), and ultrasonic-assisted extraction (0.5 h). Subsequently, a single-factor experimental design was conducted to sequentially analyze the influence of several parameters on the flavonoid yield. With baseline conditions set at a 70% ethanol concentration, 20 minutes of sonication, a 1:100 sample-to-solvent volume ratio, and a single extraction, the effects of the following factors were investigated: solvent concentration (30%, 50%, 70%, 90%), sonication time (20, 30, 40, 60 min), sample-to-solvent volume ratio (1:40, 1:60, 1:80, 1:100), and number of extractions (1, 2, or 3).

Based on the results of the single-factor experiments, an L⁹ (3⁴) orthogonal experiment was designed to investigate the three main factors that significantly affect the total flavonoid yield from Gnetum parvifolium (extraction solvent concentration, sample-to-solvent volume ratio, and number of extractions) and their interactions. The factor levels for this experiment are presented in Table 5 .

The experimental procedure was based on the method of Xu Xianghao et al. [11] . Aliquots (1 mL) of total flavonoid extracts from Gnetum parvifolium were prepared at various concentrations (0.60, 0.90, 1.20, 1.50, 2.00, and 3.00 μg/mL). Each aliquot was added to 3 mL of 0.200 mmol/L DPPH solution in a brown test tube. The mixtures were incubated for 0.5 h in the dark. Subsequently, absorbance was measured at 517 nm using a UV-Vis spectrophotometer. Vitamin C (Vc) served as the positive control, and the DPPH free radical scavenging rate was calculated using Equation (1).

$\text{DPPH radical scavenging rate}=\left[{\text{A}}_0-\left({\text{A}}_1-{\text{A}}_2\right)\right]/{\text{A}}_0\times 100\%$ (1)

In the formula: A₀ is the absorbance value of 1.0 mL anhydrous ethanol + 3.0 mL DPPH solution; A₁ is the absorbance value of 1.0 mL flavonoid extract + 3.0 mL DPPH ethanol solution; A₂ is the absorbance value of 1.0 mL flavonoid extract + 3.0 mL anhydrous ethanol.

All experiments were repeated 3 times, orthogonal experimental design and analysis were performed using software IBS SPSS Statistics 24.0, and data were processed using software Microsoft Excel.

Spectral scan analysis revealed that both the rutin standard solution and the sample solution exhibited a distinct absorption maximum at a wavelength of 345 nm ( Figure 1 ). Given the significant absorption response of the target component at this wavelength, 345 nm was selected as the determination wavelength for this experiment to ensure effective detection.

As shown in Figure 2 , linear regression analysis of the rutin standard curve yielded the equation y = 0.1042x − 0.0708 and a correlation coefficient (R²) of 0.9995. This value indicates a highly significant linear relationship between rutin mass concentration and absorbance within the 0.0032 - 0.0096 mg/mL range, which validates the standard curve as a reliable basis for quantification [12] .

As shown in Figures 3-7 , a systematic analysis of the single-factor variables indicated that the maximum yield of total flavonoids from Gnetum parvifolium was obtained using ultrasound-assisted extraction under the following conditions: a 90% ethanol solvent, a sample-to-solvent volume ratio of 1:100, sonication for 60 minutes, and three extraction cycles. However, prolonged sonication can damage the instrument, and the yield increase from 20 to 60 minutes was minimal. Therefore, to balance extraction efficiency and equipment preservation, a standard sonication time of 20 minutes was adopted for all subsequent experiments.

A statistical analysis of the orthogonal test data in Table 6 was performed to determine the influence of three variables—solvent concentration (A), sample-to-solvent volume ratio (B), and number of extractions (C)—on the total flavonoid yield from Gnetum parvifolium. The analysis indicated that the order of factor influence was A > C > B. The optimal extraction protocol was identified as A₃B₁C₃, which corresponds to using a 90% ethanol solution, a 1:60 sample-to-solvent volume ratio, and performing three extraction cycles. This combination of process parameters effectively enhances the extraction efficiency of total flavonoids from Gnetum parvifolium ( Table 7 ).

Note: P < 0.05 indicates a significant effect.

To ensure the accuracy, stability, and reproducibility of the results, validation experiments were conducted in triplicate using the optimal combination of conditions. As presented in Table 8 , the average yield of total flavonoids from Gnetum parvifolium was 47.79 mg/g, which aligns with the predicted outcome. This confirms that the process identified through the orthogonal experiment is stable and viable, establishing A₃B₁C₃ as the optimal set of conditions for the extraction process.

The DPPH free radical scavenging assay is a widely used method for the in vitro evaluation of antioxidant capacity. A DPPH ethanol solution is purple due to the selective light absorption by its stable free radical structure, exhibiting peak absorbance at a wavelength of 517 nm. When an antioxidant reacts with DPPH free radicals via electron or hydrogen atom transfer, the solution’s color fades, and its absorbance decreases. This reduction in absorbance allows for the quantitative assessment of antioxidant activity. The experimental data in Figure 8 indicate that within the concentration range of 0.60 - 3.00 μg/mL, flavonoids from Gnetum parvifolium exhibit a scavenging effect on DPPH free radicals. Furthermore, this scavenging ability increases with the concentration of total flavonoids. However, their scavenging rate was significantly lower than that of ascorbic acid at the same mass concentration.